Skip to main content
NCBI home page
Ecology and Evolution logoLink to Ecology and Evolution
. 2024 Mar 11;14(3):e10979. doi: 10.1002/ece3.10979

A pooled‐sample draft genome assembly provides insights into host plant‐specific transcriptional responses of a Solanaceae‐specializing pest, Tupiocoris notatus (Hemiptera: Miridae)

Jay K Goldberg 1,2,✉, Carson W Allan 3, Dario Copetti 4,5, Luciano M Matzkin 1,3,5, Judith Bronstein 1,3,5
PMCID: PMC10928254  PMID: 38476697

Abstract

The assembly of genomes from pooled samples of genetically heterogenous samples of conspecifics remains challenging. In this study, we show that high‐quality genome assemblies can be produced from samples of multiple wild‐caught individuals. We sequenced DNA extracted from a pooled sample of conspecific herbivorous insects (Hemiptera: Miridae: Tupiocoris notatus) acquired from a greenhouse infestation in Tucson, Arizona (in the range of 30–100 individuals; 0.5 mL tissue by volume) using PacBio highly accurate long reads (HiFi). The initial assembly contained multiple haplotigs (>85% BUSCOs duplicated), but duplicate contigs could be easily purged to reveal a highly complete assembly (95.6% BUSCO, 4.4% duplicated) that is highly contiguous by short‐read assembly standards (N 50 = 675 kb; Largest contig = 4.3 Mb). We then used our assembly as the basis for a genome‐guided differential expression study of host plant‐specific transcriptional responses. We found thousands of genes (N = 4982) to be differentially expressed between our new data from individuals feeding on Datura wrightii (Solanaceae) and existing RNA‐seq data from Nicotiana attenuata (Solanaceae)‐fed individuals. We identified many of these genes as previously documented detoxification genes such as glutathione‐S‐transferases, cytochrome P450s, and UDP‐glucosyltransferases. Together our results show that long‐read sequencing of pooled samples can provide a cost‐effective genome assembly option for small insects and can provide insights into the genetic mechanisms underlying interactions between plants and herbivorous pests.

Keywords: genomics, hemiptera, herbivory, plant‐insect, interactions, solanaceae

We present a draft genome assembly for Tupiocoris notatus generated from a pooled sample of individuals. We use this genome as a basis for a genome‐guided differential expression study and identify gene expression changes associated with feeding on different host plants.

graphic file with name ECE3-14-e10979-g003.jpg

1. INTRODUCTION

Despite being highly toxic due to numerous noxious defensive chemicals, plants in the nightshade family (Solanaceae) have a handful of specialized herbivores. One of these insects is the tobacco suckfly (Hemiptera: Miridae: Tupiocoris notatus), which is known to feed on tobaccos (Nicotiana sp.; Halitschke et al., 2011) and sacred Datura (Datura wrightii; Van Dam & Hare, 1998), two genera known for their alkaloid defenses. Tupiocoris notatus has been an important part of research on the ecological roles of plant defense induction in wild tobacco (Nicotiana attenuata; Heidel & Baldwin, 2004). Of note is T. notatus' ability to “vaccinate” plants against a more dangerous herbivore, the tobacco hornworm (Manduca sexta [Lepidoptera: Sphingidae]), a single individual of which can completely defoliate plants (Kessler & Baldwin, 2004). It has also been implicated in the maintenance of a natural trichome dimorphism, by selecting against a glandular trichome‐producing morph when it becomes locally common (Goldberg et al., 2020). Furthermore, its transcriptome has previously been studied and putative plant‐defense response genes have been identified (Crava et al., 2016). More recently, this species has been shown to manipulate plant defense and metabolism through cytokinins contained in its saliva (Brütting et al., 2018). Further insights into this insect and its ability to manipulate plant physiology for its own benefit would be vastly enabled with genomic resources that allow for deeper insights into the mechanistic underpinnings of its interactions with host plants. However, its small size presents difficulties for the generation of sequencing data to assemble a reference genome as it is not possible to extract enough DNA from a single individual for long‐read sequencing platforms.

More generally, the assembly of genomes for small species of insect remains challenging due to problems associated with using pooled samples of individuals to generate sequencing reads. Past generations of DNA sequencing data (i.e., short reads and low‐accuracy long reads) and assembly algorithms produce a single consensus sequence rather than phased haplotypes, and high heterozygosity can introduce errors to this process (Li et al., 2012). Newer sequencing data (i.e., PacBio HiFi reads and Oxford Nanopore Q20+ chemistry) and assembly algorithms are haplotype‐aware and designed to produce assemblies of separate haplotypes for single diploid organisms (Cheng et al., 2021). The presence of more than two haplotypes due to polyploidy or to the pooling of multiple individuals leads to the parallel assembly of multiple contigs (haplotigs) representing the same genomic region. Some species, such as many aphids, do not exhibit this problem due to the presence of an asexual stage in the life cycle, which produces a generation of genetically homogenous colonies in which multiple individuals can be pooled together without the risk of excess variation causing assembly errors (Davis, 2012). Recently, bioinformatic solutions have been developed that allow for the removal of duplicated haplotigs from heterozygous genomes (Guan et al., 2020). However, their efficacy for removing duplicated contigs from pooled‐sample assemblies has not yet been assessed.

In this study, we set out to produce a T. notatus draft genome assembly using sequencing data from a pooled sample of individuals. Using the purge_dups algorithm allowed us to generate a highly complete haploid genome assembly with most of the gene content represented, but with few allelic duplication errors. We further use this assembly as the basis for differential expression analyses to investigate its host plant‐specific transcriptional responses.

2. METHODS

2.1. Sample origins

In the fall of 2021, while growing D. wrightii for other studies, our greenhouse in Tucson, Arizona became infested with thousands of T. notatus. This small hemipteran herbivore specializes on plants with glandular trichomes, especially those in the Solanaceae (Van Dam & Hare, 1998). An originally small population likely entered the greenhouse without our knowledge sometime during the summer rainy season when insects are active in Southern Arizona and reached sufficient density to be noticed in the fall. Samples were collected via aspirator directly from D. wrightii plants in the greenhouse in December 2021 and January 2022. Tupiocoris notatus often co‐occurs with a similar‐looking predatory stiltbug (Hemiptera: Beritydae; Jay Goldberg personal observations), which we were careful to exclude during collections. All collections were immediately flash frozen with liquid nitrogen and stored at −80°C.

2.2. Nucleic acid extraction and sequencing

High‐molecular‐weight DNA was extracted from a single pooled sample of insects (0.5 mL, ~50 individuals) using a previously established chloroform:isoamyl phase separation protocol (Jaworski et al., 2020). HMW DNA was size checked by Femto Pulse System (Agilent), and 10 μg of DNA was sheared to appropriate size range (15–20 kb) using Megaruptor 3 (Diagenode). The sheared DNA was concentrated by bead purification using PB Beads (PacBio). The sequencing library was constructed following the manufacturer's protocols using SMRTbell Express Template Prep Kit 2.0. The final library was size selected on a Pippin HT (Sage Science) using S1 marker with a 10–25 kb size selection. The recovered final library was quantified with Qubit HS kit (Invitrogen) and sized on Femto Pulse System (Agilent). The sequencing library was sequenced with PacBio Sequel II Sequencing kit 2.0, loaded to one 8 M SMRT cell, and sequenced in CCS mode for 30 h. RNA was extracted from similar pooled‐samples (N = 3) using the ZYMO (Irvine, CA, USA) Direct‐zol RNA miniprep kit (Cat. # R2050) and sequenced using NovaSeq (Illumina, San Diego, CA, USA) paired‐end (150 bp) sequencing performed by Novogene (Sacramento, CA, USA). RNA libraries were prepared by Novogene following their standard mRNA‐seq services (polyA capture followed by cDNA reverse transcription).

2.3. Genome assembly and annotation

CCS output (i.e., HiFi reads; 3,583,689 reads; 17.69 Gb at mean Q35 score; mean length = 13,847 bp) were assembled using hifiasm‐0.16.0 (Cheng et al., 2021). The initial assembly had numerous duplicated allelic contigs (Table 1, Figure 1) and was subjected to two rounds of the standalone purge_dups algorithm (v1.2.6; Guan et al., 2020). Assemblies were visualized using Bandage v0.8.1 (Wick et al., 2015), which also provided contiguity statistics. Contigs assembled from contaminant reads were identified and filtered from our assembly using the blobtools v1.1 pipeline (Laetsch & Blaxter, 2017). Jellyfish v2.2.10 (Marcais & Kingsford, 2012) was used for k‐mer counting before using the GenomeScope2.0 web portal (Ranallo‐Benavidez et al., 2020) to estimate genome size (Figure S1). Polishing of the contaminant‐filtered assembly was carried out using Inspector v1.0.2 (Chen et al., 2021). Gene content completeness was assessed via BUSCO v5.4.7 (Seppey et al., 2019) using the hemipteran odb10 dataset (Figure 1, Table 2). Repeat content of the final (twice purged, contaminant filtered, and polished) assembly was assessed using RepeatMasker v4.1.3 (Tarailo‐Graovac & Chen, 2009; Table S2) before structurally annotating gene content with the Helixer v0.3.1 algorithm pipeline (Holst et al., 2023; Stiehler et al., 2021) using the pre‐made invertebrate training dataset. Functional annotation was done using InterProScan v5.45–80.0 (Jones et al., 2014) and blastp (using blast v2.13.0; Camacho et al., 2009) comparisons to the UniProt‐Swissprot database (Boutet et al., 2007). Functional annotation outputs were combined into a single gff using the manage_functional_annotation.pl script in the AGAT v 1.2.0 toolkit (Dainat et al., 2023).

TABLE 1.

Summary statistics of Tupiocoris notatus assemblies before and after haplotig purging.

Hifiasm assembly statistics
Total length (Mb) Largest contig (Mb) N 50 (kb) No. of contigs
Raw Assembly 1067.5 4.308 179 10,061
Purged Once 405.7 4.308 542 1310
Purged Twice 296.1 4.308 665 908
Final Assembly 291.8 4.308 675 886
Open in a new tab

FIGURE 1.

FIGURE 1

Open in a new tab

(a) Example of a properly assembled contig (left) and a y‐shaped contig – indicative of unresolved repetitive elements – as found at low frequencies (<1% of total contigs) in the “raw” pseudo‐haplotype assembly graphs (right). No y‐shaped contigs were present in the raw primary assembly or the final assembly. (b) Bar graphs showing the results of odb10 Hemiptera BUSCO analysis for the three of the T. notatus genome assemblies we produced and our structural annotation of the final assembly (bottom bar). Annotation assessment was conducted in proteome mode.

TABLE 2.

BUSCO scoring results of Tupiocoris notatus assemblies and final annotation produced by Helixer.

BUSCO scores (odb10 hemiptera; 2510 total genes)
Complete – total Single copy Duplicated (N ≥ 2) Multi‐duplicated (N ≥ 3) Fragmented Missing
Raw Assembly 2440 (97.2%) 254 (10.1%) 2186 (87.1%) 1658 (66.1%) 22 (0.9%) 48 (1.9%)
Purged Once 2424 (96.6%) 1569 (62.5%) 855 (34.1%) 112 (4.5%) 29 (1.2%) 57 (2.2%)
Purged Twice 2416 (96.3%) 2308 (92%) 108 (4.3%) 10 (0.44%) 32 (1.3%) 62 (2.4%)
Final Assembly 2398 (95.6%) 2288 (91.2%) 110 (4.4%) 11 (0.40%) 30 (1.2%) 82 (3.2%)
Helixer Annotation 2365 (94.2%) 2244 (89.4%) 121 (4.8%) 9 (0.36%) 47 (1.9%) 98 (3.9%)
Open in a new tab

2.4. Differential expression analysis

RNA‐seq reads were aligned to our genome assembly and counted using STAR (Dobin et al., 2013) using default settings after being trimmed with trimmomatic v0.39 (Bolger et al., 2014). Read counts were then analyzed using the DESeq2 package in R (Love et al., 2014; R Core Team, 2021). We reanalyzed the existing dataset of tobacco‐fed T. notatus transcriptomes (Crava et al., 2016; BioProject: PRJNA343704) and used all samples from their study as the baseline/control group (N = 6) for comparison to our Datura wrightii‐fed dataset (N = 3). The ClusterProfiler v4.0 package was used for gene set enrichment analysis (GSEA; Wu et al., 2021) of differentially expressed genes. GSEA was performed on the total set of genes for which InterProScan obtained GO‐terms. Each GO ontology (biological processes, molecular functions, and cellular component) was analyzed separately. We further separated each ontology into separate up‐ and down‐regulated gene lists as prior studies have found this approach to be more robust than grouping all differentially expressed genes (DEGs) together (Hong et al., 2014). We used a significance cutoff of p adj = .05 for all gene‐wise analyses without any fold‐change cutoff for differential expression.

3. RESULTS

3.1. Genome assembly and annotation

The initial assembly was over 1 Gb in length (Table 1), making it highly duplicated (Figure 1) and far greater than the predicted size of 247 Mb. The first round of supplemental purging reduced this to 405 Mb (Table 1, Figure 1b), but left over 30% duplicated single‐copy orthologs (Figure 1b). A second round of purge_dups reduced this to a reasonable level (Tables 1, 2). The size of the twice purged assembly (296 Mb, Table 1) was closer to the predicted size (247 Mb, Figure S1). Structural annotation identified 16,067 genes and the protein dataset had 95.1% of BUSCO genes complete when compared to the hemiptera_odb10 reference dataset (Figure 1b; Table 2). Taxonomic identification via comparison to the nt database (Camacho et al., 2009) in blobtools found 22 low‐coverage contigs likely to originate from contaminant reads (Figure S2; Table 1). These were filtered out of our assembly before beginning downstream analyses. Quality assessment with Inspector yielded an initial QV‐score of 19.7, roughly 1500 structural errors, and a small‐scale error rate of 4132.5 errors per Mb. After polishing, the QV‐score increased to 21.7 and the small‐scale error rate was reduced to 116 per Mb. Structural errors were largely unchanged by the polishing process and slightly increased from 1513 to 1542. Nearly all raw reads (99%) were mapped back to the polished assembly for an average read depth of 60.1. Detailed outputs of Inspector analyses pre‐ and post‐polishing are given in Table S1. RepeatMasker found 34.98% of the final assembly to be composed of repetitive elements (6.35% retroelements, 1.09% DNA transposons, 2.09% rolling‐circle transposons, 24.27% unclassified; detailed output in Table S2). Helixer annotated 16,062 genes in our final assembly, ranging in size from 108 bp to 379 kb (mean = 11.9 kb). 13,875 of these genes were functionally annotated, including 8824 for which GO‐terms could be assigned. Detailed annotation statistics are found in Table S3. Overall, these results show that the quality of our assembly lags behind that of recent chromosome‐scale assemblies produced using combined long‐read sequencing and chromatin conformation capture technologies (e.g., Hi‐C; Wang et al., 2023) in terms of accuracy and contiguity.

3.2. Differential expression analysis

Alignment of RNA‐seq reads was consistent across samples. An average of 94.8% of raw reads were mapped to our assembly (min = 92.1%, max = 96.1%). Most reads (mean = 77.7%) were mapped uniquely, but many (mean = 17.1%) mapped to multiple loci. Few reads (mean = 0.6%) were thrown out due to excessive multi‐mapping. Most reads that were unused were too short for mapping (mean = 4.53%). No reads were found to have too many mismatches to be used. Full read‐mapping statistics can be found in Table S4. Only a small proportion of our annotated genes (N = 509) did not have any mapped reads.

The dataset used by Crava et al. (2016) used a de novo transcriptome approach to look at differentially expressed genes in T. notatus feeding on a transgenic line of wild tobacco (Nicotiana attenuata). The ability to produce jasmonic acid and induce defense production is compromised via RNAi silencing of the biosynthetic gene allene oxide cyclase (irAOC, N = 3; vs. empty vector controls, EV; N = 3). We reanalyzed their dataset using a genome‐guided approach enabled by our draft assembly (Figure 2, Table S5). Our principal component analysis (PCA) found that iAOC‐ and EV‐fed insects formed separate clusters (Figure 2a) indicating different patterns of gene expression in each sample set; however, when we examined these patterns gene‐by‐gene, we only found 11 significantly differentially expressed loci (8 down‐regulated, 3 up‐regulated; Figure 2b). None of these genes showed substantial levels of expression changes (Log2fold‐change <1). Six of these genes were functionally annotated (Table S5), but none belonged to the detoxification gene families (cytochrome P450s, glutathione‐S‐transferases, UDP‐glucuronosyltransferases) identified in their study (Crava et al., 2016). One down‐regulated gene in our dataset was associated with digestion of plant compounds and annotated as polygalacturonase (PGN1). The difference between our results could be due to the presence of split or non‐coding genes due to Trinity assembly errors (Freedman et al., 2021; Grabherr et al., 2011) in their de novo transcriptome dataset, which contained 42,610 putative genes – a far larger number that was annotated within our assembly. Given that Crava and colleagues further confirmed some of the differentially expressed genes in their dataset using real‐time PCR, it is likely that our stringent methodology introduced false negatives and was unable to detect some truly differentially expressed genes. It is also possible that the difference is an artifact of reference‐derived biases, as our genome was assembled from individuals originating in the Tucson (Arizona) area and their RNA‐seq data was collected from a population in Utah. The genetic variation of this species is not known, and it is possible that presence‐absence variation in gene content exists – should there be substantial population structuring – as it does in other herbivorous insects (Mongue & Kawahara, 2022). This is likely to only be the case if there is geographic variation in expression levels, as we did not observe substantial differences in mapping rates between the two datasets (N. attenuata, Mean = 95.15%; D. wrightii, Mean = 94.24%).

FIGURE 2.

FIGURE 2

Open in a new tab

(a) Results of PCA analysis from differential expression study that included only Nicotiana‐fed T. notatus from a previous study (Crava et al., 2016). (a) PCA analysis shows that the two treatments form distinct clusters when gene expression is viewed globally. (b) Volcano plot showing the p‐value associated with the “line” variable in differential expression analysis and the fold change of that gene. Two‐fold changes (|log2| = 1) in either direction are marked for reference, but this was not used as a testing cutoff. Genes that satisfied our cutoff for significance (p adj < .05) are shown in red (N = 50; Table S4), whereas NS genes are in black.

We found that the expression profiles of Datura‐ and Nicotiana‐fed insects were distinct (Figure 3a). This difference was driven by many significantly up‐ or down‐regulated genes (N total = 4982; N down = 2121; N up = 2861; Figure 3b, Table S2). The most drastic differences in expression had over 10‐fold changes in either direction (272 genes with |Log2FC| > 3.01). Six differentially expressed genes (DEGs) were functionally annotated as glutathione‐S‐transferases, all of which were found to be up‐regulated in Datura wrightii‐fed samples. Another seven DEGs were annotated as UDP‐glycosyltransferases, four of which were down‐regulated and the other three up‐regulated in Datura‐fed insects. 45 cytochrome P450s were identified as DEG and most of them (N = 29) were up‐regulated. Out of 22 differentially expressed serine proteases, 17 were found to be upregulated in our Datura‐fed samples. The full list of significantly differentially expressed genes, including fold‐change and adjusted p‐values, is in Table S6. Our findings are consistent with previous studies of transcriptomic responses of herbivores to host plant chemistry (Bock, 2016; Castañeda et al., 2009; Lin et al., 2022) and confirm the role of the aforementioned gene families in digestion/detoxification of plant‐derived compounds by T. notatus.

FIGURE 3.

FIGURE 3

Open in a new tab

(a) Results of PCA analysis from differential expression study. Nicotiana‐ and Datura‐fed T. notatus form distinct clusters. (b) Volcano plot showing the p‐value associated with the “plant” variable in differential expression analysis and the fold change of that gene. Four‐fold changes in either direction (|log2| = 2) are marked for reference, but this was not used as a testing cutoff. Genes that satisfied our cutoff for significance (p adj < .05) are shown in red (N = 4982; Table S3), whereas NS genes are in black.

To explore the transcriptional changes associated with host plant species beyond our handful of target genes, we used a gene set enrichment analysis to identify common themes in our set of DEGs. We found a total of 24 GO terms to be significantly enriched within our results (Figure 4, Table S7) and that these are associated with gene/protein expression, nutrient catabolism, and chemosensory functions. Digestive functions were predominantly up‐regulated, with the notable exception of aspartic‐type endopeptidases. Another notable finding is that gustatory perception is enriched in both up‐ and down‐regulated pathways, suggesting the presence of complex host‐plant associated changes to chemosensory pathways. Olfaction was found to only be enriched within up‐regulated pathways.

FIGURE 4.

FIGURE 4

Open in a new tab

Results of gene set enrichment analyses. 24 Go terms were found to be enriched in either up‐ or down‐regulated gene lists. X‐axes show the ratio of enriched genes versus the total count of genes sharing that GO term. Dot sizes represent the total number of genes sharing each GO term whereas dot colors represent the p‐value (adjusted for multiple tests) of each term.

4. DISCUSSION

Producing high‐quality reference genome assemblies for small insects remains challenging. In this study, we were able to produce a high‐quality draft assembly using PacBio HiFi reads from a pooled sample of genetically variable individuals. It is important to note that our assembly is not chromosome‐scale or error free, and should be not used as a reference in analyses such as studies of chromosomal rearrangements or other structural variants. However, the unique protein‐coding regions are well represented, indicating that our assembly is of sufficient quality for studies focused specifically on low‐copy gene content and function. Moreover, our assembly was far more contiguous and complete than genome assemblies produced using short‐read technologies. Repetitive element and gene content of our assembly is comparable to that of another mirid (Cyrtorhinus lividipennis) which was previously found to have a 345.75 Mb genome containing 14,644 protein‐coding genes and 31.1% repeat content (Bai et al., 2022). In its current state, it is a suitable reference for the gene content of this species. We were able to use this genome as the basis for multiple differential expression analyses and preliminary assessments of functional gene content. It will likely also serve as a suitable reference assembly for population genomic analyses and other read‐mapping based pipelines. This demonstrates the utility of pooled‐sample genome assemblies when working with small insects that present difficulties for extracting sufficient quantities of DNA from a single individual.

The first of our differential expression studies was a re‐analysis of a previously published dataset comparing wild tobacco plants (N. attenuata) with functional (empty vector control; EV) and compromised (via RNAi silencing of allene oxide synthase expression; irAOC) defense induction pathways. We found that few insect genes were differentially expressed between colonies fed on these two lines, and that none of the significant genes were strongly up‐ or down‐regulated. This indicates that jasmonate‐induced plant defenses may not play a role in interactions between plants and T. notatus, a stark contrast to interactions between N. attenuata and other herbivores, such as M. sexta, against which induced defenses have been shown to play a critical role (van Dam et al., 2000). This finding is also different from the study that generated these RNA‐seq data, which found dozens of significantly differentially expressed genes (Crava et al., 2016), and is likely due to the more conservative nature of our genome‐guided approach compared to the de novo transcriptome assembly used as a mapping reference in their analysis. Our pipeline may be so stringent, in fact, that it introduced the presence of false negatives by accounting for low sample sizes and adjusting for multiple statistical tests. Crava et al. (2016) were able to confirm that some of the differentially expressed genes they identified were indeed down‐regulated in irAOC‐fed insects, yet they did not appear to be significantly differentially expressed in our analysis.

In addition to our reanalysis of Crava et al.'s (2016) dataset, we also compared newly generated RNA‐Seq data from our D. wrightii‐fed greenhouse population to their N. attenuata‐fed data. Our pooled samples were collected and prepared in a similar fashion to theirs, although differences may be present due to the geography of the sampled populations as little is known about population structure in this species. Their data originated from a captive colony (maintained in Jena, Germany) started from individuals collected from a field site in Southwest Utah. Our samples were collected from a greenhouse infestation in Tucson, Arizona, roughly 615 km away from their field site. Nonetheless, we identified many putative detoxification, digestion, and chemosensory genes in our list of differentially expressed genes. This finding suggests that a substantial amount of the gene expression differences between our samples and Crava et al.'s (2016) is due to host plant‐specific responses. We consider our list of genes a suitable starting point for future studies into the genetic basis of interactions between this species and its toxic Solanaceous hosts. Future studies might examine expression differences in response to more controlled manipulation of specific plant defensive compounds, or tissue‐specific gene expression by T. notatus, to differentiate between genes involved with plant metabolism manipulation – which are likely to be expressed in salivary glands (Boulain et al., 2019) – from those involved with digestion/detoxification. Overall, our results suggest the presence of physiological differences between T. notatus feeding on Datura and Nicotiana. Many of these differences are related to perception, digestion, and detoxification of host plant‐derived compounds, but others could also be derived from population (Arizona vs. Utah) differences or responses to other factors (e.g., greenhouse conditions). Our findings nonetheless provide a valuable starting point for future targeted studies of differentially expressed genes with specific roles mediating host‐plant interactions.

In conclusion, we have demonstrated that by using a standard haplotig purging algorithm, high‐quality pooled‐sample genome assemblies of a single haplotype can be produced. We have demonstrated the utility of our assembly for RNA‐Seq read‐mapping based pipelines by conducting two genome‐guided differential expression studies. We identified differentially expressed genes associated with specific host‐plant interactions and provide an initial functional assessment of them, many of which belong to well‐known families of detoxification and digestion genes. Together, these findings show that pooled samples may be a viable option for researchers unable to sequence single individuals of their species of interest due to small size or other factors and provide a valuable starting point for future research into the interactions between specialist herbivores (Hemiptera: Miridae) and their host plants.

AUTHOR CONTRIBUTIONS

Jay K. Goldberg: Conceptualization (lead); data curation (lead); formal analysis (lead); funding acquisition (lead); investigation (lead); methodology (lead); project administration (lead); writing – original draft (lead); writing – review and editing (lead). Carson W. Allan: Methodology (supporting); writing – review and editing (supporting). Luciano M. Matzkin: Methodology (supporting); writing – review and editing (supporting). Dario Copetti: Methodology (supporting); writing – review and editing (supporting). Judith Bronstein: Methodology (supporting); writing – review and editing (supporting).

FUNDING INFORMATION

Funding was provided by a National Science Foundation post‐doctoral research fellowship to JKG (NBI‐2010772).

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest exists.

Supporting information

Figure S1.

ECE3-14-e10979-s004.png (209.2KB, png)

Figure S2.

ECE3-14-e10979-s009.tif (23MB, tif)

Table S1.

ECE3-14-e10979-s006.xlsx (9.5KB, xlsx)

Table S2.

ECE3-14-e10979-s002.txt (2.4KB, txt)

Table S3.

ECE3-14-e10979-s008.xlsx (11.5KB, xlsx)

Table S4.

ECE3-14-e10979-s001.xlsx (12KB, xlsx)

Table S5.

ECE3-14-e10979-s003.csv (4.2KB, csv)

Table S6.

ECE3-14-e10979-s005.csv (1.9MB, csv)

Table S7.

ECE3-14-e10979-s007.csv (27.9KB, csv)

ACKNOWLEDGMENTS

Computational analyses were conducted on the UArizona HPC, and we would like to thank the staff for maintaining and providing access to this resource. We would like to thank Alex Karnish and Beth‐Ann Hansen for taking care of greenhouse plants and allowing the collection of insects before eliminating the T. notatus infestation.

Goldberg, J. K. , Allan, C. W. , Copetti, D. , Matzkin, L. M. , & Bronstein, J. (2024). A pooled‐sample draft genome assembly provides insights into host plant‐specific transcriptional responses of a Solanaceae‐specializing pest, Tupiocoris notatus (Hemiptera: Miridae). Ecology and Evolution, 14, e10979. 10.1002/ece3.10979

DATA AVAILABILITY STATEMENT

Sequencing data, including raw reads and the final genome assembly, are deposited on NCBI under BioProject PRJNA971612. No new code or analyses were generated for this project, but shell and R scripts used to run existing programs/packages are located at https://github.com/caterpillar‐coevolution/Tupiocoris‐notatus‐genome‐project. Additional datasets not included as supplemental materials, such as the genome annotation files, can be found in the GitHub repository as well. Published data from Crava et al. (2016) are available under BioProject: PRJNA343704.

REFERENCES

  1. Bai, Y. , Shi, Z. , Zhou, W. , Wang, G. , Shi, X. , He, K. , Li, F. , & Zhu, Z.‐R. (2022). Chromosome‐level genome assembly of the mirid predator Cyrtorhinus lividipennis Reuter (Hemiptera: Miridae), an important natural enemy in the rice ecosystem. Molecular Ecology Resources, 22, 1086–1099. 10.1111/1755-0998.13516 [DOI] [PubMed] [Google Scholar]
  2. Bock, K. W. (2016). The UDP‐glycosyltransferase (UGT) superfamily expressed in humans, insects and plants: Animal‐plant arms‐race and co‐evolution. Biochemical Pharmacology, 99, 11–17. 10.1016/j.bcp.2015.10.001 [DOI] [PubMed] [Google Scholar]
  3. Bolger, A. M. , Lohse, M. , & Usadel, B. (2014). Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics, 30, 2114–2120. 10.1093/bioinformatics/btu170 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Boulain, H. , Legeai, F. , Jaquiéry, J. , Guy, E. , Morlière, S. , Simon, J.‐C. , & Sugio, A. (2019). Differential expression of candidate salivary effector genes in pea aphid biotypes with distinct host plant specificity. Frontiers in Plant Science, 10, 1301. 10.3389/fpls.2019.01301 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Boutet, E. , Lieberherr, D. , Tognolli, M. , Schneider, M. , & Bairoch, A. (2007). UniProtKB/Swiss‐Prot. Methods in Molecular Biology, 406, 89–112. 10.1007/978-1-59745-535-0_4 [DOI] [PubMed] [Google Scholar]
  6. Brütting, C. , Crava, C. M. , Schäfer, M. , Schuman, M. C. , Meldau, S. , Adam, N. , & Baldwin, I. T. (2018). Cytokinin transfer by a free‐living mirid to Nicotiana attenuata recapitulates a strategy of endophytic insects. eLife, 7, e36268. 10.7554/eLife.36268 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Camacho, C. , Coulouris, G. , Avagyan, V. , Ma, N. , Papadopoulos, J. , Bealer, K. , & Madden, T. L. (2009). BLAST+: Architecture and applications. BMC Bioinformatics, 10, 1–9. 10.1186/1471-2105-10-421 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Castañeda, L. E. , Figueroa, C. C. , Fuentes‐Contreras, E. , Niemeyer, H. M. , & Nespolo, R. F. (2009). Energetic costs of detoxification systems in herbivores feeding on chemically defended host plants: A correlational study in the grain aphid, Sitobion avenae. Journal of Experimental Biology, 212, 1185–1190. 10.1242/jeb.020990 [DOI] [PubMed] [Google Scholar]
  9. Chen, Y. , Zhang, Y. , Wang, A. Y. , Gao, M. , & Chong, Z. (2021). Accurate long‐read de novo assembly evaluation with inspector. Genome Biology, 22, 1–21. 10.1186/s13059-021-02527-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cheng, H. , Concepcion, G. T. , Feng, X. , Zhang, H. , & Li, H. (2021). Haplotype‐resolved de novo assembly using phased assembly graphs with hifiasm. Nature Methods, 18, 170–175. 10.1038/s41592-020-01056-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Crava, C. M. , Brütting, C. , & Baldwin, I. T. (2016). Transcriptome profiling reveals differential gene expression of detoxification enzymes in a hemimetabolous tobacco pest after feeding on jasmonate‐silenced Nicotiana attenuata plants. BMC Genomics, 17, 1005. 10.1186/s12864-016-3348-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dainat, J. , Hereñú, D. , Murray, D. K. D. , Davis, E. , Crouch, K. , Lucile, S. , Agostinho, N. , & Zollman, Z. (2023). NBISweden/AGAT: AGAT‐v1.2.0. 10.5281/zenodo.8178877 [DOI]
  13. Davis, G. K. (2012). Cyclical parthenogenesis and Viviparity in aphids as evolutionary novelties. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution, 318, 448–459. 10.1002/jez.b.22441 [DOI] [PubMed] [Google Scholar]
  14. Dobin, A. , Davis, C. A. , Schlesinger, F. , Drenkow, J. , Zaleski, C. , Jha, S. , Batut, P. , Chaisson, M. , & Gingeras, T. R. (2013). STAR: ultrafast universal RNA‐seq aligner. Bioinformatics, 29, 15–21. 10.1093/bioinformatics/bts635 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Freedman, A. H. , Clamp, M. , & Sackton, T. B. (2021). Error, noise and bias in de novo transcriptome assemblies. Molecular Ecology Resources, 21, 18–29. 10.1111/1755-0998.13156 [DOI] [PubMed] [Google Scholar]
  16. Goldberg, J. K. , Lively, C. M. , Sternlieb, S. R. , Pintel, G. , Hare, J. D. , Morrissey, M. B. , & Delph, L. F. (2020). Herbivore‐mediated negative frequency‐dependent selection underlies a trichome dimorphism in nature. Evolution Letters, 4, 83–90. 10.1002/evl3.157 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Grabherr, M. G. , Haas, B. J. , Yassour, M. , Levin, J. Z. , Thompson, D. A. , Amit, I. , Adiconis, X. , Fan, L. , Raychowdhury, R. , Zeng, Q. , Chen, Z. , Mauceli, E. , Hacohen, N. , Gnirke, A. , Rhind, N. , di Palma, F. , Birren, B. W. , Nusbaum, C. , Lindblad‐Toh, K. , … Regev, A. (2011). Full‐length transcriptome assembly from RNA‐Seq data without a reference genome. Nature Biotechnology, 29, 644–652. 10.1038/nbt.1883 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Guan, D. , McCarthy, S. A. , Wood, J. , Howe, K. , Wang, Y. , & Durbin, R. (2020). Identifying and removing haplotypic duplication in primary genome assemblies. Bioinformatics, 36, 2896–2898. 10.1093/bioinformatics/btaa025 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Halitschke, R. , Hamilton, J. G. , & Kessler, A. (2011). Herbivore‐specific elicitation of photosynthesis by mirid bug salivary secretions in the wild tobacco Nicotiana attenuata . New Phytologist, 191, 528–535. 10.1111/j.1469-8137.2011.03701.x [DOI] [PubMed] [Google Scholar]
  20. Heidel, A. J. , & Baldwin, I. T. (2004). Microarray analysis of salicylic acid‐ and jasmonic acid‐signalling in responses of Nicotiana attenuata to attack by insects from multiple feeding guilds. Plant, Cell & Environment, 27, 1362–1373. 10.1111/j.1365-3040.2004.01228.x [DOI] [Google Scholar]
  21. Holst, F. , Bolger, A. , Günther, C. , Maß, J. , Triesch, S. , Kindel, F. , Kiel, N. , Saadat, N. , Ebenhöh, O. , Usadel, B. , Schwacke, R. , Bolger, M. , Weber, A. P. M. , & Denton, A. K. (2023). Helixer–de novo prediction of primary eukaryotic gene models combining deep learning and a hidden Markov model (preprint). BioRxiv. 10.1101/2023.02.06.527280 [DOI] [Google Scholar]
  22. Hong, G. , Zhang, W. , Li, H. , Shen, X. , & Guo, Z. (2014). Separate enrichment analysis of pathways for up‐ and downregulated genes. Journal of the Royal Society Interface, 11, 20130950. 10.1098/rsif.2013.0950 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jaworski, C. C. , Allan, C. W. , & Matzkin, L. M. (2020). Chromosome‐level hybrid de novo genome assemblies as an attainable option for nonmodel insects. Molecular Ecology Resources, 20, 1277–1293. 10.1111/1755-0998.13176 [DOI] [PubMed] [Google Scholar]
  24. Jones, P. , Binns, D. , Chang, H.‐Y. , Fraser, M. , Li, W. , McAnulla, C. , McWilliam, H. , Maslen, J. , Mitchell, A. , Nuka, G. , Pesseat, S. , Quinn, A. F. , Sangrador‐Vegas, A. , Scheremetjew, M. , Yong, S.‐Y. , Lopez, R. , & Hunter, S. (2014). InterProScan 5: Genome‐scale protein function classification. Bioinformatics, 30, 1236–1240. 10.1093/bioinformatics/btu031 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kessler, A. , & Baldwin, T. (2004). Herbivore‐induced plant vaccination. Part I. The orchestration of plant defenses in nature and their fitness consequences in the wild tobacco Nicotiana attenuata . The Plant Journal, 38, 639–649. 10.1111/j.1365-313X.2004.02076.x [DOI] [PubMed] [Google Scholar]
  26. Laetsch, D. R. , & Blaxter, M. L. (2017). BlobTools: Interrogation of genome assemblies. F1000Research, 6, 1287. 10.12688/f1000research.12232.1 [DOI] [Google Scholar]
  27. Li, Z. , Chen, Y. , Mu, D. , Yuan, J. , Shi, Y. , Zhang, H. , Gan, J. , Li, N. , Hu, X. , Liu, B. , Yang, B. , & Fan, W. (2012). Comparison of the two major classes of assembly algorithms: Overlap–layout–consensus and de‐bruijn‐graph. Briefings in Functional Genomics, 11, 25–37. 10.1093/bfgp/elr035 [DOI] [PubMed] [Google Scholar]
  28. Lin, R. , Yang, M. , & Yao, B. (2022). The phylogenetic and evolutionary analyses of detoxification gene families in Aphidinae species. PLoS One, 17, e0263462. 10.1371/journal.pone.0263462 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Love, M. I. , Huber, W. , & Anders, S. (2014). Moderated estimation of fold change and dispersion for RNA‐seq data with DESeq2. Genome Biology, 15, 1–21. 10.1186/s13059-014-0550-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Marcais, G. , & Kingsford, C. (2012). Jellyfish: A fast k‐mer counter.
  31. Mongue, A. J. , & Kawahara, A. Y. (2022). Population differentiation and structural variation in the Manduca sexta genome across the United States. G3: Genes, Genomes, Genetics, 12, jkac047. 10.1093/g3journal/jkac047 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Ranallo‐Benavidez, T. R. , Jaron, K. S. , & Schatz, M. C. (2020). GenomeScope 2.0 and Smudgeplot for reference‐free profiling of polyploid genomes. Nature Communications, 11, 1432. 10.1038/s41467-020-14998-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. R Core Team (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R‐project.org/ [Google Scholar]
  34. Seppey, M. , Manni, M. , & Zdobnov, E. M. (2019). BUSCO: Assessing genome assembly and annotation completeness. In Kollmar M. (Ed.), Gene prediction: Methods and protocols, methods in molecular biology (pp. 227–245). Springer. 10.1007/978-1-4939-9173-0_14 [DOI] [PubMed] [Google Scholar]
  35. Stiehler, F. , Steinborn, M. , Scholz, S. , Dey, D. , Weber, A. P. M. , & Denton, A. K. (2021). Helixer: Cross‐species gene annotation of large eukaryotic genomes using deep learning. Bioinformatics, 36, 5291–5298. 10.1093/bioinformatics/btaa1044 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Tarailo‐Graovac, M. , & Chen, N. (2009). Using RepeatMasker to identify repetitive elements in genomic sequences. Current Protocols in Bioinformatics, 25, 4.10.1–4.10.14. 10.1002/0471250953.bi0410s25 [DOI] [PubMed] [Google Scholar]
  37. van Dam, N. M. , Hadwich, K. , & Baldwin, I. T. (2000). Induced responses in Nicotiana attenuata affect behavior and growth of the specialist herbivore Manduca sexta. Oecologia, 122, 371–379. 10.1007/s004420050043 [DOI] [PubMed] [Google Scholar]
  38. Van Dam, N. M. , & Hare, D. J. (1998). Differences in distribution and performance of two sap‐sucking herbivores on glandular and non‐glandular Datura wrightii . Ecological Entomology, 23, 22–32. 10.1046/j.1365-2311.1998.00110.x [DOI] [Google Scholar]
  39. Wang, Z. , Huang, S. , Yang, Z. , Lai, J. , Gao, X. , & Shi, J. (2023). A high‐quality, phased genome assembly of broomcorn milet reveals the features of its subgenome evolution and 3D chromatin organization. Plant Communications, 4, 100557. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Wick, R. R. , Schultz, M. B. , Zobel, J. , & Holt, K. E. (2015). Bandage: Interactive visualization of de novo genome assemblies. Bioinformatics, 31, 3350–3352. 10.1093/bioinformatics/btv383 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Wu, T. , Hu, E. , Xu, S. , Chen, M. , Guo, P. , Dai, Z. , Feng, T. , Zhou, L. , Tang, W. , Zhan, L. , Fu, X. , Liu, S. , Bo, X. , & Yu, G. (2021). clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. The Innovations, 2, 100141. 10.1016/j.xinn.2021.100141 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Figure S1.

ECE3-14-e10979-s004.png (209.2KB, png)

Figure S2.

ECE3-14-e10979-s009.tif (23MB, tif)

Table S1.

ECE3-14-e10979-s006.xlsx (9.5KB, xlsx)

Table S2.

ECE3-14-e10979-s002.txt (2.4KB, txt)

Table S3.

ECE3-14-e10979-s008.xlsx (11.5KB, xlsx)

Table S4.

ECE3-14-e10979-s001.xlsx (12KB, xlsx)

Table S5.

ECE3-14-e10979-s003.csv (4.2KB, csv)

Table S6.

ECE3-14-e10979-s005.csv (1.9MB, csv)

Table S7.

ECE3-14-e10979-s007.csv (27.9KB, csv)

Data Availability Statement

Sequencing data, including raw reads and the final genome assembly, are deposited on NCBI under BioProject PRJNA971612. No new code or analyses were generated for this project, but shell and R scripts used to run existing programs/packages are located at https://github.com/caterpillar‐coevolution/Tupiocoris‐notatus‐genome‐project. Additional datasets not included as supplemental materials, such as the genome annotation files, can be found in the GitHub repository as well. Published data from Crava et al. (2016) are available under BioProject: PRJNA343704.

Articles from Ecology and Evolution are provided here courtesy of Wiley

RESOURCES