Gene bionetworks involved in the epigenetic transgenerational inheritance of altered mate preference: environmental epigenetics and evolutionary biology
© Skinner et al.; licensee BioMed Central Ltd. 2014
Received: 24 October 2013
Accepted: 28 April 2014
Published: 16 May 2014
Mate preference behavior is an essential first step in sexual selection and is a critical determinant in evolutionary biology. Previously an environmental compound (the fungicide vinclozolin) was found to promote the epigenetic transgenerational inheritance of an altered sperm epigenome and modified mate preference characteristics for three generations after exposure of a gestating female.
The current study investigated gene networks involved in various regions of the brain that correlated with the altered mate preference behavior in the male and female. Statistically significant correlations of gene clusters and modules were identified to associate with specific mate preference behaviors. This novel systems biology approach identified gene networks (bionetworks) involved in sex-specific mate preference behavior. Observations demonstrate the ability of environmental factors to promote the epigenetic transgenerational inheritance of this altered evolutionary biology determinant.
Combined observations elucidate the potential molecular control of mate preference behavior and suggests environmental epigenetics can have a role in evolutionary biology.
KeywordsEpigenetics Brain Networks Evolution Behavior
The current molecular paradigm for neo-Darwinian evolutionary biology is that random DNA sequence mutations, gene flow, and genetic drift promote phenotype variation that allows an adaptation event to facilitate natural selection . Although environment has an important role in natural selection, environmental factors generally do not have the capacity to alter DNA sequence or mutation rates. A small group of compounds can act as mutagens, but the vast majority of nutritionally derived and environmental toxicants do not alter DNA sequence . The current genetic paradigm does not completely explain many observations such as rapid evolutionary events, environmental impacts on evolution, and the low frequency of the occurrence of useful mutations [3, 4]. The realization that epigenetics provides an additional molecular mechanism for the environment to influence genome activity and biology has suggested a potential role for environmental epigenetics in evolutionary biology [5–11]. Charles Darwin recognized sexual selection as one of two determinants in evolutionary biology, the other being natural selection . The physical attributes and courtship rituals involved in mate preference are essential for reproductive fitness and propagation of a species. The current study examines how environmental factors can promote an epigenetic event to promote an alteration in mate preference behavior.
Previously we demonstrated that exposure of a gestating female rat to an environmental compound during fetal gonadal sex determination promoted epigenetic reprogramming of the male germline [13–15]. These reprogrammed differential DNA methylation regions (DMR) in the sperm epigenome have recently been shown to be induced by a variety of different environmental toxicants with exposure specific DMR [16, 17]. The initial environmental compound used was the commonly used fungicide vinclozolin which is an anti-androgenic endocrine disruptor . The primordial germ cells during migration down the genital ridge undergo a DNA methylation erasure that then upon gonadal sex determination the DNA re-methylation is initiated in a sex-specific manner . Environmental exposures during this developmental stage modifies the epigenetic programming of the male germline that becomes re-programmed (imprinted-like) and promotes a transgenerational phenotypic variation and adult onset disease state in subsequent generations [13–16]. The epigenetic transgenerational inheritance of adult onset disease (i.e. after one year of age) in males includes infertility, prostate disease, kidney disease, immune abnormalities and spermatogenic defects [20, 21], and in females includes mammary tumor development, kidney disease, reproductive tissue abnormalities and pregnancy abnormalities . This germline mediated epigenetic transgenerational inheritance of adult onset disease is mediated in part through alterations in the sperm epigenome [13, 14]. Since the germline establishes the base line epigenome of the organism, all tissues in both the female and male progeny including the brain appear to have altered tissue specific transgenerational epigenomes, transcriptomes and phenotypes [15, 21, 23–25].
Investigation of the epigenetic transgenerational inheritance of altered brain genome activity and behaviors previously demonstrated anxiety-like behavior increased in females and decreased in males, which correlate to alterations in specific brain region transcriptomes . Altered stress responses are also detected in the transgenerational exposure lineage animals . Interestingly, previous analysis of F3 generation control and vinclozolin lineage female and male rats (i.e. prior to the onset of disease) demonstrated an alteration in mate preference behavior . The female rats, independent of control or vinclozolin lineage, prefer control lineage males if given a choice. This behavioral decision raises the possibility of an epigenetic contribution to mate preference and sexual selection. The current study was designed to directly correlate the altered mate preference behavior with gene networks in specific brain regions in both the females and males. Observations elucidate the potential molecular control of mate preference behavior and demonstrates environmental factors have the capacity to promote the epigenetic transgenerational inheritance of altered mate preference.
Systems biology analysis has allowed biological phenomena such as mate preference to be considered from the molecular to physiological level. The gene bionetwork  analysis previously developed to investigate the molecular basis of disease  was used in the current study. This approach has been used to identify gene networks associated with disease, such as obesity and diabetes . Recently, we have used this bionetwork analysis to study a normal developmental process of primordial follicle development in the ovary [31, 32]. The gene networks identified were found to contain growth factors that are known to regulate the developmental process [31, 32]. These bionetwork analyses use a large number of microarray transcriptome analyses under different perturbations to identify gene clusters and modules that are coordinately regulated [33, 34]. The gene networks observed identify the genes with the highest level of integration and connection (i.e. connectivity) that associate with the phenotype [29, 33–35]. This genomic approach was used in the current study to identify the gene bionetworks in various brain regions associated with mate preference.
Observations demonstrate an environmental compound exposure can induce an epigenetic reprogramming of the germline that promotes epigenetic transgenerational inheritance of altered mate preference behavior. Although no direct epigenetic modifications in the brain were examined, the environmentally induced epigenetic transgenerational model used indicates epigenetics can be involved in the induction of the altered behavioral phenotypes. Sex-specific effects were observed in both the male and female brain transcriptome and behavior correlations. The gene networks in specific brain regions that statistically correlate with various mate preference behaviors provides insight into this environmentally modified transgenerational behavior. This systems biology approach has elucidated novel mechanisms to be considered in mate preference biology.
Additional file 1: Movie S1: The first 19 sec of a 10-min mate-preference trial is shown. The trial is conducted under dim red light during the nocturnal (active) phase of the rats' light cycle. At the beginning of the video, the male is in the center of the chamber. The chamber is demarcated into thirds by tape on its floor. A stimulus female can be seen at the far end of the apparatus; the other stimulus female is not visible due to the position of camera. The stimulus females are free-moving in their chambers, but they are separated from the male by a wire mesh that is bounded by Plexiglas barrier. This enables the animals to communicate by olfactory, pheromonal, or behavioral cues, but physical interaction is limited to touching across the wire mesh. The trial begins with the removal of a holding box that confines the male. The male can be seen moving into the zone in front of one stimulus female and then moving across the central portion of the cage to the other stimulus female (out of sight). Several behaviors of the male can be seen on the video such as sniffing, facial investigation, walking, and standing of the female. The male is also seen investigating the various parts of the chamber, including the wire mesh, surrounding Plexiglas partition, and the glass walls of the chamber. Behaviors were scored for each male toward each pair of opposite lineage (Control- or Vinclozolin-Lineage) stimulus females. (AVI 4 MB)
Differentially expressed Signature genes and their overlap with modules generated in combined network
Combined networks modules
Separate network modules
Number of modules
Number of modules
# genes in module
# genes overlapped between module & Signature list
# genes in module
# genes overlapped between module & Signature list
Analysis of the cellular pathways and processes associated with the gene Signature lists for each brain region is shown in Figure 2 and Additional file 5. The top 36 pathways with the greatest combined number of genes associated are shown in Additional file 5. An extended list of pathways and processes with the associated genes from the different modules and tissues is presented in Additional file 6: Table S3. Several of the most highly represented pathways in the male and female were the MAPK signaling pathway, olfactory transduction, neuroactive ligand-receptor interactions and axon guidance. The Signature list genes distributed relatively evenly across the different pathways with no major over-representation identified. Most major cellular processes and pathways were represented with no major predominance of any individual specific regulatory mechanism observed (Additional file 6: Table S3).
Due to the distinct functions of each brain region and distinct gene Signature lists, the combined analysis was found not to provide the specificity needed to identify the behavioral correlations with gene modules. Therefore, a more specific network analysis using the individual Signature lists for each brain region separately was performed, termed separate networks. Each brain region differentially expressed gene Signature list was used for separate network analysis. The cluster analysis identified specific gene modules for each brain region from the male and female gene Signature lists presented (Figure 3b, c). The modularity for the specific brain regions was not as strong as the combined region analysis. Each separate brain region is shown and the gene modules are identified by the different colors. The brain regions had 1–6 different modules and associated gene networks (Table 1). The same differentially expressed gene Signature lists were used, but the network analysis was from the separate lists (Figure 3b and c). All subsequent analyses used the gene modules from this region specific network analyses.
Gene modules highly correlated to different mate preference behavior parameters
The six male brain regions and associated gene modules had a number of statistically significant correlations with the mate preference parameters (Table 2 and Additional file 9: Table S5). All the male brain regions had statistically significant correlation with at least one module and the Wire Mesh behavioral parameter. Amy and Hipp also had correlations with the behavioral Plexiglas parameter (Figure 6). Therefore, at least one gene module in nearly all brain regions statistically correlated to the mate preference parameters analyzed. These correlations can now be considered in regards to the regulatory roles of gene networks identified for mate preference behavior alterations for the female (chooser) versus the male (selected) (Figure 5 and Additional file 8: Figure S2).
The direct connection gene sub-networks for the critical male amygdala (M-Amy) and cingulate cortex (M-CngCtx) turquoise modules are shown in Figure 5. The Signature list for each brain region sub-networks demonstrate distinct networks for each region (Additional file 8: Figure S2). Since nearly all the brain regions and key modules (Figure 6) have a statistically significant correlation with the Wire Mesh for male or Plexiglas for female mate preference behavior parameters, the combined gene sub-network (Figure 4) for all male or female brain regions identifies a potentially associated molecular control of behavior. Alternately, the analysis of separate regions differentially expressed gene sets (Signature lists) identified distinct gene sub-networks that associate with the different regions (Additional file 8: Figure S2). These potential gene sub-networks correlate and potentially regulate the mate choice behavior for the female and selection behavior/phenotype for the male. In addition to the gene networks, a correlation of critical cellular pathways in specific brain regions and modules (Additional file 5) that are associated with the epigenetic transgenerational inheritance of altered mate preference behavior.
The epigenetic transgenerational inheritance of the altered mate preference behavior requires the transmission of an altered epigenome in the germline (sperm) [5, 13, 15]. Previously the altered DNA methylation of the F3 generation sperm was characterized with 48 differentially DNA methylated regions (DMR) being identified in gene promoters . These vinclozolin induced sperm DMR are in part what promotes an altered epigenome in the embryo and all developing tissues transgenerationally . Although any developing tissue (e.g. brain) will have a dramatic cascade of epigenetic and genetic steps to achieve an adult fully differentiated state [40, 41], the possibility that some of the original germline epigenetic marks (DMR) may persist was investigated. The genes associated with the 48 previously identified sperm DMR were compared with the male and female brain region gene Signature lists. The comparison demonstrated the majority of the DMR did not correspond to differentially expressed genes in various brain regions. Only Rnase1 in the male Amy, Ig6-2a in the male EnCtx, Parp9 in the female CngCtx and Rp132 in the female OlfB overlapped. Interestingly, a copy number variation (CNV) in the Fam111a site previously identified  was found in all brain regions identified with the epigenome analysis. This provides a positive control for the technology and ability to detect the DMR. Therefore, some of the original sperm DMR programmed sites may persist, but the vast majority of brain development and epigenetic programming, and potential distal regularity role of DMR in epigenetic control regions , is distinct from the original germline epigenetic marks.
A systems biology analysis of environmentally induced epigenetic transgenerational inheritance of altered mate preference behavior was performed to suggest a potential role for epigenetics in evolutionary biology. Previous research has demonstrated that environmental toxicants such as the fungicide vinclozolin can promote a reprogramming of the germline epigenome during fetal gonadal sex determination that then transmits altered phenotypes and adult onset disease states transgenerationally in the absence of future environmental exposure . This is referred to as epigenetic transgenerational inheritance [5, 15] and suggests a role for environmental epigenetics in the inheritance of phenotypic variation and disease, independent of classic genetic inheritance mechanisms. The basic molecular mechanism involved in this non-genetic form of inheritance is the ability of environmental factors to influence the epigenetic programming of the germline [15, 19]. The primordial germ cells during migration down the genital ridge undergo an erasure of DNA methylation that then is initiated to re-methylate at the time of gonadal sex determination in a sex-specific manner . An environmental toxicant such as vinclozolin appears to alter gonadal development to influence germline DNA methylation programming  and the differential DNA methylation regions (DMR) in the sperm become imprinted-like sites that appear to not get erased at fertilization so are transmitted to subsequent generations and male and female progeny [13–16]. In addition to vinclozolin, a number of other environmental toxicants such as the plastic compound bisphenol A (BPA) [16, 43], dioxin [16, 44], methoxycholor , phthalates , pesticides , hydrocarbons , and DDT  have been shown to induce transgenerational phenotypes. Other environmental factors such as nutrition and stress can also promote transgenerational phenotypes [45–48].
The vinclozolin induced epigenetic transgenerational phenotypes previously identified included adult onset rat disease after 12 months of age of male infertility, mammary tumors, prostate disease, kidney disease and immune abnormalities . Therefore, the mate preference analysis was performed prior to adult onset disease to remove the disease as a confounding factor. A brain-behavior transgenerational phenotype observed was increased female anxiety and decreased male anxiety behaviors . This transgenerational anxiety behavior was also examined on a molecular level to identify brain region specific changes in different gene expression and gene networks associated with the behavior . Similar observations were made in the analysis of transgenerational stress responses . Interestingly, in a previous study we found that vinclozolin induced alterations in mate preference behavior . Females from either control or vinclozolin F3 generation lineages prefer control lineage males over vinclozolin lineage males, whereas no altered mate preference in males was observed . While the standard argument would be that the females are the discriminating sex and distinguish between males on as yet undetermined phenotype characteristic(s), it is important to realize that the absence of evidence (in the male) does not mean the evidence of absence of male involvement since preference is only the first step in a mating sequence. That is, under natural circumstances this is followed by a mutual decision. Mating in rodents involves pheromone and auditory cues produced by both sexes and evidence suggests (see below) that it is under such unfettered conditions that the complementarity of behavior and brain are expressed.
Observations from the current study need to consider the effects on the female brain as potentially altering female discrimination and preference. The effects on the male brain are presumed to be associated with the characteristics (e.g. auditory cues and pheromone production) being selected. This altered mate preference behavior suggests the existence of an environmentally altered epigenetic transgenerational inheritance of mate preference behavior . The current study was designed to identify the gene bionetworks in various male and female brain regions that correlate with the behavior of the transgenerational inheritance model.
A novel gene bionetwork analysis was developed to identify gene networks correlated to disease . The approach was to use a large number of microarrays to identify transcriptomes in specific tissues associated with control versus disease individuals in large cohorts. Differentially regulated genes that are coordinately regulated and having connectivity  are clustered in large gene sets to identify modules of genes that associate with the disease [29, 30, 33–35] (Figure 1). More recently, we have used a similar approach to investigate a normal development process to identify gene bionetworks associated with development [31, 32]. The primordial follicle development in the ovary was investigated to identify a network of growth factors and associated signaling systems that regulate follicle development [31, 32]. This bioinformatics approach to identify regulatory gene networks was used in the current study to correlate brain gene networks to mate preference behavior (Figure 1) in an epigenetic transgenerational model . The six different brain regions isolated from F3 generation control and vinclozolin lineage females and males were used in a microarray analysis to determine the differential gene expression in each brain region. The region specific gene sets, “Signature list”, and associated gene networks were investigated.
Analysis of significant pathways and cellular processes potentially influenced by the differentially expressed gene sets and networks did not identify predominant or over represented pathways. One pathway identified that previously has been shown to correlate with sexual selection is the olfactory transduction pathway [49–52]. However, most of the brain regions and specific networks or gene modules affected similar pathways with overlap between most. Therefore, no specific pathways were identified and most major pathways were influenced. A limitation in gene expression studies is that individual genes are assigned a specific function, but the gene may be involved in functional categories. This limitation needs to be considered in any gene expression data interpretation, but genome wide transcriptome analysis for gene sets has been shown to start to address this issue . Combined observations suggest that the distinct differential expressed gene sets for the different brain regions appear to regulate common cellular processes and pathways among the brain regions and modules.
In contrast, analysis of gene networks identified unique gene sub-networks and gene modules associated with each brain region differentially expressed gene sets. The specific inter-connected genes were unique and overall networks of connected genes distinct. Therefore, the different functions associated with each brain region and associated with the altered mate preference behavior were identified. A statistical correlation of the gene modules for each brain region with the different mate preference parameters measured identified a number of statistically significant correlations. All but one female brain region (Hipp) had statistically significant correlations with the Plexiglas behavior parameter. The female behavior directly associates with the discrimination and mate preference choice. Interestingly, all the male brain regions had some modules with statistically significant correlations with the Wire Mesh behavior parameter. The male behavior and/or phenotype (e.g. pheromonal production) associates with the selected behaviors and characteristics of the non-discriminant sex. Therefore, direct correlations with the gene modules, specific brain regions and mate preference behavior parameters were identified. The specific gene modules and behavioral parameters statistically correlated were distinct between the sexes and brain regions, but strong correlations of the gene networks to the mate preference behavior was established. Interestingly, both the Plexiglas and Wire Mesh are indicators of interest and assessment of the stimulus animal.
Darwin  considered natural selection and sexual selection as distinct processes driving the evolution of traits. Natural selection results in traits that are adaptive responses to changes in the environment. The resulting variation in traits between and within species is shaped by differential survivorship. In other words, animals that survive are those with traits that are adaptive to their environment . Darwin  conceived of sexual selection as arising from aggressive interactions between males (male-male competition) and the female's selection of a mate (mate choice). Males compete amongst themselves for access to females. Aggression between males can have a direct effect on female reproduction by preventing other breeding males from having access to females or from harming the female. Importantly aggression amongst males can also have an indirect effect by inhibiting or suppressing the normal reproductive physiology of the female or even terminate a pregnancy .
Mate preference, in its simplest form, states that males compete for females and females choose between them. Although most research has focused on how females choose males, male choice of females is also important [56–58]. This point cannot be overemphasized. That is, in virtually all paradigms published to date, the choosing individual is the independent variable and the stimulus animal is the dependent variable. Although this study is similar to previous studies in that there are restrained stimulus animals and freely moving individuals that are the investigators, it differs in several important ways. First, the ‘round robin’ testing method insured that all males and all females served both as stimulus animals and experimental animals; thus, the “preferences” exhibited reflect both the males and the females. Second, this study extends to the molecular level events in the brain of the individuals, both of which have exhibited preferences, namely brain transgenerational transcriptome alterations that correlated to the opposing sex behaviors. In essence we are seeing the outcome of the complementary nature of mate choice. This has never been shown before.
Evolution favors reproductive success, and it is in the individual's interest to focus on selecting the best mate and to avoid mating with the wrong species [59, 60]. Making the correct choice of a mate has a pronounced impact on reproductive success of both partners. Except in unusual systems, in nature the mating partners choose one another [55, 57, 60]. Experiments with flies , birds , and rodents [63, 64] indicate that individuals who are allowed to select, and be selected by, their mate enjoy greater reproductive success than force-paired animals. This consent is based not only on the internal milieu that motivates each individual to seek a partner, but also on the satisfactory nature of the phenotypic traits the potential mate displays.
There are a number of sexual selection hypotheses, all of which emphasize that females choosing optimally will produce young whose viability and survivorship are enhanced by the female’s choice of mate . The most attractive, and one that takes into account that mating is a cooperative act that involves both partners, is the sensory exploitation hypothesis . This hypothesis postulates that males have evolved calls and/or pheromones to exploit the preexisting sensory biases in the female that themselves evolved for reasons independent of female choice. Male behavior then has changed to maximize stimulation of the female's sensory systems. Therefore, the coordination of the complementary signal and receiver, mounting and lordosis, coordination of egg and sperm maturation and release is required for successful completion for reproduction. These complementary processes are evident at all levels of biological organization [55, 67] and we extend it here to the level of the genome and epigenome.
The ability of an environmental factor to alter mate preference behavior suggests a critical role of environment in evolutionary biology. This is distinct from the generally accepted role of the environment in natural selection where environment is the active factor in the selection of an adaptive phenotype, but alternately here it involves the induction of phenotypes that can be acted on by natural selection. Since the majority of environmental factors can not alter DNA sequence or promote mutagenesis , an additional molecular mechanism to consider involves environmental epigenetics . Many environmental compounds and factors such as nutrition can modify the epigenome to alter phenotypic variation. The role of epigenetics in evolutionary biology has been suggested previously [5, 8–11, 15, 68], but no significant experimental evidence has been provided. The current study demonstrates an environmental factor can promote the epigenetic transgenerational inheritance of an altered mate preference behavior. The epigenetic modification of the germline (sperm) has been previously established [13, 14, 16] and will lead to epigenetic alterations in the brain transcriptomes of both females and males  to alter the mate preference behavior . Therefore, the current study provides direct experimental evidence for a potential role of environmental epigenetics in evolution by regulating a critical determinant such as mate preference on a molecular level (i.e. altered gene networks) in specific brain regions in a sex-specific manner (Figure 7). Although no direct epigenetic alterations were examined in the brain, the germline (sperm) epigenetic alterations that generate this altered male or female brain development have been documented [13, 14, 16]. This molecular mechanism does not suggest genetics will not have a critical role in evolutionary biology, but suggests environmental epigenetics will be an additional mechanism to consider. Epigenetics provides a mechanism for the environment to impact phenotypic variation and natural selection. Epigenetic and genetic mechanisms will cooperate to regulate on a molecular basis evolutionary biology. This appears to be a “neo-Lamarckian concept to facilitate neo-Darwinian evolution” [40, 41].The systems biology approach used in the current study links an environmental exposure, epigenetic transgenerational inheritance and molecular regulation of brain function to mate preference and evolutionary biology (Figure 7). Epigenetics will have a central role in how environmental factors influence how the gene networks emerge to induce phenotypic variation. Although genetics is critical for all aspects of biology, epigenetics provides the plasticity to allow the environment to alter biological events. This type of systems approach to understand complex biological traits, such as sexual selection, provides insights into how the various components (environment, phenotype and evolution) interact in a systems biology manner.
Animal housing protocol
Male and female rats of the F3 generation of Vinclozolin (Vinclozolin-Lineage) and DMSO Control (Control-Lineage) Lineages were selected out of litters from untreated F2 generation mothers in Dr. Michael Skinner’s laboratory at Washington State University according to established protocols . Briefly, gestating female F0 generation Sprague Dawley rats were injected with the fungicide vinclozolin (100 mg/kg) daily during fetal gonadal sex determination (E8-E14) and the F1 generation were bred to generate the F2 generation and then the F2 bred to generate the F3 generation . At approximately PND 10 (before weaning), each animal was injected with a small microchip (AVID Identification system Inc. Norco, CA) subcutaneously between the shoulder blades. The animals were then shipped to the University of Texas from Washington State University on postnatal day (PND) 22, one day after weaning. Upon arriving at the University of Texas, one animal from each Lineage (Control and Vinclozolin) was pair-housed (one control and one vinclozolin animal) and remained in these dyads throughout the duration of the study. Because of the natural variation in dates of breeding, there was a 4-day spread of birth date of animals in the first cohort but in the second cohort, all animals were born on the same day. However, all pair-housed animals were no more than one day apart in birth age and were paired randomly to prevent an age effect on cagemates.
Each dyad of animals was randomly placed in a six-wide, five-high metal housing rack in standard translucent polycarbonate rat cages (46 × 24 × 20.5 cm) with ad libitum access to tap water and standard rat chow (Purina rodent chow #5LL2 Prolab RMH 1800 diet). The animal room was on a 14:10 light/dark schedule. For environmental enrichment, a 7 cm diameter PVC pipe was placed in each cage.
All experimental protocols for the procedures with rats were pre-approved by the Washington State University Animal Care and Use Committee (IACUC approval # 02568–026) and by the University of Texas at Austin Animal Care and Use Committee (Public Health Service Animal Welfare Assurance Number A4107-01).
After habituation to the testing arena, each individual was tested individually (when used as an experimental subject) or in pairs (when used as stimulus animals) with all individuals; the order of the testing was rotated during the course of both the male and female trials. All tests were conducted during the dark phase of the light cycle, beginning at 1200 h, 4 h after the progesterone injection, in a room illuminated with low levels of red light. Before trials, to confirm that females were receptive, each female was placed with a sexually experienced but otherwise experimentally naïve male; all females exhibited robust lordosis (arched back and lifted head posture) in response to mounting by the male.
Partner preference tests consisted of placing an individual (male or female) in the center of a large three-chamber glass-testing arena (122 × 46 × 54 cm). At either end was a small compartment (28 × 28 × 12.5 cm) containing the stimulus rats separated by a Wire-mesh barrier to allow exchange of olfactory, visual, and tactile cues. The area directly in front of the stimulus cage was marked by tape. Tests were conducted 2 h after the onset of the dark cycle under red-light illumination and lasted 10 min; all tests were videotaped for further review and analysis. At the end of each test, all animals were removed, and the entire testing arena was washed with a household cleaner and then wiped down with 70% ethanol to remove scent marks and residual odors. All males were tested with both types of females as stimulus animals (72 trials), and all females were tested with both types of males as stimulus animals (72 trials) (Movie S1).
The videotaped trials were analyzed by using JWatcher v1.0 (http://www.jwatcher.ucla.edu) computer software to quantify the behavior of each experimental animal. Time spent with a stimulus animal was recorded as soon as all four paws of the experimental animal crossed over the line of tape marking the boundary of that stimulus animal’s compartment. As soon as one paw crossed over the tape back into the center compartment, the time recorded with the experimental animal was stopped. Preference behaviors were defined as those directed to the stimulus animal and included time spent in contact with the Wire Mesh separating the experimental and stimulus animal (Wire Mesh), during which the animals often touched noses through the Wire Mesh (facial investigation), and contacted the Plexiglas surface surrounding the front of the stimulus cage; the cumulative total time in these preference behaviors toward each stimulus animal was also calculated (Total). Other activity measured included undirected walking and sniffing (walking), standing still with minimal head movement (still). Videos demonstrating the test can be viewed as supporting information (Movie S1).
The brain was removed in less than 1 minute and placed in crushed ice to chill. The brain was then cut in half in the sagittal plane along the midline. In all cases but one the right side was blocked and then 6 areas dissected (see list below) within 3–5 min. This procedure was done on iced tissue. The dissected brain areas were placed in chilled Trizol (150 l) in 1.5 ml Eppendorf tubes according to manufacturers specifications in each tube. No tissue fragment was more than 3 mm but in those instances, but where there were multiple fragments the amount of Trizol was doubled (approximately). After all animals were dissected, the Eppendorf tubes were vortexed (15 sec) and then frozen on dry ice. The brain regions collected were according to Paxinos & Watson : olfactory bulbs (OlfB); cingulate cortex (CngCtx), anterior to POAH (Bregma 4.7 to 1.7); preoptic area-anterior hypothalamus (POAH), 4 mm rostral to anterior commissure (AC) (Bregma −0.26 to −1.40); amygdaloid nuclei (Amy), 3 mm caudal to AC (Bregma −2.3 to −3.6); hippocampus (Hipp), 6 mm caudal to AC (Bregma −2.12 to −4.52); entorhinal cortex (EnCtx) (Bregma −5.60 to −7.80).
Brain area samples from individual rats were homogenized in 150 μl Trizol and then 600 μl Trizol was added to final volume of 750 ml. Samples were stored at −80 or −20°C until RNA extraction. For microarray analysis, from 4 to 6 biological replicas (animals) were prepared as above for each brain area Control or Vinclozolin group depending on samples availability (Additional file 2: Table S1B). A total of 132 (67 Control and 65 Vinclozolin) samples/chips were analyzed: (6 brain areas) × (2 Male or Female) × (2 Control or Vinclozolin) × (4–6 biological replicas). RNA from individual animal brain area was extracted from Trizol samples according to standard Trizol extraction protocol (Invitrogen, USA) and stored in aqueous solution at −80°C until microarray analysis.
The microarray analysis was performed by the Genomics Core Laboratory, Center for Reproductive Biology, Washington State University, Pullman, WA using standard Affymetrix reagents and protocol. Briefly, mRNA was transcribed into cDNA with random primers, cRNA was transcribed, and single-stranded sense DNA was synthesized which was fragmented and labeled with biotin. Biotin-labeled ssDNA was then hybridized to the Rat Gene 1.0 ST microarrays containing more than 30,000 transcripts (Affymetrix, Santa Clara, CA, USA). Hybridized chips were scanned on Affymetrix Scanner 3000. CEL files containing raw data were then pre-processed and analyzed with Partek Genomic Suite 6.5 software (Partek Incorporated, St. Louis, MO) using an RMA, GC-content adjusted algorithm. Raw data pre-processing was performed in 12 groups, one for each male or female brain area. Comparison of array sample histogram graphs for each group showed if data for all chips were similar and appropriate for further analysis (Additional file 3: Figure S1). By this criterion, 2 microarray samples (not counted in Additional file 4: Table S2B and not shown on Additional file 3: Figure S1) were omitted from repeated group pre-processing and further analysis.
The microarray quantitative data involves signals from an average of 28 different oligonucleotides (probes) arrayed for each transcript and many genes are represented on the chip by several transcripts. The hybridization to each probe must be consistent to allow a statistically significant quantitative measure of resulting gene expression signal. Therefore, the microarray provides an unbiased and highly stringent quantitative procedure compared to other protocols . In contrast, a quantitative PCR procedure uses only two oligonucleotides and primer bias is a major factor in this type of analysis. Therefore, we did not attempt to use PCR based approaches as we feel the microarray analysis is more accurate and reproducible without primer bias such as PCR based approaches .
All microarray CEL files from this study have been deposited with the NCBI gene expression and hybridization array data repository GEO (GEO series accession number: GSE33830) and can be also accessed through http://www.skinner.wsu.edu. For gene annotation, Affymetrix annotation file RaGene1_0stv1.na31.rn4.transcript.csv was used.
The network analysis was restricted to genes differentially expressed between the control and the treatment groups based on previously established criteria: (1) fold change of group means ≥ 1.2 or ≤ 0.83; (2) T test p-value ≤ 0.05. The union of the differentially expressed genes from the different treatments resulted in 1,693 genes for males and 1833 for females being identified and used for constructing a weighted gene co-expression network [71, 72]. Unlike traditional un-weighted gene co-expression networks in which two genes (nodes) are either connected or disconnected, the weighted gene co-expression network analysis assigns a connection weight to each gene pair using soft-thresholding and thus is robust to parameter selection. The weighted network analysis begins with a matrix of the Pearson correlations between all gene pairs, then converts the correlation matrix into an adjacency matrix using a power function f(x) = x β . The parameter β of the power function is determined in such a way that the resulting adjacency matrix (i.e., the weighted co-expression network) is approximately scale-free. To measure how well a network satisfies a scale-free topology, we use the fitting index proposed by Zhang & Horvath  (i.e., the model fitting index R 2 of the linear model that regresses log(p(k)) on log(k) where k is connectivity and p(k) is the frequency distribution of connectivity). The fitting index of a perfect scale-free network is 1.
To explore the modular structures of the co-expression network, the adjacency matrix is further transformed into a topological overlap matrix . As the topological overlap between two genes reflects not only their direct interaction, but also their indirect interactions through all the other genes in the network. Previous studies [71, 73] have shown that topological overlap leads to more cohesive and biologically meaningful modules. To identify modules of highly co-regulated genes, we used average linkage hierarchical clustering to group genes based on the topological overlap of their connectivity, followed by a dynamic cut-tree algorithm to dynamically cut clustering dendrogram branches into gene modules . Such networks were generated from all combined male or female differentially expressed genes (2 combined networks) or from each individual male or female brain region Signature lists (12 separate networks). From one to ten modules were identified in combined or separate networks and the module size was observed to range from 10 to 780 genes (Table 1).
To distinguish between modules, each module was assigned a unique color identifier, with the remaining, poorly connected genes colored grey. The hierarchical clustering over the topological overlap matrix (TOM) and the identified modules is shown (Figure 1). In this type of map, the rows and the columns represent genes in a symmetric fashion, and the color intensity represents the interaction strength between genes. This TOM heatmap highlights that genes in the transcriptional network fall into distinct network modules, where genes within a given module are more interconnected with each other (blocks along the diagonal of the matrix) than with genes in other modules. Therefore, there are two types of global connectivity, adjacency-based one and TO based one. The adjacency-based connectivity (k.all) is defined as the sum of the power-function transformed correlations between the gene g and all the other genes in the whole network while the TO-based connectivity (to.all) is defined as the sum of the topological overlaps between the gene g and all the other genes. By default, connectivity used throughout the paper refers to TO-based connectivity to.all.
Gene Co-expression Network Analysis Clarification: Gene networks provide a convenient framework for exploring the context within which single genes operate. Networks are simply graphical models comprised of nodes and edges. For gene co-expression networks, an edge between two genes may indicate that the corresponding expression traits are correlated in a given population of interest. Depending on whether the interaction strength of two genes is considered, there are two different approaches for analyzing gene co-expression networks: 1) an unweighted network analysis that involves setting hard thresholds on the significance of the interactions, and 2) a weighted approach that avoids hard thresholds. Weighted gene co-expression networks preserve the continuous nature of gene-gene interactions at the transcriptional level and are robust to parameter selection. An important end product from the gene co-expression network analysis is a set of gene modules in which member genes are more highly correlated with each other than with genes outside a module. Most gene co-expression modules are enriched for GO functional annotations and are informative for identifying the functional components of the network that are associated with disease .
This gene co-expression network analysis (GCENA) has been increasingly used to identify gene sub-networks for prioritizing gene targets associated with a variety of common human diseases such as cancer and obesity [38, 76–79]. One important end product of GCENA is the construction of gene modules comprised of highly interconnected genes. A number of studies have demonstrated that co-expression network modules are generally enriched for known biological pathways, for genes that are linked to common genetic loci and for genes associated with disease [33, 38, 71, 75–78, 80, 81]. In this way, one can identify key groups of genes that are perturbed by genetic loci that lead to disease, and that define at the molecular level disease states. Furthermore, these studies have also shown the importance of the hub genes in the modules associated with various phenotypes. For example, GCENA identified ASPM, a hub gene in the cell cycle module, as a molecular target of glioblastoma  and MGC4504, a hub gene in the unfolded protein response module, as a target potentially involved in susceptibility to atherosclerosis .
Resulting lists of differentially expressed genes for each male or female brain area as well as for each module generated in the combined network and some generated in separate networks analysis were analyzed for KEGG (Kyoto Encyclopedia for Genes and Genome, Kyoto University, Japan) pathway enrichment using Pathway-Express, a web-based tool freely available as part of the Onto-Tools (http://vortex.cs.wayne.edu)  as well as KEGG website ‘Search Pathway’ tool (http://www.genome.jp/kegg/tool/search_pathway.html). Global literature analysis of various gene lists was performed using Pathway Studio 8.0 software (Ariadne Genomics, Inc., Rockville, MD).
Differential DNA methylation regions
Embryonic day 8–14
Preoptic area-anterior hypothalamus
We thank Dr. M. Manikkam and Ms. R. Tracey for technical assistance and H. Johnson for assistance in preparation of the manuscript. We thank Drs. C. Guerrero-Bosagna and E. Nilsson for critical review of the manuscript. Research was supported in part by NIH grants ES 017538 (Crews) and ES 012974 (Skinner).
- Charlesworth B, Charlesworth D: Darwin and genetics. Genetics. 2009, 183 (3): 757-766.PubMed CentralPubMedView ArticleGoogle Scholar
- Stein RA: Epigenetics and environmental exposures. J Epidemiol Community Health. 2012, 66 (1): 8-13.PubMedView ArticleGoogle Scholar
- Kalinowski ST: Evolutionary and statistical properties of three genetic distances. Mol Ecol. 2002, 11 (8): 1263-1273.PubMedView ArticleGoogle Scholar
- Wilkins AS: Genetic networks as transmitting and amplifying devices for natural genetic tinkering. Novartis Found Symp. 2007, 284: 71-86. discussion 86–79, 110–115PubMedView ArticleGoogle Scholar
- Jirtle RL, Skinner MK: Environmental epigenomics and disease susceptibility. Nat Rev Genet. 2007, 8 (4): 253-262.PubMedView ArticleGoogle Scholar
- Crews D, McLachlan JA: Epigenetics, evolution, endocrine disruption, health, and disease. Endocrinology. 2006, 147 (6 Suppl): S4-S10.PubMedView ArticleGoogle Scholar
- Damiani G: The Yin and Yang of anti-Darwinian epigenetics and Darwinian genetics. Riv Biol. 2007, 100 (3): 361-402.PubMedGoogle Scholar
- Day T, Bonduriansky R: A unified approach to the evolutionary consequences of genetic and nongenetic inheritance. Am Nat. 2011, 178 (2): E18-E36.PubMedView ArticleGoogle Scholar
- Kuzawa CW, Thayer ZM: Timescales of human adaptation: the role of epigenetic processes. Epigenomics. 2011, 3 (2): 221-234.PubMedView ArticleGoogle Scholar
- Flatscher R, Frajman B, Schonswetter P, Paun O: Environmental heterogeneity and phenotypic divergence: can heritable epigenetic variation aid speciation?. Genet Res Int. 2012, 2012: 698421-PubMed CentralPubMedGoogle Scholar
- Klironomos FD, Berg J, Collins S: How epigenetic mutations can affect genetic evolution: model and mechanism. Bioessays. 2013, 35 (6): 571-578.PubMedView ArticleGoogle Scholar
- Darwin C: The Descent of Man, and Selection in Relation to Sex. 1871, London: T. MurrayView ArticleGoogle Scholar
- Anway MD, Cupp AS, Uzumcu M, Skinner MK: Epigenetic transgenerational actions of endocrine disruptors and male fertility. Science. 2005, 308 (5727): 1466-1469.PubMedView ArticleGoogle Scholar
- Guerrero-Bosagna C, Settles M, Lucker BJ, Skinner MK: Epigenetic transgenerational actions of vinclozolin on promoter regions of the sperm epigenome. PLoS One. 2010, 5 (9): e13100-PubMed CentralPubMedView ArticleGoogle Scholar
- Skinner MK, Manikkam M, Guerrero-Bosagna C: Epigenetic transgenerational actions of environmental factors in disease etiology. Trends Endocrinol Metab. 2010, 21 (4): 214-222.PubMed CentralPubMedView ArticleGoogle Scholar
- Manikkam M, Guerrero-Bosagna C, Tracey R, Haque MM, Skinner MK: Transgenerational actions of environmental compounds on reproductive disease and epigenetic biomarkers of ancestral exposures. PLoS One. 2012, 7 (2): e31901-PubMed CentralPubMedView ArticleGoogle Scholar
- Skinner MK, Manikkam M, Tracey R, Nilsson E, Haque MM, Guerrero-Bosagna C: Ancestral DDT exposures promote epigenetic transgenerational inheritance of obesity. BMC Med. 2013, 11: 228-PubMed CentralPubMedView ArticleGoogle Scholar
- Kelce WR, Gray LE, Wilson EM: Antiandrogens as environmental endocrine disruptors. Reprod Fertil Dev. 1998, 10 (1): 105-111.PubMedView ArticleGoogle Scholar
- Morgan HD, Santos F, Green K, Dean W, Reik W: Epigenetic reprogramming in mammals. Hum Mol Genet. 2005, 14 Spec No 1: R47-R58.PubMedView ArticleGoogle Scholar
- Anway MD, Leathers C, Skinner MK: Endocrine disruptor vinclozolin induced epigenetic transgenerational adult-onset disease. Endocrinology. 2006, 147 (12): 5515-5523.PubMedView ArticleGoogle Scholar
- Anway MD, Skinner MK: Transgenerational effects of the endocrine disruptor vinclozolin on the prostate transcriptome and adult onset disease. Prostate. 2008, 68 (5): 517-529.PubMedView ArticleGoogle Scholar
- Nilsson EE, Anway MD, Stanfield J, Skinner MK: Transgenerational epigenetic effects of the endocrine disruptor vinclozolin on pregnancies and female adult onset disease. Reproduction. 2008, 135 (5): 713-721.PubMedView ArticleGoogle Scholar
- Skinner MK, Anway MD, Savenkova MI, Gore AC, Crews D: Transgenerational epigenetic programming of the brain transcriptome and anxiety behavior. PLoS One. 2008, 3 (11): e3745-PubMed CentralPubMedView ArticleGoogle Scholar
- Anway MD, Rekow SS, Skinner MK: Transgenerational epigenetic programming of the embryonic testis transcriptome. Genomics. 2008, 91 (1): 30-40.PubMed CentralPubMedView ArticleGoogle Scholar
- Skinner MK, Manikkam M, Haque MM, Zhang B, Savenkova M: Epigenetic transgenerational inheritance of somatic transcriptomes and epigenetic control regions. Genome Biol. 2012, 13 (10): R91-PubMed CentralPubMedView ArticleGoogle Scholar
- Crews D, Gillette R, Scarpino SV, Manikkam M, Savenkova MI, Skinner MK: Epigenetic transgenerational inheritance of altered stress responses. Proc Natl Acad Sci USA. 2012, 109 (23): 9143-9148.PubMed CentralPubMedView ArticleGoogle Scholar
- Crews D, Gore AC, Hsu TS, Dangleben NL, Spinetta M, Schallert T, Anway MD, Skinner MK: Transgenerational epigenetic imprints on mate preference. Proc Natl Acad Sci USA. 2007, 104 (14): 5942-5946.PubMed CentralPubMedView ArticleGoogle Scholar
- Friend SH: The need for precompetitive integrative bionetwork disease model building. Clin Pharmacol Ther. 2010, 87 (5): 536-539.PubMedView ArticleGoogle Scholar
- Schadt EE, Lamb J, Yang X, Zhu J, Edwards S, Guhathakurta D, Sieberts SK, Monks S, Reitman M, Zhang C, Lum PY, Leonardson A, Thieringer R, Metzger JM, Yang L, Castle J, Zhu H, Kash SF, Drake TA, Sachs A, Lusis AJ: An integrative genomics approach to infer causal associations between gene expression and disease. Nat Genet. 2005, 37: 710-717.PubMed CentralPubMedView ArticleGoogle Scholar
- Yang X, Deignan JL, Qi H, Zhu J, Qian S, Zhong J, Torosyan G, Majid S, Falkard B, Kleinhanz RR, Karlsson J, Castellani LW, Mumick S, Wang K, Xie T, Coon M, Zhang C, Estrada-Smith D, Farber CR, Wang SS, van Nas A, Ghazalpour A, Zhang B, Macneil DJ, Lamb JR, Dipple KM, Reitman ML, Mehrabian M, Lum PY, Schadt EE, et al: Validation of candidate causal genes for obesity that affect shared metabolic pathways and networks. Nat Genet. 2009, 41: 415-423.PubMed CentralPubMedView ArticleGoogle Scholar
- Nilsson EE, Savenkova MI, Schindler R, Zhang B, Schadt EE, Skinner MK: Gene bionetwork analysis of ovarian primordial follicle development. PLoS One. 2010, 5 (7): e11637-PubMed CentralPubMedView ArticleGoogle Scholar
- Nilsson E, Zhang B, Skinner MK: Gene bionetworks that regulate ovarian primordial follicle assembly. BMC Genomics. 2013, 14 (1): 496-PubMed CentralPubMedView ArticleGoogle Scholar
- Zhu J, Zhang B, Smith EN, Drees B, Brem RB, Kruglyak L, Bumgarner RE, Schadt EE: Integrating large-scale functional genomic data to dissect the complexity of yeast regulatory networks. Nat Genet. 2008, 40 (7): 854-861.PubMed CentralPubMedView ArticleGoogle Scholar
- Millstein J, Zhang B, Zhu J, Schadt EE: Disentangling molecular relationships with a causal inference test. BMC Genet. 2009, 10: 23-PubMed CentralPubMedView ArticleGoogle Scholar
- Pandey G, Zhang B, Chang AN, Myers CL, Zhu J, Kumar V, Schadt EE: An integrative multi-network and multi-classifier approach to predict genetic interactions. PLoS Comput Biol. 2010, 6 (9): e1000928-PubMed CentralPubMedView ArticleGoogle Scholar
- DiBenedictis BT, Ingraham KL, Baum MJ, Cherry JA: Disruption of urinary odor preference and lordosis behavior in female mice given lesions of the medial amygdala. Physiol Behav. 2012, 105 (2): 554-559.PubMed CentralPubMedView ArticleGoogle Scholar
- Anway MD, Memon MA, Uzumcu M, Skinner MK: Transgenerational effect of the endocrine disruptor vinclozolin on male spermatogenesis. J Androl. 2006, 27 (6): 868-879.PubMedView ArticleGoogle Scholar
- Ghazalpour A, Doss S, Zhang B, Wang S, Plaisier C, Castellanos R, Brozell A, Schadt EE, Drake TA, Lusis AJ, Horvath S: Integrating genetic and network analysis to characterize genes related to mouse weight. PLoS Genet. 2006, 2: e130-PubMed CentralPubMedView ArticleGoogle Scholar
- Sakata JT, Crews D: Developmental sculpting of social phenotype and plasticity. Neurosci Biobehav Rev. 2004, 28 (2): 95-112.PubMedView ArticleGoogle Scholar
- Skinner MK: Role of epigenetics in developmental biology and transgenerational inheritance. Birth Defects Res C Embryo Today. 2011, 93 (1): 51-55.PubMedView ArticleGoogle Scholar
- Skinner MK: Environmental epigenetic transgenerational inheritance and somatic epigenetic mitotic stability. Epigenetics. 2011, 6 (7): 838-842.PubMedView ArticleGoogle Scholar
- Skinner M, Guerrero-Bosagna C, Haque MM, Nilsson E, Bhandari R, McCarrey J: Environmentally induced transgenerational epigenetic reprogramming of primordial germ cells and subsequent germline. PLoS One. 2013, 8 (7): e66318-PubMed CentralPubMedView ArticleGoogle Scholar
- Salian S, Doshi T, Vanage G: Impairment in protein expression profile of testicular steroid receptor coregulators in male rat offspring perinatally exposed to Bisphenol A. Life Sci. 2009, 85 (1–2): 11-18.PubMedView ArticleGoogle Scholar
- Bruner-Tran KL, Osteen KG: Developmental exposure to TCDD reduces fertility and negatively affects pregnancy outcomes across multiple generations. Reprod Toxicol. 2011, 31 (3): 344-350.PubMed CentralPubMedView ArticleGoogle Scholar
- Waterland RA, Travisano M, Tahiliani KG, Rached MT, Mirza S: Methyl donor supplementation prevents transgenerational amplification of obesity. Int J Obes (Lond). 2008, 32 (9): 1373-1379.View ArticleGoogle Scholar
- Painter RC, Osmond C, Gluckman P, Hanson M, Phillips DI, Roseboom TJ: Transgenerational effects of prenatal exposure to the Dutch famine on neonatal adiposity and health in later life. BJOG. 2008, 115 (10): 1243-1249.PubMedView ArticleGoogle Scholar
- Pembrey ME, Bygren LO, Kaati G, Edvinsson S, Northstone K, Sjostrom M, Golding J: Sex-specific, male-line transgenerational responses in humans. Eur J Hum Genet. 2006, 14 (2): 159-166.PubMedView ArticleGoogle Scholar
- Dias BG, Ressler KJ: Parental olfactory experience influences behavior and neural structure in subsequent generations. Nat Neurosci. 2014, 17 (1): 89-96.PubMed CentralPubMedView ArticleGoogle Scholar
- Buck LB: Olfactory receptors and odor coding in mammals. Nutr Rev. 2004, 62 (11 Pt 2): S184-S188. discussion S224-141PubMedView ArticleGoogle Scholar
- Brennan PA, Zufall F: Pheromonal communication in vertebrates. Nature. 2006, 444 (7117): 308-315.PubMedView ArticleGoogle Scholar
- Dulac C, Wagner S: Genetic analysis of brain circuits underlying pheromone signaling. Annu Rev Genet. 2006, 40: 449-467.PubMedView ArticleGoogle Scholar
- Green RE, Krause J, Briggs AW, Maricic T, Stenzel U, Kircher M, Patterson N, Li H, Zhai W, Fritz MH, Hansen NF, Durand EY, Malaspinas AS, Jensen JD, Marques-Bonet T, Alkan C, Prufer K, Meyer M, Burbano HA, Good JM, Schultz R, Aximu-Petri A, Butthof A, Hober B, Hoffner B, Siegemund M, Weihmann A, Nusbaum C, Lander ES, Russ , et al: A draft sequence of the Neandertal genome. Science. 2010, 328: 710-722.PubMedView ArticleGoogle Scholar
- Kim SY, Kim YS: A gene sets approach for identifying prognostic gene signatures for outcome prediction. BMC Genomics. 2008, 9: 177-PubMed CentralPubMedView ArticleGoogle Scholar
- Williams G: Adaptation and Natural Selection: A Critique of Some Current Evolutionary Thought. 1966, Princeton, NJ: Princeton Univ. PressGoogle Scholar
- Crews D: Diversity of hormone-behavior relations in reproductive behavior. Introduction to Behavioral Endocrinology. Edited by: Becker JBM, McCarthy M, Crews D. 1992, MIT Press/Bradford Books, 143-186. 2Google Scholar
- Crews D: The evolutionary antecedents to love. Psychoneuroendocrinology. 1998, 23 (8): 751-764.PubMedView ArticleGoogle Scholar
- Gowaty PA, Anderson WW, Bluhm CK, Drickamer LC, Kim YK, Moore AJ: The hypothesis of reproductive compensation and its assumptions about mate preferences and offspring viability. Proc Natl Acad Sci USA. 2007, 104 (38): 15023-15027.PubMed CentralPubMedView ArticleGoogle Scholar
- Mattle B, Wilson AB: Body size preferences in the pot-bellied seahorse Hippocampus abdominalis: choosy males and indiscriminate females. Behav Ecol Sociobiol. 2009, 63 (10): 1403-1410.PubMed CentralPubMedView ArticleGoogle Scholar
- Carson HL: The contribution of sexual behavior to Darwinian fitness. Behav Genet. 1987, 17 (6): 597-611.PubMedView ArticleGoogle Scholar
- Carson HL: Mate choice theory and the mode of selection in sexual populations. Proc Natl Acad Sci U S A. 2003, 100 (11): 6584-6587.PubMed CentralPubMedView ArticleGoogle Scholar
- Gowaty PA, Steinichen R, Anderson WW: Mutual interest between the sexes and reproductive success in Drosophila pseudoobscura. Evolution. 2002, 56 (12): 2537-2540.PubMedView ArticleGoogle Scholar
- Stunden CE, Bluhm CK, Cheng KM, Rajamahendran R: Factors affecting reproductive performance in captive Mallard ducks. Theriogenology. 1999, 52 (3): 435-446.PubMedView ArticleGoogle Scholar
- Drickamer LC, Gowaty PA, Holmes CM: Free female mate choice in house mice affects reproductive success and offspring viability and performance. Anim Behav. 2000, 59 (2): 371-378.PubMedView ArticleGoogle Scholar
- Drickamer LCGP, Wagner DM: Free mutual mate preferences in house mice affect reproductive success and offspring performance. Animal Behav. 2003, 65: 105-114.View ArticleGoogle Scholar
- Kirkpatrick M, Ryan MJ: The evolution of mating preferences and the paradox of the lek. Nature. 1991, 350: 33-38.View ArticleGoogle Scholar
- Ryan M: Sexual selection, sensory systems, and sensory exploitation. Oxford Survey Evol Biol. 1990, 7: 157-196.Google Scholar
- Beach F: Animal models for human sexuality. Sex, Hormones and Behaviour. Edited by: Potter RWJ. 1979, Amsterdam: Ciba Foundation Symposium 62, Excerpta Medica, 113-143.Google Scholar
- Crews D: Evolution of neuroendocrine mechanisms that regulate sexual behavior. Trends Endocrinol Metab. 2005, 16 (8): 354-361.PubMedView ArticleGoogle Scholar
- Paxinos G, Watson C: The Rat Brain in Stereotaxic Coordinates. 2007, New York: AcademicGoogle Scholar
- Bosotti R, Locatelli G, Healy S, Scacheri E, Sartori L, Mercurio C, Calogero R, Isacchi A: Cross platform microarray analysis for robust identification of differentially expressed genes. BMC Bioinforma. 2007, 8 (Suppl 1): S5-View ArticleGoogle Scholar
- Zhang B, Horvath S: A general framework for weighted gene co-expression network analysis. Stat Appl Genet Mol Biol. 2005, 4: Article17Google Scholar
- Zhu J, Wiener MC, Zhang C, Fridman A, Minch E, Lum PY, Sachs JR, Schadt EE: Increasing the power to detect causal associations by combining genotypic and expression data in segregating populations. PLoS Comput Biol. 2007, 3 (4): e69-PubMed CentralPubMedView ArticleGoogle Scholar
- Ravasz E, Somera AL, Mongru DA, Oltvai ZN, Barabasi AL: Hierarchical organization of modularity in metabolic networks. Science. 2002, 297 (5586): 1551-1555.PubMedView ArticleGoogle Scholar
- Langfelder P, Zhang B, Horvath S: Defining clusters from a hierarchical cluster tree: the dynamic tree cut package for r. Bioinformatics. 2008, 24 (5): 719-720.PubMedView ArticleGoogle Scholar
- Lum PY, Chen Y, Zhu J, Lamb J, Melmed S, Wang S, Drake TA, Lusis AJ, Schadt EE: Elucidating the murine brain transcriptional network in a segregating mouse population to identify core functional modules for obesity and diabetes. J Neurochem. 2006, 97 (Suppl 1): 50-62.PubMedView ArticleGoogle Scholar
- Chen Y, Zhu J, Lum PY, Yang X, Pinto S, MacNeil DJ, Zhang C, Lamb J, Edwards S, Sieberts SK, Leonardson A, Castellini LW, Wang S, Champy MF, Zhang B, Emilsson V, Doss S, Ghazalpour A, Horvath S, Drake TA, Lusis AJ, Schadt EE: Variations in DNA elucidate molecular networks that cause disease. Nature. 2008, 452: 429-435.PubMed CentralPubMedView ArticleGoogle Scholar
- Gargalovic PS, Imura M, Zhang B, Gharavi NM, Clark MJ, Pagnon J, Yang WP, He A, Truong A, Patel S, Nelson SF, Horvath S, Berliner JA, Kirchgessner TG, Lusis AJ: Identification of inflammatory gene modules based on variations of human endothelial cell responses to oxidized lipids. Proc Natl Acad Sci U S A. 2006, 103: 12741-12746.PubMed CentralPubMedView ArticleGoogle Scholar
- Horvath S, Zhang B, Carlson M, Lu KV, Zhu S, Felciano RM, Laurance MF, Zhao W, Qi S, Chen Z, Lee Y, Scheck AC, Liau LM, Wu H, Geschwind DH, Febbo PG, Kornblum HI, Cloughesy TF, Nelson SF, Mischel PS: Analysis of oncogenic signaling networks in glioblastoma identifies ASPM as a molecular target. Proc Natl Acad Sci U S A. 2006, 103: 17402-17407.PubMed CentralPubMedView ArticleGoogle Scholar
- Emilsson V, Thorleifsson G, Zhang B, Leonardson AS, Zink F, Zhu J, Carlson S, Helgason A, Walters GB, Gunnarsdottir S, Mouy M, Steinthorsdottir V, Eiriksdottir GH, Bjornsdottir G, Reynisdottir I, Gudbjartsson D, Helgadottir A, Jonasdottir A, Styrkarsdottir U, Gretarsdottir S, Magnusson KP, Stefansson H, Fossdal R, Kristjansson K, Gislason HG, Stefansson T, Leifsson BG, Thorsteinsdottir U, Lamb JR, Gulcher , et al: Genetics of gene expression and its effect on disease. Nature. 2008, 452: 423-428.PubMedView ArticleGoogle Scholar
- Schadt EE, Molony C, Chudin E, Hao K, Yang X, Lum PY, Kasarskis A, Zhang B, Wang S, Suver C, Zhu J, Millstein J, Sieberts S, Lamb J, GuhaThakurta D, Derry J, Storey JD, Avila-Campillo I, Kruger MJ, Johnson JM, Rohl CA, van Nas A, Mehrabian M, Drake TA, Lusis AJ, Smith RC, Guengerich FP, Strom SC, Schuetz E, Rushmore TH, et al: Mapping the genetic architecture of gene expression in human liver. PLoS Biol. 2008, 6: e107-PubMed CentralPubMedView ArticleGoogle Scholar
- Zhu J, Zhang B, Schadt EE: A systems biology approach to drug discovery. Adv Genet. 2008, 60: 603-635.PubMedView ArticleGoogle Scholar
- Draghici S, Khatri P, Tarca AL, Amin K, Done A, Voichita C, Georgescu C, Romero R: A systems biology approach for pathway level analysis. Genome Res. 2007, 17 (10): 1537-1545.PubMed CentralPubMedView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.