Butyrate prevents visceral adipose tissue inflammation and metabolic alterations in a Friedreich’s ataxia mouse model

Summary Friedreich’s ataxia (FA) is a neurodegenerative disease resulting from a mutation in the FXN gene, leading to mitochondrial frataxin deficiency. FA patients exhibit increased visceral adiposity, inflammation, and heightened diabetes risk, negatively affecting prognosis. We investigated visceral white adipose tissue (vWAT) in a murine model (KIKO) to understand its role in FA-related metabolic complications. RNA-seq analysis revealed altered expression of inflammation, angiogenesis, and fibrosis genes. Diabetes-like traits, including larger adipocytes, immune cell infiltration, and increased lactate production, were observed in vWAT. FXN downregulation in cultured adipocytes mirrored vWAT diabetes-like features, showing metabolic shifts toward glycolysis and lactate production. Metagenomic analysis indicated a reduction in fecal butyrate-producing bacteria, known to exert antidiabetic effects. A butyrate-enriched diet restrained vWAT abnormalities and mitigated diabetes features in KIKO mice. Our work emphasizes the role of vWAT in FA-related metabolic issues and suggests butyrate as a safe and promising adjunct for FA management.


INTRODUCTION
Friedreich's ataxia (FA) is a rare neurodegenerative disease caused by the expansion of intronic trinucleotide repeat GAA from 8 to 33 repeats to >90 repeats in the FXN gene encoding frataxin protein (FXN).FXN resides in mitochondria and regulates mitochondrial iron transport and respiration. 1Apart from manifesting neurodegenerative signatures, FA patients are at higher risk to developing type 2 diabetes (T2D) and cardiomyopathy than general population, and these concur to aggravate the prognosis (extensively reviewed in the study by Tamarit et al.). 2 It is now well ascertained that lipid metabolism is altered at systemic and cellular level in FA.FA patients show accumulation of lipid droplets (LD) in fibroblasts. 37][8][9] Hepatic accumulation of fat (steatosis) in mice with a liver-specific FXN ablation 10 and altered lipid metabolism associated with increased LD in glial cells of the drosophila FA model 11 have been also observed.
White adipose tissue (WAT) is the tissue with the highest capacity to accumulate fats within LD and release them to other tissues in response to increased energetic demands.In addition to being a storage depot, WAT is a high active major endocrine organ impacting the metabolic function of several tissues and overall body metabolic homeostasis. 12Subcutaneous WAT is mainly involved in the buffering of circulating free fatty acids and triglycerides, thus exerting a protective function against systemic lipotoxicity and pathological accumulation of visceral WAT (vWAT). 13,14vWAT is present mainly in the abdominal region, and its expansion largely contributes to the onset of systemic low-grade inflammatory states that are at center stage of insulin resistance and T2D development. 15In addition to adipocytes, vWAT contains a great number of stromal vascular cells (SVCs) including endothelial cells, preadipocytes, and immune cells.Among the immune cells, macrophages play an important role in vWAT inflammation.Indeed, concomitant to pathological expansion, the ratio between the pro-inflammatory M1 macrophages and anti-inflammatory M2 macrophages increases, thus eliciting a major production of pro-inflammatory cytokines (e.g., tumor necrosis factor alpha (TNF-a)). 16Several evidence indicate that impaired oxidative function of mitochondria in vWAT adipocytes may be causally involved in its expansion and development of low-grade inflammation, insulin resistance, and T2D.Of note, the amphibolic organelles mitochondria take center stage in maintaining metabolic homeostasis in white adipocytes because of their involvement in fatty acid synthesis and esterification as well as lipid oxidation. 17,18Notably, increased visceral adiposity along with systemic chronic, low-grade inflammatory state was observed in FA patients. 19,20Also, obesogenic diet, which inevitably leads to WAT expansion, was demonstrated to exacerbate the metabolic dysfunctions caused by FXN deficiency in mice, indicating a role for FXN in the maintenance of WAT function. 21he recent approval of omaveloxolone for the treatment of FA underscores the progress made in addressing the disease. 22,23However, there is the continued need for research to develop other therapies to provide patients with safe and more diverse treatment options.Butyrate is a short-chain fatty acid, primarily produced through the fermentation of dietary fibers by beneficial bacteria in the gut.5][26] Moreover, butyrate ameliorates oxidative function of mitochondria and increases lipolysis in vWAT, thus limiting its expansion and restoring plasma leptin levels in diabetic mice. 27,28These findings highlight the potential of butyrate as a safe dietary supplement that could be utilized to counteract the T2D-like features in FA.
1][32][33] This model provides valuable insights that can guide the development of therapeutic approaches targeting metabolic dysfunction in FA.In this study, we aimed to assess the potential dysfunction of vWAT in KIKO mice and explore butyrate supplementation as a safe option for addressing the T2D aspects of the disease.

WAT of KIKO mice shows metabolic alterations and inflammatory hallmarks typical of type 2 diabetes
We previously showed that frataxin KIKO mice undergo weight gain starting at 8 months of age that was accompanied by an increase in circulating levels of leptin, 8 suggesting that WAT dysfunction occurs in this mouse model.Hence, we performed bulk RNA sequencing (RNA-seq) analysis of the main WAT depot involved in T2D development i.e., vWAT.We found 97 differentially expressed genes (FC > 1.5; FC < 0.5; FDR<0.05;41 up-, 56 downregulated) in vWAT of 8-month-old KIKO with respect to age-matched wild-type (WT) mice (Figure 1A).Functional enrichment analysis of the differentially expressed genes revealed angiogenesis and extracellular matrix as the modulated biological processes and cellular component, respectively (Figure 1B).Analysis of the cellular components also evidenced an enrichment of genes pertaining to collagen-containing extracellular matrix (Figure 1B).In the GO term extracellular matrix, the presence of genes typically expressed and secreted by macrophages and mast cells was found.These include i) the macrophage-specific Lyz2, involved in the onset of local WAT inflammation, 34 and ii) the mast cell-specific carboxypeptidase A3 and chimerase 1, involved in the digestion of extracellular matrix and fibrosis development. 35,36Overall data indicated a vWAT rearrangement toward hypovascularization, inflammation, and fibrosis.To find a causal relationship between differentially expressed genes, we built a protein-protein interaction network using the free STRING platform (https:// string-db.org/).The network depicted in Figure 1C highlights that some of the differentially expressed genes create an interconnected network with vascular endothelial growth factor A (VEGFA) representing a central hub.VEGFA is a growth factor inducing proliferation and migration of vascular endothelial cells and is essential for both physiological and pathological angiogenesis.
RNA-seq data were validated through qPCR analysis.As reported in Figure 2A, Vegfa mRNA levels resulted in downregulation along with Rasip1-a protein essential for the correct assembly and angiogenic migration of endothelial cells 37 -and Robo4-an endothelial receptor that is involved in the maintenance of endothelial barrier organization and function. 38Immunofluorescence analysis and western blot analysis of VEGFA confirmed its downregulation in KIKO mice also at protein level (Figures 2B and 2C).Expectedly, FXN protein was significantly reduced in vWAT of KIKO mice (Figure 2C).This was accompanied by the decrease of the protein content of the downstream effectors of VEGFA, such as HO-1 and IRP-1, corroborating the overall downregulation of VEGFA pathway in vWAT of KIKO mice (Figure 2C).We previously demonstrated that FXN deficiency causes lipid accumulation in brown adipose tissue. 8Western blot analysis of hallmarks of lipid loading such as the adipose triglyceride lipase and the transcription factor and activator of lipid synthesis PPARg indicated that lipid accumulation also occurs in vWAT (Figure 2C).
Hypovascularization and vWAT expansion can lead to lowering of oxygen availability, decrease of mitochondrial respiration, and metabolic reprogramming toward glycolysis and lactate production.Notably, increased levels of lactate in WAT are typical of T2D conditions. 39,40n order to pinpoint possible metabolic alterations of WAT, we isolated SVCs, and preadipocytes were induced to differentiate.Analysis of Figure 1.vWAT of KIKO mice shows alteration of gene expression (A) Volcano-plot representing differentially expressed genes between vWAT of 8-month-old WT and KIKO mice (upper panel).Heatmap representing the hierarchical clustering of significantly modulated genes (n = 4 male mice/group; fold change>1.5,<0.5; FDR<0.05)(bottom panel).(B) Functional enrichment analysis of modulated genes.(C) Protein-protein interaction network of modulated genes obtained by STRING platform.Nodes were colored according to the enriched GO term.lactate concentration in culture medium showed that mature adipocytes of KIKO mice had a higher rate of lactate production than WT mice (Figure 3A).Lactate was previously regarded as the waste product of glycolysis.Recently, it has emerged that lactate serves as a danger signal that promotes polarization of resident macrophages toward a pro-inflammatory M1-like state in the context of obesity. 39,40Based on this evidence and our RNA-seq results, we determined the leukocyte abundance in SVCs isolated from vWAT.After isolation of CD45 + cells by magnetic cell sorting, we found that leukocyte content was increased in vWAT of KIKO mice (Figure 3B).Next, we evaluated the presence of macrophage infiltrates by immunofluorescence analyses of vWAT sections.We used differentiation cluster 68 (CD68) as a marker for M1like macrophages.vWAT of KIKO mice displayed a higher number of CD68 + cells per adipocytes when compared to WAT of WT mice (Figure 3C).Moreover, some crown-like structures (consisting in several macrophages surrounding a single adipocyte) typical of inflamed WAT were observable in KIKO mice (Figure 3C).Thus, we investigated mRNA level of key factors involved in the regulation of the inflammatory response.vWAT of KIKO mice showed a higher expression of the pro-inflammatory Il1b and Il6 genes compared to WT mice.Consistent with the hypothesis of increased inflammatory processes within vWAT of KIKO mice, anti-inflammatory Il10 was found downregulated (Figure 3D).These data were confirmed by analyzing the expression of Adipoq gene encoding for adiponectin, an anti-inflammatory and antidiabetic hormone secreted by adipocytes. 41,42Adiponectin improves insulin sensitivity, inhibits macrophage-mediated inflammation, and is downregulated in T2D patients. 43In line with the observed T2D signatures, Adipoq mRNA expression result significantly decreased in vWAT of KIKO mice (Figure 3D).
To better elucidate the role of adipocytes in the observed macrophage recruitment and vWAT inflammation, we downregulated FXN in a cellular model of murine (3T3-L1) white adipocytes.In line with the results obtained in vWAT, FXN deficiency caused lipid accumulation (Figures 3E-3G), VEGFA and inflammatory marker alteration (Figure 3G), and extracellular lactate increase (Figure 3H).Interestingly, in (C) Western blot analysis of FXN and proteins related to blood vessel endothelial cell proliferation (VEGFA, HO-1, and IRP-1) and lipid metabolism (ATGL, PPARg).Vinculin or actin was used as loading control.Density of immunoreactive bands was normalized with respect to loading control.Data are expressed as mean G SD (n = 3 male mice/group; Student's t test, *p < 0.05, ****p < 0.0001).(C-H) Four-month-old male WT and KIKO mice were fed with normal diet or with diet supplemented with butyrate (+BTR) for 16 weeks up to 8 months of age.Heatmap representation of butyrate-producing bacteria in fecal samples whose relative abundance is modulated by butyrate in KIKO mice (n = 6 male mice/ group) (C).Oral glucose tolerance test (OGTT) was carried out after oral administration of 2 g of dextrose/kg body mass.The data are presented as Dglucose, which was calculated by subtracting the glucose concentrations in blood at the starting point (point 0) from the concentrations measured at subsequent time points (20, 30, 60, 120 min) following oral glucose administration.Data are expressed as mean G SD (n = 6 male mice/group; ANOVA test, *p < 0.05 vs. WT mice; **p < 0.01 vs. butyrate-untreated KIKO mice) (D).Plasma triglycerides and cholesterol levels.Data are expressed as mean G SD co-culturing conditions, KIKO-derived adipocytes but not WT-derived adipocytes were able to induce Il1b gene expression in RAW264.7 macrophages (Figure 3I), pointing to a central role of adipocytes in triggering macrophage activation upon FXN deficiency.

Dietary butyrate supplementation improves systemic and vWAT metabolism in KIKO mice
Among the typical events associated with low-grade inflammation, altered gut microbiota is also included. 44Importantly, a very strict cross talk exists between WAT and gut microbiota that synergistically contributes to maintaining body metabolic homeostasis. 45,46Metagenomic analysis of fecal samples showed an altered microbiota composition in KIKO mice (Figure 4A).Deeper analysis conducted at genus level highlighted an almost total absence of butyrate-producing bacteria in gut microbiota of KIKO mice (Figure 4B).Butyrate is predominantly produced by gut microbes, and it is now ascertained that the decrease of butyrate-producing bacteria inevitably leads to diminution of systemic butyrate availability. 47Prompted by this, we proceeded to evaluate the potential of butyrate supplementation in limiting vWAT alterations and exerting antidiabetic effects.We supplemented mice with sodium butyrate by adding it in food pellets (5 g $ kgÀ1 $ dayÀ1 at the normal daily rate of calorie intake) starting at 4 months of age and continuing for 16 weeks until 8 months of age, time in which KIKO mice manifest metabolic alterations and weight gain. 8The decision to use this plan of butyrate supplementation was based on previous research demonstrating the safety and beneficial antidiabetic effects of butyrate at this dose and administration schedule. 25Accordingly, we did not observe any alterations in food and water intake upon butyrate treatment (Figures S1A and S1B).Furthermore, the canonical bioclinical markers of tissue functions (i.e., GOT, GPT, LDH, creatinine) revealed no significant differences between the butyrate-treated groups and the control groups (Figure S1C), confirming its well-tolerated nature.
Metagenomic analysis revealed that butyrate supplementation restored the abundance of certain butyrate-producing bacteria in fecal samples from both WT and KIKO mice, with KIKO mice showing levels comparable to or even larger than those of untreated WT mice (Figure 4C).These data stimulated us to evaluate whether butyrate treatment was able to mitigate systemic metabolic alterations in KIKO mice.The oral glucose tolerance test revealed that butyrate supplementation was effective in restraining glucose intolerance in KIKO mice, as a significant recovery of normal glycemia levels was observed after 120 min from glucose administration (Figure 4D).Regarding lipidemia, we found that butyrate was effective in buffering hypertriglyceridemia but not total cholesterol levels, even though a tendency was noted (Figure 4E).
Histochemical evaluation displayed an increase in adipocyte size in vWAT of KIKO compared to WT mice, confirming vWAT expansion (Figure 4F).Notably, butyrate treatment was able to reduce the diameter of adipocytes in vWAT of KIKO mice, with vWAT adipocytes of butyrate-treated KIKO mice reaching a size comparable to that observed in vWAT adipocytes of untreated WT mice (Figure 4F).These results led us to evaluate whether vWAT dysfunction in KIKO mice could be recovered by butyrate treatment.Western blot and qPCR analyses showed that markers of lipid accumulation (PPARg, PLIN-1) and hypovascularization (VEGFA, Rasip, Roboa4) came back to control values upon butyrate treatment (Figures 4G and 4H).Accumulation of lipids depends on increased PPARg-mediated lipogenesis and inhibition of the hormone-sensitive lipolytic cascade activated by protein kinase A (PKA). 12 By western blot analysis, we observed a decrease of the level of PKA-phosphosubstrates, and butyrate was able to revert this event (Figure 4H).Overall, these results underline the beneficial effects of butyrate in maintaining lipid homeostasis and vascularization in vWAT of KIKO mice.

Dietary butyrate supplementation restrains vWAT inflammation in KIKO mice
Immunohistochemical analyses also proved that, along with the recovery of VEGFA levels, collagen content and immune cell infiltrates were reduced upon butyrate treatment (Figure 5A; Figures S1D-S1F).In parallel, butyrate restrained the upregulation of the pro-inflammatory Il1b and Cox2 genes and restored the mRNA level of the anti-inflammatory cytokine Il10 (Figure 5B).Accordingly, we sought to deeply characterize innate and adaptive immune cell dynamics by investigating potential recruitment of other immune cell populations in vWAT and whether butyrate was able to exert an impact on this.To this end, we isolated SVCs of vWAT, and the single-cell suspension was analyzed by high-dimensional flow cytometry.As expected, a higher percentage of total CD45 + leukocytes were observed in SVCs of KIKO compared to WT mice (Figure 5C).We then applied a consequential gating strategy to identify the percentages of the different cell subsets of leukocytes, i.e., macrophages (CD11b + F4/80 + CD64 + cells), neutrophils (Ly6G + Ly6C À cells), T cells (CD3 + cells), natural killer (NK) (NK1 + cells), and B cells (CD19 + cells).Interestingly, vWAT of KIKO mice showed a higher percentage of macrophages and neutrophils compared to WT mice (Figure 5C), corroborating our previous findings by immunohistochemical analyses (Figures 3C and 5A).By contrast, percentages of T and B lymphocytes as well as NK cells were not changed (Figure 5C), suggesting that NK cells and cells of the adaptive immunity do not contribute to the inflammation of vWAT in KIKO mice.Notably, butyrate was able to prevent leukocyte infiltration, and this was due to reduction of both macrophages and neutrophils (Figure 5C).Immune cells have higher capacity to produce inflammatory cytokines than adipocytes; hence, .Continued (at least n = 5 male mice/group; ANOVA test, *p < 0.01, ***p < 0.001, ****p < 0.001) (E).Representative histology images of vWAT from 4-month-old WT and KIKO mice fed with normal diet or treated with butyrate (+BTR) for 16 weeks up to 8 months of age after staining with H&E (F, left panel).Adipocyte size was represented as lipid droplet diameters (F, right panel).Data are expressed as mean G SD (n = 6 male mice/group, ANOVA test, ****p < 0.0001).RT-qPCR analysis of Pparg and angiogenic mRNAs.Data are expressed as mean G SD (n = 5 male mice/group; ANOVA test, *p < 0.05; **p < 0.01; ****p < 0.0001) (G).Western blot analysis of proteins involved in lipid metabolism (PPARg, PLIN1, phospho-PKA substrates) and VEGFA.Vinculin was used as loading control.Density of immunoreactive bands was normalized with respect to related loading control.Data are expressed as mean G SD (n = 3 male mice/group; ANOVA test, *p < 0.05, ****p < 0.001) (H).See also Figure S1.
iScience Article once recruited in vWAT by adipocyte-derived signals, immune cells could enhance the production of inflammatory mediators.Hence, we questioned whether FXN deficiency could also influence the inflammatory response in immune cells and butyrate to restrain this event.
To this end, we isolated bone-marrow-derived macrophages (BMDMs) from WT and KIKO mice.BMDMs were treated with lipopolysaccharide (LPS) to reproduce an inflammatory insult.Upregulation of the expression of the Tnfa gene was obtained after LPS stimulation in both WT and KIKO BMDMs (Figure 5D); however, even though FXN deficiency did not influence basal Tnfa expression, upon LPS treatment, Tnfa upregulation was more marked in KIKO than in WT cells.Notably, butyrate was able to significantly buffer the LPS-mediated inflammatory challenge (Figure 5D), suggesting that this molecule has an anti-inflammatory action also in myeloid cells and likely at systemic level.

Butyrate supplementation ameliorates adipocyte metabolism in KIKO mice
To have a comprehensive view of the effects of butyrate on WAT metabolism, we then performed targeted metabolomic analyses of several metabolites (about 100) pertaining to glycolysis, pentose phosphate pathway, urea and Krebs cycle, and other metabolites such as carnitines, nucleotides, amino acids, and catecholamines.Only few metabolites were significantly affected in vWAT of KIKO mice (i.e., lactate, oxaloacetate, and succinate) (Figure 6A).Among these, only lactate underwent significant reduction upon butyrate treatment.We then performed XF Seahorse real-time monitoring of cell metabolism in SVCs isolated from vWAT of KIKO mice to understand whether lactate hyperproduction observed with metabolomic analyses was associated with augmented glycolytic rate, and whether butyrate treatment was able to mitigate such phenomenon in vivo.Lactate production of KIKO SVCs was attenuated following 16 weeks treatment with butyrate (Figure 6B).Expectedly, in SVCs of KIKO mice, we found a significant decrease of spare respiratory capacity, which is a measure of the ability of mitochondria to respond to increased energy demand (Figure 6C).The monitoring of extracellular acidification rate, after addition of saturating amount of glucose, revealed that KIKO SVCs have higher glycolytic rate than WT adipocytes (Figure 6C).The addition of the mitochondrial respiration inhibitor oligomycin revealed that also the maximum glycolytic capacity was higher in KIKO than WT adipocytes (Figure 6C).Notably, butyrate was able to recover mitochondrial respiration capacity and to reduce glycolytic metabolism in KIKO SVCs (Figure 6C).We also tested the effects of butyrate on lactate production of SVCs after induction of adipocyte differentiation.As reported in Figure 6D, treatment with butyrate lowered the concentration of lactate in culture medium of KIKO adipocytes.These results point to the ability of the in vivo treatment with butyrate to shift cellular metabolism from glycolysis to mitochondrial respiration, thus avoiding the release of anti-lipolytic and pro-inflammatory lactate and cytokines in WAT.

DISCUSSION
In this study, we demonstrated that FXN deficiency leads to vWAT dysfunction, which consists of the expansion of adipocyte size, hypovascularization, production of pro-inflammatory adipokines, immune cell recruitment, and fibrosis.These findings recapitulate what is observed under T2D conditions in vWAT.
We previously demonstrated that FXN deficiency causes lipid accumulation and lower thermogenic capacity in brown adipocytes, 8 indicating a possible involvement of decreased antidiabetic activity of brown adipose tissue (BAT) in the onset of metabolic complications observed in FA patients.As in BAT, in vWAT of our FA mouse model, we found altered lipid metabolism due to increased expression of lipogenic markers and impaired lipolytic activity.It appears that vWAT is more affected than BAT in FA mice; inasmuch as besides fat accumulation, we found altered expression of genes related to angiogenesis, including VEGFA, the master regulator of this process also in vWAT. 48,49nsufficient angiogenic potential is a dominant contributor to the dysfunctional WAT, as hypoxic condition triggers chronic low-grade inflammation predominantly characterized by pro-inflammatory macrophage infiltration. 48Intriguingly, VEGFA also exerts an antidiabetic action by functioning as a promoter of insulin sensitivity and as an anti-inflammatory M2 macrophage attractant in WAT, 50 suggesting that the VEGFA downregulation observed in vWAT may participate in the development of insulin resistance and inflammation.
We demonstrated that FXN-deficient white adipocytes are per se more prone to produce pro-inflammatory cytokines than normal adipocytes, arguing that they can play an active role in macrophage recruitment and activation.In vWAT of KIKO mice, we also disclosed high degree of immune cell infiltrates, mostly macrophages and neutrophils, in association with the upregulation of fibrosis markers (e.g., collagen) and pro-inflammatory cytokine production.These findings recapitulate the fibrotic inflammatory phenotype typical of vWAT in T2D. 49,51nlarged vWAT can release hormones and other substances that may contribute to insulin resistance, cardiomyopathy, and low-grade inflammation. 52,53Apart from T2D and cardiomyopathy, signs of low-grade inflammation with high levels of inflammatory cytokine expression (B) RT-qPCR analysis of inflammatory genes.Data are expressed as mean G SD (n = 3 male mice/group; ANOVA test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).(C) SVCs were isolated from vWAT and analyzed through high-dimensional flow cytometry by using specific antibodies to detect total leukocytes, neutrophils, macrophages, T cells, NK cells, and B cells.Gating strategies to detect the immune cell subpopulations are illustrated.Data are expressed as mean percentage of positive cells GSD (n = 4 male mice/group; ANOVA test, *p < 0.05, **p < 0.01, ***p < 0.001).(D) RT-qPCR analysis of Tnfa mRNA in BMDM isolated from 8-month-old female WT and KIKO mice, and treated with LPS (500 ng/mL, 16 h) alone or in combination with butyrate (500 mM).Data are expressed as mean SD (n = 3 biological replicates; ANOVA test, ****p < 0.0001).See also Figure S1.
in circulating leukocytes were found in FA patients. 20Additionally, FXN downregulation enhanced inflammatory response in macrophages, indicating that a positive loop of inflammation occurs in vWAT of FXN-deficient mice, with adipocytes favoring the infiltration of immune cells that are more responsive to inflammatory stimuli.This evidence and our results point to the role of FXN in the maintenance of overall body immune homeostasis.
Augmented lactate levels in adipose tissue and plasma are widely reported in obese/diabetic subjects. 39,54Metabolically, we found that FXN-deficient adipocytes have increased glycolytic activity and consequent hyperproduction of lactate.The switch toward glycolytic metabolism may originate from defective mitochondrial respiration and/or scarce oxygen availability due to reduced angiogenesis and vascularization.Lactate is a redox molecule that, in adipose tissues, displays a wide range of biological effects both through its binding to membrane receptors and its transport and subsequent effect on intracellular metabolism. 55Lactate produced by adipocytes inhibits lipolysis in an autocrine/paracrine manner through inhibition of PKA. 56Accordingly, we observed a decrease of PKA activity in vWAT of KIKO mice.High lactate levels increase cardiovascular risk 54,57,58 and promote macrophage recruitment and WAT inflammation. 39We found that co-culturing macrophages with FXN-deficient adipocytes promotes a more efficient inflammatory response.This result suggests that, apart from the increased production of inflammatory cytokines, FXN-deficient adipocytes produce higher level of lactate that could be responsible for macrophage activation.
0][61][62] Some FA individuals with insulin resistance exhibit increased body weight and altered adipose tissue distribution, with increased fat deposition in the visceral region. 602][63] These changes in body weight and composition highlight the complex metabolic implications of FA and the need for comprehensive studies to better understand their underlying mechanisms and potential impact on the disease.(E) Nitrite concentration in culture medium of RAW264.7 macrophages co-cultured with adipocytes isolated from 8-month-old WT or KIKO female mice and treated with BTR (500 mM, 16 h).LPS treatment (500 ng/mL, 16 h) of RAW264.7 macrophages was used as positive control.Data are expressed as mean G SD (n = 3 biological replicates; ANOVA test, ***p < 0.001, ****p < 0.0001).See also Table S1.
In mouse models of FA, phenotypic outcomes can vary based on specific genetic modifications, tissue-specific FXN deletion, and other introduced factors, recapitulating what described previously in FA patients.What observed in this work on KIKO mice adheres with the phenotype observed in FA patients with increased body weight, visceral fat, and insulin resistance. 60By contrast, recently developed FA models such as UCLA and YG8-800 mice, featuring inducible FXN knockdown and larger triplets' expansion, exhibit early onset, severe neurodegeneration, and cardiomyopathy, but do not report T2D-like signs and experience weight loss. 64The conditional MCK FXN knockout mouse, targeting cardiac and skeletal muscle, develops pronounced cardiomyopathy and significant weight loss, 65 which is likely attributed to the decrease in lean mass due to cardiomyopathy/skeletal muscle dysfunction rather than a decrease in fat mass.
In line with KIKO mice, other models characterized by late-onset and mild neurological defects, such as YG8R and YG22R, have reported weight gain and certain aspects of T2D. 66Based on our findings, it is reasonable to propose that these animals may experience expansion and inflammation of WAT.Considering the complex metabolic implications observed in FA patients, the mild and slowly progressing neurological phenotype of the KIKO model allows to specifically investigate the metabolic aspects of FA without the confounding effects of severe neurodegeneration and cardiomyopathy.This focus on the metabolic manifestations of the disease provides valuable insights into the contribution of WAT dysfunction to the overall pathology of FA.Mouse models with mild neurological defects and cardiomyopathy, but also featuring increased body weight and visceral adiposity as KIKO mouse, emerge as more suitable models for understanding potential therapeutic targets in a significant subset of FA patients with these specific characteristics.This is particularly relevant given the adverse impact of T2D on the prognosis of FA patients.
8][69] A very strict cross talk exists between vWAT and gut microbiota that synergistically contribute to maintaining body metabolic homeostasis.Indeed, altered gut microbial ecosystems have been associated with vWAT dysfunction (i.e., expansion, inflammation, and insulin resistance), lowgrade chronic inflammation, and systemic metabolic perturbations, including T2D. 70 A decrease in butyrate-producing bacteria has been causally involved in vWAT expansion and inflammation, as well as in T2D development. 71,72To our knowledge, no attempts have been made to unravel whether the microbiota is altered in FA.Herein, we show that our FA mouse model has an altered gut microbiota composition compared to healthy mice, with a reduction in the abundance of butyrate-producing bacteria.The causes of such reduction have not been explored in the present work and the possible occurrence of gut inflammation deserves further and deeper investigation.Mitochondrial metabolism of colonocytes, by consuming O 2 , maintains the predominance of anaerobic bacteria in the gut, including butyrate-producing bacteria. 73It can be argued that mitochondrial dysfunction, which is likely to occur in FXN-deficient colonocytes, increases oxygen availability and the proliferation of facultative anaerobic bacteria, while reducing the abundance of butyrate-producing bacteria (dysbiosis).This dysbiosis could contribute to the establishment of a T2D-like inflammatory state in vWAT.
While further research is needed to fully understand the role of the microbiota in FA, our findings suggest that alterations in the microbiota may be involved in the pathophysiology of the disease and provide a potential target for therapeutic interventions.Strategies that target the gut microbiota, such as butyrate supplementation, probiotics, or prebiotics, may have potential for improving symptoms and disease progression in FA.
The decrease in butyrate-producing bacteria strongly indicates that systemic butyrate availability could be reduced in KIKO mice.Butyrate has been studied for its potential anti-inflammatory and neuroprotective effects, as well as its role in diabetes management. 74,75Our data demonstrate that butyrate supplementation may be effective in counteracting vWAT dysfunction and T2D-like symptoms in the context of FA.Specifically, at the metabolic level and similarly to what was previously described in colonocytes, 76 butyrate increased respiratory capacity of mitochondria with FXN deficiency in vWAT, while decreasing glycolytic activity and lactate production.In parallel, butyrate hindered the accumulation of fats, and this was accompanied by the maintenance of angiogenic VEGFA levels to levels comparable to those of healthy mice.As also described in other models of T2D, 77,78 butyrate supplementation ameliorated glycemic profile in KIKO mice.Regarding inflammatory signatures, butyrate reduced leukocyte infiltration (i.e., macrophages, neutrophils), and production of pro-inflammatory cytokines.
Butyrate has a wide range of pleiotropic effects and mechanisms of action.Although the precise mechanisms by which butyrate acts in our FA models have not been deeply investigated in this study, it is likely that the beneficial effects of butyrate can be mediated by the inhibition of histone deacetylases, thus epigenetically modulating the expression of genes involved in inflammation and energy metabolism. 79Butyrate can also have a direct action on mitochondria.For instance, being a short-chain fatty acid, butyrate can be directly funneled into mitochondria and enhance mitochondrial respiration and fatty acid oxidation and impede lipid accumulation. 76It can be hypothesized that the observed decrease in glycolysis and subsequent recovery of mitochondrial respiratory capacity in FXN-deficient adipocytes may be due to the redirection of reducing equivalents from butyrate b-oxidation toward succinate dehydrogenase (CII activity).Notably, measurement of CII activity in KIKO mice revealed a partial reduction of approximately 40%, compared to controls. 31A complete abrogation of complex II activity in both patients and experimental models of FA has not been reported. 31,80Notably, a compensatory activation of CII has even been found in cerebellar neurons of the YG8R mouse model. 81These findings lend support to the broad applicability of butyrate for restoring mitochondrial respiration upon FXN deficiency.
We cannot exclude that the beneficial effects of butyrate supplementation could be also dependent on the recovery of gut butyrateproducing bacteria as disclosed in FA mice.This result aligns with existing evidence that exogenous butyrate has the potential to improve gut microbiota dysbiosis in animal models of obesity with high-fat diet or systemic lupus erythematosus.This improvement is achieved by increasing the abundance of butyrate-producing bacteria. 82,83It can be hypothesized that the presence of exogenous butyrate may impact the competitive interactions among different microbial species by regulating the pH of the gut lumen. 84By maintaining a mildly acidic pH, butyrate may create a favorable environment for butyrate-producing bacteria, enabling them to outcompete with other bacteria. 85verall, our results suggest that vWAT is dysfunctional and microbiota altered in FA.Butyrate supplementation prevents vWAT expansion and inflammation as well as the development of T2D-like features in FA animals.To validate these results, further analysis of the microbiota and adipokines in the feces and plasma of FA patients is warranted.These analyses on patients and the completion of the identification of the molecular mechanisms underlying the butyrate-mediated beneficial effects will hopefully pave the way for its safe usage as an adjuvant for treating T2D-related symptoms in FA.
dose-response experiments conducted on primary adipocytes or BMDM stimulated with LPS (500 ng/mL, 16 h).These experiments demonstrated the anti-inflammatory action of the 500 mM concentration while preserving cell viability.

Bio-clinical analyses
Prior to bio-clinical analyses, male mice were starved for 12 h.After blood collection, bio-clinical analyses were performed by colorimetric methods.In particular, cholesterol, triglyceride, GOT, GPT, LDH and creatinine levels were measured in plasma through the automatized KeyLab analyser (BPCBioSed, Italy) using specific assay kits (BPCBioSed).
For the glucose tolerance test (OGTT), male mice were subjected to fasting for 12 h, followed by oral gavage with 2 g of dextrose/kg body mass.At the indicated time points, blood was collected from the tail vein and glycemia measured using a glucometer (Bayer Countur XT, Bayer Leverkusen, Germany).

Histochemical and immunohistochemistry analysis
vWAT was stained with H&E or Trichrome staining to visualize general morphology and collagen deposition, respectively.For morphometric analyses, individual slides were digitized using the NanoZoomer Digital Microscope (Hamamatsu, Japan), and digital images were analyzed using ImageJ to measure the diameter of adipocytes.Values are the means of 10 fields taken from different tissue sections per mouse.Immunohistochemical detection of VEGFA or S100A8 was performed on 3-to 5-lm-thick sections obtained from formalin-fixed tissue embedded in paraffin.Antigen retrieval was performed with Citrate Buffer (pH 6) (Dako, Glostrup, Denmark).Immunohistochemical staining was performed with anti VEGFA or anti-S100A8.Incubations with primary antibodies were carried out for 2 h.Negative controls were obtained by omitting primary antibodies.The immunohistochemical procedure was performed using the MACH 4 Universal HRP-Polymer Kit with DAB as chromogen (Biocare Medical, Concord, CA, USA).
For quantification of fibrosis area with Trichrome staining, image analysis using Qpath and ImageJ (open source) was performed in four randomly selected 10X fields.For immunohistochemistry, quantification of the percentage of positive cells in four randomly selected 5X fields was performed with Qpath.

Immunofluorescence analyses
Sections of frozen vWAT were incubated with permeabilization solution (PBS/Triton X-100 0.2% [v/v]), blocked for 1 h by a blocking solution (PBS/BSA 5% [v/v]), and then incubated for 18 h with CD68 or VEGFA primary antibodies.After washing with cold PBS, sections were incubated 1 h with Alexa Fluor 488 or -568-conjugated secondary antibodies (ThermoFisher Scientific).Nuclei were stained with 10 mg/mL Hoechst 33342 (ThermoFisher Scientific).Alexa Fluor 488 Phalloidin (ThermoFisher Scientific) was used to stain actin.Images were acquired using an Olympus IX-81 confocal microscope at 603 magnitude.Representative regions of interest were acquired using a digital 33 zoom.Fluorescence intensities were set for the control samples and were maintained for all samples.To evaluate macrophage infiltration, CD68 + cells around adipocytes were counted in each field (10 fields/sample).

Immunoblotting
Tissues or cells were lysed in RIPA buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 12 mM deoxycholic acid, 0.5% Nonidet P-40, and protease and phosphatase inhibitors).Five mg proteins were loaded on SDS-PAGE and subjected to Western blotting.Nitrocellulose membranes were incubated with primary antibodies at 1:1000 dilution.Successively, membranes were incubated with the appropriate horseradish peroxidaseconjugated secondary antibodies.Immunoreactive bands were detected by a FluorChem FC3 System (Protein-Simple, San Jose, CA, USA) after incubation of the membranes with ECL Prime Western Blotting Reagent (GE Healthcare, Pittsburgh, PA, USA).Densitometric analyses of the immunoreactive bands were performed by the FluorChem FC3 Analysis Software.

Bulk RNA-sequencing and functional enrichment analysis
Total vWAT RNA was isolated using TRIzol reagent (Invitrogen, Waltham, MA, USA) and purified using the RNeasy mini kit protocol (Qiagen, Hilden, Germany) according to the manufacturer's instructions.Isolated RNA was sequenced using an Illumina NextSeq500, and the indexed libraries were prepared from 1 mg of purified RNA with TruSeq-stranded mRNA (Illumina) Library Prep Kit according to the manufacturer's instructions.The quality of the single-end reads was evaluated using FastQC version 0.11.5 (https://www.bioinformatics.babraham.ac.uk/ projects/fastqc).All FastQC files were filtered to remove low-quality reads and adapters using Trimmomatic version 0.36. 86The resulting reads were mapped to the Mus musculus genome (GRCm38) using HISAT2 version 2.1.0 87using default parameters, and StringTie version 1.3.4days 88was applied to the BAM files obtained using HISAT2 to generate expression estimates and to quantify the transcript abundance as transcripts per kilobase per million of mapped reads.The count matrices generated by StringTie were imported in R, in which differential expression analysis was performed using Deseq2 to compare the two different conditions.Functional annotation was performed using the AnnotationDbi R library (http://bioconductor.org).Differentially expressed genes were selected with a threshold of FC > 1.5 or <0.5 (FDR <0.05).Functional enrichment analyses were performed using Rosalind version 3.36.1.2.

Targeted metabolomics
Metabolomic data were obtained using liquid chromatography coupled to tandem mass spectrometry.We used an API-3500 triple quadrupole mass spectrometer (AB Sciex, Framingham, MA, USA) coupled with an ExionLC AC System (AB Sciex).vWAT specimens were disrupted in a tissue lyser for 1 min at maximum speed in 250 mL ice-cold methanol:acetonitrile 1:1 (v/v) containing 1 ng/mL [U- 13 C 6 ]-glucose and 1 ng/mL [U- 13 C 5 ]-glutamine as internal standards.Lysates were spun at 15,000 g for 15 min at 4 C. Samples were then dried under N 2 flow at 40 C and resuspended in 5 mM ammonium acetate in methanol:water 1:1 (v/v) for subsequent analyses.
Quantification of amino acids, their derivatives, and biogenic amines was performed through previous derivatization. 92Briefly, 25 mL of each 125 mL sample were collected and dried separately under N 2 flow at 40 C. Dried samples were resuspended in 50 mL phenyl-isothiocyanate, EtOH, pyridine, and water 5%:31.5%:31.5%:31.5%,then incubated for 20 min at RT, dried under N 2 flow at 40 C for 90 min, and finally resuspended in 100 mL 5 mM ammonium acetate in MeOH/H 2 O 50:50.Quantification of different amino acids was performed using a C18 column (Biocrates, Innsbruck, Austria) maintained at 50 C.The mobile phases for positive ion mode analysis were phase A: 0.2% formic acid in water and phase B: 0.2% formic acid in acetonitrile.The gradient was T 0 : 100%A, T 5.5 : 5%A, T 7 : 100%A with a flow rate of 500 mL/min.All metabolites analyzed in the described protocols were previously validated by pure standards and internal standards were used to check instrument sensitivity.
Quantification of energy metabolites and cofactors was performed using a cyano-phase LUNA column (50 mm 3 4.6 mm, 5 mm; Phenomenex) with a 5.5 min run in negative ion mode with two separated runs.Protocol A: mobile phase A was water and phase B was 2 mM ammonium acetate in MeOH, with a gradient of 10% A and 90% B for all analyses and a flow rate of 500 mL/min.Protocol B: mobile phase A was water and phase B was 2 mM ammonium acetate in MeOH, with a gradient of 50% A and 50% B for all analyses and a flow rate of 500 mL/min.
Acylcarnitine quantification was performed on the same samples using a Varian Pursuit XRs Ultra 2.8 Diphenyl column (Agilent).Samples were analyzed in a 9 min run in positive ion mode.Mobile phases were A: 0.1% formic acid in H 2 O, B: 0.1% formic acid in MeOH, and the gradient was T 0 : 35%A, T 2.0 : 35%A, T 5.0 : 5%A, T 5.5 : 5%A, T 5.51 : 35%A, T 9.0 : 35%A with a flow rate of 300 mL/min.
MultiQuant software (version 3.0.3,AB Sciex) was used for data analysis and peak review of chromatograms.Raw areas were normalized to the areas' median.Obtained data were then compared to controls and expressed as fold change.Raw data are reported in Table S1.

16s sequencing analysis
Upon collection, fecal pellets were immediately preserved by freezing in liquid nitrogen and then stored at À80 C. To extract fecal nucleic acid, an E.Z.N.A. stool DNA kit (OMEGA, Bio-tek) was used.The bacterial 16S rRNA gene was amplified from total DNA following the Illumina 16S Metagenomic Sequencing Library Preparation instructions, targeting the V3-V4 hypervariable region amplicon by PCR with universal primers containing Illumina adapters reported in Klindworth et al. 93 The resulting amplicon was purified and subjected to a second PCR to barcode the libraries using the Illumina dual-index system before a final purification step.The pooled libraries were then sequenced using paired-end sequencing (2 3 300 cycles) on an Illumina MiSeq device according to the manufacturer's specifications.The resulting sequence data obtained as FASTq files were analyzed using 16S Metagenomics GAIA 2.0 software, which performs quality control on the reads/pairs (i.e., trimming, clipping, and adapter removal) through FastQC and BBDuk before mapping them with BWA-MEM against NCBI databases.The average number of reads per sample was 220,507.1 (SD G 104,675.1).

Statistical analysis
Data were expressed as the mean G SD.The exact numbers of replicates are given in each figure legend.A two-tailed unpaired Student's t test was performed to assess the statistical significance between two groups.ANOVA analysis of variance followed by Dunnett's (comparisons relative to controls), or Tukey's (multiple comparisons among groups) post hoc tests was used to compare three or more groups.Statistical analyses were performed using GraphPad Prism 9 (GraphPad Software Inc., San Diego, CA, USA).In all cases, a p value of 0.05 was set as the significance threshold.

Figure 2 .
Figure 2. vWAT of KIKO mice shows alteration of angiogenesis and lipid metabolism (A) RT-qPCR analysis of angiogenic genes in 8-month-old WT and KIKO mice.Data are expressed as mean G SD (n = 4 male mice/group; Student's t test, ***p < 0.001).(B) Representative immunofluorescence analysis of VEGFA (red).Phalloidin was used to stain F-actin (green) and Hoechst 33342 (blue) to visualize nuclei (Magnification 2003; scale bar 10 mm).(C)Western blot analysis of FXN and proteins related to blood vessel endothelial cell proliferation (VEGFA, HO-1, and IRP-1) and lipid metabolism (ATGL, PPARg).Vinculin or actin was used as loading control.Density of immunoreactive bands was normalized with respect to loading control.Data are expressed as mean G SD (n = 3 male mice/group; Student's t test, *p < 0.05, ****p < 0.0001).

Figure 3 .
Figure 3. vWAT of KIKO mice shows signs of macrophage infiltration and inflammation (A) Spectrophotometric analysis of lactate released in culture medium of adipocytes differentiated from SVCs isolated from vWAT of 8-month-old WT and KIKO mice.Data are expressed as mean G SD (n = 4 female mice/group; Student's t test, **p < 0.01).(B) Immune cells (CD45 + ) in stromal vascular fractions isolated from vWAT of 8-month-old WT and KIKO mice by magnetic cell sorting.Data are expressed as mean G SD (n = 7 mice/group; 3 male, 4 female; Student's t test, ****p < 0.001).(C) Representative immunofluorescence analysis of CD68 + M1 macrophage infiltrates (red) in vWAT of 8-month-old WT and KIKO mice.Phalloidin was used to stain F-actin (green) and Hoechst 33342 (blue) to visualize nuclei (Magnification 2003; scale bar 10 mm).White arrows indicate the presence of crown-like structures around adipocytes.Quantification of macrophage number per adipocytes is reported (right panel).Data are expressed as mean G SD (n = 4 male mice/group; Student's t test, **p < 0.01).(D) RT-qPCR analysis of inflammatory genes in vWAT of 8-month-old WT and KIKO mice.Data are expressed as mean G SD (n = 4 male mice/group; Student's t test, ***p < 0.001, ****p < 0.0001).(E-H) Murine 3T3-L1 adipocytes were transfected with a shRNA against FXN (FXN-) or with a Scr shRNA.Western blotting analysis of FXN and actin (loading control).Immunoblot reported is representative of three giving similar results (E).Intracellular lipid content determined by spectrophotometric measurement of eluted oil red O. Data are expressed mean G SD (n = 6 biological replicates; Student's t test, *p < 0.05) (F).qPCR analysis of Fxn, inflammatory (Il1b, Il6, Il10, Adipoq), and Vegfa mRNAs.Data are expressed as mean G SD (n = 4 biological replicates; Student's t test, *p < 0.05, ***p < 0.0001, ****p < 0.0001) (G).Spectrophotometric analysis of lactate released in culture medium.Data are expressed as mean G SD (n = 4 biological replicates; Student's t test, **p < 0.01) (H).(I) qPCR analysis of Il1b mRNA in RAW264.7 macrophages co-cultured with adipocytes differentiated form SVCs isolated from vWAT of 8-month-old female WT or KIKO mice.LPS treatment was used as positive control.Data are expressed as mean G SD (n = 3 biological replicates; Student's t test, ***p < 0.001, ****p < 0.0001).

Figure 4 .
Figure 4. Butyrate supplementation counteracts development of T2D-like symptoms in KIKO miceFour-month-old male WT and KIKO mice were fed with normal diet or with diet supplemented with butyrate (+BTR) for 16 weeks up to 8 months of age.(A and B) Relative abundance of microbiota phyla (SILVA database analysis) (A) and heatmap representation of butyrate-producing bacteria (genera in terms of relative abundance) (B) in fecal samples of 8-month-old WT and KIKO mice (n = 12 male mice/group).(C-H) Four-month-old male WT and KIKO mice were fed with normal diet or with diet supplemented with butyrate (+BTR) for 16 weeks up to 8 months of age.Heatmap representation of butyrate-producing bacteria in fecal samples whose relative abundance is modulated by butyrate in KIKO mice (n = 6 male mice/ group) (C).Oral glucose tolerance test (OGTT) was carried out after oral administration of 2 g of dextrose/kg body mass.The data are presented as Dglucose, which was calculated by subtracting the glucose concentrations in blood at the starting point (point 0) from the concentrations measured at subsequent time points(20, 30, 60, 120 min) following oral glucose administration.Data are expressed as mean G SD (n = 6 male mice/group; ANOVA test, *p < 0.05 vs. WT mice; **p < 0.01 vs. butyrate-untreated KIKO mice) (D).Plasma triglycerides and cholesterol levels.Data are expressed as mean G SD

Figure 5 .
Figure 5. Butyrate treatment prevents the onset of inflammation in vWAT of KIKO mice Four-month-old male WT and KIKO mice were fed with normal diet or with diet supplemented with butyrate (+BTR) for 16 weeks up to 8 months of age.(A) Representative histology images of vWAT to detect collagen (Masson trichrome staining), VEGFA (staining with VEGFA antibody), or immune cell infiltrates (staining with S100a8 antibody).Arrowheads indicate fibrotic septa, VEGFA positive staining, or inflammatory infiltrates with scattered cells (Magnification 2003; scale bar 10 mm).(B) RT-qPCR analysis of inflammatory genes.Data are expressed as mean G SD (n = 3 male mice/group; ANOVA test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).(C) SVCs were isolated from vWAT and analyzed through high-dimensional flow cytometry by using specific antibodies to detect total leukocytes, neutrophils, macrophages, T cells, NK cells, and B cells.Gating strategies to detect the immune cell subpopulations are illustrated.Data are expressed as mean percentage of positive cells GSD (n = 4 male mice/group; ANOVA test, *p < 0.05, **p < 0.01, ***p < 0.001).(D) RT-qPCR analysis of Tnfa mRNA in BMDM isolated from 8-month-old female WT and KIKO mice, and treated with LPS (500 ng/mL, 16 h) alone or in combination with butyrate (500 mM).Data are expressed as mean SD (n = 3 biological replicates; ANOVA test, ****p < 0.0001).See also FigureS1.