TYPE Original Research PUBLISHED 19 April 2023 DOI 10.3389/fendo.2023.1146454 Overexpressing high levels of OPEN ACCESS human vaspin limits high fat EDITED BY Xi Li, Chongqing Medical University, China diet-induced obesity and REVIEWED BY Hayley Maree O’Neill, enhances energy expenditure Bond University, Australia Mourad Ferdaoussi, Alberta Education, Canada in a transgenic mouse *CORRESPONDENCE John T. Heiker Inka Rapöhn 1, Ivet Elias 2,3, Juliane Weiner 4, Anna Pujol 2,3, john.heiker@helmholtz-muenchen.de Jana Breitfeld Stephanie Kehr 5, Alexandra Chadt 6,7, Hadi Al-Hasani 6,7, jana.breitfeld@medizin.uni-leipzig.de Ralph Burkhardt 8, Nora Klöting 1, Michael Stumvoll 1,4, SPECIALTY SECTION Fatima Bosch 2,3, Peter Kovacs 4,7, John T. Heiker 1,4* This article was submitted to Obesity, and Jana Breitfeld 4* a section of the journal 1 Helmholtz Institute for Metabolic, Obesity and Vascular Research (HI-MAG) of the Helmholtz Frontiers in Endocrinology Zentrum München at the University of Leipzig and University Hospital Leipzig, Leipzig, Germany, RECEIVED 17 January 2023 2 Center of Animal Biotechnology and Gene Therapy (CBATEG) and Department of Biochemistry and ACCEPTED 30 March 2023 Molecular Biology, Universitat Autònoma de Barcelona, Bellaterra, Spain, 3 Centro de Investigación PUBLISHED 19 April 2023 Biomédica en Red de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Madrid, Spain, 4 Medical Department III - Endocrinology, Nephrology, Rheumatology, University of Leipzig Medical CITATION Center, Leipzig, Germany, 5 Bioinformatics Group, Department of Computer Science and Rapöhn I, Elias I, Weiner J, Pujol A, Interdisciplinary Center for Bioinformatics, Leipzig University, Leipzig, Germany, 6 Institute for Clinical Kehr S, Chadt A, Al-Hasani H, Burkhardt R, Biochemistry and Pathobiochemistry, German Diabetes Center (DDZ), Leibniz Center for Diabetes Klöting N, Stumvoll M, Bosch F, Kovacs P, Research at Heinrich Heine University Düsseldorf, Düsseldorf, Germany, 7 German Center for Diabetes Heiker JT and Breitfeld J (2023) Research (DZD), München-Neuherberg, Germany, 8 Institute of Clinical Chemistry and Laboratory Overexpressing high levels of human Medicine, Transfusion Medicine, University Hospital Regensburg, Regensburg, Germany vaspin limits high fat diet-induced obesity and enhances energy expenditure in a transgenic mouse. Front. Endocrinol. 14:1146454. doi: 10.3389/fendo.2023.1146454 Adipose tissue inflammation and insulin resistance are hallmarks in the COPYRIGHT development of metabolic diseases resulting from overweight and obesity, © 2023 Rapöhn, Elias, Weiner, Pujol, Kehr, Chadt, Al-Hasani, Burkhardt, Klöting, such as type 2 diabetes and non-alcoholic fatty liver disease. In obesity, Stumvoll, Bosch, Kovacs, Heiker and adipocytes predominantly secrete proinflammatory adipokines that further Breitfeld. This is an open-access article promote adipose tissue dysfunction with negative effects on local and distributed under the terms of the Creative Commons Attribution License (CC BY). The systemic insulin sensitivity. Expression of the serpin vaspin (SERPINA12) is also use, distribution or reproduction in other increased in obesity and type 2 diabetes, but exhibits compensatory roles in forums is permitted, provided the original author(s) and the copyright owner(s) are inflammation and insulin resistance. This has in part been demonstrated using credited and that the original publication in vaspin-transgenic mice. We here report a new mouse line (h-vaspinTG) with this journal is cited, in accordance with transgenic expression of human vaspin in adipose tissue that reaches vaspin accepted academic practice. No use, distribution or reproduction is permitted concentrations three orders of magnitude higher than wild type controls (>200 which does not comply with these terms. ng/ml). Phenotyping under chow and high-fat diet conditions included glucose- tolerance tests, measurements of energy expenditure and circulating parameters, adipose tissue and liver histology. Also, ex vivo glucose uptake in isolated adipocytes and skeletal muscle was analyzed in h-vaspinTG and littermate controls. The results confirmed previous findings, revealing a strong reduction in diet-induced weight gain, fat mass, hyperinsulinemia, -glycemia and -cholesterolemia as well as fatty liver. Insulin sensitivity in adipose tissue and muscle was not altered. The h-vaspinTG mice showed increased energy expenditure under high fat diet conditions, that may explain reduced weight Frontiers in Endocrinology 01 frontiersin.org Rapöhn et al. 10.3389/fendo.2023.1146454 gain and overall metabolic improvements. In conclusion, this novel human vaspin-transgenic mouse line will be a valuable research tool to delineate whole-body, tissue- and cell-specific effects of vaspin in health and disease. KEYWORDS serpin, metabolic disease, obesity, adipose tissue, mouse model, inflammation Introduction mice revealed a strong reduction in diet-induced weight gain, fat mass, hyperinsulinemia, -glycemia and -cholesterolemia. This Obesity is the major risk factor for type 2 diabetes (T2D), mouse represents an ideal model for future investigations of atherosclerosis and nonalcoholic fatty liver disease (1). vaspin function in obesity-related diseases, such as atherosclerosis Inflammation of the adipose tissue (AT) is a frequent and T2D. consequence of adipocyte hypertrophy, increased AT stresses and immune cell infiltration, and contributes to the development of insulin resistance (2, 3). Materials and methods Visceral adipose tissue derived serine protease inhibitor (vaspin), also known as SERPINA12, is a member of the serpin Generation of vaspin-transgenic family (4), that was first identified in visceral (omental) AT of the mice (h-vaspinTG) Otsuka Long-Evans Tokushima fatty (OLETF) rat, an animal model of obesity and T2D (5). In humans, serum vaspin levels are A pET16b plasmid containing the human SERPINA12 sequence increased in individuals with obesity and T2D (6). Treatment of (excluding the signal sequence and cloned using NdeI and BamHI) obese and insulin resistant mice with recombinant vaspin was provided by Prof. Jun Wada (Okoyama University Graduate counteracts obesity-induced inflammation and dysfunction of School of Medicine) and served as basis for the construction of adipose tissue and liver and partly restores glucose tolerance (7– further plasmids. First, the N-terminal sequence of the signal 9). In line, metabolic health of mice overexpressing vaspin is more peptide was added by PCR and the whole construct was cloned resistant to obesogenic conditions while a vaspin knockout into the EcoRI and XhoI sites of the pcDNA 3.1(-) expression aggravates metabolic dysfunction in obesity (10). Together, these plasmid. The correctness of the sequence was checked by findings have established the general view of vaspin as a beneficial sequencing. In the following steps, the whole SERPINA12 gene and compensatory player in obesity-related disorders and diseases was cloned into the SmaI site of a plasmid containing the aP2- (11, 12). promotor and a SV40polyA to achieve a vaspin overexpression in The serpin vaspin inhibits two kallikrein-related peptidases, the AT. Finally, transgenic mice expressing the aP2- SERPINA12 kallikrein 7 (KLK7) and 14 (KLK14) (8, 13). Knockout of KLK7 in chimeric gene were obtained by microinjection of oocytes from AT dampened obesity-induced AT inflammation and preserved B6SJL (F2) mice, which was conducted in the Unitat d´Animals insulin sensitivity in mice (14). Also, a whole-body knockout of Transgènics (UAT-CBATEG) of the Universitat Autònoma de KLK7 improved diet-induced metabolic alterations with more Barcelona, Bellaterra, Spain. Twelve founders (1 male, 11 females) pronounced reduction of diet-induced weight gain (15). Anti- were obtained and Southern blot analysis was performed using inflammatory effects of vaspin in the liver have been in part EcoRI digested genomic DNA. Serum vaspin concentrations were ascribed to vaspins’ interaction with the ER chaperon GRP78 on measured by ELISA (Adipogen, Korea). Further genotyping of the cell surface of hepatocytes (10). offspring mice was done by PCR using human specific The protective effects of vaspin in obesity-induced metabolic SERPINA12 sequence primers, available upon request. Resulting dysfunctions induced by obesity have been demonstrated using a h-vaspinTG mouse lines were then backcrossed on the C57Bl/6N transgenic mouse model (m-vaspinTG) overexpressing mouse background for 5 generations before phenotyping. vaspin in AT (10). This model exhibited an increase of circulating vaspin levels by 3-fold (0.6 ng/ml in vaspin transgenic mice vs 0.2 ng/ml in wildtype controls). A mouse model to specifically assess Animals the function of human vaspin in metabolic disease has not been reported. We here report a second mouse line (h-vaspinTG) with Mice were housed under pathogen-free conditions (3-5 mice transgenic expression of human vaspin under the aP2 promotor per cage) at 23°C on a 12 h light/dark cycle in the local animal that reaches vaspin concentrations three orders of magnitude facility (MEZ - Medizinisch-Experimentelles Zentrum, University higher than wild type controls (>200 ng/ml). In line with of Leipzig, Leipzig). All mice had ad libitum access to water and previous findings, phenotypic characterization of the h-vaspinTG food and were either fed a standard chow diet (V1534, 9 kJ% from Frontiers in Endocrinology 02 frontiersin.org Rapöhn et al. 10.3389/fendo.2023.1146454 fat, Ssniff®, Soest, Germany) or kept on a high fat diet (HFD; Ex vivo glucose uptake into isolated containing 59 kJ% from fat; E15772-34, Ssniff®, Soest, Germany) skeletal muscle and mature adipocytes starting at an age of six weeks for up to 30 weeks. Non-transgenic littermates served as controls. After sacrificing, serum was taken Glucose uptake was assessed in intact isolated Extensor and organs were removed, weighed (only liver, subcutaneous digitorum longus (EDL) skeletal muscles by the accumulation of inguinal (iWAT), visceral epididymal (eWAT), brown adipose [3H]2-deoxyglucose (Hartmann Analytic; Braunschweig, Germany) tissue (BAT)), snap-frozen in liquid nitrogen and stored at -80°C and using [14C]mannitol (PerkinElmer, Waltham, MA) as an until further use. All animal experiments were approved by the local extracellular marker as described (16). Briefly, mice were authorities (for Leipzig: Landesdirektion Leipzig, State of Saxony, sacrificed and EDL muscles were removed. After recovery, Germany, reference number TVV26-16; for Düsseldorf: Ethics muscles were treated with or without 120 nM insulin (Actrapid, Committee of the State Ministry of Agriculture, Nutrition and Novo Nordisk) for 30 min. Finally, muscles were treated with 1 mM Forestry, State of North Rhine-Westphalia, Germany, reference [3H]2-deoxy-glucose (2.5 mCi/mL) and 19 mM [14C]mannitol (0.7 number 84-02.04.2016.A430; for Barcelona: Ethics Committee in mCi/mL) for 20 minutes, subsequently frozen in liquid nitrogen. Animal and Human Experimentation of the Universitat Autònoma Cleared protein lysates were used to determine incorporated de Barcelona, Spain), as recommended by the responsible local radioactivity by scintillation counting. animal ethics review boards. Glucose uptake into isolated white adipocytes was assessed ex vivo by measuring the incorporation of radioactively labelled [14C]- D-glucose (Hartmann Analytic, Braunschweig, Germany) as Phenotypic characterization of described (17). Briefly, epididymal fat pads were dissected, h-vaspinTG mice primary adipocytes were isolated by collagenase digestion and treated with or without 120 nM insulin for 30 min, followed 0.1 Male mice were studied from the age of six up to 29 weeks on µCi/µl [14C]-D-glucose for 30 min. Cells were centrifuged in tubes either chow or HFD. Body weight was recorded weekly. containing dinonyl phthalate oil and the incorporated radioactivity Intraperitoneal (i.p.) glucose tolerance tests (GTT; at 25 weeks of was determined by scintillation counting. The resulting counts were age, 2 g glucose per kg body weight), body composition analysis and normalized to the lipid weight of the samples using overnight indirect calorimetry (at 28 weeks of age, for 72 h) were performed as heptane extraction. previously described (14). GTTs were performed after an overnight fast. GTT area under the curve (AUC) was calculated for each individual mouse after subtraction of the basal glucose level. Blood Statistical analysis glucose was determined from whole-venous blood samples using an automated glucose monitor (FreeStyle Mini; Abbott GmbH, Data are presented as means ± SEM. Statistical analyses were Ludwigshafen, Germany). performed using GraphPad Prism 9 (GraphPad, San Diego, CA, Serum adiponectin, leptin, insulin, C-peptide and vaspin were USA), except where otherwise indicated. Methods of statistical measured by standard ELISA; serum HbA1c, triglycerides (TG) and analyses were chosen according to the design of each experiment cholesterol (chol) were analyzed by an automated chemical analyzer and are indicated in the figure legends. If not otherwise stated, an at the Institute of Laboratory Medicine and Clinical Chemistry, adjusted p < 0.05 was considered statistically significant. Medical Department, University of Leipzig (14). AT and liver histology and measurements of adipocyte size distribution was performed as described previously (14). Results Gene expression analysis Vaspin transgenic mice present highly supraphysiological levels of Total RNA was isolated from various mouse tissues using circulating vaspin TRIzol (Life Technologies, Grand Island, NY). RNA clean-up was done using RNeasy MinElute Cleanup (Qiagen) according to the As reported above, 12 founders (1 male and 11 females) were manufacturer’s instructions. Following, 500 ng RNA was reverse obtained after microinjection of oocytes. Quantification of AT transcribed with standard reagents (Life Technologies). Human and mRNA expression in these h-vaspinTG mice revealed a high-level mouse gene expression was measured by quantitative real-time q- expressing mouse, five intermediate expressing and six low PCR using TaqMan methodology and fluorescence was detected on expressing founders (data not shown). For further analysis, male an ABI PRISM 7500 sequence detector (Applied Biosystems, mice of the high-level vaspin expressing line were phenotyped in Darmstadt, Germany). The following assays were used: detail. Expression of human vaspin (SERPINA12) was only Hs00545180_m1 for the human SERPINA12 gene; detectable in all AT depots and highest in BAT, followed by Mm00471557_m1 for endogenous mouse serpinA12 gene. iWAT and rather low in eWAT (Figure 1A). Non-transgenic Adipocyte gene expression was calculated by DDCT method and controls were negative for SERPINA12 mRNA expression in all normalized to 36b4 levels in each sample. tissues. Gene expression of endogenous vaspin (serpinA12) was not Frontiers in Endocrinology 03 frontiersin.org Rapöhn et al. 10.3389/fendo.2023.1146454 different between h-vaspinTG and controls, with highest expression Vaspin transgenic mice are less sensitive to in skin, stomach, liver and BAT (Figure 1B). A previously reported diet-induced obesity vaspin transgenic mouse line exhibited a 2-3-fold increase (0.4-0.6 vs. 0.2 ng/ml) in endogenous (mouse) circulating vaspin levels At the beginning of the study, h-vaspinTG and controls had compared to controls (10). Here, our high-level expressing h- comparable body weights (Figure 1D) and body length (Table 1). vaspinTG mice exhibited similar expression levels of serum Both lines responded to HFD feeding, gaining significant body weight endogenous vaspin (data not shown) and high circulating levels compared to their chow-fed controls (Figure 1E). After 29 weeks on the of human vaspin (200 ng/ml; Figure 1C). Non-transgenic controls respective diets, h-vaspinTG and controls did not significantly differ in were negative for human vaspin in the serum. H-vaspinTG mice fed body weight under chow diet conditions, though h-vaspinTG were a HFD for 29 weeks showed an additional significant increase to heavier (33.2 ± 0.9 g vs. 31.6 ± 0.6 g; Figure 1F). When fed a HFD, h- ~400 ng/ml of circulating vaspin (Figure 1C), demonstrating vaspinTG responded much less to the diet with significantly lower highly supraphysiological levels of vaspin in our h-vaspinTG mice body weights after 10 weeks on the diet. At the end of the study, h- that correspond to a 4.000 – 8.000-fold increase compared vaspinTG had gained ~5.5 g less body weight compared to controls to controls. (39.1 ± 0.9 g vs. 46.0 ± 0.7 g; Figures 1E, F). A B C D E F G H I J K L M FIGURE 1 Characterization of h-vaspinTG mice and controls under chow and high fat diet (HFD). (A, B) relative mRNA expression of (A) human (SERPINA12) and (B) endogenous (SerpinA12) vaspin in AT depots and peripheral tissues as determined by RT-qPCR (n = 3 per group). Gene expression was normalized to 36b4. (C) Human vaspin serum concentration (ng/ml) as measured by ELISA (n = 7-10), at the end of the study. (D) Initial body weight, (E) body weight gain and (F) final body weight of h-vaspinTG and control mice fed a chow or a HFD for 29 weeks, starting at the age of 6 weeks (n = 5-10). (G) Percentage of fat mass (%) and lean body mass (%) as determined by EchoMRI at the end of the study (n = 5-10). (H, I) Absolute and relative organ weights of liver, epididymal (eWAT), inguinal (iWAT) and brown adipose tissue (BAT) at the end of the study (n =5-10). (J-M) Indirect calorimetry of h-vaspinTG and control mice under chow and HFD during light and dark phase (n = 4-5 per group). (J) Oxygen consumption (VO2 in ml/kg/h) and (K) energy expenditure (EE in kcal/h/kg) were recorded over a period of 72 h. (L) Physical activity assessed via locomotor activity as movement in cage (XY-axis), rearing (Z-axis) and (M) running in wheel (cm) were recorded for 72 h (n = 4-5). Data are presented as mean ± SEM. Statistical significance was calculated by two-way ANOVA followed by Š idá k’s (A, B, D) or Tukey’s (G-L) post-hoc test or by one-way ANOVA with Fisher’s LSD test (C) (*p ≤ 0.05, ** p ≤ 0.01, ***p ≤ 0.001). Frontiers in Endocrinology 04 frontiersin.org Rapöhn et al. 10.3389/fendo.2023.1146454 TABLE 1 Anthropometric and metabolic parameters in male chow and high-fat diet (HFD)-fed h-vaspinTG and control mice. Chow HFD Control h-vaspinTG Control h-vaspinTG (n) (n) (n) (n) Body length (cm) 9.7 ± 0.1 (10) 9.6 ± 0.1 (7) 10.0 ± 0.1 (5) 9.8 ± 0.1 (5) Tail length (cm) 8.2 ± 0.1 (10) 8.3 ± 0.1 (7) 8.0 ± 0.1 (5) 7.8 ± 0.2 (5) Body temp. (°C) 35.2 ± 0.6 (7) 35.2 ± 0.4 (7) 35.4 ± 0.6 (5) 35.3 ± 0.5 (5) Food intake (kcal/d) 20.6 ± 1.6 (4) 21.3 ± 4.8 (4) 20.7 ± 2.7 (5) 24.5 ± 2.7 (4) RER 0.83 ± 0.02 (4) 0.84 ± 0.02 (4) 0.73 ± 0.01 (4) 0.73 ± 0.01 (4) Serum parameters HbA1c (%) 4.4 ± 0.1 (10) 4.6 ± 0.1 (7) 4.6 ± 0.1 (7) 4.6 ± 0.0 (7) All results are expressed as means ± SEM. Numbers of animals are given in brackets. Body composition reflected body weight development. Under accumulation in livers of h-vaspinTG and control mice. Serum chow conditions, h-vaspinTG had a significantly higher percentage triglycerides were similar in h-vaspinTG and controls, but total of fat mass (19.6 ± 0.8% vs. 13.3 ± 1.2%), though differences in circulating cholesterol, while increased after HFD in both individual AT-depot masses were not significant (Figures 1G, H). genotypes, was significantly lower in h-vaspinTG mice On the contrary, HFD-fed h-vaspinTG had significantly less body (Figure 2D). Furthermore, liver histology revealed substantial fat (42.4 ± 1.6% vs. 36.1 ± 1.3%) and absolute as well as relative AT fatty liver and ectopic lipid accumulation in in HFD-fed control fat pad masses were significantly lower (Figures 1G-I). Also, liver mice (Figure 2E), which was clearly reduced in vaspin TG mice. In weights were significantly lower in h-vaspinTG, while BAT weights obese mice of both genotypes, glycogen levels were not different were not different between genotypes, irrespective of diet conditions (Figure 2E, PAS stain). (Figures 1H, I). We next assessed energy intake and expenditure as well as locomotor activity of h-vaspinTG in metabolic chambers. Food Vaspin transgenic mice show improved intake was not different and both genotypes consumed less energy metabolic profile after HFD dense HFD food compared to chow (Table 1). There were no differences in respiratory exchange rate, which clearly reflected the Both, under chow- and HFD-fed conditions, h-vaspinTG mice switch to lipid substrate metabolism under HFD in both genotypes had significantly lower fasting blood glucose levels (Figure 3A). (Table 1). Under chow diet, h-vaspinTG exhibited a trend for lower Also, free insulin levels were significantly lower (by ~70%) in h- oxygen consumption (VO2, Figure 1J) and energy expenditure (EE, vaspinTG mice compared to controls. Furthermore, while in Figure 1K). In control mice, HFD-feeding significantly suppressed control mice insulin levels increased by ~50% (p=0.08) under VO2 and EE compared to chow, as expected. This effect did not HFD, insulin levels in h-vaspinTG doubled, but levels were still occur in h-vaspinTG mice. VO2 and EE during the light and dark below chow-fed controls and the increase did not reach statistical phase were significantly higher in HFD-fed h-vaspinTG mice significance (Figure 3B). This was clearly due to reduced insulin compared to HFD-fed controls (Figures 1J, K). Body secretion, as C-peptide levels were mirroring insulin levels in h- temperatures were not increased in h-vaspinTG (Table 1). Finally, vaspinTG (Figure 3C), resulting in similar C-peptide/insulin ratios there were no differences in locomotor activity (Figures 1L, M; in both genotypes, irrespective of diet (Figure 3D). HbA1c was not movement in cage (XY), rearing (Z) and running wheel) between different under all conditions (Table 1). Circulating levels of the genotypes, and both groups of mice showed reduced activity adiponectin, a marker of insulin sensitivity and metabolic health, under HFD. were significantly higher in h-vaspinTG under chow diet, but were significantly decreased in both genotypes in response to the HFD (Figure 3E). Leptin on the other hand, was not different in chow-fed Vaspin transgenic mice exhibit less mice of both genotypes, and highly increased (~5-fold) in HFD-fed adipocyte hypertrophy and ectopic lipid controls (Figure 3F). Reflecting the significantly reduced fat mass in deposition in liver h-vaspinTG after HFD, leptin levels in these mice were significantly lower compared to controls (by ~60%), albeit higher than in chow- AT histology revealed that the reduced fat mass in white AT fed controls (~2-fold) (Figure 3F). This resulted in a much healthier depots of HFD-fed h-vaspinTG mice also featured smaller leptin/adiponectin ratio, which was highly increased in controls adipocytes, especially in eWAT (Figures 2A-C). As we also after HFD, but not significantly different in h-vaspinTG, when observed significantly lower liver weights in h-vaspinTG mice compared to chow-fed controls (Figure 3G). The beneficial serum (Figure 1H), we assessed circulating lipids and ectopic lipid profile of h-vaspinTG did not translate into improved glucose Frontiers in Endocrinology 05 frontiersin.org Rapöhn et al. 10.3389/fendo.2023.1146454 tolerance, and both genotypes showed obesity-induced impaired vascular inflammation (reviewed in (11, 12)). Important tools to glucose clearance in the GTT (Figure 3H). Yet, h-vaspinTG mice study and establish the function of this intriguing serpin were had improved glucose tolerance under chow-fed conditions vaspin-knockout and vaspin transgenic mouse models (10, 19, 20). (Figure 3H). To assess insulin sensitivity more directly in muscle Similar to the previously described vaspinTG mouse (m- and AT, we performed ex vivo glucose uptake in isolated muscle vaspinTG) (10), the new mouse line reported here features AT and epididymal white AT. There were no differences in glucose specific overexpression using the aP2 promotor. In contrast, the uptake in both tissues (Figures 3I, J). As effects of vaspin on human new line expresses the human SERPINA12 transgene, while the myotubes have been reported before (18), we also assessed effects of initial model overexpressed the mouse serpinA12 gene. It is recombinant vaspin on insulin-stimulated glucose uptake in EDL important to note that key residues regulating vaspin function muscle and epididymal adipocytes of lean and obese controls ex and specificity identified studying human vaspin (21, 22) are all vivo (Figures 3K, L). In muscle, vaspin had no effects on insulin- conserved in the mouse ortholog. Among these are the reactive stimulated glucose uptake independent of obesity, but reduced basal center loop with the protease cleavage site, regulatory exosites as uptake in muscle of lean mice. In eWAT adipocytes, vaspin did dose well as the heparin binding site. Also, both ortholog proteins are dependently increase insulin-stimulated glucose uptake, but only in glycoproteins, but differ in some glycosylation sites (23). They share lean mice. one glycosylation site and while human vaspin exhibits two more These data indicate an improved metabolic profile based on glycosylation sites located on the top of the serpin molecule, mouse serum markers, but did not translate into improved glucose vaspin features only one additional glycosylation site located on the metabolism and responsiveness to insulin in h-vaspinTG mice. back of the molecule. Protease inhibition, heparin binding and thermal stability are comparable for both orthologs (23). Vaspin expression levels are significantly higher in the new h- Discussion vaspinTG model. While initial m-vaspinTG mice had a 2-3xfold increased circulating vaspin levels, circulating vaspin levels in the h- Vaspin exerts multiple beneficial effects in the context of obesity vaspinTG mouse are three orders of magnitudes higher than in and especially obesity-related comorbidities, as well as liver and controls. Vaspin serum levels in humans are typically in the range of A B C D E FIGURE 2 h-vaspinTG mice exhibit less hypertrophy in white adipose tissue (AT) and reduced fatty liver under high fat diet (HFD). (A) Representative histological H&E stained sections of inguinal (iWAT) and epididymal (eWAT) adipose tissue of HFD fed h-vaspinTG and control mice. Scale bar: 50 µm. (B, C) Adipocyte size distribution in iWAT (B) and eWAT (C) sections (n = 3-5 per group). (D) Cholesterol (Chol) and serum triglycerides (TG) (mg/dl) were measured in chow and HFD-fed h-vaspinTG and control mice. (E) Two representative histological PAS stained liver sections of HFD- fed h-vaspinTG and control mice. Scale bar: 100 µm. (F) Liver glycogen levels (µg) of chow and HFD-fed h-vaspinTG and control mice. (n = 5-10 per group). Data are presented as mean ± SEM. Statistical significance was calculated by multiple t-tests (B, C) or two-way ANOVA followed by Tukey’s post-hoc test (D) (*p ≤ 0.05, ** p ≤ 0.01, ***p ≤ 0.001). Frontiers in Endocrinology 06 frontiersin.org Rapöhn et al. 10.3389/fendo.2023.1146454 0.5 - 1.5 ng/ml (6) and subpopulations have been reported with m-vaspinTG mouse, in part with larger effect size. Yet, there are also vaspin levels up to >30 ng/ml (24). Furthermore, there is significant interesting differences in the high level-vaspin expressing h-vaspinTG diurnal variation in vaspin serum levels with the nocturnal peak mouse. First of all, both vaspinTG mouse lines are at least partly reaching 250% of the daily minimum (25). In addition, vaspin protected from HFD-induced obesity, but the final differences in interactions with cell surface receptors (10) or proteoglycans (21) weight gain after 20+ weeks on the HFD were much more will have significant effects on local concentrations of vaspin at or pronounced in h-vaspinTG mice. On average, h-vaspinTG mice had near the cell surface, that may be very different from the circulating gained ~15% less body weight after the HFD compared to controls (vs. levels. This is also reflected in concentrations of recombinant vaspin ~8% in m-vaspinTG). This was mainly due to a significant reduction in used in the literature to study cellular or tissue specific effects of less total body fat (~6%), and here especially of the iWAT depot. Both vaspin (ranging from 1ng/ml up to 1µg/ml, with most of the studies vaspin transgenic lines showed reduced hypertrophy under HFD- investigating effects of vaspin in in the 100-500 ng/ml range). Thus, conditions. In the m-vaspinTG mouse, no direct explanation was the h-vaspinTG mouse model clearly exhibits supraphysiologic found for reduced weight gain under HFD. There were no obvious levels of vaspin, but considering the variation observed in the differences in energy intake, energy expenditure (EE) and activity human populations and also with respect to the in vitro data, compared to controls, only a slightly increased RER may indicate these do not represent such extreme levels, which may result in alterations in metabolic substrate utilization in m-vaspinTG (10). This effects that may not be physiologically relevant. is different in the new model, where we did observe significant Together, the phenotype of the new h-vaspinTG mouse differences in VO2 and EE, albeit no changes in activity and RER. recapitulates and confirms the improved metabolic phenotype of the HFD-fed h-vaspinTG exhibited higher EE, both during the light and A B C D E F G H K I J L FIGURE 3 h-vaspinTG mice display healthier serum profile under high fat diet (HFD). (A) Blood glucose (mmol/L), (B) insulin (ng/ml) and (C) C-peptide levels (ng/ml), as well as (D) C-peptide/insulin ratio, (E) adiponectin (µg/ml) and (F) leptin levels (ng/ml), as well as (G) adiponectin/leptin ratio of h- vaspinTG and control mice fed a chow or HFD for 23 weeks. (H) Glucose tolerance tests (GTT) and area under the curve (AUC, presented in insert) in h-vaspinTG and control mice fed a chow or HFD. (I, J) Ex vivo glucose uptake in (I) EDL muscle (mmol/mg/min) and (J) epididymal adipocytes (cpm/mg lipid) of h-vaspinTG and control mice fed a chow or HFD (n = 5-10). (K, L) Relative ex vivo glucose uptake in (K) EDL muscle and (L) epididymal adipocytes from lean and obese C57BL/6N mice, treated with either 10 ng/ml or 100 ng/ml recombinant vaspin and compared to untreated controls (n = 8-12). Data are presented as mean ± SEM. Statistical significance was calculated by one-way ANOVA followed by Fisher’s LSD test (*p ≤ 0.05, ** p ≤ 0.01, ***p ≤ 0.001). Frontiers in Endocrinology 07 frontiersin.org Rapöhn et al. 10.3389/fendo.2023.1146454 dark phase compared to HFD-fed controls. Here, the reduction in EE metabolism and EE are likely affected by different genetic background, observed in mice fed a HFD (26), was not apparent in h-vaspinTG diet composition and genotype-diet interactions (32–34). mice, despite both genotypes showing the expected reduction in In conclusion, we report a novel human vaspin-transgenic activity under HFD. In HFD-fed h-vaspinTG mice, 24h EE was mouse line, which recapitulates many previous findings using a almost exactly the same as in chow-fed h-vaspinTG mice, with an low-level expressing m-vaspinTG mouse line, and in addition increase during the light phase offsetting a slight reduction during the shows an increase in EE under HFD-conditions explaining the night phase. This increase in EE may be a key mechanism limiting body weight phenotype. This new mouse model will be a valuable weight-gain under HFD, as we did not observe differences in activity tool to delineate whole-body and tissue- or cell-specific effects of and food intake. While we and others have shown that acute vaspin vaspin, not only in the context of obesity. application (intraperitoneally and intracerebroventricular) reduced food intake in obese and insulin resistant db/db mice (27–29), this Data availability statement effect seems to be lost, desensitized or compensated under conditions of chronically increased vaspin levels. Interestingly, under chow The original contributions presented in the study are included conditions the observation was the contrary, with h-vaspinTG mice in the article/supplementary material. Further inquiries can be showing a trend for lower VO2 and EE both during the light and dark directed to the corresponding authors. phase, in line with a tendency for higher body weights at the end of the study. The significant induction of vaspin expression in activated BAT may link vaspin and thermogenesis (30), but how vaspin may regulate Ethics statement whole-body EE will be the focus of future investigations. Reduced responsiveness to the HFD in h-vaspinTG mice was The animal study was reviewed and approved by Landesdirektion accompanied by a 60% reduction in circulating leptin and significant Leipzig, Sachsen, Germany, TVV26-16, State Ministry of Agriculture, improvements in fasting blood glucose and insulin levels, as well as Nutrition and Forestry, State of North Rhine-Westphalia, Germany, lower cholesterol levels at the end of the study. Notably, m-vaspinTG Reference number 84-02.04.2016.A430, Ethics Committee in Animal mice also exhibited an improved profile of circulating markers, halfway and Human Experimentation of the Universitat Autònoma de through the study (10 weeks under the HFD), however most of the Barcelona, Spain. parameters reached levels of control mice at the end of the study (10). This may in part be explained by the less pronounced difference in Author contributions body weight in the HFD-fed m-vaspinTG compared to controls. Yet, m-vaspinTG mice exhibited significantly improved glucose and JB and PK conceived, designed, and supervised the study. JB, IE, insulin tolerance in vivo after HFD compared to controls. Strikingly, AP, and FB generated vaspin transgenic mice. IR, JW, SK, JH, and we did not observe this in the h-vaspinTG model, and also ex vivo JB conducted experiments and analyzed data. RB contributed serum glucose uptake in muscle and AT of HFD-fed h-vaspinTG mice was measurements. AC and HA-H contributed ex vivo glucose uptake not different. Peripheral vaspin effects on insulin sensitivity are data. NK helped with animal experiments. IR and JH wrote the controversially discussed. Application of recombinant vaspin manuscript with input from PK and JB. All authors contributed to increases the effect of an acute insulin bolus on blood glucose levels the article and approved the submitted version. in insulin tolerance tests (7, 8). Clamp studies in db/db mice did not show increased glucose uptake in hyperinsulinemic-euglycemic clamps after acute vaspin administration (8), but chronic treatment Funding (1µg/kg for 8 weeks) increased glucose uptake in rats (9). For both studies, resulting circulating h-vaspin levels post treatment were not This work was supported by the Deutsche Forschungsgemeins reported. Clamp data from both the m-vaspinTG and h-vaspinTG chaft (DFG; German Research Foundation) through CRC1052 mice are not available, but we did not observe improvement of insulin- “Obesity Mechanisms”, project number 209933838 (B3 PK, C7 mediated glucose uptake in muscle and AT of h-vaspinTG ex vivo. In JH, Z4 SK) and by Deutsches Zentrum für Diabetesforschung human myotubes from a donor with obesity, vaspin had a significant (DZD, grant 82DZD00601). effect on insulin-stimulated glucose uptake (18). We could not confirm this finding in ex vivo glucose uptake in muscle from lean and obese Acknowledgments control mice. We did find a beneficial effect on insulin-stimulated glucose uptake in eWAT samples from lean mice incubated with We thank Manuela Quandt, Jenny Schuster, Beate Gutsmann, recombinant vaspin, in line with our findings in vaspin-overexpressing Claudia Gebhardt and Olivia Paetow for excellent technical assistance. 3T3-L1 adipocytes (31). In the liver, the h-vaspinTG mice clearly confirmed the protective effect of vaspin expression from HFD- induced fatty liver, with a significant reduction in ectopic lipid Conflict of interest deposition compared to obese controls. It has to be noted, that in addition to the higher circulating vaspin The authors declare that the research was conducted in the levels in the h-vaspinTG compared to the m-vaspinTG mice, absence of any commercial or financial relationships that could be phenotypic differences between these vaspin transgenic lines related construed as a potential conflict of interest. Frontiers in Endocrinology 08 frontiersin.org Rapöhn et al. 10.3389/fendo.2023.1146454 Publisher’s note organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or All claims expressed in this article are solely those of the authors claim that may be made by its manufacturer, is not guaranteed or and do not necessarily represent those of their affiliated endorsed by the publisher. References 1. Shoelson SE, Herrero L, Naaz A. Obesity, inflammation, and insulin resistance. 19. Nakatsuka A, Yamaguchi S, Eguchi J, Kakuta S, Iwakura Y, Sugiyama H, et al. A Gastroenterology (2007) 132(6):2169–80. doi: 10.1053/j.gastro.2007.03.059 vaspin-Hspa1l complex protects proximal tubular cells from organelle stress in diabetic 2. Hotamisligil GS. Inflammation and metabolic disorders. Nature (2006) 444 kidney disease. Commun Biol (2021) 4(1):373. doi: 10.1038/s42003-021-01902-y (7121):860–7. doi: 10.1038/nature05485 20. Nakatsuka A, Wada J, Iseda I, Teshigawara S, Higashio K, Murakami K, et al. 3. Lumeng CN, Saltiel AR. Inflammatory links between obesity and metabolic Visceral adipose tissue-derived serine proteinase inhibitor inhibits apoptosis of endothelial disease. J Clin Invest (2011) 121(6):2111–7. doi: 10.1172/Jci57132 cells as a ligand for the cell-surface Grp78/Voltage-dependent anion channel complex. Circ Res (2013) 112(5):771–80. doi: 10.1161/CIRCRESAHA.111.300049 4. Heit C, Jackson BC, McAndrews M, Wright MW, Thompson DC, Silverman GA, et al. Update of the human and mouse serpin gene superfamily. Hum Genomics (2013) 21. Ulbricht D, Oertwig K, Arnsburg K, Saalbach A, Pippel J, Strater N, et al. Basic 7(1):22. doi: 10.1186/1479-7364-7-22 residues of beta-sheet a contribute to heparin binding and activation of vaspin (Serpin A12). J Biol Chem (2017) 292(3):994–1004. doi: 10.1074/jbc.M116.748020 5. Hida K, Wada J, Zhang H, Hiragushi K, Tsuchiyama Y, Shikata K, et al. Identification of genes specifically expressed in the accumulated visceral adipose tissue 22. Ulbricht D, Pippel J, Schultz S, Meier R, Strater N, Heiker JT. A unique serpin of oletf rats. J Lipid Res (2000) 41(10):1615–22. doi: 10.1016/S0022-2275(20)31994-5 P1' glutamate and a conserved beta-sheet c arginine are key residues for activity, protease recognition and stability of Serpina12 (Vaspin). Biochem J (2015) 470(3):357– 6. Feng R, Li Y, Wang C, Luo C, Liu L, Chuo F, et al. Higher vaspin levels in subjects 67. doi: 10.1042/BJ20150643 with obesity and type 2 diabetes mellitus: a meta-analysis. Diabetes Res Clin Pract (2014) 106(1):88–94. doi: 10.1016/j.diabres.2014.07.026 23. Oertwig K, Ulbricht D, Hanke S, Pippel J, Bellmann-Sickert K, Strater N, et al. Glycosylation of human vaspin (Serpina12) and its impact on serpin activity, heparin 7. Hida K, Wada J, Eguchi J, Zhang H, Baba M, Seida A, et al. Visceral binding and thermal stability. Biochim Biophys Acta Proteins Proteom (2017) 1865 adipose tissue-derived serine protease inhibitor: a unique insulin-sensitizing (9):1188–94. doi: 10.1016/j.bbapap.2017.06.020 adipocytokine in obesity. P Natl Acad Sci USA (2005) 102(30):10610–5. doi: 10.1073/ pnas.0504703102 24. Teshigawara S, Wada J, Hida K, Nakatsuka A, Eguchi J, Murakami K, et al. Serum vaspin concentrations are closely related to insulin resistance, and Rs77060950 8. Heiker JT, Kloting N, Kovacs P, Kuettner EB, Strater N, Schultz S, et al. Vaspin at Serpina12 genetically defines distinct group with higher serum levels in Japanese inhibits kallikrein 7 by serpin mechanism. Cell Mol Life Sci (2013) 70(14):2569–83. population. J Clin Endocrinol Metab (2012) 97(7):E1202–7. doi: 10.1210/jc.2011-3297 doi: 10.1007/s00018-013-1258-8 25. Jeong E, Youn BS, Kim DW, Kim EH, Park JW, Namkoong C, et al. Circadian 9. Liu S, Duan R, Wu Y, Du F, Zhang J, Li X, et al. Effects of vaspin on insulin rhythm of serum vaspin in healthy Male volunteers: relation to meals. J Clin Endocrinol resistance in rats and underlying mechanisms. Sci Rep (2018) 8(1):13542. doi: 10.1038/ Metab (2010) 95(4):1869–75. doi: 10.1210/jc.2009-1088 s41598-018-31923-3 26. Choi MS, Kim YJ, Kwon EY, Ryoo JY, Kim SR, Jung UJ. High-fat diet decreases 10. Nakatsuka A, Wada J, Iseda I, Teshigawara S, Higashio K, Murakami K, et al. energy expenditure and expression of genes controlling lipid metabolism, Vaspin is an adipokine ameliorating er stress in obesity as a ligand for cell-surface mitochondrial function and skeletal system development in the adipose tissue, along Grp78/Mtj-1 complex. Diabetes (2012) 61(11):2823–32. doi: 10.2337/db12-0232 with increased expression of extracellular matrix remodelling- and inflammation- 11. Dimova R, Tankova T. The role of vaspin in the development of metabolic and related genes. Br J Nutr (2015) 113(6):867–77. doi: 10.1017/S0007114515000100 glucose tolerance disorders and atherosclerosis. BioMed Res Int (2015) 2015:823481. 27. Kloting N, Kovacs P, Kern M, Heiker JT, Fasshauer M, Schon MR, et al. Central doi: 10.1155/2015/823481 vaspin administration acutely reduces food intake and has sustained blood glucose- 12. Weiner J, Zieger K, Pippel J, Heiker JT. Molecular mechanisms of vaspin action - lowering effects. Diabetologia (2011) 54(7):1819–23. doi: 10.1007/s00125-011-2137-1 from adipose tissue to skin and bone, from blood vessels to the brain. Adv Exp Med Biol 28. Brunetti L, Di Nisio C, Recinella L, Chiavaroli A, Leone S, Ferrante C, et al. Effects (2019) 1111:159–88. doi: 10.1007/5584_2018_241 of vaspin, chemerin and omentin-1 on feeding behavior and hypothalamic peptide gene 13. Ulbricht D, Tindall CA, Oertwig K, Hanke S, Strater N, Heiker JT. Kallikrein- expression in the rat. Peptides (2011) 32(9):1866–71. doi: 10.1016/j.peptides.2011.08.003 related peptidase 14 is the second klk protease targeted by the serpin vaspin. Biol Chem 29. Luo X, Li K, Zhang C, Yang G, Yang M, Jia Y, et al. Central administration of (2018) 399(9):1079–84. doi: 10.1515/hsz-2018-0108 vaspin inhibits glucose production and augments hepatic insulin signaling in high-Fat- 14. Zieger K, Weiner J, Kunath A, Gericke M, Krause K, Kern M, et al. Ablation of Diet-Fed rat. Int J Obes (2016) 40(6):947–54. doi: 10.1038/ijo.2016.24 kallikrein 7 (Klk7) in adipose tissue ameliorates metabolic consequences of high fat 30. Weiner J, Rohde K, Krause K, Zieger K, Kloting N, Kralisch S, et al. Brown diet-induced obesity by counteracting adipose tissue inflammation in vivo. Cell Mol Life adipose tissue (Bat) specific vaspin expression is increased after obesogenic diets and Sci (2018) 75(4):727–42. doi: 10.1007/s00018-017-2658-y cold exposure and linked to acute changes in DNA-methylation. Mol Metab (2017) 6 15. Kunath A, Weiner J, Krause K, Rehders M, Pejkovska A, Gericke M, et al. Role of (6):482–93. doi: 10.1016/j.molmet.2017.03.004 kallikrein 7 in body weight and fat mass regulation. Biomedicines (2021) 9(2):131. 31. Zieger K, Weiner J, Krause K, Schwarz M, Kohn M, Stumvoll M, et al. Vaspin doi: 10.3390/biomedicines9020131 suppresses cytokine-induced inflammation in 3t3-L1 adipocytes Via inhibition of 16. Chadt A, Leicht K, Deshmukh A, Jiang LQ, Scherneck S, Bernhardt U, et al. nfkappab pathway. Mol Cell Endocrinol (2018) 460:181–8. doi: 10.1016/j.mce.2017.07.022 Tbc1d1 mutation in lean mouse strain confers leanness and protects from diet-induced 32. Simon MM, Greenaway S, White JK, Fuchs H, Gailus-Durner V, Wells S, et al. A obesity. Nat Genet (2008) 40(11):1354–9. doi: 10.1038/ng.244 comparative phenotypic and genomic analysis of C57bl/6j and C57bl/6n mouse strains. 17. Chadt A, Immisch A, de Wendt C, Springer C, Zhou Z, Stermann T, et al. Genome Biol (2013) 14(7):R82. doi: 10.1186/gb-2013-14-7-r82 "Deletion of both rab-Gtpase-Activating proteins Tbc1d1 and Tbc1d4 in mice 33. Yam P, Albright J, VerHague M, Gertz ER, Pardo-Manuel de Villena F, Bennett eliminates insulin- and aicar-stimulated glucose transport [Corrected]. Diabetes BJ. Genetic background shapes phenotypic response to diet for adiposity in the (2015) 64(3):746–59. doi: 10.2337/db14-0368 collaborative cross. Front Genet (2020) 11:615012. doi: 10.3389/fgene.2020.615012 18. Nicholson T, Church C, Tsintzas K, Jones R, Breen L, Davis ET, et al. Vaspin 34. Vorobyev A, Gupta Y, Sezin T, Koga H, Bartsch YC, Belheouane M, et al. Gene- promotes insulin sensitivity of elderly muscle and is upregulated in obesity. J diet interactions associated with complex trait variation in an advanced intercross outbred Endocrinol (2019) 241(1):31–42. doi: 10.1530/JOE-18-0528 mouse line. Nat Commun (2019) 10(1):4097. doi: 10.1038/s41467-019-11952-w Frontiers in Endocrinology 09 frontiersin.org