Vasoactive Intestinal Peptide Modulation of the Steroid-Induced LH Surge Involves Kisspeptin Signaling in Young but Not in Middle-Aged Female Rats

Abstract

Age-related LH surge dysfunction in middle-aged rats is characterized, in part, by reduced responsiveness to estradiol (E2)-positive feedback and reduced hypothalamic kisspeptin neurotransmission. Vasoactive intestinal peptide (VIP) neurons in the suprachiasmatic nucleus project to hypothalamic regions that house kisspeptin neurons. Additionally, middle-age females express less VIP mRNA in the suprachiasmatic nucleus on the day of the LH surge and intracerebroventricular (icv) VIP infusion restores LH surges. We tested the hypothesis that icv infusion of VIP modulates the LH surge through effects on the kisspeptin and RFamide-related peptide-3 (RFRP-3; an estradiol-regulated inhibitor of GnRH neurons) neurotransmitter systems. Brains were collected for in situ hybridization analyses from ovariectomized and ovarian hormone-primed young and middle-aged females infused with VIP or saline. The percentage of GnRH and Kiss1 cells coexpressing cfos and total Kiss1 mRNA were reduced in saline-infused middle-aged compared with young females. In young females, VIP reduced the percentage of GnRH and Kiss1 cells coexpressing cfos, suggesting that increased VIP signaling in young females adversely affected the function of Kiss1 and GnRH neurons. In middle-aged females, VIP increased the percentage of GnRH but not Kiss1 neurons coexpressing cfos, suggesting VIP affects LH release in middle-aged females through kisspeptin-independent effects on GnRH neurons. Neither reproductive age nor VIP affected Rfrp cell number, Rfrp mRNA levels per cell, or coexpression of cfos in Rfrp cells. These data suggest that VIP differentially affects activation of GnRH and kisspeptin neurons of female rats in an age-dependent manner.

Female reproductive senescence in rats is heralded by delayed and attenuated LH surges (1–3) and characterized by reduced activation of GnRH neurons (4–7). Age-related changes in the LH surge mechanism reflect reduced responsiveness of GnRH neuron excitatory and inhibitory afferent input to estradiol (E2)-positive feedback conditions (1, 4, 8). The mechanisms that give rise to age-related LH surge dysfunction are minimally understood; LH surge dysfunction does not result from disparities in ovarian steroid exposure (9), primary pituitary dysfunction (10–12), age-related changes in hypothalamic estrogen receptor expression (13, 14), or reduced sex steroid receptor binding in the hypothalamus (15). Moreover, age-related LH surge dysfunction does not result from reduced numbers of GnRH neurons (5, 16), the ability of GnRH neurons to secrete GnRH peptide in response to a depolarizing event (17), or reduced pituitary responsiveness to GnRH (10). However, female reproductive senescence is characterized by reduced noradrenergic (18), glutamatergic (9, 19–21), and kisspeptinergic (22–24) and increased γ-aminobutyric acid (GABA)ergic (21) neurotransmission. These observations suggest that characteristic age-related impairments in the LH surge result from events “upstream” of GnRH neurons and possibly upstream of kisspeptin.

The neuropeptide kisspeptin (encoded by Kiss1) potently excites GnRH neurons and is critical for reproductive function, including the LH surge (25–29). In rodents, E2-responsive kisspeptin neurons in the anteroventral periventricular (AVPV) nucleus of the hypothalamus are hypothesized to have a key role in the regulation of the LH surge (30–32). We recently reported that reproductive senescence in middle-aged female rats is characterized by reduced total Kiss1 mRNA expression in the anterior hypothalamus (containing the AVPV), and fewer detectable kisspeptin-immunoreactive neurons in the AVPV (22, 23) under E2-positive feedback conditions compared with young female rats. Additionally, we demonstrated that reverse microdialysis of kisspeptin into the medial preoptic area of the hypothalamus of middle-aged female rats restores LH surge amplitude (23). These findings suggest that age-related LH surge dysfunction results, in part, from reduced kisspeptin drive under E2-positive feedback. The mechanism resulting in reduced kisspeptin output is unclear but may involve diminished stimulatory input into kisspeptin cells (33).

The suprachiasmatic nucleus (SCN) integrates and synchronizes many diverse neuroendocrine events required for an appropriately timed GnRH-LH surge and activation of GnRH neurons (30, 34–40). Vasoactive intestinal peptide (VIP) neurons and vasopressin cells located in the ventrolateral and dorsomedial SCN, respectively, (41) are hypothesized to regulate LH release and the LH surge (42–46). Moreover, it is hypothesized that a critical level of VIP signaling is required for appropriate LH pulse frequencies and induction of an appropriately timed LH surge (38, 47–49). Consistent with this hypothesis, VIP neurons located in the SCN project to GnRH neurons (36), VIP activates GnRH neurons in a time- and E2-dependent fashion (44, 50, 51), and VIP receptors are located on both GnRH neurons and proximal to astrocytes that ensheath GnRH neurons (52, 53). Additionally, icv infusion of VIP induces GnRH (54) and LH release in E2-primed females with SCN lesions (55). Of particular interest, LH surge dysfunction in reproductively senescing female rats is characterized by reduced cfos expression (56) and VIP mRNA expression in the SCN and reduced activation of GnRH neurons with VIP contacts (49, 51). Additionally, the LH surge in young rats is delayed and attenuated, like middle-aged rats, by infusion of VIP antiserum into the third ventricle (57), infusion of VIP antisense oligonucleotides into the SCN (48), or thermal ablation of VIP neurons in the SCN (45). We recently reported that icv infusion of VIP in middle-aged females during E2-positive feedback conditions rescues the LH surge as well as increases the percentage of GnRH neurons coexpressing cfos (6). Thus, LH surge dysfunction in middle-aged rats most likely results from reduced VIP secretion rather than reduced responsiveness to VIP. We and others also found that icv infusion of VIP treatment in young females attenuated LH release (47, 58, 59) and decreased coexpression of cfos in GnRH neurons (6). These observations suggest that a critical level of VIP signaling is required to activate GnRH neurons for an appropriately timed and robust LH surge in young and middle-aged females. However, the mechanism(s) and neuronal targets underlying age-dependent responses to VIP remain unknown.

RFamide-related peptide-3 (RFRP-3; the avian ortholog for gonadotropin-inhibitory hormone) is an inhibitory regulator of LH secretion and GnRH neuronal activation (60, 61). RFRP-3 expressing neurons, found exclusively in the hypothalamic dorsal medial nucleus (DMN), express estrogen receptor α, and send projections to GnRH somata and the median eminence (60, 62, 63). RFRP-3 neurons are hypothesized to “turn off” around the time of the LH surge, thereby disinhibiting GnRH neurons and allowing enhanced secretion of GnRH and induction of the LH surge. Of note, SCN-derived fibers project to both the AVPV (the location of kisspeptin neurons) and DMN (the location of RFRP-3 neurons). Moreover, Kiss1 and Rfrp mRNA levels, and cfos coexpression in Kiss1 and Rfrp neurons, respond to circadian signals and exhibit circadian changes on proestrus that correlate with circadian changes in LH levels (25, 63–65). These observations raise the possibility that VIP regulates the timing and amplitude of the LH surge by modulating activation of RFRP-3 as well as AVPV/periventricular nucleus (PeN) kisspeptin neurons. We therefore hypothesized that altered VIP signaling in young or middle-aged female rats affects the activation of GnRH neurons and the LH surge by modulating either the number of Kiss1 and Rfrp neurons, Kiss1 and Rfrp mRNA expression per cell, or the percentage of AVPV Kiss1 and Rfrp neurons that coexpress cfos during E2-positive feedback conditions.

Materials and Methods

Animals and hormone administration

Young (2–3 months) and middle-aged (retired breeders, 9–11 months) female Sprague Dawley rats (Taconic Farms) were housed individually and maintained on a 14-hour light, 10-hour dark cycle (lights on at 6:00 am) with free access to chow and water. Only rats with at least 2 regular 4–5 day estrous cycles were used. All procedures followed the NIH Guide for the Care and Use of Laboratory Rats and were approved by the Institutional Animal Care and Use Committee at the Albert Einstein College of Medicine. To induce LH surges, E2 benzoate (EB) and progesterone (P) (Steraloids, Inc) were dissolved in peanut oil and administered sc in a volume of 0.1 mL. At 9:00 am, 7 days after ovariohysterectomy (OVX) and cannula placement, rats received the first of 2 daily injections of 2 μg of EB. At 9:00 am, 2 days after the first EB injection, rats were injected with 500 μg of P. When females fail to mount a LH surge they fail to coexpress cfos in GnRH neurons (66); therefore, based on our previous work with this model (6), we only include data from animals that have GnRH neurons that coexpress cfos.

Third ventricle cannulation and icv drug administration

For OVX and stereotaxic surgery, rats were anesthetized with im ketamine/xylazine (80 and 4 mg/kg, respectively). After OVX, females were placed in a Kopf stereotaxic apparatus, and a 22-gauge icv guide cannula (Plastics One) was placed into the third ventricle (anterior/posterior + 0.2 mm; medial/lateral +0.0 mm; dorsal/ventral −9.8 mm relative to Bregma) and plugged with a 26-gauge dummy that extended 1 mm below the guide (6). Animals recovered for 7 days before steroid priming and additional manipulations. Two days after the first EB injection, females were connected to an automatic pump (Bioanalytical System, Inc) with microinjection syringe and attached to a tether allowing rats to move freely. Human VIP (6 nmol in 45 μL; Bachem; n = 8 per group) or saline (vehicle/control; n = 8 per group) was continuously infused into the third ventricle of young and middle-aged rats for approximately 3 hours between 1:00 and 4:00 pm at 2 nmol/15 μL/h (6), a dose that rescues the LH surge in middle-aged females and a time frame coincident with the LH surge in young reproductive-aged females (6).

Brain tissue preparation

After being humanely destroyed, brains were collected, immediately frozen on dry ice, and stored at −80°C. For each brain, 5 series of 20-μm brain sections encompassing the entire forebrain and hypothalamus were cut and thaw mounted onto Superfrost plus slides. Slides were stored at −80°C until processing via single- or double-label in situ hybridization (ISH). Only sections from brains in which placement in the third ventricle was confirmed were included in our analysis.

Single-label and double-label ISH

Single-label ISH for Kiss1 or Rfrp was performed as previously described (62, 64, 67–69). Briefly, slide-mounted brain sections encompassing the entire AVPV/PeN (Kiss1) or DMN (Rfrp) were fixed in 4% paraformaldehyde, pretreated with acetic anhydride, rinsed in 2× SSC (sodium citrate, sodium chloride), delipidated in chloroform, dehydrated in ethanol, and air dried. Radiolabeled (³³P) Kiss1 or Rfrp antisense riboprobe (0.05 pmol/mL) was combined with tRNA, heat-denatured, added to hybridization buffer, and applied to each slide (100 μL/slide). Slides were coverslipped and placed in a 55°C humidity chamber overnight. The slides were then washed in 4× SSC and placed into ribonuclease (RNAse) treatment for 30 minutes at 37°C, and then in RNAse buffer without RNAse at 37°C for 30 minutes. After washing in 2× SSC at room temperature, slides were washed in 0.1× SSC at 62°C for 1 hour, dehydrated in ethanols, and air dried. Slides were then dipped in Kodak NTB emulsion, air dried, and stored at 4°C for 4–5 days (depending on the assay) before being developed and coverslipped. Previous studies in the laboratory determined that no staining was detected with sense probes.

For double-label ISH analysis of cfos induction in GnRH, Kiss1, or Rfrp neurons, slide-mounted brain sections were treated similarly to single-label ISH with the following modifications. Digoxigenin (DIG)-labeled antisense mouse GnRH, Kiss1, or Rfrp cRNA were synthesized with DIG labeling mix (Roche). Radiolabeled (³³P) antisense cfos (0.05 pmol/mL) and DIG-labeled (1:500) riboprobes were combined with tRNA, heat denatured, and dissolved together in hybridization buffer. The probe mix was applied to slides (100 μL/slide), and slides were hybridized at 55°C overnight. After the 62°C washes on day 2, slides were incubated in 2× SSC with 0.05% Triton X-100 containing 3% normal sheep serum for 1 hour at room temperature. Slides were then incubated overnight at room temperature with anti-DIG antibody conjugated to alkaline phosphatase (Roche; diluted 1:500 in buffer 1 containing 1% normal sheep serum and 0.3% Triton X-100). Slides were then washed with buffer 1 and incubated with Vector Red alkaline phosphatase substrate (Vector Labs) for 1 hour at room temperature. The slides were then air dried, dipped in emulsion, stored at 4°C, and developed 7–9 days later, depending on the assay.

Quantification and analysis of ISH data

ISH slides were analyzed with an automated image processing system (Dr. Don Clifton, University of Washington) by a person blinded to the treatment groups. For single-label experiments, the software counts the number of silver grain clusters representing cells, as well as the number of silver grains over each cell (a semiquantitative index of mRNA content per cell) (69–71). Cells were considered Kiss1 or Rfrp positive when the number of silver grains in a cluster exceeded that of background by 3-fold (72). For double-label assays, red fluorescent DIG-containing cells (GnRH, Kiss1, or Rfrp cells) were identified under fluorescence microscopy, and the grain-counting software was used to quantify silver grains (representing cfos mRNA) overlying each DIG cell. Signal-to-background ratios for individual cells were calculated, and a cell was considered double labeled if its ratio was greater than 3 (68, 73).

Statistical analysis

Data are expressed as mean ± SEM. Two-way ANOVA (age × treatment) was used to determine differences in GnRH, Rfrp, and Kiss1 cell numbers, mRNA/cell and percentage of neurons expressing cfos. Bonferroni or Tukey's post hoc tests were performed as appropriate. Nonparametric testing was used when data were not normally distributed. P < .05 was considered statistically significant.

Results

Central VIP infusion increases cfos coexpression in GnRH neurons of middle-aged females but reduces cfos coexpression in GnRH neurons of young females

This experiment determined whether icv infusion of VIP affects the activation of GnRH neurons on the day of the LH surge in young and middle-aged females. OVX, EB+P-primed rats continuously infused with saline or VIP between 1:00 and 4:00 pm on the day of the LH surge were humanely destroyed immediately upon completion of VIP infusion, a time interval that corresponds with maximal activation of GnRH neurons (6). Brains were collected and processed for double-label ISH, and the percentage of GnRH cells coexpressing cfos, a marker of neuronal activation, was quantified in the medial sepal nucleus, organum vasculosum of the lamina terminalis, and preoptic area regions of the hypothalamus. Consistent with our previous immunohistochemistry results (6), double-label ISH analysis demonstrated that approximately 55% of GnRH neurons coexpress cfos in EB+P-primed and saline-infused young rats (Figure 1A and E). Intracerebroventricular infusion of VIP in young females significantly reduced the percentage of GnRH neurons coexpressing cfos to 30% (P < .01; Figure 1, C and E). Also consistent with our previous report (6), the percentage of GnRH neurons coexpressing cfos was significantly lower in middle-aged females than young females infused with saline (P < .01; Figure 1, B and E). In contrast to young females, icv infusion of VIP in middle-aged females resulted in a 1.5-fold increase in the percentage of GnRH neurons coexpressing cfos as compared with saline-infused middle-aged controls (P < 0.01; Figure 1, D and E). Middle-aged female rats had equivalent numbers of GnRH neurons as young rats, and infusion of VIP did not affect absolute numbers of GnRH neurons (data not shown).

Figure 1.

GnRH neuronal activation in young and middle-aged females infused with VIP. Representative photomicrographs (200× magnification) showing GnRH mRNA-positive cells with or without cfos mRNA in the OVLT of OVX, EB+P-treated young female rats icv infused with saline (A) or VIP (C). Representative photomicrographs (200× magnification) showing GnRH mRNA-positive cells with or without cfos mRNA in the OVLT of OVX, EB+P-treated middle-aged female rats icv infused with saline (B) or VIP (D). GnRH-containing neurons were visualized with red fluorescence, and cfos mRNA was marked by the presence of silver grains. Yellow arrows denote examples of GnRH cells coexpressing cfos; blue arrows denote example GnRH cells without cfos. E, Mean ± SEM percentage of GnRH mRNA-containing neurons that coexpress cfos in OVX, EB+P-treated young and middle-aged female rats humanely destroyed after 3 hours of VIP or saline infusion. *, P < .001 vs age-matched saline control; **, P < .001 vs young saline; ***, P < .001 vs young VIP. Y, young; MA, middle-aged; n = 8 animals per group.

VIP-dependent regulation of GnRH neuronal activation does not involve changes in Rfrp mRNA expression or neuronal activation in young or middle-aged females

RFRP-3 neurons send neural projections to GnRH neurons (25, 60, 63), and icv RFRP-3 administration suppresses LH levels and cfos expression in GnRH neurons (74). This experiment determined whether the age-dependent effects of VIP on GnRH neuronal activation in young and middle-aged females involve RFRP-3 neurons. Two interspersed populations of Rfrp-expressing cells exist in DMN; one cell group demonstrates high Rfrp expression (HE) and the other low Rfrp expression (LE) (Figure 2C) (25). Neither the number of HE nor LE Rfrp-containing cells were affected by reproductive age or VIP treatment (data not shown). Similarly, there was no effect of reproductive age or VIP on total numbers of Rfrp expressing cells (all HE + LE cells; Figure 2, A, B, and G) or amount of Rfrp mRNA per cell (Figure 2H). Lastly, age did not affect the percentage of Rfrp neurons coexpressing cfos (Figure 2I), as assessed with double-label in situ hybridization. Likewise, VIP treatment did not affect cfos induction in Rfrp neurons in young or middle-aged females (Figure 2I). As with the single-label Rfrp assay, the mean number of detectable Rfrp cells in this double-label assay also did not differ significantly between groups (young saline: 704 ± 31; young VIP: 756 ± 34; middle-aged saline: 673 ± 25; middle-aged VIP: 671 ± 27).

Figure 2.

Rfrp gene expression levels and Rfrp neuronal activation in young and middle-aged female rats infused with saline or VIP. A, Representative dark-field photomicrographs (×100) showing Rfrp mRNA containing cells in the DMN of OVX, EB+P-treated young rats icv infused with saline. B, Representative dark-field photomicrograph (×100) showing Rfrp mRNA containing cells in the DMN of OVX, EB+P-treated middle-aged rats icv infused with saline. C, Representative high-magnification photomicrograph (×200) showing HE and LE Rfrp cells in a young saline-treated female. D and E, Representative photomicrographs (×100) showing Rfrp mRNA containing cells (red fluorescence), with or without cfos coexpression (silver grains) in the DMN of steroid-treated young and middle-aged saline-infused rats. F, Enlarged higher magnification of the yellow dotted box in panel E showing Rfrp cells coexpressing (yellow arrows) or not (blue arrows) cfos in a middle-aged saline-infused rat. G, Mean ± SEM number of Rfrp cells in the DMN of young and middle-aged female rats. H, Mean ± SEM number of silver grains per Rfrp cell in the DMN, indicative of relative Rfrp mRNA per cell. I, Mean ± SEM percentage of Rfrp cells coexpressing cfos mRNA in the DMN of young and middle-aged female rats treated with VIP or saline. Y, young; MA, middle-aged; 3V, third ventricle; n = 8 animals per group.

Intracerebroventricular infusion of VIP affects Kiss1 neurons in young but not middle-aged female rats

Kiss1 neurons in the AVPV/PeN are hypothesized to mediate E2-positive feedback effects on the GnRH/LH surge (28, 31, 64, 73, 75, 76). Given VIP's ability to differentially modulate GnRH neurons and LH release in young and middle-aged rats, we determined whether VIP affects the number, expression level, or activation of AVPV/PeN Kiss1 cells. In young females, VIP infusion did not affect the number of detectable Kiss1 cells or Kiss1 mRNA content per cell in the AVPV/PeN (Figure 3, A, B, E, and F). However, double-label analysis revealed that whereas 40% of Kiss1 neurons in saline-treated young females coexpressed cfos, VIP infusion markedly decreased the percent of Kiss1 neurons with cfos to 15% (P < 0.01 compared with saline-infused; Figure 4, A, C, and E). As with the single-label Kiss1 assay, the number of detectable Kiss1 cells in the double-label assay did not differ between groups (young saline: 156 ± 13 cells; young VIP: 157 ± 15 cells; middle-aged saline: 162 ± 20 cells; middle-aged VIP: 195 ± 35 cells).

Figure 3.

Kiss1 mRNA per cell decreases with reproductive age. Representative dark-field photomicrographs (×100) showing Kiss1 mRNA expressed in the AVPV/PeN of OVX, EB+P-treated young female rats icv infused with saline (A) or VIP (B). Representative dark-field photomicrographs (×100) showing Kiss1 mRNA expressed in the AVPV/PeN of OVX, EB+P-treated middle-aged female rats icv infused with saline (C) or VIP (D). E, Mean ± SEM total number of Kiss1 mRNA-containing cells in the AVPV/PeN of young and middle-aged females. F, Mean ± SEM level of Kiss1 mRNA per cell in the AVPV/PeN of young and middle-aged females; n = 8 animals per group. *, P < .01 vs Y saline and Y VIP. MA, middle aged; Y, young; 3V, third ventricle.

Figure 4.

Central VIP infusion inhibits Kiss1 cellular activation in young but does not affect Kiss1 cells in middle-aged female rats. Representative photomicrographs (×200) showing Kiss1 mRNA containing cells (red fluorescence) and cfos mRNA cells (silver grain clusters) in the AVPV/PeN of OVX, EB+P-treated young females infused with saline (A) or VIP (C). Representative photomicrographs (×200) showing Kiss1 mRNA-containing cells (red fluorescence) and cfos mRNA cells (silver grain clusters) in the AVPV/PeN of OVX, EB+P-treated middle-aged females infused with saline (B) or VIP (D). Yellow arrows denote examples of Kiss1 cells coexpressing cfos. Blue arrows denote examples of Kiss1 cells that do not coexpress cfos.E, Mean ± SEM percentage of Kiss1 mRNA-containing neurons in the AVPV/PeN that coexpress cfos mRNA in young and middle-aged female rats. *, P < .01 vs young saline; Y, young, MA, middle-aged; 3V, third ventricle; Sal, saline; n = 8 animals per group.

Reduced input from kisspeptin neurons in the AVPV/PeN might result in the delayed and attenuated LH surge in middle-aged rats (22). To determine whether icv VIP rescued activation of GnRH neurons and GnRH-LH release (6) in middle-aged females through effects on AVPV/PeN Kiss1 cells, we quantified the number of Kiss1 cells, levels of Kiss1 mRNA per cell, and the percentage of Kiss1 cells coexpressing cfos. Compared with young females, there was no significant difference in the total number of Kiss1-mRNA-expressing cells in middle-aged females (Figure 3, A, C, and E). However, Kiss1 mRNA per cell in saline-treated middle-aged females was significantly reduced by approximately 20%, compared with saline-treated and VIP-treated young females (Figure 3F; P < .01). Additionally, saline-treated middle-aged rats showed a 65% reduction in the proportion of Kiss1 cells coexpressing cfos compared with young saline-infused females (Figure 4, A, B, and E; P < .01). Although VIP infusion increased the percentage of GnRH neurons coexpressing cfos in middle-aged females (Figure 1), VIP did not affect numbers of AVPV/PeN Kiss1 cells or the percentage of Kiss1 cells coexpressing cfos in the same middle-aged females (Figure 3, E and F, and Figure 4E). Lastly, VIP tended to increase Kiss1 mRNA content per cell in middle-aged females such that Kiss1 mRNA content was not significantly different than saline-infused young or middle-aged controls; however, this effect was not statistically significant (P = 0.08; Figure 3F).

Discussion

Female reproductive senescence is heralded by reduced responsiveness of the reproductive neuroendocrine axis to E2-positive feedback effects. We report that when compared with young females, middle-aged females experience reduced GnRH and Kiss1 neuronal activation (as determined by cfos coexpression). Moreover, in middle-aged females, icv infusion of VIP under E2-positive feedback conditions significantly increased the percentage of GnRH neurons coexpressing cfos without significantly affecting Kiss1 mRNA expression per cell, the absolute number of Kiss1 neurons, or the percentage of Kiss1 cells coexpressing cfos. These data suggest that age-related LH surge dysfunction in middle-aged females reflect reduced availability of an appropriate level of VIP needed to activate GnRH and non-Kiss1 neurons under E2-positive feedback conditions. In contrast, icv VIP infusion in young females resulted in reduced cfos expression in GnRH cells, suggesting that in young females, extended exposure to increased VIP levels in the brain may desensitize GnRH neurons or their afferent inputs to the stimulatory effects of VIP (50, 54). Alternatively in young females extended exposure to elevated brain VIP may directly or indirectly inhibit GnRH neuron activation under E2-positive feedback conditions. Electrophysiological and pharmacodynamics studies are needed to ultimately determine how exogenous VIP affects GnRH neurons in young females during E2-positive feedback conditions. In young females, icv infusion of VIP also markedly attenuated the percentage of Kiss1 neurons that coexpress cfos, suggesting that VIP may diminish the LH surge in young females (6, 58), in part, by attenuating excitatory kisspeptin afferent input to GnRH neurons. We also determined whether reduced responsiveness to E2-positive feedback conditions in middle-aged females reflected increased inhibitory tone from Rfrp cells located in the DMN. Additionally, reduced responsiveness to E2-positive feedback in middle-aged females, as defined by decreased GnRH and Kiss1 neuronal activation, did not correlate with increased numbers of Rfrp cells, increased Rfrp mRNA levels per cell, or increased Rfrp neuronal activation. However, our studies do not rule out the possibility that advancing reproductive age disrupts Rfrp protein levels or expression of its cognate receptor. Nonetheless, age-dependent changes in E2-induction of the LH surge and activation of GnRH neurons in middle-aged females most likely does not reflect significantly increased RFRP-3 tone. Overall, these data suggest that critical levels of VIP are required to stimulate a robust LH surge in young and middle-aged female rats and support independent roles for VIP as well as kisspeptin neurotransmitter systems in female reproductive senescence.

Kisspeptin and female reproductive senescence

This study supports our previous findings that middle-aged females have significantly less Kiss1 mRNA expression in the AVPV/PeN (22) and further demonstrates that fewer numbers of Kiss1 neurons coexpress cfos in middle-aged females. Our findings are consistent with a recent study from Ishii et al (33) and suggest that fewer Kiss1 neurons in the AVPV/PeN are activated under E2-positive feedback conditions in middle-aged females. We previously demonstrated that kisspeptin infusion in middle-aged females increases glutamate and reduces GABA in middle-aged females (9). Our new data suggest that reduced activation of kisspeptin neurons in middle-aged females may result in the imbalance of excitatory and inhibitory neurotransmission observed in kisspeptin-deficient females and contribute to negative downstream effects on GnRH neuron activation and LH release (9, 23, 29).

VIP modulates GnRH neuron activation in young and middle-aged females

Middle-aged females with reduced percentages of GnRH neurons that coexpress cfos on the day of the LH surge have reduced VIP contacts (51), as well as a loss in diurnal rhythmic expression of VIP expression in the SCN (49). Moreover, infusion of antisense VIP oligonucleotide into the SCN of young females results in delayed and attenuated LH surge patterns that are reminiscent of the LH surge observed in middle-aged females (48). Lastly, unilateral lesions in the SCN result in reduced numbers of GnRH neurons that coexpress cfos neurons on the ipsilateral side of the lesion during the LH surge. Thus reduced VIP tone or responsiveness to VIP may result in the delayed and attenuated LH surge in middle-aged females (6, 48, 51). Here we demonstrate that icv infusion of VIP increased the percentage of GnRH neurons coexpressing cfos in middle-aged females. These data suggest that GnRH neurons in middle-aged rats remain appropriately responsive to VIP. They also suggest reduced availability of intrahypothalamic VIP most likely contributes, in part, to the characteristic LH surge changes observed in middle-aged females.

VIP does not modulate activation of Kiss1 neurons in middle-aged females

The presence of VIP receptors and fibers in the AVPV region, a key nucleus that projects to GnRH neurons involved in the generation of the GnRH/LH surge (3, 77), raised the possibility that VIP affects AVPV kisspeptin neurons, known modulators of the LH surge. We found that female reproductive aging is characterized by changes in the activation of kisspeptin neurons in the AVPV. The mechanisms by which reproductive senescence might affect kisspeptin neurons are minimally understood (23, 33). VIP neurons provide afferent input to E2-responsive and VIP receptor-expressing cells located in the AVPV (78, 79). Of note, circadian changes in the expression of AVPV/PeN Kiss1 mRNA expression parallel VIP as well as LH release (64, 65). We therefore determined whether VIP affected the percentage of activated Kiss1 cells located in the AVPV/PeN. However, icv infusion of VIP in EB+P-primed middle-aged females did not affect the percentage of Kiss1 neurons coexpressing cfos. Thus, it is unlikely that VIP stimulates the LH surge in middle-aged females through effects that involve activation of the kisspeptin system (80). However, because we did not measure kisspeptin protein levels or secretion, we cannot rule out the possibility that VIP affects kisspeptin protein synthesis or secretion. We previously reported that icv VIP infusion increased the absolute number of cfos -positive non-GnRH neurons in the hypothalamus of middle-aged females (6). Thus, it is possible that VIP activates nonkisspeptin cells in the AVPV, which then modulate excitatory as well as inhibitory input to GnRH neurons (23, 81). In support of this hypothesis, we visually observed higher cfos expression throughout the AVPV and preoptic area in VIP-treated compared with saline-infused middle-aged females (compare overall cfos expression in Figure 4, B and D, as well as Figure 1, A and C), suggesting that VIP activates numerous nonkisspeptin cells in these brain regions, some of which may regulate GnRH neurons. The identity of these VIP-activated cfos-expressing cells will be an important avenue for future investigation.

GnRH neurons are reported to be regulated by fast synaptic input from the neurotransmitters GABA and glutamate (82). Moreover, intrahypothalamic glutamate and increased GABA neurotransmission under E2-positive feedback conditions can induce phenotypic changes in LH surge patterns attributed to reproductive senescence (9). Thus, it is possible that some of the observed kisspeptin-independent effects of VIP infusion on GnRH neuronal activation and non-kisspeptin neurons might be mediated indirectly by VIP effects on GABA or glutamate signaling. Alternatively, dynamic interactions between neurons and astrocytes are essential for proestrus and the preovulatory LH surge (52). In vivo studies showed that VIP decreases astrocyte surface area around GnRH neurons at the time of the LH surge and could therefore indirectly enhance stimulatory inputs to GnRH neurons. Thus, reduced VIP tone in reproductively senescing females may contribute to altered GnRH activity via astrocyte-dependent mechanisms, although this remains to be tested.

VIP modulates Kiss1 neuron activation in young reproductive-age females

VIP cells located in the SCN send projections to GnRH neurons (44, 50), and VIP alters both GnRH peptide release (54) and the timing of the GnRH/LH preovulatory surge in young females (46). However, unlike middle-aged females, extended icv infusion of VIP in young females inhibits both GnRH coexpression of cfos and LH release (6, 47, 58, 59). We investigated whether increased icv VIP in young females affects Kiss1 neurons in the AVPV/PeN. We observed a significant reduction in the percentage of Kiss1 neurons coexpressing cfos in young females icv infused with VIP, correlating with similar decreases in GnRH neuronal activation in these females. This suggests that reduced GnRH activation in VIP-treated young females may be caused by direct or indirect effects of VIP on activation of kisspeptin neurons upstream of GnRH neurons. Of interest, it was recently reported that kisspeptin neurons in the AVPV of mice receive projections from vasopressin (AVP) cells located in the SCN, express AVP receptors (65), and AVP rather than VIP drives circadian rhythmicity of cfos expression in kisspeptin neurons (80). Moreover, VIP connections to AVPV/PeN kisspeptin neurons were reported to be rare in mice (83), suggesting that exogenous VIP in young females may indirectly affect kisspeptin cells, perhaps by reduced sensitivity or inhibition of afferent neurons that regulate activation of AVPV/PeN kisspeptin neurons. However, it is important to note that this VIP-kisspeptin connectivity has not been assessed in rats, nor has VIP receptor expression in kisspeptin cells been examined in any rodent models. Thus, exact effects of VIP on kisspeptin cells in rats cannot be fully described at this time. Additionally, the present study does not rule out the possibility that differences in the response of kisspeptin neurons in young and middle-aged females to VIP brain infusion reflects an age-related acquired dysfunction in the connectively between VIP and kisspeptin neurotransmitter pathways located in the SCN and the AVPV.

RFRP-3 and reproductive age

RFRP-3 cells in the DMN provide afferent inhibitory input to a subset of GnRH neurons that express RFRP-3 receptors (25, 61–63, 84, 85) and also project to some kisspeptin neurons (86, 87). Previous data in young females suggested that reduced inhibitory input from RFRP-3 may enhance GnRH neuron activation at the time of the preovulatory LH surge (63, 88, 89). We therefore hypothesized that age-related changes in the activation of GnRH neurons during the LH surge result from sustained activation of RFRP-3 neurons and increased Rfrp expression. However, there was no difference in the levels of Rfrp mRNA per cell or the percentage of Rfrp neurons coexpressing cfos in young compared with middle-aged females. Additionally, although the SCN sends VIP projections to the DMN region (whether this is specifically to RFRP-3 cells is not yet known) (25), icv VIP infusion did not change Rfrp gene expression or the activation of Rfrp cells in young or middle-aged females. Furthermore, neither age nor icv infusion of VIP affected the distribution or number of HE or LE Rfrp neurons. These data suggest that it is unlikely that altered RFRP-3 neurotransmission contributes to age-related changes in the LH surge or mediates the effect of VIP on the steroid-induced LH surge. However, we did not determine RFRP-3 protein levels, secretion, or cognate receptor levels in this study and in theory, anyone of these components might be affected by age or extended VIP treatment. Thus, although our data do not support a role for Rfrp neuronal activation in aging- or VIP-modulated LH surges, additional experiments that investigate the impact of reproductive age and ovarian steroids on RFRP-3 signaling in the hypothalamus of females are warranted.

Conclusion

We provide novel information about the effects of VIP and reproductive senescence on the activation of GnRH neurons and neuropeptide systems that excite and inhibit GnRH neuron function. We demonstrate that reproductive age determines the effects of icv VIP on GnRH and AVPV/PeN Kiss1 neurons. Specifically, icv infusion of VIP in young females reduces activation of both GnRH and Kiss1 neurons. In contrast, VIP treatment in steroid-primed middle-aged females markedly increases GnRH neuronal activation without increasing Kiss1-cfos coexpression or significantly increasing Kiss1 mRNA content per cell, suggesting that VIP elicits the LH surge in middle-aged females through kisspeptin-and RFRP-independent pathways. A recognized limitation to this conclusion is that the absence or presence of cfos may not provide insight about the ability of a specific neuron to synthesize or secrete physiologically relevant levels of neuropeptide or neurotransmitter. Therefore, future experiments, perhaps with antagonists, are required to fully investigate the functional involvement of kisspeptin and RFRP-3 in age-related deficits in the LH surge. Nonetheless, the ability of VIP to rescue the LH surge in middle-aged females suggests that reduced endogenous VIP tone under E2-positive feedback conditions contributes to the reproductive phenotype observed in middle-aged females. However, elevated levels of VIP in the brains of young females either desensitize or inhibit GnRH neurons and Kiss1 neurons, or cells that provide excitatory input to GnRH as well as kisspeptin neurons. Although RFRP-3 imposes inhibitory effects on the activation of GnRH neurons and GnRH peptide release, it is unlikely that changes in the activity or function of Rfrp neurons trigger events that result in suboptimal or abnormal VIP signaling in the neuroendocrine axis of females. Overall, our findings imply that there are independent roles for VIP as well as kisspeptin neurotransmitter systems in female reproductive senescence.