Neuronal surface P antigen (NSPA) as a novel regulator of energy homeostasis and adipose tissue metabolism

Abstract

Background

Regulation of body weight and glucose homeostasis includes the coordinated activity of hypothalamic neurons and adipocytes within a neuroendocrine network whose dysfunctions underlie obesity, insulin resistance, and type 2 diabetes (T2D). The neuronal surface P antigen (NSPA) is a plasma membrane protein with characteristics of an E3 ubiquitin ligase encoded by the unique gene Zzef1, which has been linked to T2D. NSPA’s original discovery in neurons, as a cell-surface cross-reacting autoantigen of anti-P antibodies that associate with cognitive dysfunctions in patients with systemic lupus erythematosus, focused its study on hippocampal-mediated memory processes. Anti-P effects and NSPA-KO mice revealed that NSPA contributes to glutamatergic transmission and synaptic plasticity through mechanisms involving ubiquitylation processes coupled to the stability of NMDAR at the synaptic density. However, NSPA is also expressed in hypothalamic neurons, where glutamatergic synapses and NMDAR function are pivotal to the neuroendocrine control of metabolic and energy balance. Transcriptomics suggests an extended expression of NSPA in metabolically relevant peripheral tissues, including the adipose tissue. Here, we investigated whether body weight regulation and energy homeostasis involve NSPA.

Methods

We characterized the phenotype of NSPA-KO mice under standard chow and high-fat/high-sugar (HFHS) obesogenic diet conditions, monitoring metabolic parameters and WAT’s expression of enzymes and transporters of the glucose metabolism and lipolysis.

Results

NSPA-KO mice exhibit: (i) Increased body weight gain, despite similar food intake and higher horizontal locomotion activity; (ii) A shift towards a glycolytic metabolism reflected in an increased RER, accompanied by an increased WAT mass indicating higher lipogenesis; (iii) Improved early glycemic response to glucose challenge, attenuating the acute glycemic rise induced by HFHS feeding; (iv) Reduced insulin sensitivity at 20 weeks of age; (v) Elevated Glut1 and LDH, with decreased Glut4, and HSL S563 phosphorylation in WAT, indicating altered glucose uptake, glycolysis and lipolysis; (vi) Decreased levels of phosphorylated STAT3 in the hypothalamus, suggesting attenuated leptin signaling.

Conclusions

This study identifies NSPA as a novel regulator of energy homeostasis, body weight, glucose metabolism, insulin sensitivity, and adipose tissue accumulation, presumably acting at both the hypothalamus and WAT, with potential implications for obesity and metabolic disorders.

Supplementary Information

The online version contains supplementary material available at 10.1186/s10020-026-01463-y.

Background

The mechanisms that regulate body weight and glucose homeostasis are tightly interconnected, and their dysfunction underlies obesity and type II diabetes (T2D). These disorders represent major global health challenges and leading risk factors for chronic diseases that shorten life expectancy (Bray et al. 2018; Becetti et al. 2023; Morton et al. 2014). Identifying molecular determinants common to metabolic homeostasis and lipogenesis may help to elucidate their pathogenic mechanisms and define potential therapeutic targets. Given that hypothalamic neurons and adipocytes are central components of the integrated neuroendocrine network controlling energy balance (Bray et al. 2018; Becetti et al. 2023; Morton et al. 2014), their shared cell-surface proteins deserve particular attention as potential mediators linking central and peripheral regulation of metabolism.

The hypothalamus integrates neural and peripheral humoral signals reflecting nutrient status and energy stores, generating neuroendocrine and autonomic outputs that regulate feeding behavior, energy expenditure, and glucose metabolism in response to physiological needs (Belgardt and Brüning 2010; Timper and Brüning 2017). This highly dynamic system adapts to environmental and metabolic challenges, primarily through two hormones that convey energy status: insulin, secreted by pancreatic β-cells, and leptin, produced by adipocytes (Prentki et al. 2013; Woods and Seeley 2000). These signals are interpreted by the hypothalamus through orexigenic and anorexigenic neurons that secrete distinct neuropeptides and orchestrate a complex regulatory network (Bray et al. 2018; Becetti et al. 2023; Morton et al. 2014; Varela and Horvath 2012; Schwartz et al. 2000). In turn, white adipose tissue (WAT) functions not only as the principal site of lipid storage but also as an active endocrine organ that secretes adipokines and other signaling molecules that modulate appetite, insulin sensitivity, and systemic glucose utilization (Ahmadian et al. 2010; Minemura et al. 2022; Coelho et al. 2013). Bidirectional communication between hypothalamic neurons and adipocytes is therefore essential for maintaining energy balance and adapting to nutritional or hormonal challenges (Bray et al. 2018; Becetti et al. 2023; Morton et al. 2014). Disruption of this finely tuned hypothalamic–adipose axis contributes to obesity, insulin resistance, and T2D, conditions in which impaired hypothalamic function leads to defective control of food intake and energy expenditure, whereas adipocyte dysfunction results in ectopic lipid accumulation, low-grade inflammation, and altered glucose homeostasis (Timper and Brüning 2017; Varela and Horvath 2012). Understanding how specific molecules participate in the reciprocal regulation of neuronal and adipocyte functions may thus reveal key mechanisms underlying metabolic imbalance and its systemic consequences.

The neuronal surface P antigen (NSPA) is a large plasma membrane protein originally identified as a cross-reactive antigen recognized by anti-ribosomal P antibodies (anti-P) (Matus et al. 2007), a subset of autoantibodies found in 15–30% of patients with systemic lupus erythematosus (SLE) (Gonzalez and Massardo 2018; Viana et al. 2017). In the brain, NSPA is expressed in neurons across several regions and nuclei (Matus et al. 2007; Segovia-Miranda et al. 2015). Because anti-P autoantibodies have been associated with psychosis and cognitive dysfunction in neuropsychiatric SLE (NPSLE) (Gonzalez and Massardo 2018; Viana et al. 2017; Bonfa et al. 1987), NSPA expression in brain structures involved in memory, cognition, and emotion has drawn most attention, particularly in the hippocampus (Barake et al. 2022). Studies in mice expressing a truncated NSPA protein or lacking NSPA expression (NSPA-KO mice) revealed that hippocampal NSPA contributes to glutamatergic excitatory neurotransmission, synaptic plasticity, and spatial memory (Segovia-Miranda et al. 2015; Barake et al. 2022; Espinoza et al. 2020). NSPA deficiency mainly affects the synaptic function of N-methyl-D-aspartate receptors (NMDARs) (Espinoza et al. 2020). Notably, anti-P antibodies reproduce most NSPA-deficient phenotypes, suggesting that they inhibit NSPA function (Segovia-Miranda et al. 2015; Bravo-Zehnder et al. 2015; Díaz-Valdivia et al. 2025).

Considering that both NSPA (Matus et al. 2007; Segovia-Miranda et al. 2015) and NMDARs (Cull-Candy and Leszkiewicz 2004; Rao and Finkbeiner 2007) are widely distributed in the brain, NSPA likely participates in additional processes beyond memory and cognition. NSPA is expressed in the hypothalamus (Matus et al. 2007; Segovia-Miranda et al. 2015), where glutamatergic transmission mediated by NMDARs is essential for the neuroendocrine integration of metabolism and energy balance (Ameroso et al. 2022; Khan et al. 1999; Üner et al. 2015; Meeker et al. 1994a; Liu et al. 2012; Stanley et al. 1993, 1996). Given that NMDAR-dependent synaptic plasticity is critical for hypothalamic function (Liu et al. 2012; Stanley et al. 1993, 1996), it is important to experimentally evaluate the role of NSPA in hypothalamic control of metabolism and energy balance.

Interestingly, the unique Zzef1 gene encoding NSPA harbors single-nucleotide polymorphisms (SNPs) linked to type 2 diabetes (T2D) (Cull-Candy and Leszkiewicz 2004; Scott et al. 2017; Vujkovic et al. 2020), glucose metabolic traits (Palmer et al. 2014), visceral adiposity accumulation (Karlsson et al. 2019), and body fat percentage (Martin et al. 2021), as shown by genome-wide association studies (GWAS). Moreover, transcriptomic databases, including The Human Protein Atlas, indicate that Zzef1 is expressed in neuronal and metabolic tissues such as skeletal muscle, adipose tissue, liver, and pancreas (Uhlén et al. 2015). The NSPA protein contains two ZZ-type zinc-finger domains and an anaphase-promoting complex 10 (APC10) domain characteristic of E3 ubiquitin ligases (Matus et al. 2007; Segovia-Miranda et al. 2015). Together with biochemical evidence, data from synaptosomal fractions support a role for NSPA as an E3 ubiquitin ligase that modulates NMDAR stability at the postsynaptic density through the tyrosine phosphatase PTPMEG (also named PTPN4) (Barake et al. 2022; Espinoza et al. 2020). E3 ligases regulate protein degradation through the ubiquitin–proteasome system (UPS), which is crucial for glutamatergic synaptic transmission and plasticity (Mabb and Ehlers 2010; Yi and Ehlers 2005), as well as for diverse cellular processes including intracellular trafficking, signaling, and mitochondrial metabolism (Clague and Urbé 2025; May et al. 2021). Therefore, the expression of NSPA in both brain and peripheral tissues, together with its proposed function as a plasma membrane–associated E3 ubiquitin ligase, suggests that NSPA contributes to crosstalk between neuronal and peripheral mechanisms governing energy balance and glucose metabolism.

Here, we investigated the role of NSPA in body weight regulation and energy homeostasis by evaluating the metabolic phenotype of NSPA knockout (NSPA-KO) mice on chow and on a hypercaloric high-fat/high-sugar (HFHS) diet. We provide evidence that NSPA deficiency alters body weight gain, fuel selection, glucose homeostasis, and WAT accumulation. Moreover, the absence of NSPA affects the expression of key enzymes and transporters involved in lipolysis and glucose metabolism in adipose tissue, as well as the levels of phosphorylated STAT3 as a read out of leptin signaling, in the hypothalamus. These findings identify NSPA as a novel component of the regulatory system governing body weight and energy homeostasis under both physiological and obesogenic conditions.

Methods

Animals

Three-week-old C57BL/6 mice (WT) and C57BL/6Ntac Zzef1^{tm2.1(KOMP)vlcg} (here called NSPA-KO) mice from Pontificia Universidad Católica de Chile were transported to Centro de Biología Celular y Biomedicina (CEBICEM) of Universidad San Sebastián, where they were housed in a room with 12/12 light:dark cycle and temperature control at 22ºC. At the end of protocols, mice were anesthetized with an intraperitoneal injection of 200 mg/kg Avertin followed by cervical dislocation and decapitation. Litter size in WT and NSPA-KO mice was similar over time and did not affect body weight at weaning (Table 1). WT and NSPA-KO mice were derived from the same colony; therefore, they shared the same genetic background. Animals were not randomized prior to assignment, as groups were determined by genotype, and blinding was not feasible. All protocols involving rodents were approved by the Ethical Scientific Committee of Universidad San Sebastián, protocol number CEC 07–2021-10.

Table 1.

WT and NSPA-KO mice litter size, average body weight at 3 weeks (weaning) and 21 weeks (end of protocol), and average food intake (FI) for 18 weeks

	WT				NSPA-KO
Litter size (n)		5	6	7	4	6	7
Average BW g (3w)		16,97 ± 0,56	16,44 ± 1,86	19,23 ± 0,48	12,76 ± 1,39	12,69 ± 1,10	13,38 ± 0,66
Average BW g (21w)		26,12 ± 1,29	26,18 ± 2,23	27,29 ± 1,83	27,99 ± 1,79	28,73 ± 0,38	27,76 ± 0,49
Average FI g		3,77 ± 0,79	4,23 ± 0,73	3,09 ± 0,40	4,99 ± 0,50	3,84 ± 0,76	3,95 ± 0,18

WT and NSPA-KO mice litter size, average body weight at 3 weeks (weaning) and 21 weeks (end of protocol), and average food intake (FI) for 18 weeks

Lac-Z knockout mice and genotyping

C57BL/6Ntac Zzef1^{tm2.1(KOMP)vlcg} mice were engineered in Regeneron Pharmaceuticals Inc., New York, using Velocigene technology (Valenzuela et al. 2003), replacing the entire coding region of the mouse Zzef1 gene (128 kb) with ZEN-UB1 Cassette containing the LacZ gene that encodes β-galactosidase. Details are available at the Velocigene website (http://www.velocigene.com/komp/detail/10007), and the phenotypic characterization of hippocampal function (Segovia-Miranda et al. 2015; Espinoza et al. 2020).

Diets

From weaning, 3-week-old WT and NSPA-KO mice were fed with either Prolab® RMH 3000 5P00 (Chow) diet composed by 26,1% protein, 59,5% carbohydrates and 14,4% lipids, with an energy density of 4.17 kcal/g or 45% fat diet (Research Diets, D12451) composed by 20% proteins, 35% carbohydrates and 45% lipids, with an energy density of 4.73 kcal/g supplemented with 2,5% sucrose in water from now on High-Fat/High-Sugar diet (HFHS).

Body weight

Mouse body weight was measured weekly using a mouse scale (Accuris Instruments, Model W3300-500).

Individual food intake and locomotor activity

Individual food intake and locomotor activity were measured in 10-, 15- and 20-weeks-old mice, for 12 h in the dark cycle, using a Ugo Basile Metabolic Cage (Ugo Basile S.R.L. Model 41,800–010). This metabolic cage has a feeding chamber that informs about the daily food and water consumption of each mouse, and records spontaneous motor activity and its variation over time.

Respiratory exchange ratio (RER)

The RER of 10-, 15- and 20-weeks-old mice was determined by measuring oxygen consumption (O₂) and carbon dioxide (CO₂) production. To that end, mice were individually placed into a metabolic chamber (iWorx TA, Inc.), and metabolic data were measured with a GA-200 CO₂/O₂ analyzer. Gas sensors were calibrated before experiments, and the VCO₂/VO₂ ratio was reported using LabScribe SKU-LS-METABOLIC-ANALYSIS. RER is defined as VCO₂ produced divided by VO₂ consumed.

Intraperitoneal glucose tolerance test (ipGTT)

Intraperitoneal glucose tolerance test was performed in 10-, 15- and 20-week-old WT and NSPA-KO mice. In mice fasted for 6 h, basal blood glucose was measured (time 0), and then mice were intraperitoneally injected with 2 g/kg of 20% D( +) glucose anhydrous (Winkler BM-0830) dissolved in 0.9% NaCl solution (CS-PISA). Blood glucose was recorded at 15, 30, 90 and 120 min.

Intraperitoneal insulin tolerance test (ipITT)

Intraperitoneal insulin tolerance test was performed in 10-, 15-, and 20-week-old WT and NSPA-KO mice. For this test, insulin syringes (Humulin Insulina Humana®) were prepared at a concentration of 0.5 IU/kg in 0.9% NaCl solution (CS-PISA). The mice were fasted for 3.5 h, basal blood glucose was measured (time 0), and then the mice were intraperitoneally injected with 0.5 IU/kg of insulin. Blood glucose was recorded at 15, 30, 90, and 120 min.

Extraction of adipose tissue and hypothalamus proteins for immunoblotting

After euthanasia, eWAT, and iWAT were collected and added to the glass homogenizer with 800 µl for eWAT, 500 µl for iWAT of RIPA buffer (Pierce, Thermo Fisher Scientific) with protease and phosphatase inhibitors (Pierce, Thermo Fisher Scientific). The tissue was ground until completely homogenized. Homogenates were centrifuged at 15,000xg for 15 min at 4 °C. The supernatant was removed and incubated with chloroform in a 1:1 ratio. The tubes were centrifuged at 15,000xg for 15 min at 4 °C and the upper phase was stored for quantification. After euthanasia, hypothalami were dissected on ice and homogenized in homogenization buffer (0.32 M sucrose, 0.5 mM EGTA, 5 mM Hepes, pH 7.4) supplemented with protease and phosphatase inhibitors (Pierce, Thermo Fisher Scientific) using a Potter homogenizer. Homogenates were centrifuged twice at 1,000 × g for 10 min at 4ºC. The supernatant was collected for immunoblotting. Protein concentrations were determined using the BCA assay (Pierce, Thermo Fisher Scientific). For immunoblots, 10% polyacrylamide gels (AccuGel 29:1 EC-851 National Diagnostic) were prepared. The samples were heated at 60 °C for 5 min and then 30–40 µg of proteins were loaded per well. The wet transfer was at 450 mA for 1.5 h. To block the membrane, 5% BSA in TBS 0.1% Tween20 was used for 1 h while at room temperature. The membrane was incubated with primary antibody in 1% BSA with TBS 0.1% Tween20 overnight at 4 °C. The membrane was washed 3 times for 10 min with TBS 0.1% Tween20 and then incubated with a secondary antibody for 1 h at room temperature. Primary antibodies used were rabbit anti-ZZEF1 (1:1000, Abcam ab1746529), mouse anti-PFK-1 (1:1000, Santa Cruz sc-377346), mouse anti-LDH (1:1000, Santa Cruz sc-133123), mouse anti-α Tubulin (1:3000, Invitrogen #32–2500), rabbit anti-HSL (1:1000, Invitrogen #PA5-17,196), rabbit anti-phospho-HSL S563 (1:1000, Cell Signaling #4139P), mouse anti-Vinculin (1:1000, Santa Cruz sc-73614), mouse anti-Glut4 (1:1000, Santa Cruz sc-53566), rabbit anti-Glut1 (1:1000, Alpha Diagnostic GT12A), mouse anti-STAT3 (1:1000, Cell Signaling #9139), rabbit anti-phospho-STAT3 (1:1000, Cell Signaling #9145), mouse anti-β-actin (1:3000, Abcam ab6276). Secondary antibodies used were anti-mouse HRP (1:3000, Cell Signaling #7076) and anti-rabbit HRP (1:3000, Cell Signaling #7074S).

Body fat composition

To evaluate the body fat composition of NSPA-KO and WT mice, after euthanasia, interscapular brown adipose tissue (BAT), inguinal (subcutaneous) white adipose tissue (iWAT), and epidydimal (visceral) white adipose tissue (eWAT) was extracted and weighed.

Histology

A portion of each fat depot (eWAT and iWAT) was fixed in Bouin solution for 24 h at 4ºC and dehydrated in a series of ethanol washes. Samples were cleared in xylene and mounted in paraffin. Sections of 5 and 10 μm thickness were cut and stained with hematoxylin and eosin to analyze cellular morphology.

Statistical analysis

The software GraphPad PRISM Version 10.2.0 (San Diego, CA) was used for statistical analysis. Data are presented as mean ± SEM values and differences were analyzed with unpaired t-test, one- or two-way ANOVA followed by Sídák’s multiple comparisons test, as indicated in the corresponding figure. Statistical significances correspond to *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Results

To investigate the role of NSPA in body weight balance and energy homeostasis we evaluated the metabolic phenotype of NSPA-KO mice under a standard chow diet or challenged with a hypercaloric high-fat/high-sugar (HFHS) obesogenic diet.

NSPA deficiency alters energy balance in chow-fed mice

We first evaluated the NSPA contribution to energy homeostasis by comparing body weight in WT and NSPA-KO mice weekly from weaning. NSPA-KO mice fed with a chow diet exhibited greater body weight gain over time (Fig. 1a) and cumulative weight gain at 21 weeks of age (Fig. 1b). Remarkably, NSPA-KO mice starting body weight after weaning at 3 weeks of age was significantly lower than WT (Fig.S1a). However, because they gained more weight, both groups reached approximately the same body weight by 21 weeks. Despite this increase in body weight gain, the individual food intake during 12 h in the dark cycle showed no significant differences (Fig. 1c). Moreover, energy intake during the 12-h dark cycle in chow-fed mice, expressed as kilocalories normalized by body weight was similar between both groups at all ages evaluated (Fig. 1d). Notably, NSPA-KO mice displayed a greater horizontal locomotor activity compared to WT mice at 10 and 15, but not at 20, weeks of age (Fig. 1e).

Fig. 1.

NSPA-KO mice fed a chow diet exhibit altered energy balance. A Weekly body weight gain. B Accumulative body weight gain at week 21 of WT and NSPA-KO mice fed with chow diet. C Individual food intake during the dark phase (12 h) of 10-, 15- and 20-week-old WT and NSPA-KO mice fed with chow diet. D Energy intake during the 12-h dark cycle in mice fed chow diet. Food consumption was measured over the dark phase and expressed as kilocalories normalized to body weight (kcal/g body weight). E 12 h total horizontal locomotor activity of 10-, 15- and 20-week-old WT and NSPA-KO mice fed with chow diet. F RER of 10-, 15- and 20-week-old WT and NSPA-KO mice fed with chow diet. WT n = 6–12; KO n = 7–14 biologically independent animals. Data are presented as mean ± SEM values and differences were analyzed with two-way ANOVA (A, C, D, E and F) or with one-way ANOVA (B), followed by Sídák’s multiple comparisons, test statistical significances correspond to *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001

We next assessed the ratio between CO₂ production (VCO₂) and O₂ consumption (VO₂) corresponding to the respiratory exchange ratio (RER) (Bar et al. 2020). Under normal conditions, RER values range from 0.7 to 1.0, with values closer to 0.7 indicating predominant lipid oxidation, while those nearing 1.0 reflect increased carbohydrate utilization (Simonson and DeFronzo 1990). A value of approximately 0.85 is expected for subjects that have balanced reliance on both carbohydrates and fat as energy sources on a mixed diet (Lusk 1924). NSPA-KO mice exhibited a significantly elevated RER approaching 1.0, thus indicating a shift toward carbohydrate metabolism (Fig. 1f). This can explain the increased body weight gain of NSPA-KO mice, despite their higher locomotor activity and similar feeding behavior compared to WT controls. All these results reveal an altered energy balance of NSPA-KO mice, preferentially relying on carbohydrates rather than lipids as their primary energy source (Simonson and DeFronzo 1990).

An obesogenic diet triggers similar metabolic alterations in WT and NSPA-KO mice

When challenged with an HFHS diet, both WT and NSPA-KO mice exhibited proportional body weight gain and cumulative body weight gain at 21 weeks of age, relative to their counterparts on a chow diet (Fig. 2a and b), with similar food intake at 10, 15, and 20 weeks of age and energy intake during the 12-h dark cycle (Fig. 2c and d). Therefore, in chow diet conditions, NSPA-KO mice had 30% more weight gain than WT, thus reaching levels close to WT under HFHS diet (Fig. 2a). Similar to what we observed in the chow diet-fed groups, NSPA-KO mice fed an HFHS had a significantly lower body weight than WT mice at weaning (Fig.S1a). These results highlight the role of NSPA in maintaining body weight under physiologic conditions.

Fig. 2.

NSPA-KO and WT mice exhibit similar metabolic alterations when challenged with an obesogenic diet. A Weekly body weight gain. B Accumulative body weight gain at week 21 of WT and NSPA-KO mice fed either with chow or HFHS diet. C Individual food intake during the dark phase (12 h) of 10-, 15- and 20-week-old WT and NSPA-KO mice fed with HFHS diet. D Energy intake during the 12-h dark cycle in mice fed HFHS diet. Food consumption was measured over the dark phase and expressed as kilocalories normalized to body weight (kcal/g body weight). E 12 h total horizontal locomotor activity of 10-, 15- and 20-week-old WT and NSPA-KO mice fed with HFHS diet. F RER of 10-, 15- and 20-week-old WT and NSPA-KO mice fed with HFHS diet. WT n = 7; KO n = 11; WT HFHS n = 4–8; KO HFHS n = 4–6 biologically independent animals. Data are presented as mean ± SEM values and differences were analyzed with two-way ANOVA (A, C, D, E and F) or with one-way ANOVA (B), followed by Sídák’s multiple comparisons test, statistical significances correspond to *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001

Under the HFHS diet, horizontal locomotor activity now showed no differences between groups due to an increase in WT mice and a decrease in NSPA-KO mice (Fig. 2e). RER is known to decrease over time under the HFHS diet (Marvyn et al. 2016). As expected, this effect occurred in WT mice, whereas NSPA-KO mice showed an increased RER over time, reaching significance at 20 weeks of age (Fig. 2f). NSPA deficiency seems to impede the compensatory metabolic changes induced by an obesogenic diet, maintaining a glycolytic rather than a lipolytic phenotype.

A hypercaloric diet initially reduces food intake through homeostatic feeding to maintain energy balance; however, prolonged consumption of palatable diets, such as the HFHS diet, can override homeostatic satiety signals, promoting hedonic eating and overeating (Lutter and Nestler 2009; Morales 2022). Food intake of 10 and 15-week-old WT and NSPA-KO mice was lower under the HFHS diet compared with chow-fed counterparts, but at 20 weeks of age only NSPA-KO mice still had lower food intake, with WT mice reaching similar levels of food intake under both diets (Fig.S2a and S3a). Therefore, contrasting with WT controls, the energy balance of NSPA mice is not affected by an obesogenic diet.

NSPA deficiency attenuates HFHS-induced impairment in glycemic control

The preceding results support a role for NSPA in maintaining energy homeostasis. To further explore the physiological mechanisms underlying this function, we evaluated glucose metabolism under chow feeding and after the metabolic challenge of the HFHS diet. Basal glycemia after 6 h of fasting did not differ among groups, indicating that the period of HFHS feeding did not induce overt hyperglycemia. However, under a chow diet, intra-peritoneal glucose tolerance test (ipGTT) performed in 10-, 15-, and 20-week-old mice, revealed an attenuated glycemic response to a glucose loading in NSPA-KO mice (Fig. 3a, c, and e). This suggests higher circulating insulin levels that maintain glucose levels lower than those in WT controls. However, the area under the curve (AUC) for glycemia was not significantly different among groups (Fig. 3b, d, and f), suggesting that this effect of NSPA deficiency is transient.

Fig. 3.

The absence of NSPA attenuates the acute glycemic rise induced by HFHS diet. A intraperitoneal Glucose tolerance test (ipGTT) of 10-week-old WT and NSPA-KO mice fed with chow or HFHS diet. B AUC of A. C ipGTT of 15-week-old WT and NSPA-KO mice fed with chow or HFHS diet. D Area under the curve (AUC) of C. E ipGTT of 21-week-old WT and NSPA-KO mice fed with chow or HFHS diet. F AUC of E. WT n = 7–10; KO n = 11; WT HFHS n = 7–8; KO HFHS n = 5–6 biologically independent animals. Data are presented as mean ± SEM values and differences were analyzed with two-way ANOVA (A, C, and E), and one-way ANOVA test (B, D, and F), followed by Sídák’s multiple comparisons. Statistical significances correspond to *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001

The HFHS diet consistently leads to elevated blood glucose levels following a glucose load (Fleur et al. 2011; Moreno-Fernández et al. 2018). This effect was indeed seen in HFHS-fed WT controls compared with chow-fed WT mice at all ages assessed (Fig. 3a, c, and e), a pattern also reflected in AUC measurements (Fig. 3b, d and f). NSPA-KO mice fed an HFHS diet also showed higher blood glucose levels than their chow-fed counterparts, but only at 10 and 15 weeks of age. However, at 20 weeks, NSPA-KO mice showed lower glycemia than HFHS-fed WT controls under glucose load (Fig. 3e and f). This indicates that NSPA deficiency allowed better control of glycemia in response to glucose load at 20 weeks.

Considering the difference between WT and NSPA-KO mice in the glycemic control under chow and HFHS feeding, we next assessed insulin sensitivity across groups at various ages. After a 3 h fast, basal glycemia did not differ between NSPA-KO and WT mice under either dietary condition (Fig. 4a, c, and e). Insulin tolerance tests revealed no significant differences in glycemic responses between genotypes at 10 and 15 weeks of age, regardless of diet (Fig. 4a-d). As expected, WT mice fed an HFHS diet displayed insulin resistance at 20-week-old (Fig. 4e and f). In contrast, at 20 weeks, the NSPA-KO mice showed reduced insulin sensitivity already under a chow diet compared with WT mice. This insulin resistance did not increase further under HFHS feeding (Fig. 4e and f). Thus, NSPA deficiency itself leads to insulin resistance.

Fig. 4.

NSPA-KO mice have reduced insulin sensitivity compared to chow-fed WT mice. A Intraperitoneal Insulin tolerance test (ipITT) of 10-week-old WT and NSPA-KO mice fed with chow or HFHS diet. B AUC of A. C ipITT of 15-week-old WT and NSPA-KO mice fed with chow or HFHS diet. D AUC of C. E ipITT of 20-week-old WT and NSPA-KO mice fed with chow or HFHS diet. F AUC of E. WT n = 7–10; KO n = 10–11; WT HFHS n = 7–8; KO HFHS n = 5–6 biologically independent animals. Data are presented as mean ± SEM values and differences were analyzed with two-way ANOVA (A, C, and E), and one-way ANOVA (B, D, and F), followed by Sídák’s multiple comparisons test. Statistical significances correspond to *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001

Leptin also regulates glucose homeostasis among its various metabolic functions (Lavoie et al. 2023; Myers et al. 2021; Hoek et al. 2008), and defects in leptin signaling may contribute to insulin resistance (Myers et al. 2021; Hoek et al. 2008). The best-characterized pathway of leptin signaling is mediated by Tyr705 phosphorylation of signal transducer and activator of transcription 3 (STAT3) downstream of leptin receptor B and Janus tyrosine kinase 2 (JAK2) (Liu et al. 2021). Therefore, we analyzed hypothalamic extracts by Western blot and found decreased pSTAT3 levels in NSPA-KO mice (Fig. 5), suggesting an attenuated leptin signaling pathway.

Fig. 5.

NSPA-KO mice showed reduced hypothalamic STAT3 phosphorylation. A Western blotting of pSTAT3 and total STAT3 in hypothalamus from 21-week-old WT and NSPA-KO mice. B Densitometry quantification of pSTAT3/STAT3. WT n = 6; KO n = 6 biologically independent samples. Data are presented as mean ± SEM values and differences were analyzed with unpaired t-test, statistical significances correspond to *P ≤ 0.05

The absence of NSPA increases the mass of white adipose tissue

Given that NSPA-KO mice exhibited enhanced body weight gain along with increased RER, suggesting higher adiposity, we analyzed body fat composition in 21-week-old NSPA-KO and WT mice under chow or HFHS diets. We measured subcutaneous inguinal white adipose tissue (iWAT), visceral epididymal white adipose tissue (eWAT), and interscapular brown adipose tissue (BAT). By calculating the percentage of total body weight comprised by adipose tissue, we found that NSPA-KO mice had increased adiposity, primarily due to WAT. Both iWAT and eWAT depots were significantly larger in NSPA-KO than in WT controls, whereas BAT mass was unchanged (Fig. 6a). Representative images of the extracted fat depots are shown in Fig. 6b. Consistent with gross appearance, hematoxylin and eosin staining demonstrated a modest increase in adipocyte size in NSPA-KO mice relative to WT controls in the chow diet (Fig. 6d). As expected, HFHS-feeding increased adiposity in both WT and NSPA-KO mice (Fig. 6c). These findings indicate that NSPA deficiency promotes increased adiposity under physiological dietary conditions but does not exacerbate fat accumulation over the WT mice under an HFHS diet.

Fig. 6.

NSPA-KO mice have increased mass of white adipose tissue. A Adipose tissue percentage of WT and NSPA-KO mice fed with chow diet. Total adipose tissue (AT). White adipose tissue (WAT). Inguinal white adipose tissue (iWAT). Epididymal white adipose tissue (eWAT). Interscapular brown adipose tissue (BAT). B Representative images of endpoint fat-pad obtained from 21-week-old WT and NSPA-KO male mice. eWAT, iWAT, and BAT depots are shown. C Adipose tissue percentage of WT and NSPA-KO mice fed with HFHS diet, AT, WAT, iWAT, eWAT. BAT. D Representative images of hematoxylin and eosin-stained eWAT and iWAT from 21-week-old WT and NSPA-KO male mice. Scale bar in 10X augment 200 μm and in 40X augment 50 μm. WT n = 7–8; KO n = 7–8; WT HFHS n = 5; KO HFHS n = 4 biologically independent animals (A and C). Data are presented as mean ± SEM values and differences were analyzed with unpaired t-test, statistical significances correspond to **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001

The absence of NSPA alters the levels of proteins involved in WAT metabolism in mice

The increase in white adipose tissue of chow-fed NSPA-KO mice suggests metabolic alterations that may disrupt energy homeostasis. To investigate possible underlying mechanisms, we analyzed RER at an early age, before the onset of increased body weight (Fig.S1a). At weaning, NSPA-KO mice had lower body weight compared with WT controls (Fig. 7a, Fig. S1a), a difference not attributable to litter size (Table 1). However, during the first four weeks after weaning, NSPA-KO mice exhibited accelerated weight gain relative to WT mice (Fig. 7b). This early phenotype was accompanied by a higher RER at 7 weeks of age (Fig. 7c), indicating a shift toward a glycolytic metabolism, which likely contributes to the subsequent increase in adiposity. At 7 weeks of age, chow-fed NSPA-KO mice still did not differ from WT controls in body weight or body fat composition (Fig. 7d and e). Thus, the increased adiposity observed in adult NSPA-KO mice is preceded by a shift toward reduced oxidative metabolism, as indicated by the elevated RER, which likely contributes to subsequent fat accumulation.

Fig. 7.

NSPA-KO mice show a decreased oxidative systemic metabolism before body fat accumulation. A Weekly body weight. B Weekly body weight gain. C RER of 7-week-old WT and NSPA-KO mice fed with chow diet. D Body fat composition of 7-week-old WT and NSPA-KO mice fed with chow diet, AT, WAT, and BAT. E iWAT and eWAT of 7-week-old WT and NSPA-KO mice fed with chow diet. WT n = 6; KO n = 4–6 biologically independent animals. Data are presented as mean ± SEM values and differences were analyzed with two-way ANOVA, followed by Sídák’s multiple comparisons test (A and B) and unpaired t-test (C), statistical significances correspond to ***P ≤ 0.001, ****P ≤ 0.0001

To investigate how the absence of NSPA contributes to this altered metabolic profile, whether by reducing oxidative metabolism and/or enhancing glycolytic pathways, we examined the levels of proteins involved in glucose metabolism and lipolysis in the adipose tissue. We first assessed whether NSPA is expressed in adipose tissue, as suggested by transcript data in The Human Protein Atlas (Uhlén et al. 2015). Immunoblot with the available commercial antibody revealed NSPA in eWAT and iWAT of WT mice (Fig. 8). Adipocytes predominantly express two glucose transporters, Glut1 and Glut4 (Chadt and Al-Hasani 2020). Glut1 mediates basal insulin-independent glucose uptake (Medina and Owen 2002), while Glut4 mediates insulin-stimulated glucose transport into adipocytes (Kanzaki and Pessin 2001). WAT from 7-week-old NSPA-KO mice showed increased Glut1 and decreased Glut4 levels compared to WT mice (Fig. 9a-d). These results suggest that basal glucose uptake in adipose tissue may be enhanced through increased Glut1, whereas insulin-stimulated glucose transport via Glut4 may be impaired in the absence of NSPA, potentially contributing to the systemic alterations in energy metabolism.

Fig. 8.

NSPA expression in mice eWAT and iWAT. Immunoblot to detect NSPA with a commercial antibody against ZZEF1. We used WT and NSPA-KO mice tissue samples to detect NSPA in adipose tissue

Fig. 9.

NSPA-KO mice have altered levels of proteins involved in WAT metabolism. A Western blotting of pHSL S563, total HSL, Glut1 and Glut4 in eWAT from 7-week-old WT and NSPA-KO mice. B Densitometry quantification of pHSL/HSL, Glut1 and Glut4 levels by Western blotting. C Western blotting of pHSL S563, total HSL, Glut1 and Glut4 in iWAT from WT and NSPA-KO mice. D Densitometry quantification of pHSL/HSL, Glut1 and Glut4 levels by Western blotting. E Western blotting of PFK-1 and LDH in eWAT from WT and NSPA-KO mice. F Densitometry quantification of PKF-1 and LDH levels by Western blotting. G Western blotting of PFK-1 and LDH in iWAT from WT and NSPA-KO mice. H Densitometry quantification of PKF-1 and LDH levels by Western blotting. WT n = 6; KO n = 4–6 biologically independent samples. Data are presented as mean ± SEM values and differences were analyzed with unpaired t-test, statistical significances correspond to *P ≤ 0.05 and **P ≤ 0.01

Next, we compared the expression of key enzymes involved in glucose metabolism in the WAT of NSPA-KO and WT mice. We examined phosphofructokinase-1 (PFK-1), a major regulatory enzyme of glycolysis that catalyzes ATP-dependent conversion of fructose-6-phosphate to fructose 1,6-bisphosphate and ADP (Wang et al. 2024), and lactate dehydrogenase (LDH), which converts pyruvate to lactate (Valvona et al. 2016). Immunoblot analyses revealed no significant differences in PFK-1 levels between genotypes, whereas NSPA-KO mice showed increased LDH levels (Fig. 9e-h). On the other hand, we also evaluated hormone-sensitive lipase (HSL), a key enzyme in lipolysis whose activation requires phosphorylation by protein kinase A (PKA) in serine 563 (S563) (Fortier et al. 2005). WAT of NSPA-KO mice revealed decreased levels of HSL S563 phosphorylation compared with WT controls (Fig. 9a-d), indicating reduced HSL activity and impaired lipolysis.

Discussion

Since its discovery as a neuronal surface protein (Matus et al. 2007), NSPA has been studied mainly in the context of the pathogenic role of anti-P autoantibodies in NP-SLE (Gonzalez and Massardo 2018; Schwartz et al. 2019). The association of anti-P with lupus cognitive dysfunction (Massardo et al. 2015) has focused prior studies on the hippocampus, where NSPA has been found to contribute to glutamatergic synaptic transmission and memory-related processes (Segovia-Miranda et al. 2015; Barake et al. 2022; Espinoza et al. 2020; Bravo-Zehnder et al. 2015). However, NSPA is also widely expressed in neurons of several regions in the brain, including the hypothalamus, where glutamatergic transmission contributes to the neuroendocrine integration of signals controlling energy homeostasis and food intake (Ameroso et al. 2022; Khan et al. 1999; Üner et al. 2015; Meeker et al. 1994a, 1994b). In addition, transcript data from the Human Protein Atlas suggest that NSPA may be expressed in metabolic tissues such as skeletal muscle, adipose tissue, pancreas, and liver (Uhlén et al. 2015). Furthermore, analysis of SNPs links NSPA-encoding Zzef1 gene with visceral adiposity accumulation and increased body fat percentage (Karlsson et al. 2019; Martin et al. 2021), while genome-wide association studies (GWAS) identified Zzef1 polymorphisms associated with altered glucose metabolism and increased risk of T2D (Scott et al. 2017; Suzuki et al. 2019; Vujkovic et al. 2020; Palmer et al. 2014). Here, we investigated the metabolic phenotype of NSPA-KO mice and provided the first experimental evidence for NSPA function in regulating systemic energy balance, with clear effects on body weight, glucose metabolism, insulin sensitivity, and adipose tissue accumulation. Thus, NSPA emerges as a potential modulator of metabolic health, with potential implications in obesity and metabolic disorders.

We first show that NSPA-KO mice fed chow diet gain body weight more rapidly than WT controls, despite showing no differences in food intake and exhibiting increased locomotor activity. These findings suggest that the absence of NSPA alters energy utilization rather than caloric intake in the absence of NSPA expression. The elevated RER, approaching 1.0 in NSPA-KO mice, indicates a shift toward glycolytic metabolism in the absence of NSPA, which is intrinsically less efficient than oxidative metabolism. Consistent with this profile, NSPA-KO mice have a higher percentage of total adipose tissue primarily due to increased subcutaneous inguinal and visceral epididymal WAT, moreover, hematoxylin and eosin staining revealed a slight increase in adipocyte size, particularly in eWAT, where adipocytes appear enlarged in NSPA-KO mice compared to WT.

Adipose tissue generally accumulates when energy intake exceeds expenditure (Blüher 2019). However, we found that NSPA-KO and WT mice consume similar calories. Therefore, the increased adiposity of NSPA-KO mice, which contributes to higher body weight gain, most likely reflects reduced energy expenditure. Future experiments should address this possibility by directly measuring energy expenditure using indirect calorimetry (Banks et al. 2025).

We next investigated the impact of NSPA absence on the metabolic response to an obesogenic diet. Obesity and overweight are major global health challenges, and understanding their cellular mechanisms is critical for clarifying their pathogenesis. Under a widely validated high-fat/high-sugar diet (de Moura e Dias, et al. 2021; Speakman 2019), both WT and NSPA-KO mice show comparable proportional body weight gain, with similar food intake and horizontal locomotor activity. However, as mentioned, NSPA-KO mice already exhibit increased basal weight gain on chow. These observations suggest that NSPA normally contributes to body weight regulation independently of dietary composition. NSPA-deficient mice most likely have intrinsic metabolic alterations.

Homeostatic feeding maintains proper nutrition by adjusting hunger and satiety in response to the body's internal energy demands. As a result, food intake typically decreases when consuming a hypercaloric diet (Saper et al. 2002). In contrast, non-homeostatic feeding is influenced by the rewarding and motivational properties of food, largely mediated by dopamine release in limbic brain regions, particularly in response to high-fat, high-sugar (HFHS) diets (Dallman et al. 2005; Narayanan et al. 2010). When given an HFHS diet, WT mice showed a significant reduction in food intake at 10 and 15 weeks of age. However, by 20 weeks, their intake was comparable to that of mice on a chow diet. This pattern was not observed in NSPA-KO mice, which showed consistently reduced food intake at all time points assessed.

Prolonged HFHS feeding reduces RER over time due to a metabolic shift toward lipid utilization and a more oxidative profile (Marvyn et al. 2016). In contrast, RER in NSPA-KO mice increases progressively and reaches significantly higher levels at 20 weeks of age. Although at 21 weeks of age, when we euthanized the mice, we found no differences in the percentage of WAT and brown adipose tissue between NSPA-KO and WT, it seems plausible that phenotype differences in adiposity may have arisen at longer time periods. The RER values strongly suggest that the absence of NSPA results in sustained reliance on carbohydrate metabolism, even with a lipid-rich diet. NSPA deficiency likely compromises metabolic balance, impairing the adaptive transition toward lipid utilization in response to obesogenic challenges (Chiu et al. 2019; Fuller et al. 2020).

An obesogenic HFHS diet also promotes insulin resistance and glucose intolerance, thereby contributing to T2D (Lichtenstein and Schwab 2000). We found that NSPA absence does not alter basal glycemia under a chow or HFHS diet, but it does affect the response to glucose loading. Under a chow diet, NSPA-KO mice respond with a lower glycemic peak, although without differences in the overall AUC. This suggests a transient attenuation of hyperglycemia in the absence of NSPA. Under an HFHS diet, WT mice showed the expected reduction in glucose tolerance and insulin sensitivity compared with chow-fed controls. In contrast, NSPA-KO mice showed worsened glucose tolerance at 10 and 15 weeks but performed comparably to their chow-fed counterparts at 20 weeks. The HFHS diet does not exacerbate a pre-existing insulin resistance in NSPA-KO mice. These results support the view that NSPA contributes intrinsically to the metabolic balance.

To investigate the mechanisms underlying the increased white adiposity in NSPA-KO under chow diet conditions, we examined the expression of transporters and enzymes involved in glucose metabolism and lipolysis in adipose tissue, where glycolysis contributes to fat accumulation (Muñoz et al. 2010). We analyzed 7-week-old mice, the stage at which body weight shifts. Despite having a similar body fat composition to WT controls, at this stage NSPA-KO mice already showed greater weight gain and increased RER, indicating that their metabolic alterations, particularly in fuel source preference, precede measurable changes in adipose tissue accumulation. Altered expression of glycolytic enzymes and glucose transporters has been reported as an adaptive response to conditions such as fasting (Minemura et al. 2022), hypoxia (Nagao et al. 2019), and cancer (Yang et al. 2012). The expression of LDH, which catalyzes the conversion of pyruvate into lactate at the final step of glycolysis (Valvona et al. 2016), and Glut1, which mediates basal insulin-independent glucose uptake (Medina and Owen 2002), shares common regulatory mechanisms (Nagao et al. 2019; Yang et al. 2012). Previous studies have shown that LDH deficiency in mouse adipocytes reduces Glut1 protein levels, impairing glucose uptake in WAT (Minemura et al. 2022). On the other hand, Glut4, the transporter that mediates insulin-dependent glucose uptake (Kanzaki and Pessin 2001), is down-regulated in WAT as an early indicator of insulin resistance and T2D development (Shepherd and Kahn 1999). We found increased protein levels of LDH and Glut1, and decreased levels of Glut4 in the WAT of NSPA-KO mice. This pattern resembles the adipose tissue of patients with T2D (Garvey et al. 1991). The reduced Glut4 may contribute to the impaired insulin sensitivity, whereas increased LDH and Glut1 likely reflect a compensatory mechanism that enhances insulin-independent glucose uptake and promotes lipogenesis (Guo et al. 2012).

Fat accumulation can result from various processes, including impaired lipolysis, which catabolizes triacylglycerols (TAG) stored in lipid droplets and is critical for mobilizing energy reserves within adipocytes (Ahmadian et al. 2010; Lass et al. 2011; Zimmermann et al. 2009). Phosphorylation of the hormone-sensitive lipase (HSL), particularly at serine 563 (S563) by PKA, is required to promote lipolysis (Fortier et al. 2005; Greenberg et al. 2001). Our WAT analysis reveals decreased levels of pHSL S563 in NSPA-KO mice compared with WT controls. Although other lipases involved in TAG catabolism may also be affected and deserve further investigation, the reduced phosphorylation of HSL likely lowers its enzymatic activity, thereby diminishing lipolytic capacity and contributing to fat accumulation in the absence of NSPA expression. Indeed, the mechanisms by which NSPA may modulate WAT function remain to be determined as an important direction for future research.

NSPA expression has been detected in neurons across several brain regions, including the hippocampus and hypothalamus (Matus et al. 2007; Segovia-Miranda et al. 2015). In the hippocampus, NSPA deficiency reduces basal neuronal activity, impairs synaptic plasticity associated with memory processes, and decreases NMDAR levels at the postsynaptic density (Espinoza et al. 2020). Biochemical analyses of synaptosomal fractions support a role for NSPA as an E3 ubiquitin ligase involved in stabilizing NMDARs at the postsynaptic density (Barake et al. 2022; Espinoza et al. 2020). In the hypothalamus, β-galactosidase expression driven by the Zzef1 promoter and immunohistochemical analyses have detected NSPA expression in the paraventricular nucleus (PVN) (Segovia-Miranda et al. 2015), while single-cell transcriptomic datasets further suggest that NSPA may be expressed in additional hypothalamic neuronal populations. According to data from HypoMap single cell gene expression atlas of the murine hypothalamus (Steuernagel et al. 2022), Zzef1 mRNA is expressed in the arcuate nucleus (ARC), dorsomedial hypothalamic nucleus, paraventricular nucleus (PVN) and lateral hypothalamus (LH) (Fig. S4).

The hypothalamus plays a central role in energy balance and body weight regulation through the integrated activity of neurons located in discrete nuclei and their projections to multiple brain regions and the brainstem (Friedman 2019; Caron et al. 2018; Barbosa et al. 2023; Azevedo et al. 2019; Smith and Azevedo 2025). Traditionally, anorexigenic pro-opiomelanocortin (POMC) neurons and orexigenic agouti-related peptide/neuropeptide Y (AgRP/NPY) neurons located in the arcuate nucleus (ARC) have been considered the principal regulators of feeding behavior, acting in response to leptin as a peripheral signal of energy stores (Friedman 2019). Glutamatergic transmission is a critical component of hypothalamic circuits regulating feeding behavior and body weight (Üner et al. 2015; Liu et al. 2012; Stanley et al. 1993, 1996). Pharmacological activation of glutamate or NMDA receptors within the lateral hypothalamus induces robust feeding responses even in satiated animals (Stanley et al. 1993, 2011). Consistently, selective deletion of NMDARs from AgRP neurons, but not from POMC neurons, results in reduced body weight, fat mass, and food intake (Üner et al. 2015; Liu et al. 2012). In addition, an excitatory circuit from PVN neurons to AgRP neurons has been shown to drive hunger-related behaviors (Krashes et al. 2014). Under this framework, reduced NMDAR function resulting from NSPA deficiency, as observed in hippocampal neurons (Espinoza et al. 2020), might be expected to decrease feeding behavior. However, food intake was not reduced in NSPA-KO mice. This apparent discrepancy may reflect the increasing complexity recognized within hypothalamic feeding circuits. Recent studies have identified multiple subsets of GABAergic neurons in the ARC, including non-AgRP populations, that mediate leptin’s anorexigenic effects through distinct and complementary mechanisms (Solheim et al. 2025; Qi et al. 2023; Tan et al. 2024; Lavoie et al. 2024, 2025). STAT3 phosphorylation, assessed as a downstream marker of hypothalamic leptin signaling (Williams and Elmquist 2012; Ahima and Flier 2025), showed a significant reduction in NSPA-KO mice compared to WT mice, suggesting an attenuation of leptin responsiveness in the hypothalamus lacking NSPA expression. Indeed, this observation warrants further confirmation in future studies, especially considering that reduced pSTAT3 levels may reflect diminished leptin production and/or impaired leptin sensitivity at the hypothalamic level (Liu et al. 2021). Both defective leptin signaling and adiposity may contribute to insulin resistance (Myers et al. 2021; Hoek et al. 2008). In addition, the hippocampus has also been implicated in circuits regulating food intake and susceptibility to obesity, beyond its classical role in cognition and partially through interactions with hypothalamic neuropeptide systems (Barbosa et al. 2023; Azevedo et al. 2019; Smith and Azevedo 2025). All this suggests that the functional consequences of altered NMDAR stability may vary depending on neuronal subtype, circuit context, and leptin responses, complicating straightforward predictions regarding feeding outcomes in NSPA-deficient mice.

Another possibility is that the increased adipose tissue observed in NSPA-KO mice results from glutamatergic dysfunction in hypothalamic neuronal subpopulations that regulate adipocyte metabolism via sympathetic tone. Energy storage and mobilization in white adipose tissue (WAT) are tightly controlled by the sympathetic nervous system (SNS) (Harms and Seale 2013), which directly innervates adipose depots and promotes lipolysis and fat mobilization when its outflow is increased (Caron et al. 2018; Bartness et al. 2014). Distinct central nervous system circuits, including hypothalamic nuclei such as the arcuate nucleus, dorsomedial hypothalamus, lateral hypothalamic area, and paraventricular nucleus, may coordinate adipose tissue metabolism projecting to adipocyte depots (Caron et al. 2018). Within this framework, specific neuronal populations may exert separable control over energy expenditure, thermogenesis, and feeding behavior. For example, POMC neurons regulate metabolic rate, locomotor activity, and thermogenesis in brown and inguinal white adipose tissue (Williams et al. 2014). Importantly, circuits governing thermogenesis and feeding can be functionally linked or dissociated (Sinden et al. 2019). Consistent with this notion, impaired hypothalamic regulation of sympathetic output to brown adipose tissue has been shown to reduce thermogenesis without altering food intake (Sinden et al. 2019). Accordingly, altered glutamatergic signaling resulting from NSPA deficiency may preferentially affect hypothalamic circuits that regulate sympathetic outflow to adipose tissue, thereby promoting fat accumulation independently of changes in feeding behavior. Whether NSPA contributes to regulating SNS tone toward adipose depots, through NMDAR-dependent or alternative mechanisms, is an important question for future investigation.

The Human Protein Atlas suggests that NSPA may also be expressed in metabolic tissues like skeletal muscle, adipose tissue, pancreas and liver. We confirmed that NSPA is expressed in both iWAT and eWAT. Furthermore, WAT, as an endocrine organ, contributes to energy metabolism through lipolysis, lipogenesis, and energy storage as triacylglycerols (Fonseca-Alaniz et al. 2007), as well as by secreting hormones such as leptin and adiponectin, which regulate multiple metabolic homeostatic pathways (Trayhurn et al. 2011). Therefore, the changes we observed in NSPA-KO mice may originate from a combination of metabolic dysfunctions derived from NSPA-lacking expression in WAT and hypothalamus. Indeed, a whole-body knockout does not allow us to distinguish the specific tissue(s) in which NSPA is functionally required for the observed phenotype. Previous studies have shown that NSPA is expressed in various brain regions, including the hypothalamus (Matus et al. 2007; Segovia-Miranda et al. 2015) and our data suggest its expression in adipose tissue. Analysis of SNPs links NSPA-encoding Zzef1 gene with visceral adiposity accumulation and increased body fat percentage (Karlsson et al. 2019; Martin et al. 2021). Genome-wide association studies (GWAS) have identified polymorphisms in the gene encoding NSPA that associate with altered glucose metabolism and increased risk of T2D (Scott et al. 2017; Suzuki et al. 2019; Vujkovic et al. 2020; Palmer et al. 2014). In addition, SNPs analysis in Zzef1 has linked this gene with visceral adiposity accumulation and increased body fat percentage (Karlsson et al. 2019; Martin et al. 2021). In line with those genetic analysis, our findings provide the first experimental evidence that NSPA regulates glucose metabolism, insulin sensitivity, and adiposity. The biochemical alterations found in adipocyte tissues and the decreased levels of pSTAT3 in hypothalamus suggest that NSPA has functional implications in both tissues.

This study presents the following limitations. Body composition was assessed by endpoint fat-pad dissection, a method that, while still widely used, lacks the sensitivity and precision of non-invasive approaches such as magnetic resonance imaging (MRI). Similarly, although we evaluated respiratory exchange ratio (RER) and physical activity, we did not directly measure total energy expenditure (TEE) using indirect calorimetry in metabolic cages, thus leaving the energy balance uncertain. Our use of a whole-body NSPA knockout model does not allow for precise attribution of tissue-specific functions. Although NSPA is expressed in the hypothalamus and adipose tissue, the relative contribution of central versus peripheral mechanisms remains speculative. For instance, its proposed role in modulating hypothalamic glutamatergic transmission—based on previous findings in the hippocampus—requires targeted genetic approaches for validation. All experiments were conducted in male mice. Given the well-established influence of sex hormones on metabolism, adipose biology, and insulin sensitivity, future studies should assess whether the observed phenotypes are conserved or diverge in female mice. These limitations, while significant, provide a roadmap for future investigations to build upon our initial findings.

Notably, metabolic syndrome is highly prevalent among SLE patients and correlates with disease activity (Chung et al. 2007; Demir et al. 2015). Anti-P antibodies, which cross-react with NSPA and have a prevalence of 15–30% of SLE patients, are also associated with disease activity (Gonzalez and Massardo 2018; Viana et al. 2017). Remarkably, anti-P antibodies effect in immunized mice closely resembles the phenotype observed in NSPA-deficient (Díaz-Valdivia et al. 2025). Whether these antibodies impact metabolic balance, potentially mimicking or counteracting the metabolic effects seen in NSPA-KO mice, remains an important question.

Conclusions

This study identifies NSPA as a previously unrecognized regulator of systemic energy metabolism. The metabolic phenotype of NSPA-KO mice indicates that NSPA contributes to regulating weight gain, adiposity, and insulin responses under control feeding. Additionally, under an obesogenic challenge, NSPA is required to maintain metabolic balance. Mechanistically, NSPA function is required to maintain the expression levels of key glucose transporters, glycolytic enzymes, and lipolytic markers in white adipose tissue. The presence of NSPA in metabolically active tissues suggests a dual role in both central and peripheral regulation of metabolism. These findings offer mechanistic insights linking Zzef1 gene variants to type 2 diabetes and suggest that anti-P autoantibodies in patients with SLE may disrupt metabolic balance via NSPA, opening new avenues for research at the intersection of autoimmunity and metabolism.

Abbreviations

APC10

Anaphase promoter complex 10

BAT

Brown adipose tissue

GWAS

Genome-wide association studies

HFHS

High fat high sugar

HSL

Hormone sensible lipase

ipGTT

Intraperitoneal glucose tolerance test

ipITT

Intraperitoneal insulin tolerance test

NMDAR

N-methyl-D-aspartate receptor

NSPA

Neuronal surface P antigen

RER

Respiratory exchange ratio

SLE

Systemic lupus erythematosus

SNPs

Single nucleotide polymorphisms

T2D

Type 2 diabetes

TAG

Triacylglycerols

UPS

Ubiquitin proteasome system

WAT

White adipose tissue

Authors’ contributions

Conceptualization: S.E., A.G., and B.K. Formal analysis: S.E. Funding acquisition: S.E., B.K. and A.G. Investigation: S.E., A.V., A.B., A.M., G.C., N.D-V., J.A. and L.V. Methodology: S.E. and A.V. Resources: S.E., B.K. and A.G. Supervision: S.E. and B.K. Visualization: S.E., B.K. and A.G. Writing – Original Draft Preparation: S.E. and A.V. Writing – Review & Editing: S.E., B.K. and A.G.

Funding

This work recieved financial support from FONDECYT grant #1230905 and Anillo ACT210039 (BK), Postdoctoral FONDECYT #3210493 (SE), FONDECYT grant #1221796 and Centro Científico Tecnológico de Excelencia Ciencia & Vida, Basal Project FB 210008 (AG). Agencia Nacional de Investigación y Desarrollo (ANID), Programa Becas Doctorado Nacional, Grant/Award Number: #21190474 (AV) Grant/Award Number: #21230642 (N.D-V).

Data availability

All data supporting the conclusions of this article is included in this article, raw data will be made available by the authors on request.

Declarations

Ethics approval and consent to particpate

The animal study protocol was approved by the Institutional Ethics Committee of Universidad San Sebastián (CEC 07–2021-10).

Competing interests

The authors declare no competing interests.