Analysis of Haemonchus embryos at single cell resolution identifies two eukaryotic elongation factors as intervention target candidates

Abstract

Advances in single cell technologies are allowing investigations of a wide range of biological processes and pathways in animals, such as the multicellular model organism Caenorhabditis elegans – a free-living nematode. However, there has been limited application of such technology to related parasitic nematodes which cause major diseases of humans and animals worldwide. With no vaccines against the vast majority of parasitic nematodes and treatment failures due to drug resistance or inefficacy, new intervention targets are urgently needed, preferably informed by a deep understanding of these nematodes’ cellular and molecular biology – which is presently lacking for most worms. Here, we created the first single cell atlas for an early developmental stage of Haemonchus contortus – a highly pathogenic, C. elegans-related parasitic nematode. We obtained and curated RNA sequence (snRNA-seq) data from single nuclei from embryonating eggs of H. contortus (150,000 droplets), and selected high-quality transcriptomic data for > 14,000 single nuclei for analysis, and identified 19 distinct clusters of cells. Guided by comparative analyses with C. elegans, we were able to reproducibly assign seven cell clusters to body wall muscle, hypodermis, neuronal, intestinal or seam cells, and identified eight genes that were transcribed in all cell clusters/types, three of which were inferred to be essential in H. contortus. Two of these genes (i.e. Hc-eef-1A and Hc-eef1G), coding for eukaryotic elongation factors (called Hc-eEF1A and Hc-eEF1G), were also demonstrated to be transcribed and expressed in all key developmental stages of H. contortus. Together with these findings, sequence- and structure-based comparative analyses indicated the potential of Hc-eEF1A and/or Hc-eEF1G as intervention targets within the protein biosynthesis machinery of H. contortus. Future work will focus on single cell studies of all key developmental stages and tissues of H. contortus, and on evaluating the suitability of the two elongation factor proteins as drug targets in H. contortus and related nematodes, with a view to finding new nematocidal drug candidates.

Graphical Abstract

1. Introduction

Since the 1960s, the free-living roundworm (= nematode) Caenorhabditis elegans has been used as a model organism [1], [2], [3] to extensively study many aspects of the biology of multicellular organisms generally (https://www.wormatlas.org/; http://www.wormbook.org/). This soil nematode has offered major benefits as a model, particularly because of its short (3-day) life cycle, small size (1.5 mm), mode of reproduction (selfing), and relative ease of cultivation and ‘manipulation’ in laboratory. Other key features of this nematode are its anatomical simplicity (∼ 1000 cells), including the 302-cell hermaphrodite nervous system, and its small genome (100 Mb). Complemented by the advent of cutting-edge genomic, biochemical, transcriptomic, proteomic, informatic and imaging tools, the sequencing and subsequent annotation of the C. elegans genome and availability of extensive web-based resources, such as WormBase [4], [5], have accelerated fundamental explorations using this model organism. Some of latest technologies have revolutionised discovery, and allowed the deconvolution of biological processes and pathways to the tissue and single-cell levels (e.g., [6], [7], [8]). In particular, the application of single cell (sc) and/or single nucleus (sn) RNA-sequencing technologies have allowed the creation of single cell atlases for C. elegans, providing unprecedented, ‘granular’ insights into aspects such as embryogenesis, cell differentiation, types, fate and plasticity, and neurobiology [9], [10], [11], [12], [13], and also providing a solid foundation for comparative studies of the cellular and molecular biology of related organisms (e.g., [14], [15], [16]).

Caenorhabditis elegans is in the same evolutionary clade (V) as one of the largest groups of socioeconomically important parasitic nematodes of humans and animals, including the pathogenic, blood-feeding barber’s pole worm – Haemonchus contortus [17]. Consequently, much molecular research of H. contortus has relied heavily on knowledge, data sets and information available for C. elegans [18], [19], [20]. Particularly over the past decade, intensive genomic, transcriptomic and proteomic investigations have enabled the creation of substantially improved resources and tools for H. contortus – which has now been elevated to model organism status [21], [22].

Despite these advances, there are still major knowledge gaps in many fundamental areas of H. contortus anatomy, biology and development at the cellular level, limiting our understanding of this parasite and parasitism. As a first step to gaining insights into H. contortus at this level, we conducted the first single-nucleus RNA-seq analysis of H. contortus at the stage of embryonation, as a proof-of-principle, to infer essential genes based on comparative analysis with C. elegans, as a foundation for detailed molecular investigations of the life history of this and related parasitic nematodes at the cellular level in the future.

2. Materials and methods

2.1. Production and purification of embryonated Haemonchus contortus eggs

With animal ethics approval (no. 1714374) from the University of Melbourne, a parasite-free merino lamb (Ovis aries; male; 6 months of age) was inoculated orally with 8000 infective third-stage larvae (L3s) of H. contortus (Haecon 5 strain, Australia) as described previously [23]. One month after inoculation, faeces were collected for 1 h and processed. From 100 g of faeces from an infected sheep, H. contortus eggs (n = 200,000) were purified by sucrose flotation [24] at 24 °C within 4 h of collection. By this time, the majority (95%) of eggs were at various stages of embryonation – consistent with description by Veglia [25].

2.2. Isolation of nuclei

Subsequently, cell nuclei were isolated from these embryonated eggs using a modification of the Frankenstein protocol [26] In brief, 30,000 embryonated eggs were transferred to a 1.5 ml low-DNA binding Eppendorf tube containing 200 μl of ice-cold nuclei lysis buffer (Merck Millipore, USA), and homogenised with a pellet pestle (Sigma-Aldrich, USA; 20 back-and-forth rotations). The homogenate was then transferred to 2.0 ml low-DNA binding Eppendorf tube, after which 1.3 ml of ice-cold nuclei lysis buffer were added to the tube and incubated on ice for 5 min (gently mixing with a wide-bore 1 ml tip twice during the incubation). Following the incubation, the homogenate was filtered through a 40 µm nylon mesh sieve (Thomas Scientific, USA), and then centrifuged at 500 × g and 4˚C for 5 min. The supernatant was aspirated and discarded, leaving 70 μl in the tube. The pellet of cell nuclei was resuspended in 1.5 ml of ice-cold nucleus lysis buffer and incubated on ice for 5 min. This homogenate was centrifuged at 500 × g and 4˚C for 5 min and resuspended in 1.5 ml of ice-cold nuclei wash buffer (phosphate-buffered saline [pH 7.0] containing 1% bovine serum albumin [BSA] and 0.2 U/μl RNase inhibitor); this step was repeated once and the supernatant removed. Subsequently, the pellet of nuclei was resuspended in 500 μl of ice-cold nuclei wash buffer with 10 µg/ml DAPI (4,6-diamidino-2-phenylindole; Sigma-Aldrich, USA) and an aliquot (5 μl) thereof examined at 300-times magnification using a fluorescence microscope (ECLIPSE Ti2 Nikon, Japan) to appraise the shape and size of the cell nuclei.

2.3. Purification of nuclei, library construction, sequencing and pre-processing of data

First, fluorescence-activated cell sorting (FACS) was used to separate nuclei from cell debris. Then, the nuclei were then encapsulated into droplets, cDNA libraries constructed using the standard 10x Genomics (Chromium) protocol, and sequencing (5’-end of mRNA) was conducted using the Illumina HiSeq 2500 system. Raw sequence data were assessed for quality, mapped against the reference genome of H. contortus (version WBPS17; ref. [21] and transcript-count matrix created using Cell Ranger v7.1.0 software [27]. Read counts in the matrix were processed further using the programs R v4.0.3 [28] and Seurat v3.9.9 [29]; genes with a transcript read-count of < 3 and cells with a gene count of < 50 were removed; data from droplets containing ≥ 2 nuclei were eliminated using the program scds v1.6.0 [30]; data from droplets with transcripts inferred to be ‘background’ were eliminated based on an analysis of 2000 genes with the most variable transcription levels among all cells using the program DIEM v2.4.1 [31]; only cells containing ≥ 200 transcribed genes were retained.

2.4. Data normalisation, clustering, annotation of cell types and other bioinformatic analyses

The curated snRNA-seq read counts obtained were normalised using the method SCTransform [32] and clustered using the Seurat software package. Marker-genes for individual clusters were identified based on their transcription profile employing the program Scran v1.18.7 [33]. Clusters were illustrated using the programs Seurat and UMAP [34], [35]. Cell clusters were inferred based on a comparative analysis of transcription profiles obtained from a single cell data set (GSE126954) for C. elegans [10]. In short, orthologous genes were identified by reciprocal BlastP analysis of the proteomes of C. elegans (version WBcel235; NCBI) and H. contortus (version WBP17; WormBase). In the C. elegans dataset, cell types were identified from column "cell_type1" using the program SingleCellExperiment [36], and, using exclusively orthologous genes, cell-types were inferred for the H. contortus data set using the program clustifyr v1.2.0 [37]. This approach was taken because cell types have not yet been defined in H. contortus. In addition, we performed gene ontology (GO) enrichments for the genes linked to individual cell-type clusters using the program topGO v2.42.0 [38], in which a weighted Fisher’s exact test is applied.

For individual genes transcribed at high levels in all cell clusters, genetic, physical, regulatory and/or predicted interaction networks, and their essentiality were explored with WormBase. The essentiality of genes was inferred based on RNAi-based gene knockdown in C. elegans (see WormBase) and transcription in all clusters (cf. [39]); essentiality was linked to an embryonic lethal (Emb) or larval lethal (Lvl) phenotype. Interaction networks and expression/transcription information were accessed from WormBase (15 March 2023). This approach was taken, because no large-scale functional genomic data set is available for H. contortus and RNAi does not always work reproducibly or effectively in this species [40], likely due to the absence of some components (i.e. RDE-2 or SID-2) of the RNAi machinery [21], [23].

Transcription levels of RNA-seq data represent distinct developmental stages/sexes (i.e. egg, L1, L2, L3, L4 female, L4 male, adult female and adult male) were extracted from previously published data (ref. [23]; SRP026668). Sequence reads were cleaned and trimmed using the software Trimmomatic v0.39 [41] and curated reads were mapped to all coding sequences of H. contortus (ref. [42]; WormBase ParaSite version 18) using the program RSEM v1.3.3 [43]. Resultant transcripts per million (TPM) were then used to indicate the transcription level of essential genes in the distinct developmental stages of H. contortus.

The structures of orthologous proteins encoded by selected genes were predicted using the program AlphaFold2 [44], and the confidence of domains, regions and/or overall structures assessed using the predicted local distance difference test (pLDDT) values, with the per-residue estimate of confidence given on a scale from 0–100. Evolutionary conservation was assessed at each amino acid position in a multiple sequence alignment of orthologous proteins using the program Capra07 [45]; the conservation score represented the similarity between two probability distributions (Jensen–Shannon divergence). For the phylogenetic analysis, orthologous protein sequences were selected using the program OrthoMCL v2.0.4 [46], aligned using the program MAFFT v7.407 [47] and the aligned sequences subjected to separate analyses using the programs RAxML v8.2.9 [48] and MrBayes v3.2.6 [49], in which the LG amino acid substitution model (with gamma-distribution, G, and proportion of invariable sites, F; ref. [50]) was employed.

3. Results

3.1. Single cell atlas based on an analysis of snRNA-seq data

Raw snRNA-seq data (44.6 Gb) and read-index data (5.0 Gb) (Table S1) were transferred into a droplet × feature (gene) read-count matrix. Of the sequence reads in the 150,000 droplets selected, 87.1% mapped to the H. contortus genome (including introns), and 64.2% to gene regions including introns (version WBPS17); 96.5% of the reads were within droplets; the median number of genes per droplet was 280; in total, 17,101 genes were represented (Table S2). After the removal of data derived from droplets with multiple or no nuclei, the remaining data from 14,128 droplets (each containing one nucleus) and 13,687 genes that were transcribed in ≥ 3 nuclei, were subjected to further analysis. Using transcripts from individual nuclei (Table 1; Fig. 1a), we inferred 19 distinct cell clusters (Table S3), 17 of which had marker genes (p < 0.001) (Table 1 and S4). We inferred orthologous proteins (n = 7913) between the H. contortus and C. elegans proteomes, and used 6797 of these orthologs to annotate the clusters (Table S5). Guided by C. elegans single cell data (GSE126954), we were able to reproducibly assign seven individual cell clusters to body wall muscle (cluster no. 5), hypodermis (6), neurons (2, 12, and 14), intestine (15) or seam cells (16) (Table 1; Fig. 1a).

Table 1.

Nineteen cell clusters inferred based on the analysis of single nucleus (sn)RNA-sequence data from Haemonchus contortus embryos, annotatable cell types and respective marker genes (cf. Fig. 1).

Cluster number	Number of individual nuclei	Cell type (cluster annotation)	Marker genes (top-five; p < 0.001)
1	2071	-	-
2	1591	Neurons	HCON_00176790
3	1425	-	HCON_00176310; HCON_00173800; HCON_00173380; HCON_00177780 HCON_00176280
4	1256	-	HCON_00094700
5	1068	Body wall muscle	HCON_00181850; HCON_00174260; HCON_00173460; HCON_00013360 HCON_00176775
6	784	Hypodermis	HCON_00036640; HCON_00144180; HCON_00092510; HCON_00032040 HCON_00075270
7	724	-	HCON_00089225; HCON_00176775; HCON_00186780; HCON_00091470 HCON_00050940
8	692	-	HCON_00160300; HCON_00127710
9	689	-	HCON_00096110; HCON_00034760; HCON_00013670; HCON_00049570 HCON_00078730
10	617	-	HCON_00185590; HCON_00103410; HCON_00056730; HCON_00050260 HCON_00001240
11	557	-	HCON_00190170; HCON_00190180; HCON_00027070; HCON_00150565 HCON_00104060
12	508	Neurons	HCON_00179900; HCON_00009880; HCON_00164070; HCON_00115910 HCON_00172500
13	422	-	HCON_00023690; HCON_00061920; HCON_00067190; HCON_00108700 HCON_00099280
14	376	Neurons	HCON_00017720; HCON_00017700
15	322	Intestine	HCON_00178645; HCON_00160690; HCON_00073380; HCON_00107430 HCON_00148120
16	317	Seam cells	HCON_00143140; HCON_00187800; HCON_00089910; HCON_00187910 HCON_00187890
17	287	-	-
18	273	-	HCON_00176800; HCON_00091128; HCON_00091125; HCON_00036620 HCON_00172840
19	149	-	HCON_00190170

Fig. 1.

Clusters of cell-types (n = 19; Table 1) based on the analysis of single nucleus (sn)RNA-sequence data from Haemonchus contortus eggs during embryonation (a) the annotation of five distinct cell types of H. contortus by comparative analysis with a reference single cell data set (GSE126954) for Caenorhabditis elegans[10] (b). Levels of transcription (given in transcripts per million, TPM) in eight distinct stages/sexes of H. contortus (i.e. E = egg; L1, L2, L3 and L4 = four distinct larval stages; L4f = female L4; L4m = male L4; Af = female adult; Am = male adult) of eight protein-coding genes represented in all 19 clusters of cell-types inferred for H. contortus embryos (cf. Table 2) (c).

GO enrichments for genes linked to the 19 clusters inferred for H. contortus revealed (i) molecular functions including transcription, translation, regulation and binding activities; (ii) cellular processes including metabolism, RNA splicing, energy and transport; and (iii) cellular compartments such as nucleus, organelle, protein and complexes in most clusters, but no identifiable cluster-specific enrichment(s) (Tables S6, S7 and S8).

3.2. Analysis of genes transcribed among cell clusters reveals a subset of genes/proteins with key roles in the mitochondrion, RNA biology and protein synthesis

Subsequently, we enumerated and explored transcribed genes that were common among ≥ 3 or all 19 clusters. Represented in ≥ 3 clusters were 1703 H. contortus genes (including 229 orphans), 1132 of which had known (n = 889) or orphan (n = 243) orthologs in C. elegans (Table S9). Represented in all 19 clusters were eight H. contortus genes encoding proteins matching C. elegans homologs (Table 2 and S10): an ADP, ATP carrier protein encoded by Hc-ant-1 (HCON_00082980); five RNA recognition motif (RRM) domain-containing protein homologs represented by Hc-rbm-3 (HCON_00018960, HCON_00030820, HCON_00030825, HCON_00106740 and HCON_00133560); a eukaryotic elongation factor 1-alpha encoded by Hc-eef-1A (HCON_00071960); and a eukaryotic elongation factor 1-gamma encoded by Hc-eef-1 G (HCON_00147420). Three of these genes – Hc-ant-1, Hc-eef-1A and Hc-eef-1 G – are classified as essential, with an Emb (lethal) phenotype resulting directly from RNAi knockdown in C. elegans [51], [52], [53], [54], [55], [56], [57].

Table 2.

Eight protein-coding genes transcribed in all 19 cell clusters representing Haemonchus contortus eggs during embryonation; their names, annotations, essentiality status based on information for Caenorhabditis elegans, and WormBase accession codes for orthologous genes in C. elegans.

Protein-coding gene in H. contortus	Assigned gene name	Annotation of encoded protein	Essential	WormBase accession (C. elegans gene homolog)
HCON_00082980	Hc-ant-1	ADP, ATP carrier protein	Yes	WBGene00006439 (ant-1.1), WBGene00007057 (ant-1.2), WBGene00010485 (ant-1.3), WBGene00020140 (ant-1.4)
HCON_00071960	Hc-eef-1A	Translation elongation factor 1-alpha	Yes	WBGene00001168 (eef-1A.1), WBGene00001169 (eef-1A.2)
HCON_00147420	Hc-eef-1G	Translation elongation factor 1-gamma	Yes	WBGene00008920 (eef-1G; three isoforms in C. elegans)
HCON_00018960	Hc-rbm-3	RRM domain-containing proteina	No	WBGene00011155 (rbm-3.1), WBGene00011156 (rbm-3.2), WBGene00011059 (R06C1.4)
HCON_00030820	Hc-rbm-3	RRM domain-containing protein	No	WBGene00011155 (rbm-3.1), WBGene00011156 (rbm-3.2), WBGene00011059 (R06C1.4)
HCON_00030825	Hc-rbm-3	RRM domain-containing protein	No	WBGene00011155 (rbm-3.1), WBGene00011156 (rbm-3.2), WBGene00011059 (R06C1.4)
HCON_00106740	Hc-rbm-3	RRM domain-containing protein	No	WBGene00011155 (rbm-3.1), WBGene00011156 (rbm-3.2), WBGene00011059 (R06C1.4)
HCON_00133560	Hc-rbm-3	RRM domain-containing protein	No	WBGene00011155 (rbm-3.1), WBGene00011156 (rbm-3.2), WBGene00011059 (R06C1.4)

^aRRM = RNA recognition motif.

Hc-ant-1 encodes an ANT-1 which has a higher amino acid sequence identity (88.6%) to C. elegans ANT-1.1 than to ANT-1.2 to ANT-1.4 (80.3–84.3%). The transcription profile seen here for Hc-ant-1 (Fig. 1) is consistent with that observed previously [58], with expression present in all key developmental stages, and being higher in adult females than males of H. contortus (Fig. 1c and Table S10; cf. [58]). Current evidence indicates that ANT-1.1 is located in the mitochondrion and inferred to enable ATP:ADP antiporter activity (ADP/ATP exchange between cytosol and mitochondria); it is critical for mitochondrion organisation, is required for embryogenesis and normal postembryonic body size and adult life-span, and positively regulates the apoptotic process. Previous work on transgenic lines of C. elegans containing GFP-constructs (5′-UTRs plus a region of exon 1 representing each Hc-ant-1 and C. elegans ant-1.1 [58] showed consistent expression in pharynx, body wall, tail muscle, body muscle, hypodermis and intestine, with slight variation in expression among individual worms and tissues for Hc-ant-1. The differential transcription seen between adult female and male worms of H. contortus (Fig. 1; cf. [58]) supports a key role in oogenesis, in addition to embryogenesis.

Current evidence also shows that the ant-1.1 gene has a vast network of at least 95 interactors – 41 of which are physical (protein-protein), 8 genetic (suppressive), 4 regulatory and 42 predicted – and 372 co-transcribed molecules [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], [70], [71], [72], [73], [74], [75], [76], [77], [78], [79], [80] (via WormBase – 15 March 2023). Interestingly, Shen et al. [62] identified ANT-1.1 (previously referred to as ‘WAN-1’) as an important cell death regulator. Genetic inactivation of ant-1.1 significantly suppressed both somatic and germ line cell deaths in C. elegans, the chemical inhibition of ANT-1.1 activity caused major reduction in germ line apoptosis. In the mitochondria, ANT-1.1 can form a complex with both CED-4 and CED-9, the cell death initiator EGL-1 can disrupt its interaction with CED-9 [62]. These findings suggest that ANT-1.1 (also as a dimer or oligomer) is involved in the cell-killing process, and works with the core cell death machinery to promote programmed cell death in C. elegans.

Based on evidence for C. elegans, the five RRM domain-containing proteins identified are proposed to be part of the mRNA cleavage and polyadenylation specificity factor (CPSF) complex [81], [82], which enable mRNA binding activity and pre-mRNA processing required for polyadenylation. The rbm-3.2 gene, for example, has genetic interactions with spn-4 (predicted to be involved in nervous system development and the regulation of alternative mRNA splicing, via the spliceosome) and dhc-1 (predicted to enable dynein intermediate chain-binding activity) (cf. WormBase). Collectively, these proteins are part of the cleavage–polyadenylation step required for the maturation of primary protein-encoding transcripts into functional mRNAs, which can be exported from the nucleus to the cytoplasm for translation; the 3′-end processing depends on the binding of multi-protein complexes to specific elements at the 3′-end of the pre-mRNA [83].

An analysis of function in C. elegans indicated that both proteins Hc-eEF1A and Hc-eEF1G are linked to a key step of protein synthesis (i.e. translation elongation) in H. contortus (cf. Fig. 2), with eef-1A.1 and eef1G being essential and linked to knockdown phenotypes including Emb and/or Lvl upon RNAi [51], [52], [54], [56]. In the free-living nematode, eEF1A.1 is located in the cytoplasm, is expressed predominantly in hypodermal, intestinal, muscle and neuronal cells, and promotes the GTP-dependent binding of aminoacyl-tRNA to the A-site of ribosomes during protein biosynthesis. We observed that eef1A.1 has an extensive network of at least 119 interactions – 59 of which are regulatory, 31 physical (protein-protein), one genetic and 28 predicted – and associates with hundreds of co-transcribed genes [60], [61], [69], [70], [71], [75], [84], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96], [97], [98] (via WormBase – 15 March 2023). Subsequently, we predicted the structure of the protein encoded by HCON_00071960 using AlphaFold2, with high confidence (pLDDT > 70) for most of this protein, including tr-type G domain (IPR031157), and limited confidence (pLDDT≤70) for the very N- and C-termini and the ‘disordered’ region (see Fig. 3). A comparison of this structure with that of an ortholog from a representative host animal, Ovis aries (GenBank identifier KAG5196448.1), revealed minor structural variation (TM score: 0.92) (Fig. 3).

Fig. 2.

A schematic representation of the translation elongation process in metazoans – indicating the involvement of eukaryotic elongation factors eEF1A and eEF1G. The active form of eEF1A complexed with GTP delivers an aminoacylated tRNA to the A-site of the ribosome. Upon codon-anticodon recognition, the GDP is hydrolysed, and the inactive eEF1A-GDP is released from the ribosome and is then bound by the eEF1B2GD complex, which forms a macromolecular protein aggregate, designated eEF1H (comprising the subunits eEF1B2, eEF1G and eEF1D). This complex promotes the exchange between GDP and GTP to regenerate the active eEF1A.

Image adapted from Li et al. [116].

Fig. 3.

Three-dimensional structure models for eukaryotic elongation factor proteins Hc-eEF1A and Hc-eEF1G, inferred from respective Haemonchus contortus genes HCON_00071960 (Hc-eef-1A) and HCON_00147420 (Hc-eef-1 G) (cf. Table 2), and a comparison of each of these proteins with their respective homolog/ortholog from a representative host animal – Ovis aries (accession nos. XP_042085813.1 and XP_027815168.1), or the free-living nematode – Caenorhabditis elegans (accession nos. F17C11.9 and R03G5.1). Blue (high), yellow (moderate) and red (low) indicate the confidence of the structure prediction (left), or the conservation (Capra07) upon comparison (mid and right). Template modelling (TM) scores for structural comparisons are indicated on the image.

In C. elegans, eEF1G is located within both nucleus and cytoplasm, and eef-1 G has a network of at least 75 interactions – 24 of which are regulatory (positive or negative), 27 are physical (protein-protein) 2 genetic (suppressive) and 22 predicted – and 41 co-transcribed genes [60], [61], [63], [69], [73], [75], [77], [79], [87], [95], [99], [100], [101], [102], [103], [104], [105], [106], [107]. Although eef-1 G is recognised as an essential gene in many organisms, including mammals, meaning that its function is crucial for the survival of an organism [20], the sequence and structural conservation of its gene product (eEF1G) had not been studied. Here, we investigated sequence and structural variation in eEF1G between H. contortus and representative invertebrate and vertebrate animals, to explore whether HCON_00147420, or parts thereof, might have potential as an intervention target.

Hc-eEF1G, encoded by HCON_00147420, has distinct GST C-terminal (IPR010987; amino acid positions 73–205) and an elongation factor 1B-gamma C-terminal (IPR001662; positions 253–413) domains as well as an intervening, ‘disordered’ region (with compositional bias; positions 206–252). The AlphaFold2-predicted structure of the N-terminal domain (positions 1–205) was markedly more confident (p = 0; Mann-Whitney one-tailed U-test) than the C-terminal domain (positions 253–413), and the predicted structure of the intervening region (positions 206–252) had low confidence (Fig. 3; Table S11 – with a median pLDDT value of 33). A comparison of the Hc-eEF1G sequence with an ortholog from a representative host animal, Ovis aries (GenBank accession KAG5196448.1), revealed more variation in the N-terminal (p = 0; Mann Whitney one-tailed U-test) than the C-terminal domain (Fig. 3; conservation scores given in Table S11). A subsequent phylogenetic analysis of aligned protein sequences of eEF1G orthologs from selected parasitic nematodes of the order Strongylida (including haematophagous worms – Haemonchus contortus, H. placei, Ancylostoma caninum and A. ceylanicum); free-living nematodes (including C. elegans, C. brenneri and Pristionchus pacificus); arthropods (including Drosophila melanogaster); and mammal host species (Ovis aries, Bos taurus, Homo sapiens and Canis lupus familaris) revealed a close relationship of Hc-eEF1G with full-length orthologs from (mainly clade V) nematodes (nodal support – pp = 1.00), to the exclusion of those from arthropods (pp = 1.00) and mammals (pp = 1.00) (Fig. 4; Table S11).

Fig. 4.

Phylogenetic tree showing the relationship of protein Hc-eEF1G, inferred from the Haemonchus contortus gene (HCON_00147420), with orthologous proteins from parasitic nematodes of the order Strongylida, including Ancylostoma caninum (WormBase ParaSite accession ANCCAN_22986), A. ceylanicum (WormBase ParaSite accession Acey_s0033.g1870.t2 and Acey_s0033.g1870.t3), Haemonchus placei (WormBase ParaSite accession HPLM_000530501-mRNA-1); free-living nematodes including C. elegans (WormBase accessions F17C11.9 – isoforms a to c), Caenorhabditis brenneri (WormBase accession CBN026132) and Pristionchus pacificus (WormBase ParaSite accession PPA10377); the model insect Drosophila melanogaster (NCBI accessions NP_652000.1, NP_001189308.1 and NP_733280.1); and mammalian host species of the strongylid nematodes, including Ovis aries (NCBI accession XP_027815168.1), Bos taurus (NCBI accession XP_001035577.1), Canis lupus familiaris (NCBI accession XP_853577.1) and Homo sapiens (NCBI accession NP_001395.1). Nodal support values given as posterior probability (MrBayes) and bootstrap percentage (RAxML).

4. Discussion

Parasitic nematodes cause some of the world’s most neglected diseases, affecting hundreds of millions of people and animals worldwide (e.g., [108], [109], [110], [111]). Despite efforts to control these nematodes, no highly effective, commercial vaccines are available for the majority of nematodes of animals, and treatment relies on only a small number of drug classes [112]. Even when using the native protein vaccine (Barbervax®), additional anthelmintic treatment is required to reduce infection intensity or achieve control [113]. Widespread resistance to most of these compounds seriously compromises the control of parasitic worms, particularly in livestock animals [114]. Consequently, there is an urgent need to develop new interventions against these parasites, built on a deep understanding of their biology.

The mechanisms of drug resistance in parasitic nematodes remain poorly understood at the molecular and biochemical levels, such that there is a major need for profound fundamental investigations at the cellular level. In spite of recent progress, there are distinct gaps in our knowledge and understanding of these nematodes’ biology, and precisely how they cause disease, preventing the discovery of new methods of control. Haemonchus contortus is arguably the most important and pathogenic nematodes of ruminant animals worldwide [17], and causes substantial disease as well as production losses to the livestock and associated industries and financial losses due to the costs of treatment and control [110]. For these reasons, much research effort has been focused on establishing essential genomic, transcriptomic and proteomic infrastructure for H. contortus to underpin detailed molecular and cellular biology investigations [19], [22], [115], with a view to identifying new intervention targets in this and related parasitic nematode species, particularly those of the order Strongylida.

Here, we created the first single cell atlas for H. contortus eggs undergoing embryonation through snRNA-seq and comparative analyses with C. elegans, employing advanced informatic tools. By focusing on genes that were transcribed in all clusters and types of cells in H. contortus embryos and recognised to be essential, we inferred a small subset of essential genes – Hc-ant-1, eef-1A or eef-1 G – the latter two of which are likely crucially involved in protein biosynthesis (Fig. 2; cf. [116]). Guided by information from C. elegans, we infer that Hc-eef-1A (HCON_00071960) and Hc-eef-1 G (HCON_00147420) are co-transcribed and/or interact (genetically, physically or in a regulatory manner) with large numbers (hundreds) of other genes involved in networks of critical biological processes/pathways, and show that both of these genes are highly transcribed and expressed in all key stages of H. contortus. Structure modelling and comparisons of orthologous proteins revealed relative conservation for eEF1A orthologs, but marked variation in the N-terminus of Hc-eEF1G (HCON_00147420) with respect to the host animal (sheep) and free-living nematodes, indicating that the latter protein may be more specific to H. contortus and related nematodes and, thus, might be amenable as a drug target. Nonetheless, both of these elongation factors have potential as targets, and should be considered in future evaluations.

Thus, this fundamental investigation of a first cell atlas of H. contortus has guided us toward eukaryotic protein synthesis machinery (specifically, elongation; Fig. 2) as a potential target for nematocidal drugs. An appraisal of published literature reveals that a range of natural product-derived compounds have been studied and shown to interfere with the elongation phase of eukaryotic protein biosynthesis (e.g., [117], [118]); some of these compounds have been explored and assessed as anti-cancer or ant-infective agents. Although cycloheximide, originating from Streptomyces griseus, has long been a prototype inhibitor of eukaryotic translation elongation, other natural compounds (n > 23) of various origins have been explored and shown to target the elongation step in different ways, including the inference with elongation factors eEF1 and eEF2, or binding to a particular site (A, P or E) in the ribosome [118]. Of the many compounds that have been studied, plitidepsin (Aplidin®) and homoharringtonine (Synribo®) have reached market approval. Other translation elongation inhibitors investigated to tackle infectious diseases include cycloheximide, lactimidomycin and homoharringtonine for select viruses [118] and, for example, compound DDD107498 for the malaria parasite, Plasmodium [119], which targets eEF2 in the GTP-dependent translocation of the ribosome along messenger RNA, and is essential for protein synthesis (cf. Fig. 2). DDD107498 was obtained through a phenotypic screen of a protein kinase scaffold library of 4731 compounds [120] against the blood stage of multi-drug-sensitive Plasmodium falciparum 3D7 strain; a compound series from this screen, based on a 2,6-disubstituted quinoline-4-carboxamide scaffold, had sub-micromolar potency against Plasmodium parasites, but suffered from poor physicochemical properties. Subsequent chemical optimisation resulted in DDD107498 with improved physicochemical properties and a 100-fold increase in potency [119]. This example indicates that medicinal chemistry approaches might be applicable to improve the selectivity of the effects of modified compounds to the parasite. While the possibility of “rational” design of compounds (cf. [121]) to eEF1A and eEF1G seems appealing, from an applied perspective, we would initially propose to screen known elongation inhibitors and then assess their potency on parasitic stages of H. contortus and related nematodes, and evaluate toxicity/mitochondrial toxicity in mammalian cells (representing the host animal) and in free-living nematodes (including C. elegans) to establish their selectivity/suitability for parasitic nematodes. If promising small molecules were identified that might be synthesized, we would propose medicinal chemistry studies to maximise potency and minimise toxicity. Such work would benefit from the use of a recently-established thermal proteome profiling assay [122] to identify or confirm compound-target interaction(s) as well as complementary snRNA-seq studies of different developmental stages of H. contortus to evaluate mechanism(s) of action.

5. Conclusion

Through the present investigation, our ability to identify cell populations or lineages of cells undergoing embryonation from snRNA-seq data using the present informatic workflow creates a novel paradigm for investigating H. contortus and related parasites by transferring the focus from research on whole organisms and tissues to the fine detail of cell populations within individual worms and tissues. Our plan now is to extend single nuclei analyses to other developmental stages and complement these analyses with single-cell analyses. This should allow the construction of an atlas of cell types and their differentiation for throughout H. contortus development, from embryos through to adult, blood-feeding females and males, which live within the host animal, to fill knowledge gaps in the complex biology of this worm and create critical data sets and capacity to explore related eukaryotic parasites, and to support the discovery of novel and improved interventions to combat this and related parasitic worms. These advances are also expected to open up completely new avenues to study drug resistance; drug discovery; vaccine- and drug-target discovery; host-parasite interactions and parasitism, thereby advancing the field of molecular helminthology.

Data and code availability

The project data is available through NCBI BioProject PRJNA957936 in SRA database (accession identifiers SRR24235293–6), and the scripts developed for the analysis of this snRNA data are available at https://gitlab.unimelb.edu.au/bioscience/haemonchus-contortus-snrna (version 0.01-beta).

CRediT authorship contribution statement

Pasi K. Korhonen: Curated RNA sequence data, Conducted all bioinformatic analyses using software described, Curation, Prepared results for tables and figures, Contributed to the drafting respective sections in the manuscript. Tao Wang: Collected and purified eggs of H. contortus, Isolated nuclei, Contributed to FACS and RNA sequencing, The drafting of sub-sections in the manuscript. Neil D. Young: Bioinformatics support, Discussion and comments on draft manuscript. Joe J. Byrne: Maintenance of the life cycle of H. contortus (Haecon5). Tulio L. Campos: Provision of gene essentiality data for comparison and comments on manuscript. Aya C. Taki: Maintenance of the life cycle of H. contortus (Haecon5) and contribution to figure preparation. Robin B. Gasser: Conceptualization, Supervision, Investigation and drafting of the manuscript, Editing and revision, Research funding.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary material

Supplementary material

.