SIRT1 mediates obesity- and nutrient-dependent perturbation of pubertal timing by epigenetically controlling Kiss1 expression - PubMed (original) (raw)

doi: 10.1038/s41467-018-06459-9.

M J Vazquez 1 2 3 4, J M Castellano 1 2 3 4, F Ruiz-Pino 1 2 3 4, J Roa 1 2 3 4, D Beiroa 6, V Heras 1 2 3, I Velasco 1 2 3 4, C Dieguez 4 6, L Pinilla 1 2 3 4, F Gaytan 1 2, R Nogueiras 4 6, M A Bosch 7, O K Rønnekleiv 7 8, A Lomniczi 9, S R Ojeda 8, M Tena-Sempere 10 11 12 13 14

Affiliations

SIRT1 mediates obesity- and nutrient-dependent perturbation of pubertal timing by epigenetically controlling Kiss1 expression

M J Vazquez et al. Nat Commun. 2018.

Abstract

Puberty is regulated by epigenetic mechanisms and is highly sensitive to metabolic and nutritional cues. However, the epigenetic pathways mediating the effects of nutrition and obesity on pubertal timing are unknown. Here, we identify Sirtuin 1 (SIRT1), a fuel-sensing deacetylase, as a molecule that restrains female puberty via epigenetic repression of the puberty-activating gene, Kiss1. SIRT1 is expressed in hypothalamic Kiss1 neurons and suppresses Kiss1 expression. SIRT1 interacts with the Polycomb silencing complex to decrease Kiss1 promoter activity. As puberty approaches, SIRT1 is evicted from the Kiss1 promoter facilitating a repressive-to-permissive switch in chromatin landscape. Early-onset overnutrition accelerates these changes, enhances Kiss1 expression and advances puberty. In contrast, undernutrition raises SIRT1 levels, protracts Kiss1 repression and delays puberty. This delay is mimicked by central pharmacological activation of SIRT1 or SIRT1 overexpression, achieved via transgenesis or virogenetic targeting to the ARC. Our results identify SIRT1-mediated inhibition of Kiss1 as key epigenetic mechanism by which nutritional cues and obesity influence mammalian puberty.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1

Fig. 1

Changes in hypothalamic SIRT1 content and pubertal timing induced by overnutrition. a SIRT1 content in the medial basal hypothalamus (MBH)/Preoptic area (POA) of female rats at the infantile (PND10; white bars), early juvenile (PND20; gray bars), late juvenile/pre-pubertal (PND28; gray bars), and peripubertal (PND36; gray bars) stages of reproductive development, determined by western blot analysis. *p < 0.05 vs. PND28; ***p < 0.001 vs. PND10 (one-way ANOVA followed by post-hoc Tukey test). b, c SIRT1 protein and Kiss1 mRNA content in the MBH and POA at PND10 and PND36; for data in a−c, n = 6 animals per group. d, e SIRT1 content and Kiss1 mRNA levels in the MBH of late juvenile, 29-day-old female rats fed normally (NN: normal nutrition; white bars) or subjected to nutritional excess (overnutrition, ON; light red bars) during postnatal development. f Body weight of NN and ON animals. g Cumulative percent of NN and ON animals at vaginal opening (VO). h Histological score of follicular development/ovulation and representative images of ovarian maturation in NN and ON groups; CL: corpus luteum. i Uterine weight in NN and ON animals. j Plasma LH in the same animals. The bar histograms represent the mean ± SEM. In panels b-f, i and j, *p < 0.05; **p < 0.01; ***p < 0.001 (two-sided Student’s t test). For protein analyses in panels a, b and d, three representative bands per group, run in the same original western blots, are presented. The scale bar in panel h corresponds to 600 μm. Total group sizes were: NN = 11 and ON = 12; while phenotypic and hormonal parameters were assayed in the whole groups, hypothalamic protein/RNA (d, e) and ovarian histological (h) analyses were conducted in a representative subset of randomly assigned samples from each group, with the following distribution: d, e NN = 5; ON = 6−8; h n = 6 in each group

Fig. 2

Fig. 2

Changes in hypothalamic SIRT1 and pubertal timing induced by undernutrition. a, b SIRT1 content and Kiss1 mRNA levels at PND36 in the MBH of animals fed a normal diet (NN; white bars) or given restricted access to food (undernutrition, UN; light blue bars) during prepubertal development. c Body weight of NN and UN animals. d Cumulative percent of NN and UN animals showing vaginal opening (VO). e Histological score of follicular development/ovulation and representative images of ovarian maturation in the NN and UN groups; CL: corpus luteum. f Uterine weight and g Plasma LH levels. The bar histograms represent the mean ± SEM. *p < 0.05; ***p < 0.001 vs. NN group (two-sided Student’s t test). For protein analyses in panel a, three representative bands per group, run in the same original western blots, are presented. The scale bar in panel e corresponds to 600 μm. Total group sizes were: NN = 20 and ON = 10; while phenotypic and hormonal parameters were assayed in the whole groups (in the case of LH levels, for all serum samples that were available), hypothalamic protein (a) and RNA (b), as well as ovarian histological (e) analyses were conducted in a representative subset of randomly assigned samples from each group, with the following distribution: a n = 5; b n = 6−8; e n = 5−6 determinations

Fig. 3

Fig. 3

Transgenic SIRT1 excess in female mice delays pubertal timing. SIRT1 content and indices of somatic/pubertal maturation in a transgenic model of moderate Sirt1 overexpression (Tg; cyan bars) and wild-type (WT; white bars) controls measured at PND32 are shown. a Body weight. b Hypothalamic SIRT1 content. c Hypothalamic H3K9/14Ac content. d Hypothalamic Kiss1 mRNA levels. e Cumulative percent of animals showing vaginal opening (VO). f Histological score of follicular development/ovulation and representative images of ovarian maturation; CL: corpus luteum, F3: stage-3 follicle; released oocytes are denoted by arrows in the inset. g Plasma LH levels. The bar histograms represent the mean ± SEM. *p < 0.05; ***p < 0.001 (two-sided Student’s t test). For protein analyses in panels b and c, three representative bands per group, run in the same original western blots, are presented. The scale bars correspond to 600 μm in panel f, and 200 μm in the inset. Total group sizes were: WT = 10 and TG = 8; while phenotypic parameters (BW) were assayed in the whole groups, molecular and hormonal analyses were conducted in a representative subset of randomly assigned samples from each group (n = 5), which were euthanized at the time of puberty. Note that due to a technical loss in serum samples, determinations in e were n = 4 per group

Fig. 4

Fig. 4

Central pharmacological activation of SIRT1 delays pubertal timing. Indices of somatic/pubertal maturation in immature female rats following central, pharmacological SIRT1 activation by the allosteric SIRT1-activator, SA3, are shown. a Body weight of controls (C; white bars) and SA3-treated (bars in violet) rats. b Cumulative percent of C and SA3-treated rats showing vaginal opening (VO). c Histological score of follicular development/ovulation and representative images of ovarian maturation; CL: corpus luteum, F: follicle. d Uterine weight. e Serum LH levels. f Hypothalamic SIRT1 content. g Hypothalamic H3K9/14Ac content. h Kiss1 mRNA levels. The bar histograms represent the mean ± SEM. *p < 0.05; **p < 0.01 (two-sided Student’s t test). For protein analyses in panels f and g, three representative bands per group, run in the same original western blots are presented. The scale bar in panel c corresponds to 600 μm. Total group sizes were: C = 10 and SA3 = 10; while phenotypic and hormonal parameters were assayed in the whole groups (in the case of LH levels, for all serum samples that were available), hypothalamic protein (f, g) and RNA (h) analyses were conducted in a representative subset of randomly assigned samples from each group, with the following distribution: C: n = 6−8 (protein and RNA, respectively); SA3: n = 6

Fig. 5

Fig. 5

Virogenetic overexpression of SIRT1 in the ARC delays pubertal timing. a Scheme illustrating the stereotaxic delivery of AAV to the ARC. b Representative images of bilateral targeting of the ARC, denoted by fluorescence latex microspheres (beads). c Detail of neuronal targeting in the ARC by stereotaxic injection, denoted by fluorescein. d AAV-mediated infection of ARC neurons by stereotaxic delivery, denoted by GFP labeling. e SIRT1 content and acetylated K9/14 H3 vs. total H3 (AcH3/H3) ratios in the MBH of immature rats injected with AAV overexpressing SIRT1 (AAV-SIRT1) or their respective controls (AAV-C). f, g Cumulative percent of AAV-C- and AAV-SIRT1-treated rats showing vaginal opening (VO) or first estrus (FE). h Histological score of follicular development/ovulation and representative images of ovarian maturation; CL: corpus luteum, F: follicle. i Uterine and ovarian weights in AAV-C and AAV-SIRT1 groups. The bar histograms represent the mean ± SEM. *p < 0.05; **p < 0.01 (two-sided Student’s t test). For protein analyses in panel e, three representative bands per group, run in the same original western blots, are presented. The scale bars correspond to 100 μm in panels b, c, 200 μm in panel d, and 400 μm in panel h. Total group sizes were: AAV-C = 9 and AAV-SIRT1 = 13; while phenotypic parameters were assayed in the whole groups, hypothalamic protein (e) and ovarian histological (h) analyses were conducted in a subset (n = 6) of randomly assigned samples from each group. Bar graphs of the AAV-C group are in white, while bars from AAV-SIRT1 are in blue

Fig. 6

Fig. 6

Sirt1 is expressed in KNDy neurons and SIRT1 content changes with nutritional status. a Fluorescent in situ hybridization (FISH) using a cRNA complementary to Kiss1 mRNA (red) and a cRNA probe recognizing Sirt1 mRNA (green) in KNDy neurons of the ARC. Arrows point to examples of double labeled cells. Scale bars, 100 µm. b A representative gel illustrating the presence of Sirt1 and Kiss1 mRNA in ten eGFP-tagged mouse KNDy neurons (1−10) of adult ovariectomized female mice as detected by single-cell (sc) PCR. As a negative control, the PCR reaction was performed on a Kiss1-GFP cell in the absence of reverse transcription (−RT). Other negative controls (showing no PCR product) included harvested aCSF and water blank. RNA extracted from the MBH was included as a positive control (+, with RT) and negative control (−, without RT). MM molecular markers, bp base pairs. The bar graph represents the mean ± SEM of the percentage of KNDy neurons expressing Sirt1 per animal from four mice and a total of 96 harvested _Kiss1-_positive cells. c, d Immunohistofluorescence detection of SIRT1 (green) and kisspeptin (red) from the ARC of NN (normal-nutrition) vs. ON (overnutrition) rats at PND29 (panel c), or NN vs. UN (undernutrition) animals at PND36 (panel d). Bar graphs represent the mean ± SEM of the percentage of SIRT1 signal intensity in kisspeptin-positive cells from 30 to 50 neurons per animal, normalized against NN data. *p < 0.05 (two-sided Student’s t test) (n = 4 animals per group). NN = white bars; ON = light red bars; UN = light blue bars

Fig. 7

Fig. 7

Sirt1 regulates the chromatin architecture at the Kiss1 promoter. a Chromatin immunoprecipitation (ChIP) assays showing the association of SIRT1, H3K9Ac, H4K16Ac, H3K27me3 and H3K4me3 to the Kiss1 promoter in the MBH of female rats fed normally (NN) or subjected to postnatal nutritional manipulation (ON: overnutrition, UN: undernutrition). b Lack of changes in SIRT1, H3K9Ac, H4K16Ac, H3K27me3, and H3K4me3 association to intron 2 of the Kiss1 gene. The results are expressed as % of the signal generated by input DNA. The bar histograms represent the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001 (one-way ANOVA followed by the Student−Newman−Keuls). For all panels, n = 4 animals/group. NN = white bars; ON = light red bars; UN = light blue bars

Fig. 8

Fig. 8

Kiss1 expression is repressed by an SIRT1-EED silencing complex. a SIRT1 and EED repress KISS1 promoter activity as measured by gene reporter assays performed in Neuro2A cells. The bar histograms represent the mean ± SEM. *p < 0.05; **p < 0.01 (one-way ANOVA followed by the Student−Newman−Keuls; n = 4 biological replicates per group). b SIRT1 and EED coimmunoprecipitate as determined in 293T cells transfected with SIRT1-HA and EED-V5 expression vectors; reciprocal pull-down assays are shown. c Sirt1 mRNA in rat hypothalamic R22 cells infected with a control (C) lentiviral construct expressing green fluorescent protein (LV-GFP) or SIRT1. d CHIP assay showing association of SIRT1-HA to the Kiss1 promoter. e Kiss1 mRNA in C and SIRT1 overexpressing hypothalamic R22 cells. f Detection of H3K9Ac, H3K4me3 and H3K27me3 in C or SIRT1 overexpressing R22 cells. g Recruitment of EED to the Kiss1 promoter in C and SIRT1 overexpressing R22 cells. The bar histograms represent the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001 vs. C group (two-sided Student’s t test; n = 4 per group)

Fig. 9

Fig. 9

Mode of SIRT1 action in the control of puberty and its modulation by metabolic cues. a Major events occurring during normal female pubertal maturation. Transition from late juvenile (PND29; left panel) to peripubertal (PND36; right panel) stages is defined by eviction of SIRT1 and EED, a key member of the PcG silencing complex, from the Kiss1 promoter in KNDy neurons, which changes the chromatin landscape from a predominantly repressive to a permissive histone configuration. According to this model, these changes would result in enhanced Kiss1 transcription, mandatory for puberty onset. b Predicted changes in Kiss1 transcriptional activity occurring under opposite nutritional conditions. In the left panel, early-onset obesity (caused by overnutrition) induces the premature eviction of SIRT1/EED from the Kiss1 promoter, and the rearrangement of histone configuration from repressive to permissive. This change, already evident by PND29, would allow enhancement of Kiss1 transcription, leading to precocious puberty. In contrast, prepubertal under-nutrition (right panel) prevents the eviction of SIRT1/EED from the Kiss1 promoter, which maintains a repressive histone configuration still at PND36. The model predicts that this protracted repression would lead to decreased Kiss1 transcription and delayed puberty

Similar articles

Cited by

References

    1. Parent AS, et al. The timing of normal puberty and the age limits of sexual precocity: variations around the world, secular trends, and changes after migration. Endocr. Rev. 2003;24:668–693. doi: 10.1210/er.2002-0019. - DOI - PubMed
    1. Aksglaede L, Juul A, Olsen LW, Sorensen TI. Age at puberty and the emerging obesity epidemic. PLoS ONE. 2009;4:e8450. doi: 10.1371/journal.pone.0008450. - DOI - PMC - PubMed
    1. Buck Louis GM, et al. Environmental factors and puberty timing: expert panel research needs. Pediatrics. 2008;121(Suppl 3):S192–S207. doi: 10.1542/peds.1813E. - DOI - PubMed
    1. Cesario SK, Hughes LA. Precocious puberty: a comprehensive review of literature. J. Obstet. Gynecol. Neonatal Nurs. 2007;36:263–274. doi: 10.1111/j.1552-6909.2007.00145.x. - DOI - PubMed
    1. Golub MS, et al. Public health implications of altered puberty timing. Pediatrics. 2008;121(Suppl 3):S218–S230. doi: 10.1542/peds.2007-1813G. - DOI - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources