Advertisement

Mitochondrial respiration is decreased in visceral but not subcutaneous adipose tissue in obese individuals with fatty liver disease

  • Kalliopi Pafili
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany
    Search for articles by this author
  • Sabine Kahl
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany

    Department of Endocrinology and Diabetology, Medical Faculty and University Hospital Düsseldorf, Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany
    Search for articles by this author
  • Lucia Mastrototaro
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany
    Search for articles by this author
  • Klaus Strassburger
    Affiliations
    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany

    Institute for Biometrics and Epidemiology, German Diabetes Center, Leibniz Center for Diabetes Research, Heinrich-Heine-University, 40225, Düsseldorf, Germany
    Search for articles by this author
  • Dominik Pesta
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Aerospace Center (DLR), Institute of Aerospace Medicine, 51147, Cologne, Germany

    Center for Endocrinology, Diabetes and Preventive Medicine (CEDP), University Hospital Cologne, 50931, Cologne, Germany

    Cologne Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), 50931 Cologne, Germany
    Search for articles by this author
  • Christian Herder
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany

    Department of Endocrinology and Diabetology, Medical Faculty and University Hospital Düsseldorf, Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany
    Search for articles by this author
  • Jennifer Pützer
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany
    Search for articles by this author
  • Bedair Dewidar
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany
    Search for articles by this author
  • Mona Hendlinger
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany
    Search for articles by this author
  • Cesare Granata
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany
    Search for articles by this author
  • Nina Saatmann
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany
    Search for articles by this author
  • Aslihan Yavas
    Affiliations
    Institute of Pathology, University Hospital and Heinrich-Heine-University, 40225, Düsseldorf, Germany
    Search for articles by this author
  • Sofiya Gancheva
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany

    Department of Endocrinology and Diabetology, Medical Faculty and University Hospital Düsseldorf, Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany
    Search for articles by this author
  • Geronimo Heilmann
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany
    Search for articles by this author
  • Irene Esposito
    Affiliations
    Institute of Pathology, University Hospital and Heinrich-Heine-University, 40225, Düsseldorf, Germany
    Search for articles by this author
  • Matthias Schlensak
    Affiliations
    Department of General and Visceral Surgery, Neuwerk Hospital, 41066, Mönchengladbach, Germany
    Search for articles by this author
  • Michael Roden
    Correspondence
    Corresponding author. Address: Department of Endocrinology and Diabetology, Medical Faculty and University Hospital Düsseldorf, Heinrich-Heine-University, Düsseldorf, c/o German Diabetes Center at Heinrich-Heine University, Auf‘m Hennekamp 65, 40225, Düsseldorf, Germany. Tel.: +49 211 3382 201, fax +49 211 3382 691. (M. Roden).
    Affiliations
    Institute for Clinical Diabetology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany

    German Center for Diabetes Research, Partner Düsseldorf, 85764, München-Neuherberg, Germany

    Department of Endocrinology and Diabetology, Medical Faculty and University Hospital Düsseldorf, Heinrich-Heine-University Düsseldorf, 40225, Düsseldorf, Germany
    Search for articles by this author
Open AccessPublished:August 18, 2022DOI:https://doi.org/10.1016/j.jhep.2022.08.010

      Highlights

      • Differences in mitochondrial features are observed between SAT and VAT in human obesity.
      • VAT respiration is downregulated in obese humans with fatty liver.
      • VAT respiration is also decreased in obese humans with non-alcoholic steatohepatitis.
      • Lower VAT respiration is coupled with lower adipose tissue insulin sensitivity.

      Background & Aims

      Adipose tissue dysfunction is involved in the development of insulin resistance and is responsible for excessive lipid delivery to other organs such as the liver. We tested the hypothesis that impaired mitochondrial function is a common feature of subcutaneous (SAT) and visceral adipose tissue (VAT), but may differently contribute to adipose tissue insulin resistance (IR) in obesity, non-alcoholic fatty liver (NAFL) and steatohepatitis (NASH).

      Methods

      In this cross-sectional study, we analyzed tissue-specific insulin sensitivity using stable isotope dilution and hyperinsulinemic-normoglycemic clamp tests. We also assessed mitochondrial respiration, mRNA and protein expression, and tissue morphology in biopsies of SAT and VAT from obese humans without NAFL, with NAFL or with NASH (n = 22/group).

      Results

      Compared to individuals without liver disease, persons with NAFL and NASH had about 30% (p = 0.010) and 33% (p = 0.002) lower maximal mitochondrial respiration, respectively, in VAT, but not in SAT. The lower maximal mitochondrial respiration of VAT was associated with lower adipose tissue insulin sensitivity (β = 0.985, p = 0.041) and with increased VAT protein expression of tumor necrosis factor A across all groups (β = −0.085, p = 0.040). VAT from individuals with NASH was characterized by lower expression of oxidative phosphorylation complex IV (p = 0.042) and higher mRNA expression of the macrophage marker CD68 (p = 0.002) than VAT from participants without NAFL.

      Conclusions

      Humans with non-alcoholic fatty liver disease have distinct abnormalities of VAT energy metabolism, which correlate with adipose tissue dysfunction and may favor progression of NAFL to NASH.

      Lay summary

      Adipose tissue (commonly called body fat) can be found under the skin (subcutaneous) or around internal organs (visceral). Dysfunction of adipose tissue can cause insulin resistance and lead to excess delivery of fat to other organs such as the liver. Herein, we show that dysfunction specifically in visceral adipose tissue was associated with fatty liver disease.

      Clinical trial number

      NCT01477957.

      Graphical abstract

      Keywords

      Introduction

      Adipose tissue (AT) contributes to the orchestration of whole-body metabolic homeostasis via release of free fatty acids (FFAs), hormones, cytokines and exosomes.
      • Crewe C.
      • Scherer P.E.
      Intercellular and interorgan crosstalk through adipocyte extracellular vesicles.
      ,
      • Bodis K.
      • Roden M.
      Energy metabolism of white adipose tissue and insulin resistance in humans.
      During development of obesity, AT dysfunction in association with local insulin resistance (IR) and inflammation favors ectopic fat deposition, whole-body IR and ultimately type 2 diabetes (T2D).
      • Roden M.
      • Shulman G.I.
      The integrative biology of type 2 diabetes.
      In the context of IR and fasting hypertriglyceridemia, circulating FFAs derived from AT lipolysis contribute to about 60% of hepatic triglyceride synthesis in overweight/obese humans with non-alcoholic fatty liver disease (NAFLD).
      • Donnelly K.L.
      • Smith C.I.
      • Schwarzenberg S.J.
      • Jessurun J.
      • Boldt M.D.
      • Parks E.J.
      Sources of fatty acids stored in liver and secreted via lipoproteins in patients with nonalcoholic fatty liver disease.
      Recent studies suggest that AT dysfunction with increased lipogenic substrate flux may be key to NAFLD development.
      • Ter Horst K.W.
      • Vatner D.F.
      • Zhang D.
      • Cline G.W.
      • Ackermans M.T.
      • Nederveen A.J.
      • et al.
      Hepatic insulin resistance is not pathway selective in humans with nonalcoholic fatty liver disease.
      White AT dysfunction is characterized by abnormal adipokine release and progressive fibrosis, but may also include altered mitochondrial function.
      • Kusminski C.M.
      • Scherer P.E.
      Mitochondrial dysfunction in white adipose tissue.
      Although subcutaneous AT (SAT) represents the main source of circulating FFAs and can exhibit lower expression of genes regulating the mitochondrial respiratory chain in obesity and T2D,
      • Bodis K.
      • Roden M.
      Energy metabolism of white adipose tissue and insulin resistance in humans.
      lower citrate synthase activity (CSA) has also been reported in the visceral AT (VAT) of obese humans.
      • Christe M.
      • Hirzel E.
      • Lindinger A.
      • Kern B.
      • von Flue M.
      • Peterli R.
      • et al.
      Obesity affects mitochondrial citrate synthase in human omental adipose tissue.
      VAT can also secrete hormones and proinflammatory cytokines that interact with resident liver macrophages and other immune cells involved in inflammation-induced IR.
      • Kahn D.E.
      • Bergman B.C.
      Keeping it local in metabolic disease: adipose tissue paracrine signaling and insulin resistance.
      Secretome analyses showed that proteins involved in the regulation of cellular processes, in inflammatory responses or in extracellular matrix organization are released prominently and abundantly by human VAT.
      • Alvarez-Llamas G.
      • Szalowska E.
      • de Vries M.P.
      • Weening D.
      • Landman K.
      • Hoek A.
      • et al.
      Characterization of the human visceral adipose tissue secretome.
      ,
      • Roca-Rivada A.
      • Bravo S.B.
      • Perez-Sotelo D.
      • Alonso J.
      • Castro A.I.
      • Baamonde I.
      • et al.
      CILAIR-based secretome analysis of obese visceral and subcutaneous adipose tissues reveals distinctive ECM remodeling and inflammation mediators.
      A recent study also identified distinct properties of VAT stem cells, promoting fibrotic remodeling under obesogenic stimuli in mammals.
      • Nahmgoong H.
      • Jeon Y.G.
      • Park E.S.
      • Choi Y.H.
      • Han S.M.
      • Park J.
      • et al.
      Distinct properties of adipose stem cell subpopulations determine fat depot-specific characteristics.
      Collectively, there is evidence that VAT could contribute to hepatic dysregulation in metabolic diseases. However, little is known about tissue-specific differences between SAT and VAT parameters contributing to mitochondrial functionality, oxidative and endoplasmic reticulum (ER) stress.
      NAFLD, ranging from fatty liver (NAFL), steatohepatitis (NASH) to fibrosis and cirrhosis, correlates tightly with IR.
      • Pafili K.
      • Roden M.
      Nonalcoholic fatty liver disease (NAFLD) from pathogenesis to treatment concepts in humans.
      However, hepatic mitochondrial respiration is not uniformly impaired in obese humans with NAFLD. Indeed, hepatic mitochondrial respiration is even higher in obese humans with NAFL compared to those with NASH.
      • Koliaki C.
      • Szendroedi J.
      • Kaul K.
      • Jelenik T.
      • Nowotny P.
      • Jankowiak F.
      • et al.
      Adaptation of hepatic mitochondrial function in humans with non-alcoholic fatty liver is lost in steatohepatitis.
      As VAT has been considered a major culprit in the development of NAFLD,
      • Gastaldelli A.
      • Cusi K.
      • Pettiti M.
      • Hardies J.
      • Miyazaki Y.
      • Berria R.
      • et al.
      Relationship between hepatic/visceral fat and hepatic insulin resistance in nondiabetic and type 2 diabetic subjects.
      it is interesting to examine whether oxidative capacity in VAT shows similar alterations to that observed in the liver.
      We tested the hypothesis that (i) impaired mitochondrial function is a common feature of SAT and VAT, but (ii) may differently contribute to AT IR in obesity-related NAFL and NASH. Specifically, VAT could present with abnormal mitochondrial respiration in NAFL (OBE-NAFL) and even more so in NASH (OBE-NASH) when compared to obese humans without NAFL (OBE-CON).

      Materials and methods

      Study cohort

      This cross-sectional study included all consecutive obese volunteers undergoing bariatric surgery within the BARIA_DDZ cohort (supplementary CTAT table), recruited between March 2015 and February 2020, who had a complete data set for mitochondrial respiration in SAT and VAT (Fig. S1). Based on liver histology, participants were stratified into three groups: OBE without (OBE-CON, n = 22) or with NAFL (OBE-NAFL, n = 22) or NASH (OBE-NASH, n = 22), with similar age, sex, body weight and BMI. All volunteers showing at least grade 1 steatosis were classified as having NAFL and those with at least grade 1 steatosis plus at least grade 1 ballooning and at least grade 1 lobular inflammation were classified as having NASH.
      • Bedossa P.
      • Poitou C.
      • Veyrie N.
      • Bouillot J.L.
      • Basdevant A.
      • Paradis V.
      • et al.
      Histopathological algorithm and scoring system for evaluation of liver lesions in morbidly obese patients.
      All participants with NASH exhibited profound steatosis, liver cell ballooning and lobular inflammation (Table S1, Fig. S2). Specific causes of liver disease were excluded based on medical history, lab tests and histological features. Some participants from the OBE-CON (n = 2), OBE-NAFL (n = 5) and OBE-NASH (n = 9) groups had T2D with near-normoglycemic control by lifestyle modification or antihyperglycemic medication (Table S2).
      Before inclusion, all participants gave written informed consent to the protocol, which was approved by the institutional review board of Heinrich-Heine-University Düsseldorf and conducted in accordance with the ethical standards as set down in the 1964 Declaration of Helsinki and its last amendments of 2013.

      Statistical analyses

      Statistical analyses are described in the supplementary materials and methods.
      For further details regarding the materials and methods used, please refer to the CTAT table and supplementary information.

      Results

      Individuals with NASH exhibit greater AT IR

      All three groups had comparable age, sex and body composition as well as circulating FFA levels (Table 1, Fig. S3A-C). OBE-NASH had higher fasting glycemia than both other groups and higher glycosylated hemoglobin A1c and serum alanine aminotransferase activity than OBE-CON (Table 1).
      Table 1Characteristics of the participants.
      VariableOBE-CONOBE-NAFLOBE-NASH
      Age (years)35 (31–45)40 (35–44)45 (32–51)
      Female/male (n)20/218/418/4
      Body weight (kg)155 (130–165)150 (139–162)148 (136–173)
      BMI (kg/m2)53 (46–59)51 (47–53)51 (47–55)
      Waist circumference (cm)132 (124–140)138 (128–143)137 (128–144)
      Hip circumference (cm)148 (142–162)153 (140–160)145 (141–158)
      HbA1c (%)5.4 (5.2–5.6)5.5 (5.3–5.9)6.0 (5.2–6.7)a
      Fasting glucose (mg/dl)80 (76–93)88 (83–93)98 (85–124)a,b
      Fasting insulin (mU/L)17 (11–33)21 (16–30)26 (21–33)
      Fasting triglycerides (mg/dl)99 (81–124)125 (90–158)152 (107–178)
      Fasting FFA (μmol/L)536 (418–658)635 (490–794)705 (585–792)
      hsCRP (mg/dl)0.84 (0.60–2.67)0.84 (0.48–1.32)0.74 (0.44–0.94)
      ALT (U/L)25 (17–32)30 (20–41)41 (30–55)a
      ALT, alanine aminotransferase; FFA, free fatty acids; HbA1c, hemoglobin A1c; hsCRP, high-sensitivity C-reactive protein; OBE-CON, obese humans without non-alcoholic fatty liver; OBE-NAFL, obese humans with NAFL; OBE-NASH, obese humans with NASH.
      Data are median (IQR) and absolute numbers, as applicable. a p <0.05 vs. OBE-CON, b p <0.05 vs. OBE-NAFL, one-way ANOVA corrected for multiple comparisons with Tukey-Kramer multiple comparisons test.
      Compared to OBE-NASH, fasting AT IR was 34% lower in OBE-CON and tended to be 30% lower in OBE-NAFL (p = 0.073) (Fig. 1A). Fasting hepatic IR was similar between groups (Fig. 1B).
      Figure thumbnail gr1
      Fig. 1Tissue-specific IR and substrate oxidation in OBE-CON, OBE-NAFL or OBE-NASH.
      (A) AT IR and (B) hepatic IR during fasting as well as (C) AT IS, (D) insulin-stimulated GOX and (E) insulin-suppressed LOX rates during clamp, (F) metabolic flexibility during fasting and clamp conditions. Means ± SEM, levels of significance of each significant and borderline significant difference are marked in the figure (one-way ANOVA corrected for multiple comparisons with Tukey-Kramer multiple comparisons test). AT, adipose tissue; AU, arbitrary units; FFA, free fatty acid; GOX, glucose oxidation; IR, insulin resistance; IS, insulin sensitivity; LOX, lipid oxidation; OBE-CON, obese humans without non-alcoholic fatty liver; OBE-NAFL, obese humans with NAFL; OBE-NASH, obese humans with NASH.
      During hyperinsulinemic-euglycemic clamps, reflecting postprandial metabolic conditions, AT insulin sensitivity (IS) tended to be lower in OBE-NASH than in OBE-CON (p = 0.080) (Fig. 1C), while hepatic IS was similar between groups (Fig. S3D). Whole-body IS tended to be lower in OBE-NASH than in OBE-CON (p = 0.058) (Fig. S3E), mainly due to decreased insulin-stimulated glucose oxidation rates (Fig. 1D), but not non-oxidative glucose disposal (Fig. S3F). Insulin-suppressed lipid oxidation rates were 51% and 59% lower in OBE-CON than in OBE-NAFL and OBE-NASH, respectively (Fig. 1E), resulting in decreased metabolic flexibility in OBE-NASH compared to OBE-CON (Fig. 1F).

      Individuals with NASH exhibit lower mitochondrial respiration, as well as complex IV expression in VAT, but not SAT

      Using a substrate-uncoupler-inhibitor protocol (Fig. S4), VAT oxidative phosphorylation (OXPHOS) capacity (P) with electron input through electron transferring flavoprotein (ETF) ([ETF]P), was 32% and 33% lower in OBE-NAFL and OBE-NASH, respectively, than in OBE-CON (Fig. 2A), with similar differences for ETF and complex (C)I combined ([ETF+CI]P). Maximal ADP-stimulated mitochondrial respiration with convergent electron input through ETF, CI and CII combined, [ETF+CI+II]P, was 30% and 33% lower in OBE-NAFL and OBE-NASH than in OBE-CON (Fig. 2A). Instead, in SAT, mitochondrial respiration was similar in all groups (Fig. 2B). Although mitochondrial respiration can decline with aging,
      • Phielix E.
      • Szendroedi J.
      • Roden M.
      Mitochondrial function and insulin resistance during aging: a mini-review.
      adjustment for age did not affect results of mitochondrial respiration in both AT compartments (data not shown). Both CSA and mitochondrial DNA (mtDNA) copy number were similar between groups in both compartments (Fig. 2C-D). Consequently, mitochondrial respiration normalized by mtDNA remained lower in VAT of OBE-NAFL and OBE-NASH than in OBE-CON and similar in SAT of all groups (Fig. S5A-B).
      Figure thumbnail gr2
      Fig. 2Mitochondrial respiration and content in OBE-CON, OBE-NAFL or OBE-NASH.
      Mitochondrial respiration in (A) VAT and (B) SAT, with electron flow through ETF. ADP-stimulated respiration ([ETF]P;[ETF+CI]P;[ETF+CI+II]P), leak respiration (L) ([ETF+CI+II]L) and maximal uncoupled respiration ([ETF+CI+II]E), (C) CSA and mtDNA copy number in VAT and (D) SAT, (E) protein expression of OXPHOS CI-CV in VAT and (F) SAT. Means ± SEM, levels of significance of each significant difference are marked in the figure (one-way ANOVA corrected for multiple comparisons with Tukey-Kramer multiple comparisons test). AU, arbitrary units; C, complex; CSA, citrate synthase activity; E, electron transport system capacity; ETF, electron transferring flavoprotein; mtDNA, mitochondrial DNA; OBE-CON, obese humans without non-alcoholic fatty liver; OBE-NAFL, obese humans with NAFL; OBE-NASH, obese humans with NASH; OXPHOS, oxidative phosphorylation; P, oxidative phosphorylation capacity; SAT, subcutaneous adipose tissue; VAT, visceral adipose tissue.
      To further characterize differences in mitochondrial respiration, we assessed mitochondrial coupling efficiency by calculating respiratory control ratio (RCR) and leak control ratio (LCR), which were similar between groups in both SAT and VAT (Fig. S5C-F).
      On the other hand, the protein expression of OXPHOS CIV was lower in VAT of OBE-NASH than OBE-CON, but this difference was not seen in SAT (Fig. 2E-F). Also, the sum of all bands related to OXPHOS CI-CV tended to be lower in OBE-NASH than in OBE-CON (p = 0.062) only in VAT (OBE-CON median: 8.5 [first quartile: 7.0, third quartile: 9.7] arbitrary units [AU], OBE-NAFL 9.4 [6.2, 12.2] AU, OBE-NASH 6.8 [5.7, 8.8] AU), but not SAT (OBE-CON 7.3 [4.9, 10.5] AU, OBE-NAFL 6.1 [4.3, 7.8] AU, OBE-NASH 6.6 [4.8, 7.8] AU).
      It is conceivable that adipocyte size and number drives the observed decreased mitochondrial respiration in VAT.
      • Honecker J.
      • Weidlich D.
      • Heisz S.
      • Lindgren C.M.
      • Karampinos D.C.
      • Claussnitzer M.
      • et al.
      A distribution-centered approach for analyzing human adipocyte size estimates and their association with obesity-related traits and mitochondrial function.
      ,
      • Politis-Barber V.
      • Brunetta H.S.
      • Paglialunga S.
      • Petrick H.L.
      • Holloway G.P.
      Long-term, high-fat feeding exacerbates short-term increases in adipose mitochondrial reactive oxygen species, without impairing mitochondrial respiration.
      However, analysis of AT morphology in a subgroup of the study population revealed that both adipocyte number and mean adipocyte size were similar between all groups and in both compartments (Fig. S6A-D).

      Levels of one biomarker of mitophagy are higher only in SAT of individuals with NAFL compared to those with NASH

      Maintaining mitochondrial respiration requires a balance between biogenesis, reserve capacity, mitophagy and the mitophagy-induced recruitment of the autophagic machinery.
      • Koliaki C.
      • Roden M.
      Alterations of mitochondrial function and insulin sensitivity in human obesity and diabetes mellitus.
      In both AT compartments, protein biomarkers of macro-autophagy, including the autophagy-related gene 5, p62 and the microtubule-associated protein 1 light chain 3 A/B II/I ratio, were similar between groups (Fig. S7A-B).
      Despite no differences in VAT (Fig. 3A), OBE-NAFL exhibited higher levels of the phosphorylated Parkin (pParkin) than OBE-NASH in SAT (Fig. 3B). Protein markers of mitochondrial fusion (optic atrophy 1, mitofusin 1 and 2) and fission (dynamin-related protein 1 and serine 616 phosphorylated DRP1 and their ratio) were similar between groups in both compartments (Fig. S7C-F).
      Figure thumbnail gr3
      Fig. 3Mitophagy and mitochondrial biogenesis in OBE-CON, OBE-NAFL or OBE-NASH.
      (A) Protein content of mitophagy markers (Parkin; phosphorylated Parkin (pParkin); pParkin/Parkin ratio) in VAT and (B) SAT, (C) transcript levels of regulators of mitochondrial biogenesis in VAT and (D) in SAT. Means ± SEM, levels of significance of each significant difference are marked in the figure (one-way ANOVA corrected for multiple comparisons with Tukey-Kramer multiple comparisons test). AU, arbitrary units; NRF1, nuclear respiratory factor 1; OBE-CON, obese humans without non-alcoholic fatty liver; OBE-NAFL, obese humans with NAFL; OBE-NASH, obese humans with NASH; PPARGC1A, peroxisomal proliferator activated receptor γ coactivator 1A; SAT, subcutaneous adipose tissue; TFAM, mitochondrial transcriptional factor A; VAT, visceral adipose tissue.
      The same applied for the AT mRNA expression of transcription factors regulating mitochondrial biogenesis, including nuclear respiratory factor (NRF)1, mitochondrial transcription factor A (TFAM) and PPARGC1A (peroxisomal proliferator activated receptor γ-coactivator 1A) (Fig. 3C-D).

      Selective upregulation of inflammation in VAT of NAFL and of ER stress in SAT of NASH

      Next, we examined the role of AT oxidative stress and inflammation on NAFLD.
      • Roden M.
      • Shulman G.I.
      The integrative biology of type 2 diabetes.
      There were neither differences in lipid peroxidation, as assessed by thiobarbituric acid reactive substances (TBARS) (Fig. S8A-B) nor measures of anti-oxidant defense, as assessed by the protein expression of superoxide dismutase 1 and catalase, between groups, in either VAT or SAT (Fig. S8C-F). Despite similar interleukin 6 (IL6) expression (Fig. 4A-B), the mRNA expression of tumor necrosis factor A (TNFA) was higher only in VAT of OBE-NAFL compared to OBE-CON (Fig. 4A-B).
      Figure thumbnail gr4
      Fig. 4Inflammation, ER stress and macrophage markers in OBE-CON, OBE-NAFL or OBE-NASH.
      (A) Transcript levels of TNFA and IL6 in VAT and (B) SAT, (C) protein expression of ATF4 and CHOP in VAT and (D) in SAT, (E) transcript levels of CD68, CD163 and MCP1 in VAT and (F) in SAT. Means ± SEM, levels of significance of each significant and borderline significant difference are marked in the figure (one-way ANOVA corrected for multiple comparisons with Tukey-Kramer multiple comparisons test). ATF4, activating transcription factor 4; AU, arbitrary units; CHOP, C/EBP homologous protein; ER, endoplasmic reticulum; IL6, interleukin-6; MCP1, monocyte chemoattractant protein; OBE-CON, obese humans without non-alcoholic fatty liver; OBE-NAFL, obese humans with NAFL; OBE-NASH, obese humans with NASH; SAT, subcutaneous adipose tissue; TNFA, tumor necrosis factor A; VAT, visceral adipose tissue.
      In VAT, expression of activating transcription factor 4 (ATF4) and the transcription factor C/EBP homologous protein (CHOP) were similar between the groups (Fig. 4C). On the other hand, in SAT, ATF4 was higher in OBE-NASH than in OBE-CON and CHOP tended to be higher in OBE-NASH than in OBE-NAFL (p = 0.054) (Fig. 4D). Indeed, ATF4 regulates the expression of CHOP, which upregulates pro-apoptotic protein expression.
      • Lebeaupin C.
      • Vallee D.
      • Hazari Y.
      • Hetz C.
      • Chevet E.
      • Bailly-Maitre B.
      Endoplasmic reticulum stress signalling and the pathogenesis of non-alcoholic fatty liver disease.
      Other biomarkers of ER stress, including protein kinase-like ER kinase, inositol-requiring enzyme 1α, eukaryotic translation initiation factor 2α and binding immunoglobulin protein 1 were similar across all groups and in both compartments (data not shown).
      Finally, mRNA expression of the macrophage marker, CD68, was higher in OBE-NASH than in OBE-CON, but only in VAT, not in SAT (Fig. 4E-F). Other macrophage markers including CD163 and monocyte chemoattractant protein 1 were comparable across all groups in SAT and VAT (Fig. 4E-F). Also, the immunohistochemical analysis of macrophage-specific antigens revealed similar numbers of CD68+, CD163+ and CD11c+-expressing cells in both SAT and VAT in a subgroup (n = 4 per group) (Fig. S9A-F).

      Mitochondrial respiration is higher in VAT of obese persons without NAFLD than in SAT of obese people with or without NAFLD

      Comparison of the features of SAT and VAT within and between groups showed higher [ETF]P in the VAT of OBE-CON than in SAT of all three studied groups (Fig. S10A), with similar differences for [ETF+CI]P and for maximal ADP-stimulated mitochondrial respiration ([ETF+CI+II]P) (Fig. S10A). Of note, VAT of OBE-NAFL and OBE-NASH displayed similar mitochondrial respiration as SAT of all groups (Fig. S10A).
      Accordingly, protein expression of OXPHOS CIV and CV was higher in VAT than in SAT of OBE-CON and OBE-NAFL (Fig. S10B). Protein expression of OXPHOS CIII-CV was also higher in VAT of OBE-CON compared to SAT of both OBE-NAFL and OBE-NASH.
      Of note, OBE-CON and OBE-NAFL had lower pParkin/Parkin ratio in VAT than in SAT (Fig. S10C). OBE-CON also featured reduced mRNA expression of NRF1, but not TFAM, in VAT compared to SAT (Fig. S10D).
      Finally, the mRNA expression of TNFA, was higher in SAT than in VAT of OBE-CON (Fig. S10E), despite the similar IL6 (p = 0.885, data not shown). Similarly, TBARS were increased in SAT of OBE-CON and OBE-NAFL compared to VAT of all three groups (Fig. S10E). On the other hand, VAT showed increases in ATF4 when compared to SAT of OBE-CON (Fig. S10F), and in CHOP when compared to SAT of all respective groups (Fig. S10F).

      Obese individuals with NAFLD show no differences in systemic oxidative stress, or inflammation when compared to obese humans without NAFLD

      Plasma TBARS, reflecting systemic oxidative stress, were not different between the groups (Fig. 5A), as were plasma TNFA and plasma IL6 levels (Fig. 5B-C). Only plasma fibroblast growth factor 21 (FGF21) was higher in OBE-NAFL than in OBE-CON (Fig. 5D). Finally, high-molecular weight adiponectin and leptin were similar across all groups (Fig. 5E-F).
      Figure thumbnail gr5
      Fig. 5Systemic oxidative stress and inflammation in OBE-CON, OBE-NAFL or OBE-NASH.
      Plasma concentrations of (A) TBARS, (B) TNFA, (C) IL6, (D) FGF21, (E) HMW-adiponectin and (F) leptin. Means ± SEM, levels of significance of each significant and borderline significant difference are marked in the figure (one-way ANOVA corrected for multiple comparisons with Tukey-Kramer multiple comparisons test). FGF21, fibroblast growth factor 21; HMW, high-molecular weight; IL6, interleukin-6; OBE-CON, obese humans without non-alcoholic fatty liver; OBE-NAFL, obese humans with NAFL; OBE-NASH, obese humans with NASH; TBARS, thiobarbituric acid reactive substances; TNFA, tumor necrosis factor A.

      Mitochondrial respiration in VAT associates positively with AT IS, but negatively with inflammation of VAT

      In VAT, maximal ADP-stimulated mitochondrial respiration ([ETF+CI+II]P) associated positively with AT IS (β = 0.985, p = 0.041) and tended to associate negatively with the degree of hepatic steatosis (β = −0.004, p = 0.099) across all groups combined (Table S3). The sum of OXPHOS CI-CV in VAT associated negatively with the degree of hepatic steatosis upon adjustment for T2D (β = −0.006, p = 0.048) (Table S3). Maximal ADP-stimulated respiration further associated negatively with TNFA expression in VAT (β = −0.085, p = 0.040) and with plasma FGF21 (β = −0.196, p = 0.041) (Table S3). Plasma FGF21 also associated positively with the degree of liver steatosis across all groups, only upon adjustment for T2D (β=0.009, p = 0.022) (Table S3).
      Only in OBE-NASH, maximal ADP-stimulated respiration of SAT associated with the degree of hepatic steatosis (β = −0.010, p = 0.002, p for the interaction = 0.005) (Table S3).

      Discussion

      This study showed that obese humans with NAFL and NASH exhibit marked reductions of mitochondrial respiration in VAT when compared to similarly obese humans without NAFLD. Mitochondrial respiration in VAT associated positively with insulin-stimulated IS of whole-body AT, but negatively with local inflammation. Nevertheless, compared to those without NAFL, SAT of obese humans with NAFLD showed no further alterations of mitochondrial function, but only differences in biomarkers of mitophagy and ER stress.
      The present data cannot confirm the previously reported uniform impairment of mitochondrial respiration in both fat depots of obese humans. For instance, in both VAT and SAT of obese humans, proteomic analyses revealed decreased abundance of enzymes involved in fatty acid oxidation.
      • Carruthers N.J.
      • Strieder-Barboza C.
      • Caruso J.A.
      • Flesher C.G.
      • Baker N.A.
      • Kerk S.A.
      • et al.
      The human type 2 diabetes-specific visceral adipose tissue proteome and transcriptome in obesity.
      In contrast, the present findings rather support the hypothesis of a specific downregulation of energy metabolism in VAT, at least in persons with NAFLD and when compared to a carefully matched control group with comparable whole-body and abdominal obesity. Importantly, comparison of both fat depots revealed that only VAT of obese humans without NAFLD exhibits elevated mitochondrial respiration compared to SAT of all studied groups. Interestingly, mitochondrial respiration in VAT of obese people with NAFLD was comparable to that measured in SAT of obese non-steatotic people. This is in line with studies showing that VAT is generally characterized by higher oxidative metabolism in obese humans
      • Hruska P.
      • Kucera J.
      • Pekar M.
      • Holeczy P.
      • Mazur M.
      • Buzga M.
      • et al.
      Proteomic signatures of human visceral and subcutaneous adipocytes.
      and that only mitochondrial respiration in SAT, but not VAT, associates negatively with BMI in overweight people.
      • Wessels B.
      • Honecker J.
      • Schottl T.
      • Stecher L.
      • Klingenspor M.
      • Hauner H.
      • et al.
      Adipose mitochondrial respiratory capacity in obesity is impaired independently of glycemic status of tissue donors.
      Our findings differ from previous evidence showing lower mitochondrial respiration in VAT of obese humans, whose liver histology was, however, not reported.
      • Kraunsoe R.
      • Boushel R.
      • Hansen C.N.
      • Schjerling P.
      • Qvortrup K.
      • Stockel M.
      • et al.
      Mitochondrial respiration in subcutaneous and visceral adipose tissue from patients with morbid obesity.
      Of note, the present study cannot assess whether mitochondrial respiration of both AT compartments is different between lean non-steatotic and obese humans without NAFLD, as reported for human livers.
      • Koliaki C.
      • Szendroedi J.
      • Kaul K.
      • Jelenik T.
      • Nowotny P.
      • Jankowiak F.
      • et al.
      Adaptation of hepatic mitochondrial function in humans with non-alcoholic fatty liver is lost in steatohepatitis.
      But other studies have already suggested a reduction in O2 consumption rates in both SAT and VAT when comparing overweight/obese humans with non-obese individuals.
      • Yin X.
      • Lanza I.R.
      • Swain J.M.
      • Sarr M.G.
      • Nair K.S.
      • Jensen M.D.
      Adipocyte mitochondrial function is reduced in human obesity independent of fat cell size.
      The markedly lower mitochondrial respiration in VAT of both OBE-NAFL and OBE-NASH could stem from impaired respiratory control. However, the observed comparable RCR and LCR across all groups and both fat depots suggests an intact coupling in both compartments. These findings are consistent with previous studies showing that adaptation of mitochondrial respiration coexists with unchanged RCR in SAT of murine obesity.
      • Schottl T.
      • Kappler L.
      • Fromme T.
      • Klingenspor M.
      Limited OXPHOS capacity in white adipocytes is a hallmark of obesity in laboratory mice irrespective of the glucose tolerance status.
      The rather high values of [ETF+CI+II]L compared to [ETF+CI+II]P and the relatively low RCR are in line with findings in the deep compartment of SAT,
      • Bodis K.
      • Jelenik T.
      • Lundbom J.
      • Markgraf D.F.
      • Strom A.
      • Zaharia O.P.
      • et al.
      Expansion and impaired mitochondrial efficiency of deep subcutaneous adipose tissue in recent-onset type 2 diabetes.
      and slightly higher when compared to abdominal SAT.
      • Mendham A.E.
      • Larsen S.
      • George C.
      • Adams K.
      • Hauksson J.
      • Olsson T.
      • et al.
      Exercise training results in depot-specific adaptations to adipose tissue mitochondrial function.
      Use of different methods for tissue permeabilization, such as digitonin in the chamber
      • Bodis K.
      • Jelenik T.
      • Lundbom J.
      • Markgraf D.F.
      • Strom A.
      • Zaharia O.P.
      • et al.
      Expansion and impaired mitochondrial efficiency of deep subcutaneous adipose tissue in recent-onset type 2 diabetes.
      (like in the present study) or saponin prior to addition of AT in the chamber,
      • Mendham A.E.
      • Larsen S.
      • George C.
      • Adams K.
      • Hauksson J.
      • Olsson T.
      • et al.
      Exercise training results in depot-specific adaptations to adipose tissue mitochondrial function.
      may explain the differences between studies.
      Lower mitochondrial mass may represent another possible reason for the reduced mitochondrial respiration in OBE-NAFL and NASH. In the absence of a single validated gold-standard, the present study employed 3 independent parameters, CSA, mtDNA and protein levels of OXPHOS CI-CV, as a proxy for mitochondrial content, as previously suggested.
      • Kraunsoe R.
      • Boushel R.
      • Hansen C.N.
      • Schjerling P.
      • Qvortrup K.
      • Stockel M.
      • et al.
      Mitochondrial respiration in subcutaneous and visceral adipose tissue from patients with morbid obesity.
      ,
      • Yin X.
      • Lanza I.R.
      • Swain J.M.
      • Sarr M.G.
      • Nair K.S.
      • Jensen M.D.
      Adipocyte mitochondrial function is reduced in human obesity independent of fat cell size.
      ,
      • Jokinen R.
      • Rinnankoski-Tuikka R.
      • Kaye S.
      • Saarinen L.
      • Heinonen S.
      • Myohanen M.
      • et al.
      Adipose tissue mitochondrial capacity associates with long-term weight loss success.
      ,
      • Heinonen S.
      • Buzkova J.
      • Muniandy M.
      • Kaksonen R.
      • Ollikainen M.
      • Ismail K.
      • et al.
      Impaired mitochondrial biogenesis in adipose tissue in acquired obesity.
      While reduced CSA was reported in SAT and omental AT of obese humans,
      • Yin X.
      • Lanza I.R.
      • Swain J.M.
      • Sarr M.G.
      • Nair K.S.
      • Jensen M.D.
      Adipocyte mitochondrial function is reduced in human obesity independent of fat cell size.
      the present study revealed similar CSA between groups in both compartments. No differences were detectable for mtDNA, in contrast to previous studies reporting lower mtDNA in VAT than in SAT of less obese humans.
      • Kraunsoe R.
      • Boushel R.
      • Hansen C.N.
      • Schjerling P.
      • Qvortrup K.
      • Stockel M.
      • et al.
      Mitochondrial respiration in subcutaneous and visceral adipose tissue from patients with morbid obesity.
      Despite the slightly decreased expression of OXPHOS CIV in VAT of OBE-NASH than in OBE-CON, the lack of changes in RCR, LCR and other biomarkers of mitochondrial mass suggests an adaptive response to the altered metabolic conditions in NAFL and NASH. As the respirometry analysis in this study allows us to completely assess the OXPHOS system,
      • Koliaki C.
      • Roden M.
      Alterations of mitochondrial function and insulin sensitivity in human obesity and diabetes mellitus.
      the results point to a true lower intrinsic mitochondrial functionality within VAT of humans with NAFLD. Likewise, comparison of both fat compartments revealed lower CIV-V expression in SAT of all groups than in VAT of OBE-CON, thus supporting differential adaptation of VAT in NAFLD.
      Mitochondrial dynamics cannot account for the observed differences, as biomarkers of these pathways were similar in VAT between groups. Only in SAT, lower expression of pParkin in OBE-NASH may suggest altered mitophagy. Of note, VAT of OBE-CON and OBE-NAFL featured lower pParkin and pParkin/Parkin ratios, but increased expression of ER stress markers than SAT of the respective groups. The downregulated mitophagy in these groups could contribute to the gradual increase in ER stress markers, due to the previously described negative feedback mechanisms between autophagy and ER stress.
      • Deegan S.
      • Saveljeva S.
      • Gorman A.M.
      • Samali A.
      Stress-induced self-cannibalism: on the regulation of autophagy by endoplasmic reticulum stress.
      Another novel finding of this study is the association between maximal ADP-stimulated mitochondrial respiration ([ETF+CI+II]P) in VAT with AT IS, which remains a matter of debate. While some studies found a positive correlation between mitochondrial oxidative capacity in SAT and AT glucose metabolism,
      • Jokinen R.
      • Rinnankoski-Tuikka R.
      • Kaye S.
      • Saarinen L.
      • Heinonen S.
      • Myohanen M.
      • et al.
      Adipose tissue mitochondrial capacity associates with long-term weight loss success.
      others dissociated AT respiration from tissue-specific IR.
      • Politis-Barber V.
      • Brunetta H.S.
      • Paglialunga S.
      • Petrick H.L.
      • Holloway G.P.
      Long-term, high-fat feeding exacerbates short-term increases in adipose mitochondrial reactive oxygen species, without impairing mitochondrial respiration.
      Differences among these results and our study may rely on the lack of data from human VAT.
      • Politis-Barber V.
      • Brunetta H.S.
      • Paglialunga S.
      • Petrick H.L.
      • Holloway G.P.
      Long-term, high-fat feeding exacerbates short-term increases in adipose mitochondrial reactive oxygen species, without impairing mitochondrial respiration.
      Although the current study detected elevated AT IR only in OBE-NASH, data from another cohort of BARIA_DDZ indicate that obese humans with NAFL also have higher AT IR.
      • Gancheva S.
      • Ouni M.
      • Jelenik T.
      • Koliaki C.
      • Szendroedi J.
      • Toledo F.G.S.
      • et al.
      Dynamic changes of muscle insulin sensitivity after metabolic surgery.
      These findings support a role for AT IR in the previously reported association between increased AT lipolysis and NAFLD in humans.
      • Perry R.J.
      • Samuel V.T.
      • Petersen K.F.
      • Shulman G.I.
      The role of hepatic lipids in hepatic insulin resistance and type 2 diabetes.
      Recent evidence indicates that SAT shows adequate expandability in obese people with and without NAFLD
      • Beals J.W.
      • Smith G.I.
      • Shankaran M.
      • Fuchs A.
      • Schweitzer G.G.
      • Yoshino J.
      • et al.
      Increased adipose tissue fibrogenesis, not impaired expandability, is associated with nonalcoholic fatty liver disease.
      and that SAT inflammation and increased fatty acid release in relation to fat-free mass are unrelated to IR.
      • Koh H.E.
      • van Vliet S.
      • Pietka T.A.
      • Meyer G.A.
      • Razani B.
      • Laforest R.
      • et al.
      Subcutaneous adipose tissue metabolic function and insulin sensitivity in people with obesity.
      Given the critical role of AT mitochondria in fatty acid oxidation
      • Bournat J.C.
      • Brown C.W.
      Mitochondrial dysfunction in obesity.
      and FFA release into the circulation, VAT can contribute up to 50% of the FFA delivered to the liver,
      • Klein S.
      The case of visceral fat: argument for the defense.
      as well as contributing to IR by interfering with insulin-stimulated glucose transport.
      • Roden M.
      • Shulman G.I.
      The integrative biology of type 2 diabetes.
      Of note, [ETF]P, which was reduced only in VAT of obese individuals with NAFLD, reflects oxygen consumption linked to fatty acid oxidation. Taken together, these data support that downregulation of mitochondrial respiration in VAT contributes to impaired insulin-mediated suppression of AT lipolysis, leading to increased FFA release to the portal vein and NAFLD progression in obese people.
      Despite no changes in circulating pro-inflammatory biomarkers, OBE-NAFL showed higher TNFA expression in VAT, which correlated negatively with maximal ADP-stimulated mitochondrial respiration ([ETF+CI+II]P) across all groups. This extends and supports previous findings on the TNFA-mediated downregulation of OXPHOS genes in VAT of obese females.
      • Dahlman I.
      • Forsgren M.
      • Sjogren A.
      • Nordstrom E.A.
      • Kaaman M.
      • Naslund E.
      • et al.
      Downregulation of electron transport chain genes in visceral adipose tissue in type 2 diabetes independent of obesity and possibly involving tumor necrosis factor-alpha.
      We also found greater levels of TNFA in SAT of OBE-CON and of TBARS in SAT of OBE-CON and OBE-NAFL than in VAT of the corresponding groups. Nevertheless, VAT TNFA levels were not increased in OBE-NASH, in contrast to previous evidence showing higher human VAT TNFA mRNA expression in participants with NASH, but with higher BMI, compared to those with NAFL.
      • Jorge A.S.B.
      • Andrade J.M.O.
      • Paraiso A.F.
      • Jorge G.C.B.
      • Silveira C.M.
      • de Souza L.R.
      • et al.
      Body mass index and the visceral adipose tissue expression of IL-6 and TNF-alpha are associated with the morphological severity of non-alcoholic fatty liver disease in individuals with class III obesity.
      Our findings suggest a link between VAT-derived TNFA and early hepatic damage, in line with evidence pointing to hepatocytes and Kupffer cells as the primary sources of TNFA production in the context of NASH.
      • Lu S.
      • Wang Y.
      • Liu J.
      Tumor necrosis factor-alpha signaling in nonalcoholic steatohepatitis and targeted therapies.
      Both expression and systemic IL6 concentrations were similar between groups, in line with equally high IL6 levels in obese insulin-sensitive and insulin-resistant humans.
      • Koh H.E.
      • van Vliet S.
      • Pietka T.A.
      • Meyer G.A.
      • Razani B.
      • Laforest R.
      • et al.
      Subcutaneous adipose tissue metabolic function and insulin sensitivity in people with obesity.
      Only in VAT, mRNA expression of the macrophage marker CD68 was increased in OBE-NASH, supporting previous evidence for a linear correlation between CD68+ VAT macrophages and hepatic inflammation in obese humans.
      • Cimini F.A.
      • Barchetta I.
      • Ciccarelli G.
      • Leonetti F.
      • Silecchia G.
      • Chiappetta C.
      • et al.
      Adipose tissue remodelling in obese subjects is a determinant of presence and severity of fatty liver disease.
      Finally, FGF21 also correlated negatively with maximal ADP-stimulated mitochondrial respiration ([ETF+CI+II]P) in VAT across all groups. This finding lends further support to the suggested link between FGF21 and impaired muscle mitochondrial respiration in humans,
      • Suomalainen A.
      • Elo J.M.
      • Pietilainen K.H.
      • Hakonen A.H.
      • Sevastianova K.
      • Korpela M.
      • et al.
      FGF-21 as a biomarker for muscle-manifesting mitochondrial respiratory chain deficiencies: a diagnostic study.
      but does not allow for conclusions on causality given the complex function of FGF21 in the metabolism of various tissues.
      • Henriksson E.
      • Andersen B.
      FGF19 and FGF21 for the treatment of NASH-two sides of the same coin? Differential and overlapping effects of FGF19 and FGF21 from mice to human.
      The present study benefits from the assessment of various independent features of mitochondrial function in a cohort tightly matched for measures of obesity. Our participants underwent comprehensive phenotyping of energy metabolism and detailed biopsy assessment of liver histology. Although liver biopsies were assessed by a validated score,
      • Bedossa P.
      • Poitou C.
      • Veyrie N.
      • Bouillot J.L.
      • Basdevant A.
      • Paradis V.
      • et al.
      Histopathological algorithm and scoring system for evaluation of liver lesions in morbidly obese patients.
      our data need to be carefully interpreted in the context of the limitations of the evaluation tool. As the primary aim was to examine AT mitochondrial function in obese NAFLD, the lack of a lean control group does not represent a study limitation, but means our findings are not generalizable to individuals with lean NAFLD. Also, the cross-sectional design does not allow for conclusions on causality. This study found no trend towards differences in mitochondrial respiration and content between males and females. However, the higher proportion of female volunteers, also known from other bariatric surgery cohorts,
      • Young M.T.
      • Phelan M.J.
      • Nguyen N.T.
      A decade analysis of trends and outcomes of male vs female patients who underwent bariatric surgery.
      cannot exclude that such sex differences might occur, as reported in AT of high-fat fed and obese mice.
      • MacCannell A.D.V.
      • Futers T.S.
      • Whitehead A.
      • Moran A.
      • Witte K.K.
      • Roberts L.D.
      Sexual dimorphism in adipose tissue mitochondrial function and metabolic flexibility in obesity.
      Further, the experimental setup, i.e. assessing AT respiration during the presurgical period, comprising diet, fasting and weight loss, does not necessarily reflect physiological day-to-day conditions in people with or without NAFLD.
      • Colles S.L.
      • Dixon J.B.
      • Marks P.
      • Strauss B.J.
      • O'Brien P.E.
      Preoperative weight loss with a very-low-energy diet: quantitation of changes in liver and abdominal fat by serial imaging.
      Finally, expression of mitochondrial respiration normalized to mitochondrial content remains a matter of debate.
      • Koliaki C.
      • Roden M.
      Alterations of mitochondrial function and insulin sensitivity in human obesity and diabetes mellitus.
      Nevertheless, the present study revealed similar results when expressing respiration rates either per protein or additionally per mtDNA, confirming their robustness.
      In conclusion, maximal ADP-stimulated mitochondrial respiration ([ETF+CI+II]P) in VAT is reduced in obese individuals with NAFL or NASH compared to those without NAFLD and correlates positively with AT IS, but negatively with local VAT inflammation. These data indicate an important role of compartment-specific AT energy metabolism for IR and hepatic lipid accumulation in the context of obesity. It is tempting to speculate that improvement of mitochondrial respiration in VAT could serve as a future therapeutic target to prevent the manifestation and progression of NAFLD.

      Abbreviations

      AT, adipose tissue; ATF4, activating transcription factor 4; AU, arbitrary units; C, complex; CHOP, C/EBP homologous protein; CSA, citrate synthase activity; E, electron transport system capacity; ER, endoplasmic reticulum; ETF, electron transferring flavoprotein; FFA, free fatty acids; FGF21, fibroblast growth factor 21; GOX, glucose oxidation; hsCRP, high-sensitivity C-reactive protein; IL, interleukin; IR, insulin resistance; IS, insulin sensitivity; L, leak respiration; LCR, leak control ratio; LOX, lipid oxidation; mtDNA, mitochondrial DNA; NAFL, non-alcoholic fatty liver; NAFLD, non-alcoholic fatty liver disease; NASH, non-alcoholic steatohepatitis; NOXGD, non-oxidative glucose disposal; NRF, nuclear respiratory factor; OBE-CON, obese humans without non-alcoholic fatty liver; OBE-NAFL, obese humans with NAFL; OBE-NASH, obese humans with NASH; OXPHOS, oxidative phosphorylation; P, OXPHOS capacity; POX, protein oxidation; pParkin, phosphorylated Parkin; RCR, respiratory control ratio; SAT, subcutaneous adipose tissue; T2D, type 2 diabetes; TBARS, thiobarbituric acid reactive substances; TFAM, mitochondrial transcription factor A; TNFA, tumor necrosis factor A; VAT, visceral adipose tissue.

      Financial support

      This study was supported in part by the DDZ, which is funded by the Ministry of Culture and Science of the State of North Rhine-Westphalia ( MKW NRW ) and the German Federal Ministry of Health ( BMG ), by grants of the German Federal Ministry of Education and Research ( BMBF ) to the German Center for Diabetes Research (DZD e. V., DZD Grant 2016). Parts of the study were also supported by grants from the European Funds for Regional Development (EFRE-0400191), EUREKA Eurostars-2 (E! 113230 DIA-PEP) and by grants from the German Research Foundation ( DFG , SFB 1116/2 , GRK 2576 ), the German Diabetes Association ( DDG ) and the Schmutzler-Stiftung. The funding sources had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

      Authors’ contributions

      M.R., S.K. and K.P. designed the study and analyzed the results. M.R. and K.P. drafted the manuscript. K.P., S.K., L.M., D.P., C.H., J.P., B.D., M.H., C.G, N.S., A.Y., S.G., G.H., I.E., M.S. contributed to acquisition of the data. K.S. provided statistical analysis of the data. K.P., S.K., L.M., K.S., D.P., C.H., J.P., B.D., M.H., C.G., N.S., A.Y., S.G., G.H., I.E., M.S. and M.R. revised the manuscript and approved the final version of the manuscript.

      Data availability statement

      Data are available upon reasonable request. The data sets generated during and/or analyzed during the current study are not publicly available, since they are subject to national data protection laws and restrictions imposed by the ethics committee to ensure data privacy of the study participants. However, they can be applied for through an individual project agreement with the principal investigator of the German Diabetes Study.

      Acknowledgements

      The authors would like to thank all study participants as well as Kerstin Förster, Fariba Zivehe, Michelle Reina Do Fundo, Olga Dürrschmidt, David Höhn, Jan-Marc Leonhard and Ulrike Partke from the Institute for Clinical Diabetology at the DDZ for their excellent technical assistance. Cartoons in the graphical abstract were created with BioRender.com.

      Conflict of interest

      M.R. received lecture fees or served on advisory boards for Allergan, Astra-Zeneca, Bristol-Myers-Squibb, Eli Lilly, Fishawack Group, Gilead Sciences, Intercept Pharma, Inventiva, Novartis, Novo Nordisk, Pfizer, Prosciento, Sanofi US and Target RWE and performed investigator-initiated research with support from Boehringer-Ingelheim, Nutricia/Danone and Sanofi-Aventis to the German Diabetes Center (DDZ). C.H. received grant support from Sanofi-Aventis. No conflicts of interest, financial or otherwise, are declared by the other authors.
      Please refer to the accompanying ICMJE disclosure forms for further details.

      Supplementary data

      The following are the supplementary data to this article:

      References

        • Crewe C.
        • Scherer P.E.
        Intercellular and interorgan crosstalk through adipocyte extracellular vesicles.
        Rev Endocr Metab Disord. 2021;
        • Bodis K.
        • Roden M.
        Energy metabolism of white adipose tissue and insulin resistance in humans.
        Eur J Clin Invest. 2018; 48e13017
        • Roden M.
        • Shulman G.I.
        The integrative biology of type 2 diabetes.
        Nature. 2019; 576: 51-60
        • Donnelly K.L.
        • Smith C.I.
        • Schwarzenberg S.J.
        • Jessurun J.
        • Boldt M.D.
        • Parks E.J.
        Sources of fatty acids stored in liver and secreted via lipoproteins in patients with nonalcoholic fatty liver disease.
        J Clin Invest. 2005; 115: 1343-1351
        • Ter Horst K.W.
        • Vatner D.F.
        • Zhang D.
        • Cline G.W.
        • Ackermans M.T.
        • Nederveen A.J.
        • et al.
        Hepatic insulin resistance is not pathway selective in humans with nonalcoholic fatty liver disease.
        Diabetes Care. 2021; 44: 489-498
        • Kusminski C.M.
        • Scherer P.E.
        Mitochondrial dysfunction in white adipose tissue.
        Trends Endocrinol Metab. 2012; 23: 435-443
        • Christe M.
        • Hirzel E.
        • Lindinger A.
        • Kern B.
        • von Flue M.
        • Peterli R.
        • et al.
        Obesity affects mitochondrial citrate synthase in human omental adipose tissue.
        ISRN Obes. 2013; 2013826027
        • Kahn D.E.
        • Bergman B.C.
        Keeping it local in metabolic disease: adipose tissue paracrine signaling and insulin resistance.
        Diabetes. 2022; 71: 599-609
        • Alvarez-Llamas G.
        • Szalowska E.
        • de Vries M.P.
        • Weening D.
        • Landman K.
        • Hoek A.
        • et al.
        Characterization of the human visceral adipose tissue secretome.
        Mol Cel Proteomics. 2007; 6: 589-600
        • Roca-Rivada A.
        • Bravo S.B.
        • Perez-Sotelo D.
        • Alonso J.
        • Castro A.I.
        • Baamonde I.
        • et al.
        CILAIR-based secretome analysis of obese visceral and subcutaneous adipose tissues reveals distinctive ECM remodeling and inflammation mediators.
        Sci Rep. 2015; 512214
        • Nahmgoong H.
        • Jeon Y.G.
        • Park E.S.
        • Choi Y.H.
        • Han S.M.
        • Park J.
        • et al.
        Distinct properties of adipose stem cell subpopulations determine fat depot-specific characteristics.
        Cell Metab. 2022; 34: 458-472 e456
        • Pafili K.
        • Roden M.
        Nonalcoholic fatty liver disease (NAFLD) from pathogenesis to treatment concepts in humans.
        Mol Metab. 2020; 101122
        • Koliaki C.
        • Szendroedi J.
        • Kaul K.
        • Jelenik T.
        • Nowotny P.
        • Jankowiak F.
        • et al.
        Adaptation of hepatic mitochondrial function in humans with non-alcoholic fatty liver is lost in steatohepatitis.
        Cel Metab. 2015; 21: 739-746
        • Gastaldelli A.
        • Cusi K.
        • Pettiti M.
        • Hardies J.
        • Miyazaki Y.
        • Berria R.
        • et al.
        Relationship between hepatic/visceral fat and hepatic insulin resistance in nondiabetic and type 2 diabetic subjects.
        Gastroenterology. 2007; 133: 496-506
        • Bedossa P.
        • Poitou C.
        • Veyrie N.
        • Bouillot J.L.
        • Basdevant A.
        • Paradis V.
        • et al.
        Histopathological algorithm and scoring system for evaluation of liver lesions in morbidly obese patients.
        Hepatology. 2012; 56: 1751-1759
        • Phielix E.
        • Szendroedi J.
        • Roden M.
        Mitochondrial function and insulin resistance during aging: a mini-review.
        Gerontology. 2011; 57: 387-396
        • Honecker J.
        • Weidlich D.
        • Heisz S.
        • Lindgren C.M.
        • Karampinos D.C.
        • Claussnitzer M.
        • et al.
        A distribution-centered approach for analyzing human adipocyte size estimates and their association with obesity-related traits and mitochondrial function.
        Int J Obes (Lond). 2021; 45: 2108-2117
        • Politis-Barber V.
        • Brunetta H.S.
        • Paglialunga S.
        • Petrick H.L.
        • Holloway G.P.
        Long-term, high-fat feeding exacerbates short-term increases in adipose mitochondrial reactive oxygen species, without impairing mitochondrial respiration.
        Am J Physiol Endocrinol Metab. 2020; 319: E376-E387
        • Koliaki C.
        • Roden M.
        Alterations of mitochondrial function and insulin sensitivity in human obesity and diabetes mellitus.
        Annu Rev Nutr. 2016; 36: 337-367
        • Lebeaupin C.
        • Vallee D.
        • Hazari Y.
        • Hetz C.
        • Chevet E.
        • Bailly-Maitre B.
        Endoplasmic reticulum stress signalling and the pathogenesis of non-alcoholic fatty liver disease.
        J Hepatol. 2018; 69: 927-947
        • Carruthers N.J.
        • Strieder-Barboza C.
        • Caruso J.A.
        • Flesher C.G.
        • Baker N.A.
        • Kerk S.A.
        • et al.
        The human type 2 diabetes-specific visceral adipose tissue proteome and transcriptome in obesity.
        Sci Rep. 2021; 1117394
        • Hruska P.
        • Kucera J.
        • Pekar M.
        • Holeczy P.
        • Mazur M.
        • Buzga M.
        • et al.
        Proteomic signatures of human visceral and subcutaneous adipocytes.
        J Clin Endocrinol Metab. 2021;
        • Wessels B.
        • Honecker J.
        • Schottl T.
        • Stecher L.
        • Klingenspor M.
        • Hauner H.
        • et al.
        Adipose mitochondrial respiratory capacity in obesity is impaired independently of glycemic status of tissue donors.
        Obesity (Silver Spring). 2019; 27: 756-766
        • Kraunsoe R.
        • Boushel R.
        • Hansen C.N.
        • Schjerling P.
        • Qvortrup K.
        • Stockel M.
        • et al.
        Mitochondrial respiration in subcutaneous and visceral adipose tissue from patients with morbid obesity.
        J Physiol. 2010; 588: 2023-2032
        • Yin X.
        • Lanza I.R.
        • Swain J.M.
        • Sarr M.G.
        • Nair K.S.
        • Jensen M.D.
        Adipocyte mitochondrial function is reduced in human obesity independent of fat cell size.
        J Clin Endocrinol Metab. 2014; 99: E209-E216
        • Schottl T.
        • Kappler L.
        • Fromme T.
        • Klingenspor M.
        Limited OXPHOS capacity in white adipocytes is a hallmark of obesity in laboratory mice irrespective of the glucose tolerance status.
        Mol Metab. 2015; 4: 631-642
        • Bodis K.
        • Jelenik T.
        • Lundbom J.
        • Markgraf D.F.
        • Strom A.
        • Zaharia O.P.
        • et al.
        Expansion and impaired mitochondrial efficiency of deep subcutaneous adipose tissue in recent-onset type 2 diabetes.
        J Clin Endocrinol Metab. 2020; : 105
        • Mendham A.E.
        • Larsen S.
        • George C.
        • Adams K.
        • Hauksson J.
        • Olsson T.
        • et al.
        Exercise training results in depot-specific adaptations to adipose tissue mitochondrial function.
        Sci Rep. 2020; 10: 3785
        • Jokinen R.
        • Rinnankoski-Tuikka R.
        • Kaye S.
        • Saarinen L.
        • Heinonen S.
        • Myohanen M.
        • et al.
        Adipose tissue mitochondrial capacity associates with long-term weight loss success.
        Int J Obes (Lond). 2018; 42: 817-825
        • Heinonen S.
        • Buzkova J.
        • Muniandy M.
        • Kaksonen R.
        • Ollikainen M.
        • Ismail K.
        • et al.
        Impaired mitochondrial biogenesis in adipose tissue in acquired obesity.
        Diabetes. 2015; 64: 3135-3145
        • Deegan S.
        • Saveljeva S.
        • Gorman A.M.
        • Samali A.
        Stress-induced self-cannibalism: on the regulation of autophagy by endoplasmic reticulum stress.
        Cell Mol Life Sci. 2013; 70: 2425-2441
        • Gancheva S.
        • Ouni M.
        • Jelenik T.
        • Koliaki C.
        • Szendroedi J.
        • Toledo F.G.S.
        • et al.
        Dynamic changes of muscle insulin sensitivity after metabolic surgery.
        Nat Commun. 2019; 10: 4179
        • Perry R.J.
        • Samuel V.T.
        • Petersen K.F.
        • Shulman G.I.
        The role of hepatic lipids in hepatic insulin resistance and type 2 diabetes.
        Nature. 2014; 510: 84-91
        • Beals J.W.
        • Smith G.I.
        • Shankaran M.
        • Fuchs A.
        • Schweitzer G.G.
        • Yoshino J.
        • et al.
        Increased adipose tissue fibrogenesis, not impaired expandability, is associated with nonalcoholic fatty liver disease.
        Hepatology. 2021;
        • Koh H.E.
        • van Vliet S.
        • Pietka T.A.
        • Meyer G.A.
        • Razani B.
        • Laforest R.
        • et al.
        Subcutaneous adipose tissue metabolic function and insulin sensitivity in people with obesity.
        Diabetes. 2021; 70: 2225-2236
        • Bournat J.C.
        • Brown C.W.
        Mitochondrial dysfunction in obesity.
        Curr Opin Endocrinol Diabetes Obes. 2010; 17: 446-452
        • Klein S.
        The case of visceral fat: argument for the defense.
        J Clin Invest. 2004; 113: 1530-1532
        • Dahlman I.
        • Forsgren M.
        • Sjogren A.
        • Nordstrom E.A.
        • Kaaman M.
        • Naslund E.
        • et al.
        Downregulation of electron transport chain genes in visceral adipose tissue in type 2 diabetes independent of obesity and possibly involving tumor necrosis factor-alpha.
        Diabetes. 2006; 55: 1792-1799
        • Jorge A.S.B.
        • Andrade J.M.O.
        • Paraiso A.F.
        • Jorge G.C.B.
        • Silveira C.M.
        • de Souza L.R.
        • et al.
        Body mass index and the visceral adipose tissue expression of IL-6 and TNF-alpha are associated with the morphological severity of non-alcoholic fatty liver disease in individuals with class III obesity.
        Obes Res Clin Pract. 2018; 12: 1-8
        • Lu S.
        • Wang Y.
        • Liu J.
        Tumor necrosis factor-alpha signaling in nonalcoholic steatohepatitis and targeted therapies.
        J Genet Genomics. 2022; 49: 269-278
        • Cimini F.A.
        • Barchetta I.
        • Ciccarelli G.
        • Leonetti F.
        • Silecchia G.
        • Chiappetta C.
        • et al.
        Adipose tissue remodelling in obese subjects is a determinant of presence and severity of fatty liver disease.
        Diabetes Metab Res Rev. 2021; 37: e3358
        • Suomalainen A.
        • Elo J.M.
        • Pietilainen K.H.
        • Hakonen A.H.
        • Sevastianova K.
        • Korpela M.
        • et al.
        FGF-21 as a biomarker for muscle-manifesting mitochondrial respiratory chain deficiencies: a diagnostic study.
        Lancet Neurol. 2011; 10: 806-818
        • Henriksson E.
        • Andersen B.
        FGF19 and FGF21 for the treatment of NASH-two sides of the same coin? Differential and overlapping effects of FGF19 and FGF21 from mice to human.
        Front Endocrinol (Lausanne). 2020; 11601349
        • Young M.T.
        • Phelan M.J.
        • Nguyen N.T.
        A decade analysis of trends and outcomes of male vs female patients who underwent bariatric surgery.
        J Am Coll Surg. 2016; 222: 226-231
        • MacCannell A.D.V.
        • Futers T.S.
        • Whitehead A.
        • Moran A.
        • Witte K.K.
        • Roberts L.D.
        Sexual dimorphism in adipose tissue mitochondrial function and metabolic flexibility in obesity.
        Int J Obes (Lond). 2021; 45: 1773-1781
        • Colles S.L.
        • Dixon J.B.
        • Marks P.
        • Strauss B.J.
        • O'Brien P.E.
        Preoperative weight loss with a very-low-energy diet: quantitation of changes in liver and abdominal fat by serial imaging.
        Am J Clin Nutr. 2006; 84: 304-311