Defining functional puberty in female C57BL/6J mice using endocrine, cytological and morphological markers Authors: Mélanie Chester, Noalig Wyckens, Eloïse Airaud, Raphaël Corre, Claire-Hélène Petrovic, Alice Pierre, Céline Julie Guigon Published: 2026 ABSTRACT Puberty marks the transition to reproductive competence and is driven by activation of the hypothalamo/pituitary/gonadal axis, culminating in first ovulation in females. In mice, puberty onset is commonly inferred from vaginal opening and estrous cyclicity, but the precise timing of first ovulation remains unclear. Here, we aimed to define reliable parameters of functional puberty in female C57BL/6J mice by integrating external, endocrine, and morphological measures. Vaginal opening and daily vaginal cytology were combined with daily serum luteinizing hormone (LH) and follicle-stimulating hormone (FSH) measurements and ovarian histology using the Pubertal Ovarian Maturation Score (Pub-Score), which estimates ovulation based on follicular development and corpus luteum morphology. Notably, we observed substantial inter-individual variability in puberty-related events, and no link between estrous cytology and first ovulation. Pub-Score analysis indicated that first ovulation occurred between 36 and 54 days postnatal, with a mean around 44 days, with no correlation with vaginal cytology. LH surges were detected in some of the females that had ovulated and showed limited temporal correspondence with vaginal cytology or Pub-Score estimates. Together, these findings demonstrate that puberty in female mice is a gradual, asynchronous process and that external or cytological markers alone are insufficient to define functional reproductive maturity. Introduction The concept of “puberty” has long been defined as the transitional period leading to reproductive competence, as initially described by Ojeda et al.1. This process involves profound neuroendocrine modifications within the hypothalamic circuitry, particularly involving glutamate, GABA, and kisspeptin (Kiss) neurons2. The activation of the hypothalamo-pituitary–gonadal (HPG) axis is a hallmark of puberty onset. Pulsatile secretion of gonadotropin-releasing hormone (GnRH) from hypothalamic neurons stimulates pituitary release of follicle-stimulating hormone (FSH) and luteinizing hormone (LH), thereby driving gonadal steroidogenesis and the emergence of secondary sexual characteristics. In humans, puberty typically occurs around 11–12 years of age, while in mice it takes place around 40 days postnatal (dpn)3. In females, the first ovulation represents a key event of this developmental stage, occurring under the control of the preovulatory GnRH-LH surge triggered by high estradiol (E2) levels acting through a positive feedback loop4. Defining the age at which the system becomes functional is critical in understanding key mechanisms of reproductive biology, physiology and physiopathology, and in identifying perturbations. In rodents, external indicators such as vaginal opening (VO) and subsequent stages of the estrous cycle (first estrus, E, and first diestrus 2, D2) are commonly used as indicators of puberty onset5–8. However, while vaginal opening and ovulation usually coincide in rats9, it does not appear to be the case in mice, making the accurate identification of first ovulation challenging. To date, no reliable standardized method exists to precisely define the timing of first ovulation in intact mice. Moreover, vaginal cytology alone often fails to reflect the true onset of ovarian cyclicity. The objective of this study was therefore to determine whether the integrative analysis of different parameters could mark puberty onset and first ovulation in female mice. Here, we define functional puberty as the occurrence of the first ovulation, representing the earliest physiologically effective output of the HPG axis. While this definition does not imply the establishment of stable estrous cyclicity or full reproductive maturity, it provides a biologically relevant and measurable endpoint of initial reproductive competence. We used complementary approaches combining (1) VO and cytological evaluation, (2) measurement of circulating LH and FSH levels, and (3) morphological evaluation of the ovulation on ovarian histology, using the Pubertal Ovarian Maturation Score (Pub-Score) method providing a valuable histological framework to retrospectively estimate ovulation timing described by Tena-Sempere’s group9. This allowed us to integrate, within the same experimental framework, external (VO), cytological, endocrine, and histological parameters in order to evaluate their concordance in defining functional puberty. Altogether, our findings indicate that puberty onset and first ovulation in female mice cannot be reliably inferred from a single external or hormonal marker, and that commonly used criteria may lead to misinterpretation of reproductive maturity during the peripubertal window. Instead, an integrative assessment combining external, endocrine, histological, and functional reproductive readouts is required to accurately define the acquisition of ovarian cyclicity. Results Experimental overview To investigate the temporal coordination between VO, estrous cyclicity, endocrine changes, and first ovulation in C57BL/6J female mice, three complementary experimental groups were analyzed (Fig. 1A). Mice were monitored daily for VO and estrous cyclicity, and ovaries were collected for Pub-Score analysis (Fig. 1B). Animals were euthanized either at first estrus (group 1), at first diestrus 2 (group 2), or after longitudinal blood sampling performed following VO to assess gonadotropin dynamics (group 3). Detailed experimental procedures are described in the Methods section.Fig. 1Experimental design used to characterize cues of puberty in C57BL/6J female mice. (A) Schematic representation of the experimental approaches used in this study showing the three groups of mice. All female C57BL/6J mice were monitored daily for VO and estrous cyclicity by vaginal cytology, until first estrus (1st E; n = 10 mice) and first diestrus 2 (1st D2; n = 10). In group 3, daily blood samples were collected at a fixed nocturnal time point to detect preovulatory gonadotropin surges (n = 26 mice). Created in BioRender. Guigon, C. (2026) https://BioRender.com/oegb15i. (B) Representative hematoxylin/eosin-stained ovarian sections from peripubertal mice with growing follicles at different stages of maturation and possibly corpora lutea indicating that ovulation had occurred. Sections show growing follicles at different stages of maturation and, in some cases, corpora lutea, indicating that ovulation had occurred. Ovarian histology was used to assign Pub-Scores based on established morphological criteria, as described in Gaytan et al.9. Pub-Scores were classified as negative or positive according to the absence or presence of corpora lutea, respectively, and based on antral follicle size. Shown are ovaries with negative Pub-Scores, lacking corpora lutea and displaying antral follicles of 351–400 µm (Pub-Score: − 2) or > 400 µm (Pub-Score: − 1). Ovaries with positive Pub-Scores exhibit corpora lutea with fully luteinized granulosa cells and antral follicles of 350–400 µm (Pub-Score: + 3), or corpora lutea containing apoptotic cells with large antral follicles (> 400 µm; Pub-Score: + 4). Right panels show higher-magnification views of the boxed region. Scale bars: 200 µm (left panels) and 50 µm (right panels). Experimental design used to characterize cues of puberty in C57BL/6J female mice. (A) Schematic representation of the experimental approaches used in this study showing the three groups of mice. All female C57BL/6J mice were monitored daily for VO and estrous cyclicity by vaginal cytology, until first estrus (1st E; n = 10 mice) and first diestrus 2 (1st D2; n = 10). In group 3, daily blood samples were collected at a fixed nocturnal time point to detect preovulatory gonadotropin surges (n = 26 mice). Created in BioRender. Guigon, C. (2026) https://BioRender.com/oegb15i. (B) Representative hematoxylin/eosin-stained ovarian sections from peripubertal mice with growing follicles at different stages of maturation and possibly corpora lutea indicating that ovulation had occurred. Sections show growing follicles at different stages of maturation and, in some cases, corpora lutea, indicating that ovulation had occurred. Ovarian histology was used to assign Pub-Scores based on established morphological criteria, as described in Gaytan et al.9. Pub-Scores were classified as negative or positive according to the absence or presence of corpora lutea, respectively, and based on antral follicle size. Shown are ovaries with negative Pub-Scores, lacking corpora lutea and displaying antral follicles of 351–400 µm (Pub-Score: − 2) or > 400 µm (Pub-Score: − 1). Ovaries with positive Pub-Scores exhibit corpora lutea with fully luteinized granulosa cells and antral follicles of 350–400 µm (Pub-Score: + 3), or corpora lutea containing apoptotic cells with large antral follicles (> 400 µm; Pub-Score: + 4). Right panels show higher-magnification views of the boxed region. Scale bars: 200 µm (left panels) and 50 µm (right panels). Marked inter-individual variability in estrous cycle patterns and timing of first ovulation Across the three experimental groups, VO was first observed between 27 and 38 dpn (mean ± SEM: 30.81 ± 0.28 dpn; Fig. 2A). Subsequent estrous stages, identified by daily vaginal cytology, displayed pronounced inter-individual variability. The first estrus occurred between 29 and 48 dpn (36.9 ± 0.7 dpn), while the first diestrus 2 following first estrus occurred significantly later and was observed between 34 and 50 dpn (42.95 ± 1 dpn; Fig. 2A). Consistent with this variability, the interval between vaginal opening and first estrus ranged from 0 to 17 days (6.5 ± 0.7 days), and the interval to first diestrus 2 ranged from 5 to 19 days (11.75 ± 0.98 days; Fig. 2B).Fig. 2Timing of vaginal opening, estrous cycle progression, and first ovulation in C57BL/6J female mice. (A) Age at vaginal opening (VO), first estrus (1st E), first diestrus 2 (1st D2) and first ovulation. Data of each studied parameter was obtained from mice of groups 1–3 (n = 46 analyzed mice). First ovulation was estimated by the Pub-Score method. (B) Interval (in days) between vaginal opening and first estrus, and between vaginal opening and first diestrus 2 or first ovulation calculated for each female from groups 1–3 (n = 46 analyzed mice). First ovulation was estimated by Pub-Score analysis. (C) Percentage of mice having already ovulated on the day of first estrus or first diestrus 2 (n = 10 mice/estrous stage). Data illustrate marked inter-individual variability in the timing of pubertal progression and first ovulation. Bar plots represent mean values ± SEM, with individual animals shown as dots. Data were analyzed by one-way ANOVA, with ***P < 0.001 and ****P < 0.0001. (D) Body weights at the time of vaginal opening (VO) (groups 1–3; n = 46 mice), day of first estrus (group 1, n = 10 mice) or first diestrus 2 (group 2, n = 10 mice), and in mice having ovulated at the time of dissection (groups 1–3, n = 19 mice). Timing of vaginal opening, estrous cycle progression, and first ovulation in C57BL/6J female mice. (A) Age at vaginal opening (VO), first estrus (1st E), first diestrus 2 (1st D2) and first ovulation. Data of each studied parameter was obtained from mice of groups 1–3 (n = 46 analyzed mice). First ovulation was estimated by the Pub-Score method. (B) Interval (in days) between vaginal opening and first estrus, and between vaginal opening and first diestrus 2 or first ovulation calculated for each female from groups 1–3 (n = 46 analyzed mice). First ovulation was estimated by Pub-Score analysis. (C) Percentage of mice having already ovulated on the day of first estrus or first diestrus 2 (n = 10 mice/estrous stage). Data illustrate marked inter-individual variability in the timing of pubertal progression and first ovulation. Bar plots represent mean values ± SEM, with individual animals shown as dots. Data were analyzed by one-way ANOVA, with ***P < 0.001 and ****P < 0.0001. (D) Body weights at the time of vaginal opening (VO) (groups 1–3; n = 46 mice), day of first estrus (group 1, n = 10 mice) or first diestrus 2 (group 2, n = 10 mice), and in mice having ovulated at the time of dissection (groups 1–3, n = 19 mice). These results indicate that although VO represents an early marker of pubertal activation, it precedes the establishment of a complete estrous cycle by approximately two weeks, with substantial variability among individuals. Ovarian histology revealed that only 1 out of 10 mice (10%) studied on the day of first estrus (group 1) displayed corpora lutea, indicating that ovulation had occurred in a very small fraction of mice at this stage (Fig. 2C). In contrast, 8 out of 10 mice (80%) euthanized on the day of first diestrus 2 (group 2) had ovulated, suggesting that most females achieved functional puberty during this interval. When we examined the estrus pattern of all mice having reached diestrus 2, we found that the mean delay between first estrus and first diestrus 2 was 4.8 days ± 0.8, ranging from 1 day up to 10 days. To further characterize ovarian maturation, ovaries from all mice (n = 46) were analyzed using the Pub-Score method, which integrates corpus luteum morphology, and follicular size. This analysis classified ovaries as Pub-Score positive (presence of corpora lutea) or negative (absence of corpora lutea). Overall, 22 out of 46 of examined mice (47.8%) exhibited histological evidence of ovulation. Based on Pub-Score estimates, their first ovulation may have occurred between 36 and 51 dpn with a median age around 43.5 dpn (mean ± SEM: 44.7 ± 0.7 dpn). In mice that had not yet ovulated at the time of sacrifice, retrospective estimation predicted first ovulation between 36 and 54 dpn with a median age of approximately 43 dpn (mean ± SEM: 44.0 ± 1.2 dpn). When all animals were considered together, first ovulation was confined to a developmental window spanning 36–54 dpn with a median age around 44.5 dpn, corresponding to 5–21 days after vaginal opening (mean ± SEM: 44.4 ± 0.7 dpn; 10.07 ± 0.75 days post-VO; Fig. 2A). These results place the functional onset of puberty in C57BL/6J females at approximately 44 dpn, with marked inter-individual variability. Given the temporal resolution of the Pub-Score method of 1–2 days, these values should be interpreted as approximate estimates rather than exact time points. Body weight was also analyzed at the different pubertal landmarks. It progressively increased between vaginal opening, first estrus, first diestrus 2, and first ovulation (Fig. 2D), although substantial inter-individual variability was observed at each stage. Relationship between the profile of estrous cycle and first ovulation To determine whether first ovulation coincided with a specific pattern of estrous cycle, vaginal cytology data were integrated with Pub-Score analysis. Among mice that had ovulated, no strict association was observed between the timing of first ovulation and a particular stage or pattern of the estrous cycle (Fig. 3A). Most mice exhibited irregular estrous cycles, lacking the canonical succession of PE, E, D1, and D2 typically observed in adult females. Pub-Score analysis indicated that first ovulation could occur at any stage of the estrous cycle. Nevertheless, first ovulation most frequently coincided with estrus, as determined by vaginal cytology (16 of 22 mice; 72.7%), whereas 18.2% and 9% of ovulations were predicted to occur during PE and D1, respectively.Fig. 3Estrous cycles are highly variable after vaginal opening in C57BL/6J female mice and do not show a predictive pattern of ovulation. (A) Vaginal smears patterns of proestrus (P), estrus (E), diestrus 1 (D1 and diestrus 2 (D2) in four female mice from vaginal opening to the first D2, with time of ovulation predicted by the Pub-Score and shown by an * on the graph. (B) Duration of estrus episodes before first ovulation, expressed as consecutive days of estrus during the first and second estrus events. Each dot represents one animal (n = 6–10 mice/group); horizontal bars indicate median. (C) Shown is the percentage of mice exhibiting either a regular or irregular estrous cycle pattern during the four days or more of vaginal cytology monitoring following first ovulation (n = 19 mice). Estrous cycles are highly variable after vaginal opening in C57BL/6J female mice and do not show a predictive pattern of ovulation. (A) Vaginal smears patterns of proestrus (P), estrus (E), diestrus 1 (D1 and diestrus 2 (D2) in four female mice from vaginal opening to the first D2, with time of ovulation predicted by the Pub-Score and shown by an * on the graph. (B) Duration of estrus episodes before first ovulation, expressed as consecutive days of estrus during the first and second estrus events. Each dot represents one animal (n = 6–10 mice/group); horizontal bars indicate median. (C) Shown is the percentage of mice exhibiting either a regular or irregular estrous cycle pattern during the four days or more of vaginal cytology monitoring following first ovulation (n = 19 mice). Quantitative analysis of estrous profiles further highlighted the marked irregularity of early cyclicity. First ovulation most often occurred several days after the first estrus (10 of 22 mice; 45.5%) or after a second estrus (6 of 22 mice; 27.2%). In addition, estrus phases were frequently prolonged, with multiple consecutive days of estrus observed prior to ovulation. At first estrus, 18.2% of mice displayed short estrus episodes (1–2 days), whereas 27.2% exhibited prolonged estrus lasting 3 to 9 consecutive days. A similar pattern was observed at second estrus, with a subset of animals showing extended estrus phases (up to 9 consecutive days). To further assess whether first ovulation was associated with the establishment of regular estrous cyclicity, we analyzed vaginal cytology in mice for which at least four days of post-ovulatory monitoring were available. Among these animals, 8 out of 14 (57%) exhibited a regular succession of estrous phases following first ovulation, whereas the remaining mice continued to display irregular patterns as shown for instance by the occurrence of 3 to 4 consecutive days in estrus (Fig. 3C). These findings indicate that the establishment of stable cyclicity is not immediate after first ovulation and further support the notion that reproductive maturation is a progressive process. Integration of endocrine, cytological, and morphological parameters To assess endocrine dynamics following vaginal opening, daily blood samples were collected in a separate cohort of mice undergoing longitudinal endocrine profiling (see Methods; Fig. 1A, right panel), at time points corresponding to the expected preovulatory LH surge, approximately one hour after lights off. Gonadotropin concentrations were measured using Luminex assays. Sampling began four days after VO and continued for 12 consecutive days. Basal LH concentrations remained below 1 ng/mL and fluctuated over time, consistent with the pulsatile release mode of this hormone (Fig. 4). Among the eight mice that ovulated during the sampling period, a distinct LH surge was detected in five individuals. One mouse (female 93) even exhibited two LH surges (Fig. 4). LH peaks occurred as early as four days after vaginal opening in two mice (females 93 and 104) and between 12 and 15 days post-VO in the remaining animals (females 91, 93, 94, and 105). FSH profiles showed substantial variability, with no consistent increase coinciding with LH surges. This may reflect the biphasic nature of FSH secretion around ovulation, with surges occurring outside the nocturnal sampling window. When gonadotropin profiles were combined with Pub-Score analysis for each individual, no clear temporal correspondence between LH surge detection and predicted ovulation timing was observed, except in one mouse (female 105; Fig. 4). Overall, first ovulation occurred within a 3 days range following the LH peak between 12- and 15-days post-VO, for the 5 females with a distinct LH surge, with the date of the first ovulation predicted 14–16 days post-VO for all the eight females. Integration with vaginal cytology indicated that LH surges predominantly occurred during estrus (Fig. 4). Finally, to evaluate the potential impact of repeated handling and blood sampling on pubertal timing, the estimated age of first ovulation was compared between mice subjected to blood sampling and those that were not. There was no significant effect of blood sampling when we considered age at first ovulation, but a small delay when considering the number of days after VO (Fig. 5), suggesting that blood sampling could also interfere with puberty onset.Fig. 4Serum LH and FSH profiles in peripubertal mice following vaginal opening. Daily serum LH and FSH levels were measured from 4 to 15 days after VO in eight female mice that ovulated during this period. Hormone profiles are shown together with the corresponding estrous stage and the predicted day of first ovulation, as determined using the Pub-Score method9.Fig. 5Effect of daily blood sampling on the timing of first ovulation. Left graph shows predicted age of first ovulation, and right graph the number of days between VO and first ovulation in mice subjected to daily blood sampling (+) or not (−; no blood sampling). Data are presented as bar plots of means ± SEM, and each dot represents individual values (n = 9–17 mice). Statistical significance was assessed using Student’s t-test with *, P < 0.05 and ns, not significant. Serum LH and FSH profiles in peripubertal mice following vaginal opening. Daily serum LH and FSH levels were measured from 4 to 15 days after VO in eight female mice that ovulated during this period. Hormone profiles are shown together with the corresponding estrous stage and the predicted day of first ovulation, as determined using the Pub-Score method9. Effect of daily blood sampling on the timing of first ovulation. Left graph shows predicted age of first ovulation, and right graph the number of days between VO and first ovulation in mice subjected to daily blood sampling (+) or not (−; no blood sampling). Data are presented as bar plots of means ± SEM, and each dot represents individual values (n = 9–17 mice). Statistical significance was assessed using Student’s t-test with *, P < 0.05 and ns, not significant. Discussion Accurately defining the timing of puberty is key to understand the mechanisms regulating reproductive maturation and their susceptibility to genetic, metabolic, or environmental perturbations. In this study, we sought to identify parameters defining functional puberty in C57BL/6J female mice by integrating external, cytological, endocrine, and morphological readouts. By integrating external, cytological, endocrine, and histological parameters within the same experimental framework, our study extends previous work based solely on ovarian morphology and provides a comprehensive evaluation of the concordance between commonly used pubertal markers. Our data demonstrate pronounced inter-individual variability across all pubertal landmarks examined, including the timing of VO, first E, and first D2. Histological analysis suggests that first ovulation occurs on average around 44 dpn, within a surprisingly broad developmental window spanning approximately 36–54 dpn. The Pub-Score method developed by Gaytán et al. reported a relatively narrow developmental window for this event. While the mean timing observed in our study is consistent with these previous findings, we identified a broader range of ages at first ovulation, indicating substantial inter-individual variability. This apparent discrepancy may be explained by differences in experimental design, including longitudinal monitoring, integration of multiple pubertal markers, and environmental factors such as housing conditions, handling, and early-life influences. Our results therefore extend previous work by demonstrating that, even within an inbred strain, pubertal maturation is more variable than previously appreciated. In parallel, detection of preovulatory LH surges was achieved in a subset of animals and showed limited temporal correspondence with Pub-Score-based estimates of ovulation. Together, these findings indicate that puberty in female mice is not a discrete event but rather a prolonged and asynchronous developmental process during which external, cytological, endocrine, and ovarian markers are uncoupled (Fig. 6).Fig. 6Integrated markers of puberty onset and first ovulation in female C57BL/6J mice. Schematic summary illustrating inter-individual variability in puberty onset and the timing of first ovulation in female mice. Each horizontal line represents an individual mouse positioned along the postnatal time axis (days postnatal). Colored symbols indicate the different pubertal events assessed in this study: vaginal opening (VO), first estrus and diestrus stages identified by vaginal cytology, detection of preovulatory luteinizing hormone (LH) surges, and morphological criteria of ovulation determined by Pub-Score analysis (presence of preovulatory follicles and/or corpora lutea). The figure highlights the temporal dissociation between vaginal opening, commonly used as an external marker of puberty, and the occurrence of the first ovulation confirmed by morphological and endocrine criteria. While vaginal opening and initiation of estrous cyclicity display marked inter-individual variability, the timing of first ovulation is confined to a narrower developmental window. LH surge (when detected) occurs transiently and inconsistently relative to vaginal cytology and first ovulation. Created in BioRender. Guigon, C. (2026) https://BioRender.com/oegb15i. Integrated markers of puberty onset and first ovulation in female C57BL/6J mice. Schematic summary illustrating inter-individual variability in puberty onset and the timing of first ovulation in female mice. Each horizontal line represents an individual mouse positioned along the postnatal time axis (days postnatal). Colored symbols indicate the different pubertal events assessed in this study: vaginal opening (VO), first estrus and diestrus stages identified by vaginal cytology, detection of preovulatory luteinizing hormone (LH) surges, and morphological criteria of ovulation determined by Pub-Score analysis (presence of preovulatory follicles and/or corpora lutea). The figure highlights the temporal dissociation between vaginal opening, commonly used as an external marker of puberty, and the occurrence of the first ovulation confirmed by morphological and endocrine criteria. While vaginal opening and initiation of estrous cyclicity display marked inter-individual variability, the timing of first ovulation is confined to a narrower developmental window. LH surge (when detected) occurs transiently and inconsistently relative to vaginal cytology and first ovulation. Created in BioRender. Guigon, C. (2026) https://BioRender.com/oegb15i. The marked variability observed between VO, estrous cyclicity, and first ovulation highlights the limitations of relying on external or cytological markers to define functional puberty. Together, these findings indicate that vaginal cytology alone is not a reliable predictor of first ovulation in peripubertal mice. Unlike in rats, in which VO and ovulation occur in close temporal proximity9, our data confirm that in mice VO precedes ovulation by approximately 10–15 days on average, with substantial inter-individual variation. Although the majority of mice had ovulated by the time of the first D2, only a minority had ovulated on the day of first E, indicating that estrous cytology alone does not reliably reflect ovarian competence during the peripubertal period. Consistent with this, Pub-Score analysis revealed no strict association between estrous stage and ovulation, with first ovulation most frequently occurring after several consecutive days of E (first or second estrus) but also detected at proestrus or diestrus I. This dissociation likely reflects the immaturity and instability of hypothalamo-pituitary-ovarian axis coordination during puberty, when estrus-like cytological patterns can emerge before the establishment of fully functional ovulatory cycles. Consistent with this, analysis of post-ovulatory estrous patterns revealed that about half of the mice displayed a regular sequence of estrous stage within the days following first ovulation, whereas the remaining animals continued to exhibit irregular cyclicity. These findings indicate that the establishment of stable reproductive function occurs progressively after the initial ovulatory event. Among the approaches tested, the Pub-Score method emerged as the most reliable indicator of first ovulation. By integrating corpus luteum morphology with follicular development, this method allows retrospective estimation of ovulation timing independently of transient endocrine or cytological fluctuations. The mean age of first ovulation identified in our study supports the notion that functional ovarian activation is substantially delayed relative to VO9. Notably, nearly half of the mice analyzed had not ovulated at the time of ovary collection despite exhibiting vaginal opening and, in some cases, repeated estrus-like cytology, underscoring the risk of misclassifying pubertal status when relying solely on these criteria. It would be of interest to determine whether the altered timing of puberty onset reported in numerous mouse models10–13 is systematically associated with alterations in first ovulation, and thus with the functional onset of puberty. The study of LH dosage on the evening of ovulatory discharge has already been carried out in adult females with tail sampling and appears to be a good method of analysis14–16. The limited detection of LH surges in our study emphasizes the technical and biological challenges of capturing transient endocrine events during pubertal development. The brief duration of the LH surge15,17, once-daily sampling, and the potential inhibitory effects of repeated handling with stress induction18 likely contributed to the modest detection rate. Moreover, the imperfect alignment between LH surges and Pub-Score-predicted ovulation suggests that endocrine dynamics during puberty may not fully conform to the canonical adult estrous cycle on which Pub-Score staging is based. Nevertheless, the observation that LH surges were detected exclusively in animals with histological evidence of ovulation supports the conclusion that Pub-Score positivity reflects physiologically meaningful activation of the hypothalamic-pituitary-ovarian axis. Our findings are consistent with previous studies demonstrating that VO and first E precede functional ovarian activation in mice by a variable interval. The Pub-Score method, originally described by Gaytán et al.9, established that neither VO nor estrous cytology can reliably define the timing of first ovulation. By integrating cytological, endocrine, and morphological readouts within the same experimental framework, our study extends these observations and highlights the magnitude of inter-individual variability even within a single inbred strain. Several limitations of our study should be acknowledged. Detection of LH surges was constrained by sampling frequency and potential stress-related suppression of gonadotropin secretion. Pub-Score analysis, while robust, is inherently cross-sectional and retrospective, preventing precise longitudinal tracking of ovulation in individual animals. Importantly, the occurrence of first ovulation does not imply the immediate establishment of regular estrous cycles or fertility. Rather, it represents an early functional milestone, preceding the progressive stabilization of reproductive cyclicity. Future studies will be required to determine how rapidly fully mature ovarian function is achieved following this initial ovulatory event. In addition, this study was limited to the C57BL/6J strain, and strain-specific differences in pubertal timing and ovarian physiology may limit generalization. The substantial inter-individual variability observed in this study, despite the use of an inbred strain, likely reflects the contribution of non-genetic factors such as in utero conditions, litter size, early postnatal nutrition, maternal care, and social interactions within the cage. In addition, puberty represents a highly dynamic and sensitive developmental window that may be influenced by stochastic neuroendocrine fluctuations. Taken together, our results underscore the need for integrative approaches when defining puberty in female mice. Increased sampling frequency, development of minimally invasive longitudinal methods, and incorporation of additional biomarkers of ovarian maturation may further refine the temporal resolution of pubertal staging. From a methodological standpoint, the use of Pub-Score analysis as a reference endpoint is strongly recommended when accurate identification of first ovulation is required. Importantly, experimental designs should explicitly recognize that vaginal opening and estrous cytology do not reliably predict ovulation and should avoid equating these markers with reproductive maturity. In conclusion, this study highlights the complexity and variability inherent in pubertal maturation in C57BL/6J female mice. External and cytological markers alone are insufficient to precisely determine the timing of first ovulation. Among the approaches tested, histological assessment using the Pub-Score method provides the most reliable estimate of functional puberty. These findings have important implications for experimental designs in reproductive and developmental biology and emphasize the necessity of rigorous, multi-parameter strategies when studying the onset of reproductive competence in mice. Methods Animals and treatments All experiments were conducted on C57BL/6J female mice bred in our animal facility from parents obtained from Janvier Labs (Le Genest St Isle, France). The day of birth was designated as postnatal day 0 (dpn 0), and weaning occurred at 21 dpn. Animal care and handling followed established procedures19. Mice were housed in groups of 3–5 per cage under a 12 h light/dark cycle, at 22 ± 1 °C, with ad libitum access to standard chow and water. From 10 dpn, female mice were manipulated daily to limit the stress related to vaginal smears and tail-blood collection. Animals were randomly assigned to experimental cohorts corresponding to the two complementary approaches described above (Fig. 1A). All sample collection and histological analyses were performed by investigators blinded to the experimental group. After VO, blood samples (20 µL) were collected daily for 12 consecutive days in one group of mice (n = 17), about one hour after lights off (8:30 PM), via tail sampling using capillaries. Prior to terminal cardiac puncture, mice were weighted and then anesthetized with ketamine (Imalgene® 1000; 100 mg/kg BW) and xylazine (Rompun® 2%; 10 mg/kg BW) injected intra-peritoneally with a 1 mL sterile syringe mounted with a 26-gauge needle (Terumo). Euthanasia was performed by cervical dislocation. At the time of death, body weights were as follows (mean ± SEM): Group 1 (1st E), 16.2 ± 0.3 g; group 2 (1st D2), 17.1 ± 0.3 g; and group 3,16.6 ± 0.16 g. Serum was obtained after clotting for 15 min at room temperature and centrifugation at 5000 g for 5 min. Ovaries were examined post-mortem at the time of dissection to record the presence of corpora lutea and give a first assessment of the occurrence of ovulation, and one ovary/female (left side) was fixed in Bouin’s solution (HT10132, Sigma-Aldrich) for histological processing. All procedures were approved by the Institutional Animal Care and Use Committee of Université Paris-Cité and the French Ministry of Agriculture (CEB#13-2021) and were performed in accordance with relevant guidelines and regulations. Experimental design In all groups (Fig. 1A), vaginal opening (VO) was monitored daily starting at 26 dpn, an age at which C57BL/6J females are still sexually immature. Vaginal cytology was assessed daily, and ovaries were collected at the end of the experiment for histological dating of ovulation using the Pub-Score method9. This semi-quantitative histological method is based on ovarian morphology, integrating antral follicle development in pre-ovulatory animals and corpora lutea aging in post-ovulatory animals (Fig. 1B). Negative scores indicate progressive pre-ovulatory follicular maturation and number of days before ovulation, score 0 corresponds to the first ovulation, and positive scores reflect the number of days since first or second ovulation. The age at first ovulation was calculated from the Pub-Score and the age at sample collection. Experiments were terminated at defined developmental milestones to capture key transitions: mice were euthanized either at the first estrus (group 1, 1st E; n = 10), at the first diestrus 2 following the first estrus (group 2, 1st D2; n = 10), or at variable time points between 9 and 16 days after vaginal opening (VO) following daily blood sampling to detect preovulatory gonadotropin surges (group 3; n = 26) (Fig. 1A). This design enabled an integrated analysis of vaginal cytology, endocrine dynamics, and ovarian maturation across the peripubertal period. Determination of serum LH and FSH levels Serum LH and FSH concentrations were determined using the Milliplex MAP rat pituitary magnetic bead panel on 10 µL of sample (Merck-Millipore, Nottingham, UK), following the manufacturer’s instructions19. The assay sensitivities were 32 pg/mL for FSH and 3.2 pg/mL for LH. Inter-assay coefficients of variation were 5.4% (FSH) and 3.2% (LH), and intra-assay coefficients were 7.6% (FSH) and 5.9% (LH). Determination of estrous cyclicity Vaginal smears were collected daily following VO to monitor the onset of the estrous cycle, up to first estrus, first diestrus 2 or after 9–16 days depending on the group of mice (Fig. 1). Smears were obtained using 0.9% NaCl solution, stained with hematoxylin and eosin (HE), and analyzed under light microscopy. The stages of the cycle were identified based on the predominant cell types: nucleated epithelial cells (proestrus, PE), cornified cells (estrus, E), and leukocytes (diestrus 2, D2) as previously described20,21. Histological processing and Pub-Score analysis Ovaries (n = 46) were fixed in Bouin’s solution for 48 h, rinsed in PBS, and stored in 70% ethanol. Samples were dehydrated, embedded in paraffin (HISTIM core facility, Cochin Institute, Paris), and sectioned at 5 μm. Every fifth section was mounted and stained with HE. Slides were scanned using a Pannoramic 250 Flash III system (3D HISTECH, Bicêtre Biomedical Institute). The Pub-Score method9 was used to determine ovulation timing based on CL morphology and largest follicular size. Ovaries were classified as pre-ovulatory (no CL; largest follicle used to estimate − 1 day), ovulatory (presence of newly formed CL), or post-ovulatory (regressing CL and secondary follicular growth, + 1 day). Statistical analysis Sample sizes are provided in the text. Data were analyzed using GraphPad Prism 10. Data are shown as mean ± SEM or %, and statistical analyses were performed using one-way ANOVA or Student t-test, when appropriate. Graphical design All schematic figures were created using BioRender (BioRender.com).