Genomic dissection of an extended phenotype: Oak galling by a cynipid gall wasp

Jack Hearn; Mark Blaxter; Karsten Schönrogge; José-Luis Nieves-Aldrey; Juli Pujade-Villar; Elisabeth Huguet; Jean-Michel Drezen; Joseph D Shorthouse; Graham N Stone

doi:10.1371/journal.pgen.1008398

. 2019 Nov 4;15(11):e1008398. doi: 10.1371/journal.pgen.1008398

Genomic dissection of an extended phenotype: Oak galling by a cynipid gall wasp

Jack Hearn ^1,^2,^*, Mark Blaxter ², Karsten Schönrogge ³, José-Luis Nieves-Aldrey ⁴, Juli Pujade-Villar ⁵, Elisabeth Huguet ⁶, Jean-Michel Drezen ⁶, Joseph D Shorthouse ⁷, Graham N Stone ^2,^*

Editor: Gregory P Copenhaver⁸

PMCID: PMC6855507 PMID: 31682601

Abstract

Galls are plant tissues whose development is induced by another organism for the inducer's benefit. 30,000 arthropod species induce galls, and in most cases the inducing effectors and target plant systems are unknown. Cynipid gall wasps are a speciose monophyletic radiation that induce structurally complex galls on oaks and other plants. We used a model system comprising the gall wasp Biorhiza pallida and the oak Quercus robur to characterise inducer and host plant gene expression at defined stages through the development of galled and ungalled plant tissues, and tested alternative hypotheses for the origin and type of galling effectors and plant metabolic pathways involved. Oak gene expression patterns diverged markedly during development of galled and normal buds. Young galls showed elevated expression of oak genes similar to legume root nodule Nod factor-induced early nodulin (ENOD) genes and developmental parallels with oak buds. In contrast, mature galls showed substantially different patterns of gene expression to mature leaves. While most oak transcripts could be functionally annotated, many gall wasp transcripts of interest were novel. We found no evidence in the gall wasp for involvement of third-party symbionts in gall induction, for effector delivery using virus-like-particles, or for gallwasp expression of genes coding for plant hormones. Many differentially and highly expressed genes in young larvae encoded secretory peptides, which we hypothesise are effector proteins exported to plant tissues. Specifically, we propose that host arabinogalactan proteins and gall wasp chitinases interact in young galls to generate a somatic embryogenesis-like process in oak tissues surrounding the gall wasp larvae. Gall wasp larvae also expressed genes encoding multiple plant cell wall degrading enzymes (PCWDEs). These have functional orthologues in other gall inducing cynipids but not in figitid parasitoid sister groups, suggesting that they may be evolutionary innovations associated with cynipid gall induction.

Author summary

Plant galls are induced by organisms that manipulate host plant development to produce novel structures. The organisms involved range from mutualistic (such as nitrogen fixing bacteria) to parasitic. In the case of parasites, the gall benefits only the gall-inducing partner. A wide range of organisms can induce galls, but the processes involved are understood only for some bacterial and fungal galls. Cynipid gall wasps induce diverse and structurally complex galls, particularly on oaks (Quercus). We used transcriptome and genome sequencing for one gall wasp and its host oak to identify genes active in gall development. On the plant side, when compared to normally developing bud tissues, young gall tissues showed elevated expression of loci similar to those found in nitrogen-fixing root nodules of leguminous plants. On the wasp side, we found no evidence for involvement of viruses or microorganisms carried by the insects in gall induction or delivery of inducing stimuli. We found that gall wasps express many genes whose products may be secreted to the host, including enzymes that degrade plant cell walls. Genome comparisons between galling and non-galling relatives showed cell wall-degrading enzymes are restricted to gall inducers, and hence potentially key components of a gall inducing lifestyle.

Introduction

Galls are plant tissues whose development is induced by other organisms, including viruses, bacteria, protozoa, fungi, nematodes, mites and insects [1,2]. Galls are not simply wound responses, but are specific plant growth responses to infection that provide the inducing organism with nutrition and protection from external abiotic and biotic challenges [1,2]. Galls represent sinks for plant nutrients and metabolites, diverting resources towards gall inducer development that could otherwise be invested in plant maintenance, growth and reproduction [3–6]. While some galling interactions are clearly mutualistic (e.g. legume root nodules induced by nitrogen-fixing Rhizobium bacteria [7] and the galls induced in figs by their pollinating wasps [8–10]), most have no apparent benefit for the host plant. That the impact of galling can be strongly negative for the host plant is shown by the fact that gall inducers include many serious pests of plants in agriculture and forestry [11–13], and important biocontrol agents of invasive plants [14,15].

The molecular mechanisms underpinning gall induction are only well understood in a small number of microbial systems, including Rhizobium induction of root nodules, crown gall induction by Agrobacterium tumefaciens, and the parasitic fungus Ustilago maydis of maize [7,16,17]. These systems are characterised by (i) release of host plant molecules that are detected by the gall inducing organism and (ii) release by the gall inducer of molecules, termed effectors, that (iii) interact with specific plant receptors. These interactions (iv) result in characteristic plant growth responses to produce gall tissues.

In contrast to microbial systems, gall induction by animals is poorly understood for all but a tiny minority of economically important species, including root knot and cyst nematodes (Heterodera, Globodera and Meloidogyne species) [18–20], Hessian fly (Mayetiola destructor) [21–24] and Phylloxera [25,26]. Nevertheless, with an estimated 30,000 species [27], gall inducing arthropods (and particularly insects) are abundant and ecologically important components of many biological communities [27]. A key feature of insect-induced galls is that, relative to microbial galls, most are structurally both more defined and more complex, with clear differentiation between distinct tissues associated with inducer nutrition and protection [1,28,29]. Because gall traits are characteristic of the inducing species rather than the plant host, Dawkins [30] and many others [26,31–34] have viewed plant galls as the extended phenotypes of gall inducer genes, stimulating research into the adaptive significance of insect gall morphologies. Similarity in the gall morphologies induced by phylogenetically divergent insect lineages also raises the question of whether different inducers use similar effectors to target similar plant systems to reset or divert normal plant development [6,26,35,36].

The oak gall wasps (Hymenoptera: Cynipidae: Cynipini) are one of the most species-rich lineages of galling insects, with ca.1500 species that induce diverse and structurally complex galls on oaks (Quercus) and related Fagaceae [37–39]. The lifecycles of most oak gall wasps involve obligate alternation between a sexual and an asexual generation, each of which induces a different gall phenotype on a specific organ (leaf, bud, flower, fruit, shoot or root) on specific host Fagaceae [40,41]. Evolutionary shifts between host plant lineages have been extremely rare in gall wasps [39,42]. These patterns suggest that the association between gall wasps and their hosts is physiologically intimate, involving gall wasp manipulation of specific plant developmental pathways. Several properties of oak-gall wasp interactions make this system an interesting one for exploration of processes associated with gall induction and development: (i) the developmental anatomy of some oak galls has been studied in considerable detail, allowing division of gall development into recognisable stages [1,28,41,43–46] between which patterns of gene expression can be compared; (ii) genomic resources are increasingly available for some oaks [47,48] and oak gall wasps [49–51]; (iii) oak gall diversity makes this an excellent system in which to study the adaptive significance of gall phenotypes [29,38,52,53]; and (iv) the ancestral life history strategy of cynipids prior to evolution of gall induction is known. Gall wasps evolved from parasitoids similar to extant figitid cynipoids, which develop within the bodies of herbivorous insects living within plant tissues [39,54,55]. Genomic comparisons between cynipid gall inducers, and between cynipids and their figitid sister group, thus provide a means to identify innovations associated with diversification of gall phenotypes and the origins of gall induction.

Initiation of cynipid gall induction begins when a female oviposits into a specific location in meristematic tissues on a specific host plant [40,41,53]. When the first instar gall wasp larva (ca. 0.3mm long) hatches from the egg, the surrounding plant cells disintegrate and/or de-differentiate to form a chamber lined with callus-like tissue into which the larva moves [1,28]. While initiation of cynipid gall development may involve maternal secretions in the egg [1,49], it has long been known that if the gall wasp larva is killed, gall development ceases, implying that larval production of inducing effectors is required for continued gall growth [41,56]. Cynipid gall development following initiation can be divided into three stages—Early, Growth, and Mature [28,57]—that represent natural sampling points for associated processes. These stages are illustrated for the sexual generation gall induced by our study species, Biorhiza pallida, on leaf buds of pedunculate oak, Quercus robur, in Fig 1. This gall type was selected because its developmental anatomy [1,44,45,58] and physiology [59,60] have been relatively well studied, and also because a single founding female lays 50–400 asexually-produced eggs into a single bud [61,62]. The resulting larvae develop in synchrony, facilitating the detection of transcripts from multiple larvae of the same age from whole gall RNA extractions. In Early stage galls (Fig 1A, 1B and 1C), larvae are < 1 mm long. Host tissues lining the larval chamber divide rapidly and differentiate into specialised nutritive cells that provide all larval food [63]. A thin wall of sclerenchyma separates each inner gall (larval chamber) from non-nutritive outer gall tissues [37,43,63,64]. In Growth stage galls (Fig 1A, 1D and 1E) the larvae remain very small, but development and differentiation of outer gall tissues is extensive [1,65]. The nutritive cells multiply (hyperplasy) and enlarge (hypertrophy), become multinucleate, undergo endoreduplication of their chromosomes, and accumulate high concentrations of lipids, carbohydrates and nitrogenous compounds [44,45,59,60,62,63,66]. Outside the nutritive core, gall growth proceeds rapidly by hyperplasy and hypertrophy of layers of parenchyma [37,46]. Extensive development of vascular tissues facilitates import of nutrients and water from the host (Fig 1A, 1E and 1G). In the transition to Mature stage galls (Fig 1A, 1F and 1G), the larvae grow rapidly until they exhaust the nutritive tissues and reach the surrounding sclerenchyma layer. Tissues outside the larval chamber become lignified and desiccated. The larvae then pupate and adult B. pallida emerge two to three weeks later.

Normal (ungalled) buds develop very differently (Fig 1A), through six developmental stages recognised in previous work on Quercus robur and other trees [67] that span the transition from dormant buds (S0) to fully open leaves (S5). The sexual generation galls of Biorhiza pallida develop [1,28] slowly in comparison to ungalled buds, such that fully flushed leaves develop from ungalled buds in the same time taken for galled buds to reach the Early and Growth stages (Fig 1A).

Very little is known about the plant metabolic processes underlying cynipid gall development, the nature of the effectors involved, or their source(s) [68–70]. The aim of this study was to provide the first characterisation of oak and gall wasp gene expression in gall development immediately after initiation, and to discriminate between alternative hypotheses for the source(s) of gall wasp-associated effectors. To identify plant processes involved through gall development, we identified oak genes with strong stage-specific patterns of differential expression in Early, Growth and Mature stage galls. To identify differences between galled and normally developing tissues, we identified oak genes with contrasting expression trajectories in development of galled and ungalled (control) buds. We identified gall wasp genes with strong stage-specific patterns of differential expression in the same gall stages, and used functional annotation to infer their potential roles in gall wasp development and interaction with their oak host. Finally, we use de novo genomic data for a panel of gall inducing cynipids and parasitoid figitids to identify novel gall wasp genes associated with a gall inducing life history. We use these approaches to address four general questions in cynipid gall development, outlined in detail below.

What, in plant terms, is a cynipid gall?

Cynipid galls include multiple, organized tissue types, and can be considered novel plant organs [59,60]. We tested two existing and mutually compatible hypotheses based on previous work (see Discussion): ‘galls as ectopic food storage organs’, and ‘galls as modified somatic embryos’ (Fig 2A) [59,60]. The first hypothesis stems from the observation that cynipid gall nutritive tissues express biotin carboxylase carrier protein (BCCP) [59,69], a component of the triacylglycerol lipid synthesis pathway associated with lipid food storage in seeds and tubers [59,60,70]. Cynipids could manipulate lipid metabolism to enhance the development of tissues on which the larvae feed, and galls could thus be modelled as ectopic plant food storage organs. The second hypothesis stems from the observation that Early stage cynipid galls display cellular dedifferentiation and rapid cell proliferation [45,59,60], phenomena observed during normal somatic embryogenesis [71]. Cynipid galls may thus also be modelled as developing from ectopic somatic embryos (Fig 2A). We explored the nature of plant-derived gall tissue by transcriptomic analysis, in particular seeking to identify genes and systems typical of storage organs and somatic embryos that were highly (or only) expressed in Early versus later gall tissues, and genes whose trajectory of expression through gall development contrasted with patterns observed in normally developing oak buds.

Fig 2 — A. Oak gene expression patterns predicted under the ectopic food storage organ and somatic embryo hypotheses. B. Alternative hypotheses for the origin(s) and delivery of gall development effectors. C. Alternative possible effectors. In each of A-C, boxed texts summarise expectations for patterns in transcriptome data if the hypotheses are supported (see main text Introduction and Discussion for explanation). The circled ‘S’ within the gall wasp larva represents a possible internal symbiont, such as a virus or bacterium.

What organism induces the B. pallida oak gall?

The observation that cynipid gall development requires a living larva [43,56] could indicate endogenous gall wasp production of effectors required for gall development. However, the same observation could indicate a key role for one or more endosymbionts (viruses, bacteria or other microorganisms [72]) in effector production and/or delivery (Fig 2B). Host plant manipulation by a Wolbachia symbiont has been shown for a leaf mining moth [73], and bacterial symbionts are required for successful gall development in several non-cynipid systems [74–76]. Symbionts can fulfil other crucial roles in gall tissues—for example, the larvae of ambrosia gall midges (Diptera; Cecidomyiidae) feed on fungal symbionts whose spores are inoculated into host plant tissues by the ovipositing female [77,78]. Microbial symbionts could thus be involved in cynipid gall induction, development or physiology. We hypothesised that involvement of coinfecting and gall wasp-transmitted symbionts should be associated with repeatable detection of (i) symbiont genomes in assemblies generated for female gall wasps, and (ii) symbiont transcripts in transcriptomes for relevant gall developmental stages.

We also assessed the potential for gall wasps to use genes derived from viral symbionts to deliver endogenous or symbiont effectors (Fig 2B). Parasitoid wasps export DNA or proteins that suppress insect host immune responses [79,80] by packaging them within virus-like-particles (VLPs) [79,81], the components of which are coded for by viral genes that have been incorporated into the parasitoid genome [80,81]. Leptopilina species in the Figitidae sister group of cynipids use VLPs to deliver host immune-suppressing factors [82], and gall wasps could in principle use a similar delivery system. Involvement of VLPs in signal delivery should be associated with detection of virally-derived genes in the gall wasp genome (as indicated by flanking of putative VLP sequences by unambiguously insect genes and/or presence of introns) and detection of VLP transcripts during gall development.

What processes are involved in gall wasp manipulation of plant development?

Gall wasp larvae remain very small during much of gall growth, and the assumption for over a century [83] has been that they excrete effectors into the liquid medium that surrounds them early in gall development (Fig 1A). Gall wasp larvae have a sealed anus and enlarged salivary glands, making salivary (and potentially also Malpighian tubule) secretions plausible candidates for the source of induction effectors [83,84]. Gall development could be induced through gall wasp (or associated symbiont) manipulation of plant hormone cascades (by secretion of hormone homologues or analogues, or enzymes that modulate host small molecule metabolism), secretion of direct protein regulators of plant developmental processes, and/or secretion of enzymes that change the physiology of plant cells (Fig 2C, and see Discussion). We used gall wasp genomic and larval transcriptome data to identify gall wasp genes with potential to interact directly with plant metabolism, and that were differentially expressed between Early and later gall stages. We also sought to identify parallel expression patterns in oak genes involved in hormone transport, receptors and target systems, and to identify gall-specific processes by comparison of expression trajectories in gall and normal (non-galled) oak tissues.

Which gall wasp genes involved in gall development are novel compared to non-galling sister groups?

We hypothesise that, since diverging from a parasitoid common ancestor shared with their Figitidae sister group approximately 80 million years ago [55], gall wasps will have evolved shared novel traits associated with herbivory and host plant manipulation. We therefore looked for homologues of Biorhiza pallida genes that were highly expressed during gall development in the genomes of exemplar species in six additional gall-inducing cynipid tribes and two species of non-galling parasitoid Figitidae. Our rationale is that genes that are highly expressed during gall development and shared by gall wasps but absent from their parasitoid sister groups represent candidates for evolutionary innovations associated with a gall-inducing lifestyle.

We show that while gall tissues of the sexual generation of Biorhiza pallida share some similarities with normally developing buds, patterns of oak gene expression show consistent differences between these two structures. We find no evidence for involvement of known symbionts in cynipid gall induction, or for export of gall wasp gene products in virus-like particles. We find the Biorhiza pallida genome to encode a suite a plant cell wall degrading enzymes, including cellulases, that are expressed at various points through gall development. These genes are shared with a phylogenetically diverse panel of gall wasps but are absent from non-galling sister groups, suggesting that they are shared evolutionary novelties associated with a gall inducing life history. Finally, we develop a hypothesis for chitinase-mediated gall wasp modification of plant arabinogalactan molecules during gall development.

Results

Generation of RNA-seq datasets for B. pallida galls and ungalled oak buds

We performed dual-RNA-seq [85] on four biologically independent replicates for each of Early, Growth and Mature stage galls (Fig 1; S1 Fig) and also for each of five of the six developmental stages of normally-developing ungalled pedunculate oak (Q. robur) buds from dormant buds (S0) through bud expansion (S1, S3, S4) to fully-opened leaves (S5) [47,67] (Fig 1G). Raw Illumina Hi-Seq data have been deposited in ENA under BioProjects PRJEB13357, PRJEB13424, and PRJEB32849. Detailed descriptions of sequence data processing are provided in Materials and Methods.

We generated a co-assembly of all the gall and non-gall RNA-seq data and then separated and filtered genes (component groups in TRINITY terminology) by taxonomic origin to identify transcripts deriving from oak, gall wasp, parasitoid, viral and fungal origins. We used a genomic approach to seek potential bacterial symbionts as the poly-A selection protocol we used is unlikely to detect bacterial transcripts (see below). The combined assembly spanned 592 megabases (Mb) in 773,991 transcripts with an N50 of 1,145 bases. Further filtering to remove ribosomal RNA sequences and retain only transcripts encoding an open reading frame resulted in a much reduced dataset, comprising 50,708 gall wasp transcripts corresponding to 24,916 genes and 178,522 oak transcripts corresponding to 53,529 genes (Table 1). Removing low-expression genes in DESeq2 (see Methods) retained 19,720 gall wasp genes and 42,926 oak genes in gall tissues, and 48,756 oak genes for normally developing oak bud tissues considered by each analysis. As we would expect given the increasing size of the gall wasp larvae, the proportion of gall wasp-derived reads in the entire gall transcriptome (the 'cecidome' [86]) rose from 2.3–4.0% in Early stage galls to 2.9–6.8% and 18–42% in Growth and Mature stages, respectively (Table 2). Oak and gall wasp transcripts comprised 78.6–99.6% of the total across the 12 gall replicates (Table 2). All galls yielded parasitoid-associated transcripts, with the lowest load in Early stage galls (Table 2, S2 Table). Viral and fungal sequences (considered further below) made up 0.1–8.2% and 0–5.3% of transcripts, respectively (Table 2).

Table 1. Assembly metrics for the separated gall wasp (Biorhiza pallida) and oak bud plus gall tissue (Quercus robur) transcriptomes.

Assembly	Gall wasp– B. pallida	Oak—Q. robur
N50 ^*	2338	1864
Number of transcripts ^**	50708	178522
Number of genes ^***	24916	53529
Transcriptome span (Mb)	78	232
BUSCO complete (partial) % ^§	97.7 (98.4)	87.1 (98.0)

Open in a new tab

* N50 is a weighted median contig length, such that 50% of the assembly is found in contigs of this length or greater.

** The total number of contigs produced by the TRINITY assembler.

*** The number of groups identified by TRINITY as representing separate loci.

§ BUSCO complete (and cumulative total including partially complete) genes after comparison to the Eukaryota database

Table 2. Taxonomic assignments of gall dual-RNA-Seq reads, by stage and replicate (1–12), expressed as a % of the total mapped read set.

Full read counts and additional taxonomic sub-categorisations are provided in S1 and S2 Tables.

Stage	Early				Growth				Mature
Taxon of origin	1	2	3	4	5	6	7	8	9	10	11	12
Q. robur	95.6	97.3	96.6	87.8	94.2	94.7	90.7	84.9	41.8	57.1	56.8	70.8
B. pallida	3.9	2.3	3.0	3.3	2.9	3.7	6.7	5.80	41.0	32.4	19.8	16.3
Parasitoids	1.0	0.2	0.2	0.7	2.5	0.7	1.0	7.1	10.1	7.1	13.6	3.7
Fungi	0.0	0.0	0.0	5.3	0.2	0.1	0.1	0.5	3.7	1.5	2.0	1.0
Virus	1.4	0.1	0.1	3.2	0.1	0.8	1.4	3.3	2.4	1.5	5.8	8.2

Open in a new tab

Gall wasp and oak genes show consistent patterns of stage-specific expression in Biorhiza pallida galls

Both gall wasp genes (Fig 3A) and oak genes (Fig 3B) in gall tissues showed largely consistent expression patterns across replicates within a given gall developmental stage. Gall stage accounted for 82% of variance in gall wasp gene expression and 50% of variance in oak gene expression. All gall wasp and oak genes showing differential expression between gall developmental stages at a log₂-fold change of plus or minus one are listed in S8 and S9 Files, respectively. The numbers of gall wasp genes differentially expressed at more than our ±1-fold log₂ difference threshold between Early and later stages increased during gall development, from 79 between Early and Growth stages to 371 between Early and Mature stages and 249 between Growth and Mature stages (Table 3). Differentially expressed gall wasp genes showed a strong bias towards higher expression in the younger gall stage in all between-stage comparisons (Table 3).

Table 3. Counts of gall wasp and oak genes differentially expressed (DE) in gall tissues.

Host oak tissues are further summarised into the numbers of genes DE between gall stages that were not differentially expressed between any stages in normal bud development.

*Gall wasp (B. pallida)*	Early vs. Growth		Early vs. Mature		Growth vs. Mature
Direction of differential gene expression in earlier stage	Up in Early	Down in Early	Up in Early	Down in Early	Up in Growth	Down in Growth
Total number of DE^* genes for this gall stage comparison	75	4	283	88	210	39
*Oak (Q. robur)*	Early vs. Growth		Early vs. Mature		Growth vs. Mature
Direction of gene expression in earlier stage	Up in Early	Down in Early	Up in Early	Down in Early	Up in Growth	Down in Growth
Total number DE* genes in galls	546	747	1045	845	63	35
Number of DE genes that were not DE between any stages in normal bud development	197	491	421	594	34	32
% DE genes that were not DE between any stages in normal bud development	36%	66%	40%	70%	54%	91%

Open in a new tab

* DE genes are defined as those showing at least a one-fold log₂ change in DESeq2 analyses. The Bud (S0-S4) vs. Leaf comparison represents results for combined developing bud stages (S0, S1, S3, and S4) versus mature leaf tissue S5. vs. = versus.

Numbers of oak genes differentially expressed between gall stages showed a somewhat contrasting pattern to gall wasp genes, with high numbers differentially expressed between Early stage galls and each of Growth (1,293) and Mature (1,890) stages, but only 98 between Growth and Mature stages (Table 3). Also in contrast to gall wasp genes, differentially expressed oak genes in gall tissues showed no consistent bias towards higher expression in younger gall tissues (Table 3).

Oak gene expression in normally developing tissues shows partial differentiation between bud stages and strong differentiation between buds and leaves

Developmental stage explained 32% of the variance in oak gene expression in normally developing buds. In comparison to gall tissues, much lower relative numbers of transcripts from normal bud samples were identified as viral or fungal in origin (S3 Table). All oak genes showing differential expression between stages in normal bud development at a log₂-fold change of plus or minus one are listed in S10 File. Only small numbers were differentially expressed between bud stages S0, S1, S3 and S4 (1–31 genes, depending on the stages compared; Table 4), and gene expression patterns of these stages were incompletely separated (Fig 3C). In contrast, many genes (4,418) were differentially expressed between normal bud stage S4 and open leaves stage S5 (Table 4), and leaf tissues were clearly distinct from all earlier bud stages (Fig 3C). Lack of clear separation between bud stages S0-S4 was maintained if S5 replicates were removed from the analysis (S2 Fig). When data for bud stages S0-S4 were combined into a single bud developmental category (supported by lack of differential expression and overlapping PCA results, Fig 3C) the number of genes differentially expressed between bud stages S0-S4 and open leaves stage S5 rose to 4,981 (Table 4). In conclusion, identification of oak genes showing similar (or contrasting) trajectories of differential expression in gall and normal bud tissues thus primarily involves identifying those showing similar directions of differential expression between Early versus Growth or Mature gall stages, and between S4 buds and S5 leaves.

Table 4. Counts of gall wasp and oak genes differentially expressed (DE) in normally developing bud developmental stages.

Oak Buds (S0-S4) and Leaves (S5)	Bud S0 vs. Bud S1		Bud S1 vs. Bud S3		Bud S3 vs. Bud S4		Bud S4 vs. Leaf (S5)		Bud (S0-S4) vs. Leaf (S5)
Direction of differential gene expression	Up in S0	Down in S0	Up in S1	Down in S1	Up in S3	Down in S3	Up in S4	Down in S4	Up in S0-S4	Down in S0-S4
Total number of DE genes	1	26	5	31	3	12	2108	2310	2000	2881

Open in a new tab

Functional annotation of gall wasp and oak genes expressed during gall and normal bud development

Functional annotation of the gall wasp larval transcriptome was significantly poorer than for oak genes expressed in gall tissues, at 44% genes annotated versus 66% respectively. The subset of B. pallida genes differentially expressed between gall stages was more poorly annotated than the gall wasp transcriptome as a whole: functional annotations were identified for only 29% (22/75) of gall wasp genes more highly expressed in Early galls versus Growth and Mature stages, and none of the top five most highly expressed of these genes were annotated. Functional interpretation of changing gall wasp gene expression thus includes only a small proportion of the differentially expressed genes. In contrast, 89% (8225/9229) of differentially expressed oak genes for all contrasts in gall tissues and in normal buds were functionally annotated, allowing better-informed comparison of expression patterns in these two structures.

Galled and ungalled buds show diverging patterns of oak gene expression during development

To compare galled and normal oak bud tissues, we identified oak genes and GO terms in gall and normal bud tissues showing similar or contrasting trajectories of expression during development of their respective tissues. A striking feature of our results is that most of the genes differentially expressed during gall development were not differentially expressed in normally developing bud tissues, and vice versa. The extent of overlap in differentially expressed oak genes declined during development of gall and normal bud development, compatible with increasing divergence between galler-induced and normal developmental trajectories.

Few genes were differentially expressed between normal bud stages S0–S4 (Table 3) and only five were differentially expressed in both gall and normal bud tissues (S4 Table). Four showed parallel expression trajectories of increasing expression through development of which two are unannotated (TRINITY_DN48733_c0_g1 and TRINITY_DN82248_c2_g1); one is a glyceraldehyde-3-phosphate dehydrogenase (TRINITY_DN55651_c1_g1) and one a cytochrome P450 (TRINITY_DN80279_c2_g1). Only one gene, a galacturonosyltransferase-like 10 (TRINITY_DN65887_c0_g1), showed increasing expression during gall development but decreasing expression during development of normal buds. Of the >4400 oak genes differentially expressed between normal bud stage S4 and fully opened leaves S5 (Table 4), 8.9–23.3% showed similar patterns of increasing or decreasing differential expression in gall tissues, depending on the direction of change and gall stages compared (Fig 4, S4 Table). Broadly similar numbers of shared differentially expressed genes were identified when expression in normal buds was compared with expression in either Early versus Growth or Early versus Mature stage galls (Fig 4). Much smaller numbers of oak genes (0.5–3.6% of those differentially expressed between S4 and S5 during normal bud development) showed contrasting trajectories of differential expression in normal bud and gall tissues. For example, only 10 of the 2108 genes (0.5%) that showed a significant increase in expression between normal bud stage S4 and fully open leaves S5 also showed a significant decrease in expression between Early and Mature stage galls. Shared differentially expressed genes comprised a higher percentage of all genes differentially expressed in gall tissues than normal oak buds (S4 Table).

Fig 4 — Venn diagrams are shown for genes that are up- or down-regulated in the earlier stage of each comparison as for Tables 3 and 4. Red numbers are the number of genes unique to that contrast, relative to ungalled buds for gall tissues, and relative to gall tissues for ungalled buds. Purple numbers show genes that are differentially expressed in the same direction in both galled and ungalled buds. For example, there are 311 genes that are upregulated in Early *versus* Growth stage galls and also in ungalled buds between S4 buds *versus* S5 leaves. Blue and green numbers are for genes with contrasting expression trajectories in gall tissues *versus* ungalled buds: blue numbers count genes that are upregulated in Early *versus* later stage galls but downregulated in S4 buds *versus* S5 leaves, while and green numbers count genes that are downregulated in Early stage galls but upregulated in ungalled S4 buds. These data are also given in S4 Table.

To compare metabolism and development of gall and normal oak tissues, we identified oak GO terms with similar and contrasting trajectories of differential expression between gall developmental stages and between S4 bud and S5 leaf stages in normal bud development (Figs 4 and 5). Biological process (BP) and molecular function (MF) GO terms that were differentially expressed only during gall development are listed in S7 and S8 Tables, while those that showed similar patterns of differentially expression in both tissues (i.e. increasing in both or decreasing in both) are listed in S9 and S10 Tables. Biological process (BP) and molecular function (MF) GO terms that were differentially expressed during normal bud development are listed in S11 and S12 Tables.

Fig 5 — A: GO terms that were upregulated only in Early *versus* Growth stage gall tissues, and not in comparisons between any normal (ungalled) bud stages. B: GO terms that were upregulated both in Early *versus* Growth stage gall tissues and in normal buds (S4) *versus* open leaves (S5). C: GO terms that were downregulated only in Early *versus* Growth stage gall tissues, and not in comparisons between any normal (ungalled) bud stages. D: GO terms that were downregulated both in Early *versus* Growth stage gall tissues and in normal buds (S4) *versus* open leaves (S5). Circle size is scaled by–log₁₀ of the enrichment p-value with scales inset next to each plot. Groups of GO terms are coloured by cluster as classified by REVIGO treemaps (S3 Fig). There are fewer GO terms than for the equivalent *TopGO* comparisons as REVIGO reduced the redundancy in enriched GO terms. Numbers in the figures refer to the following GO terms and reference numbers: 1 = response to brassinosteroid (GO:0009741), 2 = cutin biosynthetic process (GO:0010143), 3 = thymidine metabolic process (GO:0046104), 4 = leaf vascular tissue pattern formation (GO:0010305), 5 = cellulose catabolic process (GO:0030245), 6 = asymmetric cell division (GO:0008356), 7 = wax biosynthetic process (GO:0010025), 8 = lignan biosynthetic process (GO:0009807), 9 = regulation of organ growth (GO:0046620), 10 = guard cell differentiation (GO:0010052), 11 = fatty acid metabolic process (GO:0006631), 12 = auxin-activated signalling pathway (GO:0009734), 13 = regulation of seed germination (GO:0010029), 14 = cytokinin-activated signalling pathway (GO:0009736), 15 = lipid catabolic process (GO:0016042), 16 = plant-type secondary cell wall biogenesis (GO:0009834), 17 = response to zinc ion (GO:0010043), 18 = gibberellic acid homeostasis (GO:0010336), 19 = chorismate biosynthetic process (GO:0009423), 20 = L-glutamate import across plasma membrane (GO:0098712), 21 = fruit ripening (GO:0009835), 22 = carbohydrate metabolic process (GO:0005975), 23 = proteolysis (GO:0006508), 24 = response to osmotic stress (GO:0006970), 25 = wound healing (GO:0042060), 26 = oxidation-reduction process (GO:0055114), 27 = response to herbicide (GO:0009635), 28 = regeneration (GO:0031099), 29 = response to biotic stimulus (GO:0009607), 30 = lignin biosynthetic process (GO:0009809), 31 = one-carbon metabolic process (GO:0006730), 32 = S-adenosylhomocysteine catabolic process (GO:0019510), 33 = shikimate metabolic process (GO:0019632), 34 = UDP-N-acetylglucosamine biosynthetic process (GO:0006048), 35 = ether metabolic process (GO:0018904), 36 = nitrate assimilation (GO:0042128), 37 = S-adenosylmethionine biosynthetic process (GO:0006556), 38 = sodium ion homeostasis (GO:0055078), 39 = mucilage biosynthetic process involved in seed coat development (GO:0048354), 40 = aromatic amino acid family biosynthetic process (GO:0009073), 41 = gluconeogenesis (GO:0006094), 42 = ureide catabolic process (GO:0010136), 43 = uracil catabolic process (GO:0006212), 44 = oxidative phosphorylation (GO:0006119), 45 = response to water deprivation (GO:0009414), 46 = beta-alanine biosynthetic process (GO:0019483), 47 = glucuronoxylan biosynthetic process (GO:0010417), 48 = cell wall organization (GO:0071555), 49 = developmental process (GO:0032502), 50 = cell division (GO:0051301), 51 = positive regulation of mitotic recombination (GO:0045951), 52 = cellular response to gamma radiation (GO:0071480), 53 = cortical cytoskeleton organization (GO:0030865), 54 = cell proliferation (GO:0008283), 55 = guard mother cell differentiation (GO:0010444), 56 = cytokinesis by cell plate formation (GO:0000911), 57 = microtubule-based movement (GO:0007018), 58 = microtubule-based process (GO:0007017), 59 = thigmotropism (GO:0009652), 60 = stomatal complex patterning (GO:0010375), 61 = cell cycle (GO:0007049), 62 = regulation of transcription, DNA-templated (GO:0006355), 63 = cotyledon development (GO:0048825), 64 = positive regulation of ubiquitin protein ligase activity (GO:1904668), 65 = positive regulation of anthocyanin metabolic process (GO:0031539), 66 = maintenance of floral organ identity (GO:0048497), 67 = positive regulation of cell proliferation (GO:0008284), 68 = epidermal cell fate specification (GO:0009957), 69 = endosperm development (GO:0009960), 70 = anastral spindle assembly involved in male meiosis (GO:0009971), 71 = DNA endoreduplication (GO:0042023), 72 = trichome branching (GO:0010091), 73 = mitotic spindle assembly checkpoint (GO:0007094), 74 = meiotic cell cycle (GO:0051321), 75 = phragmoplast microtubule organization (GO:0080175), 76 = polarity specification of adaxial/abaxial axis (GO:0009944), 77 = plant-type secondary cell wall biogenesis (GO:0009834), 78 = response to nitrate (GO:0010167), 79 = plasmodesmata-mediated intercellular transport (GO:0010497), 80 = lignin biosynthetic process (GO:0009809), 81 = catabolic process (GO:0009056), 82 = fruit dehiscence (GO:0010047), 83 = aging (GO:0007568), 84 = cellular response to salt stress (GO:0071472), 85 = S-adenosylmethionine biosynthetic process (GO:0006556), 86 = ethylene biosynthetic process (GO:0009693), 87 = regulation of salicylic acid metabolic process (GO:0010337), 88 = 3'-UTR-mediated mRNA destabilization (GO:0061158), 89 = nitrate transport (GO:0015706), 90 = oxidation-reduction process (GO:0055114), 91 = fruit ripening (GO:0009835), 92 = glycine catabolic process (GO:0006546), 93 = nitrate assimilation (GO:0042128), 94 = oligopeptide transport (GO:0006857), 95 = glucuronoxylan biosynthetic process (GO:0010417), 96 = cell wall organization (GO:0071555).

Biological process GO terms showing similar expression trajectories in gall tissues and normal buds. Forty-seven biological process GO terms showed significantly higher expression in Early versus Growth stage galls and in normally developing S4 buds versus S5 leaves (S9 Table). These were grouped into 5 metabolic clusters by REVIGO treemap analysis (Fig 5B). As we might expect given rapid growth of both structures, GO terms elevated early in development of both galls and normal buds included many terms for cell division, cell cycle and proliferation (Fig 5B. 19/47 GO terms identified by orange circles) including DNA replication and endoreduplication/ regulation of DNA replication. Twenty-three biological GO term processes showed significantly higher expression in Growth versus Early stage galls and in normally developing S5 leaves versus S4 buds (S9 Table). These were grouped into seven metabolic clusters by REVIGO treemap analysis (Fig 5D), comprising processes associated with aging, catabolism, fruit dehiscence, lignin biosynthesis, secondary plant cell wall biogenesis, intercellular transport and response to nitrate. Fifteen of the same GO terms were shared with the 26 upregulated in Mature versus Early stage galls (S9 Table).

Biological process GO terms showing contrasting expression trajectories in gall tissues and normal buds. Fifteen biological process GO terms were upregulated in Early versus Growth stage galls, but not differentially expressed between any stages in normal bud development (S7 Table). These were grouped by REVIGO treemap analysis into 5 clusters of linked metabolic processes: response to brassinosteroid, leaf vascular tissue pattern formation, thymidine metabolism, cellulose catabolism and cutin biosynthesis (Fig 5A). Forty-five GO term processes were upregulated in Growth versus Early stage galls, but not differentially expressed between any stages in normal bud development (S7 Table). These were grouped into seven metabolic clusters by REVIGO treemap analysis (Fig 5C), comprising processes associated with carbohydrate metabolism, chorismate biosynthesis, fruit ripening, gibberellic acid homeostasis, L-glutamate transport, secondary plant cell wall biogenesis and response to Zinc ions. Fourteen of the same GO terms were shared with the 30 upregulated in Mature versus Early stage galls (S7 Table).

Our data thus suggest that galls of Biorhiza pallida share many metabolic and developmental characteristics with normally developing buds, but also that some genes show contrasting trajectories of differential expression between these tissues. Most genes that are differentially expressed between stages in gall development are also constitutively highly expressed in buds throughout development (S22 Table, mean expression of genes that are DE in galls but not buds for both experiments). In Early galls versus Growth this includes an endogenous endoglucanase and polygalacturonase (TRINITY_DN75537_c1_g1 and TRINITY_DN75938_c0_g1).

Further, many genes that are differentially expressed through gall development are not differentially expressed between any stages in normal leaf bud development. Similarity in oak gene expression patterns between gall tissues and normal buds declined as these tissues develop.

Patterns in the expression of oak genes associated with alternative hypotheses for gall development

We compared expression patterns in galled and normally developing buds for specific candidate oak genes (S6 Table) relevant to alternative hypotheses of gall development.

BCCP. The 'galls as ectopic food storage organs' hypothesis predicts high expression of the acetyl CoA component BCCP during development of nutritive tissues in Early and Growth stage galls relative to Mature stage galls (in which nutritive tissues have been entirely consumed). In contrast to this prediction, we found no differential expression of BCCP across developmental stages of either gall tissues or normally developing buds, and two of five expressed copies were expressed highly in all gall and normal bud stages sampled (S6 Table).
ENOD genes. The ‘galls as modified somatic embryos’ hypothesis predicts high expression of ENOD genes early in gall development. We found three genes coding for ENOD proteins containing phytocyanin domains to be significantly overexpressed in Early versus Growth stage galls. Of these, two were highly expressed in absolute terms (S6 Table), and were also expressed in normal bud tissues. One showed increasing expression during development of both gall tissues and normal buds (S4 to S5), while the second was not differentially expressed between any normal bud stages. These two proteins are present in eleven GO term gene groups that are enriched in Early versus later stage galls, including terms for cytokinesis, histone phosphorylation, anaphase, cell division and cell cycle. These data are compatible with a potential role for ENOD genes early in gall development.
Oak chitinases. Chitinases can protect plant tissues against fungal attack, and are also involved in release of arabinogalactan signalling molecules from the phytocyanin-containing ENOD proteins considered above. The latter mechanism could be involved in gall development through action of one or both of gall wasp or oak chitinases. Numerous oak chitinases were expressed throughout development of galled and normally developing buds. The two most highly expressed chitinases were concordantly differentially and more highly expressed in mature galls (Mature > Growth and Early stages) and in older normal bud tissues (S5 > S4). Similar late stage host-plant chitinase expression was also observed in chestnut gall tissues attacked by another cynipid, Dryocosmus kuriphilus [87].
Plant hormone-associated genes. Many auxin-related genes were expressed in B. pallida galls (n = 115) and in normally developing buds (118), and most (110) were expressed in both tissue types (S2–S4 Files and S8–S10 Files for annotations and significant genes respectively). Some genes expressed in galls and normal buds, including Auxin responsive factor 5 and Auxin responsive factor 9, showed similar differential and higher expression in Early versus Growth galls, and in normal buds (S0-4) versus leaves S5. Oak calreticulins [88] were highly expressed in all stages of gall and normal bud tissues without differential overexpression. There was also no differential expression through development in gall or normal bud tissues of the jasmonic acid-amido synthetase (JAR1) gene, which is required by the jasmonic acid mediated signalling pathway, although the gene is expressed in all stages of bud and gall development. We detected no expression in gall or normal bud tissues of homologues of the key cytokinin transporter purine permease PUP14 [89]. Ungalled Leaves (S5) and Growth stage galls both showed higher expression of the same aldehyde oxidase gene [90] than buds (S4) and Early stage galls, respectively.

However, several genes relevant to plant hormone signalling showed differential expression through development of galls but not normal buds. Early stage galls showed elevated expression of an auxin-related gene, auxin-responsive protein IAA13 (a transcription factor involved in regulation of gene responses to auxin [91]), and a cytokinin-related gene, histidine kinase cytokinin receptor (histidine kinase 5). Histidine kinase 5 fulfils a range of roles in Arabidopsis, including regulation of growth via the ethylene pathway [92], stomatal opening, and defence against pathogens [93]. Two flavin-dependent monooxygenase (FMO) [94] genes that were distinct from those differentially expressed during normal bud development were more highly expressed in Growth stage galls than in either Early or Mature stages. FMOs are involved in a range of processes in plant metabolism including synthesis of auxin (IAA), processing of glucosinolates, and defence against pathogens [95].

No demonstrated role for Symbionts and virus-like-particles in cynipid gall development

If symbiotic micro-organisms, viruses and/or virus-like particles (VLPs) play a fundamental role in cynipid gall induction, we expect to detect symbiont genomes and/or VLP coding sequence in assemblies generated from gall wasp genomic DNA libraries, and (with the exception of prokaryotes given our sequencing methodology) associated gall wasp larval gene expression.

We identified a supergroup A and B Wolbachia endosymbiont respectively in the de novo genome assemblies for B. pallida and the rose gall wasp D. spinosa (S13 Table). However, we found no bacterial or fungal symbiont genomes in any other gall inducing cynipids (or outgroup species; Table 5). These results, together with the patchy distribution of Wolbachia within and among gall wasp species [96,97], do not support a fundamental role for Wolbachia, other bacterial or fungal symbionts in cynipid gall induction. We also found no evidence for insertion of viral coat protein genes in the genomes of B. pallida, other gall wasps, or outgroup species, arguing against involvement of VLPs with recognisable homology to known viruses in transducing cynipid gall-inducing signals.

Table 5. Assembly metrics for gall inducing cynipid and parasitoid figitid wasp genomes.

Cynipidae and Figitidae are sister groups within the Cynipoidea that are estimated to have diverged 80 million years ago [55].

Species and trophic group	N50 (bp) ^*	No. of contigs ^**	Assembly size (Mb)	BUSCO complete (%) ^§	BUSCO partial (%) ^§§	Assembly software
Cynipidae
Aylax hypecoi (gall inducer on Hypecoum, Papaveraceae)	10635	119258	294	92	1	SPAdes
Diastrophus kincaidii (gall inducer on Rubus)	3366	125551	295	83	11	SPAdes
Diplolepis spinosa (gall inducer on rose, Rosa)	1747	1169600	511	86	8	SPAdes
Eschatocerus acaciae (gall inducer on acacia)	11180	25287	158	89	6	MaSuRCA
Pediaspis aceris (gall inducer on sycamore, Acer)	2703	389833	492	84	11	SPAdes
Synergus japonicus (inquiline in oak cynipid galls)	61843	12893	224	95	1	MaSuRCA
Figitidae (parasitoids, sister group to the Cynipidae)
Alloxysta arcuata	11631	76465	301	92	4	MaSuRCA
Parnips nigripes	2915	343555	470	86	10	SPAdes

Open in a new tab

* As defined in Table 1.

** Assemblies were filtered to remove all contigs less than 200 bp long.

§ The percentage of BUSCO core Eukaryota genes found in the assembly that are complete.

§§ The percentage of BUSCO core Eukaryota genes found in the assembly that are fragmented.

We sought transcripts that could derive from co-bionts in the dual RNA-seq data, noting that our RNA-seq methodology was not designed to detect expression of bacterial genes. Fungal gene expression was detected in all Growth and Mature stage galls of B. pallida, but was not detected in three of four Early stage gall replicates (Table 2), arguing against a role for fungal gene expression early in gall development. A small number of viral transcripts (782) were expressed at low levels in Early and Growth gall stages, and comprised a variable proportion of Mature gall transcripts reaching 8% in one sample (sample 12, Table 2). Viral sequences were identified as deriving principally from plant single stranded RNA viruses and derive from at least five different orders of viruses, with 'unidentified' the most prevalent grouping at the order level (S1 Table). However, no recognisably viral transcripts were expressed across all replicates of any gall developmental stage, providing no evidence as yet for viral symbionts in cynipid gall induction. The possibility of roles for currently unrecognised viral transcripts in cynipid gall induction cannot be excluded (see Discussion).

Patterns of gall wasp gene expression during gall development

As noted above, functional annotation of the gall wasp transcriptome was relatively poor and functional interpretation of changing gall wasp gene expression through gall development is based on incomplete information. Notably however, gene products with predicted secretory peptide motifs and no transmembrane domain (44%) were abundant among differentially expressed genes including those otherwise un-annotated, a pattern that is compatible with their export from the gall wasp larva and roles in external manipulation and exploitation of the plant host. However, no overrepresented motifs other than transmembrane or signal peptide domains were found in Early genes versus the rest of the gall wasp gene set by MEME analysis [98], although we would not detect a motif through inter-stage comparisons if gall wasp virulence genes are necessarily constitutively expressed throughout gall-development.

Perhaps unsurprisingly given the low percentage of transcripts annotated, and the similar developmental state of the gall wasp larvae, Early versus Growth stage B. pallida transcripts were enriched for only two biological process and four molecular function GO terms (S14–S15 Tables). The four molecular function processes include pectin lyase (see below) and carbonate dehydrase activity. This increased to 38 biological process and 15 molecular function GO terms for Early versus Mature stage gall comparisons, including five chitin-associated terms for biological processes and chitinase and chitin-binding activity and pectin lyase activity for molecular functions (S14–S15 Tables). Few GO terms were enriched for genes more highly expressed in Mature versus earlier gall tissues stages (two and one biological process GO term and three and one molecular function GO term for Mature versus Early and Mature versus Growth stage comparisons, respectively); these included terms for digestion and nutrient reservoir activity, but there were no overrepresented motifs by MEME analysis [72].

As recorded in other herbivorous insects [99–102], we hypothesised that the genome of Biorhiza pallida would encode enzymes associated with metabolic breakdown of plant structural materials. In addition to roles in gall wasp nutrition, degradation products of cell wall components can act as plant signalling molecules [103,104] and so potentially play a part in gall development. We found fourteen plant cell wall degrading enzyme (PCWDE) loci coded within the gall wasp genome to be expressed through gall development, comprising six pectin/pectate lyases, four cellulases and four rhamnogalacturonan lyases (Table 6; S16 Table). Larvae from Early stage galls showed differentially greater expression of a pectin/pectate lyase versus both Growth and Mature stages, and three further pectin/pectate lyase genes were more highly expressed in Early versus Mature stages. None of the four gall wasp cellulases were differentially expressed across gall stages; two were highly expressed throughout larval development, while one (orthologous to other arthropod cellulases by blast annotation) was expressed only at low levels. The four rhamnogalacturonan lyase genes were also not differentially expressed across gall stages.

Table 6. Plant cell wall degrading enzyme (PCWDE) genes in the nuclear genomes of cynipid gall wasps, their Figitidae sister group, and Nasonia vitripennis (Chalcidoidea).

Species	PCWDE type
Species	Rhamnogalacturonan lyase	Pectin/Pectate lyase	Cellulase
Cynipidae
Biorhiza pallida	Present^*	Present	Present
Aylax hypecoi	Present	Present	Present
Diastrophus kincaidii	Present	Present	Present
Diplolepis spinosa	Present	Present	Present
Eschatocerus acaciae	Present	Present	Present
Pediaspis aceris	Present	Present	Absent
Synergus japonicus	Present	Present	Present
Other species
Alloxysta arcuata (Figitidae, Cynipoidea)	Absent	Absent	Absent
Parnips nigripes (Figitidae, Cynipoidea)	Absent	Pseudogene	Pseudogene
Nasonia vitripennis (Pteromalidae Chalcidoidea)	Absent	Absent	Absent

Open in a new tab

* As the genomes are drafts it was not possible to accurately assign copy number for each gene in each genome. Counts of sequences for each enzyme type per species are given in S16 Table. Introns were predicted in all species and genes where PCWDEs were present except D. spinosa rhamnogalacturonan lyase and pectin lyase.

Early galls had elevated expression of seven gall wasp chitinases; two were highly expressed in absolute terms (S6 Table) and were differentially expressed in both Early versus Mature and Growth versus Mature comparisons. Of these seven, five contained a signal peptide and only one encoded a signal peptide and transmembrane domain. All seven were also annotated with the GO term GO:0005576 for extracellular region. Taken together these are compatible with cellular export of these chitinases. Other gall wasp genes showing significantly elevated expression in Early stage galls (S6 Table) included a tyrosine-protein phosphatase, eleven carbonic anhydrases (involved in maintenance of pH balance and transport of carbon dioxide by hydration to bicarbonate [105]), venom acid phosphatase, and a glycine N-acyltransferase-like protein. We found no evidence for gall wasp production of plant hormone homologues in the differentially expressed genes, and neither did we detect expression of BmIAO1 (or any close homologue), a gene involved in endogenous insect production of auxin [106]. We found no differential expression of putative xanthine dehydrogenase/aldehyde oxidase homologues for BmIAO1 and only one expressed at low-level in this experiment (S6 Table). The Hessian fly, Mayetiola destructor, expresses genes that manipulate host plant hormones [107], but we found no homologues for these in the gene set for B. pallida. More generally, none of the B. pallida genes differentially expressed in Early galls were homologues of loci expressed by Hessian fly larvae during gall induction in wheat.

Which gall wasp genes involved in gall development are novel compared to non-galling sister groups?

We searched the genomes of other gall-inducing and parasitoid cynipoids for orthologues of B. pallida genes identified as differentially expressed during gall development. Of the 75 protein-coding genes differentially expressed more highly in Early than Growth B. pallida galls (Table 3), 27 had orthologues in at least one other sampled cynipoid species (S17 Table). Of these 27, 14 (52%) had orthologues both in other gall-inducing cynipids and at least one parasitoid Figitid non-galling outgroup (Alloxysta arcuata and Parnips nigripes). One group incorporates all B. pallida Early stage differentially expressed carbonic anhydrase genes, and another incorporated venom acid phosphatases. These genes are not unique to gall-inducers as each orthrogroup included orthologs from both galling cynipid and non-galling figitid genomes. Five orthologue groups were restricted to the cynipid gall-inducers: one encodes pectate lyase PCWDEs, one a MBF2 transcription activator, and the remaining three are unannotated.

All three classes of plant cell wall degrading enzyme (PCWDE) found in B. pallida were also found in the nuclear genomes of other gall wasps, including an inquiline species (Synergus japonicus; Table 6; S16 Table) that induces its own nutritive tissues inside galls initiated by gall-inducing Cynipinae [54,108]. All of the PCWDE genes identified in B. pallida and other cynipoids show significant sequence similarity to PCWDE from plant pathogenic bacteria (S18 Table). However, three lines of evidence support the conclusion that these genes are within the insect genomes: (i) detection of transcripts in poly(A)-selected RNA-seq data is consistent with derivation from eukaryotic (rather than prokaryotic) transcripts; (ii) predicted PCWDE genes in gall wasp genomes include eukaryotic introns; and (iii) predicted PCWDE genes are flanked by one or more genes of unambiguously arthropod origin on the same contig (S19 Table). Orthologs to gall wasp PCWDE genes were not identified in nuclear genomes for the figitid parasitoids P. nigripes and A. arcuata, or the distantly related chalcid parasitoid Nasonia vitripennis (Table 6). However, for P. nigripes a BLAST search identified potential homologous sequence to B. pallida pectin lyase and cellulase. These are likely pseudogenes or fragments of a functional domain as, in contrast to all cynipid PCWDEs, the three putative loci identified all contain internal stop codons and were not predicted by AUGUSTUS [109] (Table 6). The PCWDE loci initially identified in Biorhiza pallida are thus almost or entirely restricted to the genomes of gall-inducing cynipids.

Discussion

The galls induced on oak by B. pallida are an example of a common life history strategy amongst insects, mites, nematodes and other organisms. By manipulating the plant host to produce a protective structure that also supplies nutrients for growing larvae or sedentary adults, galling animals exploit and subvert plant developmental pathways to produce novel structures that benefit them, sometimes at measurable cost to their hosts. We have used genomic and transcriptomic analyses to explore the B. pallida–Quercus robur oak gall system to identify gene expression changes through gall development that are informative regarding the activity of the wasp within the gall and the response of the tree. Gall development involves expression of oak and gall wasp genes in repeatable, growth stage-specific patterns [26,86,87]. We compared oak gene expression in gall tissues and ungalled developing buds of Q. robur to identify genes and processes restricted to, and hence characteristic of, developing galls. To explore the evolutionary origins of genes involved in the galling phenotype we generated and analysed de novo draft genomes for a biologically diverse set of galling and non-galling Cynipoidea, looking for orthologues of candidate genes identified in Biorhiza pallida.

We sampled biologically independent replicates of three morphologically-defined gall stages, corresponding to Early, Growth, and Mature stage galls. Gall stage was a major driver of variance in the whole-gall RNA-seq data, suggesting that our sampling captured strong biological signal.

We used de novo genomes for the oak and gall wasp to separate the sampled whole gall transcriptome into oak, gall wasp and other co-biont compartments, and thus to assess, independently, the patterns of gene expression through gall development in each player. Oak transcripts dominated all gall stages, with the gall wasp component increasing as larvae developed. Gall wasp transcripts made up only ~3% of the Early gall transcriptome, but ~22% of the Mature gall. Fungal and viral infection of the galls was also detected and variable between replicates. This was in contrast to normally developing buds, which had negligible viral and fungal infection levels. Analyses of the patterns of oak and gall wasp gene expression accounted for these co-bionts by controlling for their effect during differential expression testing (see Methods).

We found marked divergence in oak gene expression between galled and ungalled host tissues, as observed in cynipid-induced chestnut galls and phylloxera-induced galls on grape vines [25,87]. The clear developmental trajectory of gall tissues also contrasted strongly with the pattern observed in normally developing ungalled buds, which showed little transcriptomic differentiation between recognised morphological stages through budburst, but clear differentiation between opening buds and fully open leaves.

We were able to functionally annotate most oak transcripts based on the large body of data and analysis available for model and crop plants. However, the gall wasp transcripts were relatively poorly functionally annotated, with many fewer sequence similarity matches and known domains. Functional annotation was particularly scarce for gall wasp genes that were differentially expressed between Early gall and later developmental stages. The annotational novelty in the gall wasp larval transcriptome parallels that for gall wasp ovaries and venom glands, which represent possible sources of maternal stimuli involved in the initiation stage of gall induction [49]. Around 90% of differentially expressed venom and ovary-specific transcripts are novel in B. pallida and the rose gall wasp Diplolepis rosae while equivalently expressed genes are better annotated. This suggests that lack of annotation for differentially expressed genes reflects their novelty rather than a general lack of annotation for gall wasp genomes [49]. Similarly low levels of annotation and high annotational novelty have been reported in the immature stages of other gall inducers, including fig wasps and Hessian fly [8,21,22]. The long divergence of gall wasps from non-galling ancestors [55] coupled with potentially rapid (co-)evolution of genes underlying the galling interaction could well leave little signal of orthology between contemporary lineages. We now return to the four questions originally posed in the Introduction.

What, in plant terms, is a cynipid gall?

Different galling organisms induce galls on different plant tissues, including meristematic tissues and developing organs of both aerial and root systems. These different targets, and the different galls induced, may result from both shared and taxon-specific manipulation of core plant processes. Cynipid galls have morphologies that are specific to the inducing wasp species and generation, and thus represent a specific response of the host to the particular infection experienced. Biorhiza pallida sexual generation galls are derived from oak bud meristematic tissues which, in normal development, would generate leaves or flower parts, but (in contrast to some other insect-induced galls [25]) the galls do not obviously resemble these plant organs. Two related hypotheses for the oak organ similarities of cynipid galls have been proposed previously from anatomical and developmental studies [1,59,60]: ‘galls as ectopic food storage organs’ and ‘galls as modified somatic embryos’. We used dual RNA-seq expression data to identify genetic systems expressed in the oak tissues, and compared these to known plant organ systems to assess expression phenotype resemblances. Overall, we found gene expression and inferred developmental process in galled and ungalled buds to diverge substantially through development, and more support for ‘galls as modified somatic embryos’ than ‘galls as ectopic storage organs’.

Developing buds and Early stage gall oak tissues share many processes of cell division, cell cycle, and chromosomal organisation (Fig 5) that are fundamental to organogenesis in plants. As galled and ungalled tissues develop, their trajectories of differential oak gene expression diverge, to the point that genes differentially expressed in Mature galls relative to earlier stages show little or no overlap to genes differentially expressed in leaves relative to earlier stages in normal bud development. While the sexual generation gall of Biorhiza pallida is clearly very different from the leaf into which the host bud would otherwise develop, more in depth analysis of gene expression is required to query whether cynipid gall development involves pathways involved in normal plant organogenesis, such as the flower development pathways recently demonstrated in phylloxera galls [25].

Some differential oak gene expression between gall developmental stages can be related directly to observed developmental or homeostatic processes in the galls. For example, differentially high expression of genes involved in cell wall growth and vascularization, such as Auxin responsive factor 5 in Early stage galls, and of other genes later in gall development, matches observed development of outer gall tissues and of vascular bundles supplying the larval chambers. Given that cynipid galls expand rapidly during the Growth stage, the absence of elevated expression of cell division-associated genes in Growth versus Early stage galls is perhaps surprising. However, anatomical studies show that much of the growth of oak cynipid galls is accomplished by increase in cell volume (hypertrophy) rather than by cell division [37,46]. This parallels the similar transition from cell division to cell expansion observed as normally developing buds flush to fully developed leaves [110–112].

Both developing galls and ungalled buds expressed oak gene systems associated with cell division and nuclear endoreduplication. In terms of oak cynipid gall development these findings are consistent with previous in situ hybridization analysis of gall development in B. pallida, which suggested that endoreduplication and somatic embryogenesis are primarily associated with nutritive tissues immediately surrounding the larva, while high rates of cell division are associated with rapid expansion of the outer parenchyma [59]. That these patterns may be common in cynipids is suggested by transcriptomic evidence from galls induced in chestnut by another gall wasp, Dryocosmus kuriphilus [113]. Transcriptomic sampling of oak genes in specific gall tissue compartments, rather than the entire gall, is required to test these ideas.

The ‘galls as ectopic food storage organs’ hypothesis was based on detection in earlier metabolomic studies of BCCP expression by gall tissues, associated with seed development in Arabidopsis and food storage organ development in other plants [59,60,70]. We found BCCP expression throughout bud development into mature leaves, and across all three stages of gall development, confirming previous work [59,60]. While our data do not rule out a role for BCCP expression in nutrient processing in cynipid galls as originally hypothesised, they confirm a more general role for products of these genes during both gall and normal bud and leaf development.

Plant somatic embryogenesis is characterized by expression of phytocyanin domain-containing arabinogalactan proteins (AGPs), proteoglycans that carry sets of arabinosyl and galactosyl residues. AGPs are naturally cleaved by action of plant chitinase enzymes and then act as signal molecules [114–118]. These signalling-associated AGPs are present in a range of plant tissues, including seeds [119,120] and we found them to be highly expressed in all normally developing bud stages (S0-S4) except leaves (S5). Signalling AGPs are encoded by a subset of the ENOD genes [121,122] involved in legume root nodule development [123,124]). We observed upregulation of ENOD AGPs in Early stage galls, and suggest that the somatic embryogenesis-like state observed in young cynipid galls may be specifically induced by signals derived from oak AGPs by action of gall wasp chitinases (see below).

Proteomic work on leaf galls induced by other oak cynipids (Cynips and Neuroterus species) identified overproduction of the enzyme S-adenosyl methionine (SAM) synthase in gall tissues [105]. We found four genes encoding the same enzyme to be differentially over-expressed in Growth stage relative to Early stage galls, two of which are only differentially expressed in gall tissues (TRINITY_DN55180_c4_g1 and TRINITY_DN69171_c2_g1). SAM synthase catalyses the production of SAM from methionine, and the SAM in turn is a substrate for the synthesis of a range of important plant metabolites, including ethylene and polyamines. Possible roles for these metabolites discussed in [105] include a range of processes relevant to gall induction that includes meristem development [125], regulation of Rhizobium-induced root nodulation in legumes [126], organogenesis and fruit maturation [127]. Further work is required to understand the role(s) of these genes in gall tissues.

What organism induces the B. pallida oak gall?

We found no support for symbiont involvement in B. pallida gall development. Although gall development is known to require a living gall wasp larva, it has remained possible that the inducing organism is not the wasp but rather a symbiont. Symbiotic bacteria such as Wolbachia are present in many galling insects, and have been shown to influence the development of plants attacked by leaf miners [73]. Bacterial endosymbionts are abundant in the guts of both larval and adult Hessian flies, though their role in the Hessian fly-wheat interaction is unknown [74]. While our poly(A)-selected transcriptomic analysis was not intended to quantify bacterial gene expression, our analyses of adult gall wasp holobiont genomic data found no consistent association between gall wasps and specific bacterial endosymbionts. We identified Wolbachia in genome sequence data from B. pallida (a supergroup A strain) and the rose gall wasp Diplolepis spinosa (a supergroup B strain), but not in the other gall wasp or Figitidae sister group genomes analysed. Wolbachia prevalence and incidence vary substantially within and among gall wasp species, suggesting multiple gains and losses of infection [96,97] and arguing against a fundamental role for Wolbachia in cynipid gall development. We detected transcripts from plant viruses and plant-associated fungi in gall tissues, but neither comprised taxa that were consistently detected in cynipid genomic libraries, nor showed expression patterns across gall stages consistent with a causal role in gall development. However, new viral taxa continue to be discovered [128,129] and it remains possible that unrecognised viral symbionts contribute to the large numbers of differentially expressed but unannotated genes attributed to the gall wasp in our experiment. A role for viruses (or any other currently unrecognisable symbiont) in gall induction thus cannot be wholly excluded. However, our data are consistent with phylogenetic patterns in gall traits that support a major role of gall wasp nuclear genes in gall development.

What processes are involved in gall wasp manipulation of plant development?

Our data are compatible with secretion of effectors from the gallwasp into surrounding host plant tissues, we hypothesise from the enlarged larval salivary glands. We find no evidence for effector delivery via virus-like particles (VLPs).

(i) How does the gall wasp larva export stimuli to the plant? In other plant pathogenic animals, including nematodes, gall-inducing fig wasps and Hessian flies, highly expressed novel proteins with secretory signal peptides have been proposed to represent putative effectors [21,46,130]. Most Early stage gall wasp transcripts had no informative functional annotation, but many highly-expressed loci encoded proteins with secretory signal peptides. Of the 75 genes more highly expressed in Early stage gall wasp larvae, 44% (33) had signal peptides and no transmembrane helix, compatible with secretion of these proteins into surrounding host tissues. These are obvious candidates for further study. Whether these proteins are secreted or excreted by the wasp is unknown, but they could include novel effectors. Their general lack of informative annotation, and the absence of any conserved motifs among them, will make analysis challenging. A striking feature of the unannotated but highly expressed genes of Early stage gall wasp larvae is their lack of homology with similarly unannotated protein coding genes expressed during gall induction and development by Hessian flies or any known genes (via comparison to the non-redundant protein database). This lack of homology suggests that even if these gall inducers are targeting similar developmental pathways in their plant hosts, they are doing so in taxon-specific way. This lack of homology is also concordant with the lineage-specificity of genes involved in plant manipulation by the aphid Myzus persicae relative to other arthropods [131].

Parasitoid wasps deliver DNA and protein to their hosts through virus-like particles, produced from viral genes incorporated into the wasp nuclear genome [79,81]. It is possible that a similar system could be used to deliver gall inducing stimuli in cynipid galls. However, we found no evidence of Early gall expression of gall wasp homologues of genes coding for viral coat proteins, arguing against export of gall inducing stimuli in virus-like particles with homology to any documented virus [128]. Cambier et al [49] also found no evidence for VLPs in the ovaries and venom gland transcriptomes of B. pallida and a rose gall wasp, D. rosae. The possibility that gallwasps use VLPs derived from currently unrecognised viral taxa cannot, however, be excluded.

(ii) What signals might be involved? Plant development is regulated by networks of peptide and other hormones, cell interactions and cell-autonomous processes. Gall wasps could drive gall development by intervening at key steps in normal plant developmental processes, initiating pathways that then autonomously generate structures that support gall wasp development. In a psyllid leaf gall system, the host gall response included transcriptional upregulation of several loci annotated as involved in responses to auxin, but the biological annotation of the insect genes was not reported [132]. Auxin (indoleacetic acid, IAA) and cytokinins have been extracted from the bodies of a range of insect gall inducers including aphids, sawflies, gall-midges and cynipids [36,94,132–138], and while the source (insect, symbiont, host plant) of these effectors has in some cases yet to be determined beyond doubt, ability to synthesise IAA may be widespread in insects [106]. It has been proposed that gall inducing insects acquired the necessary plant synthetic enzymes through microbial symbiosis or lateral gene transfer [139,140]. Indeed, it has been proposed that plants themselves acquired IAA synthesis from bacteria [141]. We found no evidence for Early gall stage upregulation of gall wasp transcripts with recognisable homology to plant genes involved in auxin- or cytokinin-related processes, including the recently discovered BmIAO1 gene [106]. We also found no sequences with recognisable affinity to cytokinin-related plant genes in cynipid genomes. We found no evidence for consistent gall wasp association with a bacterial or fungal endosymbiont that could produce plant hormone homologues or analogues. There is thus currently no evidence to support the hypothesis of horizontal transfer of genes regulating plant hormonal systems into the gall wasp genome.

This does not mean that auxins and other phytohormones are not involved in cynipid gall development, rather their involvement is likely to be downstream of the key gall inductive and maintenance signals. The oak Auxin responsive factor 5-like gene, highly expressed in Early stage galls, is typically associated with auxin-induced roles in axis formation and/or the development of gall xylem and phloem tissues [142,143]. It is also possible that gall wasps interfere with, or produce, phytohormones using enzymes that have no informative sequence similarity to the plant genes that normally regulate and produce these messengers. Direct assays of phytohormones or their precursors in the bodies of gall inducers, and evidence of endogenous production rather than concentration of plant or symbiont products [94,106,134,138,144], are more powerful tools in assessing the roles and sources of auxins, cytokinins and other plant hormones.

Which gall wasp genes involved in gall development are novel compared to non-galling sister groups?

(i) Gall wasps produce their own plant cell wall degrading enzymes (PCWDEs). To exploit gall tissues, the developing wasp larvae must access and digest plant tissues. Digestion of the complex plant cell wall is accomplished by gut microbiota in most insect herbivores, but a growing number of studies have detected endogenous PCWDE genes in insects [145–148]. These genes are derived from a combination of horizontal gene transfer events from bacteria (HGT) [149,150] and expansion of PCWDE gene families within insect genomes. Three classes of gall wasp-encoded PCWDEs were expressed during cynipid gall development: pectin lyases, rhamnogalacturonan lyases and cellulases. Gall wasp pectin lyase activity (GO:0047490) was differentially overexpressed in Early relative to later stage galls, and two distinct gall wasp cellulase genes were highly expressed throughout gall development. PCWDE genes are also expressed in the venom glands of B. pallida [49], as is a venom acid phosphatase.

Though the genomic origin of venom gland-expressed genes was not demonstrated, two were hypothesized to have been incorporated in the gall wasp genome following horizontal gene transfer from a bacterial origin [49]. Our genomic analyses confirm that the cellulases expressed in adult and larval gall wasps are encoded by gall wasp genes, through detection of (a) spliceosomal introns indicative of eukaryotic genome architecture, (b) sequence contiguous with the cellulases that contains genes of obvious insect origin in multiple gall wasp species, and (c) polyadenylated (i.e. non-bacterial) cellulase mRNAs in the B. pallida transcriptome. As hypothesised for the B. pallida venom gland cellulases, larvally-expressed cellulases are most similar to either bacterial sequences or to genes in distantly related insect herbivores for which bacterial origins have been proposed [100]. Interestingly, BLAST comparison shows that different and divergent cellulase genes are expressed in adult B. pallida venom glands and in larvae, aligning to different contigs in the B. pallida genome (S21 Table). This could reflect adaptation to differing roles and/or host plant targets, such as adaptation of specific gene copies for high and tissue-specific expression via a secretory organ-specific promoter. This is hypothesised to originate from the venom gland in the adult female [49] and is hypothesised to be the salivary glands in the larva [43,56]. It may also be attributable in part to differing methodologies of transcriptome sequencing and assembly between this study and [49].

Gall wasp PCWDEs could be involved in larval hatching and maternal venom gland PCWDEs could contribute to formation of the initial chamber adjacent to the egg into which the hatching larva moves [40]. Larval PCWDEs probably function to break down the walls of nutritive cells in the larval chamber [2,151,152], and could also facilitate passage of gall induction effector proteins or maintenance factors (potentially including cell wall breakdown products themselves) into host plant tissues by making cell walls permeable to the passage of signal macromolecules. Intracellularly-acting effectors have been identified in other galling insects including aphids and hessian flies [23,153–155]. Both have mouthparts capable of penetrating plant cells to aid effector delivery, and this process could be assisted enzymatically in gall wasps. Bacterial expression assays of gall wasp PCWDEs are required to determine their substrate relationships and identify their roles in vivo.

(ii) A PCWDE repertoire may be a synapomorphy of gall-inducing and inquiline cynipids. Orthologous genes encoding all three classes of B. pallida PCWDEs were found in the genomes of other cynipid wasps, including the inquiline S. japonicus and the gall-inducing species A. hypecoi and E. acaciae (Table 6), representing members of three deeply diverging tribal lineages within the Cynipidae [54]. The same genes were absent from (or detected only as pseudogenes in) the genomes of non-galling Figitid wasps, the sister group of gall wasps whose parasitoid life history is thought to represent that ancestral state from which cynipid gall induction is derived [39]. This suggests that a multilocus repertoire of PCWDEs is a derived trait that evolved in the shared common ancestor of gall inducing and inquiline cynipids, i.e., a synapomorphy. Gall wasps evolved from a parasitoid of insect hosts concealed within plant tissues [55], and shifted to phytophagy either prior to or concomitantly with the evolution of gall induction [54,55]. Acquisition of PCWDEs by the common ancestor of gall wasps may have helped larvae migrate through plant tissues in search of insect hosts (as a parasitoid), feeding sites (as a herbivore) or assisted in penetration of plant tissues by the adult female ovipositor prior to the evolution of gall-induction. Wider sampling of cynipids and their relatives in the Cynipoidea is required to reconstruct the origin(s) and diversification of the cynipid PCWDE repertoire.

Chitinases, arabino-galactan glycoproteins and a model for cynipid gall induction

Gall wasp chitinases are a notable exception to the general paucity of functional annotation of cynipid genes differentially overexpressed in Early and Growth stage galls. Homologues of these chitinases were present in other gall wasps and our sampled outgroups. We hypothesise that Early stage over-expression of gall wasp chitinases may play a role in gall formation through interaction with oak ENOD gene products.

Two oak ENOD (early nodulin) arabinogalactan protein (AGP) transcripts were upregulated in Early stage galls. ENOD genes were initially identified for their roles in Rhizobium nodule development in legumes [156], but are widely present in vascular plants [157,158]. Legume ENOD genes are specifically induced by nod factors during the earliest stage of Rhizobium nodule development, and have been co-opted for nutritive benefit by reniform nematode pathogens of soybeans [159]. ENOD genes could potentially offer a toolkit for host manipulation by parasites more widely through their key roles in plant developmental pathways [159]. The upregulated oak ENOD AGPs in B. pallida galls belong to a group containing a phytocyanin-like domain that have been implicated in somatic embryogenesis in plants [116]. AGPs carry glycan side chains containing glucosamine and N-acetyl-D-glucosaminyl residues, which can be cleaved from the molecule by chitinases [114,115]. Action of plant chitinases on AGPs promotes somatic embryogenesis, implying that the released glycan moieties may have a role in signalling [114,160]. We hypothesise that the wasp chitinases highly expressed in Early stage galls could manipulate oak developmental pathways by similarly cleaving ENOD AGP proteins, thus inducing somatic embryogenesis.

All of the cynipid genomes we examined contained at least one chitinase gene. These chitinases were most similar to the GH18 family of insect chitinases, exemplified by the N. vitripennis chitotriosidase-1-like protein. GH18 chitinases function in turnover of extracellular chitin-containing matrices, such as the insect cuticle [161] during larval and pupal moulting [162]. However, we suggest that chitinase expression in larval B. pallida is probably not associated with moulting because larvae grow very slowly in Early stage galls (when chitinase expression is high) and then grow rapidly and moult in Growth and Mature galls (when expression of the same chitinases is low in our data).

Additional possible roles for insect chitinases are breakdown of chitin in the eggshell, protection against attack by fungi, and suppression of plant defence responses to large chitin molecules [163–166]. Most insect egg shells do not contain chitin [167,168], though where chitin is present newly-hatched larvae do show chitinase activity [169]. However, young larval B. pallida gall wasps are separated from their eggshells by their own movement into plant tissues, and by the development of plant gall tissues around them [1]. A role in eggshell degradation for early chitinase expression is thus undemonstrated. Gall wasp chitinases could potentially play a role in protection of the gall wasp and gall tissues from fungal attack, given that fungal transcripts were detected in gall samples. However, chitinase expression in Early stage galls was not correlated with fungal infection of galls across replicates (S2 Table), and we thus consider it unlikely that the observed chitinase production was part of an induced anti-fungal defence. It is also interesting that larval chitinases are distinct from the chitinase expressed in the adult female venom gland (S21 Table), which is probably injected along with the egg during oviposition and for which an anti-fungal role has been suggested [49]. If our hypothesis is correct, we expect chitinases that interact with the host to be exported from the gall wasp larva. In Hessian flies, chitinases are expressed by the salivary glands, making them potential effectors in that system [21].

We speculate that secreted gall wasp chitinases act in Early stage galls to mimic endogenous plant chitinases. By cleaving glucosamine and N-acetyl-D-glucosaminyl residues from oak AGPs in the extracellular matrix they could drive induction of somatic embryogenesis-like dedifferentiation and cell division in host tissues. To test this model, future work could identify the larval tissue of origin of the chitinases, localise the gall wasp chitinases in the developing gall, define their enzymatic activity on oak glycoproteins in vitro, and explore expression patterns of orthologues in other galling species.

Conclusions

We have identified candidate genes associated with gall development in both the gall inducer and the host plant in the B. pallida-oak system. Functional inference from oak gene expression, most of which was annotated, correlates with observed development of gall phenotypes. In contrast, most gall wasp genes differentially expressed at the key early post-initiation stage had no informative functional annotation. However, the fact that many are predicted to be secreted suggests that they could include novel parasitic effectors of plant development. We hypothesise that high expression of host arabinogalactan proteins and of gall wasp chitinases in Early stage galls interacts to generate a somatic embryogenesis-like process in nutritive tissues surrounding the gall wasp larvae. This mechanistic framework provides testable hypotheses for future functional dissection of cynipid gall development.

Methods

Sample collection, RNA extraction and sequencing

We sampled sexual generation galls of Biorhiza pallida near Blandford Forum, Dorset in 2011 and sampled normally developing buds at Dalkeith Country Park in 2018, Midlothian, United Kingdom. All tissues were sampled from oaks morphologically identifiable as Quercus robur. Four biological replicates, each from a separate tree, were collected for each of the following three gall developmental stages, giving 12 samples in total. (i) Early stage galls (Fig 1A and 1B. Gall diameter <5mm, often fully or partly concealed within bud scales; for illustration of all 4 biological replicates, see S1 Fig). (ii) Growth stage galls (Fig 1A, 1D and 1E. Gall diameter 20-30mm, advanced larval chamber and nutritive tissue development, larval length <1mm). (iii) Mature galls (Fig 1A, 1F and 1G; epidermis brown, with a papery texture, internal tissues lignified, larvae active and ≥3mm long). Four biological replicates, again each from a separate tree, were collected for each of bud tissue developmental stages S0-S5 (Fig 1A). All galls were rapidly sliced into 1mm thick sections, bud tissues diced into small pieces, and a cross section of leaf tissue from the centre of the leaf perpendicular to the petiole of approximately 2 cm width was sliced and immediately immersed in RNAlater (Ambion) in the field. RNA extractions were made for each replicate using the RNEasy plant mini kit extraction protocol (Qiagen) and stored at -80°C until sequencing.

Presence of gall wasp mRNA in extractions was confirmed using reverse transcriptase polymerase chain reaction (RT-PCR) amplification of a pair of exon-primed intron-crossing (EPIC) gall wasp loci, Receptor for Activated C Kinase 1 (RACK1) and Ribosomal Protein L37 Rpl37 ([170], further refined for cynipids by James Nicholls (personal communication). A B. pallida genomic DNA positive control was used to indicate cDNA amplification over genomic DNA carry-over.

Sample quality was assessed for RNA purity by 260/280 and 260/230 ratios measured on a NanoDrop spectrophotometer (Thermo Scientific), followed by RNA integrity analysis by Agilent 2100 Bioanalyzer (Agilent Technologies) total RNA nano trace. No samples showed visible degradation on Bioanalyzer traces. Gall and ungalled bud/leaf samples were prepared as 100 and 150 base pair paired-end TruSeq libraries respectively by the NERC Edinburgh Genomics facility. Gall and ungalled bud libraries were multiplexed and sequenced separately on one lane of an Illumina Hi-Seq sequencer. The eight Early and Growth stage gall libraries were sequenced on an additional lane to increase numbers of gall wasp reads sequenced, since in these stages gall wasp RNA comprises only a small fraction of whole gall RNA.

Transcriptome assembly, quantification and taxonomic assignment

Reads (ENA PRJEB13357; S20 Table) were quality and adapter trimmed with default parameters in FASTP version 0.19.3 [171] and results checked in FASTQC version 0.11.7 [172]. All reads were then assembled in TRINITY version 2.6.6 [173]. Transcripts were quantified in SALMON [174] with “—seqBias”, “—gcBias”, and “—rangeFactorizationBins 4” options and converted to gene-level counts using the Bioconductor package TXIMPORT [175] (Rscript: S1 File). Transcripts were annotated in the TRINOTATE pipeline [176] (S2–S4 Files). To assign taxonomic origins a TRINITY SuperTranscript [177] was created from each group of associated transcripts representing a gene and compared against a combined reference using MINIMAP2 in ‘splice’ mode [178]. This reference consisted of the B. pallida genome sequence [50], the Q. robur haploid genome sequence (PM1N) [179] and four chalcid parasitoid genomes [51]. The four chalcid parasitoids are common gall inhabitants and were used to remove parasitoid hymenopteran sequences that were not gall wasp-derived. They were Cecidostiba fungosa (Pteromalidae), Megastigmus dorsalis (Megastigmidae), Ormyrus pomaceus (Ormyridae) and Torymus auratus (Torymidae), and details of their assembly can be found in [51]. Supertranscripts were assigned by top score to a taxon of origin, ambiguous sequences were assigned by inspection. We also independently assigned a taxonomic origin to each transcript by aligning to proteins of the ‘nr’ database (downloaded August 15^th 2018) with DIAMOND [180] and TAXONOMIZR (https://github.com/sherrillmix/taxonomizr). The transcriptome was then further filtered to remove fungal contaminant sequences identified as the top-scoring alignment per gene by DIAMOND. Ribosomal RNAs were removed after identification by RNAMMER and BLAST alignment to SILVA large subunit and small subunit rRNA databases [181]. Finally, only transcripts that encode an open reading frame as predicted by TRANSDECODER were retained for further analysis. The SALMON count matrix was filtered according to the above and split into Q. robur gall tissue and normal bud and B. pallida specific matrices for differential expression analyses (S5–S7 Files). BUSCO (v. 3.0.2) scores were computed for the species-specific transcriptomes against the Eukaryota database (Downloaded July 2017) [182].

Statistical analysis of stage specific variation in gene expression

Differential expression analysis was performed using DESeq2 (version 1.6.2) [183] for three separate analyses: gall wasps, oak gall tissues, and normal oak bud and leaf tissues on genes with a greater than 10 combined count across replicates. Count tables and DESeq2 R scripts used are provided in S1 File. For gall tissues and gall wasp contrasts a blocking factor for fungi was included where the fungi-derived reads accounted for greater than 1% of the dataset to control for the effect of fungal infection on gall wasp and oak expression (N = 4 samples; S2 Table). A further blocking factor, ‘parasitoid load’ was fitted for gall wasp transcript analyses, as parasitoid read-depth was greater than 10% of the gall wasp read-depth for nine samples. No blocking factors were fitted for normal bud and leaf comparisons as no fungal or other infection was detected. Samples were inspected for stage-specific clustering by principal components analysis (PCA) after variance-stabilising transformation (VST) of the data in DESeq2. Tests were performed sequentially with each stage contrasted with the next developmental stage for galls and normal buds. We specified the ‘apeglm’ method for effect size shrinkage [184], and imposed a log-fold change requirement of one; meaning that to qualify as differentially expressed, a gene must be expressed significantly greater than log-fold change plus or minus one. This method results in s-values which are analogous to q-values [185], and a cut-off of α < = 0.005 was applied as recommended by the ‘apeglm’ package authors (significant genes: S8–S10 Files). Significant genes were retained for GO term enrichment analysis using the “weight01” algorithm in TopGO [186] (Rscript: S1 File) and GO terms annotated by Trinotate. Genes of interest were compared against all genes expressed in the analysis, which corresponded to all genes passing initial expression filtering in DESeq2. Enriched GO terms were considered significant at a cut-off of α < = 0.01. To more easily interpret the results we clustered the resulting GO term lists to reduce redundancy and visualised them in two-dimensional semantic space using REVIGO [187]. We performed this for genes that were differentially expressed only in gall tissue contrasts and for genes that were shared and expressed in the same direction between buds (S4 only) and leaves (S5). S9 and S10 Tables for shared genes with concordant expression which corresponds to GO terms enriched for genes corresponding to cells H4:H7 in S4 Table, and for gall tissue-specific genes and GO terms detailed in S5 Table. The Arabidopsis thaliana GO term database was used as a reference, ‘SimRel’ as a similarity measure, the enrichment p-values, and ‘medium’ similarity. The B. pallida Early stage gall-upregulated gene set was compared to all other B. pallida genes to identify potential protein motifs associated with induction using MEME [98].

Orthologue identification of candidate genes

Additional gall wasp genomes were generated for six additional cynipids in six additional tribes within Cynipidae chosen to maximise cynipid diversity analysed according to the phylogeny of Ronquist et al. [54]: Aylax hypecoi (tribe Aylacini), Diastrophus kincaidii (tribe Diastrophini), Diplolepis spinosa (tribe Diplolepidini), Eschatocerus acaciae (tribe Eschatocerini), Pediaspis aceris (Tribe Pediaspidini) and Synergus japonicus (tribe Synergini), and also for two parasitoids in the Figitidae sister group of Cynipidae, Alloxysta arcuata and Parnips nigripes. Figitidae and Cynipidae are thought to have diverged around 80 million years ago [55], and within the Cynipidae the lineage which diversified into Diplolepis and Pediaspis split basally from the lineage that includes Biorhiza and other oak cynipids [54]. The cynipid and figitid genomes were sampled from entire single individuals and extracted using the DNeasy kit (Qiagen), and Nextera genomic libraries (Illumina) were created using standard protocols. Libraries were sequenced on the Illumina Hi-Seq platform by Edinburgh Genomics. Raw data were quality controlled with FastQC, quality and adapter trimmed using CUTADAPT and subsequently assembled with SPAdes [188] (version 3.5) or MaSuRCA [189] assemblers. Assembly quality was assessed by CEGMA score. Assemblies were then masked for repeats by a combination of REPEATMODELER (v1.0.11) and REPEATMASKER (v4.0.6) [190]. Assemblies were checked for contaminant sequences and endosymbionts using a blob-plotting approach [191] (https://github.com/DRL/blobtools-light). To create blob-plots each contig of the genome was annotated using the DIAMOND aligner in nucleotide versus protein search mode with an e-value cut-off of 1e-5, and retaining the five best matches per contig. Coding sequences were then predicted using AUGUSTUS (v3.3) with Nasonia vitripennis as the reference species [109]. PCWDE enzyme encoding genes were inspected for introns and genes of unambiguously hymenopteran origin in synteny on the same sequence. This would confirm their incorporation into the cynipoid (eukaryotic) genome, rather than being expressed by a previously undetected symbiont. Protein predictions were used as input for ORTHOFINDER with default parameters (v2.3.1) to identify orthologue groups [192].

Supporting information

S1 File. R scripts used for testing differential expression and GO term enrichment in all comparisons.

Each section is separated by “##” and explanatory text: 1) importing Salmon alignment results and conversion to gene counts, 2) gall wasp larvae DESeq2 script, 3) gall tissue DESeq2 script, 4) normal bud tissue DESeq2 script, and 5) TopGO script to identify enriched GO terms. DESeq2 scripts differ in blocking for fungal infection and parasitoid load in gall wasp tests and fungal infection only for gall tissues.

(TXT)

Click here for additional data file.^{(14.8KB, txt)}

S2 File. Trinotate pipeline annotation file in excel format.

Annotations for transcripts included in the differential analysis experiments created with the Trinotate pipeline. File contains transcripts TRINITY_DN0_c0_g1 to TRINITY_DN63911_c3_g1.

(CSV)

Click here for additional data file.^{(64.1MB, csv)}

S3 File. Trinotate pipeline annotation file in excel format.

Annotations for transcripts included in the differential analysis experiments created with the Trinotate pipeline. File contains transcripts TRINITY_DN63911_c3_g1 to TRINITY_DN76618_c5_g1.

(CSV)

Click here for additional data file.^{(76.1MB, csv)}

S4 File. Trinotate pipeline annotation file in excel format.

Annotations for transcripts included in the differential analysis experiments created with the Trinotate pipeline. File contains transcripts TRINITY_DN76618_c5_g1 to TRINITY_DN236370_c0_g1.

(CSV)

Click here for additional data file.^{(72.9MB, csv)}

S5 File. The Salmon read-count matrix for the gall wasp (B. pallida) data used for differential expression analysis in DESeq2.

Each column represents the count for a replicate and each row is a TRINITY output component group, taken as a proxy for a candidate gene.

(CSV)

Click here for additional data file.^{(1.3MB, csv)}

S6 File. The Salmon read count matrix for the oak gall tissue (Quercus robur) data used for differential expression analysis in DESeq2.

Each column represents the count for a replicate and each row is a TRINITY output component group, taken as a proxy for a candidate gene.

(CSV)

Click here for additional data file.^{(3.4MB, csv)}

S7 File. The Salmon read count matrix for the normal oak bud (Quercus robur) data used for differential expression analysis in DESeq2.

Each column represents the count for a replicate and each row is a TRINITY output component group, taken as a proxy for a candidate gene.

(CSV)

Click here for additional data file.^{(5.2MB, csv)}

S8 File. DESeq2 results for gall wasp genes showing significant differential expression between all gall developmental stages.

The threshold for significance was an S-value threshold of < 0.005 and a log₂-fold change of plus or minus one. Different contrasts are separated by “##” and explanatory text. Columns: 1) gene name, 2) mean expression across all replicates for that gene, 3) log-fold change in base two, 4) log-fold change standard error, and 5) S-value.

(CSV)

Click here for additional data file.^{(253.8KB, csv)}

S9 File. DESeq2 results for oak genes showing significant differential expression between all gall developmental stages.

(CSV)

Click here for additional data file.^{(309KB, csv)}

S10 File. DESeq2 results for oak genes showing significant differential expression between stages in normal bud development.

(CSV)

Click here for additional data file.^{(878.4KB, csv)}

S11 File. Gene names for all differentially expressed genes overlapping in S4 Table.

Column headers indicate which cell value the gene list corresponds to in S4 Table reproduced here. Annotations for these genes are provided in S2–S4 Files.

(XLSX)

Click here for additional data file.^{(40.1KB, xlsx)}

S12 File. Gene names for all differentially expressed genes overlapping in S5 Table.

Column headers indicate which cell value the gene list corresponds to in S5 Table reproduced here. Annotations for these genes are provided in S2–S4 Files.

(XLSX)

Click here for additional data file.^{(32.6KB, xlsx)}

S1 Fig. Images of the four replicates of each sexual generation Biorhiza pallida gall stage sampled in the RNAseq experiment.

A. Early stage galls. Galls as small as possible were sampled. Bud scales are visible around the galls. B. Growth stage galls. The pink-red epidermis is characteristic of these galls on exposure to sunlight. The lower row comprises cross-sections showing the spongy parenchyma surrounding the larval chambers (up to several hundred in this gall type), which contain small larvae. Vascularisation of tissues can be seen from the point of connection with the oak shoot most clearly in the fourth gall. C. Mature stage galls. The gall epidermis is now brown and papery. The scale bar in all images is in cm/mm.

(EPS)

Click here for additional data file.^{(15.4MB, eps)}

S2 Fig. Principle component analysis in DESeq2 of oak bud developmental stages S1-S4.

X-axis corresponds to principle component one and y-axis to principle component two, and axis labels give the percentage of variance these axes explain in each analysis. A lack of clear separation between stages S0-S4 is observed.

(EPS)

Click here for additional data file.^{(58.3KB, eps)}

S3 Fig. REVIGO treemaps of enriched biological process GO terms for oak genes expressed during development of galled and normal oak bud tissues.

These maps were used to construct Fig 5 by providing higher-order groupings of GO terms. Parts A, B, C and D correspond to A, B, C and D in Fig 5.

(EPS)

Click here for additional data file.^{(277.5KB, eps)}

S1 Table. Taxonomizr classified transcripts of viral origin.

Transcripts were classified as far as the data allow through taxonomic levels from order to species during the annotation process.

(XLSX)

Click here for additional data file.^{(41.1KB, xlsx)}

S2 Table. Read alignment totals for all gall-derived reads.

Data are summarised for each replicate (4 replicates for each of Early, Growth and Mature gall stages), and split into rows for oak (Quercus robur), gall wasp (Biorhiza pallida), and four major categories of non-target contaminants: fungi, parasitoids excluding Torymus, the dominant parasitoid genus Torymus, and virus-derived reads.

(XLSX)

Click here for additional data file.^{(9.8KB, xlsx)}

S3 Table. Read alignment totals for oak (Quercus robur) normal bud stage replicates.

Data are summarised for each replicate (4 replicates for each of five bud stages S0, S1, S3, S4, and leaf S5) and split into rows for transcripts originating from oak (Quercus robur), fungi and viruses.

(XLSX)

Click here for additional data file.^{(9.8KB, xlsx)}

S4 Table. Counts of oak (Quercus robur) genes and GO terms that are differentially expressed in both gall and normal bud tissues.

The table gives the total numbers of genes differentially expressed between specific gall developmental stages, and the numbers of these genes that are also differentially expressed between specific stages in normal bud development. The GO term numbers for molecular function (MF) and biological process (BP) categories are the overlap between GO terms enriched for each gall and bud tissue separately. For example, cell H26 represents the overlap in GO terms between genes expressed more highly in Early versus Growth stage galls and genes expressed more highly in S4 buds versus S5 leaves, in this case an overlap of 27 terms is observed. Data are provided as percentages in an otherwise similar version of the table at right. The direction of differential expression (DE) is given for gall and normal bud tissues. Gall stages compared: E = Early, G = Growth, M = Mature. Normal oak bud stages compared: S0, S1, S3, S4 are bud stages, S5 represents a fully open leaf. Gene names for genes contained in this table are given in S11 File.

(XLSX)

Click here for additional data file.^{(12.2KB, xlsx)}

S5 Table. Counts of oak (Quercus robur) genes and GO terms that are only differentially expressed during gall development.

The table gives the numbers of genes differentially expressed between specific gall developmental stages that are not differentially expressed between any stages in normal bud development, and the percentage these genes make up of all differentially expressed oak genes in the given comparison. Direction = in which direction differential expression occurred. Higher is always higher in the earlier developmental stage and vice versa for lower. Percentage = percent of total genes or terms that are only differentially expressed in gall tissues for that comparison. Gene names for genes contained in this table are given in S12 File.

(XLSX)

Click here for additional data file.^{(9.5KB, xlsx)}

S6 Table. Normalised counts for specific genes in gall wasp (Biorhiza pallida) larvae and oak (Quercus robur) gall tissues after differential expression analysis.

The table shows transcript counts across gall stage replicates for specific genes discussed in the text.

(XLSX)

Click here for additional data file.^{(13.9KB, xlsx)}

S7 Table. Oak (Quercus robur) biological process (BP) GO terms that were differentially expressed in gall tissues but not during normal bud development.

The table lists the BP GO terms that were up- or downregulated between each pair of gall developmental stages, but which were not differentially expressed during normal bud development. Annotated = number of genes annotated with that GO term in the complete dataset, Significant = observed number of significantly differentially expressed genes with that annotation, Expected = the expected number of differentially expressed genes with that annotation, P-value = test p-value. We used a significance threshold value of p≤ 0.01.

(XLSX)

Click here for additional data file.^{(18.7KB, xlsx)}

S8 Table. Oak (Quercus robur) molecular function (MF) GO terms that were differentially expressed in gall tissues but not during normal bud development.

The table lists the MF GO terms that were up- or downregulated between each pair of gall developmental stages, but which were not differentially expressed during normal bud development. Annotated = number of genes annotated with that GO term in the complete dataset, Significant = observed number of significantly differentially expressed genes with that annotation, Expected = the expected number of differentially expressed genes with that annotation, P-value = test p-value. We used a significance threshold value of p≤ 0.01.

(XLSX)

Click here for additional data file.^{(15.8KB, xlsx)}

S9 Table. Oak (Quercus robur) biological process (BP) GO terms that were differentially expressed both in gall tissues and during normal bud development.

The table lists the BP GO terms that were differentially expressed in the same direction between gall developmental stages and between S4 buds and S5 leaves in normal bud development. This is equivalent to GO term enrichment of genes numbered in cells H4:H7 of S4 Table. Annotated = number of genes annotated with that GO term in the complete dataset, Significant = observed number of significantly differentially expressed genes with that annotation, Expected = the expected number of differentially expressed genes with that annotation, P-value = test p-value. We used a significance threshold value of p≤ 0.01.

(XLSX)

Click here for additional data file.^{(61.5KB, xlsx)}

S10 Table. Oak (Quercus robur) molecular function (MF) GO terms that were differentially expressed both in gall tissues and during normal bud development.

The table lists the MF GO terms that were differentially expressed in the same direction between gall developmental stages and between S4 buds and S5 leaves in normal bud development. This is equivalent to GO term enrichment of genes numbered in cells H4:H7 of S4 Table. Annotated = number of genes annotated with that GO term in the complete dataset, Significant = observed number of significantly differentially expressed genes with that annotation, Expected = the expected number of differentially expressed genes with that annotation, P-value = test p-value. We used a significance threshold value of p≤ 0.01.

(XLSX)

Click here for additional data file.^{(15.7KB, xlsx)}

S11 Table. Oak (Quercus robur) biological process (BP) GO terms that were differentially expressed between stages in normal bud development.

Terms enriched in each comparison and each direction, either up- or down-regulated are included. Annotated = number of genes annotated with that GO term in the complete dataset, Significant = observed number of significantly differentially expressed genes with that annotation, Expected = the expected number of differentially expressed genes with that annotation, P-value = test p-value. We used a significance threshold value of p≤ 0.01.

(XLSX)

Click here for additional data file.^{(38.9KB, xlsx)}

S12 Table. Oak (Quercus robur) molecular function (MF) GO terms that were differentially expressed between stages in normal bud development.

Terms enriched in each comparison and each direction, either up- or down-regulated are included. Annotated = number of genes annotated with that GO term in the complete dataset, Significant = observed number of significantly differentially genes with that annotation, Expected = the expected number of differentially expressed genes with that annotation, P-value = test p-value. We used a significance threshold value of p≤ 0.01.

(XLSX)

Click here for additional data file.^{(29.4KB, xlsx)}

S13 Table. Nucleotide blast results of Biorhiza pallida and Diplolepis spinosa genomes against Wolbachia genomes.

The best match per cynipid contig by bit score is shown and results are ordered by bit score; an e-value threshold of 1x10^-50 was applied. Biorhiza pallida Wolbachia contigs have best homology to supergroup A Wolbachia, in particular Wmel of Drosophila melanogaster, whereas D. spinosa Wolbachia contigs are most similar to supergroup B Wolbachia.

(XLSX)

Click here for additional data file.^{(114.6KB, xlsx)}

S14 Table. Gall wasp (Biorhiza pallida) biological process (BP) GO terms that were differentially expressed between stages in gall development.

Terms enriched in each comparison and each direction, either up- or down-regulated are included. Annotated = genes annotated with that GO term in the complete dataset, Significant = observed number of significantly differentially expressed genes with that annotation, Expected = the expected number of genes with that annotation, P-value = test p-value. We used a significance threshold value of p≤ 0.01.

(XLSX)

Click here for additional data file.^{(14KB, xlsx)}

S15 Table. Biorhiza pallida molecular function (MF) GO terms.

Terms enriched in each comparison and each direction, either up- or down-regulated are included. Annotated = genes annotated with that GO term in the complete dataset, Significant = observed number of significantly differentially expressed genes with that annotation, Expected = the expected number of genes with that annotation. We used a significance threshold value of p≤ 0.01.

(XLSX)

Click here for additional data file.^{(13.3KB, xlsx)}

S16 Table. Intron sequences identified in Cynipid PCWDE genes.

Gene models were predicted using Augustus version 3.3 with Nasonia vitripennis as the reference species. The number of genes and number of introns predicted per species is shown for each of the three classes of PCWDE enzyme.

(XLSX)

Click here for additional data file.^{(9.6KB, xlsx)}

S17 Table. OrthoFinder orthogroups of genes more highly-expressed in Early gall larvae with at least one orthologue in another Cynipid species.

(XLSX)

Click here for additional data file.^{(18.1KB, xlsx)}

S18 Table. Plant Cell Wall Degrading Enzyme Trinotate annotations.

Annotations are given for all transcripts/isoforms that were annotated as a cellulase, pectin/pectate lyase or rhamnogalacturonan lyase.

(XLSX)

Click here for additional data file.^{(17.1KB, xlsx)}

S19 Table. Nucleotide versus protein blast results of contigs encoding PCWDEs of bacterial origin or most homologous to PCWDEs present in D. ponderosae.

PCWDE sequences for the cynipids Aylax hypecoi, Eschatocerus acaciae and Synergus japonicus, the three most contiguous gall wasp species draft genomes in this study. The table shows best hits for non-overlapping regions of each contig in tabular output format. In addition to annotations for PCWDEs in bacteria and identified as homologous to those in the pine beetle Dendroctonus ponderosae, other regions of the same contigs are annotated as genes with homologues in the Hymenoptera. This confirms that cynipid PCWDE genes are integrated into these species’ nuclear genomes alongside ancestrally hymenopteran genes.

(XLSX)

Click here for additional data file.^{(46.3KB, xlsx)}

S20 Table. Combined Illumina read statistics for raw and filtered data for each transcriptome sample.

All sequencing was paired end. Sample Name = original sample identification; Stage = developmental stage of a replicate in gall or bud; Raw Reads = number of read pairs sequenced; Trimmed Reads = number of read-pairs remaining after quality filtering and trimming; Salmon Mapped Reads = number of read pairs mapped by Salmon; Mapping Rate = percentage of trimmed reads that mapped to the transcriptome.

(XLSX)

Click here for additional data file.^{(10.8KB, xlsx)}

S21 Table. Comparison of cynipid venom gland and larval expressed cellulases, showing these to be distinct.

The table shows results of a nucleotide blast comparison between cellulases and chitinases identified in the Biorhiza pallida venom gland [49], in larvally expressed transcripts and in the B. pallida genome assembly. Results were filtered to included matches with an e-value score of 1e^-5 or less.

(XLSX)

Click here for additional data file.^{(11.9KB, xlsx)}

S22 Table. Mean expression of uniquely DE genes in gall tissue contrasts for gall- and bud tissues.

Genes are sorted by mean expression in the gall comparison, note that mean expression levels are not directly comparable between experiments due to normalisation of gene counts during separate DESEq2 analyses. Where no annotation was made a “.” Is present in the description column.

(XLSX)

Click here for additional data file.^{(107.5KB, xlsx)}

Acknowledgments

We thank Jack Schultz for his detailed and insightful feedback on our paper. We also thank Kingston Lacy House, National Trust and Dalkeith Country Park for permission to collect samples at these sites.

Data Availability

Raw Illumina Hi-Seq data have been deposited in the European Nucleotide Archive under BioProjects PRJEB13357, PRJEB13424, and PRJEB32849.

Funding Statement

This study was supported by a NERC studentship awarded to Jack Hearn, a NERC NBAF grant (NBAF375 2009-2010) awarded to Graham Stone, and NERC grant NE/J010499. Jose-Luis Nieves Aldrey was supported by project CGL2015-66571-P (MINECO, Spain).The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Meyer J, Maresquelle HJ. Anatomie des galles. Berlin: Gebrüder Borntraeger; 1983. [Google Scholar]
2.Rohfritsch O, Shorthouse JD, Braun AC, Kahl G, Schell JS. Insect galls. Academic Press; 1982. [Google Scholar]
3.Dorchin N, Cramer MD, Hoffmann JH. Photosynthesis and sink activity of wasp-induced galls in Acacia pycnantha. Ecology. Wiley Online Library; 2006;87: 1781–1791. 10.1890/0012-9658(2006)87[1781:pasaow]2.0.co;2 [DOI] [PubMed] [Google Scholar]
4.Larson KC, Whitham TG. Manipulation of food resources by a gall-forming aphid: the physiology of sink-source interactions. Oecologia. Springer; 1991;88: 15–21. 10.1007/BF00328398 [DOI] [PubMed] [Google Scholar]
5.Abrahamson WG, McCrea KD. Nutrient and biomass allocation in Solidago altissima: effects of two stem gallmakers, fertilization, and ramet isolation. Oecologia. Springer; 1986;68: 174–180. 10.1007/BF00384784 [DOI] [PubMed] [Google Scholar]
6.Oates CN, Denby KJ, Myburg AA, Slippers B, Naidoo S. Insect gallers and their plant hosts: from omics data to systems biology. Int J Mol Sci. Multidisciplinary Digital Publishing Institute; 2016;17: 1891. [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Poole P, Ramachandran V, Terpolilli J. Rhizobia: from saprophytes to endosymbionts. Nat Rev Microbiol. Nature Publishing Group; 2018;16: 291 10.1038/nrmicro.2017.171 [DOI] [PubMed] [Google Scholar]
8.Martinson EO, Hackett JD, Machado CA, Arnold AE. Metatranscriptome Analysis of Fig Flowers Provides Insights into Potential Mechanisms for Mutualism Stability and Gall Induction. PLoS One. 2015;10: e0130745 10.1371/journal.pone.0130745 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Cook JM, Rasplus J-Y. Mutualists with attitude: coevolving fig wasps and figs. Trends Ecol Evol. Elsevier; 2003;18: 241–248. [Google Scholar]
10.Cook JM, West SA. Figs and fig wasps. Curr Biol. Elsevier; 2005;15: R978—R980. 10.1016/j.cub.2005.11.057 [DOI] [PubMed] [Google Scholar]
11.Wesemael W, Viaene N, Moens M. Root-knot nematodes (Meloidogyne spp.) in Europe. Nematology. Brill; 2011;13: 3–16. [Google Scholar]
12.Csóka G, Stone GN, Melika G. Non-native gall-inducing insects on forest trees: a global review. Biol Invasions. Springer; 2017;19: 3161–3181. [Google Scholar]
13.Escobar MA, Dandekar AM. Agrobacterium tumefaciens as an agent of disease. Trends Plant Sci. Elsevier; 2003;8: 380–386. 10.1016/S1360-1385(03)00162-6 [DOI] [PubMed] [Google Scholar]
14.Hoffmann JH, Impson FAC, Moran VC, Donnelly D. Biological control of invasive golden wattle trees (Acacia pycnantha) by a gall wasp, Trichilogaster sp.(Hymenoptera: Pteromalidae), in South Africa. Biol Control. Elsevier; 2002;25: 64–73. [Google Scholar]
15.Story JM, Smith L, Corn JG, White LJ. Influence of seed head-attacking biological control agents on spotted knapweed reproductive potential in western Montana over a 30-year period. Environ Entomol. Oxford University Press Oxford, UK; 2008;37: 510–519. [DOI] [PubMed] [Google Scholar]
16.Lacroix B, Citovsky V. Pathways of DNA Transfer to Plants from Agrobacterium tumefaciens and Related Bacterial Species. Annu Rev Phytopathol. Annual Reviews; 2019;57: 231–251. 10.1146/annurev-phyto-082718-100101 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Le Fevre R, Evangelisti E, Rey T, Schornack S. Modulation of host cell biology by plant pathogenic microbes. Annu Rev Cell Dev Biol. Annual Reviews; 2015;31: 201–229. 10.1146/annurev-cellbio-102314-112502 [DOI] [PubMed] [Google Scholar]
18.Abad P, Gouzy J, Aury J-M, Castagnone-Sereno P, Danchin EGJ, Deleury E, et al. Genome sequence of the metazoan plant-parasitic nematode Meloidogyne incognita. Nat Biotechnol. 2008;26: 909–15. 10.1038/nbt.1482 [DOI] [PubMed] [Google Scholar]
19.Mejias J, Truong NM, Abad P, Favery B, Quentin M. Plant proteins and processes targeted by parasitic nematode effectors. Front Plant Sci. Frontiers Media SA; 2019;10. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Rehman S, Gupta VK, Goyal AK. Identification and functional analysis of secreted effectors from phytoparasitic nematodes. BMC Microbiol. BioMed Central; 2016;16: 48 10.1186/s12866-016-0632-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Stuart JJ, Chen M-S, Shukle R, Harris MO. Gall midges (Hessian flies) as plant pathogens. Annu Rev Phytopathol. 2012;50: 339–57. 10.1146/annurev-phyto-072910-095255 [DOI] [PubMed] [Google Scholar]
22.Chen M-S, Fellers JP, Zhu YC, Stuart JJ, Hulbert S, El-Bouhssini M, et al. A super-family of genes coding for secreted salivary gland proteins from the Hessian fly, Mayetiola destructor. J Insect Sci. 2006;6: 12 10.1673/2006.06.12.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Zhao C, Shukle R, Navarro-Escalante L, Chen M, Richards S, Stuart JJ. Avirulence gene mapping in the Hessian fly (Mayetiola destructor) reveals a protein phosphatase 2C effector gene family. J Insect Physiol. 2016;84: 22–31. 10.1016/j.jinsphys.2015.10.001 [DOI] [PubMed] [Google Scholar]
24.Aggarwal R, Subramanyam S, Zhao C, Chen M-S, Harris MO, Stuart JJ. Avirulence effector discovery in a plant galling and plant parasitic arthropod, the Hessian fly (Mayetiola destructor). PLoS One. Public Library of Science; 2014;9: e100958 10.1371/journal.pone.0100958 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Schultz JC, Edger PP, Body MJA, Appel HM. A galling insect activates plant reproductive programs during gall development. Sci Rep. 2019;9: 1833 10.1038/s41598-018-38475-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Nabity PD, Haus MJ, Berenbaum MR, DeLucia EH. Leaf-galling phylloxera on grapes reprograms host metabolism and morphology. Proc Natl Acad Sci. National Acad Sciences; 2013;110: 16663–16668. 10.1073/pnas.1220219110 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Raman A, Schaefer CW, Withers TM, others. Biology, ecology, and evolution of gall-inducing arthropods Boca Raton: CRC Press; 2005. [Google Scholar]
28.Rohfritsch O. Patterns in gall development In: Shorthouse JD, Rohfritsch O, editors. Biology of insect-induced galls. Oxford University Press: New York; 1992. pp. 60–86. [Google Scholar]
29.Stone GN, Schönrogge K. The adaptive significance of insect gall morphology. Trends Ecol Evol. 2003;18: 512–522. 10.1016/S0169-5347(03)00247-7 [DOI] [Google Scholar]
30.Dawkins R. The Extended Phenotype. Oxford: Oxford University Press; 1989. [Google Scholar]
31.Crespi B, Worobey M. Comparative Analysis of Gall Morphology in Australian Gall Thrips: The Evolution of Extended Phenotypes. Evolution (N Y). 1998; 1686–1696. [DOI] [PubMed] [Google Scholar]
32.Stone GN, Cook JM. The structure of cynipid oak galls: patterns in the evolution of an extended phenotype. Proc R Soc B Biol Sci. 1998;265 10.1098/rspb.1998.0387 [DOI] [Google Scholar]
33.Kutsukake M, Meng X-Y, Katayama N, Nikoh N, Shibao H, Fukatsu T. An insect-induced novel plant phenotype for sustaining social life in a closed system. Nat Commun. Nature Publishing Group; 2012;3: 1187 10.1038/ncomms2187 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Stern DL. Phylogenetic evidence that aphids, rather than plants, determine gall morphology. Proc R Soc London Ser B Biol Sci. The Royal Society London; 1995;260: 85–89. [Google Scholar]
35.Giron D, Huguet E, Stone GN, Body M. Insect-induced effects on plants and possible effectors used by galling and leaf-mining insects to manipulate their host-plant. J Insect Physiol. 2015;84: 70–89. 10.1016/j.jinsphys.2015.12.009 [DOI] [PubMed] [Google Scholar]
36.Tooker JF, Helms AM. Phytohormone Dynamics Associated with Gall Insects, and their Potential Role in the Evolution of the Gall-Inducing Habit. J Chem Ecol. 2014;40: 742–753. 10.1007/s10886-014-0457-6 [DOI] [PubMed] [Google Scholar]
37.Csóka G, Stone GN, Melika G. The biology, ecology and evolution of Gall-inducing Cynipidae In: Raman C, Schaefer W, Withers TM, editors. Biology, ecology and evolution of gall inducing insects. Boca Raton: CRC Press; 2005. pp. 573–642. [Google Scholar]
38.Bailey R, Schönrogge K, Cook JM, Melika G, Csóka G, Thúroczy C, et al. Host niches and defensive extended phenotypes structure parasitoid wasp communities. PLoS Biol. 2009;7: e1000179 10.1371/journal.pbio.1000179 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Ronquist F, Liljeblad J. Evolution of the gall wasp—host plant association. Evolution (N Y). 2001;55: 2503–2522. [DOI] [PubMed] [Google Scholar]
40.Shorthouse JD, Leggo JJ, Sliva MD, Lalonde RG. Has egg location influenced the radiation of Diplolepis (Hymenoptera: Cynipidae) gall wasps on wild roses? Basic Appl Ecol. 2005;6: 423–434. 10.1016/j.baae.2005.07.006 [DOI] [Google Scholar]
41.Hough JS. Studies on the common spangle gall of oak. New Phytol. 1953;52: 149–177. [Google Scholar]
42.Stone GN, Hernandez-Lopez A, Nicholls JA, di Pierro E, Pujade-Villar J, Melika G, et al. Extreme host plant conservatism during at least 20 million years of host plant pursuit by oak gallwasps. Evolution (N Y). 2009;63: 854–869. [DOI] [PubMed] [Google Scholar]
43.Hough JS. Studies on the common spangle gall of oak: II. A general consideration of past work on gall induction. New Phytol. 1953;52: 218–228. [Google Scholar]
44.Rey L. La galle de Biorhiza pallida Bosc. Sur la racine de Quercus pedunculata Ehrh.: observation de la ponte et de les premières réactions de l’hôte. Comptes rendus l’Académie des Sci. 1967;264: 2891–2894. [Google Scholar]
45.Rey L. La galle de Biorhiza pallida Ol. Etude préliminaire du bourgeon de Quercus pedunculata Ehrh. au moment de l’attaque du parasite. Bull la Société Bot Fr. 1966;113: 503–514. [Google Scholar]
46.Garbin L, Durfort M, Diaz NB, Pujade-Villar J. Histological Modifications on Quercus suber Twigs (Fagaceae) Caused by the Gallwasp Plagiotrochus suberi (Hymenoptera: Cynipidae). Entomol Gen. 2005;28: 91–102. [Google Scholar]
47.Derory J, Léger P, Garcia V, Schaeffer J, Hauser MT, Salin F, et al. Transcriptome analysis of bud burst in sessile oak (Quercus petraea). New Phytol. 2006;170: 723–738. 10.1111/j.1469-8137.2006.01721.x [DOI] [PubMed] [Google Scholar]
48.Plomion C, Aury J-M, Amselem J, Leroy T, Murat F, Duplessis S, et al. Oak genome reveals facets of long lifespan. Nat Plants. Nature Publishing Group; 2018;4: 440 10.1038/s41477-018-0172-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Cambier S, Ginis O, Moreau SJM, Gayral P, Hearn J, Stone GN, et al. Gall wasp transcriptomes unravel potential effectors involved in molecular dialogues with oak and rose bushes. Front Physiol. 2019;10: 926 10.3389/fphys.2019.00926 [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Hearn J, Stone GN, Bunnefeld L, Nicholls JA, Barton NH, Lohse K. Likelihood-based inference of population history from low-coverage de novo genome assemblies. Mol Ecol. 2014;3643: 198–211. 10.1111/mec.12578 [DOI] [PubMed] [Google Scholar]
51.Bunnefeld L, Hearn J, Stone GN, Lohse K. Whole-genome data reveal the complex history of a diverse ecological community. Proc Natl Acad Sci. 2018;115: E6507–E6515. 10.1073/pnas.1800334115 [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Nicholls JA, Melika G, Stone GN. Sweet Tetra-Trophic Interactions: Multiple Evolution of Nectar Secretion, a Defensive Extended Phenotype in Cynipid Gall Wasps. Am Nat. 2017;189: 67–77. 10.1086/689399 [DOI] [PubMed] [Google Scholar]
53.Cook JM, Rokas A, Pagel M, Stone GN. Evolutionary shift between host oak section and host-plant organs in Andricus gallwasps. Evolution (N Y). 2002;56: 1821–1830. [DOI] [PubMed] [Google Scholar]
54.Ronquist F, Nieves-Aldrey J-L, Buffington ML, Liu Z, Liljeblad J, Nylander JAA. Phylogeny, evolution and classification of gall wasps: the plot thickens. PLoS One. 2015;10: e0123301 10.1371/journal.pone.0123301 [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Phylogeny Ronquist F., classification and evolution of the Cynipoidea. Zool Scr. 1999;28: 139–164. 10.1046/j.1463-6409.1999.00022.x [DOI] [Google Scholar]
56.Beyerinck MW. Beobachtungen ueber die ersten Entwicklungsphasen einiger Cynipidengallen Amsterdam: Mueller; 1882; [Google Scholar]
57.Brooks SE, Shorthouse JD. Developmental morphology of stem galls of Diplolepis nodulosa (Hymenoptera: Cynipidae) and those modified by the inquiline Periclistus pirata (Hymenoptera: Cynipidae) on Rosa blanda (Rosaceae). Can J Bot. NRC Research Press; 1998;76: 365–381. [Google Scholar]
58.Rey L. Particular aspects of the endoplasmic reticulum observed during the evolution of the nutrient cells of the gall caused by Biorhiza pallida [Quercus]. Comptes Rendus Hebd des Seances l’Academie des Sci Ser D. 1975; [Google Scholar]
59.Harper LJ, Schönrogge K, Lim KY, Francis P, Lichtenstein CP. Cynipid galls: insect-induced modifications of plant development create novel plant organs. Plant Cell Environ. 2004;27: 327–335. [Google Scholar]
60.Schönrogge K, Harper LJ, Lichtenstein CP. The protein content of tissues in cynipid galls (Hymenoptera: Cynipidae): Similarities between cynipid galls and seeds. Plant Cell Environ. 2000;23: 215–222. [Google Scholar]
61.Atkinson RJ, Brown GS, Stone GN. Skewed sex ratios and multiple founding in galls of the oak apple gall wasp Biorhiza pallida. Ecol Entomol. 2003;28: 14–24. [Google Scholar]
62.Folliot R. Les insectes cécidogènes et la cécidogenèse. Trait Zool Anat systématique, Biol. 1977;8: 389–429. [Google Scholar]
63.Bronner R. The role of nutritive cells in the nutrition of cynipids and cecidomyiids In: Shorthouse JD, Rohfritsch O, editors. Biology of insect-induced galls. New York: Oxford University Press; 1992. pp. 118–140. [Google Scholar]
64.Shorthouse JD, Rohfritsch O. Biology of insect-induced galls New York, USA: Oxford University Press; 1992. [Google Scholar]
65.Bagatto G, Paquette LC, Shorthouse JD. Influence of galls of Phanacis taraxaci on carbon partitioning within common dandelion, Taraxacum officinale. Entomol Exp Appl. 1996;79: 111–117. [Google Scholar]
66.Rey L. La galle de Biorhiza pallida Ol.: stades ultérieurs de développement. Marcellia. 1971;37: 193–218. [Google Scholar]
67.Sinclair FH, Stone GN, Nicholls JA, Cavers S, Gibbs M, Butterill P, et al. Impacts of local adaptation of forest trees on associations with herbivorous insects: implications for adaptive forest management. Evol Appl. Wiley Online Library; 2015;8: 972–987. 10.1111/eva.12329 [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Hartley SE, Lawton JH. Host-plant manipulation by gall-insects: a test of the nutrition hypothesis. J Anim Ecol. JSTOR; 1992; 113–119. [Google Scholar]
69.Allison SD, Schultz JC. Biochemical responses of chestnut oak to a galling cynipid. J Chem Ecol. 2005;31: 151–166. [DOI] [PubMed] [Google Scholar]
70.Elborough K, Winz R, Deka R, Markham J, White A, Rawsthorne S, et al. Biotin carboxyl carrier protein and carboxyltransferase subunits of the multi-subunit form of acetyl-CoA carboxylase from Brassica napus: cloning and analysis of expression during oilseed rape embryogenesis. Biochem J. 1996;315: 103–112. 10.1042/bj3150103 [DOI] [PMC free article] [PubMed] [Google Scholar]
71.Wilhelm E. Somatic embryogenesis in oak (Quercus spp.). Vitr Cell Dev Biol—Plant. 2000;36: 349–357. 10.1007/s11627-000-0062-y [DOI] [Google Scholar]
72.Gibson CM, Hunter MS. Extraordinarily widespread and fantastically complex: Comparative biology of endosymbiotic bacterial and fungal mutualists of insects. Ecol Lett. 2010;13: 223–234. 10.1111/j.1461-0248.2009.01416.x [DOI] [PubMed] [Google Scholar]
73.Kaiser W, Huguet E, Casas J, Commin C, Giron D. Plant green-island phenotype induced by leaf-miners is mediated by bacterial symbionts. Proc R Soc B Biol Sci. 2010;277: 2311–2319. 10.1098/rspb.2010.0214 [DOI] [PMC free article] [PubMed] [Google Scholar]
74.Bansal R, Hulbert S, Schemerhorn B, Reese JC, Whitworth RJ, Stuart JJ, et al. Hessian Fly-Associated Bacteria: Transmission, Essentiality, and Composition. PLoS One. 2011;6: e23170 10.1371/journal.pone.0023170 [DOI] [PMC free article] [PubMed] [Google Scholar]
75.Joy JB. Symbiosis catalyses niche expansion and diversification. Proc Biol Sci. 2013;280: 20122820 10.1098/rspb.2012.2820 [DOI] [PMC free article] [PubMed] [Google Scholar]
76.Nelson LA, Davies KA, Scheffer SJ, Taylor GS, Purcell MF, Giblin-Davis RM, et al. An emerging example of tritrophic coevolution between flies (Diptera: Fergusoninidae) and nematodes (Nematoda: Neotylenchidae) on myrtaceae host plants. Biol J Linn Soc. 2014;111: 699–718. 10.1111/bij.12237 [DOI] [Google Scholar]
77.Borkent A, Bissett J. Gall midges (Diptera: Cecidomyiidae) are vectors for their fungal symbionts. Symbiosis (USA). 1985; [Google Scholar]
78.Joy JB. Symbiosis catalyses niche expansion and diversification. Proc R Soc B Biol Sci. 2013;280. [DOI] [PMC free article] [PubMed] [Google Scholar]
79.Cornell HV. The Secondary Chemistry and Complex Morphology of Galls Formed by the Cynipinae (Hymenoptera): Why and How? Am Midl Nat. 1983;110: 225–234. 10.2307/2425263 [DOI] [Google Scholar]
80.Bézier A, Annaheim M, Herbinière J, Wetterwald C, Gyapay G, Bernard-Samain S, et al. Polydnaviruses of Braconid Wasps Derive from an Ancestral Nudivirus. Science (80-). 2009;323: 926–930. 10.1126/science.1166788 [DOI] [PubMed] [Google Scholar]
81.Drezen J-M, Leobold M, Bezier A, Huguet E, Volkoff A-N, Herniou EA. Endogenous viruses of parasitic wasps: variations on a common theme. Curr Opin Virol. Elsevier; 2017;25: 41–48. 10.1016/j.coviro.2017.07.002 [DOI] [PubMed] [Google Scholar]
82.Rizki RM, Rizki TM. Parasitoid virus-like particles destroy Drosophila cellular immunity. Proc Natl Acad Sci. National Acad Sciences; 1990;87: 8388–8392. 10.1073/pnas.87.21.8388 [DOI] [PMC free article] [PubMed] [Google Scholar]
83.H R. Von welchen Organen der Gallwespenlarve geht der Reiz zur Bildung der Pflanzengalle aus? Zoologischer Jahrbücher. Abteilung für Syst Geogr und Biol der Tiere. 1904;20: 19–90. [Google Scholar]
84.Roth P. Beiträge zur Biologie der Gallwespen. Verhandlungen der Naturforscher Gesellschaft Basel. 1949;60: 104–178. [Google Scholar]
85.Westermann AJ, Gorski SA, Vogel J. Dual RNA-seq of pathogen and host. Nat Rev Microbiol. 2012;10: 618–630. 10.1038/nrmicro2852 [DOI] [PubMed] [Google Scholar]
86.Nabity PD. Insect-induced plant phenotypes: Revealing mechanisms through comparative genomics of galling insects and their hosts. Am J Bot. 2016;103: 979–981. 10.3732/ajb.1600111 [DOI] [PubMed] [Google Scholar]
87.Zhu C, Shi F, Chen Y, Wang M, Zhao Y, Geng G. Transcriptome Analysis of Chinese Chestnut (Castanea mollissima Blume) in Response to Dryocosmus kuriphilus Yasumatsu Infestation. Int J Mol Sci. Multidisciplinary Digital Publishing Institute; 2019;20: 855. [DOI] [PMC free article] [PubMed] [Google Scholar]
88.Jia XY, He LH, Jing RL, Li RZ. Calreticulin: Conserved protein and diverse functions in plants. Physiol Plant. 2009;136: 127–138. 10.1111/j.1399-3054.2009.1223.x [DOI] [PubMed] [Google Scholar]
89.Zürcher E, Liu J, di Donato M, Geisler M, Müller B. Plant development regulated by cytokinin sinks. Science (80-). 2016;353: 1027–1030. 10.1126/science.aaf7254 [DOI] [PubMed] [Google Scholar]
90.Seo M, Akaba S, Oritani T, Delarue M, Bellini C, Caboche M, et al. Higher Activity of an Aldehyde Oxidase in the Auxin-Overproducing superroot 1 mutant of Arabidopsis thaliana. Plant Physiol. Am Soc Plant Biol; 1998;116: 687–693. 10.1104/pp.116.2.687 [DOI] [PMC free article] [PubMed] [Google Scholar]
91.Rhoades MW, Reinhart BJ, Lim LP, Burge CB, Bartel B, Bartel DP. Prediction of plant microRNA targets. Cell. Elsevier; 2002;110: 513–520. 10.1016/s0092-8674(02)00863-2 [DOI] [PubMed] [Google Scholar]
92.Iwama A, Yamashino T, Tanaka Y, Sakakibara H, Kakimoto T, Sato S, et al. AHK5 histidine kinase regulates root elongation through an ETR1-dependent abscisic acid and ethylene signaling pathway in Arabidopsis thaliana. Plant cell Physiol. Oxford University Press; 2007;48: 375–380. 10.1093/pcp/pcl065 [DOI] [PubMed] [Google Scholar]
93.Tran L-SP, Urao T, Qin F, Maruyama K, Kakimoto T, Shinozaki K, et al. Functional analysis of AHK1/ATHK1 and cytokinin receptor histidine kinases in response to abscisic acid, drought, and salt stress in Arabidopsis. Proc Natl Acad Sci. National Acad Sciences; 2007;104: 20623–20628. 10.1073/pnas.0706547105 [DOI] [PMC free article] [PubMed] [Google Scholar]
94.Suzuki H, Yokokura J, Ito T, Arai R, Yokoyama C, Toshima H, et al. Biosynthetic pathway of the phytohormone auxin in insects and screening of its inhibitors. Insect Biochem Mol Biol. 2014;53: 66–72. 10.1016/j.ibmb.2014.07.008 [DOI] [PubMed] [Google Scholar]
95.Römer P, Hahn S, Jordan T, Strauß T, Bonas U, Lahaye T. Plant pathogen recognition mediated by promoter activation of the pepper Bs3 resistance gene. Science (80-). American Association for the Advancement of Science; 2007;318: 645–648. [DOI] [PubMed] [Google Scholar]
96.Field LM, James AA, Plantard O, Rasplus J-Y, Mondor G, Clainche I, et al. Distribution and phylogeny of Wolbachia inducing thelytoky in Rhoditini and Aylacini (Hymenoptera: Cynipidae). Insect Mol Biol. 1999;8: 185–191. [DOI] [PubMed] [Google Scholar]
97.Rokas A, Atkinson RJ, Nieves-Aldrey JL, West SA, Stone GN. The incidence and diversity of Wolbachia in gallwasps (Hymenoptera; Cynipidae) on oak. Mol Ecol. 2002;11: 1815–1829. 10.1046/j.1365-294x.2002.01556.x [DOI] [PubMed] [Google Scholar]
98.Bailey TL, Williams N, Misleh C, Li WW. MEME: discovering and analyzing DNA and protein sequence motifs. Nucleic Acids Res. 2006;34: 369–373. 10.1093/nar/gkl198 [DOI] [PMC free article] [PubMed] [Google Scholar]
99.Keeling CI, Henderson H, Li M, Yuen M, Clark EL, Fraser JD, et al. Transcriptome and full-length cDNA resources for the mountain pine beetle, Dendroctonus ponderosae Hopkins, a major insect pest of pine forests. Insect Biochem Mol Biol. 2012;42: 525–36. 10.1016/j.ibmb.2012.03.010 [DOI] [PubMed] [Google Scholar]
100.Keeling CI, Yuen MM, Liao NY, Docking TR, Chan SK, Taylor GA, et al. Draft genome of the mountain pine beetle, Dendroctonus ponderosae Hopkins, a major forest pest. Genome Biol. BioMed Central Ltd; 2013;14: R27 10.1186/gb-2013-14-3-r27 [DOI] [PMC free article] [PubMed] [Google Scholar]
101.Pauchet Y, Heckel DG. The genome of the mustard leaf beetle encodes two active xylanases originally acquired from bacteria through horizontal gene transfer. Proc Biol Sci. 2013;280: 20131021 10.1098/rspb.2013.1021 [DOI] [PMC free article] [PubMed] [Google Scholar]
102.Pauchet Y, Wilkinson P, Chauhan R, Ffrench-Constant RH. Diversity of beetle genes encoding novel plant cell wall degrading enzymes. PLoS One. 2010;5: e15635 10.1371/journal.pone.0015635 [DOI] [PMC free article] [PubMed] [Google Scholar]
103.De Lorenzo G, Ferrari S, Giovannoni M, Mattei B, Cervone F. Cell wall traits that influence plant development, immunity, and bioconversion. Plant J. Wiley Online Library; 2019;97: 134–147. 10.1111/tpj.14196 [DOI] [PubMed] [Google Scholar]
104.Vallarino JG, Osorio S. Signaling role of oligogalacturonides derived during cell wall degradation. Plant Signal Behav. Landes Bioscience; 2012;7: 1447–1449. 10.4161/psb.21779 [DOI] [PMC free article] [PubMed] [Google Scholar]
105.Pawłowski TA, Staszak AM, Karolewski P, Giertych MJ. Plant development reprogramming by cynipid gall wasp: proteomic analysis. Acta Physiol Plant. Springer; 2017;39: 114. [Google Scholar]
106.Takei M, Kogure S, Yokoyama C, Kouzuma Y, Suzuki Y. Identification of an aldehyde oxidase involved in indole-3-acetic acid synthesis in Bombyx mori silk gland. Biosci Biotechnol Biochem. Taylor & Francis; 2019;83: 129–136. 10.1080/09168451.2018.1525275 [DOI] [PubMed] [Google Scholar]
107.Zhao CY, Escalante LN, Chen H, Benatti TR, Qu JX, Chellapilla S, et al. A massive expansion of effector genes underlies gall-formation in the wheat pest Mayetiola destructor. Curr Biol. 2015;25 10.1016/j.cub.2014.12.057 [DOI] [PubMed] [Google Scholar]
108.Abe Y, Ide T, Wachi N. Discovery of a New Gall-Inducing Species in the Inquiline Tribe Synergini (Hymenoptera: Cynipidae): Inconsistent Implications from Biology and Morphology. Ann Entomol Soc Am. 2011;104: 115–120. 10.1603/AN10149 [DOI] [Google Scholar]
109.Stanke M, Morgenstern B. AUGUSTUS: a web server for gene prediction in eukaryotes that allows user-defined constraints. Nucleic Acids Res. Oxford University Press; 2005;33: W465—W467. 10.1093/nar/gki458 [DOI] [PMC free article] [PubMed] [Google Scholar]
110.Volkenburgh E Van. Leaf expansion—an integrating plant behaviour. Plant Cell Environ. Wiley Online Library; 1999;22: 1463–1473. [Google Scholar]
111.Maksymowych R. Analysis of leaf development. CUP Archive; 1973. [Google Scholar]
112.Becraft PW. 1 Development of the Leaf Epidermis Current topics in developmental biology. Elsevier; 1999. pp. 1–40. [DOI] [PubMed] [Google Scholar]
113.Botta R, Sartor C, Torello Marinoni D, Quacchia A, Alma A. Differential gene expression in chestnut buds following infestation by gall-wasp (Dryocosmus kuriphilus Yasumatsu, Hymenoptera: Cynipidae). IV International Chestnut Symposium 844. 2008. pp. 405–410.
114.van Hengel AJ, Tadesse Z, Immerzeel P, Schols H, van Kammen A, de Vries SC. N-acetylglucosamine and glucosamine-containing arabinogalactan proteins control somatic embryogenesis. Plant Physiol. 2001;125: 1880–90. 10.1104/pp.125.4.1880 [DOI] [PMC free article] [PubMed] [Google Scholar]
115.De Jong AJ, Cordewener J, Lo Schiavo F, Terzi M, Vandekerckhove J, Van Kammen A, et al. A carrot somatic embryo mutant is rescued by chitinase. Plant Cell. 1992;4: 425–33. 10.1105/tpc.4.4.425 [DOI] [PMC free article] [PubMed] [Google Scholar]
116.Poon S, Heath RL, Clarke AE. A chimeric arabinogalactan protein promotes somatic embryogenesis in cotton cell culture. Plant Physiol. 2012;160: 684–95. 10.1104/pp.112.203075 [DOI] [PMC free article] [PubMed] [Google Scholar]
117.Fehér A, Pasternak TP, Dudits D. Transition of somatic plant cells to an embryogenic state. Plant Cell, Tissue Organ Cult. 2003;74: 201–228. 10.1023/A:1024033216561 [DOI] [Google Scholar]
118.Nguema-Ona E, Vicré-Gibouin M, Cannesan M-A, Driouich A. Arabinogalactan proteins in root-microbe interactions. Trends Plant Sci. 2013;18: 440–449. 10.1016/j.tplants.2013.03.006 [DOI] [PubMed] [Google Scholar]
119.Van Hengel AJ, Van Kammen A, De Vries SC. A relationship between seed development, Arabinogalactan-proteins (AGPs) and the AGP mediated promotion of somatic embryogenesis. Physiol Plant. 2002;114: 637–644. [DOI] [PubMed] [Google Scholar]
120.Costa ML, Sobral R, Costa MMR, Amorim MI, Coimbra S. Evaluation of the presence of arabinogalactan proteins and pectins during Quercus suber male gametogenesis. Ann Bot. Annals Botany Co; 2014; mcu223. [DOI] [PMC free article] [PubMed] [Google Scholar]
121.De Jong AJ, Heidstra R, Spaink HP, Hartog MV, Meijer EA, Hendriks T, et al. Rhizobium Lipooligosaccharides Rescue a Carrot Somatic Embryo Mutant. Plant Cell. 1993;5: 615–620. 10.1105/tpc.5.6.615 [DOI] [PMC free article] [PubMed] [Google Scholar]
122.Röhrig H, Schmidt J, Walden R, Czaja I, Miklaševičs E, Wieneke U, et al. Growth of Tobacco Protoplasts Stimulated by Synthetic Lipo-Chitooligosaccharides. Science (80-). 1995;269: 841–843. 10.1126/science.269.5225.841 [DOI] [PubMed] [Google Scholar]
123.Schultze M, Kondorosi A. Regulation of symbiotic root nodule development. Annu Rev Genet. 1998;32: 33–57. 10.1146/annurev.genet.32.1.33 [DOI] [PubMed] [Google Scholar]
124.Cannesan M-A, Nguema-Ona E. Arabinogalactan proteins in root–microbe interactions. Trends Plant Sci. 2013;18: 440–449. 10.1016/j.tplants.2013.03.006 [DOI] [PubMed] [Google Scholar]
125.Vandenbussche F, Vaseva I, Vissenberg K, Van Der Straeten D. Ethylene in vegetative development: a tale with a riddle. New Phytol. Wiley Online Library; 2012;194: 895–909. 10.1111/j.1469-8137.2012.04100.x [DOI] [PubMed] [Google Scholar]
126.Gresshoff PM, Lohar D, Chan P-K, Biswas B, Jiang Q, Reid D, et al. Genetic analysis of ethylene regulation of legume nodulation. Plant Signal Behav. Taylor & Francis; 2009;4: 818–823. 10.4161/psb.4.9.9395 [DOI] [PMC free article] [PubMed] [Google Scholar]
127.Tiburcio AF, Altabella T, Bitrián M, Alcázar R. The roles of polyamines during the lifespan of plants: from development to stress. Planta. Springer; 2014;6: 1–18. [DOI] [PubMed] [Google Scholar]
128.Volkoff A-N, Jouan V, Urbach S, Samain S, Bergoin M, Wincker P, et al. Analysis of virion structural components reveals vestiges of the ancestral ichnovirus genome. PLoS Pathog. 2010;6: e1000923 10.1371/journal.ppat.1000923 [DOI] [PMC free article] [PubMed] [Google Scholar]
129.Webster CL, Waldron FM, Robertson S, Crowson D, Ferrari G, Quintana JF, et al. The discovery, distribution, and evolution of viruses associated with Drosophila melanogaster. PLoS Biol. Public Library of Science; 2015;13: e1002210 10.1371/journal.pbio.1002210 [DOI] [PMC free article] [PubMed] [Google Scholar]
130.Quentin M, Abad P, Favery B. Plant parasitic nematode effectors target host defense and nuclear functions to establish feeding cells. Front Plant Sci. 2013;4: 53 10.3389/fpls.2013.00053 [DOI] [PMC free article] [PubMed] [Google Scholar]
131.Boulain H, Legeai F, Guy E, Morlière S, Douglas NE, Oh J, et al. Fast Evolution and Lineage-Specific Gene Family Expansions of Aphid Salivary Effectors Driven by Interactions with Host-Plants. Genome Biol Evol. Oxford University Press; 2018;10: 1554–1572. 10.1093/gbe/evy097 [DOI] [PMC free article] [PubMed] [Google Scholar]
132.Bailey S, Percy DM, Hefer CA, Cronk QCB. The transcriptional landscape of insect galls: psyllid (Hemiptera) gall formation in Hawaiian Metrosideros polymorpha (Myrtaceae). BMC Genomics. 2015;16: 943 10.1186/s12864-015-2109-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
133.Tanaka Y, Okada K, Asami T, Suzuki Y. Phytohormones in Japanese Mugwort Gall Induction by a Gall-Inducing Gall Midge. Biosci Biotechnol Biochem. 2013;77: 1942–1948. 10.1271/bbb.130406 [DOI] [PubMed] [Google Scholar]
134.Takei M, Yoshida S, Kawai T, Hasegawa M, Suzuki Y. Adaptive significance of gall formation for a gall-inducing aphids on Japanese elm trees. J Insect Physiol. 2015;72: 43–51. 10.1016/j.jinsphys.2014.11.006 [DOI] [PubMed] [Google Scholar]
135.Ohkawa M. Isolation of zeatin from larvae of Dryocosmus kuriphilus Yasumatsu [Castanea crenata, gall initiation]. HortScience (USA). 1974;9: 458–459. [Google Scholar]
136.Percy DM, Page RDM, Cronk QCB. Plant-insect interactions: double-dating associated insect and plant lineages reveals asynchronous radiations. Syst Biol. 2004;53: 120–127. 10.1080/10635150490264996 [DOI] [PubMed] [Google Scholar]
137.Wood BW, Payne JA. Growth regulators in chestnut shoot galls infected with oriental chestnut gall wasp, Dryocosmus kuriphilus (Hymenoptera: Cynipidae). Environ Entomol. 1988;17: 915–920. [Google Scholar]
138.Brütting C, Crava CM, Schäfer M, Schuman MC, Meldau S, Adam N, et al. Cytokinin transfer by a free-living mirid to Nicotiana attenuata recapitulates a strategy of endophytic insects. Elife. eLife Sciences Publications Limited; 2018;7: e36268 10.7554/eLife.36268 [DOI] [PMC free article] [PubMed] [Google Scholar]
139.Bartlett L, Connor EF. Exogenous phytohormones and the induction of plant galls by insects. Arthropod Plant Interact. 2014;8: 339–348. 10.1007/s11829-014-9309-0 [DOI] [Google Scholar]
140.John M, Röhrig H, Schmidt J, Walden R, Schell J. Cell signalling by oligosaccharides. Trends Plant Sci. Elsevier; 2015;2: 111–115. 10.1016/S1360-1385(97)01005-4 [DOI] [Google Scholar]
141.Yue J, Hu X, Huang J. Origin of plant auxin biosynthesis. Trends Plant Sci. Elsevier; 2014;19: 764–770. 10.1016/j.tplants.2014.07.004 [DOI] [PubMed] [Google Scholar]
142.Nieminen KM, Kauppinen L, Helariutta Y. A weed for wood? Arabidopsis as a genetic model for xylem development. Plant Physiol. Am Soc Plant Biol; 2004;135: 653–659. 10.1104/pp.104.040212 [DOI] [PMC free article] [PubMed] [Google Scholar]
143.Bar M, Ori N. Leaf development and morphogenesis. Development. 2014;141: 4219–4230. 10.1242/dev.106195 [DOI] [PubMed] [Google Scholar]
144.Body MJA, Zinkgraf MS, Whitham TG, Lin C-H, Richardson RA, Appel HM, et al. Heritable Phytohormone Profiles of Poplar Genotypes Vary in Resistance to a Galling Aphid. Mol Plant-Microbe Interact. 2019;32: 654–672. 10.1094/MPMI-11-18-0301-R [DOI] [PubMed] [Google Scholar]
145.Kunieda T, Fujiyuki T, Kucharski R, Foret S, Ament SA, Toth AL, et al. Carbohydrate metabolism genes and pathways in insects: insights from the honey bee genome. Insect Mol Biol. 2006;15: 563–576. 10.1111/j.1365-2583.2006.00677.x [DOI] [PMC free article] [PubMed] [Google Scholar]
146.Mckenna DD, Scully ED, Pauchet Y, Hoover K, Kirsch R, Geib SM, et al. Genome of the Asian longhorned beetle (Anoplophora glabripennis), a globally significant invasive species , reveals key functional and evolutionary innovations at the beetle–plant interface. Genome Biol. 2016; 1–18. 10.1186/s13059-015-0866-z [DOI] [PMC free article] [PubMed] [Google Scholar]
147.Kirsch R, Gramzow L, Theißen G, Siegfried BD, Richard H, Heckel DG, et al. Horizontal gene transfer and functional diversification of plant cell wall degrading polygalacturonases: Key events in the evolution of herbivory in beetles. Insect Biochem Mol Biol. 2014;52: 33–50. 10.1016/j.ibmb.2014.06.008 [DOI] [PubMed] [Google Scholar]
148.Kirsch R, Wielsch N, Vogel H, Svatoš A, Heckel DG, Pauchet Y. Combining proteomics and transcriptome sequencing to identify active plant-cell-wall-degrading enzymes in a leaf beetle. BMC Genomics. 2012;13: 587 10.1186/1471-2164-13-587 [DOI] [PMC free article] [PubMed] [Google Scholar]
149.Busch A, Danchin EGJ, Pauchet Y. Functional diversification of horizontally acquired glycoside hydrolase family 45 (GH45) proteins in Phytophaga beetles. BMC Evol Biol. 2019;19: 100 10.1186/s12862-019-1429-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
150.Brand P, Lin W, Johnson BR. The draft genome of the invasive walking stick, Medauroidea extradendata, reveals extensive Lineage-Specific gene family expansions of Cell wall degrading enzymes in phasmatodea. G3 Genes, Genomes, Genet. 2018;8: 1403–1408. [DOI] [PMC free article] [PubMed] [Google Scholar]
151.Bronner R. Proprieties lytiques des œufs de Biorrihza pallida. Comptes Rendus l’Académie des Sci. 1973;276: 189–192. [Google Scholar]
152.Bronner R. Contribution a l’etude histochimique des tissus nourriciers des zoocecidies. Marcellia. 1977;40: 1–134. [Google Scholar]
153.Rodriguez PA, Stam R, Warbroek T, Bos JIB. Mp10 and Mp42 from the aphid species Myzus persicae trigger plant defenses in Nicotiana benthamiana through different activities. Mol Plant-Microbe Interact. Am Phytopath Society; 2014;27: 30–39. 10.1094/MPMI-05-13-0156-R [DOI] [PubMed] [Google Scholar]
154.Jaouannet M, Rodriguez PA, Thorpe P, Lenoir CJG, MacLeod R, Escudero-Martinez C, et al. Plant immunity in plant–aphid interactions. Front Plant Sci. 2014;5: 663 10.3389/fpls.2014.00663 [DOI] [PMC free article] [PubMed] [Google Scholar]
155.Hogenhout SA, Bos JIB. Effector proteins that modulate plant–insect interactions. Curr Opin Plant Biol. 2011;14: 422–428. 10.1016/j.pbi.2011.05.003 [DOI] [PubMed] [Google Scholar]
156.Sanchez F, Padilla J E, Perez H, and Lara M. Control of Nodulin Genes in Root-Nodule Development and Metabolism. Annu Rev Plant Physiol Plant Mol Biol. 1991;42: 507–528. 10.1146/annurev.pp.42.060191.002451 [DOI] [Google Scholar]
157.Ma Y, Yan C, Li H, Wu W, Liu Y, Wang Y, et al. Bioinformatics Prediction and Evolution Analysis of Arabinogalactan Proteins in the Plant Kingdom. Front Plant Sci. 2017;8: 66 10.3389/fpls.2017.00066 [DOI] [PMC free article] [PubMed] [Google Scholar]
158.Mashiguchi K, Asami T, Suzuki Y. Genome-Wide Identification, Structure and Expression Studies, and Mutant Collection of 22 Early Nodulin-Like Protein Genes in Arabidopsis. Biosci Biotechnol Biochem. 2009;73: 2452–2459. 10.1271/bbb.90407 [DOI] [PubMed] [Google Scholar]
159.Redding NW, Agudelo P, Wells CE. Multiple Nodulation Genes Are Up-Regulated During Establishment of Reniform Nematode Feeding Sites in Soybean. Phytopathology. Am Phytopath Society; 2018;108: 275–291. 10.1094/PHYTO-04-17-0154-R [DOI] [PubMed] [Google Scholar]
160.Domon J-M, Neutelings G, Roger D, David A, David H. A basic chitinase-like protein secreted by embryogenic tissues of Pinus caribaea acts on arabinogalactan proteins extracted from the same cell lines. J Plant Physiol. 2000;156: 33–39. [Google Scholar]
161.Arakane Y, Muthukrishnan S. Insect chitinase and chitinase-like proteins. Cell Mol Life Sci. 2010;67: 201–216. 10.1007/s00018-009-0161-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
162.Zhu Q, Arakane Y, Beeman RW, Kramer KJ, Muthukrishnan S. Functional specialization among insect chitinase family genes revealed by RNA interference. Proc Natl Acad Sci U S A. 2008;105: 6650–5. 10.1073/pnas.0800739105 [DOI] [PMC free article] [PubMed] [Google Scholar]
163.Taper ML, Case TJ. Interactions between oak tannins and parasite community structure: unexpected benefits of tannins to cynipid gall-wasps. Oecologia. Springer; 1987;71: 254–261. 10.1007/BF00377292 [DOI] [PubMed] [Google Scholar]
164.Taper ML, Zimmerman EM, Case TJ. Sources of mortality for a cynipid gall-wasp (Dryocosmus dubiosus (Hymenoptera: Cynipidae)): the importance of the tannin/fungus interaction. Oecologia. Springer; 1986;68: 437–445. 10.1007/BF01036752 [DOI] [PubMed] [Google Scholar]
165.Wilson D, Carroll GC. Avoidance of high-endophyte space by gall-forming insects. Ecology. Wiley Online Library; 1997;78: 2153–2163. [Google Scholar]
166.Wilson D. Fungal endophytes which invade insect galls: insect pathogens, benign saprophytes, or fungal inquilines? Oecologia. Springer; 1995;103: 255–260. 10.1007/BF00329088 [DOI] [PubMed] [Google Scholar]
167.Tichomiroff A. Chemische Studien über die Entwicklung der Insecteneier. Zeitschrift für Physiol Chemie. 1885;9: 518–532. [Google Scholar]
168.Furneaux PJS, Mackay AL. The composition, structure and formation of the chorion and the vitelline membrane of the insect egg-shell The insect integument. Amsterdam: Elsevier; 1976. pp. 157–176. [Google Scholar]
169.Moreira MF, Dos Santos AS, Marotta HR, Mansur JF, Ramos IB, Machado EA, et al. A chitin-like component in Aedes aegypti eggshells, eggs and ovaries. Insect Biochem Mol Biol. 2007;37: 1249–61. 10.1016/j.ibmb.2007.07.017 [DOI] [PubMed] [Google Scholar]
170.Lohse K, Sharanowski B, Nicholls JA, Blaxter M, Stone GN. Developing EPIC markers for chalcidoid Hymenoptera from EST and genomic data. Mol Ecol Resour. 2011;3: 521–529. [DOI] [PubMed] [Google Scholar]
171.Chen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. Cold Spring Harbor Laboratory; 2018;34: i884–i890. 10.1093/bioinformatics/bty560 [DOI] [PMC free article] [PubMed] [Google Scholar]
172.Andrews S. FastQC: A quality control application for high throughput sequence data. Available at: http://www.bioinformatics.babraham.ac.uk/projects/fastqc. 2010. [Google Scholar]
173.Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, et al. Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotech. 2011;29: 644–652. [DOI] [PMC free article] [PubMed] [Google Scholar]
174.Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C. Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods. Nature Publishing Group; 2017;14: 417–419. 10.1038/nmeth.4197 [DOI] [PMC free article] [PubMed] [Google Scholar]
175.Soneson C, Love MI, Robinson MD. Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Research. Faculty of 1000 Ltd; 2015;4. [DOI] [PMC free article] [PubMed] [Google Scholar]
176.Haas BJ, Papanicolaou A, Yassour M, Grabherr M, Blood PD, Bowden J, et al. De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis. Nat Protoc. 2013;8: 1494–1512. 10.1038/nprot.2013.084 [DOI] [PMC free article] [PubMed] [Google Scholar]
177.Davidson NM, Hawkins ADK, Oshlack A. SuperTranscripts: a data driven reference for analysis and visualisation of transcriptomes. Genome Biol. BioMed Central; 2017;18: 148 10.1186/s13059-017-1284-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
178.Li H. Minimap2: pairwise alignment for nucleotide sequences Bioinformatics. Oxford University Press; 2018;1: 7. [DOI] [PMC free article] [PubMed] [Google Scholar]
179.Plomion C, Aury J-M, Amselem J, Alaeitabar T, Barbe V, Belser C, et al. Decoding the oak genome: public release of sequence data, assembly, annotation and publication strategies. Mol Ecol Resour. Wiley Online Library; 2016;16: 254–265. 10.1111/1755-0998.12425 [DOI] [PubMed] [Google Scholar]
180.Buchfink B, Xie C, Huson DH. Fast and sensitive protein alignment using DIAMOND. Nat Methods. Nature Publishing Group; 2015;12: 59 10.1038/nmeth.3176 [DOI] [PubMed] [Google Scholar]
181.Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. Oxford University Press; 2012;41: D590—D596. 10.1093/nar/gks1219 [DOI] [PMC free article] [PubMed] [Google Scholar]
182.Simão FA, Waterhouse RM, Ioannidis P, Kriventseva E V, Zdobnov EM. BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics. Oxford University Press; 2015;31: 3210–3212. 10.1093/bioinformatics/btv351 [DOI] [PubMed] [Google Scholar]
183.Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-Seq data with DESeq2. Genome Biol. 2014;15: 550 10.1186/s13059-014-0550-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
184.Zhu A, Ibrahim JG, Love MI. Heavy-tailed prior distributions for sequence count data: removing the noise and preserving large differences bioRxiv. Cold Spring Harbor Laboratory; 2018; 303255. [DOI] [PMC free article] [PubMed] [Google Scholar]
185.Stephens M. False discovery rates: a new deal. Biostatistics. Oxford University Press; 2016;18: 275–294. [DOI] [PMC free article] [PubMed] [Google Scholar]
186.Alexa A, Rahnenfuhrer J. topGO: Enrichment Analysis for Gene Ontology. Available at: https://bioconductor.org/packages/release/bioc/html/topGO.html. 2016. 10.3390/ijms17030353 [DOI] [Google Scholar]
187.Supek F, Bošnjak M, Škunca N, Šmuc T. REVIGO summarizes and visualizes long lists of gene ontology terms. PLoS One. 2011;6: e21800 10.1371/journal.pone.0021800 [DOI] [PMC free article] [PubMed] [Google Scholar]
188.Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, et al. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol. 2012;19: 455–77. 10.1089/cmb.2012.0021 [DOI] [PMC free article] [PubMed] [Google Scholar]
189.Zimin A V, Marçais G, Puiu D, Roberts M, Salzberg SL, Yorke JA. The MaSuRCA genome assembler. Bioinformatics. 2013;29: 2669–77. 10.1093/bioinformatics/btt476 [DOI] [PMC free article] [PubMed] [Google Scholar]
190.Smit AFA, Hubley R, Green P. RepeatMasker Open-3.0. 1996–2010.
191.Kumar S, Jones M, Koutsovoulos G, Clarke M, Blaxter M. Blobology: exploring raw genome data for contaminants, symbionts and parasites using taxon-annotated GC-coverage plots. Front Genet. Frontiers; 2013;4: 237 10.3389/fgene.2013.00237 [DOI] [PMC free article] [PubMed] [Google Scholar]
192.Emms DM, Kelly S. OrthoFinder: solving fundamental biases in whole genome comparisons dramatically improves orthogroup inference accuracy. Genome Biol. BioMed Central; 2015;16: 157 10.1186/s13059-015-0721-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS Genet. doi: 10.1371/journal.pgen.1008398.r001

Decision Letter 0

Gregory P Copenhaver

30 Sep 2019

* Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. *

Dear Dr Hearn,

Thank you very much for submitting your Research Article entitled 'Genomic dissection of an extended phenotype: oak galling by a cynipid gall wasp' to PLOS Genetics. Your manuscript was fully evaluated at the editorial level and by two of the independent peer reviewers who evaluated the prior related submission. The reviewers appreciated the attention to an important topic but identified some aspects of the manuscript that should be improved.

Both reviewers were complementary of the effort extended to address the prior critiques. Reviewer #1's comments are primarily focused on the organization of the manuscript and providing the proper context so that readers will be able to understand the importance of the work. Reviewer #1 offers a lengthy set of suggestions for how to improve the manuscript along those lines. While many (if not all) of the reviewers comments are potentially helpful, we also understand that you will want the manuscript to reflect your voice and your priorities as authors. As such, we strongly encourage you to improve the contextualization and logical flow, use standard terminology, and pay very close attention to citing the relevant important literature, but we will leave it up to you to decide how to accomplish that. Reviewer #2's comments can also be addressed by textual/figure revision. In particular we ask that you pay close attention to Reviewer #2's comments regarding figure clarity, providing sufficient detail in the methods descriptions, and accurately describing the biology of cytokinin production in the system. The editors anticipate that addressing these concerns will require a significant overhaul of the text and figures, but will not require any additional experimentation.

We therefore ask you to modify the manuscript according to the review recommendations before we can consider your manuscript for acceptance. Your revisions should address the specific points made by each reviewer.

In addition we ask that you:

1) Provide a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript.

2) Upload a Striking Image with a corresponding caption to accompany your manuscript if one is available (either a new image or an existing one from within your manuscript). If this image is judged to be suitable, it may be featured on our website. Images should ideally be high resolution, eye-catching, single panel square images. For examples, please browse our archive. If your image is from someone other than yourself, please ensure that the artist has read and agreed to the terms and conditions of the Creative Commons Attribution License. Note: we cannot publish copyrighted images.

We hope to receive your revised manuscript within the next 30 days. If you anticipate any delay in its return, we would ask you to let us know the expected resubmission date by email to plosgenetics@plos.org.

If present, accompanying reviewer attachments should be included with this email; please notify the journal office if any appear to be missing. They will also be available for download from the link below. You can use this link to log into the system when you are ready to submit a revised version, having first consulted our Submission Checklist.

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

Please be aware that our data availability policy requires that all numerical data underlying graphs or summary statistics are included with the submission, and you will need to provide this upon resubmission if not already present. In addition, we do not permit the inclusion of phrases such as "data not shown" or "unpublished results" in manuscripts. All points should be backed up by data provided with the submission.

PLOS has incorporated Similarity Check, powered by iThenticate, into its journal-wide submission system in order to screen submitted content for originality before publication. Each PLOS journal undertakes screening on a proportion of submitted articles. You will be contacted if needed following the screening process.

To resubmit, you will need to go to the link below and 'Revise Submission' in the 'Submissions Needing Revision' folder.

[LINK]

Please let us know if you have any questions while making these revisions.

Yours sincerely,

Gregory P. Copenhaver

Editor-in-Chief

PLOS Genetics

Gregory Barsh

Editor-in-Chief

PLOS Genetics

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: This is a newly-submitted manuscript. I reviewed an earlier version, which was also reviewed by two scientists who clearly know much more than I do about genetics, genomics, and molecular techniques. Both reviewers stated that conclusions from the study would be strengthened by the addition of control data showing how oak tissues targeted for gall induction by the insect Biorhiza pallida develop in the absence of the insect. These data have been added in this version, which is a new submission. (I did wonder why the authors changed in the new submission. One former author is now missing?)

The new submission begins with author responses to comments of the three reviewers. My comments were adequately addressed by the authors in this new version.

In hopes that the same two reviewers will be reviewing this new version I leave it to them to comment on the possible limits of this experimental design, which as I understand was to add the requested control plant developmental treatments after having completed the plant-insect treatments earlier. If this is the case, a better experimental design would have been to collect control plant material at the same time as the insect-infested test plant material. I may have misunderstood the design the authors used for this new version.

I will limit my comments to more general topics, specifically revision of the Introduction to make a better argument for their study system. The Abstract similarly needs improvement but I leave it to the authors to do that. At the end of my review, I list specific points that can easily be addressed. In many cases the authors use terminology that is not in common use, being specific to people who study galls.

This is the problem with the Introduction: it fails to make strong arguments for: 1) galls as an interesting biotic interaction in which the host alters its development to accommodate the needs of a symbiont, 2) oak trees as an interesting case of plants that are proficient in making galls, and 3) oak gall wasps as a particularly interesting group of gall inducers. The Introduction should carefully build these arguments. It also needs to make clear what the overall endeavor is and what this particular study achieves for the endeavor.

Many of the comments of the two reviewers seemed to be about the fact that many of the aspects of the interaction were not resolved. The Introduction needs to make clear how complicated the interaction is, not only having many moving parts at any one point in time but also having all these moving parts changing over as the gall and insect develop and mature. This study is an entry point for the endeavor. That should be clear in the Introduction, as should questions that are resolved by this study.

Another problem for the Introduction is the confusion created by presenting a grab-bag of ideas. In many cases, instead of spending some time to explain each idea so that the reader can see its relevance for the experimental work, the idea is simply mentioned and then another idea pops up. One idea the authors seem especially fond of is "extended phenotype". If this is their concept of galls it must be explained. I find its application to plant galls facile or even misleading. The plant makes the gall. It is part of its own phenotype. Moreover, while the concept is perhaps cute for galls made for plant parasites it does not seem to apply for galls that plants make for mutualists like nitrogen-fixing bacteria.

Another idea that is mentioned but not explained is the assertion that understanding galls offer a unique perspective on normal growth and development. Either don't mention this or explain it better. Professor Minelli, an animal developmental biologist, recently turned his attention to plant development biology wrote a book in which he claims that plant galls deserve far more attention as interesting developmental phenomena, independently from the organisms associated with their creation. Minelli pointed to two aspects of plant development that were understudied. Galls were one of them. For me, this idea fits in better with the experimental work in this study.

In terms of the Introduction I found unsatisfying the hypotheses about what galls represent to the plant. Why not just mention these hypotheses in the Discussion? My sense from the manuscript is that the authors know more about insects than plants. I am an insect person rather than a plant person so I cannot tell if the hypotheses they raise about what the gall is from the plant's perspective make sense to a whole organism plant person. Not being a plant person, I wanted an explanation of why plants normally do the two things erected as hypotheses: 1) produce ectopic storage organs and 2) produce somatic embryos. Are these two functions entirely separate or are they simply things that can be identified because of a particular convenient molecular signature? I looked up somatic embryogenesis and it said it is an artificial process in plants not a natural process? And it is not clear how the ENOD system relates to one or both of these plant functions? Or does ENOD only relate to somatic embryos?

Below is an outline of how I would revise the Introduction. Galls are not a subject known by most scientists. Readers who know a little about them see galls as oddities having little relevance for big questions in plant development or host-parasite interactions. An argument must be made.

Right now, the argument seems to be "we have perfect understanding of galls plants make for bacteria but, oh yes, plant make galls for insects too and so let's investigate them too but without referencing what is known about how plants make galls for bacteria." That seems weak to me. As a reader I think: why bother with insect galls? So argue: Insects offer things that bacteria (and fungi) do not. What they offer needs to be made clear. Otherwise readers will think understanding galls made for bacteria and fungi is enough.

Paragraph One: Introduce galls and explain why they are interesting. Make it clear that plants make galls. Say plants make galls in response to many types of foreigners- list them (Redfern 2011 does this). Say plants sometimes make galls for mutualists - explain N-fixing bacteria and fig wasps. Say plants mostly make galls for foreigners that seemingly provide no benefits. We know that many of these no-benefit-for plants foreigners actually harm the plant because, across the vast phylogenetic range of gall inducers, there are many organisms considered to be pests, of agriculture, forestry, and landscapes. The idea is that the foreigner induces a sink at the colonization site and the plant makes a gall there. Resources meant for other plant processes are diverted to the sink. The gall can replace flowers or seeds, thus harming reproduction. Structural problems created at the colonization site can harm plant survival.

Paragraph Two: Use next paragraph to introduce a simple framework and terms that are relevant for your study. Do this by introducing reader to what is known about the galls that are best understood in terms of genetics, including nodules made for mutualist N-fixing bacteria of two types, crown gall made for parasitic Agrobacterium bacteria, and maize smut made for a parasitic fungus Ustilago maydis. Cite major references on these interactions (there are many reviews in Annual Reviews and Nature Reviews). Three important features are 1) release of plant molecules recognized by the foreigner, 2) release of foreigner molecules recognized by the plant, and 3) subsequent changes in plant cellular growth and development that lead to the creation of new accommodations, e.g., a gall, for the foreigner. In the case of the parasites, there are molecules intentionally made in the body of the foreigner and secreted onto or into plant cells that effect a change in host cells that benefits the foreigner. These molecules are called 'effectors'. They are encoded by 'effector genes', candidates of which can be found in the foreigner's genome. Plant molecules that act to recognize foreigner-produced molecules are 'receptors' and are encoded by 'receptor genes', candidates of which can be found in the plant's genome. For both foreigners acting as mutualists or parasites, the plant has growth and developmental traits involved in accommodating the foreigner. They are encoded by various genes found in the plant genome. If the foreigner has the ability to harm the plant, the plant may also have traits that allow it to resist colonization. These are called 'resistance traits' and they are encoded by 'resistance gene', candidates of which can be found in the plant's genome.

Paragraph Three: Explain that, with the exception of root knot and cyst nematodes, understanding of genetics and plant galling responses has lagged behind for insects. And yet plant galls cannot be understood without understanding galls induced for insects. Now make the argument. For example: vast majority of galls made by plants are made for insects, co-evolution and co-speciation, galls made for insects are more elaborate in terms of tissue organization, galls are more species- specific and even sex- or generation- specific, insects interact with plants in different ways to induce galls - stylets, egg-laying females, etc (doubtless the authors are aware of many other reasons, stress ecology and especially evolution). Also, it is interesting that two insect orders with greatest numbers of plant-feeding species have so few species that induce galls.

Paragraph Four: Make the argument for oaks being an interesting plant for galls. Explain the variety of galls oaks make for diverse gall inducers. Is oak a special target of certain types of gall inducers? Explain that genome of oaks was recently published (Plomian et al. 2018- why is this paper not discussed?) What do we learn from oak genome? It has lots of Resistance genes. How does genome make endeavor at hand easier?

Paragraph Five: Make the argument for oak gall wasps (not more generally for cynipids). Explain evolution from parasitoids. Explain relationship between galls and diversification. Explain lineages with and without gall inducers. Make the case for this being an evolutionarily interesting group for plant galls. Explain concept of phylogenetic signal and its relevance for this group of insects (see review in Ann Review Phytopathology). Is this the case for nematodes? I doubt it.

Paragraph Six: Explain exactly how the insect interacts with the plant. Hostfinding. Oviposition and how it sets up idea that female injects something into plant that is first to induce the gall. Do not forget to talk about physical wounding. What happens in plant if female wounds and egg is laid and larva is killed- nothing or the beginnings of the gall? gall made specifically for your gall wasp. Explain you are looking at sexual generation but also explain how this differs for asexual generations. Explain why it is difficult in this system to catch what happens during the very first moments of the larval- plant interaction. Explain if all those larvae developing in the gall are offspring of a single female. Explain how synchronized their development is. Give a figure (a drawing would be best) that shows side by side the development of the larva and the gall (you do not have to show all the larvae). Why does the larva stay so small for so long as the gall first develops?

Paragraph Seven: Explain general methods and objectives for this study. Keep it simple. If you want to get complicated, do it in the Discussion not the Introduction.

I final comment is that often the references are not good choices. In many cases an Annual Review article would be more appropriate than a single research article. Also all the major books on galls should be cited- Redfern, Williams (edited volume) etc.

Simple suggestions:

General: it is often not clear if the term "gall wasp" refers to the species (adults and larvae), the adults, or can also refer to larvae. Specify wasp adult, egg, larva and pupa.

Line 87: define gall induction

Line 108: How many eggs does female lay in this one location and how long does this take? Are all cohabitants of gall siblings? What is known about glands associated with ovipositor in this gall wasp species versus relatives that are not gall wasps?

Lines 108- 131: wounding is not discussed along with many other details relevant to this research (see comments above)

Line 109: "as the egg hatches…" timing of insect versus plant events is not clear- what happens if larvae dies immediately after eclosion?

Line 118: what is size of newly hatched larva relative to larva in Early Stage Galls?

Line 114: explain sexual versus asexual generation

Line 116: not clear - you chose this wasp species because there are lots of larvae inside a single gall?

Line 120: is this usual for early stage larvae to grow very little while the gall itself grows? When do larvae have their growth spurt- when nutritive tissue is first made?

Lines 108-131: is anything known about internal anatomy of first instar larvae, in particular size of salivary glands?

Line 133- 136: It would be nice to have a drawing that shows along one timeline the growth of the larva, the growth of the gall, and the growth of ungalled tissue.

Line 149: the questions could be greatly simplified, 2-3 sentences for each one.

Line 163: "exploited" is a strange word to use for rhizobia which work together with the plant?

Line 170: background for this question should be introduced earlier when explaining oviposition and larval interactions. For each mention how microbial symbionts could be involved?

Line 175: "effector" should be introduced almost at the beginning of the Introduction.

Line 180: not clear who receives the benefit of the microbial symbiont manipulating the plant? All this seems rather vague. Strip it down to basics and discuss in more detail in Discussion.

Line 210: CLAVATA system is mentioned several times in Introduction but we are not told what its relevance is. Why not keep these details for the Discussion?

Line 242: parasitoids were present in the gall? The impact of this on the plant-wasp interaction is not discussed. This should be mentioned in the Introduction. Also gall tissue showed evidence of viral and fungal infections while ungalled tissue did not (Line 676). What does this mean?

Line 296: meaning of term "cecidome" is not explained. Do not use terms like this. Also Line 660.

Line 655: is there a cost to the oak tree of making galls for gall wasps? Discuss the evidence.

Line 716: conclusion that gall is a "novel amalgam" is not really explained. Are the two hypotheses being abandoned?

Line 782: this info should appear in the Introduction

Line 782: in what way are they similar- give an example. Is "many" an exaggeration?

Line 784: this info should be given in the Introduction.

Line 786: "suggesting that the wasp is the source of the galling stimuli" - imprecise use of word "wasp" - be specific about life stage you are talking about.

Line 789: relevance of leaf miners to galls is not clear.

Line 791: section should start with this sentence "We found no support…" the earlier part of this section having been given in the Introduction.

Line 813: effectors should be introduced in Introduction. In Hessian fly, because of gene-for-gene interactions we know the molecules identified as effectors play a role in interactions with plants -read Aggarwal et al. 2014. Genome paper on Hessian fly also provides evidence Zhao et al. 2015.

Line 824: this info about larvae having enlarged salivary glands should be presented in Introduction when effectors are first discussed. Note that Hessian fly larvae also have enlarged salivary glands and much research has been done showing many salivary genes are strongly unregulated during first days of plant interactions. Citing of literature on Hessian fly is spotty.

Line 829: statement "or more generally any genes from other gall-inducers" should have support with references.

Line 841: "On the basis of these data…" this sentence should be the first sentence of the section not the last.

Line 842: I think you mean saliva of gall wasp larvae?

Line 909-915: Discussion of function of PCWDEs seems pretty vague. How about aiding in effectors entering plant cells? What do cytological studies by Bronner indicate is the likely role of these enzymes? Bronner did excellent work and it should not be ignored. Creating a solid story for function of PCWDEs is important if you are going to go on to claim that a PCWDE repertoire is a synapomorphy (please define this term) of gall-inducing cynipids (Line 916)

Line 937-947: you do not really explain the role of ENODs in nodules made for rhizobia and how it might be relevant for gall wasps.

Line 982: Conclusions - Abstract should be more like this - straightforward in terms of what experimental work was done, what were major conclusions, and where the research will go in the future.

Line 996- 1000: this info should be in the Introduction. The reader is left in the dark about so many aspects of this plant-insect system.

Reviewer #2: Summary: This refined manuscript represents a solid omic dissection of gall development by cynipid wasps. The authors have substantially addressed prior reviews to refine the prose, identify specific hypotheses, and correlate (with allowable speculation) changes in plant and insect gene expression at different developmental stages. There remain a few limitations to this study, of which some are explained (not hidden) and thus justifiable, and others require some polishing. There do not appear any lingering major limitations or concerns.

Below are specific comments per section or questions to address:

Response to reviewers.

Overall, the authors did a thorough analysis of the review comments and incorporated changes that should suffice for each reviewer. Having been one of those reviewers (#2), I feel the authors really listened to the criticism and produced an enhanced analysis that reduces speculation and confusion by better correlating their data or revealing new patterns. They explicitly describe the hypotheses being tested, the limits of their data, and how these results advance the literature.

abstract.58. if the comparison includes galling, inquiline, and non galling (non herbivore) species, could the authors extrapolate that PCWDEs are one path to herbivory, given several other herbivores evolved these. Hessian fly, another gall did not have these. The fig wasp does not appear to either. Thus maybe these enzymes are not key to galling, unless you delineate it as key to cynipid wasp galling

Wybouw et al 2016 GBE discusses PCWDE from herbivores if it helps

author.summary.66. this seems a strong statement, assuming the interaction must either negatively affect the host or benefit both. There are many cases where gall formation does not induce fitness costs, so while the insect may live on the plant, reorganizing sink and other signaling may enhance resource acquisition. Perhaps the authors would change this to "The organisms involved range from mutualistic to parasitic" to better encompass the gradient of host interactions.

intro.81. "interactions ranging from mutualist to parasitic"

99. "gall communities contribute significantly to biodiversity" of what? of species in general? of gall species? this is unclear.

108. what is a generation specific tissue? does this mean the generation of the insect or plant?

136. thus, cynipids are maintaining a meristem or delaying its maturity. It is interesting to think about the mechanism for this.

178. “some gall-inducing insects have been shown to produce endogenous auxins and/or cytokinins (see below).”

208. “Endogenous production of plant hormones (such as auxin and cytokinins) has been shown in a range of galling and non-galling herbivorous insects [],” In the first sentence the authors indicate insects produce these compounds (but do not differentiate if their co-bionts are responsible). In the later sentence the presence of these compounds is indicated (not the production) suggesting a co-biont may be responsible (even if this isn’t stated). The authors would be best served by removing the misleading intimation in the first sentence that insects synthesize cyotkinins until data show up that substantiate this claim.

This point was raised in the review and the authors claimed to have corrected it in the response to reviewers. Indeed, the discussion reads as if the hypothesis that insects synthesize cytokinin remains possible yet unvalidated, although evidence thus far provides no support, questioning if it should still remain a viable hypothesis. In the intro, I would expect a similar statement, and not one that purports that some galling insects produce cytokinins. This paper would benefit (and be more cited) if they recognized this throughout all sections.

226. what is the evolutionary age of the split? for insect feeding parasitoids (highly specialized) and plant feeders (highly specialized) the divergence may be so long that rapidly evolving genes underlying a potentially antagonistic interaction (assuming the most negative of parasitism) diverged quickly so signatures or orthology are so low they may be missed. Thus, some caution should be used in this comparison.

results.

table.3. can the numbers be parsed into up(down) or something that incorporates more context?

b. what is a "number of DE genes that were not DE between any stages in normal bud development" could this be "unique DE genes in galls not found in DE of normal bud development"

c. this is confusing. is this table the same as saying 1(26) up(down) in S0 vs S1? if so, using this notation would decrease the table size. It seems intuitive if in a pairwise comparison you have up and down genes, then the higher expression will be based on the direction of the comparison where up are genes up in 0 vs 1 and down are genes up in 1 vs 0.

416 may be constitutively expressed. did you check counts?

462 if no JA response, how antagonistic is this interaction? This is in reference to what level of negative response defines a parasite vs a commensal.

488 how was the microbial profiling done?

524. yes constitutive expression may prevent detection.

570-572. this is not surprising given the lineage specificity of genes involved in plant manipulation (eg., Boulain et al 2018)

table.4 can you put a dated phylogeny on the edge of the table so we know how different these insects are?

606. what about the inquiline? in table 5 the non galler retains PCWDEs indicating it may also be important for herbivory but not galling

table.5 can this be in the same order as table 4.

830. yes see boulain reference above

874. yes this is how to discuss cytokinins. why is the intro set up differently?

924. maybe the step was not direct to galling but to herbivory.

methods.

where is the analysis for detecting microbial sequences? STable 13 suggests you only blasted your assembly against Wolbachia, but did not survey for other microbes.

fig.2. This could be clearer, 1) perhaps with reduced text and better delineation of which organism’s genes you are making the prediction on. 2) seems like export through active secretion could apply to any organism (insect or microbe) living in the gall and not just the insect. 3) hypotheses are not true, just supported, so perhaps reword this in the caption. 4) Perhaps indicate A represents the expected gene expression patterns if the induce phenotype results in an ectopic storage organ or a modified somatic embryo. 5) counting up now I expect 3 to be similar in concept to 2, but it is not because 1&2 represent the expectations of the plant transcriptome and 3-5 represent what genes are expressed that are not of plant origin. I don't think the number should be continuous between letters. 6) Perhaps change the title from a question to something that captures all the boxes are transcriptome patterns, ie “Expected (or hypothetical) gene expression patterns for galls and associated gallers during development” 7) export is repeated a lot so maybe use an arrow? 8) why are processes different from stimuli – it seems a secretory protein, enzyme, or rna could easily result in the alteration of gene expression of processes. How these are different is unclear.

fig.4. This is not clear. I gather that 747 genes are up in early vs growth and 2310 are up in S4 vs S5 of buds, but what is not clear is the 311 that increased in both gall and normal buds. Are you saying that 311 of the same genes were up in the first and latter comparison I just described (as a venn diagram would indicate). This would suggest 311 belong in the set of 747 and of 2310. If you wanted to make more contrast between the genes that are unique you could label circles as “unique genes” found DE for the specific comparison and boxes as “shared” genes. This would result in a circle of 436 unique to early vs growth stage galls and 1999 unique to normal bud S4 vs S5. I think identifying what is unique is a stronger result than combining and having the reader subtract to get the number. Total gene numbers is more insightful than percents, and in an already complex figure the fewer the numbers the better. You could also present this in a more standard venn diagram way with overlapping circles. If my interpretation of this graph was wrong, then you may need to reconsider it completely, because it is very complex.

fig.5. is this to be oriented differently in the manuscript? Fig4 is aligned well.

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the PLOS Genetics data availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

Reviewer #1: Yes

Reviewer #2: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

PLoS Genet. doi: 10.1371/journal.pgen.1008398.r002

Decision Letter 1

Gregory P Copenhaver

23 Oct 2019

Dear Dr Hearn,

We are pleased to inform you that your manuscript entitled "Genomic dissection of an extended phenotype: oak galling by a cynipid gall wasp" has been editorially accepted for publication in PLOS Genetics. Congratulations!

Before your submission can be formally accepted and sent to production you will need to complete our formatting changes, which you will receive in a follow up email. Please be aware that it may take several days for you to receive this email; during this time no action is required by you. Please note: the accept date on your published article will reflect the date of this provisional accept, but your manuscript will not be scheduled for publication until the required changes have been made.

Once your paper is formally accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you’ve already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at plosgenetics@plos.org.

In the meantime, please log into Editorial Manager at https://www.editorialmanager.com/pgenetics/, click the "Update My Information" link at the top of the page, and update your user information to ensure an efficient production and billing process. Note that PLOS requires an ORCID iD for all corresponding authors. Therefore, please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager.

If you have a press-related query, or would like to know about one way to make your underlying data available (as you will be aware, this is required for publication), please see the end of this email. If your institution or institutions have a press office, please notify them about your upcoming article at this point, to enable them to help maximise its impact. Inform journal staff as soon as possible if you are preparing a press release for your article and need a publication date.

Thank you again for supporting open-access publishing; we are looking forward to publishing your work in PLOS Genetics!

Yours sincerely,

Gregory P. Copenhaver

Editor-in-Chief

PLOS Genetics

Gregory Barsh

Editor-in-Chief

PLOS Genetics

www.plosgenetics.org

Twitter: @PLOSGenetics

----------------------------------------------------

Comments from the reviewers (if applicable):

----------------------------------------------------

Data Deposition

If you have submitted a Research Article or Front Matter that has associated data that are not suitable for deposition in a subject-specific public repository (such as GenBank or ArrayExpress), one way to make that data available is to deposit it in the Dryad Digital Repository. As you may recall, we ask all authors to agree to make data available; this is one way to achieve that. A full list of recommended repositories can be found on our website.

The following link will take you to the Dryad record for your article, so you won't have to re‐enter its bibliographic information, and can upload your files directly:

http://datadryad.org/submit?journalID=pgenetics&manu=PGENETICS-D-19-01463R1

More information about depositing data in Dryad is available at http://www.datadryad.org/depositing. If you experience any difficulties in submitting your data, please contact help@datadryad.org for support.

Additionally, please be aware that our data availability policy requires that all numerical data underlying display items are included with the submission, and you will need to provide this before we can formally accept your manuscript, if not already present.

----------------------------------------------------

Press Queries

If you or your institution will be preparing press materials for this manuscript, or if you need to know your paper's publication date for media purposes, please inform the journal staff as soon as possible so that your submission can be scheduled accordingly. Your manuscript will remain under a strict press embargo until the publication date and time. This means an early version of your manuscript will not be published ahead of your final version. PLOS Genetics may also choose to issue a press release for your article. If there's anything the journal should know or you'd like more information, please get in touch via plosgenetics@plos.org.

PLoS Genet. doi: 10.1371/journal.pgen.1008398.r003

Acceptance letter

Gregory P Copenhaver

28 Oct 2019

PGENETICS-D-19-01463R1

Genomic dissection of an extended phenotype: oak galling by a cynipid gall wasp

Dear Dr Hearn,

We are pleased to inform you that your manuscript entitled "Genomic dissection of an extended phenotype: oak galling by a cynipid gall wasp" has been formally accepted for publication in PLOS Genetics! Your manuscript is now with our production department and you will be notified of the publication date in due course.

The corresponding author will soon be receiving a typeset proof for review, to ensure errors have not been introduced during production. Please review the PDF proof of your manuscript carefully, as this is the last chance to correct any errors. Please note that major changes, or those which affect the scientific understanding of the work, will likely cause delays to the publication date of your manuscript.

Soon after your final files are uploaded, unless you have opted out or your manuscript is a front-matter piece, the early version of your manuscript will be published online. The date of the early version will be your article's publication date. The final article will be published to the same URL, and all versions of the paper will be accessible to readers.

Thank you again for supporting PLOS Genetics and open-access publishing. We are looking forward to publishing your work!

With kind regards,

Matt Lyles

PLOS Genetics

On behalf of:

The PLOS Genetics Team

Carlyle House, Carlyle Road, Cambridge CB4 3DN | United Kingdom

plosgenetics@plos.org | +44 (0) 1223-442823

plosgenetics.org | Twitter: @PLOSGenetics

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 File. R scripts used for testing differential expression and GO term enrichment in all comparisons.

(TXT)

Click here for additional data file.^{(14.8KB, txt)}

S2 File. Trinotate pipeline annotation file in excel format.

Annotations for transcripts included in the differential analysis experiments created with the Trinotate pipeline. File contains transcripts TRINITY_DN0_c0_g1 to TRINITY_DN63911_c3_g1.

(CSV)

Click here for additional data file.^{(64.1MB, csv)}

S3 File. Trinotate pipeline annotation file in excel format.

Annotations for transcripts included in the differential analysis experiments created with the Trinotate pipeline. File contains transcripts TRINITY_DN63911_c3_g1 to TRINITY_DN76618_c5_g1.

(CSV)

Click here for additional data file.^{(76.1MB, csv)}

S4 File. Trinotate pipeline annotation file in excel format.

Annotations for transcripts included in the differential analysis experiments created with the Trinotate pipeline. File contains transcripts TRINITY_DN76618_c5_g1 to TRINITY_DN236370_c0_g1.

(CSV)

Click here for additional data file.^{(72.9MB, csv)}

S5 File. The Salmon read-count matrix for the gall wasp (B. pallida) data used for differential expression analysis in DESeq2.

Each column represents the count for a replicate and each row is a TRINITY output component group, taken as a proxy for a candidate gene.

(CSV)

Click here for additional data file.^{(1.3MB, csv)}

S6 File. The Salmon read count matrix for the oak gall tissue (Quercus robur) data used for differential expression analysis in DESeq2.

Each column represents the count for a replicate and each row is a TRINITY output component group, taken as a proxy for a candidate gene.

(CSV)

Click here for additional data file.^{(3.4MB, csv)}

S7 File. The Salmon read count matrix for the normal oak bud (Quercus robur) data used for differential expression analysis in DESeq2.

Each column represents the count for a replicate and each row is a TRINITY output component group, taken as a proxy for a candidate gene.

(CSV)

Click here for additional data file.^{(5.2MB, csv)}

S8 File. DESeq2 results for gall wasp genes showing significant differential expression between all gall developmental stages.

(CSV)

Click here for additional data file.^{(253.8KB, csv)}

S9 File. DESeq2 results for oak genes showing significant differential expression between all gall developmental stages.

(CSV)

Click here for additional data file.^{(309KB, csv)}

S10 File. DESeq2 results for oak genes showing significant differential expression between stages in normal bud development.

(CSV)

Click here for additional data file.^{(878.4KB, csv)}

S11 File. Gene names for all differentially expressed genes overlapping in S4 Table.

Column headers indicate which cell value the gene list corresponds to in S4 Table reproduced here. Annotations for these genes are provided in S2–S4 Files.

(XLSX)

Click here for additional data file.^{(40.1KB, xlsx)}

S12 File. Gene names for all differentially expressed genes overlapping in S5 Table.

Column headers indicate which cell value the gene list corresponds to in S5 Table reproduced here. Annotations for these genes are provided in S2–S4 Files.

(XLSX)

Click here for additional data file.^{(32.6KB, xlsx)}

S1 Fig. Images of the four replicates of each sexual generation Biorhiza pallida gall stage sampled in the RNAseq experiment.

(EPS)

Click here for additional data file.^{(15.4MB, eps)}

S2 Fig. Principle component analysis in DESeq2 of oak bud developmental stages S1-S4.

(EPS)

Click here for additional data file.^{(58.3KB, eps)}

S3 Fig. REVIGO treemaps of enriched biological process GO terms for oak genes expressed during development of galled and normal oak bud tissues.

These maps were used to construct Fig 5 by providing higher-order groupings of GO terms. Parts A, B, C and D correspond to A, B, C and D in Fig 5.

(EPS)

Click here for additional data file.^{(277.5KB, eps)}

S1 Table. Taxonomizr classified transcripts of viral origin.

Transcripts were classified as far as the data allow through taxonomic levels from order to species during the annotation process.

(XLSX)

Click here for additional data file.^{(41.1KB, xlsx)}

S2 Table. Read alignment totals for all gall-derived reads.

(XLSX)

Click here for additional data file.^{(9.8KB, xlsx)}

S3 Table. Read alignment totals for oak (Quercus robur) normal bud stage replicates.

(XLSX)

Click here for additional data file.^{(9.8KB, xlsx)}

S4 Table. Counts of oak (Quercus robur) genes and GO terms that are differentially expressed in both gall and normal bud tissues.

(XLSX)

Click here for additional data file.^{(12.2KB, xlsx)}

S5 Table. Counts of oak (Quercus robur) genes and GO terms that are only differentially expressed during gall development.

(XLSX)

Click here for additional data file.^{(9.5KB, xlsx)}

S6 Table. Normalised counts for specific genes in gall wasp (Biorhiza pallida) larvae and oak (Quercus robur) gall tissues after differential expression analysis.

The table shows transcript counts across gall stage replicates for specific genes discussed in the text.

(XLSX)

Click here for additional data file.^{(13.9KB, xlsx)}

S7 Table. Oak (Quercus robur) biological process (BP) GO terms that were differentially expressed in gall tissues but not during normal bud development.

(XLSX)

Click here for additional data file.^{(18.7KB, xlsx)}

S8 Table. Oak (Quercus robur) molecular function (MF) GO terms that were differentially expressed in gall tissues but not during normal bud development.

(XLSX)

Click here for additional data file.^{(15.8KB, xlsx)}

S9 Table. Oak (Quercus robur) biological process (BP) GO terms that were differentially expressed both in gall tissues and during normal bud development.

(XLSX)

Click here for additional data file.^{(61.5KB, xlsx)}

S10 Table. Oak (Quercus robur) molecular function (MF) GO terms that were differentially expressed both in gall tissues and during normal bud development.

(XLSX)

Click here for additional data file.^{(15.7KB, xlsx)}

S11 Table. Oak (Quercus robur) biological process (BP) GO terms that were differentially expressed between stages in normal bud development.

(XLSX)

Click here for additional data file.^{(38.9KB, xlsx)}

S12 Table. Oak (Quercus robur) molecular function (MF) GO terms that were differentially expressed between stages in normal bud development.

Terms enriched in each comparison and each direction, either up- or down-regulated are included. Annotated = number of genes annotated with that GO term in the complete dataset, Significant = observed number of significantly differentially genes with that annotation, Expected = the expected number of differentially expressed genes with that annotation, P-value = test p-value. We used a significance threshold value of p≤ 0.01.

(XLSX)

Click here for additional data file.^{(29.4KB, xlsx)}

S13 Table. Nucleotide blast results of Biorhiza pallida and Diplolepis spinosa genomes against Wolbachia genomes.

(XLSX)

Click here for additional data file.^{(114.6KB, xlsx)}

S14 Table. Gall wasp (Biorhiza pallida) biological process (BP) GO terms that were differentially expressed between stages in gall development.

(XLSX)

Click here for additional data file.^{(14KB, xlsx)}

S15 Table. Biorhiza pallida molecular function (MF) GO terms.

Terms enriched in each comparison and each direction, either up- or down-regulated are included. Annotated = genes annotated with that GO term in the complete dataset, Significant = observed number of significantly differentially expressed genes with that annotation, Expected = the expected number of genes with that annotation. We used a significance threshold value of p≤ 0.01.

(XLSX)

Click here for additional data file.^{(13.3KB, xlsx)}

S16 Table. Intron sequences identified in Cynipid PCWDE genes.

(XLSX)

Click here for additional data file.^{(9.6KB, xlsx)}

S17 Table. OrthoFinder orthogroups of genes more highly-expressed in Early gall larvae with at least one orthologue in another Cynipid species.

(XLSX)

Click here for additional data file.^{(18.1KB, xlsx)}

S18 Table. Plant Cell Wall Degrading Enzyme Trinotate annotations.

Annotations are given for all transcripts/isoforms that were annotated as a cellulase, pectin/pectate lyase or rhamnogalacturonan lyase.

(XLSX)

Click here for additional data file.^{(17.1KB, xlsx)}

S19 Table. Nucleotide versus protein blast results of contigs encoding PCWDEs of bacterial origin or most homologous to PCWDEs present in D. ponderosae.

(XLSX)

Click here for additional data file.^{(46.3KB, xlsx)}

S20 Table. Combined Illumina read statistics for raw and filtered data for each transcriptome sample.

(XLSX)

Click here for additional data file.^{(10.8KB, xlsx)}

S21 Table. Comparison of cynipid venom gland and larval expressed cellulases, showing these to be distinct.

(XLSX)

Click here for additional data file.^{(11.9KB, xlsx)}

S22 Table. Mean expression of uniquely DE genes in gall tissue contrasts for gall- and bud tissues.

(XLSX)

Click here for additional data file.^{(107.5KB, xlsx)}

Data Availability Statement

Raw Illumina Hi-Seq data have been deposited in the European Nucleotide Archive under BioProjects PRJEB13357, PRJEB13424, and PRJEB32849.

[pgen.1008398.ref001] 1.Meyer J, Maresquelle HJ. Anatomie des galles. Berlin: Gebrüder Borntraeger; 1983. [Google Scholar]

[pgen.1008398.ref002] 2.Rohfritsch O, Shorthouse JD, Braun AC, Kahl G, Schell JS. Insect galls. Academic Press; 1982. [Google Scholar]

[pgen.1008398.ref003] 3.Dorchin N, Cramer MD, Hoffmann JH. Photosynthesis and sink activity of wasp-induced galls in Acacia pycnantha. Ecology. Wiley Online Library; 2006;87: 1781–1791. 10.1890/0012-9658(2006)87[1781:pasaow]2.0.co;2 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref004] 4.Larson KC, Whitham TG. Manipulation of food resources by a gall-forming aphid: the physiology of sink-source interactions. Oecologia. Springer; 1991;88: 15–21. 10.1007/BF00328398 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref005] 5.Abrahamson WG, McCrea KD. Nutrient and biomass allocation in Solidago altissima: effects of two stem gallmakers, fertilization, and ramet isolation. Oecologia. Springer; 1986;68: 174–180. 10.1007/BF00384784 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref006] 6.Oates CN, Denby KJ, Myburg AA, Slippers B, Naidoo S. Insect gallers and their plant hosts: from omics data to systems biology. Int J Mol Sci. Multidisciplinary Digital Publishing Institute; 2016;17: 1891. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref007] 7.Poole P, Ramachandran V, Terpolilli J. Rhizobia: from saprophytes to endosymbionts. Nat Rev Microbiol. Nature Publishing Group; 2018;16: 291 10.1038/nrmicro.2017.171 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref008] 8.Martinson EO, Hackett JD, Machado CA, Arnold AE. Metatranscriptome Analysis of Fig Flowers Provides Insights into Potential Mechanisms for Mutualism Stability and Gall Induction. PLoS One. 2015;10: e0130745 10.1371/journal.pone.0130745 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref009] 9.Cook JM, Rasplus J-Y. Mutualists with attitude: coevolving fig wasps and figs. Trends Ecol Evol. Elsevier; 2003;18: 241–248. [Google Scholar]

[pgen.1008398.ref010] 10.Cook JM, West SA. Figs and fig wasps. Curr Biol. Elsevier; 2005;15: R978—R980. 10.1016/j.cub.2005.11.057 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref011] 11.Wesemael W, Viaene N, Moens M. Root-knot nematodes (Meloidogyne spp.) in Europe. Nematology. Brill; 2011;13: 3–16. [Google Scholar]

[pgen.1008398.ref012] 12.Csóka G, Stone GN, Melika G. Non-native gall-inducing insects on forest trees: a global review. Biol Invasions. Springer; 2017;19: 3161–3181. [Google Scholar]

[pgen.1008398.ref013] 13.Escobar MA, Dandekar AM. Agrobacterium tumefaciens as an agent of disease. Trends Plant Sci. Elsevier; 2003;8: 380–386. 10.1016/S1360-1385(03)00162-6 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref014] 14.Hoffmann JH, Impson FAC, Moran VC, Donnelly D. Biological control of invasive golden wattle trees (Acacia pycnantha) by a gall wasp, Trichilogaster sp.(Hymenoptera: Pteromalidae), in South Africa. Biol Control. Elsevier; 2002;25: 64–73. [Google Scholar]

[pgen.1008398.ref015] 15.Story JM, Smith L, Corn JG, White LJ. Influence of seed head-attacking biological control agents on spotted knapweed reproductive potential in western Montana over a 30-year period. Environ Entomol. Oxford University Press Oxford, UK; 2008;37: 510–519. [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref016] 16.Lacroix B, Citovsky V. Pathways of DNA Transfer to Plants from Agrobacterium tumefaciens and Related Bacterial Species. Annu Rev Phytopathol. Annual Reviews; 2019;57: 231–251. 10.1146/annurev-phyto-082718-100101 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref017] 17.Le Fevre R, Evangelisti E, Rey T, Schornack S. Modulation of host cell biology by plant pathogenic microbes. Annu Rev Cell Dev Biol. Annual Reviews; 2015;31: 201–229. 10.1146/annurev-cellbio-102314-112502 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref018] 18.Abad P, Gouzy J, Aury J-M, Castagnone-Sereno P, Danchin EGJ, Deleury E, et al. Genome sequence of the metazoan plant-parasitic nematode Meloidogyne incognita. Nat Biotechnol. 2008;26: 909–15. 10.1038/nbt.1482 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref019] 19.Mejias J, Truong NM, Abad P, Favery B, Quentin M. Plant proteins and processes targeted by parasitic nematode effectors. Front Plant Sci. Frontiers Media SA; 2019;10. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref020] 20.Rehman S, Gupta VK, Goyal AK. Identification and functional analysis of secreted effectors from phytoparasitic nematodes. BMC Microbiol. BioMed Central; 2016;16: 48 10.1186/s12866-016-0632-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref021] 21.Stuart JJ, Chen M-S, Shukle R, Harris MO. Gall midges (Hessian flies) as plant pathogens. Annu Rev Phytopathol. 2012;50: 339–57. 10.1146/annurev-phyto-072910-095255 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref022] 22.Chen M-S, Fellers JP, Zhu YC, Stuart JJ, Hulbert S, El-Bouhssini M, et al. A super-family of genes coding for secreted salivary gland proteins from the Hessian fly, Mayetiola destructor. J Insect Sci. 2006;6: 12 10.1673/2006.06.12.1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref023] 23.Zhao C, Shukle R, Navarro-Escalante L, Chen M, Richards S, Stuart JJ. Avirulence gene mapping in the Hessian fly (Mayetiola destructor) reveals a protein phosphatase 2C effector gene family. J Insect Physiol. 2016;84: 22–31. 10.1016/j.jinsphys.2015.10.001 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref024] 24.Aggarwal R, Subramanyam S, Zhao C, Chen M-S, Harris MO, Stuart JJ. Avirulence effector discovery in a plant galling and plant parasitic arthropod, the Hessian fly (Mayetiola destructor). PLoS One. Public Library of Science; 2014;9: e100958 10.1371/journal.pone.0100958 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref025] 25.Schultz JC, Edger PP, Body MJA, Appel HM. A galling insect activates plant reproductive programs during gall development. Sci Rep. 2019;9: 1833 10.1038/s41598-018-38475-6 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref026] 26.Nabity PD, Haus MJ, Berenbaum MR, DeLucia EH. Leaf-galling phylloxera on grapes reprograms host metabolism and morphology. Proc Natl Acad Sci. National Acad Sciences; 2013;110: 16663–16668. 10.1073/pnas.1220219110 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref027] 27.Raman A, Schaefer CW, Withers TM, others. Biology, ecology, and evolution of gall-inducing arthropods Boca Raton: CRC Press; 2005. [Google Scholar]

[pgen.1008398.ref028] 28.Rohfritsch O. Patterns in gall development In: Shorthouse JD, Rohfritsch O, editors. Biology of insect-induced galls. Oxford University Press: New York; 1992. pp. 60–86. [Google Scholar]

[pgen.1008398.ref029] 29.Stone GN, Schönrogge K. The adaptive significance of insect gall morphology. Trends Ecol Evol. 2003;18: 512–522. 10.1016/S0169-5347(03)00247-7 [DOI] [Google Scholar]

[pgen.1008398.ref030] 30.Dawkins R. The Extended Phenotype. Oxford: Oxford University Press; 1989. [Google Scholar]

[pgen.1008398.ref031] 31.Crespi B, Worobey M. Comparative Analysis of Gall Morphology in Australian Gall Thrips: The Evolution of Extended Phenotypes. Evolution (N Y). 1998; 1686–1696. [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref032] 32.Stone GN, Cook JM. The structure of cynipid oak galls: patterns in the evolution of an extended phenotype. Proc R Soc B Biol Sci. 1998;265 10.1098/rspb.1998.0387 [DOI] [Google Scholar]

[pgen.1008398.ref033] 33.Kutsukake M, Meng X-Y, Katayama N, Nikoh N, Shibao H, Fukatsu T. An insect-induced novel plant phenotype for sustaining social life in a closed system. Nat Commun. Nature Publishing Group; 2012;3: 1187 10.1038/ncomms2187 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref034] 34.Stern DL. Phylogenetic evidence that aphids, rather than plants, determine gall morphology. Proc R Soc London Ser B Biol Sci. The Royal Society London; 1995;260: 85–89. [Google Scholar]

[pgen.1008398.ref035] 35.Giron D, Huguet E, Stone GN, Body M. Insect-induced effects on plants and possible effectors used by galling and leaf-mining insects to manipulate their host-plant. J Insect Physiol. 2015;84: 70–89. 10.1016/j.jinsphys.2015.12.009 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref036] 36.Tooker JF, Helms AM. Phytohormone Dynamics Associated with Gall Insects, and their Potential Role in the Evolution of the Gall-Inducing Habit. J Chem Ecol. 2014;40: 742–753. 10.1007/s10886-014-0457-6 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref037] 37.Csóka G, Stone GN, Melika G. The biology, ecology and evolution of Gall-inducing Cynipidae In: Raman C, Schaefer W, Withers TM, editors. Biology, ecology and evolution of gall inducing insects. Boca Raton: CRC Press; 2005. pp. 573–642. [Google Scholar]

[pgen.1008398.ref038] 38.Bailey R, Schönrogge K, Cook JM, Melika G, Csóka G, Thúroczy C, et al. Host niches and defensive extended phenotypes structure parasitoid wasp communities. PLoS Biol. 2009;7: e1000179 10.1371/journal.pbio.1000179 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref039] 39.Ronquist F, Liljeblad J. Evolution of the gall wasp—host plant association. Evolution (N Y). 2001;55: 2503–2522. [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref040] 40.Shorthouse JD, Leggo JJ, Sliva MD, Lalonde RG. Has egg location influenced the radiation of Diplolepis (Hymenoptera: Cynipidae) gall wasps on wild roses? Basic Appl Ecol. 2005;6: 423–434. 10.1016/j.baae.2005.07.006 [DOI] [Google Scholar]

[pgen.1008398.ref041] 41.Hough JS. Studies on the common spangle gall of oak. New Phytol. 1953;52: 149–177. [Google Scholar]

[pgen.1008398.ref042] 42.Stone GN, Hernandez-Lopez A, Nicholls JA, di Pierro E, Pujade-Villar J, Melika G, et al. Extreme host plant conservatism during at least 20 million years of host plant pursuit by oak gallwasps. Evolution (N Y). 2009;63: 854–869. [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref043] 43.Hough JS. Studies on the common spangle gall of oak: II. A general consideration of past work on gall induction. New Phytol. 1953;52: 218–228. [Google Scholar]

[pgen.1008398.ref044] 44.Rey L. La galle de Biorhiza pallida Bosc. Sur la racine de Quercus pedunculata Ehrh.: observation de la ponte et de les premières réactions de l’hôte. Comptes rendus l’Académie des Sci. 1967;264: 2891–2894. [Google Scholar]

[pgen.1008398.ref045] 45.Rey L. La galle de Biorhiza pallida Ol. Etude préliminaire du bourgeon de Quercus pedunculata Ehrh. au moment de l’attaque du parasite. Bull la Société Bot Fr. 1966;113: 503–514. [Google Scholar]

[pgen.1008398.ref046] 46.Garbin L, Durfort M, Diaz NB, Pujade-Villar J. Histological Modifications on Quercus suber Twigs (Fagaceae) Caused by the Gallwasp Plagiotrochus suberi (Hymenoptera: Cynipidae). Entomol Gen. 2005;28: 91–102. [Google Scholar]

[pgen.1008398.ref047] 47.Derory J, Léger P, Garcia V, Schaeffer J, Hauser MT, Salin F, et al. Transcriptome analysis of bud burst in sessile oak (Quercus petraea). New Phytol. 2006;170: 723–738. 10.1111/j.1469-8137.2006.01721.x [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref048] 48.Plomion C, Aury J-M, Amselem J, Leroy T, Murat F, Duplessis S, et al. Oak genome reveals facets of long lifespan. Nat Plants. Nature Publishing Group; 2018;4: 440 10.1038/s41477-018-0172-3 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref049] 49.Cambier S, Ginis O, Moreau SJM, Gayral P, Hearn J, Stone GN, et al. Gall wasp transcriptomes unravel potential effectors involved in molecular dialogues with oak and rose bushes. Front Physiol. 2019;10: 926 10.3389/fphys.2019.00926 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref050] 50.Hearn J, Stone GN, Bunnefeld L, Nicholls JA, Barton NH, Lohse K. Likelihood-based inference of population history from low-coverage de novo genome assemblies. Mol Ecol. 2014;3643: 198–211. 10.1111/mec.12578 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref051] 51.Bunnefeld L, Hearn J, Stone GN, Lohse K. Whole-genome data reveal the complex history of a diverse ecological community. Proc Natl Acad Sci. 2018;115: E6507–E6515. 10.1073/pnas.1800334115 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref052] 52.Nicholls JA, Melika G, Stone GN. Sweet Tetra-Trophic Interactions: Multiple Evolution of Nectar Secretion, a Defensive Extended Phenotype in Cynipid Gall Wasps. Am Nat. 2017;189: 67–77. 10.1086/689399 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref053] 53.Cook JM, Rokas A, Pagel M, Stone GN. Evolutionary shift between host oak section and host-plant organs in Andricus gallwasps. Evolution (N Y). 2002;56: 1821–1830. [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref054] 54.Ronquist F, Nieves-Aldrey J-L, Buffington ML, Liu Z, Liljeblad J, Nylander JAA. Phylogeny, evolution and classification of gall wasps: the plot thickens. PLoS One. 2015;10: e0123301 10.1371/journal.pone.0123301 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref055] 55.Phylogeny Ronquist F., classification and evolution of the Cynipoidea. Zool Scr. 1999;28: 139–164. 10.1046/j.1463-6409.1999.00022.x [DOI] [Google Scholar]

[pgen.1008398.ref056] 56.Beyerinck MW. Beobachtungen ueber die ersten Entwicklungsphasen einiger Cynipidengallen Amsterdam: Mueller; 1882; [Google Scholar]

[pgen.1008398.ref057] 57.Brooks SE, Shorthouse JD. Developmental morphology of stem galls of Diplolepis nodulosa (Hymenoptera: Cynipidae) and those modified by the inquiline Periclistus pirata (Hymenoptera: Cynipidae) on Rosa blanda (Rosaceae). Can J Bot. NRC Research Press; 1998;76: 365–381. [Google Scholar]

[pgen.1008398.ref058] 58.Rey L. Particular aspects of the endoplasmic reticulum observed during the evolution of the nutrient cells of the gall caused by Biorhiza pallida [Quercus]. Comptes Rendus Hebd des Seances l’Academie des Sci Ser D. 1975; [Google Scholar]

[pgen.1008398.ref059] 59.Harper LJ, Schönrogge K, Lim KY, Francis P, Lichtenstein CP. Cynipid galls: insect-induced modifications of plant development create novel plant organs. Plant Cell Environ. 2004;27: 327–335. [Google Scholar]

[pgen.1008398.ref060] 60.Schönrogge K, Harper LJ, Lichtenstein CP. The protein content of tissues in cynipid galls (Hymenoptera: Cynipidae): Similarities between cynipid galls and seeds. Plant Cell Environ. 2000;23: 215–222. [Google Scholar]

[pgen.1008398.ref061] 61.Atkinson RJ, Brown GS, Stone GN. Skewed sex ratios and multiple founding in galls of the oak apple gall wasp Biorhiza pallida. Ecol Entomol. 2003;28: 14–24. [Google Scholar]

[pgen.1008398.ref062] 62.Folliot R. Les insectes cécidogènes et la cécidogenèse. Trait Zool Anat systématique, Biol. 1977;8: 389–429. [Google Scholar]

[pgen.1008398.ref063] 63.Bronner R. The role of nutritive cells in the nutrition of cynipids and cecidomyiids In: Shorthouse JD, Rohfritsch O, editors. Biology of insect-induced galls. New York: Oxford University Press; 1992. pp. 118–140. [Google Scholar]

[pgen.1008398.ref064] 64.Shorthouse JD, Rohfritsch O. Biology of insect-induced galls New York, USA: Oxford University Press; 1992. [Google Scholar]

[pgen.1008398.ref065] 65.Bagatto G, Paquette LC, Shorthouse JD. Influence of galls of Phanacis taraxaci on carbon partitioning within common dandelion, Taraxacum officinale. Entomol Exp Appl. 1996;79: 111–117. [Google Scholar]

[pgen.1008398.ref066] 66.Rey L. La galle de Biorhiza pallida Ol.: stades ultérieurs de développement. Marcellia. 1971;37: 193–218. [Google Scholar]

[pgen.1008398.ref067] 67.Sinclair FH, Stone GN, Nicholls JA, Cavers S, Gibbs M, Butterill P, et al. Impacts of local adaptation of forest trees on associations with herbivorous insects: implications for adaptive forest management. Evol Appl. Wiley Online Library; 2015;8: 972–987. 10.1111/eva.12329 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref068] 68.Hartley SE, Lawton JH. Host-plant manipulation by gall-insects: a test of the nutrition hypothesis. J Anim Ecol. JSTOR; 1992; 113–119. [Google Scholar]

[pgen.1008398.ref069] 69.Allison SD, Schultz JC. Biochemical responses of chestnut oak to a galling cynipid. J Chem Ecol. 2005;31: 151–166. [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref070] 70.Elborough K, Winz R, Deka R, Markham J, White A, Rawsthorne S, et al. Biotin carboxyl carrier protein and carboxyltransferase subunits of the multi-subunit form of acetyl-CoA carboxylase from Brassica napus: cloning and analysis of expression during oilseed rape embryogenesis. Biochem J. 1996;315: 103–112. 10.1042/bj3150103 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref071] 71.Wilhelm E. Somatic embryogenesis in oak (Quercus spp.). Vitr Cell Dev Biol—Plant. 2000;36: 349–357. 10.1007/s11627-000-0062-y [DOI] [Google Scholar]

[pgen.1008398.ref072] 72.Gibson CM, Hunter MS. Extraordinarily widespread and fantastically complex: Comparative biology of endosymbiotic bacterial and fungal mutualists of insects. Ecol Lett. 2010;13: 223–234. 10.1111/j.1461-0248.2009.01416.x [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref073] 73.Kaiser W, Huguet E, Casas J, Commin C, Giron D. Plant green-island phenotype induced by leaf-miners is mediated by bacterial symbionts. Proc R Soc B Biol Sci. 2010;277: 2311–2319. 10.1098/rspb.2010.0214 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref074] 74.Bansal R, Hulbert S, Schemerhorn B, Reese JC, Whitworth RJ, Stuart JJ, et al. Hessian Fly-Associated Bacteria: Transmission, Essentiality, and Composition. PLoS One. 2011;6: e23170 10.1371/journal.pone.0023170 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref075] 75.Joy JB. Symbiosis catalyses niche expansion and diversification. Proc Biol Sci. 2013;280: 20122820 10.1098/rspb.2012.2820 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref076] 76.Nelson LA, Davies KA, Scheffer SJ, Taylor GS, Purcell MF, Giblin-Davis RM, et al. An emerging example of tritrophic coevolution between flies (Diptera: Fergusoninidae) and nematodes (Nematoda: Neotylenchidae) on myrtaceae host plants. Biol J Linn Soc. 2014;111: 699–718. 10.1111/bij.12237 [DOI] [Google Scholar]

[pgen.1008398.ref077] 77.Borkent A, Bissett J. Gall midges (Diptera: Cecidomyiidae) are vectors for their fungal symbionts. Symbiosis (USA). 1985; [Google Scholar]

[pgen.1008398.ref078] 78.Joy JB. Symbiosis catalyses niche expansion and diversification. Proc R Soc B Biol Sci. 2013;280. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref079] 79.Cornell HV. The Secondary Chemistry and Complex Morphology of Galls Formed by the Cynipinae (Hymenoptera): Why and How? Am Midl Nat. 1983;110: 225–234. 10.2307/2425263 [DOI] [Google Scholar]

[pgen.1008398.ref080] 80.Bézier A, Annaheim M, Herbinière J, Wetterwald C, Gyapay G, Bernard-Samain S, et al. Polydnaviruses of Braconid Wasps Derive from an Ancestral Nudivirus. Science (80-). 2009;323: 926–930. 10.1126/science.1166788 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref081] 81.Drezen J-M, Leobold M, Bezier A, Huguet E, Volkoff A-N, Herniou EA. Endogenous viruses of parasitic wasps: variations on a common theme. Curr Opin Virol. Elsevier; 2017;25: 41–48. 10.1016/j.coviro.2017.07.002 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref082] 82.Rizki RM, Rizki TM. Parasitoid virus-like particles destroy Drosophila cellular immunity. Proc Natl Acad Sci. National Acad Sciences; 1990;87: 8388–8392. 10.1073/pnas.87.21.8388 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref083] 83.H R. Von welchen Organen der Gallwespenlarve geht der Reiz zur Bildung der Pflanzengalle aus? Zoologischer Jahrbücher. Abteilung für Syst Geogr und Biol der Tiere. 1904;20: 19–90. [Google Scholar]

[pgen.1008398.ref084] 84.Roth P. Beiträge zur Biologie der Gallwespen. Verhandlungen der Naturforscher Gesellschaft Basel. 1949;60: 104–178. [Google Scholar]

[pgen.1008398.ref085] 85.Westermann AJ, Gorski SA, Vogel J. Dual RNA-seq of pathogen and host. Nat Rev Microbiol. 2012;10: 618–630. 10.1038/nrmicro2852 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref086] 86.Nabity PD. Insect-induced plant phenotypes: Revealing mechanisms through comparative genomics of galling insects and their hosts. Am J Bot. 2016;103: 979–981. 10.3732/ajb.1600111 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref087] 87.Zhu C, Shi F, Chen Y, Wang M, Zhao Y, Geng G. Transcriptome Analysis of Chinese Chestnut (Castanea mollissima Blume) in Response to Dryocosmus kuriphilus Yasumatsu Infestation. Int J Mol Sci. Multidisciplinary Digital Publishing Institute; 2019;20: 855. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref088] 88.Jia XY, He LH, Jing RL, Li RZ. Calreticulin: Conserved protein and diverse functions in plants. Physiol Plant. 2009;136: 127–138. 10.1111/j.1399-3054.2009.1223.x [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref089] 89.Zürcher E, Liu J, di Donato M, Geisler M, Müller B. Plant development regulated by cytokinin sinks. Science (80-). 2016;353: 1027–1030. 10.1126/science.aaf7254 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref090] 90.Seo M, Akaba S, Oritani T, Delarue M, Bellini C, Caboche M, et al. Higher Activity of an Aldehyde Oxidase in the Auxin-Overproducing superroot 1 mutant of Arabidopsis thaliana. Plant Physiol. Am Soc Plant Biol; 1998;116: 687–693. 10.1104/pp.116.2.687 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref091] 91.Rhoades MW, Reinhart BJ, Lim LP, Burge CB, Bartel B, Bartel DP. Prediction of plant microRNA targets. Cell. Elsevier; 2002;110: 513–520. 10.1016/s0092-8674(02)00863-2 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref092] 92.Iwama A, Yamashino T, Tanaka Y, Sakakibara H, Kakimoto T, Sato S, et al. AHK5 histidine kinase regulates root elongation through an ETR1-dependent abscisic acid and ethylene signaling pathway in Arabidopsis thaliana. Plant cell Physiol. Oxford University Press; 2007;48: 375–380. 10.1093/pcp/pcl065 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref093] 93.Tran L-SP, Urao T, Qin F, Maruyama K, Kakimoto T, Shinozaki K, et al. Functional analysis of AHK1/ATHK1 and cytokinin receptor histidine kinases in response to abscisic acid, drought, and salt stress in Arabidopsis. Proc Natl Acad Sci. National Acad Sciences; 2007;104: 20623–20628. 10.1073/pnas.0706547105 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref094] 94.Suzuki H, Yokokura J, Ito T, Arai R, Yokoyama C, Toshima H, et al. Biosynthetic pathway of the phytohormone auxin in insects and screening of its inhibitors. Insect Biochem Mol Biol. 2014;53: 66–72. 10.1016/j.ibmb.2014.07.008 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref095] 95.Römer P, Hahn S, Jordan T, Strauß T, Bonas U, Lahaye T. Plant pathogen recognition mediated by promoter activation of the pepper Bs3 resistance gene. Science (80-). American Association for the Advancement of Science; 2007;318: 645–648. [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref096] 96.Field LM, James AA, Plantard O, Rasplus J-Y, Mondor G, Clainche I, et al. Distribution and phylogeny of Wolbachia inducing thelytoky in Rhoditini and Aylacini (Hymenoptera: Cynipidae). Insect Mol Biol. 1999;8: 185–191. [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref097] 97.Rokas A, Atkinson RJ, Nieves-Aldrey JL, West SA, Stone GN. The incidence and diversity of Wolbachia in gallwasps (Hymenoptera; Cynipidae) on oak. Mol Ecol. 2002;11: 1815–1829. 10.1046/j.1365-294x.2002.01556.x [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref098] 98.Bailey TL, Williams N, Misleh C, Li WW. MEME: discovering and analyzing DNA and protein sequence motifs. Nucleic Acids Res. 2006;34: 369–373. 10.1093/nar/gkl198 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref099] 99.Keeling CI, Henderson H, Li M, Yuen M, Clark EL, Fraser JD, et al. Transcriptome and full-length cDNA resources for the mountain pine beetle, Dendroctonus ponderosae Hopkins, a major insect pest of pine forests. Insect Biochem Mol Biol. 2012;42: 525–36. 10.1016/j.ibmb.2012.03.010 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref100] 100.Keeling CI, Yuen MM, Liao NY, Docking TR, Chan SK, Taylor GA, et al. Draft genome of the mountain pine beetle, Dendroctonus ponderosae Hopkins, a major forest pest. Genome Biol. BioMed Central Ltd; 2013;14: R27 10.1186/gb-2013-14-3-r27 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref101] 101.Pauchet Y, Heckel DG. The genome of the mustard leaf beetle encodes two active xylanases originally acquired from bacteria through horizontal gene transfer. Proc Biol Sci. 2013;280: 20131021 10.1098/rspb.2013.1021 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref102] 102.Pauchet Y, Wilkinson P, Chauhan R, Ffrench-Constant RH. Diversity of beetle genes encoding novel plant cell wall degrading enzymes. PLoS One. 2010;5: e15635 10.1371/journal.pone.0015635 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref103] 103.De Lorenzo G, Ferrari S, Giovannoni M, Mattei B, Cervone F. Cell wall traits that influence plant development, immunity, and bioconversion. Plant J. Wiley Online Library; 2019;97: 134–147. 10.1111/tpj.14196 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref104] 104.Vallarino JG, Osorio S. Signaling role of oligogalacturonides derived during cell wall degradation. Plant Signal Behav. Landes Bioscience; 2012;7: 1447–1449. 10.4161/psb.21779 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref105] 105.Pawłowski TA, Staszak AM, Karolewski P, Giertych MJ. Plant development reprogramming by cynipid gall wasp: proteomic analysis. Acta Physiol Plant. Springer; 2017;39: 114. [Google Scholar]

[pgen.1008398.ref106] 106.Takei M, Kogure S, Yokoyama C, Kouzuma Y, Suzuki Y. Identification of an aldehyde oxidase involved in indole-3-acetic acid synthesis in Bombyx mori silk gland. Biosci Biotechnol Biochem. Taylor & Francis; 2019;83: 129–136. 10.1080/09168451.2018.1525275 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref107] 107.Zhao CY, Escalante LN, Chen H, Benatti TR, Qu JX, Chellapilla S, et al. A massive expansion of effector genes underlies gall-formation in the wheat pest Mayetiola destructor. Curr Biol. 2015;25 10.1016/j.cub.2014.12.057 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref108] 108.Abe Y, Ide T, Wachi N. Discovery of a New Gall-Inducing Species in the Inquiline Tribe Synergini (Hymenoptera: Cynipidae): Inconsistent Implications from Biology and Morphology. Ann Entomol Soc Am. 2011;104: 115–120. 10.1603/AN10149 [DOI] [Google Scholar]

[pgen.1008398.ref109] 109.Stanke M, Morgenstern B. AUGUSTUS: a web server for gene prediction in eukaryotes that allows user-defined constraints. Nucleic Acids Res. Oxford University Press; 2005;33: W465—W467. 10.1093/nar/gki458 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref110] 110.Volkenburgh E Van. Leaf expansion—an integrating plant behaviour. Plant Cell Environ. Wiley Online Library; 1999;22: 1463–1473. [Google Scholar]

[pgen.1008398.ref111] 111.Maksymowych R. Analysis of leaf development. CUP Archive; 1973. [Google Scholar]

[pgen.1008398.ref112] 112.Becraft PW. 1 Development of the Leaf Epidermis Current topics in developmental biology. Elsevier; 1999. pp. 1–40. [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref113] 113.Botta R, Sartor C, Torello Marinoni D, Quacchia A, Alma A. Differential gene expression in chestnut buds following infestation by gall-wasp (Dryocosmus kuriphilus Yasumatsu, Hymenoptera: Cynipidae). IV International Chestnut Symposium 844. 2008. pp. 405–410.

[pgen.1008398.ref114] 114.van Hengel AJ, Tadesse Z, Immerzeel P, Schols H, van Kammen A, de Vries SC. N-acetylglucosamine and glucosamine-containing arabinogalactan proteins control somatic embryogenesis. Plant Physiol. 2001;125: 1880–90. 10.1104/pp.125.4.1880 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref115] 115.De Jong AJ, Cordewener J, Lo Schiavo F, Terzi M, Vandekerckhove J, Van Kammen A, et al. A carrot somatic embryo mutant is rescued by chitinase. Plant Cell. 1992;4: 425–33. 10.1105/tpc.4.4.425 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref116] 116.Poon S, Heath RL, Clarke AE. A chimeric arabinogalactan protein promotes somatic embryogenesis in cotton cell culture. Plant Physiol. 2012;160: 684–95. 10.1104/pp.112.203075 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref117] 117.Fehér A, Pasternak TP, Dudits D. Transition of somatic plant cells to an embryogenic state. Plant Cell, Tissue Organ Cult. 2003;74: 201–228. 10.1023/A:1024033216561 [DOI] [Google Scholar]

[pgen.1008398.ref118] 118.Nguema-Ona E, Vicré-Gibouin M, Cannesan M-A, Driouich A. Arabinogalactan proteins in root-microbe interactions. Trends Plant Sci. 2013;18: 440–449. 10.1016/j.tplants.2013.03.006 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref119] 119.Van Hengel AJ, Van Kammen A, De Vries SC. A relationship between seed development, Arabinogalactan-proteins (AGPs) and the AGP mediated promotion of somatic embryogenesis. Physiol Plant. 2002;114: 637–644. [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref120] 120.Costa ML, Sobral R, Costa MMR, Amorim MI, Coimbra S. Evaluation of the presence of arabinogalactan proteins and pectins during Quercus suber male gametogenesis. Ann Bot. Annals Botany Co; 2014; mcu223. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref121] 121.De Jong AJ, Heidstra R, Spaink HP, Hartog MV, Meijer EA, Hendriks T, et al. Rhizobium Lipooligosaccharides Rescue a Carrot Somatic Embryo Mutant. Plant Cell. 1993;5: 615–620. 10.1105/tpc.5.6.615 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref122] 122.Röhrig H, Schmidt J, Walden R, Czaja I, Miklaševičs E, Wieneke U, et al. Growth of Tobacco Protoplasts Stimulated by Synthetic Lipo-Chitooligosaccharides. Science (80-). 1995;269: 841–843. 10.1126/science.269.5225.841 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref123] 123.Schultze M, Kondorosi A. Regulation of symbiotic root nodule development. Annu Rev Genet. 1998;32: 33–57. 10.1146/annurev.genet.32.1.33 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref124] 124.Cannesan M-A, Nguema-Ona E. Arabinogalactan proteins in root–microbe interactions. Trends Plant Sci. 2013;18: 440–449. 10.1016/j.tplants.2013.03.006 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref125] 125.Vandenbussche F, Vaseva I, Vissenberg K, Van Der Straeten D. Ethylene in vegetative development: a tale with a riddle. New Phytol. Wiley Online Library; 2012;194: 895–909. 10.1111/j.1469-8137.2012.04100.x [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref126] 126.Gresshoff PM, Lohar D, Chan P-K, Biswas B, Jiang Q, Reid D, et al. Genetic analysis of ethylene regulation of legume nodulation. Plant Signal Behav. Taylor & Francis; 2009;4: 818–823. 10.4161/psb.4.9.9395 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref127] 127.Tiburcio AF, Altabella T, Bitrián M, Alcázar R. The roles of polyamines during the lifespan of plants: from development to stress. Planta. Springer; 2014;6: 1–18. [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref128] 128.Volkoff A-N, Jouan V, Urbach S, Samain S, Bergoin M, Wincker P, et al. Analysis of virion structural components reveals vestiges of the ancestral ichnovirus genome. PLoS Pathog. 2010;6: e1000923 10.1371/journal.ppat.1000923 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref129] 129.Webster CL, Waldron FM, Robertson S, Crowson D, Ferrari G, Quintana JF, et al. The discovery, distribution, and evolution of viruses associated with Drosophila melanogaster. PLoS Biol. Public Library of Science; 2015;13: e1002210 10.1371/journal.pbio.1002210 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref130] 130.Quentin M, Abad P, Favery B. Plant parasitic nematode effectors target host defense and nuclear functions to establish feeding cells. Front Plant Sci. 2013;4: 53 10.3389/fpls.2013.00053 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref131] 131.Boulain H, Legeai F, Guy E, Morlière S, Douglas NE, Oh J, et al. Fast Evolution and Lineage-Specific Gene Family Expansions of Aphid Salivary Effectors Driven by Interactions with Host-Plants. Genome Biol Evol. Oxford University Press; 2018;10: 1554–1572. 10.1093/gbe/evy097 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref132] 132.Bailey S, Percy DM, Hefer CA, Cronk QCB. The transcriptional landscape of insect galls: psyllid (Hemiptera) gall formation in Hawaiian Metrosideros polymorpha (Myrtaceae). BMC Genomics. 2015;16: 943 10.1186/s12864-015-2109-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref133] 133.Tanaka Y, Okada K, Asami T, Suzuki Y. Phytohormones in Japanese Mugwort Gall Induction by a Gall-Inducing Gall Midge. Biosci Biotechnol Biochem. 2013;77: 1942–1948. 10.1271/bbb.130406 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref134] 134.Takei M, Yoshida S, Kawai T, Hasegawa M, Suzuki Y. Adaptive significance of gall formation for a gall-inducing aphids on Japanese elm trees. J Insect Physiol. 2015;72: 43–51. 10.1016/j.jinsphys.2014.11.006 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref135] 135.Ohkawa M. Isolation of zeatin from larvae of Dryocosmus kuriphilus Yasumatsu [Castanea crenata, gall initiation]. HortScience (USA). 1974;9: 458–459. [Google Scholar]

[pgen.1008398.ref136] 136.Percy DM, Page RDM, Cronk QCB. Plant-insect interactions: double-dating associated insect and plant lineages reveals asynchronous radiations. Syst Biol. 2004;53: 120–127. 10.1080/10635150490264996 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref137] 137.Wood BW, Payne JA. Growth regulators in chestnut shoot galls infected with oriental chestnut gall wasp, Dryocosmus kuriphilus (Hymenoptera: Cynipidae). Environ Entomol. 1988;17: 915–920. [Google Scholar]

[pgen.1008398.ref138] 138.Brütting C, Crava CM, Schäfer M, Schuman MC, Meldau S, Adam N, et al. Cytokinin transfer by a free-living mirid to Nicotiana attenuata recapitulates a strategy of endophytic insects. Elife. eLife Sciences Publications Limited; 2018;7: e36268 10.7554/eLife.36268 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref139] 139.Bartlett L, Connor EF. Exogenous phytohormones and the induction of plant galls by insects. Arthropod Plant Interact. 2014;8: 339–348. 10.1007/s11829-014-9309-0 [DOI] [Google Scholar]

[pgen.1008398.ref140] 140.John M, Röhrig H, Schmidt J, Walden R, Schell J. Cell signalling by oligosaccharides. Trends Plant Sci. Elsevier; 2015;2: 111–115. 10.1016/S1360-1385(97)01005-4 [DOI] [Google Scholar]

[pgen.1008398.ref141] 141.Yue J, Hu X, Huang J. Origin of plant auxin biosynthesis. Trends Plant Sci. Elsevier; 2014;19: 764–770. 10.1016/j.tplants.2014.07.004 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref142] 142.Nieminen KM, Kauppinen L, Helariutta Y. A weed for wood? Arabidopsis as a genetic model for xylem development. Plant Physiol. Am Soc Plant Biol; 2004;135: 653–659. 10.1104/pp.104.040212 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref143] 143.Bar M, Ori N. Leaf development and morphogenesis. Development. 2014;141: 4219–4230. 10.1242/dev.106195 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref144] 144.Body MJA, Zinkgraf MS, Whitham TG, Lin C-H, Richardson RA, Appel HM, et al. Heritable Phytohormone Profiles of Poplar Genotypes Vary in Resistance to a Galling Aphid. Mol Plant-Microbe Interact. 2019;32: 654–672. 10.1094/MPMI-11-18-0301-R [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref145] 145.Kunieda T, Fujiyuki T, Kucharski R, Foret S, Ament SA, Toth AL, et al. Carbohydrate metabolism genes and pathways in insects: insights from the honey bee genome. Insect Mol Biol. 2006;15: 563–576. 10.1111/j.1365-2583.2006.00677.x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref146] 146.Mckenna DD, Scully ED, Pauchet Y, Hoover K, Kirsch R, Geib SM, et al. Genome of the Asian longhorned beetle (Anoplophora glabripennis), a globally significant invasive species , reveals key functional and evolutionary innovations at the beetle–plant interface. Genome Biol. 2016; 1–18. 10.1186/s13059-015-0866-z [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref147] 147.Kirsch R, Gramzow L, Theißen G, Siegfried BD, Richard H, Heckel DG, et al. Horizontal gene transfer and functional diversification of plant cell wall degrading polygalacturonases: Key events in the evolution of herbivory in beetles. Insect Biochem Mol Biol. 2014;52: 33–50. 10.1016/j.ibmb.2014.06.008 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref148] 148.Kirsch R, Wielsch N, Vogel H, Svatoš A, Heckel DG, Pauchet Y. Combining proteomics and transcriptome sequencing to identify active plant-cell-wall-degrading enzymes in a leaf beetle. BMC Genomics. 2012;13: 587 10.1186/1471-2164-13-587 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref149] 149.Busch A, Danchin EGJ, Pauchet Y. Functional diversification of horizontally acquired glycoside hydrolase family 45 (GH45) proteins in Phytophaga beetles. BMC Evol Biol. 2019;19: 100 10.1186/s12862-019-1429-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref150] 150.Brand P, Lin W, Johnson BR. The draft genome of the invasive walking stick, Medauroidea extradendata, reveals extensive Lineage-Specific gene family expansions of Cell wall degrading enzymes in phasmatodea. G3 Genes, Genomes, Genet. 2018;8: 1403–1408. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref151] 151.Bronner R. Proprieties lytiques des œufs de Biorrihza pallida. Comptes Rendus l’Académie des Sci. 1973;276: 189–192. [Google Scholar]

[pgen.1008398.ref152] 152.Bronner R. Contribution a l’etude histochimique des tissus nourriciers des zoocecidies. Marcellia. 1977;40: 1–134. [Google Scholar]

[pgen.1008398.ref153] 153.Rodriguez PA, Stam R, Warbroek T, Bos JIB. Mp10 and Mp42 from the aphid species Myzus persicae trigger plant defenses in Nicotiana benthamiana through different activities. Mol Plant-Microbe Interact. Am Phytopath Society; 2014;27: 30–39. 10.1094/MPMI-05-13-0156-R [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref154] 154.Jaouannet M, Rodriguez PA, Thorpe P, Lenoir CJG, MacLeod R, Escudero-Martinez C, et al. Plant immunity in plant–aphid interactions. Front Plant Sci. 2014;5: 663 10.3389/fpls.2014.00663 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref155] 155.Hogenhout SA, Bos JIB. Effector proteins that modulate plant–insect interactions. Curr Opin Plant Biol. 2011;14: 422–428. 10.1016/j.pbi.2011.05.003 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref156] 156.Sanchez F, Padilla J E, Perez H, and Lara M. Control of Nodulin Genes in Root-Nodule Development and Metabolism. Annu Rev Plant Physiol Plant Mol Biol. 1991;42: 507–528. 10.1146/annurev.pp.42.060191.002451 [DOI] [Google Scholar]

[pgen.1008398.ref157] 157.Ma Y, Yan C, Li H, Wu W, Liu Y, Wang Y, et al. Bioinformatics Prediction and Evolution Analysis of Arabinogalactan Proteins in the Plant Kingdom. Front Plant Sci. 2017;8: 66 10.3389/fpls.2017.00066 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref158] 158.Mashiguchi K, Asami T, Suzuki Y. Genome-Wide Identification, Structure and Expression Studies, and Mutant Collection of 22 Early Nodulin-Like Protein Genes in Arabidopsis. Biosci Biotechnol Biochem. 2009;73: 2452–2459. 10.1271/bbb.90407 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref159] 159.Redding NW, Agudelo P, Wells CE. Multiple Nodulation Genes Are Up-Regulated During Establishment of Reniform Nematode Feeding Sites in Soybean. Phytopathology. Am Phytopath Society; 2018;108: 275–291. 10.1094/PHYTO-04-17-0154-R [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref160] 160.Domon J-M, Neutelings G, Roger D, David A, David H. A basic chitinase-like protein secreted by embryogenic tissues of Pinus caribaea acts on arabinogalactan proteins extracted from the same cell lines. J Plant Physiol. 2000;156: 33–39. [Google Scholar]

[pgen.1008398.ref161] 161.Arakane Y, Muthukrishnan S. Insect chitinase and chitinase-like proteins. Cell Mol Life Sci. 2010;67: 201–216. 10.1007/s00018-009-0161-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref162] 162.Zhu Q, Arakane Y, Beeman RW, Kramer KJ, Muthukrishnan S. Functional specialization among insect chitinase family genes revealed by RNA interference. Proc Natl Acad Sci U S A. 2008;105: 6650–5. 10.1073/pnas.0800739105 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref163] 163.Taper ML, Case TJ. Interactions between oak tannins and parasite community structure: unexpected benefits of tannins to cynipid gall-wasps. Oecologia. Springer; 1987;71: 254–261. 10.1007/BF00377292 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref164] 164.Taper ML, Zimmerman EM, Case TJ. Sources of mortality for a cynipid gall-wasp (Dryocosmus dubiosus (Hymenoptera: Cynipidae)): the importance of the tannin/fungus interaction. Oecologia. Springer; 1986;68: 437–445. 10.1007/BF01036752 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref165] 165.Wilson D, Carroll GC. Avoidance of high-endophyte space by gall-forming insects. Ecology. Wiley Online Library; 1997;78: 2153–2163. [Google Scholar]

[pgen.1008398.ref166] 166.Wilson D. Fungal endophytes which invade insect galls: insect pathogens, benign saprophytes, or fungal inquilines? Oecologia. Springer; 1995;103: 255–260. 10.1007/BF00329088 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref167] 167.Tichomiroff A. Chemische Studien über die Entwicklung der Insecteneier. Zeitschrift für Physiol Chemie. 1885;9: 518–532. [Google Scholar]

[pgen.1008398.ref168] 168.Furneaux PJS, Mackay AL. The composition, structure and formation of the chorion and the vitelline membrane of the insect egg-shell The insect integument. Amsterdam: Elsevier; 1976. pp. 157–176. [Google Scholar]

[pgen.1008398.ref169] 169.Moreira MF, Dos Santos AS, Marotta HR, Mansur JF, Ramos IB, Machado EA, et al. A chitin-like component in Aedes aegypti eggshells, eggs and ovaries. Insect Biochem Mol Biol. 2007;37: 1249–61. 10.1016/j.ibmb.2007.07.017 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref170] 170.Lohse K, Sharanowski B, Nicholls JA, Blaxter M, Stone GN. Developing EPIC markers for chalcidoid Hymenoptera from EST and genomic data. Mol Ecol Resour. 2011;3: 521–529. [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref171] 171.Chen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. Cold Spring Harbor Laboratory; 2018;34: i884–i890. 10.1093/bioinformatics/bty560 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref172] 172.Andrews S. FastQC: A quality control application for high throughput sequence data. Available at: http://www.bioinformatics.babraham.ac.uk/projects/fastqc. 2010. [Google Scholar]

[pgen.1008398.ref173] 173.Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, et al. Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotech. 2011;29: 644–652. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref174] 174.Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C. Salmon provides fast and bias-aware quantification of transcript expression. Nat Methods. Nature Publishing Group; 2017;14: 417–419. 10.1038/nmeth.4197 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref175] 175.Soneson C, Love MI, Robinson MD. Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Research. Faculty of 1000 Ltd; 2015;4. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref176] 176.Haas BJ, Papanicolaou A, Yassour M, Grabherr M, Blood PD, Bowden J, et al. De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis. Nat Protoc. 2013;8: 1494–1512. 10.1038/nprot.2013.084 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref177] 177.Davidson NM, Hawkins ADK, Oshlack A. SuperTranscripts: a data driven reference for analysis and visualisation of transcriptomes. Genome Biol. BioMed Central; 2017;18: 148 10.1186/s13059-017-1284-1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref178] 178.Li H. Minimap2: pairwise alignment for nucleotide sequences Bioinformatics. Oxford University Press; 2018;1: 7. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref179] 179.Plomion C, Aury J-M, Amselem J, Alaeitabar T, Barbe V, Belser C, et al. Decoding the oak genome: public release of sequence data, assembly, annotation and publication strategies. Mol Ecol Resour. Wiley Online Library; 2016;16: 254–265. 10.1111/1755-0998.12425 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref180] 180.Buchfink B, Xie C, Huson DH. Fast and sensitive protein alignment using DIAMOND. Nat Methods. Nature Publishing Group; 2015;12: 59 10.1038/nmeth.3176 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref181] 181.Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. Oxford University Press; 2012;41: D590—D596. 10.1093/nar/gks1219 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref182] 182.Simão FA, Waterhouse RM, Ioannidis P, Kriventseva E V, Zdobnov EM. BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. Bioinformatics. Oxford University Press; 2015;31: 3210–3212. 10.1093/bioinformatics/btv351 [DOI] [PubMed] [Google Scholar]

[pgen.1008398.ref183] 183.Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-Seq data with DESeq2. Genome Biol. 2014;15: 550 10.1186/s13059-014-0550-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref184] 184.Zhu A, Ibrahim JG, Love MI. Heavy-tailed prior distributions for sequence count data: removing the noise and preserving large differences bioRxiv. Cold Spring Harbor Laboratory; 2018; 303255. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref185] 185.Stephens M. False discovery rates: a new deal. Biostatistics. Oxford University Press; 2016;18: 275–294. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref186] 186.Alexa A, Rahnenfuhrer J. topGO: Enrichment Analysis for Gene Ontology. Available at: https://bioconductor.org/packages/release/bioc/html/topGO.html. 2016. 10.3390/ijms17030353 [DOI] [Google Scholar]

[pgen.1008398.ref187] 187.Supek F, Bošnjak M, Škunca N, Šmuc T. REVIGO summarizes and visualizes long lists of gene ontology terms. PLoS One. 2011;6: e21800 10.1371/journal.pone.0021800 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref188] 188.Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, et al. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol. 2012;19: 455–77. 10.1089/cmb.2012.0021 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref189] 189.Zimin A V, Marçais G, Puiu D, Roberts M, Salzberg SL, Yorke JA. The MaSuRCA genome assembler. Bioinformatics. 2013;29: 2669–77. 10.1093/bioinformatics/btt476 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref190] 190.Smit AFA, Hubley R, Green P. RepeatMasker Open-3.0. 1996–2010.

[pgen.1008398.ref191] 191.Kumar S, Jones M, Koutsovoulos G, Clarke M, Blaxter M. Blobology: exploring raw genome data for contaminants, symbionts and parasites using taxon-annotated GC-coverage plots. Front Genet. Frontiers; 2013;4: 237 10.3389/fgene.2013.00237 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pgen.1008398.ref192] 192.Emms DM, Kelly S. OrthoFinder: solving fundamental biases in whole genome comparisons dramatically improves orthogroup inference accuracy. Genome Biol. BioMed Central; 2015;16: 157 10.1186/s13059-015-0721-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Genomic dissection of an extended phenotype: Oak galling by a cynipid gall wasp

Jack Hearn

Mark Blaxter

Karsten Schönrogge

José-Luis Nieves-Aldrey

Juli Pujade-Villar

Elisabeth Huguet

Jean-Michel Drezen

Joseph D Shorthouse

Graham N Stone

Roles

Abstract

Author summary

Introduction

Fig 1. Stages in the development of the sexual generation galls of Biorhiza pallida and of ungalled buds of Quercus robur.

What, in plant terms, is a cynipid gall?

Fig 2. Patterns of oak, gall wasp and symbiont gene expression expected under alternative hypotheses for cynipid gall development.

What organism induces the B. pallida oak gall?

What processes are involved in gall wasp manipulation of plant development?

Which gall wasp genes involved in gall development are novel compared to non-galling sister groups?

Results

Generation of RNA-seq datasets for B. pallida galls and ungalled oak buds

Table 1. Assembly metrics for the separated gall wasp (Biorhiza pallida) and oak bud plus gall tissue (Quercus robur) transcriptomes.

Table 2. Taxonomic assignments of gall dual-RNA-Seq reads, by stage and replicate (1–12), expressed as a % of the total mapped read set.

Gall wasp and oak genes show consistent patterns of stage-specific expression in Biorhiza pallida galls

Fig 3. Overall similarity in patterns of gene expression within and among developmental stages of B. pallida galls and normally developing oak buds.

Table 3. Counts of gall wasp and oak genes differentially expressed (DE) in gall tissues.

Oak gene expression in normally developing tissues shows partial differentiation between bud stages and strong differentiation between buds and leaves

Table 4. Counts of gall wasp and oak genes differentially expressed (DE) in normally developing bud developmental stages.

Functional annotation of gall wasp and oak genes expressed during gall and normal bud development

Galled and ungalled buds show diverging patterns of oak gene expression during development

Fig 4. Venn diagrams of numbers of oak genes that were differentially expressed in each, and both, of gall and normal bud tissues.

Fig 5. REVIGO scatter plots of enriched biological process GO terms for oak genes expressed during development of galled and normal oak bud tissues.

Patterns in the expression of oak genes associated with alternative hypotheses for gall development

No demonstrated role for Symbionts and virus-like-particles in cynipid gall development

Table 5. Assembly metrics for gall inducing cynipid and parasitoid figitid wasp genomes.

Patterns of gall wasp gene expression during gall development

Table 6. Plant cell wall degrading enzyme (PCWDE) genes in the nuclear genomes of cynipid gall wasps, their Figitidae sister group, and Nasonia vitripennis (Chalcidoidea).

Which gall wasp genes involved in gall development are novel compared to non-galling sister groups?

Discussion

What, in plant terms, is a cynipid gall?

What organism induces the B. pallida oak gall?

What processes are involved in gall wasp manipulation of plant development?

Which gall wasp genes involved in gall development are novel compared to non-galling sister groups?

Chitinases, arabino-galactan glycoproteins and a model for cynipid gall induction

Conclusions

Methods

Sample collection, RNA extraction and sequencing

Transcriptome assembly, quantification and taxonomic assignment

Statistical analysis of stage specific variation in gene expression

Orthologue identification of candidate genes

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Gregory P Copenhaver

Roles

Decision Letter 1

Gregory P Copenhaver

Roles

Acceptance letter

Gregory P Copenhaver

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases