Sexual congruency in the connectome and translatome of VTA dopamine neurons

The ventral tegmental area (VTA) dopamine system is important for reward, motivation, emotion, learning, and memory. Dysfunctions in the dopamine system are linked to multiple neurological and neuropsychiatric disorders, many of which present with sex differences. Little is known about the extent of heterogeneity in the basic organization of VTA dopamine neurons with regard to sex. Here, we characterized the cell-specific connectivity of VTA dopamine neurons, their mRNA translational profile, and basic electrophysiological characteristics in a common strain of mice. We found no major differences in these metrics, except for differential expression of a Y-chromosome associated mRNA transcript, Eif2s3y, and the X-linked, X-inactivation transcript Xist. Of note, Xist transcript was significantly enriched in dopamine neurons, suggesting tight regulation of X-linked gene expression to ensure sexual congruency. These data indicate that the features that make dopamine neurons unique are highly concordant and not a principal source of sexual dimorphism.

dimorphism. These data suggest that the intrinsic properties of the VTA dopamine system in male and female mice is not a contributing factor to sexual mosaicism of the brain, and support the larger premise recently demonstrated in the human nervous system that there is no explicitly male or female brain 17 and meta-analysis of data from male and female rats demonstrating a lack of differential variability 18 .

Dopamine neuron connectivity.
To examine the inputs to VTA dopamine neurons, we performed conditional rabies viral tracing 14 . To achieve cell-specificity, a Cre-dependent adeno-associated viral vector (AAV1-EF1α-FLEX-GTB) containing the tumor virus A (TVA) and the rabies glycoprotein (RG) was injected into the VTA of mice expressing Cre recombinase from the endogenous dopamine transporter locus (Slc6a3 Cre/+ ) 19 . Two weeks following AAV injection, mice were injected with the glycoprotein gene deleted virus containing the avian sarcoma leucosis virus glycoprotein EnvA (EnvA-SAD-ΔG-mCherry; Fig. 1A-C). After nine days, pseudotyped rabies virus injection inputs to VTA dopamine neurons were assessed by immunohistochemical analysis of mCherry expression. We observed no significant difference in the expression of GFP or mCherry in the VTA of male (n = 5) and female (n = 7) of Slc6a3 Cre/+ mice ( Fig. 1D and Supplementary Figure S1A and B). Analysis of inputs to VTA dopamine neurons identified numerous brain regions, similar to those previously described 20 , and revealed an equivalent number of projections in female and male mice (962.4 ± 89.2, female vs. 802.2 ± 86.9, male; P = 0.24). Inputs to VTA dopamine neurons from specific brain regions were significantly correlated between sexes ( Fig. 1E; Slope = 1.0 ± 0.06, Pearson's r = 0.96, P < 0.0001); no interaction between brain region and sex was observed ( Fig. 1F and G; two-way ANOVA, P = 0.55).
In addition to cell-specific mapping using the rabies viral approach, we also mapped inputs to the VTA using a non-cell specific approach. To achieve this, we utilized the retrograde transducing virus canine adenovirus (CAV2) containing an expression cassette for Cre recombinase (CAV2-Cre) 21 . CAV2-Cre was injected into the midbrain of male (n = 5) and female (n = 5) Ai14 reporter mice (Gt(ROSA)26Sor tm(CAG-tdTomato)Hze -22 ; Fig. 2A-D), Cre-mediated expression of tdTomato was quantified across multiple brain regions, similar to our quantification of rabies virus labelling ( Fig. 2E-H). Although CAV2-Cre labelled a larger number of cells, the results were similar with regard to sex. Region-specific inputs were significantly correlated between sexes ( Figure S1H; Slope = 1.08 ± 0.08, Pearson's r = 0.95, P < 0.0001); no interaction between brain region and sex was observed ( Fig. 2G; two-way ANOVA, P = 0.22).
To establish the extent to which the basic connectivity of VTA dopamine neuron projections are similar, or dissimilar between male and female mice, we activated dopamine producing neurons using the stimulatory DREADD receptor HM3Dq 15 and assayed for induction of the immediate early gene Fos in downstream target structures. Two-weeks following injection of AAV1-FLEX-HM3Dq-YFP into the VTA of Slc6a3 Cre/+ female (n = 4) and male (n = 5) mice (Fig. 3A) we injected mice with the DREADD receptor agonist clozapine-N-oxide (CNO, 1 mg/kg i.p.) to induce neuronal activation. CNO induced robust Fos expression in the VTA along the rostral-caudal axis that did not differ between male and female mice ( Fig. 3B and C; two-way ANOVA, P = 0.75). Similarly, we observed significantly correlated Fos expression across multiple downstream brain regions between sexes ( Fig. 3D; Slope = 0.9 ± 0.04, Pearson's r = 0.99, P < 0.0001); no interaction between sex and brain region was observed ( Fig. 3E and F; two-way ANOVA, P = 0.99). Projections to the Fos-positive areas were confirmed through expression of the synaptic marker synaptophysin-GFP in VTA dopamine neurons by injecting Cre-dependent virus (AAV1-FLEX-Synapto-GFP) 23 into Slc6a3 Cre/+ mice ( Figure S2A and B).
Translational profile of VTA dopamine neurons. To determine the extent to which actively translating mRNA in dopamine neurons is similar between male and female mice, we utilized a RiboTag strategy that allows for cell-specific immuno-isolation of polyribosomal mRNA 16 . Slc6a3 Cre/+ mice were injected into the VTA with an AAV vector containing a Cre-dependent expression cassette for the affinity tagged ribosomal protein Rpl22 (AAV1-EF1α-Rpl22-HA 24 ; Fig. 4A). Three weeks following viral injection to allow for Rpl22-HA expression (Fig. 4B), the midbrain was microdissected and polyribosomes were immunoprecipitated (IP). Polyribosome-associated mRNA was reverse transcribed and cDNAs from male (n = 4) and female (n = 3) mice were analysed by Illumina microarray. Analysis of all IP transcripts from male and female mice revealed significantly correlated translatomes ( Fig. 4C; Slope = 0.98 ± 0.05, Pearson's r = 0.96, P < 0.0001). Of the > 18,000 probes, only two, the X-linked Xist and the Y-chromosome gene Eif2s3y, were identified as significantly different between female and male translatomes ( Fig. 4D and E). Of these, only Xist was significantly enriched in dopamine neurons (Fig. 4D). Analysis of enrichment (IP/input) showed 52 genes with greater than 4-fold enrichment, including canonical markers for dopamine synthesis and release (Fig. 4F). There was no significant interaction between gene and sex of enriched markers (two-way ANOVA, P = 0.99).
Intrinsic electrophysiological properties of dopamine neurons. Our connectivity and translatome analysis are consistent with equivalent organization of the VTA dopamine system in male and female mice. To explore this further, we analysed the translational profile of genes encoding ion channels that are known to regulate dopamine neuron activity and intrinsic excitability, as well as genes encoding neurotransmitter and neuropeptide receptors. Numerous ion channel subunits were expressed in dopamine producing neurons, but none were found to be differentially expressed in female and male mice (Fig. 5A). To establish which of these genes have specific expression in dopamine neurons we pooled male and female samples; twelve genes showed significant enrichment (Supplementary Figure S3A; Wilcoxon signed rank test, theoretical median 1; p < 0.05). Of these, six showed 4-fold or greater enrichment (Clcn6, Kcnd3, Kcnn3, Scn3b, Trpc4, and Trpc6).
Scientific REpORTS | 7: 11120 | DOI:10.1038/s41598-017-11478-5 To confirm these observations, we recorded the intrinsic electrophysiological properties of midbrain dopamine neurons using whole-cell patch clamp recordings from acute brain slices. Adult female and male Slc6a3 Cre/+ ;Gt(RO SA)26Sor tm(CAG-tdTomato)Hze mice (8 weeks of age) were utilized to genetically identify dopamine producing neurons by tdTomato fluorescence. Basic features including capacitance and input resistance did not differ between females and males ( Fig. 5B and C). Afterhyperpolarization induced current (I h ), a prominent feature of many, but not all dopamine neurons 25 , also did not differ ( Fig. 5D; Supplementary Figure S4A), consistent with equivalent expression of the hyperpolarization activated, cyclic nucleotide-gated potassium channel HCN2 (Hcn2, Fig. 5A). We also observed equivalent expression of the small conductance calcium-activated potassium channel, SK3 (Kcnn3, Fig. 5A), known to be enriched in dopamine neurons 26,27 . We did not observe differences between sexes in tail currents ( Fig. 5E; Supplementary Figure S4B), known to be principally mediated by SK3 channels 26 . Intrinsic pacemaker firing ( Fig. 5F; Supplementary Figure S4C) and current-induced spike firing ( Fig. 5G; Supplementary Figure S4D) also did not differ between sexes. Similar to ion channel subunits, analysis of neurotransmitter and neuropeptide receptors did not reveal sex-specific differences in actively translating mRNA (Fig. 6A). Twelve genes were found to be enriched (>2-fold); two neurotransmitter receptor subunits (Chrna6 and Chrnb3) and four neuropeptide receptors (Oprk1, Ghsr, Ntsr1, and Tacr3) were highly enriched (Supplementary Figure S3B). Consistent with our input mapping, analysis of excitatory and inhibitory synaptic connectivity, as measured by miniature inhibitory and excitatory postsynaptic currents (mIPSC and mEPSC), showed equivalent frequency and amplitude of mIPSCs (

Discussion
In this study we have analysed the cell-type specific connectivity of the VTA midbrain dopamine system, the translational profile of these neurons, and their basic electrophysiological properties in both male and female mice. Our study was specifically designed to not account for potential effects of circulating sex steroids, or sexual dimorphism in brain regions projecting to the VTA in order to establish whether analysis of this system in an unbiased manner would reveal differences in variability that may subsequently be accounted for by sex steroid differences or dimorphic organization of inputs to the regions. Our analysis reveals a highly congruent organization with no measurable differences between sexes relating to within subjects variability. Our cell-specific input analysis using Cre-dependent viral rabies tracing is consistent with numerous previous reports; however, previous studies did not investigate the extent of sex differences 20, 28-31 . Our findings match previous observations that the highest density of inputs to dopamine neurons arise from the DRN, BNST, NAc, PPTg, and LH. Further, our output analysis using conditional expression of the stimulatory DREADD receptor   HM4Dq in dopamine neurons confirmed previous observations of Fos expression following chemogenetic activation of dopamine neurons 32 . Interestingly, although we did not observe sex-specific differences in the input and output relationships of VTA dopamine neurons we did identify numerous brain regions projecting to the VTA and receiving input from the VTA that have been previously described as being sexually dimorphic in their organization. These include the BNST, medial amygdala, prefrontal cortex, and hypothalamus 13 . These findings are consistent with an organization in which sexually dimorphic nodes intersect with the non-dimorphic VTA dopamine system that are likely to give rise to sex-specific motivated behaviours, such as those described recently for MPA regulation of motivated behaviours in female mice 33 .
Similar to our analysis of the connectome of dopamine neurons, examination of the translatome of these neurons using the RiboTag method revealed enrichment of actively translating mRNAs that are consistent with previous analysis of gene expression in dopamine producing neurons [34][35][36][37][38] . We found no sex-specific differences in genes that were preferentially expressed in dopamine neurons; however, our analysis did reveal two sexually dimorphic genes: Xist and Eif2s3y. Xist is an X-linked gene whose purpose is to inactivate one of the two X chromosomes in females, to prevent gene dosage effects 39 . The reason for our observation of Xist transcript being associated with Rpl22 is unclear, but likely reflects the previous observations that Xist interacts with ribonuclear proteins involved in RNA splicing 40 and the recently identified role of Rpl22 in RNA splicing 41 . We find Xist to be >2-fold enriched in VTA dopamine neurons, suggesting that this cell type potently regulates X-linked gene expression to ensure a lack of gene dosage effects, further supporting the notion that the dopamine neurons of the VTA are programmed to ensure highly concordant gene expression. It is not surprising that Eif2s3y is only expressed in males. We find the X-linked homolog of Eif2s3y, Eif2s3x is expressed in both male and female dopamine neurons and is not enriched within these cells. Although the broad expression of Eif2s3y and Eif2s3x in the brain has been reported, the exact function of these genes is largely unknown 42 .
In agreement with our findings that the translational profile of dopamine neurons is equivalent between male and female mice, we did not find sex-specific differences in transcripts for ligand and non-ligand gated ion channels, and neurotransmitter and neuropeptide associated G-protein coupled receptors. Additionally, we did not find differences in the intrinsic electrophysiological properties of dopamine neurons. Although our findings indicate that the basic organization of the dopamine system is not different between male and female mice, it does not preclude the existence of differences within the system. It has been previously shown that psychostimulant drugs such as cocaine result in differential behavioural outcomes in male and female rats 43 and mice 11 . These behavioural differences likely reflect differential regulation of dopamine release and reuptake by sex steroids 11,44,45 , or differences in the in vivo electrophysiological properties of dopamine neurons in response to sex hormones 11,12 . Hormonal regulation of dopamine release in the PFC of male rats has also demonstrated 46 .
In summary, our data support the conclusion that the organizational principles of the VTA dopamine neurons are not sexually dimorphic, but rather likely reflect a fundamentally basic function of this system for reward processing, motivation, and emotional regulation. Based on the framework that sexual dimorphism can arise from several principal sources, including X-and Y-linked gene expression, sex steroids, sexually dimorphic inputs, and developmental programming, we conclude that the major sources that contribute to the previously ascribed dimorphism of the VTA dopamine system arise from sources outside the developmental organization of VTA dopamine neurons. These sources include hormonal regulation and circuits nodes upstream of dopamine neurons. Our data support the conclusion that the organization of the VTA dopamine system is invariant with regard to sex role allowing it to function as a basic facilitator of reward and motivational processes. Such an organization allows for flexibility in the control of sex-specific behaviours without the need for evolving multiple, intendent motivational systems.

Animals. All methods and experiments were approved by the University of Washington Institutional
Animal Care and Use Committee. All experiments were performed in accordance to guidelines and regulations. Gt(ROSA)26Sor tm(CAG-tdTomato)Hze mice aged 10 weeks were used for all CAV2-Cre experiments. Slc6a3 Cre/+ mice aged 10 weeks were used for all rabies virus, HM3Dq, and RiboTag experiments. Slc6a3 Cre/+ ;Gt(ROSA)26Sor tm(CAG-tdTomato)Hze mice aged 8 weeks were used for electrophysiology experiments. Both male and female mice were used in these experiments. Viral Injection. Connectivity. CAV2-Cre: All mice were anesthetized using isoflurane and stereotaxically injected bilaterally with CAV2-Cre (0.5 uL/side) into the VTA. Stereotaxic injection coordinates from bregma in mm, A-P: −3.25*x, M-L: ± 0.5, D-V: −4.5 (x = lambda:bregma distance/4.21) for the VTA. Mice were allowed to recover for two weeks before perfusion.
HM3Dq. All mice were anesthetized using isoflurane and stereotaxically injected bilaterally with AAV1-HM3Dq-YFP (0.5 uL/side) into the VTA. Stereotaxic injection coordinates from bregma in mm, A-P: −3.25*x, M-L: ± 0.5, D-V: −4.5 (x = lambda:bregma distance/4.21) for the VTA. Mice were allowed to recover for 2 weeks. Mice were then habituated to syringe injection with saline injections daily for 3 days. On the fourth day, mice were injected with 1 mg/kg of CNO and perfused 2 hours later.
Histology. Mice were anesthetized with 50 mg/kg of Beuthenasia and perfused with phosphate-buffered saline (PBS) and 4% paraformaldehyde. Whole brains were dissected and fixed overnight in paraformaldehyde, followed by immersion in a 30% sucrose solution for at least 48 hours. Brains were frozen in OCT at −20 degrees Celsius and sectioned coronally on a cryostat in 30 um sections. Sections were then stored in PBS and 0.1% sodium azide until immunostaining and/or mounting onto slides for imaging.
Immunostaining. cFos: Every other section from the entire brain was washed in 1x tris buffered solution (TBS) + 0.3% TritonX 100 (TBST) with 3% donkey serum for 30 minutes. Sections were then incubated overnight at 4 degrees Celsius or for 4 hours at room temperature in primary antibody (rabbit anti cfos, 1:2000, CalBiochem). This was followed by a 3x wash in TBS for 10 minutes and a 1 hour incubation at room temperature in secondary antibody conjugated to Cy3 or AF-488 at a 1:200 dilution. Finally, sections were washed in 1x TBS 3 more times before mounting onto slides.
Image Analysis. Fluorescent images of whole sections were acquired at 10x magnification (Keyence BZ-X710) and organized based on corresponding atlas reference figures (Mouse Brain Atlas, Franklin and Paxinos). Cells were then counted manually with the exception of cFos, which was counted using ImageJ software. This was done by analysing particles within individually drawn outlines of brain structures based on atlas reference figures.
RiboTag. Brain tissue from the VTA area was collected using a tissue punch and homogenized, as previously described 16 . Tissue was then incubated with 5 ul of anti-HA primary antibody (Covance) for four hours at 4 degrees Celsius, followed by overnight incubation with 200 ul of magnetic beads (Pierce). Next, RNA-conjugated beads were washed using a high salt buffer and the RNA was extracted from the magnetic beads. RNA was then purified using a RNeasy Plus Micro kit (Qiagen).
To confirm efficient enrichment for dopaminergic markers, qRT-PCR analysis was performed for Slc6a3. mRNA in IP versus input was quantified using a Ribogreen RNA kit (Invitrogen) and converted to cDNA using Superscript IV and oligo dT primers (Invitrogen). TaqMan primers (Applied Biosystems) for Slc6a3 were used to measure gene expression. Expression was quantified using the Ct values normalized to Actb (ΔCt). Fold enrichment of IP over input was calculated for using 2 −ΔΔCt . Of the four male and four female mice injected and processed, one female did not shown enrichment for Slc6a3 mRNA and was excluded from further analysis.
For recording miniature excitatory postsynaptic currents, electrodes were filled with an internal solution containing (in mM): 130 K-gluconate, 10 HEPES, 5 NaCl, 1 EGTA, 5 Mg-ATP, 0.5 Na-GTP, pH 7.3, 280 mOsm, and 200 μM picrotoxin was bath applied through the ACSF to block inhibit GABAA receptor-mediated events. For recording spontaneous inhibitory postsynaptic currents, electrodes were filled with an internal solution containing (in mM): 135 KCl, 12 NaCl, 0.05 EGTA, 100 HEPES, 0.2 Mg-ATP, 0.02, Na-GTP (include pH and osmolarity here); 2 mM kynurenic acid was bath applied through the ACSF to block glutamatergic synaptic transmission. For all miniature current recordings, cells were clamped at a holding potential of −60 mV for a minimum of 5 minutes and were recorded in the presence of 1 mM tetrodoxin (TTX) to block action potentials. Access resistance was monitored throughout all experiments.
Statistical Analysis. All statistical analysis was done using Prism (GraphPad). Data Availability. All data are available upon request.