Genome-Wide Temporal Gene Expression Reveals a Post-Reproductive Shift in the Nematode Caenorhabditis briggsae

Abstract

The nematodes Caenorhabditis briggsae and its well-known cousin Caenorhabditis elegans offer many features for comparative investigations of genetic pathways that affect physiological processes. Reproduction is one such process that directly impacts longevity due to its significant energetic demands. To study gene expression changes during reproductive and post-reproductive phases in both these nematodes, we conducted whole-genome transcriptome profiling at various adult stages. The results revealed that the majority of differentially expressed (DE) genes were downregulated during the reproductive period in both species. Interestingly, in C. briggsae, this trend reversed during post-reproduction, with three-quarters of the DE genes becoming upregulated. Additionally, a smaller set of DE genes showed an opposite expression trend, i.e. upregulation followed by post-reproductive downregulation. Overall, we termed this phenomenon the “post-reproductive shift”. In contrast, the post-reproductive shift was much less pronounced in C. elegans. In C. briggsae, DE genes were enriched in processes related to the matrisome, muscle development and function during the reproductive period. Post-reproductive downregulated genes were enriched in DNA damage repair, stress response, and immune response. Additionally, terms related to fatty acid metabolism, catabolism, and transcriptional regulation exhibited complex patterns. Experimental manipulations in C. briggsae to affect their reproductive status predictably altered gene expression, providing in vivo support for the post-reproductive shift. Overall, our study reveals novel gene expression patterns during reproductive and post-reproductive changes in C. briggsae. The data provide a valuable resource for cross-sectional comparative studies in nematodes and other animal models to understand evolution of genetic pathways affecting reproduction and aging.

Significance.

The interplay between reproduction and aging is a fundamental aspect of biology, yet the molecular details remain incompletely understood. This study reveals a significant post-reproductive shift in gene expression in Caenorhabditis briggsae, a nematode closely related to Caenorhabditis elegans. We discovered that gene expression patterns dramatically change as the animals transition from a reproductive to post-reproductive stage, highlighting major genetic networks and processes. Interestingly, our comparative analysis with C. elegans has uncovered significant evolutionary changes underlying the reproductive process. The findings provide new insights into how reproduction shapes aging and offers a foundation for exploring the evolution of genetic mechanisms underlying reproductive aging across animals.

Introduction

The process of reproduction is energetically costly and has a major impact on lifespan, particularly in females, where a negative relationship between reproductive activity and longevity is well-documented (Duggal 1978; Partridge et al. 2005; Maklakov and Immler 2016). Aging affects all biological processes, with reproduction being intimately intertwined with and influenced by changes that occur in animals as they get older. Germ cell quality diminishes with age, leading to reduced reproductive success and an increased incidence of birth defects (Brieno-Enriquez et al. 2022; Chu et al. 2023). In several mammal species, menopause marks the cessation of reproductive activity, initiating a post-reproductive life stage that is the topic of much scientific debate (Morton et al. 2013; Croft et al. 2015; Takahashi et al. 2016; Monaghan and Ivimey-Cook 2023; Gems and Kern 2024). Caenorhabditis elegans has been instrumental in broadening our understanding of genetic pathways across organisms in general, including more specifically pathways involved in reproductive aging (McCarroll et al. 2004; Girardot et al. 2006; van der Goot et al. 2012; Gems and Partridge 2013; Wang et al. 2014).

Hermaphroditic nematodes like C. elegans also experience an extended post-reproductive lifespan, though likely for different evolutionary reasons than menopausal mammals (Scharf et al. 2021). In these animals, reproduction is generally limited by the self-sperm supply, though mating with males can extend the reproductive phase until age-related impairments become more pronounced (Ward and Carrel 1979). Even after the cessation of reproduction which typically happens between day 6 and day 9 of adulthood in virgin hermaphrodites (Scharf et al. 2021), oocytes and yolk proteins continue to be produced and are eventually expelled without fertilization (Hansen and Schedl 2006; Kern et al. 2023).

As a close relative of C. elegans, Caenorhabditis briggsae serves as both an independent and a comparative evolutionary model system for exploring physiological processes and genetic pathways (Alliance of Genome Resources Consortium 2024; Gupta et al. 2007). This comparative approach allows for understanding traits that are conserved versus species-specific, as well as the molecular basis of these traits. Although both species are androdioecious and morphologically very similar, notable differences have been described in several physiological processes including reproduction (Kiontke et al. 2004; Gupta et al. 2007; Guo et al. 2009; Jhaveri et al. 2025).

Many longitudinal studies in C. elegans have focused on transcriptome changes in the context of aging or looked at the long-term effects of experimental interventions (e.g. Budovskaya et al. 2008; Golden et al. 2008; Schmeisser et al. 2013; Li et al. 2019; Tarkhov et al. 2019). These studies have revealed gene expression changes at different adult stages that cover both reproductive and post-reproductive ages. Other transcriptomic studies have focused on reproduction-related genes, specifically those involved in germline development and function. For example, Reinke et al. identified germline genes in C. elegans through differential transcriptomics in germline-impaired mutants (Reinke et al. 2000; Reinke et al. 2004). These sets include genes directly involved in reproduction and downstream factors, including transcription factors (TFs) lin-35, E2F heterodimer factors efl-1 and dpl-1, and spe-44, which regulate reproduction-related processes (Ceol and Horvitz 2001; Chi and Reinke 2006; Kulkarni et al. 2012; Ragle et al. 2022; Mikeworth et al. 2023). The evolutionary conservation of germline genes across nematodes has been studied, revealing rapid evolution of coding and regulatory sequences compared to somatic genes (Cutter and Ward 2005; Artieri et al. 2008). However, genome-wide expression studies are needed to facilitate a better understanding of differences and similarities of germline processes and reproduction in general.

Besides the role of aging in the reproductive capacity of nematodes, longitudinal studies have determined that many different physiological processes change and become disrupted with age including lipid and protein metabolism, mitochondrial function, and collagen metabolism (Golden et al. 2008). Aging affects the expression of extracellular matrix proteins, collectively referred to as the matrisome, which are constantly adjusted to maintain homeostasis but decline after reproduction (Ewald 2020). Lipid metabolism operates in a balancing act, allocating resources between survival and reproductive investment (Hansen et al. 2013). Movement and muscle function assays in C. elegans have also reported a gradual decline as age progresses (Herndon et al. 2002). Moreover, movement velocity decreases with age, and the decline becomes steeper between day 5 and day 10 as the reproductive period ends (Hahm et al. 2015; Koopman et al. 2020).

As mentioned earlier, longitudinal studies on aging often include time points that cover the reproductive as well as the post-reproductive stage, however the specific effects of this transition have remained underexplored. In this study, we took a genome-wide approach to examine gene expression patterns in C. briggsae in a longitudinal manner, across its reproductive and post-reproductive stages. Our analysis revealed a post-reproductive shift in the transcriptome, leading us to investigate affected processes, germline gene activity, and reproduction-related TFs. The findings showed post-reproductive upregulation of the matrisome and muscle processes, and downregulation of DNA repair, immunity, and stress response. A comparison of our transcriptome data between C. briggsae and C. elegans revealed similarities as well as differences in the transcriptome pattern. We observed a post-reproductive shift in C. briggsae but not in C. elegans, whereas expression changes in matrisome-related and several germline genes were partially conserved. Among other analyses, a comparison of downstream genes of a set of germline-related TFs revealed both conserved and species-specific expression trends. We experimentally manipulated the reproductive status of animals using Cbr-glp-4 mutants and mating, and observed that both approaches disrupted the post-reproductive shift, emphasizing its biological significance. Our study provides significant insights into gene expression changes during reproductive and post-reproductive stages in C. briggsae and serves as the basis for future research in comparative genomics. Similar studies in other nematodes and metazoans will help understand the evolution of genetic networks during reproduction and aging.

Results

Transcriptome Profiling During Reproductive and Post-Reproductive Stages

To investigate gene expression changes in C. briggsae hermaphrodites across reproductive and post-reproductive periods, we conducted whole-body transcriptome analyses on unmated populations at days 1, 3, 6, and 9 of adulthood. Figure 1 shows the reproductive span of AF16 unmated worms. The majority of fertilized eggs were laid within the first three days of adulthood. By day 6, egg laying was nearly complete, with a sharp decline in the number of fertilized eggs produced. By this time, most individuals were laying their last few eggs (<10), while some had already exhausted their capacity. By day 9, unmated individuals had fully stopped reproducing. The average reproductive span was 5.9 ± 0.31 d, with a mean lifetime brood size of 353.8 ± 47.7 (n = 9). Based on these observations, we defined D1-3 and D3-6 as early and late reproductive phases, respectively, and day 9 and beyond as the post-reproductive phase. Thus, the first three time points of transcriptome analysis overlapped with the reproductive period of C. briggsae, allowing us to compare DEGs between these four stages (i.e. day 1, day 3, day 6, and day 9) to determine gene expression changes during and after reproduction.

Fig. 1.

Reproductive span and brood size of C. briggsae AF16 unmated hermaphrodites. Bars display mean number of viable eggs laid by unmated C. briggsae hermaphrodites per day of adulthood. Data are plotted as the mean ± standard error. Line displays the percentage of animals that remain reproductively active. N = 23.

Our analysis revealed distinct patterns in DEGs across the sampled time points. When examined without a log2 fold-change cutoff, the proportion of upregulated and downregulated genes remained roughly equal at all intervals (supplementary file S1, Supplementary Material online; supplementary table S1, Supplementary Material online). However, the size of expression changes was less neutral after application of a log2 fold of 1.0 cutoff (Table 1). A comparison between day 1 and day 3 (D1-3) showed that the majority of DEGs were downregulated by day 3 (Fig. 2a to d; Table 1). A similar trend was observed when comparing days 1 to 6 (D1-6). In contrast, the comparison between days 3 and 6 (D3-6) revealed very few DEGs, with a greater proportion of genes being upregulated. This upregulation trend continued from day 6 to day 9 (D6-9). The strong bias toward downregulation by day 3, around the mid-reproductive stage, followed by a shift toward upregulation as the worms transitioned into the post-reproductive stage, suggests a shift in transcriptome associated with this transition.

Table 1.

C. briggsae age-based differentially expressed genes

Timeframe	Total DEG	Upregulated	Downregulated
D1-3	3,663	846	2,817
D1-6	5,074	1,188	3,886
D3-6	377	258	119
D1-9	2,879	1,624	1,255
D6-9	5,909	4,168	1,742

DE genes obtained from pairwise comparison of RNA-sequencing data. Inclusion criteria were relative log2fold expression ± 1 and P < 0.01.

Fig. 2.

MA plots of pairwise age-based differentially expressed genes in C. briggsae and C. elegans. a to e) Differentially expressed genes for C. briggsae D1-3 a), D3-6 b), D1-6 c), D6-9 d), and D1-9 e) sets of genes. f to h) Differentially expressed genes for C. elegans D1-5 f), D5-10 g), and D1-10 h) sets of genes. Positive log fold change indicates higher expression in the later time point, negative log fold change indicates lower expression in the later time point.

Analysis of DE Genes During Reproductive and Post-Reproductive Stages

As mentioned earlier, C. briggsae hermaphrodites stop reproducing after day 6. Hence, days 3 and 6 can be compared to day 1 to examine gene expression changes within the reproductive period, while comparisons with day 9 should reflect post-reproductive expression changes. Our analysis revealed a substantial overlap between the genes differentially expressed (DEGs) on D1-3 and D1-6, with 3,117 out of 3,663 genes (85%) shared between these two comparisons (Fig. 3a). All but four of the overlapping genes maintain the same directionality, further confirming that genes activated during reproduction maintain their expression trends until day 6. Conversely, the D1-6 set contains 1,957 genes (608 up and 1,349 down) that are not part of the D1-3 set (Fig. 3a). The majority of these genes (89%, 1,736) underwent a gradual expression change based on the fact that they were absent from the D3-6 set (supplementary file S1, Supplementary Material online).

Fig. 3.

Caenorhabditis briggsae differentially expressed genes and comparison of gene expression direction changes with time. a, c) Bar graphs of D1-3 versus D1-6 a) and D1-6 versus D6-9 c). Bars on the left indicate the number of DEGs in x-axis indicated set. Bars on the right are labeled to differentiate between genes not included in the left time point set (Non-overlapping DEG), genes included in the left set and having the same expression change direction, i.e. both increasing or decreasing (Overlapping DEG, same direction), and genes included in the left set and having an opposite expression change direction (Overlapping DEG, opposite direction). Inclusion criteria were relative log2fold expression ± 1 and P < 0.01. b) Heatmap of all 687 C. briggsae genes with significant expression changes at each pairwise measured interval between adult days 1, 3, 6, and 9, P < 0.01.

The D3-6 set consists of very few genes, and we found that nearly all of these genes show very small expression differences compared to the D1-3 and D6-9 sets (Fig. 3b, Table 1, supplementary file S1, Supplementary Material online). These data suggest that animals remain in a relatively transcriptionally stable state during this period. Overall, our analysis revealed that the D1-6 set represents the most inclusive DEG set for the reproductive period of C. briggsae. Comparing DEGs between this and the D6-9 set revealed a striking trend of expression inversion (Fig. 3c). Specifically, 52% (617 out of 1,188) of upregulated genes and 77% (2,995 out of 3,886) of downregulated genes inverted in expression direction. When taken together, this represents 71% of all DEGs.

To visualize the major patterns of gene expression changes, we carried out hierarchical clustering and identified nine distinct clusters that meaningfully represent genes in distinct groups (Fig. 4a to e). Genes that were downregulated during reproduction but upregulated post-reproduction (termed “down-invert” genes) make up 75% of the genes represented in three clusters followed by 22% with an “up-invert” pattern in four clusters. In line with there being very few DEGs in the D3-6 set, the largest clusters of these two categories consist of genes showing small expression changes between days 3 and 6. Finally, the two remaining clusters consist of genes, 2% and 1%, respectively, that on average do not show a reversal in expression trend.

Fig. 4.

Hierarchically clustered DEGs sorted into directional categories. a) Dendrogram displaying distance between genes which are differentially expressed in C. briggsae between days 1 and 6. The horizontal line indicates the cutting point in the tree, generating nine clusters. b to e) Hierarchically generated clusters, based on the 5,074 genes in C. briggsae with D1-6 differential expression. f to j) Similar dendrogram generated for C. elegans differentially expressed genes between days 1 and 5 f) and hierarchically generated clusters, based on the 3,535 genes (g to j). The horizontal line in f) indicates the cutting point in the tree, generating nine clusters. For all clusters in b) to e) and g) to j), the y axis shows transcript counts of each gene represented by relative change in its expression over the time period studied, where counts were transformed using variance stabilizing transformation and scaled by gene mean and standard deviation. Each thin line represents an individual gene, and the thick line represents the median. Clusters are subdivided into (b, g) up-invert, (c, h) down-invert, (d, i) down, and (e, j) up.

Enrichment Analysis of Processes and Pathways that Correlate With Post-Reproductive Shift

To elucidate the physiological changes associated with the transition from the reproductive (D1-6) to post-reproductive (day 9) stages of life, we performed Gene Ontology (GO) analysis on all C. briggsae DEGs and their C. elegans orthologs for biological processes, complemented by Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis on the latter. While both approaches produced largely similar results, C. elegans ortholog analyses provided a larger set of annotation terms and a higher coverage than C. briggsae and are therefore featured foremost in our analysis.

Independent analyses were performed on up- and downregulated gene sets to determine which processes were becoming more or less active. These analyses identified a large number of enriched annotation terms for each pairwise comparison. Because many of these terms referred to somewhat similar processes, we manually consolidated them into seven major function categories: matrisome, muscle development and function, DNA damage repair, stress and immune response, fatty acid metabolism, non-lipid catabolism, and regulation of transcription. Figure 5 shows general expression trends of associated genes and supplementary file S2, Supplementary Material online contains a full list of the grouped GO and KEGG terms with enrichment fold changes and statistical significance.

Fig. 5.

Caenorhabditis briggsae longitudinal transcriptome GO and KEGG term categories. GO term and KEGG pathway enrichment overview, categorized on expression change direction between days 1, 6, and 9 of adulthood, based on statistically significant, log2fold cutoff ± 1 DE genes list. Positive enrichment of upregulated genes and negative enrichment of downregulated genes both refer to related processes becoming more active. Conversely, positive enrichment of downregulated genes and negative enrichment of upregulated genes refer to related processes becoming less active. Boxes with both colors indicate a mixture of GO terms that are divided between both categories. For details on GO and KEGG annotation analysis, see Materials and Methods, and for enrichment term statistics, see supplementary file S2, Supplementary Material online.

The above categories could further be sorted into three broad expression trends: down-invert (matrisome and muscle development and function), up-invert (DNA repair and stress and immune response), and mixed direction changes (fatty acid metabolism, lipid and non-lipid catabolism and transcription, and transcription factors). In general, GO terms and sets of associated genes showed similar enrichment trends between the two species. Below, we discuss these categories and their associated GO terms, with additional descriptions available in the Supplementary Text (supplementary file S3, Supplementary Material online).

Matrisome

We observed a significant enrichment of sets of matrisome genes that were upregulated post-reproduction. Comparison of DEGs between the reproductive (D1-6) and post-reproductive (D6-9) periods revealed that genes linked to categories such as “basement membrane”, “extracellular matrix organization”, and “collagen-containing extracellular matrix” were significantly downregulated during the reproductive phase (D1-6) but upregulated post-reproductively (supplementary fig. S1a, Supplementary Material online). This gene set includes laminins, fos family member Cbr-fos-1, collagens (e.g. Cbr-emb-9, Cbr-let-2, and Cbr-cle-1), papilin Cbr-mig-6, nidogen Cbr-nid-1, SPARC, and the proteoglycan Cbr-unc-52. Since collagens were strongly represented among the genes that show upregulation after reproduction, we analyzed these in further detail through comparison with previous findings. Of the 181 C. elegans cuticular collagens (Teuscher et al. 2019), 149 were present in the C. briggsae genome as orthologs (supplementary file S4, supplementary fig. S1b, Supplementary Material online), with 77 of these showing the down-invert pattern, including the four with human orthologs (Cbr-cle-1, Cbr-col-99, Cbr-emb-9, and Cbr-let-2).

Muscle Development and Function

Similar to the matrisome category, we observed gene expression profiles for several terms related to muscle development and function. These genes showed the down-invert pattern and were involved in “regulation of striated muscle cell differentiation”, “muscle cell cellular homeostasis”, “muscle system process”, “muscle contraction”, “myofilament”, “sarcomere”, “contractile fiber”, and “striated muscle dense body”. Schorr et al. (2023) recently published a comprehensive list of C. elegans body-muscle-expressed genes. Although these were in mixed stage animals, we found that C. briggsae orthologs of body-muscle-specific genes, Cbr-unc-54, Cbr-mlc-2.2, Cbr-unc-27, and Cbr-lev-11, all fit in the down-invert pattern, as did most muscle-specific genes with RNA-binding domains (supplementary file S5, Supplementary Material online).

DNA Damage Repair

Unlike the above two categories, GO terms related to DNA damage repair enriched genes that were post-reproductively downregulated. The pattern of expression changes was unique in that the upregulated genes in the D1-3 set were enriched for terms involved in DNA damage repair or DNA replication but not in the D3-6 and D1-6 sets, indicating that these processes are most active early in reproduction. Enrichment terms included “DNA repair” and “double-strand break repair via break-induced replication”, with several MCM (minichromosome maintenance) complex components including mcm family members (Cbr-mcm-2, Cbr-mcm-3, Cbr-mcm-4, Cbr-mcm-5, and Cbr-mcm-6), Cbr-lem-3, and DNA repair protein Cbr-rfs-1/RAD51. Additionally, there were many terms involved in DNA replication, some of which overlapped with DNA repair.

Stress and Immune Response

Genes involved in stress and immune response showed a largely similar trend as the DNA damage repair set, except that the expression trend was significantly upregulated in both the D1-3 and D1-6 sets, followed by downregulation post-reproductively. This suggests that stress and immune response processes are active throughout reproduction. Representative genes and gene families that showed a post-reproductive shift include three caenacins (Cbr-cnc-4, Cbr-cnc-5.2, Cbr-cnc-11), seven lysozyme-like protein genes (lys and ilys classes), and systemic stress signaling Cbr-sysm-1. Other examples included the hsp-16.41 ortholog CBG04607 and four hsp-16.2 (HSP16 family) orthologs that displayed the post-reproductive shift. KEGG pathway annotation for D1-6 DEGs was enriched for the lysosome pathway including three cpr genes (Cbr-cpr-1, Cbr-cpr-3, Cbr-cpr-4).

Fatty Acid Metabolism

Fatty acid metabolism consisted of GO terms that displayed a mix of expression trends with some GO terms being enriched among genes with down-invert expression while others represented processes that showed the opposite trend. Within this group fits the set of terms involving metabolism and catabolism of fatty acids. The downregulated genes during the reproductive phase were enriched for four fatty acid elongation GO terms, each containing the same seven long chain fatty acid elongation genes (Cbr-elo-1, Cbr-elo-2, Cbr-elo-4, Cbr-elo-5, Cbr-elo-7, Cbr-elo-8, and Cbr-elo-9). Post-reproductively, annotation was enriched for fatty acid and lipid metabolism terms. The genes that invert expression upwards after reproduction were enriched for terms such as “lipid metabolic process”, “lipid catabolic process”, and “fatty acid catabolic process”. These included 20 lipase genes with seven lips-class members and Cbr-lipl-7, as well as fatty acid hydrolases Cbr-faah-1, Cbr-faah-3, and three acyl COA dehydrogenases (CBG18107, CBG15945, CBG02501). Among the post-reproductively downregulated genes, we found enrichment for several terms related to lipid metabolism as well. The terms included “negative regulation of lipid metabolic process”, “lipid metabolic process”, and “lipid biosynthetic process”, including Cbr-elo-3, as well as Cbr-faah-2, Cbr-faah-4, Cbr-faah-5, Cbr-lipl-1, Cbr-lipl-2, and CBG10449, which are all members of the same gene classes in the reproductively downregulated set but not the same genes. Under “lipid localization”, we found all five C. briggsae vitellogenin genes (CBG16767, Cbr-vit-2, CBG14203, CBG14234, Cbr-vit-6). In agreement with GO annotation, KEGG pathway analysis of C. elegans 1-to-1 orthologs revealed that “fatty acid metabolism” and “fatty acid biosynthesis” were enriched among both up and downregulated genes during the reproductive period. However, for the post-reproductive period, enrichment was observed only among upregulated genes. Furthermore, the entire set of post-reproductive DE genes were enriched for the “mTOR signaling pathway”, which plays a role in lipid metabolism (Lamming and Sabatini 2013), with genes like mom-2 (Wnt) and dsh-2 downregulated, leading to reduced inhibition of GSK3β and thereby reducing inhibition of mTORC1. Similarly, strd-1 (STRAD) was downregulated, impairing phosphorylation of aak-1/aak-2 (AMPK). Phosphorylation factor aap-1 (PI3K), which affects mTORC2, was also among the downregulated genes.

Non-Lipid Catabolism

Beyond lipid catabolism, many other catabolic processes were affected by the reproductive period expression shift, and that the genes involved rarely overlapped with the lipid catabolism sets, indicating that these were distinct processes. The GO terms “small molecule catabolic process”, “amino acid catabolic process”, and “carboxylic acid catabolic process” were all enriched among down-invert genes.

Transcriptional Regulation

The final category entails regulation of gene transcription. TFs are important to consider in transcriptomic analysis, given that changes in their expression could have profound downstream consequences. Finding germline-associated TFs was especially relevant considering the post-reproductive shift in expression. Our analysis revealed that for D-3, upregulated genes were enriched for several terms related to general regulation of transcription. Interestingly, D1-6 GO terms did not show significant enrichment among upregulated genes. However, when examining the post-reproductive period, enrichment was observed for genes involved in both positive regulation and negative regulation of transcription (supplementary file S2, Supplementary Material online).

To further investigate the genes affecting transcription, we analyzed the expression of the RNA polymerase II core complex members in our dataset. During reproduction, expression of these genes either remained unchanged or increased significantly whereas post-reproductively, all were decreased (supplementary file S6, Supplementary Material online). A similar analysis of corresponding C. elegans genes from a published dataset (Schmeisser et al. 2013) revealed increases during reproduction for only three core complex genes, and no significant downregulation post-reproductively. Thus, the effect appears to be specific to C. briggsae.

It was reported recently that RNA processing fidelity in C. elegans declines with age (Ham et al. 2022), with overall mRNA levels decreasing as various non-exon protein-coding and noncoding RNA categories become more prevalent. We analyzed the RNA processing-related terms in C. briggsae and found these to be enriched among up-invert genes (supplementary file S2, Supplementary Material online), consistent with the process becoming less active in post-reproductive age. This included genes enriched for GO terms “mRNA processing” and “RNA surveillance” among others that show subtle but significant changes in transcript levels (supplementary fig. S2, Supplementary Material online). It is important to mention that there was no reduction in RNA quality scores or mappable reads between samples obtained from differentially aged nematodes (supplementary file S7, Supplementary Material online; also see Methods). These results are expanded upon in the Supplementary Text (supplementary file S3, Supplementary Material online).

Comparison of Gene Expression Trends Between C. elegans and C. briggsae

We compared our longitudinal C. briggsae transcriptome with published C. elegans transcriptome data that feature reproductive and post-reproductive time points (supplementary file S1, Supplementary Material online) (Schmeisser et al. 2013). Our analysis revealed a similar reproductive period downregulation bias, with 2.8 time more downregulated than upregulated genes (Figs. 2e to h and 6, supplementary table S2, Supplementary Material online; see Materials and Methods for details). Unexpectedly, we found that the post-reproductive shift in C. elegans was not as stark as in C. briggsae: During the post-reproductive period, the downregulation bias disappeared and no major upregulation bias was seen. Hierarchical clustering of C. elegans DE genes (Fig. 4f to j) allowed us to compare expression clusters with those of C. briggsae. This analysis revealed a strikingly different distribution, with much fewer genes showing down-invert (18%) and up-invert (11%) patterns of post-reproductive expression shift (Fig. 4g and h). By contrast, the majority of genes maintain the same direction of expression change or remain unchanged post-reproductively (Fig. 4i and j).

Fig. 6.

C. elegans comparison of gene expression direction changes with time. Bar graph of D1-5 versus D5-10 DEGs in C. elegans. Bars on the left indicate number of DEGs in x-axis indicated set. Bars on the right are labeled to differentiate between genes not included in the left time point set (Non-overlapping DEG), genes included in the left set and having the same expression change direction, i.e. both increasing or decreasing (Overlapping DEG, same direction), and genes included in the left set and having an opposite expression change direction (Overlapping DEG, opposite direction). Inclusion criteria were relative log2fold expression ± 1 and P < 0.01.

GO annotation was performed on all C. elegans DE genes which revealed enrichment patterns that differed from C. briggsae (supplementary file S2, Supplementary Material online). The only similarity was an up-invert enrichment pattern for stress and immune response. The differences in expression patterns between the two species lead us to conclude that the C. briggsae post-reproductive shift, in C. elegans is either very different in nature, or absent.

Interestingly, we found that some of the annotation categories associated with the C. briggsae post-reproductive shift were previously reported in C. elegans. This included matrisome collagens and other genes involved in various types of metabolism including lipid metabolism (Golden et al. 2008). Of the 20 collagens for which Golden et al. reported a down-invert pattern, our C. elegans analysis revealed that these were similarly downregulated on day 5 but showed no upregulation on day 10. In C. briggsae, of the 13 orthologs present, there was a conserved, down-invert expression pattern for 11 of these collagens (Cbr-rol-6, Cbr-col-34, Cbr-col-49, Cbr-col-73, Cbr-col-122, Cbr-col-184, Cbr-dpy-5, Cbr-dpy-13, CBG18259, CBG23158, and CBG24927). Only Cbr-col-101 and Cbr-col-158 showed species-specific patterns, with the former decreasing expression by day 9 and the latter increasing on both days 6 and 9.

Muscle development-related terms were enriched among down-invert genes, but there was no significant enrichment for muscle function-related terms in C. elegans. This is different from C. briggsae, where a down-invert pattern was seen for both types of processes, suggesting that genes related to muscle function are regulated differently in these two species. In addition, we found that orthologs of the C. briggsae upregulated muscle genes were not upregulated in the C. elegans transcriptome of aged (day 10) animals. However, a comparison of our C. briggsae post-reproductive shift genes to orthologs of muscle-tissue-expressed genes in C. elegans adults (Kaletsky et al. 2018) yielded an overlap of 97/330 with up-invert genes (29.4%) and 395/2,236 down-invert genes (17.7%) (supplementary file S8, Supplementary Material online). Thus, 19.2% of all muscle-related DEGs in C. briggsae have C. elegans orthologs that are enriched in muscle tissue and a majority of these showed a down-invert expression pattern.

Germline Genes Analysis

We hypothesized that genes expressed during reproduction and showing a post-reproductive shift would correlate with reproductive processes. To test this, we examined expression data of C. elegans germline genes (Reinke et al. 2004). Our analysis revealed a significant enrichment of germline genes among the C. briggsae genes that were upregulated during reproduction and downregulated post-reproductively (103 of 222 genes, P < 0.001 using hypergeometric test with Benjamini and Hochberg false discovery rate correction) (supplementary fig. S3, Supplementary Material online, supplementary file S9, Supplementary Material online). Thus, a significant proportion of up-invert genes are involved in reproduction. These overlapping genes are linked to a broad range of processes, including oocyte development and maturation, embryogenesis, and embryonic development. The up-invert germline set includes Cbr-puf-8, for which the C. elegans ortholog is known to inhibit the overproliferation of germline stem cells in the germline allowing for stem cells to enter meiosis and become oocytes (Racher and Hansen 2012).

Reproductive processes in hermaphroditic nematodes involve many genes, which can be split up into components such as oogenesis, spermatogenesis, and intrinsic germline-related genes (Reinke et al. 2004). Work in C. elegans has shown that the TFs lin-35, efl-1, and dpl-1 are necessary for proliferation of germ cells and affect expression of several genes, although only efl-1 and dpl-1 are essential for oogenesis (Chi and Reinke 2006). Our expression analysis revealed that these three TFs are expressed and upregulated in both species during the reproductive period, however post-reproductive downregulation is observed only in C. briggsae (supplementary file S1, supplementary fig. S4, Supplementary Material online). Analysis of the downstream genes affected by these three factors showed that their genetic networks were largely conserved, with some species-specific differences (supplementary file S10, Supplementary Material online). In the case of lin-35, genes affected during reproduction were equally upregulated and downregulated in C. elegans (9 of 24 genes in each category). In contrast, in C. briggsae, the majority were upregulated (7 of 14 genes) with few being downregulated (two genes). Post-reproductively, only a couple of lin-35-affected genes were differentially expressed in C. elegans, whereas, in C. briggsae, a larger set of genes were affected and most were downregulated (six of nine genes).

Another TF, spe-44, promotes spermatogenesis (Ragle et al. 2022) and may also have a conserved role in C. briggsae (Kulkarni et al. 2012). Interestingly, we found that the Cbr-spe-44 expression has an up-invert pattern, unlike its C. elegans ortholog which remains unchanged through adulthood. Overall, these expression patterns of TFs and downstream genes suggest specific differences between C. elegans and C. briggsae reproductive gene expression, which might help explain the exclusive presence of the post-reproductive shift effect in C. briggsae.

Manipulating the Reproductive State of C. briggsae Hermaphrodites Affects Gene Expression

To determine whether altering the reproductive stage affects the post-reproductive shift, we selected six genes, of which five are down-invert genes with strong expression changes in C. briggsae: Cbr-col-129, Cbr-col-139, Cbr-mpz-4, Cbr-oac-9, and Cbr-spe-11 as well as non-invert gene Cbr-clec-266 which provides contrast (supplementary file S1, supplementary fig. S4c and d, Supplementary Material online). These genes are associated with immune response, matrisome, and reproductive processes. Expression of these genes was analyzed in two contexts: Cbr-glp-4(v273) hermaphrodites which have a temperature sensitive sterility phenotype (Velayudhan and Ellis 2022) and in AF16 that had reproduction extended by mating with males. In both these conditions, the post-reproductive shift is not expected to occur. Our qPCR analysis supported these expectations (supplementary file S11, Supplementary Material online). For all six genes, expression levels in day 2 Cbr-glp-4(v273) were comparable to post-reproductive AF16 animals (Fig. 7a). Additionally, the transcripts of all but Cbr-clec-266 were significantly higher in day 2 Cbr-glp-4 mutants in comparison to same stage AF16 (Fig. 7b). In control experiments, a post-reproductive shift was observed for most genes in AF16 adults (supplementary fig. S5a, Supplementary Material online). We also compared gene expression in day 2 and day 6 Cbr-glp-4 mutants. There was no consistent trend in transcript levels (supplementary file S11, Supplementary Material online), which may be due to genes becoming stochastically regulated in nonreproductive animals as they age (Herndon et al. 2002).

Fig. 7.

qPCR analysis of gene expression in reproduction impaired Cbr-glp-4(v273) and reproduction extended mated AF16. Of the six genes tested, Cbr-clec-266 is involved in immunity, Cbr-col-129 and Cbr-col-139 in cuticle formation, and Cbr-mpz-4, Cbr-oac-9, and Cbr-spe-11 in germline processes. Data are plotted as mean ± standard deviation. Statistically significant comparisons are indicated by stars (*P < 0.05). a, b) Cbr-glp-4(v473) hermaphrodites and AF16 controls were grown at 26°C and tested on day 2 (reproductive) and day 6 (post-reproductive). See Materials and Methods for details. c, d) Days 3 and 9 of animals, grown at 20°C, were mated on day 1 and day 7 of adulthood, respectively, and compared with unmated animals.

In unmated AF16 controls, gene expression generally increased from day 3 to day 9 as expected from our RNA-seq results (supplementary fig. S5b, Supplementary Material online). However, mating abolished the post-reproductive shift in all genes between day 3 mated and day 9 mated animals, except Cbr-clec-266 (Fig. 7c). Notably, Cbr-clec-266 appeared as an outlier in both experimental conditions (Fig. 7b and c), suggesting that immune genes might have a less straightforward relationship to the post-reproductive shift than the collagen and germline genes. It is also worth noting that mating itself altered gene expression, as three of the six genes tested showed expression differences between mated and unmated day 3 worms (Fig. 7d). Collectively, these findings demonstrate that the majority of the target genes are affected by the reproductive status of the animal.

Discussion

We performed transcriptome analysis across several time points, spanning both the reproductive and post-reproductive periods in C. briggsae and C. elegans. In C. briggsae, results revealed significant patterns of gene expression, most notably the post-reproductive shift. We observed that many genes, which were differentially expressed during the reproductive period, underwent marked changes in expression as animals aged and reproduction ceased. Relatively few genes showed further increase or decrease in the same direction after reproduction ended. This suggests that the cessation of reproduction plays a major role in the observed gene expression patterns, besides the process of aging itself.

The pronounced shift in gene expression change that we observed upon entry into the post-reproductive phase of C. briggsae is remarkable, in that more than two-thirds of the genes change the direction of their expression. The vast majority of these show the up-invert pattern, which suggests that these genes are likely to be directly involved in, or indirectly supporting, reproduction. This is supported by our analysis of germline genes. The down-invert genes may be either involved in physiological processes during the post-reproductive phase or subject to stochastic expression due to the force of natural selection declining after reproduction has ceased (de Magalhaes and Church 2005; Monaghan and Ivimey-Cook 2023).

Interestingly, while the gene expression pattern during the reproductive phase appears to be similar between C. briggsae and C. elegans, presence of the post-reproductive shift as seen in C. briggsae is not obvious in C. elegans. Several factors could explain these observed differences. While the time points of sample generation in the two species correspond to the reproductive and post-reproductive stages, chronologically they are not identical. Variation in sample collection locations may be another contributing factor. Based on our data, we argue that post-reproductive changes in gene expression are a biological reality which are more pronounced in, or unique to, C. briggsae. It is important to consider that C. briggsae differs from C. elegans in several significant ways, including developmental, genetic, and genomic factors that underlie differences in phenotypic traits such as temperature and stress sensitivity (Gupta et al. 2007; Jhaveri et al. 2025). Although both species are androdioecious, this trait has evolved convergently since the common ancestor was not (Kiontke et al. 2004). Thus, despite physiological similarities, gene expression patterns are expected to differ to some degree. Our findings of transcriptomic differences further emphasize the distinct physiologies of these two species. Follow-up work on species-specific differences in the post-reproductive shift might focus on the genetic underpinnings that lead to this expression pattern. Sequence similarity of TFs between C. elegans and C. briggsae is high (Haerty et al. 2008), leaving a small number of non-conserved TFs as candidates and potential TF binding sites for a focused investigation, e.g. their effect on downstream genes affected by the post-reproductive shift. Another line of investigation could focus more broadly on a comparative study of cis-regulatory sequences of such genes.

Through annotation analysis, we associated several major processes to gene expression changes in C. briggsae. The most consistent patterns included down-invert regulation of matrisome and muscle-related genes and up-invert regulation of DNA damage repair, stress and immune response-related genes. Collagen is a major component of the matrisome, and expression of some collagen genes increases with age (Halaschek-Wiener et al. 2005). Phenotypically, cuticle thickness also increases, which has been suggested to be the result of more lax expression regulation after the reproductive phase (Herndon et al. 2002). This phenomenon can be attributed to reduced evolutionary pressure on post-reproductive gene expression, as gene expression at this stage may no longer directly benefit offspring fitness (Rose 1991). Our analysis of collagens in C. elegans revealed a similar expression trend to that observed in C. briggsae. Intertwined with the progression from reproductive to post-reproductive stage are also age-related effects on expression of certain collagen and cuticle-related genes that were reported to decline in older adults (Budovskaya et al. 2008; Ewald et al. 2015; Palani et al. 2023). Cuticular collagens have been found to confer resistance to pathogen infection and oxidative stress (Ewald et al. 2015; Sellegounder et al. 2019), traits which wane as animals age (Lopez-Otin et al. 2013).

Similar to matrisome genes, muscle-related genes in C. briggsae exhibited increased expression post-reproductively. Proper muscle function depends on anchoring to other cells and to matrisome components. It is therefore not surprising to see analogous gene expression changes between muscle and matrisome genes. Unlike in C. briggsae, C. elegans muscle-related genes decreased during the reproductive period but did not increase afterward. Previous studies have shown that C. elegans muscle-specific gene transcripts, particularly those involved in muscle contraction, progressively and consistently become downregulated until day 7 of adulthood (Mergoud dit Lamarche et al. 2018). Another study reported that gene expression in C. elegans muscle tissue did not increase with age (Roux et al. 2023). Thus, while C. briggsae exhibits a post-reproductive upregulation pattern for muscle-related genes, the lack of a similar pattern in C. elegans is intriguing.

DNA repair-related genes were upregulated in C. briggsae during the early reproductive period (D1-3), followed by a smaller decline until day 6 and a significant decline during the post-reproductive phase. No enrichment for relevant terms was found in our C. elegans analysis. Previous studies have suggested that increases in chromatin condensation and DNA damage in older C. elegans may lead to decreased gene expression fidelity, with compensatory mechanisms potentially involving transcription factor expression (Golden et al. 2008). DNA repair is known to decrease with age, both at the transcriptional level and in effectiveness, across various species, including humans (Li et al. 2016; Lidzbarsky et al. 2018). This decline, measured as genomic instability and telomere attrition, is a common factor limiting lifespan, though more programmatic theories of aging should also be considered (Medvedev 1990; de Magalhaes and Church 2005; Lopez-Otin et al. 2013).

Cellular stress, such as mitochondrial stress and increased reactive oxygen species generation, is thought to contribute to the physiological decline associated with aging (Shpilka and Haynes 2018). Our analysis showed that, at the whole-transcriptome level, stress response processes decline in older, post-reproductive animals. Individual stress response genes varied: Among the post-reproductively downregulated genes we found caenacins and lysozyme-like proteins, of which C. elegans orthologs are involved in immune response against bacterial infection (Nicholas and Hodgkin 2004). We further found several orthologs of cpr genes which, in C. elegans, are shown to be upregulated following overexpression of the immune response activating gene clec-47 (Pan et al. 2021). Our results also revealed post-reproductive downregulation of proteasome-mediated catabolism, which is consistent with previous studies showing cellular machinery becoming less efficient in dealing with stress with age (Ben-Zvi et al. 2009; Rodriguez et al. 2013). The upregulation of immune and stress response genes during the reproductive phase and subsequent downregulation is consistent with evolutionary pressure to maximize survival during the reproductive period and reproductive success.

Lipid metabolism and catabolism-related genes exhibited varying expression patterns throughout the stages we measured. We found post-reproductive downregulation of vitellogenins, which are involved in the balance between reproductive success and lifespan as they traffic lipids from the intestine to eggs (Ranawade et al. 2018; Kern et al. 2023). Vitellogenin gene expression aligns with the initial high energy investment in egg production. This expression first decreases moderately as egg production tapers off and, after oocyte stores are depleted, continues to traffic lipids for some time to produce ventable yolk to support offspring (Kern et al. 2021; Kern and Gems 2022).

Additionally, our analysis of germline transcription factors Cbr-lin-35, Cbr-efl-1, Cbr-dpl-1, and Cbr-spe-44 showed an up-invert expression in C. briggsae, while no such pattern was found in C. elegans. Furthermore, the expression of genes downstream of these factors suggests that the species-specific post-reproductive shift has a large effect on reproduction-related factors in C. briggsae. We used Cbr-glp-4 sterile and mated wild-type worms with altered reproductive states to examine changes in the expression of a set of genes involved in matrisome and reproductive processes. As expected, the results showed an absence of a post-reproductive shift in these animals. However, we also found expression patterns that did not fit our hypothesis and could not be explained solely by the reproductive state. For instance, older Cbr-glp-4 adults (day 6) showed no consistent trend in gene expression compared to day 2 adults of the same genotype, suggesting a relaxed pressure on gene expression in post-reproductive animals (Rose 1991). Additionally, mating altered expression of some of the genes, in agreement with previous studies reporting gene expression changes in mated hermaphrodites (Booth et al. 2022; Shi and Murphy 2023). Future experiments involving a larger set of genes and manipulations affecting the reproductive state are needed to gain deeper insight into how reproduction and the post-reproductive shift influence the various processes identified in this study.

While other longitudinal transcriptome studies were performed under similar conditions as our study (Golden et al. 2008; Schmeisser et al. 2013; Tarkhov et al. 2019), it is important to acknowledge that the transcriptome, as extracted from hermaphrodites during the reproductive phase contains transcripts originating from both mothers and developing oocytes within the uterus. Many longitudinal transcriptome studies have worked around this issue by inhibiting reproduction through chemicals like fluorodeoxyuridine (FUdR) (e.g. Rangaraju et al. 2015; Li et al. 2019; Ham et al. 2022) or using strains with condition-specific induced sterility such as rrf-3 and fem mutants (e.g. Barton et al. 1987; Garigan et al. 2002; Lund et al. 2002; Murphy et al. 2003). While these methods effectively prevent progeny from mixing with the adult population when extracting RNA, they move the animals away from natural conditions, potentially affecting the transcriptome (Feldman et al. 2014; Anderson et al. 2016). One study also showed that FUDR exposure alters Escherichia coli (E. coli) amino acid biosynthesis, which can affect gene transcription of feeding nematodes (McIntyre et al. 2021). In generating the transcriptome for this study, we opted not to interfere with reproduction. Although the oocyte-related transcripts in actively reproducing hermaphrodites should be acknowledged, it represents a tiny portion (Schafer 2005; Corsi et al. 2015) and is therefore unlikely to meaningfully affect our conclusions. Moreover, based on PFAM and GO annotation, the DEGs in our RNA-seq dataset contained no genes exclusively involved in embryonic processes.

In conclusion, we have demonstrated the occurrence of a post-reproductive shift in the transcriptome of C. briggsae. We found that the genes affected by this shift are highly represented by several biological processes, most notably reproduction. Additionally, our findings reveal that genes involved in transcription and regulation of gene expression are downregulated post-reproductively. One possible explanation for this change may be an overall reduction in gene regulation in older adults, as predicted by the run-on effect of a quasi-program (Gems and Kern 2024). Presenting the first adult longitudinal transcriptome in C. briggsae, we believe that our data and analysis provide a valuable resource for the field, enabling more in-depth study of this species and Caenorhabditis nematodes more broadly. Furthermore, this work lays the groundwork for future research into the conserved mechanisms governing reproductive biology and aging, with broader implications across different animal models.

Materials and Methods

Worm Culture Maintenance

Worms were grown and maintained according to established methods (Wood 1988). Cultures were propagated on the Nematode Growth Medium (NGM) plates seeded with E. coli OP50 bacteria and maintained at 20°C, unless otherwise stated. Caenorhabditis briggsae AF16 wild-type strain was obtained from the Caenorhabditis Genetic Center (CGC). Cbr-glp-4(v473) was kindly provided by Ronald Ellis (Rowan University) (Velayudhan and Ellis 2022). For reproductive span determination, adult hermaphrodites were allowed to lay eggs for 1 h. The eggs were allowed to grow and the resulting young nongravid adults were transferred to individual plates. The adults were moved to new plates every 24 h until the cessation of egg-laying and the progeny were counted.

RNA Isolation and Transcriptome Profiling

For RNA isolation, age-synchronized populations of hermaphrodites were obtained by two successive rounds of bleaching of young gravid adults by standard methods using NaOCl and NaOH (Stiernagle 2006). Animals were left to pass through larval stages and live to specified ages: days 1, 3, 6, and 9 of adulthood. Every second day, adults and larvae were column separated by washing in an upright 15 ml tube filled with M9 for 5 min to achieve size separation, after which the water column was pipetted off to remove larvae. This was repeated at least three more times to increase removal of larvae. Worms were pipetted on fresh seeded plates, and remaining larvae were removed with a platinum worm pick and burned. Populations were screened to assure the absence of males, and any males found were removed. When populations reached the target age, worms were washed in the same way and frozen at −80°C followed by RNA isolation.

RNA was isolated using Trizol-reagent (Sigma, USA, catalog number T9424), chloroform, and isopropanol (Chomczynski and Sacchi 1987). The quality of total purified RNA was confirmed using Nanodrop 1000 bioanalyzer (Thermofisher). cDNA libraries were constructed from 100 to 200 ng RNA using an Illumina-specific commercial kit (TruSeq RNA Sample Preparation Kit v2, Set A, catalog number RS-122-2001). RNA-sequencing was carried out using an Illumina NovaSeq PE100 system at Génome Québec. For the day 1, 3, and 6 age categories, three biological replicates were used, consisting of full large plates of worms. For the day 9 age category, two biological replicates were used due to one sample not passing quality control. Paired-end reads were obtained for each cDNA library. Sequencing adapters were used with the following sequences: read 1 “AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC”; read 2 “AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGT”.

RNA sample quality was vetted for inclusion based on Bioanalyzer RIN scores (supplementary file S7, Supplementary Material online). Sequence data were extracted in FASTQ format and further processed and analyzed on the Galaxy platform (https://usegalaxy.org). RNA-sequencing was performed with the following software and settings: MultiQC was used to aggregate multiple results into a single report (Ewels et al. 2016). Reads were aligned to the genome using RNA STAR, set to paired-end sequences and default settings (Dobin et al. 2013). The reference genome used was C. briggsae CB4 (Ross et al. 2011), in combination with WS279 annotation obtained from WormBase. To count hits per gene, featureCounts was used with default settings for paired-end reads (Liao et al. 2014). Average input read counts were 33.0 M per sample (range 22.7 M to 39.2 M) and average percentage of uniquely aligned reads was 87.1% (range 85.8% to 88.1%) (supplementary file S7, Supplementary Material online). Differential gene expression analysis was performed with DESeq2 in a pairwise manner with default settings (Love et al. 2014). In each comparison, the earlier time point was used as baseline.

The day 1 samples of the first and second sequencing runs were compared and found to be negligibly different. Further analyses involving the day 9 samples of the second sequencing run were done in comparison with the day 1, 3, and 6 samples of the first sequencing run. Genes were considered significantly differentially expressed on the later time point of a pairwise comparison for adjusted P-values <0.01. For those analyses where log2fold cutoff values were applied, limits were set for log2fold changes of ±1.

DEG counts were tracked and graphed in Microsoft Excel. Further full time-course analysis and hierarchical clustering were performed in R with hclust, set to find complete linkage on transcript numbers transformed with variance stabilizing transformation (vst). Graphs of Figs. 2 and 4 and supplementary figs. S2 and S5, Supplementary Material online were produced in R using ggplot2 (Wickham 2016). Venn diagram of germline-related gene analysis was made with Venny (Oliveros, 2007-2015 Last accessed March 11 ).

Gene Annotation

WormBase (http://www.wormbase.org) was used as a resource for information about worm genes and pathways (Sternberg et al. 2024). Caenorhabditis elegans homologs of C. briggsae genes were identified based on known orthology data (Inparanoid 8 https://inparanoidb.sbc.su.se/, Wormbase-Compara, OrthoMCL https://orthomcl.org/orthomcl/app, OMA https://omabrowser.org/oma/home/, and Hillier et al. 2007) (supplementary file S12, Supplementary Material online) and a custom Perl script. Analyses were done on both the C. briggsae genes and C. elegans orthologs of the equivalent sets of orthologous genes, i.e. genes without known homologs between the two species were excluded from annotation analysis. The sets of upregulated and downregulated genes were analyzed separately. GO analysis was performed through the 17.0 release of PANTHER at https://archive.pantherdb.usc.edu/ (Thomas et al. 2022). KEGG analysis was performed on C. elegans orthologs, using version 106.0 of the KEGG database (Kanehisa and Goto 2000). The tool KOBAS-i was used to perform KEGG analysis in a high-throughput manner on the top 3,000 genes with greatest expression change per dataset (Bu et al. 2021). GO and KEGG P-values are provided in two formats: uncorrected and Benjamini and Hochberg FDR corrected, with inclusion cutoffs set at P < 0.05 for both. GO and KEGG enriched terms were grouped into categories, each represented by multiple annotation terms at most or all time points. This was further manually trimmed to seven major categories based on gene set overlap and physiological relatedness.

C. elegans Comparison

Aging-associated expression patterns in C. briggsae were compared to C. elegans. Published microarray and RNA-seq datasets (Golden et al. 2008; Schmeisser et al. 2013) were used as comparative longitudinal expression data in C. elegans. Main expression comparisons were performed using the Schmeisser et al. (2013) dataset that describes RNA-seq analysis of C. elegans N2 hermaphrodites grown at 20°C, aged to days 1, 5, and 10 of adulthood. These time points correspond to reproductive (day 1 and day 5) and post-reproductive (day 10) stages of animals and, therefore, compared with C. briggsae samples of similar reproductive (day 1 and day 6) and post-reproductive (day 9) stages. For the Golden et al. (2008) dataset, comparisons were done using expression values for day 1, day 5, and day 9 of adulthood.

qPCR Analysis of Reproduction-Related Genes

Cbr-glp-4(v473) were bleach-synchronized and grown at 25°C until L4. AF16 were grown alongside but started 12 h later to compensate for developmental delay in Cbr-glp-4 mutants. Since Cbr-glp-4 mutants are fully sterile at 26°C, experiments were performed at this temperature. Day 2 and day 6 time points were chosen based on the reproductive states of AF16, which were reproductively active on day 2 and had stopped producing progeny by day 6.

For extended reproduction assays, AF16 hermaphrodites were mated to males for 48 h from day 1 and day 7 of adulthood. Mating success was confirmed by observation of copulatory plug on all animals after approximately 24 h, after which males were removed. After reaching the designated ages of day 3 and day 9 of adulthood, hermaphrodite populations were washed 1× in M9, 1× in water to get rid of bacteria. After washing, worms were left in 18 μl water, 50 μl RNAzol (MRC Inc.) and 2 μl precipitation carrier (MRC Inc.) were added.

RNA was extracted using isopropanol and followed standard methods. Samples were generated in at least three biological replicates. cDNA synthesis was performed using a SensiFAST cDNA synthesis kit (BIOLINE) with equalized amounts of RNA between samples. SensiFAST SYBR Green (BIOLINE) quantitative RT-qPCR was performed using the BIORAD CFX-96 Real Time system and following the BIORAD-CFX Software manual. Gene expression levels were normalized to housekeeping gene Cbr-iscu-1. Primers used are listed in supplementary table S3, Supplementary Material online. The results were analyzed using CFX Maestro 3.1 software (Bio-Rad, Canada), from three technical replicates for each run using the comparative 2ΔΔCt method and significance assessed by one-way ANOVA or t-test.

Supplementary Material

Supplementary material is available at Genome Biology and Evolution online.

Funding

This research was supported by a Natural Sciences and Engineering Research Council of Canada Discovery grant to B.P.G.

Data Availability

All analyzed results and data files underlying this article are available in the article and in its online Supplementary material. The RNA-seq raw files are available at NCBI GEO under accession number GSE285675. Differential expression analysis data are in supplementary file S1, Supplementary Material online. GO and KEGG pathway enrichment results are available in supplementary file S2, Supplementary Material online. All scripts are publicly available on the GitHub repository: https://github.com/Echodonut/Post-reproductive-shift.git.