Sex-Specific Role of NPVF Signalling in Homeostatic Control

Abstract

Neuropeptide VF (NPVF) is a member of the RFamide family of peptides and is suggested to be involved in homeostatic regulations. However, direct evidence is sparse. Here, we generated a NPVF knockout mouse model to comprehensively investigate its role in energy and glucose homeostasis controls. We show that while male Npvf^−/− mice on chow were WT-like at both room temperature (RT 22 °C) and thermoneutrality (TN 28 °C) with regards to body weight, body composition, and the parameters involved in energy homeostasis, female Npvf^−/− mice exhibit significantly reduced water intake at RT and TN regardless of food access, significantly increased the femur bone mineral content at RT and reduced the adiposity at TN. Strikingly, sex differences are absent under high-fat diet (HFD) conditions, with Npvf deletion leading to hyperphagia and increased weight gain in both sexes. Furthermore, Npvf^−/− mice on chow at RT exhibit normal glucose tolerance and insulin action for both sexes. On a HFD or at TN, Npvf^−/− mice display improved and impaired insulin action in females and males, respectively, with female Npvf^−/− mice at TN further showing an improved glucose tolerance. Collectively, these findings establish NPVF as a key regulator of energy and glucose metabolism with sex dimorphism, and are critically dependent on environmental and nutritional factors.

1. Introduction

Maintaining energy homeostasis represents a fundamental biological challenge that requires precise coordination between peripheral metabolic signals and central neural circuits. In mammals, this integration occurs primarily within hypothalamic and brainstem nuclei, which serve as central processing hubs for diverse inputs including circulating hormones, nutrient availability, and visceral sensory information [1,2,3]. While the roles of classical neuropeptides like NPY and POMC-derived peptides in energy balance are well-established [4], recent research has highlighted the importance of less-characterised neuropeptidergic systems, particularly the RFamide peptide family [5,6,7]. Among these, neuropeptide VF (NPVF, also known as RFRP-3) has emerged as a particularly intriguing candidate for metabolic regulation due to its unique expression patterns and sensitivity to both metabolic and environmental cues [5].

The NPVF system originates from the Npvf gene, which encodes two bioactive peptides—the octapeptide NPVF and the 37-amino acid neuropeptide SF (NPSF). These peptides signal primarily through the NPFFR1 receptor, with secondary affinity for NPFFR2 [5]. Anatomically, NPVF-expressing neurons are concentrated in the dorsomedial hypothalamus [5]. While initially characterised for its role as the mammalian ortholog of avian GnIH and its inhibitory effects on reproductive neuroendocrine function [8,9], accumulating evidence suggests that NPVF plays a much broader role in physiological regulation.

Recent studies have revealed that NPVF neurons express receptors for key metabolic hormones including leptin and insulin, positioning them as potential nutrient sensors [10,11]. The system demonstrates remarkable plasticity in response to metabolic challenges, with Npvf expression being significantly downregulated during positive energy balance states such as high-fat diet feeding, while remaining relatively stable during fasting conditions [12]. Most strikingly, NPVF has been identified as one of the most temperature-sensitive neuropeptides, with its expression dramatically downregulated during cold exposure and increased under thermoneutral or warm conditions [13]. Studies have further demonstrated that NPVF neurons can modulate nociception and anxiety, providing a potential mechanistic link between this system and energy expenditure regulation [14,15].

The NPVF system also exhibits notable sexual dimorphism in both its regulation and function. Sex steroids including estrogen and testosterone exert repressive effects on Npvf expression, and recent work has identified the sex-specific projection patterns of NPVF neurons [16,17,18]. These findings raise important questions about whether the metabolic effects of NPVF represent primary functions or secondary consequences of its reproductive modulation, and to what extent these roles may differ between sexes.

To address these questions, our study employs a comprehensive approach using germline Npvf knockout mice to systematically investigate the consequences of NPVF/NPSF deficiency across different dietary and environmental conditions. By examining body weight dynamics, adiposity distribution, glucose homeostasis, and thermogenic capacity in both male and female animals, here we aimed to delineate the system’s role in energy balance regulation and identify potential sex-specific phenotypes. This work not only advances our understanding of NPVF biology but also contributes to the broader effort to map the complex neuropeptidergic networks governing energy homeostasis.

2. Material and Methods

2.1. Animals

All animal studies were approved by the Garvan Institute of Medical Research/St. Vincent’s Hospital Animal Ethics Committee (Project No. 19_22) and conducted in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes. Mice were housed at either room temperature (22 °C) or thermoneutral temperature (28 °C) under a 12 h light/dark cycle (lights on at 07:00). They were provided ad libitum access to water and either a standard chow diet (6% fat, 23% protein, 51% carbohydrate; Cat #27 mod KI, Gordon’s Speciality Stock Feeds) or a high-fat diet (HFD) (43% fat, 17% protein, 40% carbohydrate; Cat #SF03-020, Specialty Feeds). Body weight was monitored weekly from 4 weeks of age, and the health status of all animals was assessed at regular six-month intervals.

Npvf^−/− mice were generated by eliminating all 3 coding exons of the Npvf gene using CRISPR technology in house (Figure 1A). Successful deletion of the Npvf gene was confirmed via PCR using genomic DNA isolated from tail tips. For this, three primers (NPVF-A: 5’-ATTAAGCACATTCCTTGTGTGC-3’; NPVF-B: 5’-GAACTCATCTGTACAGAA GGTG-3’; and NPVF-C: 5’-CCAGTTTACATTACAAGCATCC-3’) were designed to allow for distinguishing WT, heterozygote, and homozygote knockout genotypes in one reaction via melt curve analysis (Figure 1B,C). Deletion of the Npvf gene was also confirmed via RNAScope analysis (Figure 1D). For all experiments, male and female Npvf^−/− and littermate WT were used.

2.2. Mice Monitoring and Tissue Analyses at Room Temperature (RT 22 °C)

WT and Npvf^−/− mice of both sexes were maintained at room temperature (RT, 22 °C). Body weight was monitored weekly from 4 weeks of age (Figure 1E), and several metabolic and physiological parameters were assessed according to the timeline outlined in Figure 1E.

At 10 weeks of age, glucose metabolism was evaluated using an intraperitoneal glucose tolerance test (IPGTT) as described below.

Whole-body composition (fat and lean mass) was measured at 13 weeks using an EchoMRI™ system (Model 2016 E26-277-RM, EchoMRI LLC, Houston, TX, USA). This method quantifies lean mass as the water-containing tissue mass, excluding fat, bone minerals, and NMR-inactive materials (e.g., hair, claws).

Body and brown adipose tissue temperatures were recorded at 15 weeks via infrared imaging.

At 16 weeks, mice were placed in a Promethion^® metabolic monitoring system (Sable Systems, NV) for assessment of food and water intake, energy metabolism, and locomotion. Mice were acclimatised for 72 h under ad libitum conditions, followed by a 24 h fast and a 48 h refeeding period.

At 18 weeks of age, mice were euthanised between 14:00 and 17:00 via cervical dislocation and decapitation. White adipose tissue depots (inguinal, gonadal, mesenteric, and retroperitoneal), brown adipose tissue, and major organs (heart, liver, kidneys, spleen, pancreas, gonads, and seminal vesicles in males) were dissected and weighed as previously described [12].

Femurs were excised, cleaned of soft tissue, and measured for length using a digital micrometre (Mitotokyo, Tokyo, Japan). Bone mineral density and content were analysed via dual-energy X-ray absorptiometry (DEXA; Lunar PIXImus2, GE Healthcare, Madison WI, USA) as previously described [12].

2.3. Mice Monitoring and Tissue Analyses Under Thermoneutrality (TN 28 °C)

A separate set of wild-type and Npvf^−/− mice of both sexes were transferred to a thermoneutral environment (TN, 28 °C) at 4 weeks of age and maintained under TN for 14 weeks. During this time, body weight was monitored weekly. Glucose metabolism, whole-body lean and fat masses, body and brown adipose tissue temperatures, food and water intakes, locomotion, and energy metabolism were examined as described above following the timeline outlined in Figure 1E. Tissue analyses were conducted at 18 weeks of age as described above.

2.4. High-Fat Diet Study

A separate cohort of WT and Npvf^−/− mice of both sexes held at RT were fed on a high-fat diet (HFD) from 4 weeks of age for 14 weeks. Body weight was monitored weekly. Glucose metabolism, whole-body lean and fat masses, and body and brown adipose tissue temperatures were examined after 6, 9, and 11 weeks on the HFD, respectively, as described above. After 12 weeks on the HFD, mice were monitored in the Prometihon system under ad libitum conditions to examine water intake, energy metabolism, and locomotion as described below. Due to the texture of the HFD, which does not permit the accurate assessment of HFD intake in the Promethion system, HFD intake was measured manually after the completion of the Promethion recording. For this, mice were transferred into individual cages with paper towel bedding and allowed to acclimatise for two nights. Daily food intake was adjusted for spillage and determined as the average of duplicate readings taken over 2 consecutive days. Tissue analysis was performed at the end of 14 weeks of HFD feeding as described above.

2.5. Metabolic and Behavioural Profiling Using Promethion System

Metabolic and behavioural profiling was performed using a Promethion^® system (Sable Systems, Las Vegas, NV, USA). Following a 3-day acclimatisation period in the metabolic cages, mice were transferred to the recording system for data collection. Each cage was equipped with a water bottle, food hopper, and habitat, each connected to weight sensors for continuous monitoring of water intake, food intake, and body weight, respectively. Ambulatory activity and position were tracked using XYZ beam arrays (BXYZ-R, Sable Systems; beam spacing: 1 cm), from which the locomotor travel distance was derived.

Energy metabolism was assessed using indirect calorimetry. Air from the cage was sampled via a microperforated stainless steel manifold and analysed using a gas analyser (GA3, Sable Systems). The incurrent flow was maintained at 2000 mL/min using a pull-mode negative pressure system (FR8, Sable Systems). Oxygen consumption (VO₂) and carbon dioxide production (VCO₂) were measured for each cage in 30 s sampling epochs at 5 min intervals, with incurrent air references taken after every fourth cage.

The respiratory exchange ratio (RER) was calculated as VCO₂/VO₂, where an RER of 1 indicates 100% carbohydrate oxidation and 0.7 indicates 100% fat oxidation [19,20]. Energy expenditure (kcal/h) was derived using the Weir equation: 60 × (0.003941 × VO₂ + 0.001106 × VCO₂) [21].

Data acquisition and instrument control were managed using MetaScreen v1.6.2. Raw data were processed using ExpData v1.4.3 (Sable Systems) with custom analysis scripts that performed all necessary transformations and calculations; these scripts are available upon request to the corresponding author.

2.6. Determination of Body and Brown Adipose Tissue Temperatures

Body and brown adipose tissue (BAT) temperatures were measured using high-sensitivity infrared imaging, as described previously [12,22,23]. Two days prior to imaging, the interscapular and lumbar back regions were shaved under light isoflurane anaesthesia to expose the skin for thermal readings. A high-sensitivity infrared camera (ThermoCAM T640, FLIR Systems Australia Pty Ltd.; Mulgrave Victoria 3170, Australia), (sensitivity 0.04 °C, resolution 640 × 480 pixels) was mounted on a tripod 90 cm above the mice. Videos (30–60 s, 30 Hz) of freely moving animals were recorded. From these recordings, thermal frames were selected in which the mouse was standing still or walking slowly, with the body naturally extended and both shaved regions oriented vertically toward the camera lens. Frames showing rapid movement or running were excluded.

Analysis was performed using FLIR ResearchIR MAX 4.40 software, with emissivity set to 0.98 to match rodent skin [22]. The hottest pixel within each defined region—interscapular (BAT) and lumbar (body)—was identified per frame. The BAT temperature was indexed from the interscapular region, and the body temperature from the lumbar region. The difference between these two temperatures (BAT–lumbar) was used as an indicator of BAT thermogenesis [22].

2.7. Cold Exposure Study

A separate cohort of 17–18-week-old wild-type and Npvf^−/− mice of both sexes was used for cold exposure studies. Mice were transferred to individual cages three days before testing. At 09:00 on the test day, basal body and brown adipose tissue (BAT) temperatures were first assessed via infrared imaging (as described above), followed by the measurement of the rectal temperature according to established protocols [7,24,25]. Briefly, a rectal probe (Thermocouple Thermometer QM1601; Protec, Broadmeadow NSW 2292, Australia) connected to a thermometer was inserted and retained until the reading stabilised (~3 s); measurements were completed within 10 s of removing the mouse from its cage.

Following baseline measurements, food was removed and mice were transferred to a temperature-controlled chamber at 4 °C for 6 h. Body and BAT temperatures were monitored hourly via infrared imaging. Rectal temperature was measured again after 3 and 6 h of cold exposure. To ensure reproducibility, all rectal temperature measurements were performed by the same skilled researcher using a consistent procedure throughout the study.

2.8. mRNA Detection via RNAscope

We used the RNAscope fluorescent assay (Advanced Cell Diagnostics, Inc., Newark, CA, USA)—an in situ hybridisation (ISH) technique for visualising multiple cellular mRNA targets in fresh frozen tissues [26]—to detect the expression of Npvf in the adult WT and Npvf^−/− mouse brains. Brain coronal sections (25 µm) containing the dorsomedial hypothalamic nuclei—the primary region in the brain expressing Npvf [5,27]—were cut on a cryostat, thaw-mounted on Superfrost^® slides (Menzel-Glaser, Braunschweig, Germany), and labelled for the mRNA of Npvf (ACD #520361-C3) using the RNAscope^® Multiplex Fluorescent Detection Kit following the manufacturer’s protocol (#ADV320851, Advanced Cell Diagnostics, Inc.). Section staining was photographed using a DM 5500 microscope (Leica, Wetzlar, Germany).

2.9. Intraperitoneal Glucose Tolerance Test

IPGTT was performed as described previously [28]. Briefly, food was removed from cage hoppers at 09:00 hr, and 6 h later a dose of glucose (1 g/kg body weight) was injected into the peritoneal cavity. Tail vein blood was collected at 0, 15, 30, 60, and 90 min after the glucose bolus injection for the determination of glucose concentrations using a glucometer (Accu-Chek^® Go, Roche) and blood insulin levels using an ELISA kit from Crystal Chem (Crystal Chem, Elk Grove Village, IL, USA). Glucose tolerance curves for glucose and insulin are presented as absolute values. Area under the glucose or insulin concentration curves between 0 and 90 min after glucose administration were calculated.

2.10. Statistical Analyses

All data are expressed as means ± SEM. Time course data for energy expenditure, the respiratory exchange ratio, and locomotion were binned forward hourly (e.g., 7 h data were binned from data collected between 07:00 and 08:00, and so on). The differences between two genotypes in body composition and femur analysis were evaluated using Student’s t-test. Differences between different genotypes in the time course data, data averaged/summed over different conditions (i.e., light phase, dark phase, and 24 h period), and the weights of multiple tissues and organs were assessed using repeated measures ANOVA with post-hoc analysis where appropriate to assess the group difference at each time point, condition, or tissue/organ, using the Sidak method to correct for multiple comparisons. Data were tested for normal distribution using the Shapiro–Wilk test. Parametric tests were used for data meeting the normality assumption; non-parametric equivalents were applied where appropriate. Comparisons of energy expenditure (kcal/h) were carried out via ANCOVA with the metabolically active tissue (MAT) weight [6,12,29,30,31] as the covariate to determine the genotype effect on energy expenditure that is independent of body composition. MAT was defined as the whole-body lean mass plus 20% of the whole-body fat mass based on the estimation that the fat mass makes a contribution to energy expenditure that is ~20% of that of the equivalent lean mass, as discussed, and the specific metabolic activity of adipose tissue is ~20% of the lean tissue [32,33]. In ANCOVA analysis, the slopes of the regression lines were compared via a test of the homogeneity of slopes. The adjusted means of energy expenditure at a common MAT weight were generated using ANCOVA and presented. ANCOVA analysis was performed with SPSS version 29 (SPSS, Chicago, IL). Other statistical analyses were conducted using PRISM, version 10.0 (GraphPad Software Inc., San Diego, CA, USA). Analyses of variance (ANOVAs) were performed using Type III sums of squares, which is appropriate for designs with unequal group sizes. Statistical significance was defined as p < 0.05.

3. Results

3.1. Lack of NPVF Signalling Does Not Impact Reproductive Fitness

To assess the systemic impact of NPVF loss, we evaluated general health and reproductive fitness. Npvf^−/− mice were viable and displayed no overt physical abnormalities. Heterozygous crosses produced offspring at the expected Mendelian frequency, and the litter size, mortality, and sex ratios from both heterozygous and homozygous breeding pairs were comparable to those of wild-type controls (

Table 1. Breeding profiles from WT, Npvf^−/−, and Npvf^+/− breeding pairs.

Type of Breeding Pair	WT × WT	Npvf−/− × Npvf−/−	Npvf+/− × Npvf+/−
Offspring genotype composition from Npvf+/− breeding pairs (% total born)
WT	-	-	25.76 ± 2.94%
Npvf−/−	-	-	22.54 ± 3.30%
Npvf+/−	-	-	51.69 ± 3.85%
Litter size (N/Litter)	5.9 ± 0.46	6.3 ± 0.37	7.1 ± 0.29
Birth mortality rate (% total born)	1.88 ± 1.37%	3.94 ± 1.49%	1.75 ± 0.98%
Male % total born	53.3 ± 4.25%	48.2 ± 4.32%	53.8 ± 3.88%

Twenty-three litters produced from ten Npvf^+/− breeding pairs were used for the analysis of offspring genotype composition from heterogeneous breeding. For the analysis of litter size, mortality rate, and sex ratio, the numbers of litters were 20, 28, and 23 for WT, Npvf^−/−, and Npvf^+/− breeding pairs, respectively. Data are Mean ± SEM. One-way ANOVA was used to determine difference among groups. One-sample t test was used to test whether the frequency of genotypes from heterozygous breeding pairs is statistically different from expected Mendelian ratio (i.e., 25% for WT or homozygous offsprings, and 50% for heterozygous offsprings), and whether sex ratio (Male % total born) is statistically different from 50%.

). These data indicate that NPVF deficiency does not compromise overall viability or reproductive fitness.

3.2. Sex-Dependent Effects of NPVF Signalling on Bone Metabolism, Oxidative Fuel Selection, and Water Intake in Mice on Chow at Room Temperature (RT 22 °C)

Baseline characterisation of chow-fed male and female Npvf^−/− mice at room temperature (RT, 22 °C) revealed comparable growth in body weight to that of sex-matched wild-type (WT) mice (Figure 2A,D). Body composition analysis using EchoMRI at 13 weeks of age showed no significant differences in whole-body lean (Figure 2B,E) or fat (Figure 2C,F) masses between genotypes in either sex. Consistent with these findings, the weights of individual white adipose tissue (WAT) depots (inguinal, gonadal, mesenteric, and retroperitoneal) (Figure S1A,I) and their combined mass (Figure S1B,J), dissected at 18 weeks of age, were also not statistically different between Npvf^−/− and WT mice in both sexes. In addition, no significant genotype differences were found in the weights of brown adipose tissue (Figure S1C,K) or other organs examined (Figure S1D,L). Examination of the femur reveals no significant genotype differences in the femur length (Figure 2G,J) or the bone mineral density (Figure 2H,K) in either sex. Interestingly, female but not male Npvf^−/− mice had a significantly higher femur mineral content than their sex-matched controls (Figure 2I,L), indicating the sex-dependent role of NPVF signalling in the regulation of bone metabolism.

Metabolic and behavioural profiling using the Promethion system revealed that under ad libitum feeding male and female Npvf^−/− mice did not significantly differ from their sex-matched WT controls with regards to energy expenditure (Figure 2M,O), respiratory exchange ratio (Figure 2N,P), locomotion (Figure S1E,M), or food intake (Figure 2Q,S), either expressed as hourly time course or summarised over dark and light phases and a 24 h period (

Table 2. Effects of NPVF deletion on respiratory exchange ratio, energy expenditure, locomotion, and food and water intakes during dark and light phases and 24 h period in male and female mice at room temperature (RT 22 °C) on chow, thermoneutral environment (TN 28 °C) on chow, or RT on a high-fat diet (HFD).

		Male		Female
		WT	Npvf−/−	WT	Npvf−/−
RT (22 °C) chow
Respiratory exchange ratio	Dark	0.929 ± 0.011	0.942 ± 0.006	0.973 ± 0.014	0.966 ± 0.013
	Light	0.830 ± 0.008	0.841 ± 0.008	0.850 ± 0.015	0.872 ± 0.014
	24 h	0.880 ± 0.007	0.891 ± 0.006	0.911 ± 0.014	0.919 ± 0.013
Energy expenditure (kcal/h)	Dark	0.528 ± 0.009	0.528 ± 0.009	0.485 ± 0.010	0.489 ± 0.014
	Light	0.421 ± 0.005	0.433 ± 0.004	0.386 ± 0.006	0.399 ± 0.008
	24 h	0.474 ± 0.005	0.480 ± 0.006	0.436 ± 0.008	0.444 ± 0.010
Locomotion (m/phase)	Dark	82.66 ± 7.34	91.55 ± 9.39	127.7 ± 8.29	123.8 ± 10.4
	Light	38.50 ± 2.90	42.63 ± 4.15	41.25 ± 2.96	42.34 ± 4.32
	24 h	121.1 ± 9.11	134.2 ± 11.5	169.0 ± 6.50	166.1 ± 13.3
Food intake (g/phase)	Dark	4.863 ± 0.610	4.571 ± 0.330	5.741 ± 0.461	5.893 ± 0.213
	Light	1.173 ± 0.302	1.102 ± 0.193	1.256 ± 0.318	1.584 ± 0.209
	24 h	6.036 ± 0.795	5.763 ± 0.419	6.997 ± 0.601	7.477 ± 0.289
Water intake (g/phase)	Dark	2.049 ± 0.110	3.081 ± 0.823	2.141 ± 0.112	1.725 ± 0.078 *
	Light	0.417 ± 0.060	1.220 ± 0.668	0.627 ± 0.100	0.552 ± 0.027
	24 h	2.466 ± 0.101	4.301 ± 1.479	2.768 ± 0.179	2.276 ± 0.075
TN (28 °C) chow
Respiratory exchange ratio	Dark	0.860 ± 0.011	0.849 ± 0.009	0.840 ± 0.022	0.842 ± 0.010
	Light	0.826 ± 0.006	0.807 ± 0.007	0.766 ± 0.022	0.770 ± 0.012
	24 h	0.843 ± 0.008	0.828 ± 0.007	0.803 ± 0.021	0.806 ± 0.010
Energy expenditure (kcal/h)	Dark	0.377 ± 0.012	0.352 ± 0.006 (p = 0.06)	0.337 ± 0.003	0.354 ± 0.006 (p = 0.06)
	Light	0.255 ± 0.007	0.256 ± 0.006	0.222 ± 0.007	0.229 ± 0.005
	24 h	0.316 ± 0.009	0.304 ± 0.005	0.280 ± 0.005	0.292 ± 0.004
Locomotion (m/phase)	Dark	98.04 ± 14.7	94.13 ± 11.4	140.3 ± 21.4	143.7 ± 13.4
	Light	29.35 ± 2.11	40.64 ± 9.32	40.67 ± 3.45	45.88 ± 7.61
	24 h	127.4 ± 16.8	124.8 ± 20.5	181.0 ± 24.1	189.5 ± 20.6
Food intake (g/phase)	Dark	4.189 ± 0.619	3.413 ± 0.315	4.077 ± 0.376	3.828 ± 0.382
	Light	1.077 ± 0.127	0.931 ± 0.229	0.668 ± 0.124	0.682 ± 0.110
	24 h	5.266 ± 0.713	4.345 ± 0.381	4.735 ± 0.422	4.509 ± 0.441
Water intake (ml/phase)	Dark	1.933 ± 0.106	1.931 ± 0.222	1.961 ± 0.261	1.457 ± 0.078
	Light	0.563 ± 0.077	0.808 ± 0.117	0.789 ± 0.067	0.638 ± 0.043
	24 h	2.496 ± 0.180	2.739 ± 0.241	2.750 ± 0.320	2.095 ± 0.085
RT (22 °C) HFD
Respiratory exchange ratio	Dark	0.842 ± 0.012	0.853 ± 0.006	0.829 ± 0.013	0.846 ± 0.006
	Light	0823 ± 0.012	0.839 ± 0.008	0.804 ± 0.020	0.822 ± 0.008
	24 h	0.833 ± 0.011	0.846 ± 0.006	0.817 ± 0.016	0.834 ± 0.008
Energy expenditure (kcal/h)	Dark	0.543 ± 0.009	0.566 ± 0.005	0.513 ± 0.008	0.519 ± 0.007
	Light	0.452 ± 0.005	0.460 ± 0.008	0.422 ± 0.007	0.410 ± 0.005
	24 h	0.498 ± 0.007	0.513 ± 0.006	0.468 ± 0.007	0.464 ± 0.004
Locomotion (m/phase)	Dark	91.62 ± 3.87	103.8 ± 9.76	123.4 ± 10.3	121.8 ± 15.4
	Light	36.13 ± 6.23	33.91 ± 9.54	31.40 ± 4.72	26.15 ± 3.45
	24 h	127.8 ± 9.14	137.7 ± 17.7	154.9 ± 14.1	147.9 ± 16.0
Water intake (g/phase)	Dark	1.544 ± 0.21	1.895 ± 0.386	1.344 ± 0.092	1.538 ± 0.121
	Light	1.030 ± 0.293	0.808 ± 0.171	1.037 ± 0.369	0.625 ± 0.027
	24 h	2.573 ± 0.497	2.702 ± 0.544	2.380 ± 0.405	2.163 ± 0.131

Data are Mean ± SEM. For RT chow cohort, n = 5–6 and 9 for WT and Npvf^−/−, respectively, in males; n = 7 and 7–8 for WT and Npvf^−/−, respectively, in females. For TN chow cohort, n = 4 and 6 for WT and Npvf^−/−, respectively, in males; n = 6 and 6–7 for WT and Npvf^−/−, respectively, in females. For RT HFD cohort, n = 9 and 7 for WT and Npvf^−/−, respectively, in males; n = 6 for WT and Npvf^−/− in females. Comparisons of energy expenditure were carried out using ANCOVA with the metabolically active tissue (MAT) weight as covariate. MAT was defined as whole-body lean mass plus 20% of whole-body fat mass. The adjusted means of energy expenditure (kcal/h) at common MAT weight were generated using ANCOVA and presented. For RT chow, the common MAT weights were 23.97 g and 17.59 g for males and females, respectively. For TN chow, the common MAT weights were 23.0 g and 16.8 g for males and females, respectively. For RT HFD, the common MAT weights were 23.9 g and 17.73 g for males and females, respectively. All other parameters were assessed for genotype differences using repeated measures ANOVA with post-hoc analysis where appropriate to assess group difference at each time point or condition, using Sidak method to correct for multiple comparisons. * p < 0.05 or value as indicated vs. sex-matched WT under the same condition.

). Consistently, the cumulative locomotion (Figure S1F,N) and food intake (Figure 2R,T) over the monitoring period were not statistically different between WT and Npvf^−/− mice in both sexes. Water intake on the other hand, while comparable between WT and Npvf^−/− mice in males (Figure 2U,V, Table 2), showed a significant decrease in female Npvf^−/− mice (Figure 2W,X, Table 2; main effect of genotype: F(1,12) = 7.51, p < 0.05 in Figure 2W; F(1,12) = 7.70, p < 0.05 in Figure 2X; F(1,12) = 6.41, p < 0.05 in Table 2), mostly due to a decrease during the dark phase (Figure 2W, Table 2).

Mice were further examined in the Promethion system during 24 h fasting followed by 48 h refeeding. Both male and female Npvf^−/− mice showed comparable energy expenditures to their sex-matched counterparts during the fasting–refeeding challenge (Figure 2Y,AA). Male but not female Npvf^−/− mice exhibited a lower respiratory exchange ratio during the last 3 h of fasting and first 9 h of refeeding compared to the sex-matched WT controls (Figure 2Z,AB; main effect of genotype during indicated period in Figure 2Z: F(1,13) = 4.75, p < 0.05), indicating a greater usage of lipid as oxidative fuel in male Npvf^−/− versus WT mice under and during the early recovery from a negative energy balance. Locomotion during the fasting–refeeding challenge (Figure S1G,H,O,P) and food intake during refeeding (Figure 2AC,AE) did not significantly differ between WT and Npvf^−/− mice in either sex. Furthermore, water intake in female but not male Npvf^−/− mice showed a significant decrease compared to their sex-matched WT during both fasting and refeeding periods (Figure 2AD,AF; main effect of genotype in Figure 2AF: F(1,11) = 10, p < 0.01 for fasting, F(1,11) = 8.78, p < 0.05 for refeeding period), similar to observations under ad libitum feeding (Figure 2U–X, Table 2).

Taken together, these data suggest the important roles of NPVF signalling in the regulation of bone metabolism, oxidative fuel selection during a challenge to energy balance, and fluid homeostasis under RT chow conditions. Furthermore, these data suggest that NPVF signalling exerts these regulations in a sex-dependent manner.

3.3. Sex-Dependent Effects of NPVF Signalling on Body Composition, Energy Expenditure, and Water Intake in Mice on Chow at Thermoneutrality (28 °C)

Given the known influence of ambient temperature on Npvf expression [13], we next evaluated male and female Npvf^−/− mice maintained on a chow diet under thermoneutral conditions (TN, 28 °C). Both male and female Npvf^−/− mice had comparable body weights relative to their sex-matched counterparts at the time of transfer to the TN condition and displayed similar growth curves during the following 14 weeks maintained at TN (Figure 3A,D). Body composition—determined at 13 weeks of age using EchoMRI—revealed comparable whole-body lean masses between genotypes for both sexes (Figure 3B,E), whilst the whole-body fat mass showed a significant reduction in female but not male Npvf^−/− versus WT mice (Figure 3C,F). Consistent with this reduced adiposity, weights across the white adipose tissue depots in female Npvf^−/− mice dissected at 18 weeks of age were significantly reduced compared to female WT mice (Figure 3I, main effect of genotype F(1,9) = 7.32, p < 0.05), and the summed weights of these dissected white adipose tissue depots showed a strong trend towards decreasing (Figure 3J). No significant genotype difference in the weights of white adipose tissue depots (Figure 3G) or the summed weight of these depots (Figure 3H) were observed in male mice. Furthermore, no significant differences were found in brown adipose tissue weight (Figure S2A,J), the weights of other examined major organs (Figure S2B,K), femur length (Figure S2C,L), femur bone mineral density (Figure S2D,M), or mineral content (Figure S2E,N) between Npvf^−/− and WT mice for either sex.

Mice were further examined for energy metabolism and behaviours in the Promethion system. At TN, whilst male Npvf^−/− mice exhibited comparable energy expenditure (Figure 3K, Table 2), female Npvf^−/− mice showed an increase in energy expenditure during the dark phase (Figure 3M, Table 2; main effect of genotype during indicated dark periods: F(1,9) = 5.68, p < 0.05), which was likely a main contributor to the reduced fat mass observed in these mice (Figure 3F,I,J). The respiratory exchange ratio (Figure 3L,N, Table 2), locomotion (Figure S2F,O, Table 2), and food intake (Figure 3O,P,Q,R, Table 2), either expressed as hourly time course or summarised over light, dark, and 24 h periods did not significantly differ between Npvf^−/− and WT mice for either sex. Moreover, the cumulative locomotion (Figure S2G,P) and food intake (Figure 3P,R) over the monitoring period were comparable between genotypes for both males and females. Furthermore, whilst similar between Npvf^−/− and WT mice in males (Figure 3S,T, Table 2), the water intake in females was reduced in Npvf^−/− compared to WT mice (Figure 3U,V, Table 2; main effect of genotype: F(1,10) = 3.79, p = 0.08 in Figure 3U, F(1,10) = 5.37, p < 0.05 in Figure 3V), F(1,10) = 3.92, p = 0.08 in Table 2), significantly so when represented as accumulated data (Figure 3V). We further challenged mice at TN with a fasting–refeeding paradigm. Like the observations with ad libitum feeding, the energy expenditure was comparable between genotypes in male mice (Figure 3W, Table 2) and showed an increase in female Npvf^−/− versus WT mice (Figure 3Y, main effect of genotype: F(1,9) = 5.31, p < 0.05 for indicated period during fasting; F(1,9) = 5.95, p < 0.05 for indicated period during refeeding). The respiratory exchange ratio (Figure 3X,Z) and locomotion (Figure S2H,I,Q,R) during the fasting–refeeding challenge, and the food intake during the refeeding period (Figure 3AA,AC) did not significantly differ between genotypes for either sex. Additionally, the water intake in male mice showed no genotype differences during the fasting or subsequent refeeding period (Figure 3AB), whereas in female mice, it showed a significant reduction in Npvf^−/− compared to WT mice during the fasting period (Figure 3AD; main effect of genotype: F(1,10) = 5.75, p < 0.05 during fasting; F(1,10) = 4.35, p = 0.06 over whole period), consistent with observations in these mice under ad libitum feeding conditions (Figure 3U,V).

Taken together, these results are consistent with the sex-dependent role of NPVF signalling in the regulation of body composition, energy expenditure, and water intake under thermoneutral conditions.

3.4. Greater Weight Gain and Food Intake in Npvf^−/− Mice on a High-Fat Diet

When challenged with a high-fat diet (HFD) from 4 weeks of age in RT conditions, both male and female Npvf^−/− mice exhibited greater weight growth compared to their sex-matched WT controls (Figure 4A,D; main effect of genotype: F(1,14) = 17.58, p < 0.001 for males in Figure 4A; F(1,10) = 9.63, p < 0.05 for females in Figure 4D). Interestingly, this greater weight growth appears to be mainly due to a higher lean mass rather than fat mass, in that the whole-body lean mass examined at 13 weeks of age was significantly greater in Npvf^−/− versus WT mice for both sexes (Figure 4B,E), whereas the whole-body fat mass was comparable between genotypes (Figure 4C,F). In addition, there was no significant difference in the weights of white adipose tissue depots dissected at 18 weeks of age (Figure 4G,I) or the summed weight of these dissected white adipose tissue depots (Figure S3A,H) between Npvf^−/− and WT mice for either sex. Interestingly, liver weight significantly increased in both male and female Npvf^−/− mice compared to their sex-matched WT mice (Figure 4H,I; for males in Figure: 4H main effect of genotype F(1,14) = 12.99, p < 0.01; genotype x tissue type interaction F(1.78, 24.92) = 8.93, p < 0.01; for females in Figure 4I: main effect of genotype F(1,10) = 21.69, p < 0.001, genotype x tissue type interaction F(1.45, 14.22) = 12.56, p < 0.01), whereas other organs (Figure 4H,J) and brown adipose tissues (Figure S3B,I) showed comparable weights between genotypes for both sexes. Furthermore, whilst the femur length was comparable between WT and Npvf^−/− mice for both sexes (Figure 4K,N), the femur bone mineral density significantly increased in male but not female Npvf^−/− mice (Figure 4L,O), and the femur bone mineral content significantly increased in female but not male Npvf^−/− mice (Figure 4M,P) compared to their sex-matched counterparts, indicating a more pronounced bone phenotype in Npvf^−/− versus WT mice in both sexes on a HFD.

While energy expenditure did not significantly differ between Npvf^−/− and WT mice for both sexes (Figure 4Q,S), Npvf^−/− mice exhibited an increase in HFD intake, although this increase did not reach statistical significance in females (p = 0.06) (Figure 4R,T). This increase in food intake is likely a main contributor to the greater weight gain in Npvf^−/− mice on a HFD (Figure 4A,D). The respiratory exchange ratio (Figure S3C,J), locomotion (Figure S3D,F,K,L), and water intake (Figure S3F,G,M,N) were comparable between genotypes for both sexes. The lack of impact of NPVF deletion on water intake in female mice on the HFD (Figure S3M,N) contrasts with the pronounced reduction in water intake observed on chow under both ambient temperatures (Figure 2W,X and Figure 3U,V), suggesting a diet-dependent modulation of fluid homeostasis through NPVF signalling. In summary, these data suggest the sex-consistent role of NPVF signalling in regulating body weight and food intake under HFD conditions.

3.5. Effects of NPVF Deletion on Brown Adipose Tissue Thermogenesis

Brown adipose tissue (BAT) is a key thermogenic organ in rodents, essential in cold-induced thermogenesis to defend body temperature and a primary site of diet-induced adaptive thermogenesis [34,35]. Consequently, BAT thermogenesis constitutes a major component of total energy expenditure.

To examine the impact of disrupted NPVF signalling on energy metabolism, we assessed BAT thermogenesis in wild-type and Npvf^−/− mice under varying ambient temperatures and in response to a high-fat diet (HFD) challenge using infrared imaging. BAT thermogenesis was quantified by measuring the skin surface temperature over the interscapular BAT depot and the lumbar back region in freely moving mice. The temperature difference between these regions (ΔT_BAT-Back) served as an indicator of thermogenic activity, based on the principle that increased BAT heat production directly elevates the local skin temperature, thereby enlarging the (ΔT_BAT-Back) [12,22].

In male mice, BAT thermogenesis was comparable between WT and Npvf^−/− mice at the baseline RT chow condition (Figure 5A). In response to the HFD, both WT and Npvf^−/− mice showed a significant increase in BAT thermogenesis compared to those in basal chow conditions (Figure 5A), consistent with a diet-induced thermogenesis. When held in the TN environment, where there is less thermogenic demand to maintain body temperature, both male WT and Npvf^−/− mice showed an expected reduction in BAT thermogenesis compared to their genotype-matched baseline controls, with no significant difference between genotypes (Figure 5A, main effect of conditions F(2,35) = 39.44, p < 0.0001). On the other hand, in females, Npvf^−/− mice displayed significantly higher BAT thermogenesis compared to WT mice at baseline (Figure 5B). In response to a HFD, while an increase in BAT thermogenesis from baseline chow conditions was observed in WT mice (Figure 5B), this was absent in Npvf^−/− mice (Figure 5B), indicating impaired diet-induced thermogenesis in female mice lacking NPVF signalling. When held at TN, both female WT and Npvf^−/− mice showed an expected reduction in BAT thermogenesis compared to their genotype-matched RT chow baseline, with no significant genotype effects (Figure 5B, main effect of conditions: F(2,39) = 25.88, p < 0.0001, genotype and condition interaction: F(2,39) = 2.74, p = 0.077).

We further challenged WT and Npvf^−/− mice with cold exposure, where mice were held at 4 °C for 6 h and assessed for BAT thermogenesis hourly. BAT thermogenesis showed a marked and significant increase from the RT baseline value at 1 h after exposure to 4 °C, maintained at elevated levels throughout the remaining cold exposure challenge with no significant genotype differences in either male or female mice (Figure 5C,D; main effect of time: F(4.29,94.38) = 49.10, p < 0.0001 for males in Figure 5C; F(3.69,59.04) = 42.70, p < 0.0001 for females in Figure 5D). The rectal temperature was comparable between genotypes at baseline before cold exposure and showed a significant reduction at 3 h after the induction of cold exposure, with no significant genotype effects for either sex (Figure 5E,F; main effect of time: F(1.54,33.95) = 91.37, p < 0.0001 for male in Figure 5E; F(1.60,25.58) = 20.56, p < 0.0001 for female in Figure 5F). The rectal temperature measured at 6 h after the cold exposure induction, while still significantly lower than baseline, did not significantly differ from previous measurements at 3 h after the commencement of cold exposure for both genotypes and sexes (Figure 5E,F), indicating that both male and female mice were able to induce enough BAT thermogenesis to maintain a stable body temperature, albeit at a lower level during the 4 °C cold challenge, and more importantly, this ability was not impacted by lack of NPVF signalling.

Taken together, these data suggest that NPVF signalling plays a role in the regulation of diet-induced adaptive BAT thermogenesis in female mice, but may not be involved in the regulation of cold-induced BAT thermogenesis in either sex.

3.6. Sex-Dependent Effects of NPVF Signalling on Glucose Metabolism

Given the recent evidence suggesting the involvement of NPVF and its cognate receptor NPFFR1 in the control of glucose metabolism [17,36], we investigated the impact of a lack of NPVF signalling on glucose homeostasis by examining the blood glucose and insulin responses to an intraperitoneal glucose bolus (1 g/kg body weight) in WT and Npvf^−/− mice. Under RT chow conditions, both male and female Npvf^−/− mice exhibited a comparable blood glucose excursion following an i.p. glucose bolus relative to their sex-matched WT controls (Figure 6A,C), leading to an unaltered area under the blood glucose curves (Figure 6B,D). Furthermore, the blood insulin levels following the i.p. glucose bolus (Figure 6E,G) and area under the insulin curves (Figure 6F,H) did not significantly differ between WT and Npvf^−/− mice for either sex. These data suggest that a lack of NPVF signalling in RT chow conditions may not have a significant impact on glucose tolerance and insulin responses.

Since ambient temperature has been shown to impact glucose metabolism [37,38] as well as alter Npvf expression [13], we examined glucose metabolism in Npvf^−/− and WT mice under TN conditions. Interestingly, while male Npvf^−/− mice exhibited a similar blood glucose excursion to WT mice (Figure 7A,B) (— indicative of an unaltered glucose tolerance), they had significantly elevated insulin levels (Figure 7C, main effect of genotype: F(1,16) = 5.055, p < 0.05) and area under the insulin curves (Figure 7D), indicating impaired insulin action in male Npvf^−/− mice at TN. On the other hand, a lack of NPVF in female mice resulted in a significantly lower glucose excursion (Figure 7E, main effect of genotype: F(1,12) = 7.49, p < 0.05) and a strong trend towards decreasing in the area under the glucose curves (Figure 7F, p = 0.05 by Mann-Whitney U), following the i.p. glucose bolus compared to WT mice, indicating a greater glucose tolerance. Furthermore, this greater glucose tolerance was associated with a significantly reduced insulin excursion (Figure 7G, main effect of genotype: F(1,10) = 7.51, p < 0.05) and area under the insulin curves (Figure 7H), suggesting enhanced insulin action in female Npvf^−/− versus WT mice in TN conditions.

Since the HFD is also known to impact glucose metabolism [39], we examined glucose tolerance in WT and Npvf^−/− mice fed on a HFD. Both male and female Npvf^−/− mice on the HFD had comparable glucose tolerance to their sex-matched WT mice in that the blood glucose curves (Figure 8A,C) and resulting area under the blood glucose curves (Figure 8B,D) were comparable between WT and Npvf^−/− mice for both sexes. Interestingly, the blood insulin curves in male Npvf^−/− mice were significantly elevated from that of WT mice (Figure 8E, main effect of genotype: F(1,15) = 5.94, p < 0.05), leading to a significantly greater area under the glucose curve in male Npvf^−/− vs. WT mice (Figure 8F), suggesting impaired insulin action in male Npvf^−/− mice fed a HFD. By contrast, female Npvf^−/− mice on the HFD showed a lower blood insulin curve compared to WT mice, significantly so from 30 min onwards after the i.p. glucose challenge (Figure 8G, main effect of genotype from 30 min onwards: F(1,10) = 8.44, p < 0.05). The area under the insulin curve over the 90 min glucose tolerance test showed a strong trend towards decreasing (p = 0.06) in female Npvf^−/− vs. WT mice (Figure 8H). These data suggest the sex-dependent effect of NPVF signalling on insulin action in mice on the HFD, with a lack of NPVF signalling in mice on the HFD leading to impaired and enhanced insulin action in males and females, respectively. In summary, these results suggest that NPVF signalling is an important player in the regulation of glucose homeostasis, and that this regulation interacts with diet and ambient temperature in a sex-dependent manner.

4. Discussion

Our findings reveal the complex and sexually dimorphic role of NPVF signalling in the regulation of energy homeostasis and glucose metabolism that is significantly influenced by both ambient temperature and dietary conditions. While dispensable for basal survival and reproduction, NPVF exerts nuanced, sexually dimorphic control over energy partitioning, glucose homeostasis, and related metabolic processes, with its influence heavily dependent on the environmental and dietary context.

The lack of an overt phenotype in development, Mendelian distribution, reproductive fitness, and reproductive organ weight in Npvf^−/− mice is a critical finding. This is consistent with Mamgain et al. [40], showing that NPVF-ablated mice exhibited apparent normal fertility and unaltered weight of seminal vesicles in males and uteri in females. It suggests that NPVF signalling may not be part of the core, non-redundant circuitry required for gestation, parturition, or postnatal survival under standard conditions. While in contrast with other neuropeptide systems (e.g., leptin, melanocortin), where deletion often causes severe metabolic or reproductive deficits [41,42], our findings suggest that NPVF may function more as a fine-tuning modulator rather than a primary driver of these essential functions. This may be somewhat surprising since data show that NPVF negatively regulates GnRH neurons and thus appears to act as a ‘brake’ on the neuroendocrine reproductive axis [9], and pharmacological experiments have also shown that NPVF/RFRP-3 has a dose-dependent stimulatory effect on LH secretion when administered centrally to both intact and castrated male mice [18]. The lack of any major effects on reproduction in our Npvf^−/− animals under normal conditions therefore may suggest a possible compensatory action from other systems such as kisspeptin neurons to overcome the functional loss of NPVF signalling [43] and give weight to the emerging modulatory role of NPVF in reproductive function and stress responses [40,44,45].

The most consistent and striking finding across this study is the profound sexual dimorphism in the phenotypic consequences of Npvf deletion. This dimorphism in metabolic phenotypes manifests best in thermoneutral conditions, a state that reveals intrinsic metabolic regulation by removing the confounding variable of cold stress. At thermoneutrality, female Npvf^−/− mice exhibited a distinct lean phenotype characterised by a reduced fat mass and increased energy expenditure during the dark phase. This suggests that NPVF in females acts as a tonic brake on thermogenic energy dissipation when there is minimum thermal demand for heat generation to maintain body temperature. By contrast, males showed no such phenotype at thermoneutrality, highlighting the sex-dependent role of NPVF signalling in this regulation.

Furthermore, our findings reveal the complex and sexually dimorphic role of NPVF signalling in the regulation of glucose homeostasis that is influenced by both the environmental temperature and dietary composition. While both male and female Npvf^−/− mice were WT-like with regards to glucose tolerance and insulin responses to i.p. glucose bolus in RT chow conditions, under thermoneutrality or challenged with a HFD at RT, male Npvf^−/− mice developed a profile suggestive of insulin resistance, requiring hyperinsulinemia to maintain normal glucose tolerance. Conversely, female Npvf^−/− mice displayed enhanced insulin action, achieving better glucose clearance with lower insulin secretion. This positions NPVF as a negative regulator of insulin action in females but a supportive factor in males, a divergence with significant implications for understanding sex-biased susceptibility to type 2 diabetes.

Another intriguing finding is the development of positive energy balance and obesity in HFD-fed KO mice despite their lean phenotype under chow-fed conditions. This suggests that the protective metabolic role of NPVF signalling is overridden by the potent obesogenic stimulus of a HFD. One plausible mechanism is that the inherent hyper-thermogenic drive in KO females (evident at RT) may be suppressed or maximised at thermoneutrality, eliminating this energy-dissipating pathway and unmasking a latent susceptibility to diet-induced obesity. Future studies directly measuring brown adipose tissue function and sympathetic tone under thermoneutral HFD conditions are needed to test this hypothesis and resolve the context-dependent role of NPVF in energy balance.

Our results also reveal the novel role of NPVF signalling in the regulation of bone metabolism. Thus, in females but not males, a lack of NPVF signalling in mice on chow at RT results in an increased femur bone mineral content, and this effect is abolished at thermoneutrality. When fed on a HFD at RT, both male and female Npvf^−/− mice displayed a high-bone phenotype. These findings suggest that NPVF signalling may have a generally inhibitory action on bone accretion, an effect potentially linking energy status to skeletal re-modelling, and that this action of NPVF signalling interacts with environmental temperature and dietary conditions in a sex-dependent manner. It is interesting to note that NPFF—a RFamide family peptide closely related to NPVF—has also been shown to play a role in the regulation of bone metabolism specifically in females under RT chow conditions [6], highlighting the recurrent theme of the sex dimorphic involvement of RFamide peptides in bone homeostasis. Moreover, in keeping with the involvement of the RFamide peptides in skeletal health, NPFFR2—the cognate receptor for NPFF with an affinity to NPVF—has been reported as an important player in bone metabolic control [12]. The possible role of NPFFR1—the cognate receptor for NPVF—in bone homeostasis may warrant future investigation.

Interestingly, our data show a profound reduction in water intake in female but not male Npvf^−/− mice regardless of ambient temperature or food accessibility. This reveals NPVF signalling as a novel player in the regulation of fluid homeostasis, which is possibly linked to its involvement in stress and pressor responses [45,46,47,48,49]. Interestingly, NPFF has also been shown to play a role in the regulation of fluid homeostasis [6]. In conjunction with these findings, it may point to the closely related NPFF and NPVF systems as an integrated unit to regulate fluid balance. In keeping with this notion, NPFF and its associated receptors NPFFR2 and NPFFR1 have emerged as novel players in the regulation of blood pressure and renal function [50,51]. Moreover, our finding highlights a sex dimorphism in the action of NPVF signalling in the control of fluid balance, in contrast to the sex-consistent role of NPFF in this control [6]. Furthermore, the lack of effect of NPVF deletion on water intake in mice of either sex on a HFD indicates the interaction of NPVF signalling with dietary conditions in this regulation.

The sexually dimorphic metabolic effects of NPVF signalling align with the dual modes of sex hormone actions: the permanent organisational effects on neural circuits during development and the acute activation effects in adulthood. A key mechanistic hypothesis is that NPVF neurons in the dorsomedial hypothalamus (DMH) are direct targets for estradiol and testosterone, which could drive sex-specific circuit connectivity or neuronal excitability, whereby estradiol in females specifically enhances the functional link between NPVF activation and the suppression of sympathetic outflow to thermogenic tissues, making this energy-conserving “brake” more potent [18,52]. This may be an adaptive mechanism for safeguarding energy stores. Conversely, in males, the lack of this estrogen-mediated potentiation likely underlies a weaker tonic inhibition, facilitating a more flexible thermogenic response. More detailed work is needed to confirm this.

Another controlling factor for the sex difference in phenotypes with a functional interaction with NPVF signalling could be leptin. In females, leptin is a critical driver of thermogenesis, and anatomical studies confirm an overlap between leptin receptor (LepR)-expressing neurons and NPVF neurons within the DMH [11]. This colocalisation supports a model where NPVF acts as a downstream mediator or parallel modulator of leptin’s effects on energy expenditure. In wild-type females, NPVF may provide a tonic inhibitory brake that restrains leptin’s stimulatory drive on sympathetic thermogenic pathways. This could create a balanced, homeostatic tone [17,52]. Consequently, the genetic removal of this brake in Npvf-deficient females may lead to a disinhibited, hyper-thermogenic phenotype, consistent with our experimental observations. This hypothesis is supported by the established, sexually dimorphic role of leptin in driving female thermogenesis [53] and the anatomical overlap of LepR-expressing neurons in the DMH [13,54], a site of Npvf expression.

In addition, thyrotropin-releasing hormone (TRH) neurons in the paraventricular nucleus (PVH), which are principal controllers of basal metabolic rate and thermogenesis, could be differently influenced by NPVF signalling in males and females. This is supported by the known projections from the DMH to TRH neurons in the PVH, whereby sex-specific inhibitory inputs from DMH Npvf neurons could tonically restrain TRH release, particularly in females. The ablation of Npvf in knockout females would disinhibit the PVH, leading to increased TRH drive, elevated thyroid hormone levels, and a consequent hypermetabolic state—directly contributing to the observed lean phenotype [55,56].

Overall, our study highlights that the physiological role of NPVF cannot be defined in isolation; it is shaped by external cues such as ambient temperature and/or dietary composition. For example, the female-specific lean and thermogenic phenotype was unmasked only at thermoneutrality, illustrating how chronic mild cold stress at room temperature can mask intrinsic metabolic regulation. Similarly, the sex-divergent glucose/insulin phenotypes were most pronounced at TN, linking thermal environment to pancreatic function via NPVF. Furthermore, the functional role of NPVF shifted dramatically when on the HFD. The sex-specific effects on adiposity and water intake observed on chow were overridden, and a new, sex-consistent phenotype emerged: hyperphagia and increased lean mass gain in both male and female knockouts. This suggests that under the potent metabolic pressure of a HFD, NPVF’s orexigenic drive becomes dominant and generalised. The loss of the female-specific hypodipsia on the HFD further indicates that nutrient-sensing pathways can override NPVF’s role in fluid balance. However, like with all germline knockout models, developmental changes that may compensate for some of the sex-specific effects cannot be excluded.

5. Conclusions

Taken together, these findings support a model where NPVF signalling acts as a context-sensitive integrator, coordinating responses across energy balance, glucose metabolism, and hydromineral homeostasis in a sex-specific manner. In females, it appears to favour a more “thrifty” phenotype under standard conditions, promoting energy storage (fat), restraining energy expenditure (thermogenesis), and modulating bone and fluid balance, all while negatively modulating insulin sensitivity under metabolic challenge. In males, its role is more subtle under chow conditions, but becomes evident in promoting insulin action on the HFD and oxidative fuel selection during fasting.