Sexual dimorphism in cortical excitability: repetitive nitroglycerin unmasks maladaptive spreading depolarization in female mice

Abstract

Background

Cortical excitability is maintained by a finely regulated balance between excitatory and inhibitory neurotransmission. This balance is sexually dimorphic and shapes vulnerability to pathological events such as cortical spreading depolarization (CSD), the electrophysiological correlate of migraine aura and a key phenomenon in stroke, traumatic brain injury, and seizures. Although migraine and several CSD-related disorders disproportionately affect women, most preclinical studies rely on acute, male-only paradigms, limiting understanding of how biological sex shapes cortical responses to recurrent migraine-relevant stimuli. Here, we examined whether repetitive exposure to the migraine trigger nitroglycerin (NTG) unmasks sex-specific alterations in cortical excitability and CSD recurrence dynamics.

Methods

Male and female C57BL/6J mice received five systemic NTG or vehicle injections every other day to induce sustained sensitization. After the development of mechanical orofacial allodynia, in vivo electrophysiological recordings were performed to quantify CSD frequency, induction latency, and waveform characteristics. Generalized Poisson and mixed-effects models were used to assess the effects of sex, treatment, and their interaction, with model selection based on likelihood ratio tests and residual diagnostics performed using the DHARMa package.

Results

Repetitive NTG induced progressive mechanical allodynia in both sexes (χ²(2) = 110.05, p < 0.001). In contrast, CSD metrics revealed marked sexual dimorphism in the adaptive response to repeated NTG stimulation. NTG increased CSD frequency by 41% in females (IRR [95% CI] = 1.41 [1.15–1.73], p = 0.001) but not in males (sex × group: p = 0.043). NTG also reduced first CSD induction latency (χ²(1) = 4.76, p = 0.029) and attenuated the prolongation of latency across successive CSD events (χ²(1) = 4.54, p = 0.033) selectively in females indicating sustained cortical hyperexcitability. CSD amplitude and duration remained unchanged across groups.

Conclusions

These findings demonstrate that repetitive NTG exposure unmasks a female-specific failure of cortical homeostatic adaptation to recurrent noxious stimuli in the context of CSD-related readouts. This maladaptive response is consistent with the involvement of sex-dependent regulatory pathways, potentially involving hormonal or epigenetic mechanisms, as candidate determinants of the cortical threshold for depolarization; however, causal mechanistic attribution awaits direct experimental testing. Our results highlight sex as a fundamental determinant of cortical excitability and underscore the need for sex-inclusive, repetitive paradigms to better understand female-biased vulnerability and chronification in migraine.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s10194-026-02449-1.

Background

Cortical excitability is modulated by a fine-tuned equilibrium between excitatory and inhibitory neurotransmission. This homeostatic regulation is not uniform; sexual dimorphism shapes these physiological thresholds through fluctuating gonadal hormones, sex chromosomes, and epigenetic mechanisms [1]. Deciphering these sex-specific constraints is necessary to understand why certain neurological conditions exhibit significant differences in prevalence and phenotypic expression between sexes.

Migraine affects women three times more frequently than men, offering a unique clinical lens into sex-specific brain regulation [2]. Cortical spreading depolarization (CSD), the electrophysiological correlate of migraine aura, serves as a highly sensitive readout of cortical excitability and homeostatic failure [3]. While aura affects approximately 30% of migraine patients, its unpredictable and short-lasting nature has historically limited in vivo human electrophysiological and functional imaging recordings [4, 5]. Preclinical models have bridged this gap, demonstrating that CSD activates trigeminal nociceptors and induces cutaneous allodynia, providing a mechanistic link between cortical events and migraine headache [6]. Interestingly, CSD occurs in other neurological diseases, including ischemic, haemorrhagic and traumatic brain injury, seizures, vascular malformations and brain tumours [7], highlighting the relevance of studying this phenomenon to better understand the sexual dimorphism in cortical excitability. Notably, sex differences are key determinants of CSD susceptibility in rodents. Female mice exhibit lower induction thresholds and increased susceptibility compared to males, an effect modulated by hormonal fluctuations across the estrous cycle and experimental hormonal manipulations [8–14]. Hence, it could be argued that the female brain is more susceptible to cortical excitability.

Nitroglycerin (NTG), a gold-standard migraine trigger in both clinical and preclinical settings [15, 16], reliably induces migraine-like attacks in migraine patients but possesses a poor efficacy for triggering aura in humans, a limitation that is also found when using other agents in migraine provoking studies [17, 18]. In rodent models, the repetitive NTG paradigm offers superior translational validity compared to the single application NTG one by mimicking the paroxysmal nature of the disease [15, 19]. The application of a repetitive NTG stimulating protocol offers an opportunity to study the ability of the brain to adapt to repetitive stimuli and the sexual dimorphism of this adaptation, in the context of CSD-related readouts. Despite its widespread use to study allodynia, CSD susceptibility within the repetitive NTG model remains critically under-explored. To date, only two studies have examined CSD following NTG exposure, both using a single NTG application model in male rodents only, finding no significant modulation of CSD susceptibility [20, 21]. Hence, the effects of repetitive NTG exposure, particularly in the female brain, remain unknown.

In this study, we sought to elucidate sex-specific differences in cortical excitability within the repetitive NTG model. By quantifying CSD dynamics, including induction latency, wave morphology, and event frequency, we demonstrate how repetitive NTG exposure differentially impacts the homeostatic thresholds of the male and female cortex in mice.

Methods

Animals

All experiments were conducted in accordance with the Vall d’Hebron Institute of Research Animal Ethics Committee, were approved by the competent regional institution (Project #11777) and are reported in agreement with the “Animal Research: Reporting of in vivo Experiments (ARRIVE) guidelines for animal research” [22].

C57BL/6J 9-weeks old female and male mice (n = 32, Charles River) were housed under standard light and climate-controlled conditions with standard chow and water available ad libitum. Mice were randomly assigned to experimental groups (n = 8 / group / sex) and sample sizes were calculated based on previous in-house studies using G*Power [23] based on two-way ANOVA approximation with an expected large effect size of Cohen’s f = 0.63, α = 0.050, and a target power of 0.80, yielding a minimum of 8 animals per group per sex. Behavioural analyses and electrophysiological recordings were conducted by an observer blinded to the experimental grouping. All experiments were performed between 9:00 and 15:00 to minimize potential confounding effects of the diurnal cycle.

Repetitive NTG model

Nitroglycerin (NTG) was purchased at a concentration of 5 mg/ml in ethanol (Kern Pharma). NTG was freshly diluted before each injection in physiological saline to a concentration of 1 mg/ml for a dose of 10 mg/kg. The vehicle used was physiological saline, as in-house studies showed no significant mechanical thresholds difference between physiological saline and ethanol. As previously established, the repetitive NTG model was established by intraperitoneal administration of vehicle or NTG (10 mg/kg) every other day for a total of five injections [19].

Orofacial mechanical allodynia testing

Mice were tested in low light (30–50 lx) and low noise conditions and were habituated to the testing apparatus during one hour for two consecutive days before the first baseline test and on each test day for 30 min before the first measurement (Fig. 1A). To assess orofacial mechanical allodynia, calibrated von Frey filaments (force ranging from 0.008 to 2 g) were applied to the periorbital region caudal to the eyes, starting with the 0.4 g filament (Fig. 1B). The threshold for responses to mechanical stimuli were assessed following the up-down method and four filaments were applied after the first positive response [24]. Three baseline responses were recorded every three days to avoid potential sensitization from excessive orofacial stimuli, followed by two recordings on days five and ten post-NTG treatment onset (Fig. 1A). All measurements were conducted prior to NTG or vehicle administration.

Fig. 1.

Experimental design. (A) Mice (n = 8 / group / sex) were habituated to the von Frey (VF) testing apparatus on days − 8 and − 7. Baseline VF measurements were obtained on days − 6, -3 and 0. Subsequently, 10 mg/kg NTG or saline was administered on days 1, 3, 5, 7 and 9, and VF testing was conducted on days 5 and 10, prior to each injection. (B) Orofacial mechanical allodynia was assessed in the periorbital region caudal to the eyes. (C) Representation of the direct current (DC) shift recording site and the KCl application site. Image created in https://BioRender.com

Surgical preparation: CSD induction and recording

24 h after the last vehicle or NTG injection, mice were anesthetised with isoflurane (5% for induction, 0.5–1.5% for maintenance; Aerrane, Baxter) and were placed in a stereotaxic frame on a thermal pad to keep the body temperature constant at 37 °C. Blood pressure and heart rate were monitored throughout the experiments using the non-invasive system CODA Monitor (Kent Scientific). As previously described [25, 26], two craniotomies of approximately 2 × 2 mm were performed in the right hemisphere using a saline-cooled drill, one in the parietal bone for KCl induction and the other in the frontal bone for cortical steady state potential recordings (direct current shift) (Fig. 1C). For these, a glass pipette with a tip diameter of 10 μm and filled with 3 M NaCl was placed 500 μm below the cortical surface and was connected to a high-impedance headstage through an Ag/AgCl pellet. An Ag/AgCl reference electrode was placed subcutaneously on the neck for differential recording. The signal was filtered and amplified (Neurolog system, Cambridge Electronic Design) and was sent to an analogue-to-digital converter (Power 1401, Cambridge Electronic Design) connected to a computer, where the recordings were processed and saved using Spike2 v7.03 software.

Baselines were recorded for 30 min, after which a cotton ball soaked in 1 M KCl was placed on the cortex surface and CSDs were recorded for 1 h, during which KCl was refreshed every 15 min. CSD events were identified by a negative DC shift exceeding 5 mV in amplitude with a minimum duration of 30 s. The frequency, latency to first CSD, amplitude, and duration of KCl-induced CSDs were quantified for subsequent group comparisons. After the experiment, mice were euthanized by an overdose of isoflurane.

Statistical analysis

All statistical analyses were performed in R (4.5.0) using the glmmTMB package (v1.1.12). Six outcome variables were analysed: (a) mechanical allodynia thresholds (von Frey test), (b) number of CSDs, (c) latency to the first CSD after KCl induction, (d) latency of recurrent CSDs after induction, (e) CSD amplitude, and (f) CSD duration. In all cases, model-based approaches were used to evaluate the effects of sex (male, female) and experimental group (vehicle, NTG), as well as their potential interactions.

For mechanical thresholds, a mixed-effects model was fitted using the square-root-transformed von Frey score to reduce skewness and stabilize variance. Fixed effects included sex, experimental group, time (Day 0, 5, 10), and their interactions as specified in the final model, with mouse included as a random intercept to account for repeated measurements. For CSD-related outcomes, model structures were tailored to the nature of each endpoint. CSD frequency, as overdispersed count data, was analysed using a generalized Poisson model with a log link, including sex, experimental group, and their interaction; results were expressed as incidence rate ratios (IRRs). This family was selected over the negative binomial on the basis that CSD counts are bounded by the experimental protocol and the generalised Poisson provided substantially better fit (ΔAIC = 26.85; Supplementary Table S2). Latency to the first CSD was analysed using a Gaussian model on square-root-transformed values with sex, group, and their interaction. Recurrent CSD latency was modelled longitudinally including a linear CSD order, sex, group, and their interactions, with a random intercept for mouse. CSD amplitude was analysed using a mixed-effects model including sex, group, and a quadratic polynomial term for CSD order, and their interactions, with random intercepts and random slopes for CSD order by mouse. The quadratic term was adopted following sensitivity analyses showing that a linear CSD order term produced mild residual non-uniformity that was resolved by the quadratic specification. CSD duration was analysed using mixed-effects models including sex, group, a linear CSD order term, and their interactions, with random intercepts and random slopes for CSD order by mouse. Duration, as a positive and right-skewed variable, was modelled using a Gamma distribution with a log link.

For each outcome, additive and interaction model structures were compared using likelihood ratio tests based on maximum likelihood (ML) estimation, as required for valid fixed-effect comparison. Final parameter estimates for models with random effects were subsequently obtained under restricted maximum likelihood (REML) to ensure unbiased estimation of variance components. Additive models assumed parallel NTG effects across sexes, whereas interaction models allowed NTG effects to differ by sex and, for longitudinal outcomes, across CSD order. The better-fitting model was retained for inference. Following analysis of deviance (Type III Wald χ² tests when interaction terms were present and Type II tests otherwise) and model estimation, estimated marginal means (EMMs) were derived from fitted models using the emmeans package (v1.11.2-8). Post hoc contrasts were used to examine group differences within sex (and, where applicable, across time or CSD order). Where applicable, p-values from post hoc comparisons were adjusted for multiple testing using Benjamini–Hochberg false discovery rate (FDR) procedure. Simulation-based residual diagnostics were performed for all models using the DHARMa package. Full model specifications, distributional family selection justification, model diagnostics, variance components, AIC/BIC values for all competing models, and back-transformed estimated marginal means with 95% confidence intervals are provided in Supplementary Methods and Supplementary Tables S1–S8.

Results

Repetitive NTG increases orofacial mechanical allodynia

Repetitive NTG induced a progressive, time-dependent reduction in orofacial mechanical thresholds in both sexes, with no sex-dependent modulation of this effect (group × time interaction: χ²(2) = 110.05, p < 0.001; Table 1a). NTG-treated mice showed significantly lower thresholds than vehicle controls at Day 5 (β [SE] = − 0.386 [0.062], p < 0.001) and Day 10 (− 0.650 [0.062], p < 0.001) (Supplementary Table S9), with significant progressive decreases at both timepoints (adj.p < 0.001), whereas vehicle-treated mice remained stable throughout (adj.p ≥ 0.458; Fig. 2).

Table 1.

Wald χ² tests for fixed effects from generalized and linear mixed-effects models assessing behavioral and electrophysiological outcomes across sex and experimental group

Term	χ²(df)	p
(a) Mechanical allodynia thresholds (von Frey tests)
Sex	0.443 (1)	0.506
Group	0.087 (1)	0.769
Time	2.87 (2)	0.238
Sex × Time	0.241 (2)	0.886
Group × Time	110.1 (2)	< 0.001
(b) Number of CSDs
Sex	3.63 (1)	0.057
Group	1.56 (1)	0.212
Sex × Group	4.10 (1)	0.043
(c) Latency to the first CSD after induction
Sex	3.28 (1)	0.070
Group	0.342 (1)	0.558
Sex × Group	4.76 (1)	0.029
(d) Recurrent CSD latency after induction
Sex	0.143 (1)	0.706
Group	0.004 (1)	0.950
CSDorder	750.7 (1)	< 0.001
Sex × Group	0.293 (1)	0.588
Sex × CSDorder	5.37 (1)	0.020
Group × CSDorder	3.06 (1)	0.080
Sex × Group × CSDorder	4.54 (1)	0.033
(e) CSD amplitude
Sex	0.062 (1)	0.804
Group	0.429 (1)	0.513
CSDorder (quadratic term)	20.093 (2)	< 0.001
Sex × CSDorder (quadratic term)	0.210 (2)	0.901
Group × CSDorder (quadratic term)	0.061 (2)	0.970
(f) CSD duration
Sex	1.43 (1)	0.231
Group	0.728 (1)	0.393
CSDorder (linear term)	1.40 (1)	0.236
Sex × CSDorder (linear term)	3.34 (1)	0.067
Group × CSDorder (linear term)	0.066 (1)	0.798

Statistically significant variables are shown in bold

Abbreviations: df = degrees of freedom; CSD = cortical spreading depression

Fig. 2.

Mechanical threshold over time in males and female mice following chronic NTG administration. Mean (± SEM) von Frey thresholds (g) are shown for NTG-treated and vehicle control mice at baseline (Day 0), Day 5, and Day 10 (n = 8 / group / sex). Both sexes exhibited significant time-dependent decreases in mechanical threshold after NTG treatment, whereas VEH mice showed no change. Asterisks indicate significant within-group comparisons after false discovery rate (FDR) correction. Significance: * adj.p < 0.05; ** adj.p < 0.010; *** adj.p < 0.001. Abbreviations: NTG = nitroglycerin; VEH = vehicle; SEM = standard error of the mean

Repetitive NTG increases CSD frequency only in female mice

Repetitive NTG selectively increased CSD frequency in female mice, as indicated by a significant sex × group interaction (χ² (1) = 4.11, p = 0.043; Table 1b). NTG was not significantly associated with CSD frequency in males (β [SE] = 0.090 [0.073], p = 0.212), whereas NTG-treated females showed a 41% higher CSD count compared with vehicle-treated females (IRR [95% CI] = 1.41 [1.15, 1.73], p = 0.001; Fig. 3B; Supplementary Table S10).

Fig. 3.

Electrophysiological characterization of cortical spreading depolarizations (CSD) after chronic NTG exposure. Representative traces of CSDs recorded after KCl induction in vehicle and nitroglycerin-treated mice (A). Panels show the number of CSDs recorded for one hour (B), latency to the first CSD (C), CSD latencies (D), CSD amplitude (E), and CSD duration (F) across sexes and experimental groups (n = 8 / group / sex). Data are presented as mean ± SEM. Statistical analyses were performed using generalized Poisson or linear mixed-effects models, including sex and experimental group as fixed factors; random intercepts for mouse identity were included for all longitudinal outcomes, with additional random slopes for CSD order in the amplitude and duration models. Where applicable, post hoc contrasts were adjusted for multiple comparisons using the Benjamini-Hochberg false discovery rate procedure. Significance: *p < 0.05; **p < 0.010; ***p < 0.001. Abbreviations: NTG = nitroglycerin; VEH = vehicle; SEM = standard error of the mean

Repetitive NTG reduces first CSD latency only in female mice

NTG selectively reduced the latency to the first CSD in female mice, as supported by a significant sex × group interaction (χ²(1) = 4.76, p = 0.029; Table 1c). First CSD latency was unaffected by NTG in males (p = 0.562), whereas females exposed to NTG exhibited a significantly shorter latency relative to vehicle-treated females (p = 0.029; Fig. 3C; Supplementary Table S11).

Repetitive NTG reduces the progression of recurrent CSD latency only in female mice

NTG attenuated the progressive increase in inter-CSD latency selectively in females, as indicated by a significant sex × group × CSD order interaction (χ²(1)= 4.54, p = 0.033; Table 1d). Across all groups, CSD latency increased significantly with each successive event (CSD order: χ²(1) = 750.69, p < 0.001; all within-group consecutive contrasts adj.p < 0.001). In males, NTG produced only a modest attenuation of this increase. In contrast, females exposed to NTG exhibited a markedly flatter latency trajectory across successive CSDs relative to vehicle-treated females (Fig. 3D; Supplementary Table S12), indicating sustained cortical hyperexcitability.

Repetitive NTG does not change CSD’s amplitude or duration

Neither NTG nor sex influenced CSD amplitude or duration (all group and sex effects: p > 0.51; Table 1e-f). CSD amplitude showed a significant progressive decline across successive events regardless of group or sex (Quadratic term CSD order: χ²(2) = 20.09, p < 0.001; Table 1e; model coefficients in Supplementary Table S13), while duration remained stable throughout (Supplementary Table S14; Fig. 3E-F). These results confirm that NTG-induced changes in CSD dynamics are specific to susceptibility-related parameters and do not reflect a global alteration of waveform properties.

Discussion

This study provides the first evaluation of cortical spreading depolarization (CSD) susceptibility in both male and female mice using a repetitive nitroglycerin (NTG) paradigm, a model that more closely reflects the recurrent nature of migraine attacks. By examining cortical excitability under conditions of sustained sensitization, we reveal sex-specific shifts in CSD susceptibility and recurrence dynamics that were not detectable at baseline. These findings extend current understanding of how biological sex shapes cortical responses to repeated noxious stimuli and offer a broader framework for interpreting brain homeostasis across neurological disorders characterized by spreading depolarization, including migraine, ischemic and haemorrhagic stroke, traumatic brain injury, and epilepsy.

Repetitive NTG exposure induced robust mechanical allodynia in both sexes, consistent with previous chronic NTG studies. However, only females exhibited increased CSD frequency, reduced induction latency, and an increase in latency across successive events; features indicative of sustained cortical hyperexcitability. In contrast, CSD amplitude and duration were unaffected by NTG or sex, supporting their role as stable physiological descriptors rather than primary markers of susceptibility. These results demonstrate that repetitive NTG selectively modulates susceptibility-related parameters of CSD rather than globally altering waveform properties. The absence of sex differences in NTG-induced allodynia in our study contrasts with previously reported sex-stratified findings in which allodynia lasted one day longer in female than in male mice following direct epidural KCl application [27]. This apparent discrepancy is best explained by the distinct mechanisms underlying each model: CSD-induced allodynia reflects acute propagation of cortical depolarization to trigeminal nociceptive circuits, whereas NTG-induced allodynia arises from sustained sensitization through systemic nitric oxide signalling, a process that does not require CSD as an obligatory intermediate step and that appears to affect both sexes equivalently.

Our findings challenge earlier studies conducted exclusively in male rodents in the acute NTG paradigm only, which concluded that NTG does not influence CSD susceptibility [20, 21]. The present data show that such negative results likely reflect sex-restricted experimental designs rather than the absence of a biological effect. By revealing a female-specific maladaptive response to repeated NTG stimulation in the context of CSD readouts, our study underscores the translational limitations of male-only paradigms [28] and provides empirical support for sex-inclusive research frameworks such as Sex Inclusive Research Framework (SIRF [29]), Sex as a Biological Variable (SABV [30]), and Sex and Gender Equity in Research (SAGER [31]).

A key implication of our results is the emergence of a sex-dependent (mal)adaptive pattern in cortical excitability. Based on our data, we hypothesise that male brains are capable of adapting to repetitive stimuli, such as NTG injections, stressors, or environmental challenges, whereas female brains are not. This differential capacity for adaptation may render the female cortex more susceptible to CSD and, more broadly, to central sensitization and disease chronification (Fig. 4). This hypothesis aligns with clinical observations of female-biased vulnerability to chronic migraine and other disorders involving recurrent neural stress. While the mechanistic basis of this sex-dependent adaptation remains to be established, candidate pathways that warrant future experimental investigation include hormonal modulation of cortical excitability, sex-specific epigenetic regulation, and differential excitatory-inhibitory balance; none of which can be inferred from the present data alone but each of which represents a tractable target for hypothesis-driven mechanistic work.

Fig. 4.

Sexual dimorphism in brain adaptation to repetitive stimuli. Single stimulus elicit an expected cortical excitability in both sexes (represented with a single vertical arrow). Repetitive stimuli unmask a sexual dimorphic brain response where males are able to adapt to repetitive NTG maintaining an unchanged cortical excitability in terms of CSD readouts (represented with a single horizontal arrow), whereas females are not able to adapt presenting an increased cortical excitability (represented with two vertical arrows). This figure is a conceptual schematic and does not contain statistical data

The use of a repetitive NTG model is central to these observations. Prior work using single NTG injections typically assessed CSD within hours of administration, capturing early vascular effects rather than migraine-relevant sensitization [20, 21]. Clinical evidence indicates that NTG induces a delayed migraine-like phase associated with central sensitization, and our chronic paradigm more accurately reflects this state. The divergence from earlier reports likely arises from this shift from acute to sensitized conditions, which unmasks latent sex differences in cortical regulation.

We did not observe baseline sex differences in CSD susceptibility in vehicle-treated animals, a finding that warrants discussion considering prior reports. A significantly reduced CSD induction threshold has been demonstrated in naïve female C57BL/6 mice compared to males under acute, non-sensitized conditions [14]. The absence of a comparable baseline difference in our vehicle-treated animals does not contradict those findings; rather, it suggests that the female-specific vulnerability unmasked in our study emerges specifically under conditions of sustained nociceptive challenge, rather than reflecting a simple amplification of a pre-existing threshold difference [11, 14]. Estrous cycle stages were not monitored, representing a limitation; hormonal fluctuations across the estrous cycle are known modulators of cortical excitability and CSD susceptibility, and their potential influence on our findings cannot be excluded. Hormonal modulation is therefore proposed here as a candidate mechanistic hypothesis rather than an established explanation. However, co-housing conditions have been described as potential promoters of estrous synchrony among females, which may have reduced within-group hormonal variability and contributed to the consistency of the treatment-induced differences we observed [32, 33]. Future studies incorporating estrous cycle staging, ovariectomy, or hormonal replacement paradigms will be necessary to determine the extent to which cyclic hormonal variation drives the female-specific cortical response observed here. Systematic investigation of CSD susceptibility across defined estrous phases represents an important priority for future work.

Overall, these findings highlight a sex-dependent maladaptive state in cortical excitability that is consistent with, but does not establish causation for, female-biased risk of central sensitization and migraine chronification. They also position the migraine brain as a valuable model for studying sexual dimorphism in neural plasticity and excitability regulation. The present study establishes a robust phenotypic foundation upon which future mechanistic work can build, with the aim of generating and testing hypotheses about sex-specific processes relevant to therapeutic targets in migraine and related disorders. However, limitations regarding the translatability of our results must be taken into account, including the limited ability of NTG to provoke aura in humans and the lack of current evidence to extrapolate the obtained results to stroke and traumatic brain injury.

Conclusions

Repetitive nitroglycerin exposure selectively increases CSD susceptibility and alters recurrence dynamics only in female mice, revealing a female-specific failure of cortical homeostatic adaptation to repeated noxious stimuli. While both sexes developed mechanical allodynia, only females exhibited increased CSD frequency, reduced induction latency, and impaired latency progression across events, indicating a sustained state of cortical hyperexcitability. These effects were not present at baseline, demonstrating that chronic sensitization unmasks a latent sex-dependent vulnerability rather than amplifying pre-existing differences.

The study provides clear evidence that sex is a fundamental determinant of neurophysiological response, with direct implications for understanding migraine chronification and female-biased susceptibility to disorders involving spreading depolarization. By showing that male and female brains adapt differently to repeated NTG stimuli in CSD-related readouts, this work underscores the necessity of sex-inclusive and repetitive experimental designs to ensure translational relevance in headache research and neurology more broadly.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

BOLD

Blood oxygenation level-dependent

CSD

Cortical spreading depression

EMMs

Estimated marginal means

FDR

False discovery rate

IRR

Incidence rate ratios

MRI

Magnetic resonance imaging

NTG

Nitroglycerin

VF

Von Frey test

Author contributions

OG: performed the experiments; VJG: performed the statistical analysis and was a major contributor in writing the manuscript; PPR: participated in the conceptualization of the study and in the acquisition of funding; MVP: participated in the conceptualization of the study, performed the experiments, acquired the funding and wrote the manuscript. All authors read and approved the final manuscript.

Funding

This study was supported by the “Proyectos de Generación de Conocimiento” #PID2022-137287NA-I00 from the Ministerio de Ciencia e Innovación – Spain and by the Junior Leader Fellowship #LCF/BQ/PI23/11970015 from la Caixa Foundation. OG was supported by the PFIS contract #FI23/00149 and MVP by the Miguel Servet contract #CP25/00099 from the Instituto de Salud Carlos III.

Data availability

Data are available upon request from the corresponding author.

Declarations

Ethical approval

All experiments were conducted in accordance with the Vall d’Hebron Institute of Research Animal Ethics Committee were approved by the competent regional institution (#11777; Departament d’Acció Climàtica, Alimentació i Agenda Rural, Generalitat de Catalunya).

Competing interests

The authors declare no competing interests.