Transcriptomic changes in the testes and brain associated with sperm competition risk in male zebra finches

doi:10.21203/rs.3.rs-3456692/v1

Download PDF

Article

Transcriptomic changes in the testes and brain associated with sperm competition risk in male zebra finches

https://doi.org/10.21203/rs.3.rs-3456692/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

We investigated how phenotypic plasticity induced by experimentally manipulated sperm competition risk is reflected in transcriptomic profiles of testes and two brain areas (posterior pallium and optic tectum) in breeding male zebra finches (Taeniopygia guttata). Based on an exceptionally large sample of 59 individual transcriptomes, we compared two experimental groups: males held in single pairs versus two pairs per breeding cage. We found tissue-specific differential gene expression in all three tissues at the level of gene co-expression network modules. The treatment affected multiple cellular functions in the testes, which can be associated with germ cell production and modification. Moreover, neural and glial brain cell changes indicate phenotypic adjustment of male zebra finch behaviour to the social environment. Finally, individual co-expression modules correlated with individual behavioural phenotypes and hormonal profiles. Our pre-registered study shows the importance of transcriptomic plasticity for phenotypic adjustment to male-male competition in zebra finches.

Peter Korsten, Uwe Mayer and Tim Schmoll share the last authorship.

Biological sciences/Genetics/Gene expression

Biological sciences/Ecology/Behavioural ecology

Biological sciences/Neuroscience/Social behaviour

Aggression

extra-pair paternity

male-male competition

pre-registration

sexual selection

sperm competition

RNA-Seq

transcriptomics

weighted correlation network analysis

To maximise fitness in heterogeneous environments, animals need to adjust their phenotypes to ever-changing conditions along a multitude of environmental dimensions^1–3. This includes matching their morphological, physiological, and behavioural phenotypes to the current social environment. Such social niche conformance⁴ is likely to strongly affect natural variation among individuals in phenotypic trait expression, especially in response to social conflicts^4,5. A critical feature of the social niche that can generate phenotypic variation is sexual competition^6–8, which is shaped by the density of potential mating partners and same-sex competitors, and leads to the sexual selection of competitive and/or ornamental traits^9–12. In mating systems that are not strictly genetically monogamous, sexual selection also operates after mating through sperm competition^13,14 and/or cryptic mate choice^15,16. Furthermore, males represent the more competitive sex in most animal mating systems^17,18. Male investment into competitive traits is costly¹⁹, promoting the evolution of individual phenotypic plasticity³ of male competitive traits in response to variation in relevant socio-ecological conditions. For example, the level of (perceived) sperm competition affects ejaculate size²⁰ and composition²¹ and has been hypothesised to affect costly behavioural traits such as aggression^22,23. However, while substantial evidence suggests that males are sensitive to key socio-sexual gradients like rival density, the underlying molecular mechanisms enabling these phenotypically plastic responses remain largely unclear.

Changes in morphological, physiological or behavioural phenotypes in response to the social environment are mediated by mechanisms that alter how genes are expressed without altering the underlying genetic sequence^24,25. The recent advent of omics- technology, particularly differential gene expression studies based on RNA sequencing, has allowed direct investigation of these phenomena at an unprecedented scale²⁶. For example, differential gene expression studies were successfully used to analyse the neurogenomic architecture of female mating preferences²⁷, the neurogenomic response to elevated nest site competition in females²⁸, the molecular mechanics of sex-differential ornament expression²⁹, the re-modelling of the phenotypes of males engaging in alternative mating tactics³⁰, the neurogenomic plasticity associated with male territory defence³¹ and the molecular basis of individual variation in male sexual competitive ability³². However, it is largely unclear how differential gene expression allows for individual phenotypic plasticity in male competitive traits (see e.g.^33–35).

Studying the molecular basis of phenotypic plasticity in male competitiveness can be especially informative in socially monogamous mating systems with male parental care, where males have to trade off investment in seeking additional fertilisation opportunities versus investment in caring for zygotes they have already fertilised^36,37. For example, up-regulating aggression towards sexual competitors in territory defence or mate guarding may negatively affect the provision of paternal care. Thus, under social monogamy and male parental care (as in zebra finches), the integration of conflicting demands of competitiveness versus care may constrain the evolution and plastic up-regulation of sexual competitive male traits. As a result, any footprints of post-mating sexual selection may be subtle and remain unrecognised or under-appreciated in their ecological and evolutionary significance. By adopting a reverse-engineering perspective, studying differential gene expression in relation to experimentally controlled mating regimes may help recognise more subtle phenotypic changes resulting from post-mating sexual selection. Moreover, this approach can help to uncover the molecular mechanisms involved in balancing the conflicting demands of sexual competitiveness versus parental care³⁸.

Here, we report on socially sensitive gene expression in male zebra finches (Taeniopygia guttata), a socially monogamous passerine bird with male parental care and female extra-pair mating³⁹, and a key model in studies on mate choice^39,40 and sperm competition^41,42. Based on exceptionally large samples of up to 59 individual transcriptomes, in this pre-registered study⁴³, we explored whether phenotypic plasticity due to variation in sperm competition risk and associated social stimulation is reflected in gene expression. We conducted RNA-sequencing comparing the gene expression profiles of male zebra finches across two experimental groups exposed to different social environments (Single-pair versus Double-pair males, see Fig. 1). The effects of this treatment on the behavioural and hormonal profiles of zebra finch males have been reported in a complementary paper, which revealed individual phenotypic plasticity to be induced by the social environment⁴⁴. For example, males decreased within-pair interactions (i.e., time spent courtship singing and time spent near their social mates) and aggression towards an unfamiliar intruder under the Double-pair treatment⁴⁴.

The current paper focuses on the gene expression profiles in different tissues of the same zebra finch males. For each male, we obtained the transcriptome from one testis and two brain regions (posterior pallium and optic tectum). We decided to study the testes because of their central role in traits linked to sperm competition (plasticity in testosterone profiles and ejaculate traits⁴⁵). The posterior pallium we selected because it contains large portions of amygdala-homologue brain regions (arcopallium, posterior amygdala and nucleus taenia of the amygdala)^46,47. These regions are part of the social decision-making network⁴⁸, which modulates multiple social responses, including fear/aggression towards conspecifics. The optic tectum, the first station of the primary visual pathway of birds, we selected as a control region, which we expected to be unaffected by the social treatment. We thus studied the key tissues (testis and brain) implicated in regulating male competitiveness to shed light on the molecular mechanisms underlying the plasticity of complex and highly integrated male reproductive phenotypes. We thereby capitalised on three significant advantages of using a high number of individual transcriptomes. First, we conducted a differential expression analysis of individual genes between the two treatment groups on a large sample without pooling the data of multiple individuals. Additionally, this large sample allowed us to perform a gene co-expression network analysis to study groups of genes with highly correlated expression (co-expression modules) affected by the experimental treatment⁴⁹. Lastly, we related the individuals’ expression of these modules to their hormone profiles (plasma testosterone and plasma corticosterone) and key behavioural traits (aggression and song).

Distinct gene expression between the testis, posterior pallium and optic tectum

We collected 180 tissue samples from the testis, posterior pallium and optic tectum of 60 male zebra finches (n = 30 males per treatment level). Not all samples provided sufficient RNA, though, resulting in overall 173 tissues samples from 60 individuals (testis Single-pair n = 29, Double-pair n = 30; posterior pallium Single-pair: n = 28, Double-pair: n = 29; optic tectum Single-pair: n = 28, Double-pair n = 29). Across the 173 tissue samples, a total of 21,040 genes annotated in the zebra finch genome^50,51 were expressed. As expected, the overall gene expression patterns summarised by a principal component analysis clearly distinguished the three tissues (Fig. 2), revealing highly tissue-specific expression patterns. We thus performed gene expression and gene co-expression analyses separately for each tissue. Specifically, we compared Single-pair and Double-pair males to identify genes and gene networks whose expression levels were sensitive to our experimental treatment.

Treatment-dependent expression of gene networks in the testis of Single- and Double-pair males

Overall, 20,380 expressed genes were identified in the testis (TES). While the transcriptomic profiles differed substantially among individual males, analyses restricted to individual genes did not reveal any differential expression between Single- versus Double-pair males in the testis. However, any complex remodelling of integrated phenotypes may be caused by concerted and potentially subtle expression changes of networks of co-expressed genes rather than by dramatic expression changes in single genes. The genes forming such co-expression networks are likely involved in the same or similar functional pathways^52,53. We thus applied weighted co-expression gene network analysis (WGCNA) to identify groups of genes (co-expression modules) that changed expression in response to the experimental treatment⁴⁹. We detected 66 modules in the testis representing clusters of genes with highly correlated expression (Supplementary Data Table S1a; note that modules are named by arbitrary colours assigned by WGCNA software). Five of these modules were differently expressed between the two treatment groups (Fig. 3a-e, Table 1). These treatment-sensitive modules contained between 40 and 265 genes, of which 8–46 were classified as hub genes, a class of highly connected genes within a module (Table 1). The statistical results for all modules, including lists of identified genes and hub genes of each treatment-sensitive module, are summarised in Supplementary Table S1a-c.

Table 1

Treatment-sensitive, differentially expressed modules in the testis (TES), posterior pallium (PAL) and optic tectum (OT).
Modules	Number of genes	Number of hub genes	Statistics
TES Paleturquoise	128	18	F_(1,57) = 7.180, p = 0.010
TES Orangered3	40	8	F_(1,57) = 5.663, p = 0.021
TES Cyan	265	38	F_(1,57) = 4.654, p = 0.035
TES Lightgreen	210	46	F_(1,57) = 4.581, p = 0.037
TES Saddlebrown	140	24	F_(1,57) = 7.472, p = 0.008
PAL Floralwhite	108	28	F_(1,55) = 8.327, p = 0.006
PAL Saddlebrown	173	59	F_(1,55) = 7.251, p = 0.009
PAL Blue	1986	179	F_(1,55) = 6.180, p = 0.016
PAL Darkolivegreen	160	160	F_(1,55) = 4.506, p = 0.038
PAL Darkturquoise	193	47	F_(1,55) = 4.137, p = 0.047
PAL Darkgreen	210	28	F_(1,55) = 4.066, p = 0.049
OT Salmon	255	40	F_(1,55) = 5.054, p = 0.029
OT Thistle1	48	10	F_(1,55) = 4.430, p = 0.040
OT Lightyellow	190	5	F_(1,55) = 4.066, p = 0.049

To specify the biological processes associated with the genes of the five modules significantly affected by the experimental treatment, we performed gene ontology (GO) enrichment analysis⁵⁴. This analysis is based on functional annotations databases, reflecting the scientific community's current understanding of the functions associated with each gene⁵⁵. Four of the five modules contained significantly enriched GO terms in the testis. The top five functions identified by enriched GO terms are summarised in Table 2 (for other functions, see Supplementary Table S1d-h).

Table 2

Top five significant gene ontology (GO) terms for biological processes within the modules that were significantly differentially expressed between treatment levels. The number of significantly enriched GO terms are indicated in brackets for each module. Testis (TES), Posterior pallium (PAL) and Optic Tectum (OT).
Module	Top five terms	Term identity
TES Paleturquoise (1)	Cytoplasm	GO:0005737
TES Orangered3 (213)	Positive regulation of transcription by RNA polymerase II Regulation of MAPK cascade Regulation of transcription from RNA polymerase II promoter in response to stress Regulation of DNA-templated transcription in response to stress MAPK cascade	GO:0045944 GO:0043408 GO:0043618 GO:0043620 GO:0000165
TES Cyan (7)	Plus-end-directed vesicle transport along microtubule Plus-end-directed organelle transport along microtubule Transport along microtubule Cytoskeleton-dependent intracellular transport Vesicle transport along microtubule	GO:0072383 GO:0072386 GO:0010970 GO:0030705 GO:0047496
TES Lightgreen (5)	Positive regulation of endocytosis Regulation of phagocytosis, engulfment Positive regulation of phagocytosis, engulfment Regulation of membrane invagination Positive regulation of membrane invagination	GO:0045807 GO:0060099 GO:0060100 GO:1905153 GO:1905155
TES Saddlebrown (0)	-
PAL Floralwhite (0)	-	-
PAL Saddlebrown (31)	Oligodendrocyte differentiation Glial cell differentiation Gliogenesis Glial cell development Ensheathment of neurons	GO:0048709 GO:0010001 GO:0042063 GO:0021782 GO:0007272
PAL Blue (55)	Neuroactive ligand-receptor interaction Intrinsic component of membrane Integral component of membrane Signaling Cell communication	KEGG:04080 GO:0031224 GO:0016021 GO:0023052 GO:0007154
PAL Darkolivegreen (0)	-	-
PAL Darkturquoise (5)	Semaphorin receptor activity Transmembrane signaling receptor activity Molecular transducer activity Signaling receptor activity Neurotransmitter receptor activity involved in regulation of postsynaptic cytosolic calcium ion concentration	GO:0017154 GO:0004888 GO:0060089 GO:0038023 GO:0099583
PAL Darkgreen (6)	Meiosis I Meiosis I cell cycle process Reciprocal homologous recombination Homologous recombination Reciprocal meiotic recombination	GO:0007127 GO:0061982 GO:0140527 GO:0035825 GO:0007131
OT Salmon (3)	Protein-DNA complex Nucleosome DNA packaging complex	GO:0032993 GO:0000786 GO:0044815
OT Thistle1 (50)	Axoneme assembly Microtubule bundle formation Cilium or flagellum-dependent cell motility Cilium movement involved in cell motility Cilium-dependent cell motility	GO:0035082 GO:0001578 GO:0001539 GO:0060294 GO:0060285
OT Lightyellow (317)	DNA-binding transcription factor activity Transcription regulator activity DNA-binding transcription factor activity, RNA polymerase II-specific DNA binding Sequence-specific DNA binding	GO:0003700 GO:0140110 GO:0000981 GO:0003677 GO:0043565

All 217 GO terms identified for the Orangered3 module in the TES tissue were significantly enriched. The many enriched functional annotations of the TES Orangered3 module are summarised in a semantic space scatter plot (Fig. 4a). Their functions are mostly related to diverse cellular processes, such as regulation of the MAPK cascade, cell response to stress, cell communication, cell differentiation, signalling and other metabolic processes. All seven GO terms identified for the TES Cyan module were significantly enriched. These GO terms are exclusively related to intracellular transport functions along the microtubule and cytoskeleton (Table 2). In addition, all five GO terms identified for the TES Lightgreen module were significantly enriched. These GO terms are related to functions such as endo- and phagocytosis and membrane invagination (Table 2). The only significantly enhanced GO term of the TES Paleturquoise module (out of 3,487) was cytoplasm (Table 2). The TES Saddlebrown module, containing 13,686 GO terms, did not present significantly enriched ones.

Treatment-dependent expression of genes and gene networks in the posterior pallium and optic tectum

In the posterior pallium (PAL), 10 of 18,917 identified genes were differentially expressed between Single-pair and Double-pair males (Table 3). Four genes had higher expression in Double- compared to Single-pair males, while six genes showed lower expression levels in the Double-pair treatment. In the optic tectum (OT), 6 out of 19,030 genes had lower expression in Double-pair males than in Single-pair males. One gene (Predicted: MTTP) showed lower expression in both tectum and posterior pallium in Double-pair compared to Single-pair males (Table 3).

Table 3

Differentially expressed genes between Single- *versus* Double-pair males in the posterior pallium and optic tectum. The magnitude of the difference in expression is represented by log 2-fold change (log₂FC). Higher values indicate higher expression in the Double-pair relative to the Single-pair treatment level. False discovery rate (FDR) refers to p-values that were adjusted for multiple comparisons. The significance level was set to FDR < 0.1
Gene symbol	Gene name	log₂FC	FDR
Pallium: higher expression in the Double-pair treatment level
MC5R	Melanocortin 5 receptor	0.519	0.031
GNA11	G Protein Subunit Alpha 11	0.163	0.044
RND3	Rho Family GTPase 3	0.138	0.044
Predicted: UGT8 (LOC100228968)	2-hydroxyacylsphingosine 1-beta galactosyltransferase	0.813	0.099
Pallium: higher expression in the Single-pair treatment level
Predicted: MTTP (LOC105759397)	Microsomal triglyceride transfer protein large subunit	-2.596	0.031
Predicted: SMAD6 (LOC105758838)	Mothers against decapentaplegic homolog 6	-0.297	0.031
ZNRD2	Zinc Ribbon Domain Containing 2	-0.496	0.031
VAMP8	Vesicle Associated Membrane Protein 8	-0.388	0.099
TVP23A	Trans-Golgi Network Vesicle Protein 23 Homolog A	-0.165	0.099
MXD1	MAX Dimerization Protein 1	-0.353	0.099
Tectum: higher expression in the Single-pair
Predicted: MTTP (LOC105759397)	Microsomal triglyceride transfer protein large subunit	-3.244	< 0.001
Uncharacterised (LOC115492340)	-	-1.937	0.005
PNMT	Phenylethanolamine N-methyltransferase	-1.715	0.055
Predicted: OASL1 (LOC100224927)	2'-5'-oligoadenylate synthase-like protein 1	-1.792	0.099
Predicted: IFIT5 (LOC100229421)	Interferon-induced protein with tetratricopeptide repeats 5	-1.628	0.099
Uncharacterised (LOC115492673)	-	-1.917	0.099

We identified 49 gene co-expression modules in the posterior pallium (Supplementary Data Table S2a). Six modules showed significantly different expressions between the two treatment groups (Fig. 3f-k). These modules contained between 108 and 1986 genes, of which 28–179 were classified as hub genes (see Table 1). The results for all modules in the posterior pallium, including the list of all genes contained in the modules and hub genes of treatment-sensitive modules, are summarised in Supplementary Table S2a-c.

Four out of the six significantly affected modules contained significantly enriched GO terms. All 55 GO terms identified for the Blue module in pallium (PAL Blue module) were significantly enriched. PAL Saddlebrown had 31 significantly enriched GO terms out of 4832 identified terms. PAL Darkturquoise contained five (out of 5,376), and PAL Darkgreen 6 (out of 3,870) significantly enriched GO terms. The other modules (PAL Floralwhite, containing overall 1,887 GO terms and PAL Darkolivegreen, containing 3,045 GO terms) did not present significantly enriched GO terms. The functional annotations of the top five significantly enriched GO terms for each module are summarised in Table 2 (for all other GO terms, see Supplementary Table S2d-i).

The functions of the two modules with the highest numbers of significantly enriched GO terms for pallial tissue (PAL Saddlebrown and PAL Blue) have been further summarised as semantic space scatter plots (Fig. 4b-c). This revealed a macro cluster of functions related to the G protein coupled receptor signalling pathway in the PAL Blue module. In contrast, in the PAL Saddlebrown module, a predominance of functions related to glial cell differentiation was evident. Cell communication and signalling were significantly enriched in the PAL Blue but not in the PAL Saddlebrown module. Overall, most gene functions of the PAL Blue module seem to be related to neuronal processes and signalling. On the contrary, the PAL Saddlebrown module mainly identifies gene functions related to glial cells (see Table 2). The main functions shared by both modules are related to cell migration and multicellular organismal processing. As for the PAL Darkturquoise module genes, the main functions seem to be associated with the activity of different neural receptors. The PAL Darkgreen module’s functions mainly relate to meiosis (see Table 2).

In the optic tectum, 40 modules were identified (Supplementary Data Table S3a). Three modules were differently expressed between the treatment groups (Fig. 3l-n). These modules contained between 48 and 255 genes, of which 5–40 were classified as hub genes (see Table 1 for a summary of the treatment-sensitive modules and Supplementary Table S3a-c for an overview of all results, including the list of genes and hub genes).

All three significantly affected modules contained significantly enriched GO terms. OT Salmon included three (out of 4,304), OT Thistle1 contained 50 (out of 1,651), and OT Lightyellow had 317 (out of 4,688) significantly enriched GO terms. See Table 2 for the functional annotation of the top five significantly enriched GO terms (for all other GO terms, see Supplementary Data Table S3d-f).

The OT Salmon module showed three enriched GO terms related to cellular component functions, including protein-DNA complex and nucleosome DNA packaging complex (Table 2). The many enriched functions of the OT Thistle1 and OT Lightyellow modules are further summarised as semantic space scatter plots (Fig. 4d-e). The OT Thistle1 module revealed a predominance of cilia-related functions, such as axoneme assembly and cilia-dependent cell motility (cilia are abundant in brain cells⁵⁶). The OT Lightyellow module had many enriched functional terms (Fig. 4e). The most prominent functions are related to DNA transcription. Other functions are linked to biological regulations of the central nervous system, such as brain development, neural differentiation and neural metabolism. Interestingly, some enriched functions are related to behaviour, such as locomotion (Fig. 4e). Indeed, also the complete list of significantly enriched GO terms (Supplementary Table S3f) includes behaviour (GO:0007610), social behaviour (GO:0035176), adult behaviour (GO:0030534), eating behaviour (GO:0042755) and locomotion (GO:0040011).

Correlations of individual phenotypic trait variation with gene co-expression modules

To explore whether individual variation in gene expression is related to the expression of individual phenotypic traits, we tested for correlations of all identified gene modules with male courtship singing behaviour, aggression, testosterone and corticosterone profiles.

In the testis, four modules correlated positively with aggression (TES Lightsteelblue: r = 0.4005, p = 0.002; TES Navajowhite2: r = 0.263, p = 0.044; TES Mediumorchid: r = 0.320, p = 0.014; TES Purple: r = 0.266, p = 0.042) and three negatively (TES Cyan: r=-0.281, p = 0.031; TES Darkred: r=-0.323, p = 0.013; TES Coral2: r=-0.275, p = 0.035). Another three modules correlated positively with testosterone levels (TES Plum1: r = 0.283, p = 0.030; TES Honeydew1: r = 0.270, p = 0.038; TES Ivory: r = 0.258, p = 0.049) and three negatively (TES Greenyellow: r=-0.323, p = 0.013; TES Skyblue3: r=-0.322, p = 0.013; TES Mediumpurple2: r=-0.285, p = 0.029). Moreover, four modules showed negative correlations with the amount of time males spent singing (TES Skyblue: r=-0.316, p = 0.015; TES Skyblue2: r=-0.310, p = 0.017; TES Darkmagenta: r=-0.287, p = 0.028; TES Lavenderblush3: r=-0.265, p = 0.042). No correlations with the corticosterone levels were present (significant correlations are illustrated in Fig. 5a-q. For all results, see Supplementary Table S1i).

In the posterior pallium, two modules correlated positively with aggression (PAL Darkseagreen4: r = 0.326, p = 0.013; PAL Darkgrey: r = 0.296, p = 0.025) and one negatively (PAL Orangered4: r=-0.436, p = 0.001). Moreover, the module PAL Lightcyan1 correlated positively with testosterone (r = 0.301, p = 0.023). No correlations with the corticosterone levels or time males spent singing were present (significant correlations are illustrated in Fig. 5r-u. For all results, see Supplementary Table S2j).

In the optic tectum, two modules correlated positively with aggression (OT Purple: r = 0.349, p = 0.008; OT Darkorange: r = 0.290, p = 0.029), two modules correlated negatively with corticosterone (OT Greenyellow: r=-0.436, p = 0.019; OT Sienna3: r= -0.341, p = 0.022) and only the module OT Floralwhite correlated positively with testosterone (r = 0.306, p = 0.021). No correlations were present with song (significant correlations are illustrated in Fig. 5v-z. For all results, see Supplementary Table S3g).

In this pre-registered study⁴³, we hypothesised that elevated sperm competition risk experienced by Double-pair males would affect gene expression in the testis and brain compared to Single-pair males. In line with our expectations, we uncovered substantial differences in the expression of gene co-expression network modules in the testis, indicating treatment-driven plasticity of testis physiology. We also found differences between the treatment groups in several gene co-expression network modules in the two brain tissues, posterior pallium and optic tectum. In addition, tissue from both these brain areas contained a small number of individual genes differentially expressed between the treatment groups. Finally, we detected significant correlations between the expression of several gene network modules in individual transcriptomes and among-individual variation in aggression and hormone levels. In the following, we discuss our results for each target tissue and in relation to the hormonal and behavioural results, which were reported in detail in our complementary paper⁴⁴.

The effects we found on gene co-expression modules in the testis are particularly interesting in light of our pre-registered (non-directional) hypothesis. We expected our analysis to reveal differential gene expression between the two treatment groups in the testis, especially for candidate genes previously implicated in male-male competition⁴³. The testes are responsible for sperm but also androgen production, with downstream effects on tissues elsewhere, particularly the brain^57,58. The testes, thus, arguably constitute a target tissue of major importance for uncovering the effects of variation in sperm competition risk on gene expression. Contrary to our expectations, the treatment did not affect the expression of individual genes in the testis. Given the general exploratory framework of this study, we thus applied weighted gene co-expression networks, which revealed that the social treatment strongly affected five (out of 66) gene co-expression networks. The five highly correlated expression clusters contained 783 genes altogether. Our study thus implies that any phenotypically plastic response of traits involved in sperm competition is based on polygenic processes rather than a few individual genes.

These modules' many enriched functional annotations suggest that the effects we found on testis physiology were mostly related to germ cells. The module TES Orangered3 is particularly interesting in this regard. This module has been related to many cellular functions, including cell proliferation, which could influence germ cell functions and production. Socially sensitive expression of such gene networks could, for example, allow speeding up sperm production or sperm maturation under elevated sperm competition risk. In the complex process of spermatogenesis, stem cells undergo mitosis, meiosis, and cell differentiation to generate mature spermatozoa, which involves many cellular functions. One of the primary functions of the TES Orangered3 module is the regulation of the mitogen-activated protein kinases (MAPK) cascade. The MAPK cascade transmits various extracellular signals to their intracellular targets and regulates cell cycle progression and cell proliferation in eukaryotic cells⁵⁹. Interestingly, however, MAPK has also been implicated in spermatogenesis and different sperm functions⁶⁰, including sperm maturation⁶¹. In addition, traits like sperm morphology are prime candidates for phenotypically plastic responses to sperm competition^62,63. In the zebra finch, for example, longer (and thereby faster) sperm have an advantage in competitive fertilisation success⁶⁴. Furthermore, while avian sperm length is, moderately heritable⁶⁵ phenotypic plasticity in avian sperm length is well documented (e.g.^65,66). It is thus plausible that the many cellular functions of the TES Orangered3 module genes were related to sperm production and modification.

It is interesting to note that TES Orangered3 contained only eight hub genes (APOLD1, AFT3, CCN1, CCN2, DUSP1, EGR4, NR4A1 and OTUD1), which are promising candidates for the future study of sperm competition effects. Some of these genes have already been documented to play a role in germ cell production. For example, DUSP1 is involved in spermatogonia differentiation in rats⁶⁷, and EGR4 is important for mammalian fertility and is expressed in both testicular and germ cells⁶⁸. The gene NR4A1, on the other hand, is a nuclear receptor important for hormone-induced steroidogenesis in the Leydig cells of the testes in mammals⁶⁹. The Leydig cells are considered the main source of testicular androgens, primarily testosterone, in mammals⁷⁰ and have similar functions in the gonads of birds⁷¹. Testicular androgens play a crucial role in regulating male reproductive success in birds by promoting courtship⁷² and the production of sperm⁷³. Gonadal androgens act as body-wide plasticity mediators and are produced in feedback loops with the brain, where they regulate brain functions related to sexual and aggressive behaviours⁷⁴. Thus, the uncovered hub genes of the TES Orangered3 module further support our hypothesis that the social treatment in our study will have had an impact on traits related to sperm competition in zebra finch males.

Our pre-registered experimental design involved the analysis of differential gene expression in the septum⁴³. For technical reasons, we could not extract a sufficient amount of RNA from this brain area. We thus analysed the transcriptome of the posterior pallium and the optic tectum, the latter serving as a control region. Our treatment affected the transcriptomes in both brain tissues, indicating responses in neural functions that regulate behavioural phenotypes. As reported in our accompanying paper⁴⁴, zebra finch males adjusted their behavioural phenotypes to the manipulated social environment indicating social niche conformance⁴. However, contrary to our pre-registered expectations⁴³, there was no evidence that these behavioural adjustments were related to the elevated sperm competition risk. We expected the Double-pair males to show more courtship displays, including songs directed towards their social mate, more affiliative and copulation behaviour with their social mate and more aggression towards competitors than the control group⁴³. On the contrary, males of the Double-pair treatment group decreased courtship rates (song), and when confronted with an unfamiliar intruder, they responded less aggressively than Single-pair males⁴⁴. No other behavioural measures revealed significant treatment effects (see⁴⁴ for all details). These unexpected effects suggest that in our study, we observed adjustment of social behaviour to social factors other than the elevated sperm competition risk, e.g., reactions to mild social isolation. The presence of only one other male in a large breeding cage may have been too weak a stimulus to upregulate competitive behavioural traits in the Double-pair treatment. Instead, males in this condition probably habituated to the presence of the other breeding pair and became more tolerant towards novel male intruders. On the contrary, Single-pair males were deprived of the exposure to other males that is typical for this highly social colonial breeder. Consequently, Single-pair males reacted with increased aggression to the appearance of a new male. This is in line with studies showing that zebra finches with reduced exposure to social partners react more aggressively towards competitors than those held in larger groups^75,76 (for similar results in mammals and fish, see, e.g.^77,78). Nevertheless, the social treatment induced behavioural phenotypic plasticity reflected in the brain transcriptome. This provides valuable evidence on the brain gene expression patterns underlying behavioural adjustment to different social environments.

Both brain tissues contained some differently expressed individual genes. Although their number was rather low (ten in the posterior pallium and six in the optic tectum), this provided a first indication of social treatment-dependent transcriptomic changes in the brain. However, like in the testis, differences were much more substantial at the level of gene co-expression networks. Six (out of 49) co-expression modules in the posterior pallium, containing 2,830 genes overall, showed differences between treatments. The enriched functional annotations of two gene co-expression modules, PAL Blue and PAL Darkturquoise, indicate modifications related to neural functions in the posterior pallium. The enriched functions of the module PAL Saddlebrown showed changes related to the development and differentiation of glial and endothelial cells. Together this indicates that the social treatment induced modifications of neural processes, which were accompanied by changes also in the glial cells. Changes in endothelial cells may also indicate cerebrovascular plasticity, which is needed to adjust blood supply to changes in the metabolic demands of neural and glial cells⁷⁹.

Such social experience-dependent modulation of gene expression in the posterior pallium was expected as many of the functions of this large brain region regulate social behaviours at different levels. Specifically, the samples from the posterior pallium contained a large portion of the posterior nidopallium, almost the entire arcopallium, the posterior amygdala and the nucleus taenia of the amygdala. At large, the arcopallium and nucleus taenia of the amygdala in birds regulate fear⁸⁰. This region responds to novel stimuli, such as exposures to novel environments⁸¹ or novel objects⁸² and first encounters with conspecifics in naïve birds⁸³. Moreover, the nucleus taenia of the amygdala is part of the social behaviour network, which is shared among all vertebrates. It comprises interconnected areas rich in sex steroid receptors and is implicated in a range social behaviours, including aggression^48,84,85.

While the results we observed when analysing this large section of the brain are extremely helpful for our explorative purpose, it remains largely unclear which specific brain processes and regions were affected by our treatment. The arcopallium is a large brain region, which in zebra finches has been divided into six major domains with twenty distinct sub-regions⁸⁶. The arcopallium receives inputs from numerous brain areas and is a major source of descending sensory and motor projections. It can thus be considered a key brain region of the avian forebrain⁸⁶. Likewise, the avian posterior nidopallium is a large brain region supporting many functions, from working memory⁸⁷, executive functions⁸⁸ and visual categorisation⁸⁹ to sexual imprinting in zebra finches⁹⁰. Which of these brain functions were affected by our treatment needs further investigation.

We also found changes in gene expression in the optic tectum, which was unexpected. The optic tectum, located in the dorsal midbrain of birds, is the primary recipient of around 80% of retinal inputs. Its main function is the generation of orienting responses to stimuli of interest, especially when they are moving⁹¹. These responses are considered innate or reflexive⁹². At this early stage of visual processing, we did not expect to find plastic adjustments to our social treatment. Contrary to our expectation, three (out of 40) gene co-expression modules were differently expressed between the treatment levels, indicating that some lasting adjustment to the social environment is also present in the optic tectum. This is an interesting finding, which may be explained by the organisation of the tectofugal visual pathway. In birds, the tectofugal visual pathway stretches from the retina to the optic tectum, then to the nucleus rotundus in the thalamus, before reaching the entopallium in the forebrain. The entopallium sends projections to higher telencephalic regions, including the arcopallium⁹³. The arcopallium, in turn, projects back on the optic tectum, completing a tecto-tectal loop⁹⁴. It is thus possible that the plastic changes found in the posterior pallium induced changes in the optic tectum through this loop. The tectofugal visual pathway in birds is involved, among other things, in perception and attention to object and shape information^89,95,96. The projections from arcopallium to tectum could thus mediate preferential attention to visual social stimuli, such as the visual appearance of male and female conspecifics. Our study provides the first evidence of tectal adjustment to the social environment in zebra finches, which could support alterations in basic mechanisms of visual attention and the detection of social stimuli. The neural basis of this interesting phenomenon needs to be addressed explicitly in future neurobiological studies on the optic tectum.

Our study's large sample of individual transcriptomes also allowed us to analyse correlations between the individuals’ transcriptomes and their behavioural and hormonal phenotypes according to our pre-registered design⁴³. As expected, our results suggest that individual variation in gene expression was related to among-individual differences in hormonal and behavioural phenotypes. All three tissues revealed correlations with the eigengene expression of several co-expression modules. Aggression correlated with seven co-expression modules in the testis, three modules in the pallium and two in the optic tectum. Testosterone levels correlated with six modules in the testis, one in the pallium and one in the optic tectum. Amount of song correlated with four modules in the testis. While this analysis is highly explorative at this stage, it indicates that individual behavioural and hormonal phenotypes are reflected in the gene expression patterns.

In conclusion, as predicted, manipulating the opportunity for extrapair mating and, thereby, the risk of sperm competition induced changes in gene expression in the testes and brains of male zebra finches. These social treatment effects were first and foremost apparent at the level of gene co-expression networks, indicating that biological traits subject to phenotypic adjustment to the social environment are based on polygenic processes rather than a few individual genes with large effects. The changes in the testes can be specifically associated with sexual competition. In contrast, changes in the posterior pallium and optic tectum can be associated with behavioural adjustments of male zebra finches to other concurrent changes in the social environment. Our study indicates the importance of social environment and intrasexual competition for the phenotypic plasticity of zebra finch males. Many questions remain, but the present study opens new doors for expanding our understanding of the mechanisms behind social niche conformance.

Ethical statement

All procedures complied with the regulations covering animal experimentation within Germany (Animal Welfare Act) and the EU (European Community Council Directive 2010/63/EU), and were approved by the Landesamt für Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen (case number 81‐02.04.2018.A311).

Experimental approach and general methods

This study has been pre-registered at the Open Science Framework⁴³, and its behavioural procedures and methods have been described in detail in a complementary paper⁴⁴. Here we only summarise the main aspects in reference to⁴⁴. In brief, zebra finch breeding pairs were randomly allocated to one of two experimental treatment groups, Single-pair versus Double-pair (Fig. 1). Thus, we exposed the breeding pairs to two different social environments reflecting two different levels of sperm competition risk and social stimulation. Only the males of the Double-pairs were exposed to sperm competition risk and direct physical interaction with another male.

At the start of the experimental treatments (ca. 89 days after pair formation), we transferred Single-pair males to new breeding cages together with their established social mates. Double-pair males were transferred to a double-sized breeding cage, together with their social mate and an additional unrelated and unfamiliar social pair. All birds were fitted with three additional leg rings of the same colour to allow for individual identification during behavioural observations. The provision of nest boxes and nesting material stimulated the birds to start breeding. We stimulated breeding pairs to produce two consecutive clutches to give the focal males enough time to respond to the experimental treatment. For both clutches, plastic dummy eggs replaced the eggs on the day they were laid. We removed the first dummy clutch 15 days after the first egg was laid, to induce the production of the second clutch (replacement clutch). Data and sample collection from the males were scheduled with regard to the timing of laying of the replacement clutch. This procedure ensured that the timing of data collection was standardised according to the reproductive cycle of each breeding pair, and that females were still receptive when male competitive behavioural traits were recorded. Multiple individual phenotypic traits were recorded before sacrificing the males. Finally, on the fourth day after the first egg of the replacement clutch was laid, immediately after the intruder test, we sacrificed 30 males from each treatment level (Double-pair and Single-pair) for gene expression analyses targeting testes and two brain areas (posterior pallium and optic tectum).

Measurement of phenotypes

Behavioural recordings. One day after the first egg of the replacement clutch was laid, we recorded the birds’ behaviours in the breeding cage (2 hours). We quantified male courtship behaviour as total singing duration (min) consisting of song directed to the social mate (or the non-focal female in the case of the Double-pair treatment group) and undirected (for details, see Supplementary Table S5 in⁴⁴). Four days after the replacement clutch was initiated, we staged a standardised intruder test, in which we measured the focal males’ aggressive responses against an unfamiliar male introduced into their breeding cage (10 min). Aggression was defined as the total time (min) the focal and the stimulus male spent in any aggressive interactions (for details, see Supplementary Table S6 in⁴⁴). Behavioural recordings (i.e. singing and aggression) were used for correlation analyses in relation to individual-specific patterns of gene expression (see below).

Hormone sampling. We took one blood sample from each focal male two days after the first egg of the replacement clutch was laid. The sample was analysed for plasma concentrations of testosterone and corticosterone (see⁴³). We used hormone concentrations for correlation analyses in relation to individual-specific patterns of gene expression (see below).

Tissue dissection

Focal males were sacrificed by decapitation. The brain was extracted from the skull within 3–5 minutes, snap-frozen on dry ice and stored at -80°C until further processing. This was followed by extraction of the left testis (which usually represents the bigger testis⁹⁷). After the mass of the extracted testis was measured on a standard laboratory scale (Sartorius Quintex 124-1S) to a precision of 0.1 µg, it was cut into two approximately equal pieces. One half was incubated for 24 hours in RNAlater™ Stabilization Solution (Invitrogen, catalogue number: AM7021) and stored at -80°C until RNA extraction.

We extracted the posterior pallium and optic tectum, which deviates from what was declared in the pre-registration⁴³, as we initially planned to analyse the transcriptome of the septum (a core region of the social behaviour network). However, the small septum did not provide sufficient RNA for individual-level transcriptome analysis. We thus extracted larger brain regions with similar functions and added a control region from the sub-pallium (the optic tectum). Before extraction, the brains were stored for 2 hours at -20°C. This preserved brain anatomical structures while softening the frozen tissue to allow dissection. The brain regions of interest were then extracted on ice using binoculars. The anatomical areas were located and delineated based on visual observation of anatomical landmarks as referenced in the zebra finch brain atlas⁹⁸. The pallial samples were taken from a region approximating the one between anterior A 1.08 to posterior P 0.18 and including the entire tissue lateral to L 1.0. It thus included a large portion of the posterior nidopallium, almost the whole arcopallium, the posterior amygdala and the nucleus taenia of the amygdala. We did not include the most dorsal parts of the telencephalon to avoid inadvertently sampling from the hippocampal formation (HF) and the high vocal centre (HVc). The tectal samples were extracted from the right hemisphere approximating a region from A 2.43 to A 1.35 and including the entire tissue lateral to L 3.0. The samples thus included all the layers of the optic tectum as well as parts of the torus semicircularis (TOS), substantia grisea et fibrosa periventriculare (SGP), nucleus mesencephalicus lateralis (MLd, MLv) and nucleus isthmi (IPC, IM). The isolated brain regions were incubated in RNAlater™ for 24 hours at 4°C and then stored at -80°C until RNA extraction.

RNA isolation and sequencing

Total RNA was extracted using the RNeasy Mini Kit (Qiagen, catalogue number 74106) for total RNA extractions. In addition to the protocol supplied with this kit, RNA samples were purified with RNase-Free DNase Set (Qiagen, catalogue number 79254) to eliminate possible DNA contamination. After RNA extraction, the quality of extracted RNA samples was tested with the 2100 Bioanalyzer System (Agilent). This revealed that for all of the samples the RNA integrity numbers (RIN) were > 8, indicating high-quality RNA. Extracted total RNA samples were subsequently sent to the Beijing Genomics Institute (BGI, www.genomics.org.cn) for complementary DNA library construction and mRNA sequencing using the DNBseq platform. mRNA-seq libraries were prepared using the BGI in-house library preparation kit (BGI, www.genomics.org.cn) and sequenced with 100bp paired-end reads. Read filtering was performed by SOAPnuke⁹⁹. Quality control of raw reads was performed with FastQC v0.11.9¹⁰⁰. On average 74 million clean reads were obtained for each sample, with an average Phred quality score of 37 after trimming.

Genome mapping

We used hisat2 v.2.2.1⁵¹ for mapping the reads to the zebra finch reference genome (GenBank assembly accession: GCF_008822105.2). We used strict mapping options (--no-discordant --no-mixed) to avoid mismapping. On average 85% of the reads per sample were uniquely mapped to the genome. The gtf-file of the zebra finch reference genome, in combination with the mapped reads, was used to generate the reads count using htseq-count v.0.6.1⁵⁰, a Python-based framework to analyse sequencing data (settings used: --format bam --order pos --mode union --stranded no --minaqual 1 --type exon).

Differential gene expression

Differential expression analyses (DEGs) concerning the experimental treatment were conducted with the R package DEseq2 (v.1.26.0)¹⁰¹ separately for each tissue (using the model: ~Treatment). The false discovery rate (FDR) q-value for the DEseq2 was set as default (FDR < 0.1). Genes with fewer than ten total reads across all samples for any tissue were filtered out. For visualisation, the data were normalised using the standard formula (log2(counts + 1)), as suggested by the manual of the DEseq2 (v.1.26.0)¹⁰¹ package. We used the clustering of samples by Principal Component Analysis to visualise gene expression in different target tissue types. Volcano plots were used to visualise differential gene expression between treatments separately for each tissue.

Weighed correlation network analysis (WGCNA)

Weighted correlation network analysis (WGCNA)⁴⁹ was used to identify clusters (modules) of highly co-expressed genes. Modules are composed of genes with a similar co-expression pattern, creating a network centred on the so-called hub genes (genes with a relatively high number of connections with other genes within the co-expression network are denoted as network hubs⁵³). Since genes act together in shared pathways and networks, this analysis provides additional insights into even subtle but network-wide expression changes related to the experimental treatment^52,53. Also, the genes with fewer than ten total reads across all samples for any tissue were filtered out. We used the variance-stabilizing transformed data (generated with the VST function of the DEseq2 R package) as input for WGCNA.

The number of genes included in this analysis for the posterior pallium, optic tectum and testis were 18,917, 19,030 and 20,380, respectively. We created signed networks by selecting soft threshold power (β) using the pickSoftThreshold function (for posterior pallium β = 7, for tectum β = 6, for testis β = 7). We kept modules with a minimum module size of 30 genes. A threshold of 0.25 was used to merge modules. We identified hub genes in significantly associated modules as those with gene significance values above 0.2 and module memberships above 0.6.

We tested for differences in module eigengene expression between treatments using ANOVAs. Moreover, to identify potential correlations between gene co-expression modules and individual phenotypic traits (male song duration, aggression, testosterone and corticosterone plasma levels), we used Pearson correlations. All statistical analyses were performed using R¹⁰².

Functional annotation analysis

To specify the biological process associated with the genes of the modules significantly affected by the experimental treatment we performed gene ontology (GO)⁵⁴ and KEGG (Kyoto Encyclopaedia of Genes and Genomes)¹⁰³ enrichment analysis, also called over-representation analysis (ORA). This was done using the WEB-based toolkit g:Profiler¹⁰⁴ (https://biit.cs.ut.ee/gprofiler/gost; g:Profiler version e104_eg51_p15_3922dba [October 2021]). The analysis is based on gene annotation terms, to determine differences in the processes/functions associated with a subset of genes (in this case, those of each co-expression module), compared to the functional annotations of the full list of all known zebra finch genes for the same tissue. To do so, we ran a functional enrichment analysis on the list of candidate genes (un-ordered list) for each significant module from the WGCNA, in relation to the experimental treatment, with an FDR < 0.05. The main results were further summarised by semantic similarity scatterplots visualised by REVIGO¹⁰⁵.

Data availability

RNA sequencing data will be available through the European Nucleotide Archive (https://www.ebi.ac.uk/ena/browser/home) after acceptance for publication.

Code availability

Only standard R packages, described in the methods section with references, were used.

Author contributions

PK and TS conceived the study in consultation with UM. PK and TS obtained funding for the study. PK, TS, NDLL and SR designed and planned the experiment in consultation with UM. NDLL and SR coordinated and performed the experiment. SR and UM collected tissue samples with the help of NDLL and PK. AJ and SR analysed the data. UM organised the results. UM, TS and SR wrote the manuscript with support from PK, AJ and NDLL. All authors reviewed, commented on and approved the final version of the manuscript. PK, UM and TS jointly supervised the project.

Acknowledgements

This project was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) as part of the SFB TRR 212 A Novel Synthesis of Individualization across Behaviour, Ecology and Evolution: Niche Choice, Niche Conformance, Niche Construction (NC³) — project numbers 316099922 and 396777869. We are thankful to the members of the NC³ community for the discussion. We thank Stefanie Bölting and Kathrin Engel for sharing their knowledge on zebra finch biology and providing administrative support during the preparation of the experiment. We thank Arne Kalinowski, Stephen Salazar, Jonathan Grone, Elke Hippauf and Anna Hamker for technical support during and after the experiment. We thank Katrin Lehmann and Tony Rinaud for assistance with RNA extraction and quality assessment. We thank Sonja Schindler for administrative assistance in the RNA lab. We thank Peter Liu from BGI Genomics for his guidance regarding tissue sampling and RNA extraction. We thank Christian Klein, Anabel Kröhnert, Sophia Marie Quante, Katja de Rooy, and Carolin Stober for their help with analysing the behavioural recordings. We thank Sylvia Kaiser and Sabine Kruse for analysing the hormone samples and sharing their knowledge on endocrinology. We are thankful to the animal caretakers, especially Ursula Kodytek, Kristina Walkenfort, and Jakob Besgen, who diligently cared for the study animals.

Competing interests

The authors declare no competing interests

Reporting Summary

Correspondence and material requests should be addressed to SR or UM or TS

Via, S. Adaptive phenotypic plasticity: target or by-product of selection in a variable environment? Am Nat 142, 352–365 (1993).
Pigliucci, M. Phenotypic Plasticity. (2001). doi:10.56021/9780801867880.
Nussey, D. H., Wilson, A. J. & Brommer, J. E. The evolutionary ecology of individual phenotypic plasticity in wild populations. J Evol Biol 20, 831–844 (2007).
Trappes, R. et al. How Individualized Niches Arise: Defining Mechanisms of Niche Construction, Niche Choice, and Niche Conformance. Bioscience 72, 538–548 (2022).
Bergmüller, R. & Taborsky, M. Animal personality due to social niche specialisation. Trends Ecol Evol 25, 504–511 (2010).
Schradin, C. Intraspecific variation in social organization by genetic variation, developmental plasticity, social flexibility or entirely extrinsic factors. Philos Trans R Soc Lond B Biol Sci 368, 20120346 (2013).
Montiglio, P.-O., Wey, T. W. & Sih, A. Effects of the group’s mix of sizes and personalities on the emergence of alternative mating systems in water striders. Behavioral Ecology 28, 1068–1074 (2017).
Fraser, B. A., Janowitz, I., Thairu, M., Travis, J. & Hughes, K. A. Phenotypic and genomic plasticity of alternative male reproductive tactics in sailfin mollies. Proc Biol Sci 281, 20132310 (2014).
Darwin, C., Bonner, J. T. & May, R. M. The Descent of Man, and Selection in Relation to Sex. (Princeton University Press, 1981).
Andersson, M. Sexual Selection. (Princeton University Press, 1994).
Andersson, M. & Simmons, L. W. Sexual selection and mate choice. Trends Ecol Evol 21, 296–302 (2006).
Jones, A. G. & Ratterman, N. L. Mate choice and sexual selection: what have we learned since Darwin? Proc Natl Acad Sci U S A 106 Suppl 1, 10001–10008 (2009).
Parker, G. A. Sperm Competition and Its Evolutionary Consequences in the Insects. Biological Reviews 45, 525–567 (1970).
Pitnick, S. S., Hosken, D. J. & Birkhead, T. R. Sperm Biology: An Evolutionary Perspective. (Academic Press, 2008).
Eberhard, W. Female Control: Sexual Selection by Cryptic Female Choice. (Princeton University Press, 1996).
Firman, R. C., Gasparini, C., Manier, M. K. & Pizzari, T. Postmating Female Control: 20 Years of Cryptic Female Choice. Trends Ecol Evol 32, 368–382 (2017).
Parker, G. A. The sexual cascade and the rise of pre-ejaculatory (Darwinian) sexual selection, sex roles, and sexual conflict. Cold Spring Harb Perspect Biol 6, a017509 (2014).
Janicke, T., Häderer, I. K., Lajeunesse, M. J. & Anthes, N. Darwinian sex roles confirmed across the animal kingdom. Sci Adv 2, e1500983 (2016).
Dougherty, L. R. Meta-analysis reveals that animal sexual signalling behaviour is honest and resource based. Nat Ecol Evol 5, 688–699 (2021).
Kelly, C. D. & Jennions, M. D. Sexual selection and sperm quantity: meta-analyses of strategic ejaculation. Biol Rev Camb Philos Soc 86, 863–884 (2011).
Hopkins, B. R. et al. Divergent allocation of sperm and the seminal proteome along a competition gradient in Drosophila melanogaster. Proc Natl Acad Sci U S A 116, 17925–17933 (2019).
Tuni, C., Weber, S., Bilde, T. & Uhl, G. Male spiders reduce pre- and postmating sexual investment in response to sperm competition risk. Behavioral Ecology 28, 1030–1036 (2017).
Bretman, A., Fricke, C., Hetherington, P., Stone, R. & Chapman, T. Exposure to rivals and plastic responses to sperm competition in Drosophila melanogaster. Behavioral Ecology 21, 317–321 (2010).
Wang, H., Duclot, F., Liu, Y., Wang, Z. & Kabbaj, M. Histone deacetylase inhibitors facilitate partner preference formation in female prairie voles. Nat Neurosci 16, 919–924 (2013).
Bird, A. Perceptions of epigenetics. Nature 447, 396–398 (2007).
Todd, E. V., Black, M. A. & Gemmell, N. J. The power and promise of RNA-seq in ecology and evolution. Mol Ecol 25, 1224–1241 (2016).
Bloch, N. I. et al. Early neurogenomic response associated with variation in guppy female mate preference. Nat Ecol Evol 2, 1772–1781 (2018).
Bentz, A. B. et al. Experimental competition induces immediate and lasting effects on the neurogenome in free-living female birds. Proc Natl Acad Sci U S A 118, e2016154118 (2021).
Anderson, A. P., Rose, E., Flanagan, S. P. & Jones, A. G. The Estrogen-Responsive Transcriptome of Female Secondary Sexual Traits in the Gulf Pipefish. J Hered 111, 294–306 (2020).
Schunter, C., Vollmer, S. V., Macpherson, E. & Pascual, M. Transcriptome analyses and differential gene expression in a non-model fish species with alternative mating tactics. BMC Genomics 15, 167 (2014).
Bukhari, S. A. et al. Temporal dynamics of neurogenomic plasticity in response to social interactions in male threespined sticklebacks. PLoS Genet 13, e1006840 (2017).
Eswine, S. L., Pontinen, J. K. & Heimovics, S. A. Competitive ability during mate competition relates to unique patterns of dopamine-related gene expression in the social decision-making network of male zebra finches. Neurosci Lett 706, 30–35 (2019).
Connahs, H. et al. The yellow gene regulates behavioural plasticity by repressing male courtship in Bicyclus anynana butterflies. Proc Biol Sci 289, 20212665 (2022).
Aubin-Horth, N., Landry, C. R., Letcher, B. H. & Hofmann, H. A. Alternative life histories shape brain gene expression profiles in males of the same population. Proc Biol Sci 272, 1655–1662 (2005).
Ramm, S. A. et al. Sex allocation plasticity on a transcriptome scale: Socially sensitive gene expression in a simultaneous hermaphrodite. Mol Ecol 28, 2321–2341 (2019).
Magrath, M. J. L. & Komdeur, J. Is male care compromised by additional mating opportunity? Trends in Ecology & Evolution 18, 424–430 (2003).
Fromhage, L. & Jennions, M. D. Coevolution of parental investment and sexually selected traits drives sex-role divergence. Nat Commun 7, 12517 (2016).
Bukhari, S. A. et al. Neurogenomic insights into paternal care and its relation to territorial aggression. Nat Commun 10, 4437 (2019).
Forstmeier, W., Martin, K., Bolund, E., Schielzeth, H. & Kempenaers, B. Female extrapair mating behavior can evolve via indirect selection on males. Proc Natl Acad Sci U S A 108, 10608–10613 (2011).
Forstmeier, W. & Birkhead, T. R. Repeatability of mate choice in the zebra finch: consistency within and between females. Animal Behaviour 68, 1017–1028 (2004).
Birkhead, T. R., Fletcher, F., Pellatt, E. J. & Staples, A. Ejaculate quality and the success of extra-pair copulations in the zebra finch. Nature 377, 422–423 (1995).
Birkhead, T. R., Hunter, F. M. & Pellatt, J. E. Sperm competition in the zebra finch, Taeniopygia guttata. Animal Behaviour 38, 935–950 (1989).
Lilie, N. D., Riyahi, S., Korsten, P. & Schmoll, T. Male social niche conformance in zebra finches. OSF Preregistration (2019) doi:10.17605/OSF.IO/84Z5R.
Lilie, N. D. et al. Male social niche conformance? Effects of manipulated opportunity for extra-pair mating on behavior and hormones of male zebra finches. Horm Behav 146, 105243 (2022).
Hurley, L. L. et al. Longitudinal covariation of testosterone and sperm quality across reproductive stages in the zebra finch. Horm Behav 153, 105388 (2023).
Herold, C., Paulitschek, C., Palomero-Gallagher, N., Güntürkün, O. & Zilles, K. Transmitter receptors reveal segregation of the arcopallium/amygdala complex in pigeons (Columba livia). J. Comp. Neurol. 526, 439–466 (2018).
Abellán, A., Legaz, I., Vernier, B., Rétaux, S. & Medina, L. Olfactory and amygdalar structures of the chicken ventral pallium based on the combinatorial expression patterns of LIM and other developmental regulatory genes. J. Comp. Neurol. 516, 166–186 (2009).
O’Connell, L. A. & Hofmann, H. A. The vertebrate mesolimbic reward system and social behavior network: a comparative synthesis. J. Comp. Neurol. 519, 3599–3639 (2011).
Langfelder, P. & Horvath, S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 9, 559 (2008).
Anders, S., Pyl, P. T. & Huber, W. HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics 31, 166–169 (2015).
Kim, D., Langmead, B. & Salzberg, S. L. HISAT: a fast spliced aligner with low memory requirements. Nat Methods 12, 357–360 (2015).
Iancu, O. D., Colville, A., Darakjian, P. & Hitzemann, R. Coexpression and cosplicing network approaches for the study of mammalian brain transcriptomes. Int Rev Neurobiol 116, 73–93 (2014).
Zhang, B. & Horvath, S. A general framework for weighted gene co-expression network analysis. Stat Appl Genet Mol Biol 4, Article17 (2005).
Ashburner, M. et al. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet 25, 25–29 (2000).
Tomczak, A. et al. Interpretation of biological experiments changes with evolution of the Gene Ontology and its annotations. Sci Rep 8, 5115 (2018).
Guemez-Gamboa, A., Coufal, N. G. & Gleeson, J. G. Primary cilia in the developing and mature brain. Neuron 82, 511–521 (2014).
Ball, G. F. & Balthazart, J. Seasonal and hormonal modulation of neurotransmitter systems in the song control circuit. J Chem Neuroanat 39, 82–95 (2010).
Frankl-Vilches, C. & Gahr, M. Androgen and estrogen sensitivity of bird song: a comparative view on gene regulatory levels. J Comp Physiol A Neuroethol Sens Neural Behav Physiol 204, 113–126 (2018).
Cargnello, M. & Roux, P. P. Activation and function of the MAPKs and their substrates, the MAPK-activated protein kinases. Microbiol Mol Biol Rev 75, 50–83 (2011).
Sun, Q. Y., Breitbart, H. & Schatten, H. Role of the MAPK cascade in mammalian germ cells. Reprod Fertil Dev 11, 443–450 (1999).
Li, M. W. M., Mruk, D. D. & Cheng, C. Y. Mitogen-activated protein kinases in male reproductive function. Trends Mol Med 15, 159–168 (2009).
Knief, U. et al. A sex-chromosome inversion causes strong overdominance for sperm traits that affect siring success. Nat Ecol Evol 1, 1177–1184 (2017).
Kim, K.-W. et al. A sex-linked supergene controls sperm morphology and swimming speed in a songbird. Nat Ecol Evol 1, 1168–1176 (2017).
Bennison, C., Hemmings, N., Slate, J. & Birkhead, T. Long sperm fertilize more eggs in a bird. Proc Biol Sci 282, 20141897 (2015).
Edme, A. et al. Moderate heritability and low evolvability of sperm morphology in a species with high risk of sperm competition, the collared flycatcher Ficedula albicollis. J Evol Biol 32, 205–217 (2019).
Schmoll, T., Kleven, O. & Rusche, M. Individual phenotypic plasticity explains seasonal variation in sperm morphology in a passerine bird. Evol Ecol Res 19, 547–560 (2018).
Ryser, S. et al. Gene expression profiling of rat spermatogonia and Sertoli cells reveals signaling pathways from stem cells to niche and testicular cancer cells to surrounding stroma. BMC Genomics 12, 29 (2011).
Hogarth, C. A., Mitchell, D., Small, C. & Griswold, M. EGR4 displays both a cell- and intracellular-specific localization pattern in the developing murine testis. Dev Dyn 239, 3106–3114 (2010).
Daems, C., Martin, L. J., Brousseau, C. & Tremblay, J. J. MEF2 is restricted to the male gonad and regulates expression of the orphan nuclear receptor NR4A1. Mol Endocrinol 28, 886–898 (2014).
Wang, Y., Chen, F., Ye, L., Zirkin, B. & Chen, H. Steroidogenesis in Leydig cells: effects of aging and environmental factors. Reproduction 154, R111–R122 (2017).
Estermann, M. A., Major, A. T. & Smith, C. A. Genetic Regulation of Avian Testis Development. Genes (Basel) 12, 1459 (2021).
Fusani, L. Testosterone control of male courtship in birds. Horm Behav 54, 227–233 (2008).
Hau, M. Regulation of male traits by testosterone: implications for the evolution of vertebrate life histories. BioEssays 29, 133–144 (2007).
Mhaouty-Kodja, S. Role of the androgen receptor in the central nervous system. Mol Cell Endocrinol 465, 103–112 (2018).
Ruploh, T., Bischof, H.-J. & von Engelhardt, N. Adolescent social environment shapes sexual and aggressive behaviour of adult male zebra finches (Taeniopygia guttata). Behav Ecol Sociobiol 67, 175–184 (2013).
Ruploh, T., Bischof, H.-J. & von Engelhardt, N. Social experience during adolescence influences how male zebra finches (Taeniopygia guttata) group with conspecifics. Behav Ecol Sociobiol 68, 537–549 (2014).
Zimmermann, T. D., Kaiser, S., Hennessy, M. B. & Sachser, N. Adaptive shaping of the behavioural and neuroendocrine phenotype during adolescence. Proceedings of the Royal Society B: Biological Sciences 284, 20162784 (2017).
Arnold, C. & Taborsky, B. Social experience in early ontogeny has lasting effects on social skills in cooperatively breeding cichlids. Animal Behaviour 79, 621–630 (2010).
Bogorad, M. I., DeStefano, J. G., Linville, R. M., Wong, A. D. & Searson, P. C. Cerebrovascular plasticity: Processes that lead to changes in the architecture of brain microvessels. J Cereb Blood Flow Metab 39, 1413–1432 (2019).
Phillips, R. E. & Youngren, O. M. Unilateral kainic acid lesions reveal dominance of right archistriatum in avian fear behavior. Brain Res. 377, 216–220 (1986).
Morandi-Raikova, A. & Mayer, U. The effect of monocular occlusion on hippocampal c-Fos expression in domestic chicks (Gallus gallus). Sci Rep 10, 7205 (2020).
Perez, E. C. et al. Object and food novelty induce distinct patterns of c-fos immunoreactivity in amygdala and striatum in domestic male chicks (Gallus gallus domesticus). Behav Brain Res 381, 112453 (2020).
Mayer, U., Rosa-Salva, O. & Vallortigara, G. First exposure to an alive conspecific activates septal and amygdaloid nuclei in visually-naïve domestic chicks (Gallus gallus). Behav. Brain Res. (2016) doi:10.1016/j.bbr.2016.09.031.
Goodson, J. L. The vertebrate social behavior network: evolutionary themes and variations. Horm Behav 48, 11–22 (2005).
Newman, S. W. The medial extended amygdala in male reproductive behavior. A node in the mammalian social behavior network. Ann. N. Y. Acad. Sci. 877, 242–257 (1999).
Mello, C. V., Kaser, T., Buckner, A. A., Wirthlin, M. & Lovell, P. V. Molecular architecture of the zebra finch arcopallium. J Comp Neurol 527, 2512–2556 (2019).
Hahn, L. A., Balakhonov, D., Fongaro, E., Nieder, A. & Rose, J. Working memory capacity of crows and monkeys arises from similar neuronal computations. Elife 10, e72783 (2021).
Güntürkün, O., von Eugen, K., Packheiser, J. & Pusch, R. Avian pallial circuits and cognition: A comparison to mammals. Curr Opin Neurobiol 71, 29–36 (2021).
Pusch, R., Clark, W., Rose, J. & Güntürkün, O. Visual categories and concepts in the avian brain. Anim Cogn 26, 153–173 (2023).
Lieshoff, C., Große-Ophoff, J. & Bischof, H.-J. Sexual imprinting leads to lateralized and non-lateralized expression of the immediate early gene zenk in the zebra finch brain. Behavioural Brain Research 148, 145–155 (2004).
Wylie, D. R. W., Gutierrez-Ibanez, C., Pakan, J. M. P. & Iwaniuk, A. N. The optic tectum of birds: mapping our way to understanding visual processing. Can J Exp Psychol 63, 328–338 (2009).
Rosa Salva, O., Mayer, U. & Vallortigara, G. Roots of a social brain: developmental models of emerging animacy-detection mechanisms. Neurosci Biobehav Rev 50, 150–168 (2015).
Clark, W. J. & Colombo, M. The functional architecture, receptive field characteristics, and representation of objects in the visual network of the pigeon brain. Prog Neurobiol 195, 101781 (2020).
Bischof, H. J. & Watanabe, S. On the structure and function of the tectofugal visual pathway in laterally eyed birds. Eur J Morphol 35, 246–254 (1997).
Watanabe, S., Mayer, U. & Bischof, H.-J. Visual Wulst analyses ‘where’ and entopallium analyses ‘what’ in the zebra finch visual system. Behav. Brain Res. 222, 51–56 (2011).
Knudsen, E. I. Evolution of neural processing for visual perception in vertebrates. J Comp Neurol 528, 2888–2901 (2020).
Birkhead, T. R., Fletcher, F. & Pellatt, E. J. Sexual selection in the zebra finch Taeniopygia guttata : condition, sex traits and immune capacity. Behav Ecol Sociobiol 44, 179–191 (1998).
Nixdorf-Bergweiler, B. E., Bischof, H.-J., Nixdorf-Bergweiler, B. E. & Bischof, H.-J. A Stereotaxic Atlas Of The Brain Of The Zebra Finch, Taeniopygia Guttata. (National Center for Biotechnology Information (US), 2007).
Chen, Y. et al. SOAPnuke: a MapReduce acceleration-supported software for integrated quality control and preprocessing of high-throughput sequencing data. Gigascience 7, 1–6 (2018).
Andrews, S. et al. FastQC A Quality Control tool for High Throughput Sequence Data. https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ (2010).
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15, 550 (2014).
R Core Team (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/
Kanehisa, M. & Goto, S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res 28, 27–30 (2000).
Raudvere, U. et al. g:Profiler: a web server for functional enrichment analysis and conversions of gene lists (2019 update). Nucleic Acids Res 47, W191–W198 (2019).
Supek, F., Bošnjak, M., Škunca, N. & Šmuc, T. REVIGO summarizes and visualizes long lists of gene ontology terms. PLoS One 6, e21800 (2011).

There is NO Competing Interest.

TableS1Testis.xlsx
Table S1: Supplementary information for genes and gene networks expressed in testis tissue. a) Detected gene networks, including ANOVA statistics for comparing Single-pair versus Double-pair males. Significantly differentially expressed modules are denoted in red. b) Full lists of genes of detected gene networks. Significantly differentially expressed modules are denoted in red. c) Hub genes of gene networks with significantly differential expression in relation to the experimental treatment. d) Gene ontology terms for the module TES Saddlebrown. e) Gene ontology terms for the module TES Paleturquoise. f) Gene ontology terms for the module TES Orangered3. g) Gene ontology terms for the module TES Cyan. h) Gene ontology terms for the module TES Lightgreen. i) Statistics for Pearson correlations of the expression of gene networks with behavioural (aggression, courtship song duration) and hormonal (testosterone, corticosterone) phenotypic variation. Significant correlations are denoted in red.
TableS2PosteriorPallium.xlsx
Table S2: Supplementary information for genes and gene networks expressed in posterium pallium tissue. a) Detected gene networks, including ANOVA statistics for comparing Single-pair versus Double-pair males. Significantly differentially expressed modules are denoted in red. b) Full lists of genes of detected gene networks. Significantly differentially expressed modules are denoted in red. c) Hub genes of gene networks with significantly differential expression in relation to the experimental treatment. d) Gene ontology terms for the module PAL Floralwhite. e) Gene ontology terms for the module PAL Saddlebrown. f) Gene ontology terms for the module PAL Blue. g) Gene ontology terms for the module PAL Darkolivegreen. h) Gene ontology terms for the module PAL Darkturquoise. i) Gene ontology terms for the module PAL Darkgreen. j) Statistics for Pearson correlations of the expression of gene networks with behavioural (aggression, courtship song duration) and hormonal (testosterone, corticosterone) phenotypic variation. Significant correlations are denoted in red.
TableS3OpticTectum.xlsx
Table S3: Supplementary information for genes and gene networks expressed in optic tectum tissue. a) Detected gene networks, including ANOVA statistics for comparing Single-pair versus Double-pair males. Significantly differentially expressed modules are denoted in red. b) Full lists of genes of detected gene networks. Significantly differentially expressed modules are denoted in red. c) Hub genes of gene networks with significantly differential expression in relation to the experimental treatment. d) Gene ontology terms for the module OT Salmon. e) Gene ontology terms for the module OT Thistle1. f) Gene ontology terms for the module OT Lightyellow. g) Statistics for Pearson correlations of the expression of gene networks with behavioural (aggression, courtship song duration) and hormonal (testosterone, corticosterone) phenotypic variation. Significant correlations are denoted in red.

Download PDF

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Transcriptomic changes in the testes and brain associated with sperm competition risk in male zebra finches

Archived Versions:

Version 1

Abstract

Figures

Introduction

Results

Distinct gene expression between the testis, posterior pallium and optic tectum

Treatment-dependent expression of gene networks in the testis of Single- and Double-pair males

Treatment-dependent expression of genes and gene networks in the posterior pallium and optic tectum

Correlations of individual phenotypic trait variation with gene co-expression modules

Discussion

Methods

Declarations

References

Additional Declarations

Supplementary Files

Archived Versions:

Version 1