Integrated Investigation of Sex Differences in the Human Nociceptin/Orphanin FQ Receptor System: Postmortem Analyses and In Vivo [¹¹C]NOP-1A Positron Emission Tomography Imaging

Abstract

Background

The nociceptin/orphanin FQ (NOP) system modulates pain, stress, and reward, with emerging evidence for sex-dependent biology. Despite NOP’s hypothesized role in psychopathologies with pronounced sex differences, little is known about sex-associated distribution of NOP receptors.

Methods

To fill this gap, we combined postmortem molecular analyses with in vivo [¹¹C]NOP-1A PET imaging. Western blot analyses examined four NOP receptor isoforms in two postmortem cohorts, and OPRL1 mRNA expression was quantified by qPCR. Guided by these findings, [¹¹C]NOP-1A PET data from 68 healthy adults were analyzed in corticolimbic and striatal regions implicated in affective and stress-related processes. Regional total distribution volume (V_T) was estimated using Logan analysis and sex effects were assessed using mixed-effects models adjusting for age, body mass index, nicotine use, and race.

Results

Postmortem analyses highlighted higher 31.8 kDa isoform expression in men in the nucleus accumbens and amygdala, higher 39.9 kDa expression in the entorhinal cortex, and higher 42.6 kDa expression in women in the nucleus accumbens. OPRL1 mRNA expression was also higher in men, most notably in ventromedial prefrontal cortex. In vivo, PET analyses showed a significant main effect of sex, with higher V_T in men, across basal ganglia regions and a significant sex-by-region interaction in ventromedial prefrontal cortex. Post hoc model-based contrasts indicated higher [¹¹C]NOP-1A V_T in men in putamen, amygdala, and rostral anterior cingulate cortex in adjusted analyses.

Conclusions

Together, these multimodal findings highlight region-specific sex-related variation in the human NOP system across corticolimbic-striatal circuits and provide a basis for future studies of sex effects in NOP-related neuropsychiatric research.

INTRODUCTION

The nociceptin/orphanin FQ (NOP) receptor, encoded by the OPRL1 (Opioid Receptor-Like 1) gene, is a member of the opioid receptor family with distinct pharmacology (1,2). Its endogenous ligand, nociceptin/orphanin FQ, modulates a wide range of physiological processes, particularly those related to pain modulation, stress response, mood regulation, and reward processing (3–5).

Preclinical studies have highlighted robust sex-specific differences in NOP signaling. For instance, NOP receptor knockout models show sex-dependent effects on stress-related hyperalgesia and anxiety (6). NOP-mediated stress regulation has also been linked to hypothalamic-pituitary-adrenal axis modulation and coping styles (7,8). These findings align with broader evidence that men and women differ in their susceptibility and response to pain and stress-srelated affective disorders (9).

Despite preclinical evidence for sex-dependent modulation of NOP receptor signaling, direct assessments of these differences in the human brain remain scarce. This gap has largely reflected the historical absence of selective imaging tools capable of quantifying NOP receptor availability in vivo. The development of [¹¹C]NOP-1A, a high-affinity positron emission tomography (PET) radioligand, has enabled reliable characterization of NOP receptor distribution and availability in the human brain (10–13). [¹¹C]NOP-1A demonstrates high specific-to-nonspecific binding ratios, excellent test–retest reproducibility, and well-characterized kinetic behavior (11,12). Autoradiographic and in vivo findings with [¹¹C]NOP-1A indicate dense NOP receptor expression in cortical, limbic, and striatal regions (3,11,14).

Prior [¹¹C]NOP-1A studies have largely examined receptor distribution and clinical populations, including post-traumatic stress disorder, and substance use disorders, where V_T has been linked to symptom severity, craving, and relapse risk (15–18). While these findings underscore clinical relevance, putative effects of sex and other demographic factors on [¹¹C]NOP-1A binding in healthy individuals have not been systematically evaluated.

Given the widespread expression of NOP receptors in brain regions involved in stress, emotion, and reward processing (3,11), understanding how sex and demographic variables modulate NOP receptor availability is critical. This knowledge could inform the interpretation of clinical PET findings and motivate future investigation of sex as biological variable in NOP-targeted research (19–21).

To address this gap, we employed a multimodal approach combining postmortem molecular analyses and in vivo PET data. We first examined sex differences in NOP receptor isoform expression and OPRL1 mRNA levels in human brain tissue, focusing on striatal, limbic, and prefrontal regions. These molecular findings informed our region-of-interest selection for PET analyses using [¹¹C]NOP-1A in a relatively large healthy adult cohort. We hypothesized that men and women would show region-specific differences in NOP expression and binding, particularly in circuits implicated in stress and reward regulation. These findings may clarify normative NOP biology and inform future translational studies.

METHODS AND MATERIALS

Postmortem Analysis

We analyzed postmortem human brain data on NOP receptor peptide isoform expression (Cohorts 1–2) and OPRL1 transcript expression (Cohort 1). Cohort 1 included 6 men and 6 women with samples from the amygdala, cerebellum, dorsolateral prefrontal cortex (dlPFC), entorhinal cortex (ECx), insula, orbitofrontal cortex (OFC), ventromedial prefrontal cortex (vmPFC), putamen, and nucleus accumbens (NAc); in addition, NAc and putamen samples from 7 additional men donors were available. Cohort 2 comprised 7 men and 7 women with amygdala samples. These cohorts were closely matched for age and sex and excluded donors with psychiatric or neurodegenerative disorders, suicide, substance use, or neuropathological evidence of neurodegeneration, stroke, or brain cancer. Medical records and extensive family questionnaires were available for each donor and did not support regular or chronic use of neuroactive pharmacological agents. Cohorts’ characteristics are presented in Table 1A. Tissue was obtained from the Harvard Brain Tissue Resource Center (HBTRC). Brains were processed using standardized procedures under approvals from the Partners Human Research Committee, with written informed consent obtained from donors’ legal next-of-kin.

Table 1. Cohorts’ characteristics.

(A) Demographic and clinical characteristics of the postmortem brain donors included in the molecular analyses. Cohort 1 consisted of 6 men and 6 women with samples available from the amygdala, cerebellum, dorsolateral prefrontal cortex (dlPFC), entorhinal cortex (ECx), insula, orbitofrontal cortex (OFC), ventromedial prefrontal cortex (vmPFC), putamen, and nucleus accumbens (NAc). Additional putamen and NAc samples from 7 male donors were available for selected analyses. Cohort 2 consisted of 7 men and 7 women with amygdala samples. Age and postmortem interval (PMI) are presented as mean ± standard deviation (SD). Categorical variables are presented as number (percentage). (B) Demographic characteristics of the 68 healthy participants included in the PET study, stratified by sex. Data are presented as mean ± standard deviation (SD) for continuous variables and as number (percentage) for categorical variables. Groups were similar in age, BMI, and racial distribution. Nicotine use was present in approximately one-third of participants.

A. Postmortem Cohorts
Cohort	Brain region(s) available	Men, n	Women, n	Total, n	Age, mean ± SD (years)	PMI, mean ± SD (hours)	Benzodiazepine use, n (%)	Substance dependence, n (%)	Alcohol use, n (%)
Cohort 1	Amygdala, cerebellum, dlPFC, ECx, insula, OFC, vmPFC, putamen, NAc	6	6	12	62.6 ± 9.7	21 ± 3.5	0 (0%)	0 (0%)	0 (0%)
Additional donors	Putamen and NAc	7	0	7	44.3 ± 12.3	20.0 ± 6.1	0 (0%)	0 (0%)	0 (0%)
Cohort 2	Amygdala	7	7	14	64.7 ± 7.3	22.5 ± 3.2	0 (0%)	0 (0%)	0 (0%)

B. In vivo PET cohort
Group	Age (y)	BMI	Caucasian, n (%)	African American, n (%)	Asian, n (%)	Mixed, n (%)	Nicotine status, n (%)	Benzodiazepine use, n (%)	Substance dependence, n (%)	Alcohol use, n (%)
Healthy Men (n = 24)	32.4 ± 9.5	28.5 ± 5.9	18 (75.0)	4 (16.7)	1 (4.2)	1 (4.2)	10 (41.7)	0 (0)	0 (0)	0 (0)
Healthy Women (n = 44)	29.5 ± 9.5	26.1 ± 4.2	31 (70.5)	4 (9.1)	6 (13.6)	3 (6.8)	13 (29.5)	0 (0)	0 (0)	0 (0)

Abbreviations: dlPFC = dorsolateral prefrontal cortex; ECx = entorhinal cortex; OFC = orbitofrontal cortex; vmPFC = ventromedial prefrontal cortex; NAc = nucleus accumbens; PMI = postmortem interval

Western Blot analyses

Flash-frozen human brain tissue was lysed in RIPA buffer (50 mM Tris-HCl, pH 7.4; 150 mM NaCl; 1% Triton X-100; 0.5% sodium deoxycholate; 0.1% SDS; 10 mM EDTA) supplemented with protease and phosphatase inhibitor cocktails. Samples were sonicated (3× at 50% power) and centrifuged (20-min, 16,000×g) to remove debris. Supernatants were collected and protein concentration was measured using a DC assay (Bio-Rad). Equal protein (20 μg/lane) was separated on 12% NuPAGE Bis-Tris gels (MOPS, 200 V, 1h, on ice) and transferred overnight at 4°C onto 0.45 μm PVDF-FL membranes (NuPAGE transfer buffer with 10% methanol and antioxidant; constant current 50 mA). Membranes were washed, air-dried, methanol reactivated, stained with LI-COR Revert 700 Total Protein Stain, and imaged on an Odyssey CLx system. Membranes were then blocked (Odyssey Blocking Buffer) and incubated with anti-NOP/OPRL1 antibody (rabbit, Proteintech; 1:1000). NOP signal was detected using an IRDye 680–conjugated secondary antibody, while VCP (loading control) was detected using an IRDye 800–conjugated secondary antibody (LI-COR; 1:20,000). Band intensities were quantified in Empiria Studio and normalized to total protein signal, providing a global measure of sample input and transfer efficiency that is less sensitive to variation in a single control protein. VCP was also probed as an additional loading/transfer control, with stability empirically assessed in pilot studies. Inter-gel normalization used log10-transformed signals with gel-median alignment. Technical duplicates were assessed using a 95th-percentile QC cutoff derived from pooled reference samples; failed samples were rerun, and mean values from accepted duplicates were used for downstream analyses.

qPCR analyses

OPRL1 mRNA expression was quantified by quantitative real-time PCR (qPCR) in selected cortical and subcortical regions. Total RNA was extracted from ~50–100 mg snap-frozen brain tissue using TRIzol followed by chloroform phase separation and RNeasy column purification with on-column DNase treatment. RNA integrity was assessed on an Agilent 2100 Bioanalyzer, and RIN values were included as covariates; 1.2 μg RNA was reverse-transcribed using SuperScript III with oligo(dT) priming. OPRL1 transcripts were amplified using a validated SYBR Green primer pair targeting a conserved region across OPRL1 transcript variants. Real-time PCR was performed on a Bio-Rad CFX Connect system using SYBR Green chemistry with 1 μL cDNA input (57 ng RNA equivalent). Samples were run in triplicates, with serial dilutions used to verify assay performance. No reverse-transcriptase controls [RT(−)] were included for each RNA sample or representative samples from each tissue group, with no-template controls in every run. Quantification was based on cycle threshold (Ct/Cq) values. Data were normalized to stable, well-validated reference genes (RPLPO, HPRT1)

Statistical analyses

Normalized values were analyzed using two-tailed unpaired t-tests to compare men and women within each region. When normality assumptions were violated, nonparametric Mann–Whitney U tests were applied. Outliers exceeding ±2.5 standard deviations from the group mean were excluded. As secondary analysis, linear mixed-effects models were applied with sex as the main fixed effect and age, postmortem interval (PMI), substance use and pharmacological treatment (if any) as additional fixed effects, and subject ID as a random effect. For transcript analyses, between-group differences in OPRL1 mRNA expression were tested using independent t-tests.

P-values were adjusted for multiple testing using the Bonferroni correction, with statistical significance defined as p<0.05. Sex-related effects from the postmortem results were used to guide region-of-interest selection and statistical modeling for in vivo PET analyses.

PET Imaging Analysis

Participant characteristics

We analyzed PET data from 68 healthy adults (24 men, 44 women) (Table 1B), collected as part of prior imaging studies (15–17). The groups were demographically balanced for age (men: 32.4 ± 9.5 years; women: 29.5 ± 9.5 years), BMI (28.5 ± 5.9 vs. 26.1 ± 4.2), and racial composition. Nicotine use was present in approximately one-third of participants, with no differences between sexes.

Imaging Acquisition and Processing

Participants underwent 70-min dynamic scan following injection of approximately 10 mCi [¹¹C]NOP-1A on a Siemens Biograph64 mCT scanner as previously described (15–18). Data were motion-corrected by frame-to-frame rigid registration, with PET co-registered to the high-resolution T1-weighted MPRAGE. Combined transforms were applied inversely to resample the MRI into native PET space. Using FreeSurfer v7, we extracted regional time-activity curves from cortical and subcortical ROIs involved in pain and emotion processing, with cerebellar gray matter as a low-binding region.

Tissue TACs were converted to standardized uptake value (SUV) by normalizing to injected activity and body weight. Arterial blood was collected and selected plasma samples were analyzed by radio-HPLC to derive metabolite-corrected plasma input functions. Regional total volume of distribution (V_T), a measure of tracer distribution at equilibrium, was estimated with Logan graphical analysis (22) using the metabolite-corrected arterial parent-plasma input function (t* = 30-min).

Statistical Analyses

To examine sex- and region-specific patterns of [¹¹C]NOP-1A binding in relation to the postmortem findings, we performed mixed-effects models with Region as a repeated measure and Sex as a between-subject factor. ROIs were selected a priori based on the postmortem analyses and grouped into three sets: basal ganglia (NAc, caudate, putamen, pallidum), limbic regions (amygdala, hippocampus, ECx), and vmPFC (rostral anterior cingulate cortex [rACC], medial OFC [mOFC]) (Figure 1). For each ROI group, V_T was modeled with fixed effects of Sex, Region, and Sex × Region, and a random intercept for participant. Primary models adjusted for age, BMI, nicotine use, and race, with all covariates entered simultaneously; age and BMI were modeled as continuous variables, nicotine use as a dichotomous variable (yes/no), and race as a categorical factor using dummy-coded indicator variables. Unadjusted models were also estimated as sensitivity analyses. Overall effects were assessed using likelihood-ratio tests, and post hoc model-based contrasts were used to test sex differences within regions, with Bonferroni correction applied within each ROI group. Primary analyses were performed without partial-volume correction. As a sensitivity analysis, region-based voxel-wise (RBV) partial-volume correction was performed using the PETPVC toolbox with FreeSurfer-derived aparc+aseg segmentations and a scanner point spread function of 4.4 mm isotropic full-width at half maximum. Given that RBV PVC introduces additional assumptions, these analyses are presented as supportive sensitivity analyses rather than the primary results.

Figure 1. Regions of interest used for primary PET analyses.

Sagittal, coronal, and axial views illustrate the regions of interest (ROIs) included in the primary in vivo [¹¹C]NOP-1A PET analyses. ROIs were defined based on regions showing isoform expression in the postmortem analyses and were grouped into three anatomical sets: basal ganglia (nucleus accumbens, caudate, putamen, pallidum), limbic regions (amygdala, hippocampus, entorhinal cortex), and ventromedial prefrontal cortex (rostral anterior cingulate cortex and medial orbitofrontal cortex). Colors indicate individual ROIs as shown in the legend.

Exploratory ROI-level analyses examined sex differences in V_T across 15 predefined cortical and subcortical regions (Supplemental Figure 1) using two-sided Welch’s t-tests and separate multiple linear regression models including the same covariates. Bonferroni correction was applied across exploratory ROI-wise comparisons, with corrected p<0.05 considered significant. Whole-brain voxel-wise analyses were performed in SPM12 using spatially normalized V_T images smoothed with a 3 mm FWHM Gaussian kernel. Men and women were compared using a two-sample t-test, with maps thresholded at voxel-wise p<0.001 (uncorrected) and cluster-level FWE-corrected p<0.05.

RESULTS

Postmortem Findings

Across all cohorts, four NOP receptor isoforms were consistently detected (31.8, 39.9, 40.6, and 42.6 kDa) (Supplemental Figure 2). Western blots showed the highest signal for the 42.6 kDa isoform across brain regions, whereas the 31.8 kDa isoform demonstrated more region- and sex-specific variability.

In Cohort 1, NOP receptor isoform expression varied across cortical and subcortical regions, with clear regional differences and sex-related effects. The 31.8 kDa isoform showed the greatest variability, with relatively high levels in prefrontal and limbic areas, including dlPFC, vmPFC, OFC, insula, and putamen, in both sexes (Supplemental Figure 3). Relative to women, men showed higher 31.8 kDa isoform expression in the amygdala (p=0.020) and NAc (p=0.010), consistent with mesocorticolimbic involvement (Figure 2A). The 39.9 kDa isoform was more uniform across cerebellum, dlPFC, OFC, and vmPFC, with modest, non-significant sex differences (Supplemental Figure 3) but elevated expression in men in the ECx (p=0.020) (Figure 2A). The 40.6 kDa isoform showed lower overall abundance but displayed a cortical distribution similar across sexes and regions (Supplemental Figure 3). The 42.6 kDa isoform (putative mature receptor) was broadly expressed, with relatively non-significant greater levels in OFC in men (Supplemental Figure 3) and higher expression in the NAc in women (p=0.020) (Figure 2A). Samples with insufficient protein yield for Western blot quantification were excluded (amygdala: women n=1; NAc: women n=4; ECx: women n=3 and men n=4; NAc 31.8 kDa isoform: men n=2). Covariate analyses did not reveal any significant associations, indicating that none of the tested demographic or behavioral variables acted as confounders.

Figure 2: Sex Differences in NOP Receptor Isoform and Transcript Expression in Human Postmortem Brain.

(A) Additional regional analyses revealed higher 31.8 kDa NOP receptor isoform in men in the amygdala and nucleus accumbens (NAc), while women showed greater expression of the 42.6 kDa isoform in the NAc and lower 39.9 kDa expression in the entorhinal cortex (ECx). Bars represent mean relative expression ± SEM. Samples with insufficient protein yield for Western blot quantification were excluded (amygdala: women, n = 1 [16.7%]; NAc: women, n = 4 [66.7%]; ECx: women, n = 3 [50.0%] and men, n = 4 [66.7%]; NAc 31.8 kDa isoform: men, n = 2 [15.4%]. Between-sex comparisons were performed using two-tailed unpaired t-tests within each region; when normality assumptions were violated, Mann–Whitney U tests were used. (B) Quantitative PCR (qPCR) analysis of OPRL1 mRNA across cortical and subcortical regions indicated higher transcript levels in men, with a trend toward male-biased expression in the orbitofrontal cortex (OFC; p = 0.051) and significantly higher expression in the ventromedial prefrontal cortex (vmPFC; p = 0.003). For transcript analyses, between-group differences in OPRL1 mRNA expression were tested using independent two-tailed t-tests. P-values were Bonferroni-corrected for multiple comparisons, with significance set at p < 0.05. (C) Expression of the 31.8 kDa NOP receptor isoform was significantly higher in the amygdala of men donors compared to women (p = 0.0146).

OPRL1 mRNA levels showed regionally varying patterns between men and women across the examined cortical and subcortical regions (Figure 2B). Although most areas demonstrated comparable transcript abundance between sexes (cerebellum p=0.603, ECx p=0.282, insula p=0.633), several regions showed meaningful sex-related differences. Men exhibited non-significant but non-negligible higher OPRL1 expression in the NAc (p=0.082), putamen (p=0.063), amygdala (p=0.072) and dlPFC (p=0.088), reflecting a consistent male-biased trend in limbic-striatal circuitry and association cortices. The OFC displayed a near-significant elevation in men (p=0.051), while vmPFC mRNA levels were significantly higher in men (p=0.003), reinforcing protein-level findings and supporting sex-dependent regulation of NOP receptor expression across cortical and limbic circuits.

In Cohort 2 (Figure 2C), postmortem analysis of the amygdala revealed a significant Sex effect in NOP receptor isoform expression. The 31.8 kDa isoform was significantly higher in men compared to women (p=0.015, after outlier correction removing one man and one woman from the analysis).

In vivo [¹¹C]NOP-1A PET findings

ROI-Level Primary Analyses

Mixed-effects ROI analyses were performed across three region sets informed by the postmortem studies.

In the basal ganglia (NAc, caudate, putamen, pallidum), the mixed-effects model revealed a strong main effect of Region (χ²(3)=560.651, p<0.001) and a significant main effect of Sex (χ²(1)=5.104, p=0.024), with higher V_T in men across regions. The Sex × Region interaction was not significant (χ²(3)=5.195, p=0.158) (Table 2A). Post hoc contrasts showed higher V_T in men in the putamen (M−W=1.137, 95% CI [0.330, 1.945], p=0.023). Differences in the caudate were not significant after Bonferroni correction (Table 2B, Figure 3A).

Table 2. Mixed-effects ROI-level analyses of [¹¹C]NOP-1A V_T by sex adjusted for age, BMI, nicotine use, and race.

(A) Omnibus mixed-effects model results for each ROI set. Mixed-effects models were conducted separately for three anatomically defined ROI groups: basal ganglia (nucleus accumbens [NAc], caudate, putamen, pallidum), limbic regions (amygdala, hippocampus, entorhinal cortex [ECx]), and ventromedial prefrontal cortex (rostral anterior cingulate cortex [rACC], medial orbitofrontal cortex [mOFC]). Region was treated as a repeated measure within participants and sex as a between-subject factor. All models included age and body mass index (BMI) as continuous covariates, nicotine use as a dichotomous covariate, and race as categorical covariates. The table reports likelihood ratio χ² statistics and associated p values for the main effects of region and sex, as well as the sex × region interaction within each ROI set. Significant region effects were observed in all three ROI sets. A significant main effect of sex was observed in the basal ganglia ROI set, while sex effects in the limbic and vmPFC ROI sets were at trend level. A significant sex × region interaction was observed in the vmPFC ROI set only. (B) Post hoc pairwise sex contrasts derived from the mixed-effects models. For each region, mean V_T values are reported separately for women (W) and men (M), along with the estimated contrast (M − W), 95% confidence intervals (CI), uncorrected p values, and Bonferroni-corrected p values for regional comparisons within each ROI set. After Bonferroni correction, men showed significantly higher V_T in the putamen, amygdala, and rACC. Uncorrected p values additionally indicated higher V_T in men in the caudate. Significant associations (p < 0.05) are highlighted in bold.

A. Omnibus mixed-effects results
ROI set	Regions	Region effect χ2(df)	p	Sex effect χ2(df)	p	Sex × Region χ2(df)	p
Basal ganglia	NAc, caudate, putamen, pallidum	560.651 (3)	<0.001	5.104 (1)	0.024	5.195 (3)	0.158
Limbic	Amygdala, hippocampus, ECx	241.995 (2)	<0.001	3.392 (1)	0.066	3.884 (2)	0.143
vmPFC	rACC, mOFC	5.727 (1)	0.017	3.694 (1)	0.055	6.308 (1)	0.012

B. Post hoc sex contrasts derived from the mixed-effects models
ROI set	Region	Mean W	Mean M	Contrast (M − W)	95% CI	Uncorr_p	Bonferroni p
Basal ganglia	Accumbens	12.746	13.453	0.770	[−0.038, 1.577]	0.062	0.247
	Caudate	11.462	12.360	0.960	[0.153, 1.768]	0.020	0.079
	Putamen	12.884	13.958	1.137	[0.330, 1.945]	0.006	0.023
	Pallidum	8.633	9.350	0.780	[−0.028, 1.587]	0.058	0.233
Limbic	Amygdala	13.927	14.943	1.101	[0.205, 1.996]	0.016	0.048
	Hippocampus	10.706	11.357	0.736	[−0.160, 1.632]	0.107	0.322
	ECx	11.535	12.005	0.555	[−0.340, 1.451]	0.224	0.673
vmPFC	rACC	13.570	14.624	1.203	[0.244, 2.162]	0.014	0.028
	mOFC	13.503	14.011	0.658	[−0.301, 1.617]	0.179	0.358

Abbreviations: ECx, entorhinal cortex; mOFC, medial orbitofrontal cortex; NAc, nucleus accumbens; rACC, rostral anterior cingulate cortex; V_T, total distribution volume.

Figure 3. Sex Differences in in Vivo NOP Receptor Binding Across Brain Regions.

Bar graphs depict group means (± SEM) of [¹¹C]NOP-1A total distribution volume (V_T) across selected regions of interest: (A) Basal ganglia (nucleus accumbens, caudate, putamen, pallidum), (B) Limbic regions (amygdala, hippocampus, entorhinal cortex), (C) Ventromedial prefrontal cortex (vmPFC) including the rostral anterior cingulate cortex (rACC), and medial orbitofrontal cortex (mOFC). Data are plotted separately for men (n = 24, gray) and women (n = 44, teal), with individual data points overlaid. Statistical effects were evaluated using mixed-effects models conducted separately for each anatomical grouping, with region treated as a repeated measure and sex as a between-subject factor. Models were adjusted for age, body mass index (BMI), nicotine use, and race. Omnibus tests assessed main effects of region and sex and the sex × region interaction. Post hoc sex contrasts within individual regions were derived from the mixed-effects models and Bonferroni-corrected within each anatomical grouping (p < 0.05). Significant effects (p < 0.05) are highlighted by a red star.

In limbic regions (amygdala, hippocampus, ECx), there was a main effect of Region (χ²(2)=241.995, p<0.001), but no significant main effect of Sex (χ²(1)=3.392, p=0.066) or Sex × Region interaction (χ²(2)=3.884, p=0.143) (Table 2A). Given the significant postmortem findings in the amygdala, post hoc contrasts were conducted and probed for similar effects in the PET data; V_T was indeed higher in men in the amygdala (M−W=1.101, 95% CI [0.205, 1.996], p=0.048), whereas differences in the hippocampus and ECx were not significant (Table 2B, Figure 3B).

In vmPFC subregions (rACC, mOFC), there was a main effect of Region (χ²(1)=5.727, p=0.017) and a significant Sex × Region interaction (χ²(1)=6.308, p=0.012), but no main effect of Sex (χ²(1)=3.694, p=0.055) (Table 2A). Post hoc contrasts showed higher V_T in men in the rACC (M−W=1.203, 95% CI [0.244, 2.162], p=0.028), whereas the mOFC difference was not significant (Table 2B, Figure 3C).

Unadjusted mixed-effects models produced similar patterns of results across all ROI sets, including significant region effects and comparable regional patterns of sex differences (Supplemental Table 1). Sensitivity analyses using RBV partial-volume correction showed that the overall pattern of findings was preserved, with robust regional effects across all ROI sets and a significant Sex × Region interaction remaining in the vmPFC. PVC attenuated localized post hoc sex differences, which no longer survived Bonferroni correction (Supplementary Table 2).

Exploratory Analyses

To further explore potential sex-related effects beyond the a priori targets, exploratory analyses were conducted across additional cortical and limbic regions. These regions were selected based on their strong NOP receptor expression and their established roles in stress, emotion, and reward processing. Welch t-test identified higher V_T in men relative to women in three regions: pallidum (p_uncorr=0.040), putamen (p_uncorr=0.029), and rACC (p_uncorr=0.045). However, none of the findings survived Bonferroni correction for 15 ROIs comparisons (Supplemental Table 3).

Multiple linear regression results aligned with ROI-level findings, identifying sex as a significant predictor of V_T in amygdala, superior frontal, rACC, caudate, putamen, thalamus, insula, brainstem and pallidum (p_uncorr<0.05) (Supplemental Table 4). No significant associations were detected for age, BMI, race, or nicotine use in any ROI. After Bonferroni correction, no demographic variable retained statistical significance.

Whole-brain voxel-wise analyses conducted in SPM12 did not reveal any clusters showing a statistically significant main effect of sex in [¹¹C]NOP-1A V_T (voxel-wise p<0.001 uncorrected, cluster-level FWE-corrected p<0.05) (data not shown).

DISCUSSION

This study combined postmortem molecular data and in vivo PET imaging to provide a multimodal evaluation of sex differences in the human nociceptin receptor system. Across two postmortem cohorts and in vivo [¹¹C]NOP-1A PET imaging totaling 101 subjects, we found convergent evidence for sex- and region-specific distribution of the NOP receptor system in the human brain.

Protein analyses revealed four distinct NOP receptor isoforms (31.8, 39.9, 40.6, and 42.6 kDa), with the 42.6 kDa isoform showing the highest overall abundance and the 31.8 kDa isoform displaying the greatest region- and sex-specific variability. Men were characterized by overall higher expression of the lower molecular weight NOP receptor isoforms, particularly the 31.8 kDa form, across prefrontal, limbic, and striatal regions, with significantly elevated expression in the amygdala, entorhinal cortex, and nucleus accumbens. The increased presence of these isoforms is consistent with a possible increase in receptor synthesis or membrane-associated forms within mesocorticolimbic circuits. In contrast, women exhibited relatively higher expression of the 42.6 kDa isoform, corresponding to the higher-molecular-weight band consistent with a more glycosylated species (23), with the strongest signal in the nucleus accumbens. These findings should be interpreted cautiously given uncertainty regarding the functional meaning of these isoforms. The present study was not designed to determine the functional identity of these isoforms. Future biochemical and cellular studies will be necessary to establish whether these molecular-weight forms represent mature receptors, immature intermediates, alternatively spliced variants, or other post-translationally modified receptor species. Complementary OPRL1 mRNA analyses revealed higher transcript levels in men in the orbitofrontal and ventromedial prefrontal cortex, reinforcing protein-level evidence for sex-dependent regulation across mesocorticolimbic circuits. This pattern may reflect sex-dependent regulation of receptor maturation and trafficking within regions that integrate emotional, motivational, and cognitive functions and regulate stress responses, consistent with evidence that post-translational modifications and trafficking programs shape opioid receptor availability and signaling (24,25). Together, these data indicate convergent sex-related differences at both the protein and transcript levels. The postmortem analyses should also be interpreted in light of the relatively small sample sizes, further reducing statistical power. Consequently, these regional findings should be considered preliminary and warrant replication in larger independent postmortem cohorts.

In vivo PET results partially mirrored the postmortem findings. Thus, covariate-adjusted mixed-effects models revealed a significant main effect of Sex across basal ganglia regions and a significant Sex × Region interaction in ventromedial prefrontal cortex. Model-based regional contrasts indicated higher V_T in men in the putamen, amygdala, and rACC after Bonferroni correction, whereas other regional sex differences did not survive correction. Adjustment for demographic and behavioral covariates produced only modest changes in the estimated sex contrasts, suggesting that the overall pattern of findings was not primarily driven by differences in age, BMI, nicotine use, or race. Moreover, sensitivity analyses using RBV partial-volume correction yielded similar overall conclusions, although localized post hoc sex differences were attenuated. The partial spatial correspondence between postmortem and PET data is consistent with subtle sex-related variation in NOP receptor distribution, although these cross-modal patterns should be interpreted cautiously. The absence of significant voxel-wise effects is consistent with modest ROI effect sizes and likely reflects higher voxel-level noise. This suggests [¹¹C]NOP-1A variation is localized to discrete gray matter regions (basal ganglia, anterior cingulate cortex).

Together, the postmortem and in vivo PET results highlight sex-related differences in NOP receptor regulation across mesocorticolimbic circuitry, while underscoring that each modality captures a distinct biological layer. Because PET integrates binding across tracer-accessible receptor states, shifts in isoform balance may not translate linearly to V_T. OPRL1 mRNA reflects transcriptional regulation and thus indicates the potential capacity for receptor production, but it does not determine how much receptor protein is translated, which molecular-weight forms predominate, or whether receptors reach the membrane. Western blot analyses quantify receptor protein and reveal multiple molecular-weight forms. PET V_T reflects aggregate radiotracer binding to tracer-accessible receptors and does not distinguish among molecular-weight forms that retain the ligand-binding pocket. Within this framework, men and women differed not only in the magnitude of NOP receptor expression but also in the distribution of receptor molecular forms. Sex-specific shifts between molecular forms may therefore amplify or offset in vivo PET signal depending on regional balance. This provides a parsimonious explanation for why PET showed higher [¹¹C]NOP-1A V_T in men most clearly in the putamen, amygdala, and rACC in the adjusted analyses, while additional regions such as the caudate and pallidum showed similar directional trends that did not survive correction. In contrast, regions such as NAc, mOFC, and ECx showed sex effects more clearly in postmortem measures than in PET. In particular, opposing postmortem effects in NAc, with higher lower molecular-weight forms in men and higher 42.6 kDa expression in women, could yield comparable net tracer binding, effectively flattening detectable PET differences. However, this interpretation remains speculative and will require direct experimental validation to determine how individual NOP receptor isoforms contribute to the in vivo PET signal.

PET binding reflects the net pool of tracer-accessible receptors, which depends not only on total protein and molecular form but also on trafficking and membrane availability. Mechanistically, agonist-selective phosphorylation can drive NOP receptor desensitization and internalization (26), providing a plausible route from endogenous tone and ligand history to in vivo binding, in concert with broader post-translational regulation described across opioid receptor families (24,25). An additional possibility is that alternative splicing generates truncated GPCR isoforms that are functionally impaired yet can alter surface delivery of canonical receptors via intracellular retention and/or hetero-oligomerization, producing dominant-negative effects that shift the balance of tracer-accessible receptors in a region-dependent manner (27–29). In support of this notion, a truncated OPRL1 splice variant with poor ligand binding and impaired signaling has been reported (30). We cannot exclude that some in vivo–postmortem discrepancies reflect limited postmortem sample sizes in certain regions and inter-individual variability, such that a subset of findings may be driven by regional variability rather than stable sex differences.

Sexual dimorphism in NOP receptor expression is consistent with prior work demonstrating sex differences in pain, stress, and affect regulation circuits, as well as vulnerability to stress-related disorders, most prominently major depression (6,9,19,20,31). These same networks, including the basal ganglia, amygdala, and rACC, are involved in motivational salience, emotional appraisal, and pain modulation (32–34). The rACC, in particular, serves as an integrative hub linking limbic and prefrontal regions (35) rich in NOP receptors (36), which may explain its consistent sex-related difference across imaging and postmortem findings. Androgens can shape opioid receptor function by influencing expression, coupling, and trafficking (37). These hormonedriven processes could alter in vivo binding independently of total protein, and they may help explain sex-linked variability in [¹¹C]NOP-1A outcomes, alongside phosphorylation-dependent desensitization and internalization pathways at NOP (26).

The regionally specific sex effects we observed in striatal and limbic/prefrontal NOP measures align with prior molecular imaging evidence implicating this system in neuropsychiatric conditions. Altered [¹¹C]NOP-1A binding has been reported in post-traumatic stress disorder and substance use disorders (15–17), underscoring the receptor’s sensitivity to stress and environmental factors. Within this context, our results in healthy adults may provide a normative reference for interpreting disease-related deviations and evaluating sex as a biological variable in future NOP-targeted studies. Early translational studies demonstrating central NOP target engagement and preliminary efficacy of a selective NOP antagonist in major depression and alcohol dependence (38–40) further support the clinical relevance of the cortical–striatal–limbic sex differences identified here. Notably, in the NOP antagonist trial in alcohol dependence, subgroup analyses suggested a substantially stronger treatment effect in women than men (39). These clinical observations also motivate mechanistic hypotheses for the isoform findings observed here, including the possibility that sex differences in receptor state/trafficking (membrane-accessible vs internalized pools) could influence apparent ligand accessibility/affinity and may help contextualize reports of increased [¹¹C]NOP-1A binding following hydrocortisone (41). Hydrocortisone/cortisol regulation and bioavailability show sex-dependent differences (e.g., stress reactivity and corticosteroid-binding globulin), which could further contribute to differential modulation of the NOP system (42,43). Glucocorticoid signaling provides a plausible mechanistic link between stress exposure and NOP receptor regulation. Hydrocortisone activates glucocorticoid receptors, which can influence OPRL1 transcription (44,45) and, critically, GPCR trafficking and recycling (46).

Sex hormone fluctuations not captured in this cross-sectional design may contribute to within-sex variability. In particular, estradiol and progesterone can modulate stress-related receptor signaling and regulation in emotion-relevant circuits (20,47,48), which warrants future PET studies incorporating hormonal profiling or menstrual-cycle tracking.

The strengths of this study include the integration of protein, transcript, and receptor-availability measures, the use of arterial-input kinetic modeling, and rigorous region-based statistical analyses. Nonetheless, the moderate sample size in the postmortem dataset may have reduced power to detect small but biologically meaningful effects after Bonferroni correction, and the absence of systematic assessment of hormonal status, menstrual-cycle phase, and menopausal state may have contributed to within-group variability among women. Despite these constraints, convergence across modalities supports the robustness of the observed sex-related trends. The NOP system is implicated in neuropsychiatric disorders such as depression, post-traumatic stress disorder, and substance use disorders, which themselves exhibit sex-biased prevalence and treatment responses (9,20). Understanding normative sex variation in receptor density and isoform composition may help motivate future clinical studies to evaluate whether sex influences response to NOP-targeted ligands. In particular, separating surface from intracellular receptor pools and quantifying glycosylation and coupling efficiency will be essential to determine how isoform composition influences in vivo binding and receptor signaling, consistent with evidence that glycosylation, phosphorylation, and trafficking regulate opioid receptor maturation, signaling efficacy, and membrane availability (24–26). These findings support consideration of sex in future NOP studies, although larger longitudinal studies with hormonal profiling will be needed to clarify its effects in health and disease. In healthy individuals, these effects appear modest and may be more pronounced in clinical populations.

In conclusion, postmortem and PET findings indicate region-specific sex differences in the human NOP system, suggesting that isoform-level dimorphism may contribute to PET-measured receptor availability and warrant further study of hormone- and stress-related regulation.

Supplementary Material

Supplement Description: Figures S1-S3, Tables S1-S4