Expression of LONP1 Is High in Visceral Adipose Tissue in Obesity, and Is Associated with Glucose and Lipid Metabolism

Original Endocrinol Metab 2021;36:661-671 https://doi.org/10.3803/EnM.2021.1023 Article pISSN 2093-596X · eISSN 2093-5978

Ju Hee Lee1,2,*, Saet-Byel Jung2,*, Seong Eun Lee2, Ji Eun Kim2, Jung Tae Kim2, Yea Eun Kang1,2, Seul Gi Kang2, Hyon-Seung Yi1,2, Young Bok Ko3, Ki Hwan Lee3, Bon Jeong Ku1,2, Minho Shong1,2, Hyun Jin Kim1,2

1Department of Internal Medicine, 2Research Center for Endocrine and Metabolic Diseases, 3Department of Obstetrics and Gynecology, Chungnam National University College of Medicine, Daejeon, Korea

Background: The nature and role of the mitochondrial stress response in adipose tissue in relation to obesity are not yet known. To determine whether the mitochondrial unfolded protein response (UPRmt) in adipose tissue is associated with obesity in humans and rodents. Methods: Visceral adipose tissue (VAT) was obtained from 48 normoglycemic women who underwent surgery. Expression levels of mRNA and proteins were measured for mitochondrial chaperones, intrinsic proteases, and components of electron-transport chains. Furthermore, we systematically analyzed metabolic phenotypes with a large panel of isogenic BXD inbred mouse strains and Geno- type-Tissue Expression (GTEx) data. Results: In VAT, expression of mitochondrial chaperones and intrinsic proteases localized in inner and outer mitochondrial mem- branes was not associated with body mass index (BMI), except for the Lon protease homolog, mitochondrial, and the corresponding gene LONP1, which showed high-level expression in the VAT of overweight or obese individuals. Expression of LONP1 in VAT positively correlated with BMI. Analysis of the GTEx database revealed that elevation of LONP1 expression is associated with enhancement of genes involved in glucose and lipid metabolism in VAT. Mice with higher Lonp1 expression in adipose tissue had better systemic glucose metabolism than mice with lower Lonp1 expression. Conclusion: Expression of mitochondrial LONP1, which is involved in the mitochondrial quality control stress response, was ele- vated in the VAT of obese individuals. In a bioinformatics analysis, high LONP1 expression in VAT was associated with enhanced glucose and lipid metabolism.

Keywords: Intra-abdominal fat; Obesity; Metabolic syndrome

INTRODUCTION nutritional status. The remodeling process includes qualitative and quantitative changes in adipocytes and adipose tissue-resi- Adipose tissue undergoes dynamic remodeling in response to dent cells [1]. Chronic overnutrition leads to obesity, which is

Received: 4 March 2021, Revised: 12 April 2021, Accepted: 3 May 2021 Copyright © 2021 Korean Endocrine Society This is an Open Access article distributed under the terms of the Creative Com Corresponding author: Hyun Jin Kim mons Attribution Non-Commercial License (https://creativecommons.org/ Department of Internal Medicine, Chungnam National University College of licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribu Medicine, 282 Munhwa-ro, Jung-gu, Daejeon 35015, Korea tion, and reproduction in any medium, provided the original work is properly Tel: +82-42-280-7151, Fax: +82-42-280-8193, E-mail: [email protected] cited. *These authors contributed equally to this work.

www.e-enm.org 661 Lee JH, et al.

accompanied by stress and an adaptive response that is associat- ipants were included in this study. Blood chemistry of partici- ed with altered homeostasis of cellular organelles, such as endo- pants was assessed, and serum insulin concentrations were mea- plasmic reticulum and mitochondria, in adipocytes, liver, and sured as previously described [9]. skeletal muscle [2,3]. Mitochondrial homeostasis may play crit- ical roles in both normal and pathological adipocyte remodel- Biopsy ing, and a better understanding of the molecular links between Omental adipose tissue samples from the distal portion of the mitochondrial homeostasis and energy metabolism should help greater omentum were collected during surgery under general to identify potential targets for the treatment of obesity. anesthesia. Adipose tissue specimens were snap-frozen in liquid Mitochondrial homeostasis requires the induction of several nitrogen and stored at −70°C. pathways that are responsible for mitochondrial quality control, including the mitochondrial unfolded protein response (UPRmt) RNA isolation and quantitative real-time polymerase [4-6]. The UPRmt senses proteostatic disturbances specifically in chain reaction the mitochondria and resolves the stress by retrograde signaling The methods that were used for RNA extraction, cDNA synthe- to the nucleus, resulting in transcriptional activation of protec- sis, and real-time polymerase chain reaction (RT-PCR) analysis tive genes [7]. The damaged proteins are recognized by mito- have been described previously [10]. The primers for RT-PCR chondrial chaperones, such as 60 kDa heat shock protein, mito- are listed in Supplemental Table S1. Target mRNA expression chondrial (encoded by HSPD1) and DnaJ heat shock protein was normalized to 18S mRNA expression and determined by family (Hsp40) member A3 (encoded by DNAJA3), then un- the 2−ΔΔCt method. Mitochondrial DNA (mtDNA) content was folded. Those proteins thereby become substrates for proteases determined by quantitative RT-PCR, as previously described such as caseinolytic mitochondrial matrix peptidase proteolytic [11]. subunit (encoded by CLPP), and Lon protease homolog, mitochondrial (encoded by LONP1) [4,8], which degrade them. Protein analysis In this study, we investigated the molecular features of UPRmt Antibodies targeting cytochrome c oxidase subunit 1 (COX1), in human visceral adipose tissue (VAT) in relation to obesity. ubiquinol-cytochrome-c reductase complex core protein 2 We found that expression of the AAA+ family protease LONP1 (UQCRC2), DNAJA3, LONP1, and β-actin were purchased is associated with obesity. We performed a bioinformatics anal- from Abcam (Cambridge, UK); antibody to NADH dehydroge- ysis that determined the metabolic phenotypes that were associ- nase (ubiquinone) 1 beta subcomplex subunit 8, mitochondrial ated with high and low expression of LONP1 in human VAT. (NDUFB8) was purchased from Novex (Carlsbad, CA, USA); Furthermore, we identified an association between Lonp1 ex- and antibodies recognizing succinate dehydrogenase (ubiqui- pression and systemic glucose metabolism in murine white adi- none) iron-sulfur subunit, mitochondrial (SDHB), transcription pose tissue (WAT), by use of a genetic database. factor A, mitochondrial (TFAM), HSPD1, and CLPP were purchased from Santa Cruz Biotechnology (Dallas, TX, USA). An- METHODS ti-rabbit secondary antibodies were obtained from Santa Cruz Biotechnology, and anti-mouse secondary antibodies were ob- Participants tained from Cell Signaling Technology (Danvers, MA, USA). Patients who were scheduled for elective surgery were prospec- Western blotting images were scanned and quantified using an tively enrolled at the Department of Gynecology of Chungnam Odyssey imaging system and Image Studio DiGit software (LI- National University Hospital (CNUH), Daejeon, Republic of COR Biosciences, Lincoln, NE, USA). Lipid peroxidation was Korea. The Institutional Review Board of CNUH approved the analyzed by measuring malondialdehyde concentrations with a study (approval number: 2016-07-026), and written informed commercial thiobarbituric acid-reactive substances assay kit consent was obtained from all participants. The study was per- (Cayman Chemical, Ann Arbor, MI, USA), according to the formed in accordance with the principles of the Declaration of manufacturer’s specifications. Helsinki. Inclusion criteria were age >19 years, and elective surgery for benign uterine and ovarian disorders. Individuals Statistical analysis with cancer, prediabetes, diabetes mellitus, active liver disease, Statistical analyses were performed with SPSS software version and/or end-stage renal disease were excluded. Finally, 48 partic- 24.0 (IBM, Armonk, NY, USA). Clinical data are expressed as 662 www.e-enm.org Copyright © 2021 Korean Endocrine Society Mitochondrial LON Protease Expression in Obesity

mean±standard deviation. mRNA levels are expressed as group). Phenotypic data recorded for the mice were body mean±standard error of the mean. Continuous variables were weight in males aged 8, 16, and 28 weeks old (GN record ID: tested for normality with the Kolmogorov-Smirnov test. Clini- 17557-17562); blood glucose measured during the oral glucose- cal characteristics and mRNA levels of genes related to oxida- tolerance test (OGTT) at 17 weeks old (GN record ID: 17643- tive phosphorylation (OXPHOS) and UPRmt were compared 17660); insulin measured during OGTT at 17 weeks old (GN between groups of individuals with body mass index (BMI) record ID: 17665-17670); and lipid profiles measured in fasted <23 kg/m2 and those with BMI ≥23 kg/m2 by Student’s t test state at 29 weeks old (GN record ID:17801-17812) [14,15]. or the Mann-Whitney U test for data with or without a normal Metabolic phenotypes and Lonp1 mRNA levels in sWAT were distribution, respectively. The strengths of the relationships be- compared between Lonp1-low and Lonp1-high groups with tween variables of interest were analyzed by calculation of Student’s t test. A paired-sample comparison of Lonp1 expres- Spearman’s correlation coefficients. P<0.05 was considered to sion in the sWAT of mice fed chow versus high-fat diets was represent statistical significance. conducted in 35 mouse strains.

Gene Ontology and pathway-enrichment analysis with the RESULTS GTEx database All the available Genotype-Tissue Expression (GTEx) data for VAT mRNA levels of OXPHOS-complex and UPRmt genes VAT (n=541) were obtained from the UCSC database (http:// in relation to BMI xena.ucsc.edu/). The data used for our analyses were obtained Among 48 patients, 11 were obese (≥25 kg/m2), 11 were over- from dbGaP, accession number: phos000424.v8.02 on April 13, weight (23 to 24.9 kg/m2), and 26 were of normal or under- 2020. Of the 541 VAT samples in the GTEx database, those with weight (<22.9 kg/m2), according to the World Health Organiza- LONP1 expression in the highest (n=134) and lowest (n=134) tion Asia-Pacific Obesity Classification [16]. Clinical character- quartiles were used for Gene Ontology (GO) and Kyoto Ency- istics of the participants stratified by BMI (<23 kg/m2 vs. ≥23 clopedia of Genes and Genomes (KEGG) pathway analysis. kg/m2) are summarized in Table 1. BMI, waist circumference, Differentially expressed genes were identified by the establish- fasting blood glucose, fasting blood insulin, homeostatic model ment of two groups based on LONP1 expression with the R assessment of insulin resistance (HOMA-IR), and alanine ami- package DESeq2 [12]. The gene-set collection of KEGG was notransferase values were significantly higher in the high-BMI obtained from Enrichr (https://amp.pharm.mssm.edu/Enrichr/), than in the low-BMI group, whereas age and other variables did and the gene-set enrichment analysis was conducted with the R not differ significantly between the two groups. Among OX- package Platform for Integrative Analysis of Omics data (PIA- PHOS-complex genes (NADH:ubiquinone oxidoreductase sub- NO). P values were adjusted using Benjamini-Hochberg correc- unit A9 [NDUFA9], SDHB, UQCRC2, COX4, and ATP synthase tion for controlling false-discovery rate, and results were con- F1 subunit alpha [ATP5A1A]), UQCRC2 (from OXPHOS com- sidered statistically significant when adjusted P values were plex III) was only significantly lower in the high-BMI group <0.05. The full list of significantly enriched pathways is includ- than in the low-BMI group (Fig. 1A). The mitochondrial bio- ed in Supplemental Material S1. genesis-related genes, PPARGC1A (which encodes PPARG coactivator 1α), a key activator of mitochondrial transcription, Murine metabolic phenotype analysis with BXD TFAM expression and mtDNA content did not differ significant- recombinant inbred mouse strains ly between the groups (Fig. 1B, C). Although excess fat accu- To analyze murine metabolic phenotypes in relation to expres- mulation increases mitochondrial production of reactive oxygen sion of Lonp1, GeneNetwork (www.genenetwork.org) was used species to cause adipocyte mitochondrial dysfunction [2], cellu- for multi-omics data for BXD recombinant inbred (RI) strains lar concentrations of malondialdehyde, a naturally occurring [13]. Expression of Lonp1 in subcutaneous WAT (sWAT) of product of lipid peroxidation, did not differ significantly be- BXD mice that were fed either chow or high-fat diets (GN ac- tween the groups (Fig. 1D). In correlation analyses, expression cession numbers: GN779 and GN778, respectively) was used of these genes including UQCRC2, VAT mtDNA content, and for the analysis. Metabolic phenotypes were compared between cellular malondialdehyde concentrations did not correlate sig- mice in the upper (Lonp1-high) and lower (Lonp1-low) quar- nificantly with BMI (data not shown). tiles with respect to WAT Lonp1 expression (n=9–10 mice per We compared the expression of genes encoding mitochondri- Copyright © 2021 Korean Endocrine Society www.e-enm.org 663 Lee JH, et al.

Table 1. Baseline Characteristics of the Study Participants

Variable BMI <23.0 kg/m2 (n=26) BMI ≥23.0 kg/m2 (n=22) P value Age, yr 42±5 (31–55) 43±6 (33–55) 0.345 BMI, kg/m2 20.9±1.5 (17.4–23.0) 26.2±2.9 (23.5–33.4) <0.001 Waist circumference, cm 76.4±5.3 (64–84) 88.6±7.1 (79–111) <0.001 Fasting blood glucose, mg/dL 82.7±6.8 (72–97) 88.2±5.9 (75–99) 0.005 Fasting blood insulin, mIU/L 7.0±2.5 (3.3–14.6) 9.5±3.5 (3.5–16.7) 0.005 HOMA-IR 1.4±0.6 (0.6–3.1) 2.1±0.8 (0.8–3.7) 0.003 Triglycerides, mg/dL 91±60 120±76 0.159 Total cholesterol, mg/dL 176±24 186±36 0.288 LDL-C, mg/dL 101±22 112±28 0.123 HDL-C, mg/dL 61±11 55±14 0.080 AST, IU/L 18±8 19±5 0.410 ALT, IU/L 13±4 19±12 0.034 eGFR, mL/min 124±21 118±27 0.453

Values are expressed as mean±standard deviation (range). P values are calculated by Student’s t test or Mann-Whitney U test. BMI, body mass index; HOMA-IR, homeostatic model assessment of insulin resistance; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; AST, aspartate aminotransferase activity; ALT, alanine aminotransferase activity; eGFR, estimated glomerular filtration rate. al chaperones (HSPD1 and DNAJA3), mitochondrial matrix showed a difference in expression, which was significantly proteases (CLPP and LONP1), inner membrane protease higher in the VAT of participants with high BMI (Fig. 2A). (YME1 like 1 ATPase [YME1L1]), intramembrane protease Among the OXPHOS-complex proteins, participants with high (HtrA serine peptidase 2 [HTRA2]), outer membrane protease BMI had significantly lower expression of the OXPHOS com- (ubiquitin specific peptidase 30 [USP30]) in VAT in relation to plex I protein NDUFB8 than those with low BMI; no other dif- BMI. Among these UPRmt-related genes, LONP1 expression ferences in expression were observed (Fig. 2B). Expression of was around twice as high in the high-BMI group as in the low- the mitochondrial biogenesis-related protein TFAM was similar BMI group (Fig. 1E). LONP1 gene expression was positively in participants with low BMI and those with high BMI (Fig. correlated with BMI (ρ=0.308, P=0.050) (Fig. 1F). LONP1 ex- 2C). LONP1 was therefore the only differentially expressed pression was also positively correlated with expression of protein for which a consistent change in mRNA levels occurred HSPD1 and DNAJA3 (ρ=0.622, P<0.001 and ρ=0.409, P= in the UPRmt in the VAT of participants with high BMI, suggest- 0.012, respectively), but was not correlated with expression of ing that LONP1 is a marker for nutritional stress in the mito- OXPHOS genes, PPARGC1A or TFAM, VAT mtDNA content chondria of human VAT. or malondialdehyde concentration, participant age, serum glucose, lipid profile or HOMA-IR (data not shown). This finding GTEx gene-set-enrichment analysis of VAT, in relation to indicates that transcriptional activation of LONP1 is a signature LONP1 expression feature occurred in the course of human visceral obesity. To gain further insight into the relationship between energy metabolism and LONP1 expression in human VAT, gene-enrich- Association of BMI with expression of UPRmt proteins in ment analysis was performed with VAT expression data from VAT the GTEx database, with reference to levels of LONP1 expres- We measured the expression of mitochondrial chaperones, pro- sion. Of the 541 VAT samples in the GTEx database, samples teases, OXPHOS-complex proteins, and proteins involved in with LONP1 expression levels in the highest quartile (n=134) mitochondrial biogenesis in VAT of the five participants with and the lowest quartile (n=134) were used for GO and KEGG the lowest BMI (mean 18.6 kg/m2, range 17.4 to 19.3 kg/m2) pathway analyses (Fig. 3A). In these samples, 6,455 genes were and the five with the highest BMI (mean 29.6 kg/m2, range 27.5 upregulated in the high-LONP1 group compared with the low- to 31.8 kg/m2). Among the UPRmt proteins, only LONP1 LONP1 group (Fig. 3B). In a GO–biological-process analysis, 664 www.e-enm.org Copyright © 2021 Korean Endocrine Society Mitochondrial LON Protease Expression in Obesity

BMI <23 kg/m2 BMI <23 kg/m2 2 2 1.5 BMI ≥23 kg/m 2.0 BMI ≥23 kg/m 2,500 80

2,000 1.5 60 1.0

expression a expression 1,500 1.0 40

mRNA mRNA 1,000 0.5 0.5 20 Malondialdehyde (μm) Malondialdehyde

Relative mtDNA content Relative mtDNA 500 Relative Relative

0 0 0 0 NDUFA9 SDHB UQCRC2 COX4 ATP5A1A PPARGC1A TFAM BMI BMI BMI BMI A B <23 kg/m2 ≥23 kg/m2 C <23 kg/m2 ≥23 kg/m2 D

3 BMI <23 kg/m2 2.0 ρ=0.308 a BMI ≥23 kg/m2 P=0.050 1.5 2

expression 1.0 expression

mRNA 0.5

1 LONP1

Log 0 Relative 20 30 40 0 −0.5 HSPD1 DNAJA3 CLPP LONP1 YME1L1 HTRA2 USP30 E Body mass index (kg/m2) F

Fig. 1. Visceral adipose tissue expression of genes encoding proteins of the oxidative phosphorylation (OXPHOS) complex, mitochondrial chaperones, and proteases in relation to body mass index (BMI). Relative mRNA expression of genes of the OXPHOS complex (A) and mitochondrial biogenesis (B) in the visceral adipose tissue between a group with BMI <23 kg/m2 (n=26) and a group with BMI ≥23 kg/m2 (n=26). Gene expression relative to the mean level in the group with BMI <23 kg/m2 was determined for each sample by real-time polymerase chain reaction. (C) Relative mitochondrial DNA (mtDNA) content. (D) Malondianldehyde concentration. (E) Relative mRNA expression of genes encoding mitochondrial chaperones and proteases. (F) Correlation between Lon peptidase 1, mitochondrial (LONP1) mRNA expression in the visceral adipose tissue and BMI. Relative mRNA expression is presented as mean±standard error of the mean. P values were calculated by the Mann-Whitney U test. NDUFA9, NADH:ubiquinone oxidoreductase subunit A9; SDHB, NADH dehydrogenase (ubiquinone) 1 beta subcomplex subunit 8, mitochondrial; UQCRC2, ubiquinol-cytochrome-c reductase complex core protein 2; COX2, cytochrome c oxidase subunit II; ATP5A1A, ATP synthase F1 subunit alpha; PPARGC1A, PPARG coactivator 1 alpha; TFAM, transcription factor A, mitochondrial; HSPD1, heat shock protein family D (Hsp60) member 1; DNAJA3, DnaJ heat shock protein family (Hsp40) member A3; CLPP, caseinolytic mitochondrial matrix peptidase proteolytic subunit; YME1L1, YME1 like 1 ATPase; HTRA2, HtrA serine peptidase 2; USP30, ubiquitin specific peptidase 30. aP<0.01.

1,119 processes were upregulated in the high-LONP1 group VAT of the high-LONP1 group compared with the low-LONP1 compared with the low-LONP1 group (Fig. 3C). Among the group (Fig. 3C), including pathways related to the citrate cycle, significantly upregulated processes were cellular response to OXPHOS, insulin signaling pathway, regulation of lipolysis in nutrient level, lipid homeostasis, carbohydrate and glucose ho- adipocytes, and insulin resistance (Fig. 3G). These results indi- meostasis, and response to insulin (Fig. 3D). In addition, genes cate that LONP1 expression in human VAT is associated with involved in regulation of the lipid metabolic process, fatty acid metabolic regulation involving processes such as the TCA cy- metabolic process, and fatty acid oxidation were upregulated in cle, and glucose and lipid metabolism. the VAT of individuals with higher LONP1 expression (Fig. 3E), as were genes involved in the glucose metabolic process Murine metabolic phenotype according to the Lonp1 and response to glucose (Fig. 3F). expression level of WAT in BXD RI mouse strain In the KEGG analysis, 133 pathways were upregulated in the To attempt to validate the physiological role of LONP1, we ana- Copyright © 2021 Korean Endocrine Society www.e-enm.org 665 Lee JH, et al.

Low BMI High BMI 15 Low BMI HSPD1 High BMI b DNAJA3 10

CLPP 5 LONP1 Relative band density β-Actin 0 HSPD1 DNAJA3 CLPP LONP1 A

Low BMI High BMI 2.0 NDUFB8 Low BMI High BMI SDHB 1.5

UQCRC2 1.0 COX1 a 0.5 ATP5A Relative band density

β-Actin 0 NDUFB8 SDHB UQCRC2 COX1 ATP5A B

Low BMI High BMI 1.5 TFAM Low BMI High BMI β-Actin 1.0

0.5 Relative band density

C 0

Fig. 2. Expression of proteins of the oxidative phosphorylation (OXPHOS) complex, transcription factors, mitochondrial chaperones, and proteases in human visceral adipose tissue. (A) Expression of mitochondrial chaperones and proteases in visceral adipose tissue of the five participants with the lowest body mass index (BMI; low BMI: mean, 18.6 kg/m2; range, 17.4 to 19.3 kg/m2) and the five with the highest BMI (high BMI: mean, 29.6 kg/m2; range, 27.5 to 31.8 kg/m2). (B) Expression of proteins of the OXPHOS complex. (C) Expression of transcription factor A, mitochondrial (TFAM). Western blot band density relative to β-actin of each sample is calculated and presented as mean±standard error of the mean (n=5). P values were calculated by the Mann-Whitney U test. HSPD1, 60 kDa heat shock protein, mitochondrial; DNAJA3, DnaJ heat shock protein family (Hsp40) member A3; CLPP, caseinolytic mitochondrial matrix peptidase proteolytic subunit; LONP1, Lon protease homolog, mitochondrial; NDUFB8, NADH dehydrogenase (ubiquinone) 1 beta subcomplex subunit 8, mitochondrial (OXPHOS complex I); SDHB, succinate dehydrogenase (ubiquinone) iron-sulfur subunit, mitochondrial (OXPHOS complex II); UQCRC2, ubiquinol-cytochrome-c reductase complex core protein 2 (OXPHOS complex III); COX1, cytochrome c oxidase subunit 1 (OXPHOS complex IV); ATP5A1A, ATP synthase F1 subunit alpha (OXPHOS complex V). aP<0.01; bP<0.001. lyzed phenotypic data in the BXD RI mouse strain database ac- consistent with the high LONP1 expression that we observed in cording to Lonp1 expression in the sWAT (data for VAT were the VAT of individuals with high BMI. In comparisons between not available). Lonp1 expression was significantly higher in the chow-fed BXD strains in the highest quartile for Lonp1 expres- sWAT of mice that were fed a high-fat diet than in the corre- sion in sWAT (Lonp1-high) and those in the lowest quartile sponding mice that were fed a chow diet (Fig. 4A), which was (Lonp1-low), body weight was significantly lower in the Lonp1- 666 www.e-enm.org Copyright © 2021 Korean Endocrine Society Mitochondrial LON Protease Expression in Obesity

Differentially expressed genes GO-BP analysis KEGG analysis 200

n 100 Upregulated Upregulated

o DifferentialDlyi fefexrepressntialelyd e gxenepresss ed genes GO-BP anaGOlysis-BP analysis KEGG anaKElyGsisG analysis Upregulated i 90 200 Downregulated 1,119 Downregulated s 200 ) LONP1 high Downregulated s 80 GO-BP analysis KEGG analysis 2 Differentially expressed genes Unregulated GO-BP analysis KEGG analysis

n Unregulated 133 100 n 100 Differentially ex6,p4ress55 ed genes e Upregulated Upregulated Upregulated Upregulated Unregulated

200 g o o (upper quartile) UpregulatedUpregulated r 200 51 i 90 i 7090 o DownregulateDdownregulated DownregulateDdownregulated

p 1,119 1,119 s s l ) ) n 100 DownregulatDeodwnregulated n 100 LONP1 high LONP1 high Upregulated s s 80 x 80 Upregulated Upregulated

, Upregulated 2 2 o Unregulated Unregulated Unregulated Unregulated133 133Upregulated o 60 Upregulated i 90 6,455 6,455 e e i 90 e UnregulatedUnregulated g g (upper quartile)(upper quartile) Downregulated Downregulated 26 r r 1,119 51 s Downregulated 51 1,119 Downregulated s M ) 70 70 Downregulated ) Downregulated o

o LONP1 high p

s LONP1 high p l

l 1,9266 s 80 1 50 2 80 Unregulated

2 Unregulated 133 P Unregulated Unregulated 133 x

x 6,455 e ,

, 6,455

e Unregulated g 60 Unregulated

60 g (upper quartile) r P (upper quartile) 51 r e T 9,436 e 70 51 26 26

o 70 (

M 40 o M p 2,463 l p l N 1,9266

1 1,9266

1 50

x 50 x P , P 60 , 60 140 e P P e O 30

T 9,436 26 T 9,436 26 M

M LONP1 low ( 40 ( 40 2,463 2,463 L N 1,9266 N 1 50 1,9266 1 50 P P 20 140 140 P O 30 (lower quartile) P

O 30

T 9,436 T LONP1 low LONP1 low 9,436

( 40

( 40 2,463

L 2,463 L N 20 N 1020 (lower quartile)(lower quartile) 140 140

O 30 O 30 LONP1 low LONP1 low Total =3,5157 Total=3,633 Total=299 L L 1T0otal 541 samples A B C 1200 20 Tota(llo5w4e1rsqaumarptillee(sl)ower quartile) Total =3,5157Total =3,5157 Total=3,633Total=3,633 Total=299 Total=299 10Total 541 sa1m0ples Total 541 samTpolteasl 541 samples Total =3,515T7otal =3,5157 No. of Total=3,633Total=3,633 Total=299Total=299 No. of GO_biNool.oogf ic proceNsos. of genes No. oGf O_bioNloo. gofic process GO_biologic pGroOc_ebsisologic process GO_biologic process genes genes genes GO_biologic process genes genes No. of No. of No. of No. of Cellular response to nutrient levelsG(OG_Ob:0io0l3o1g6ic6G9p)Oro_cbeisoslogic process 53 Regulation of lipid metaboGlicOp_rboicoelossgGi(cOGp_Obr:oi0oc0le1os9gs2ic16p)rocess 100 Cellular responseCtoellnuulatrrierenst pleovnesles t(oGnOu:0tr0ie3n1t6le6v9e)ls (GO:0031669) ge5n3es gen5e3s Regulation of lRipeidgumlaetitoanboolficlipirdocmeestsab(GolOic:0p0ro1c9e2s1s6()GO:0019216) g1e0n0es ge1n0e0s ReRgeuglautlaiotinonoof fcceelllulullaarr mettaabboolilcicpprorcoecsess(sG(OG:0O0:3010332133) 23) 40 40 Lipid catabolLicipirdocceastsa(bGoOlic:0p01ro6c0e42s)s (GO:0016042) 21 RegulCateiollnulaorf creesllpuolanrsCmeeetloltuanlbauortlriceiespnprtolencvseeslssto((GGnOuOt:r:0i0e00n33t11le36v26e39l)s) (GO:0031669) 4503 53 Regulation oRfLleiippgiiuddlamctaieottnaabbooofllilicicpippdrromocceetssassb(o(GGlicOOp::0r0o00c11e69s02s41(26G))O:0019216) 12010 10201 Lipiiddhhoommeeosotsatsaissi(sG(OG:0O0:5050058580) 88) 49 49 Cellular lipidCcealtlaublaorliclipirdocceastsa(bGoOlic:0p04ro4c2e42s)s (GO:0044242) 24 Regulation of ceRllueglaurlLamitpieoitdnabhoof lmciceelploursolatcaresmsisse(t(GaGbOOo:l:0i0c00p53r51o03c8e28s3)s) (GO:0031323) 4490 40 CellularLliippiiddccaattaabboolLliiccipppirdrooccaeetssassb(o(GGlicOOp::0r0o00c41e46s20s442(2G))O:0016042) 2241 2214 CCaarrbboohhydratteehhoommeeosotsatsaissi(sG(OG:0O0:3030530305) 00) 62 62 Lipid biosynthLetipicidpbroiocseysnstLh(GieptOicd:0pb0rio0oc8se6ys1ns0t(h)GeOtic:0p00ro8c6e10s)s (GO:0008610) 72 72 CarbohydLraiptiedhhoommeeoossttaassLiisisp(i(dGGOhOo::0m000e35o35s50ta08s08i)s) (GO:0055088) 6429 49 Cellular lipidCcaeltlaublaorlilcippidroccaetsasbo(GlicOp:r0o0c4e4s2s4(2G)O:0044242) 7224 24 CCarabrboohhyyddrraatte cattaabboolilcicpprorcoecsess(sG(OG:0O0:1060015620) 52) 34 TriglycerideTmriegtalybcoelircidperomceestsa(bGoOlic:0p00ro6c6e41s)s (GO:0006641) 52 CarboChayrdbroahteydcraattaebhoColimcarpebrooshctyaedsrissa(t(eGGOhOo::0m000e13o63s05ta50s20i)s) (GO:0033500) 3642 62 34 TriglyLcieprididebiomseytnatbLhoiepltiiicdcpbpriroosccyeensstsshe((GGticOOp::0r0o00c00e68s66s41(10G))O:0008610) 5722 72 52 Glucose homeostasis (GO:0042593) 45 Fatty acid metabolic process (GO:0006631) 105 CarbohydGraltueccoCastearbhboolmihcyepdorrsoaGtcateelsusicscsao(t(GasGbeOOo:hl:0i0co00mp41r26oe50co9e5s3s2t)sa) s(GisO(:G00O1:60005422) 593) 4354 34 45 TrFigaltytycearcididemmTereitgtaalybbcooelliriccidpperromocceeFtssaassbt(ot(GyGlicOaOpc::0r0ido00c00me66s6e6s3t4a(11Gb))Ool:i0c0p0r6o6c4e1s) s (GO:00066311)5025 52 105 Cellular glucose homeostasis (GO:0001678) 33 Fatty acid catabolic process (GO:0009062) 65 Cellular GglluuccoosseeChheoolmlmueleaoorsGstgtalaulsusciisocsso((eGsGeOhOo:h:0m0o00em04o12es65toa79ss83ti)sa) s(GisO(:G00O4:20509031)678) 3435 45 33 FFaattttyy aacciidd mcaeFttaabtbtoyollaiicccpipdrromocceetsFsassab(ot(GtGlyicOOap::0cr0o0i0dc00e9c6s0a6s6t3a2(1Gb))Ool:i0c0p0r6o6c3e1s) s (GO:000906216)055 105 65 Response to insulin (GO:0032868) 70 Fatty acid oxidFaattiotyna(cGidOo:0x0id1a9ti3o9n5()GO:0019395) 50 Cellular gRluecsopsoeCnhseoellmutoleaorinsgstlauslcinioss((eGGOhOo::0m000e30o21s86ta67s88i)s) (GO:0001678) 7303 33 Fatty acid caFtaabttoyliaccpidroccaetsasbo(GlicOp:r0o0Fc0ae9stt0sy6(2aG)cOid:0o0x0i9d0a6t2io)n (GO:001939556)05 65 50 Response to insulin (GO:0032868) 70 Fatty acid beta-oxidation (GO:0006635) Cellular responseCtoelliunlsaurlirnessptiomnusleusto(GinOsu:0li0n3s2ti8m6u9l)us (GO:0032869) 110 Fatty acid beta-oxFidaatttiyonac(iGdOox:0id0a0t6io6n35(G) O:0019395) 50 5500 CReleluslpaornrseestpooinsRueelisntop(oGinOseu:0lt0ion3i2ns8stiu6ml8inu) l(uGsO(:G00O3:20806382) 869) 1710 70 Fatty acid oxidation (FGaOtt:y00a1c9i3d9b5e) ta-oxidation (GO:00066355)0 50 Insulin receptor signaling pathway (GO:0008286) 67 110 Fatty acid biosFyanttthyeaticcidpbroiocseysnsth(GetOic:0p0ro0c6e6s3s3()GO:0006633) 47 CeInllsuulalinr rreescpeopntosCreestliolgunilnaasrliurnelgisnpsoatnitmhsweulautoysi(n(GGsOuOl:i:0n000s03ti8m228u86l6u9)s) (GO:0032869) 67 110 Fatty acid beFta-totyxaidcaidtiobnet(aG-oOx:i0d0a0ti6o6n3(5G)O:0006635) 4570 50 Insulin receptor signaling pathway (GO:0008286) 110 67 Fatty acid biosynthetic process (GO:0006633) 47 Insulin receptor signaling pathway (GO:0008286) 0 1 2 3 4 Fatty acid biosynthetic process (GO:0006633) 0 1 2 3 4 47 Insulin receptor signaling pathway (GO:0008286)0 1 2 3 4 67 67 Fatty acid biosynthetic process (GO:0006633)0 1 2 3 4 47 -Log10 (P adj0) 1 2 3 4 0 -Log110 (P adj2) 3 4D -Log100(P adj)1 2 3 E4 -Log10 (0P adj) 1 2 3 4 0 1 2 3 4 0 1 2 3 4 -Log10 (P adj) -Log10 (P adj) -Log (P adj) -Log10 (P adj) No. of -Log (P adj) -Log10 (P adj) No. of 10 No. of 10 GO_biologiNcop. roof cess KEGG pathwKaEyGG pathway genes GO_biologic process genes genes No. of No. of genes No. of No. of KEGG pathNwo.aoyf KEGG pathway GO_biologicGpOro_cbGeiosOslo_gbiicoNploor.ogocficespsrocess Citrate cycle (TCAKEcyGclGe) pathway gen3e0s genes Glucose metabolic process (GO:0006006) genes gen6e3s genes Citrate cycle (TCA cycle) ge3n0es Glucose metabolic process (GO:0006006) 63 Oxidative phosphorylation 104 104 Positive regulation of glucose transport (GO:0010828) 25 OCxiidtraattievecpychloes(CpThiCtoraAryteclayctciyolcenl)e (TCACictryactle)cycle (TCA cycle) 30 30 30 Positive reguGlaluticoonsoefmgleutcaobsoGelillcutrcapconrososecpeeomsmrset t(e(aGGtbaOOob:l:i0oc00l0ip1c0r0o6p8c0r2e0o8s6c)s)e(sGsO(:G00O0:60000066)006) 2653 63 63 AMPK signaling pathway 120 Regulation of glucose import (GO:0046324) 31 OAxMidPaKtivseigpnhaolOisnpxgihdpoaartytivhlaewtipaohynosphOoxryidlaatitoivne phosphorylation112004 104 104 Positive reRgeuglaPutlPioaootsniositoinitvfivoegeflurgrecelugocsuoelasttetriioaoinmnsoppofoofgrrtgltu(l(cGuGocOOsoe::0s0t0e0ra41tn6r0as38p2n2o4s8r)p)t o(GrtO(:G00O1:00802180)828) 3215 25 25 PPAR signaling pathway 74 PAPMAPRKssiiggnnaalliinAnggMppPaaKtthhswwigaanyyaling pathway 74 120 PositivReeregguulalatiotionnoof fgglulucocoseseimimppoortrt(G(GOO:0:004466332246) ) 26 Insulin signalingApMatPhwKasyignaling pathway120 137 120 PositiveRreegguullaattiioonnooffgglluuccoosseeiimmppoorrtt((GGOO::00004466332264)) 2361 31 Insulin signalinPgPpAaRthswiganyaling pathway 137 74 RegulatioCnanoofngicluacl oglsyecoilmyspiso(rGt (OG:0O0:6010642613) 24) 31 PPAR AsidginpaolcinygtopkiantehwsiagynalingPpPatAhwRasyignaling pathway 74 69 Positive regulation of glucose import (GO:0046326) 2264 69 74 Positive regulatCioannoofngicluaclogslyeciomlypsoisrt((GGOO::00006416632216)) 2246 AdipocyItnoskuinlienssiiggnnaalliinInnggsupplaiantthhswwigaanyyaling pathway 137 137 PGoluscitoivse creatgaubloalticiopnroocfegsslutcoopsyeruimvapteo(rGt (OG:0O0:6010741683) 26) 24 26 Regulation of lipolysis in adInipsouclyintessignaling pathway 55 137 Canonical glycolysis (GO:0061621) 24 Regulation of lipoAlydsipisoicnytaodkiipnoecsyitgensaling pathway 55 69 Glucose catabolic Cpraoncoensiscatol gplycruovlyasteis((GGOO::00006611761281)) 2244 Adipocytokine signGalliyncgoplyastihsw/Galyuconeogenesis 69 Glycolytic process through gCluacnoosne-ic6a-plhgolsypchoalytesi(sG(OG:0O0:6010662106) 21) 24 Adipocytokine signaling pathway 67 69 Glucose catabolic process to pyruvate (GO:0061718) 24 24 GlycolyRseisg/uGlaluticoonnoefolgipeonlyessiiss in adipocytes 67 55 Glycolytic pGrloucceosssethcarotaubgohligclupcroocsees-6s-tpohpoysrpuhvaattee((GGOO::00006611672108)) 2244 Regulation of lipolysis in adipPoycryutevsate metabolism 55 39 Glucose catabolicRpersopcoenssse to gplyurcuovsaet(eG(OG:0O0:0090764197) 18) 45 PyruvaGtelymcoelytasRbisoe/lGigslmulcaotnioenogoefnleipsoislysis in adipocytes39 67 55 Glycolytic pRroecsepsosntsherotuogghlugcluocsoes(eG-6O-:p0h0o0s9p7h4a9te) (GO:0061620) 45 24 24 Glycolysis/GluconeogeneIsnisulin resistance 67 107 Glycolytic process through glucose-6-phosphate (GO:0061620) 24 Insulin resPisytarunvcaeteGmleytcaobloylsisims/Gluconeogenesis107 39 67 Glycolytic prCoecleluslasr trhersopuognsheRgtelouspgcolunscseoes-6eto-spgthaluorvcsaoptsihoean(tG(eGO(O:G0:0O0:4902701464919)6) 20) 4352 24 Non-aPlcyorhuovalictefamtteytalivbeorlidsmisease (NAFLD) 39 144 Cellular response tRoegslpuocnose tsotagrlvuactoiosne((GGOO::00004029174499)) 3425 Non-alcoholic fatty liver disease (NAFILnDsu) lin resistance 144 107 Gluconeogenesis (GO:0006094) 40 Insulin resistancTehermogenPeysrisuvate metabolism107 202 39 CellulGarlurecospnoenosgeentRoeesgislsupc(oGonsOes:e0s0tta0or6vg0al9tuio4cn)o(sGeO(:G00O4:20104099)749) 40 32 45 Non-alcoholic fatty liver disease (NAFLD) 144 Cellular response to glucose starvation (GO:0042149) 32 Non-alcoholic fatty liver diTsheearsmeo(gNeAnFeLsDis) Insulin resistance210424 107 Gluconeogenesis (GO:0006094) 0 1 2 3 4 40 Thermogenesis 0 1 2 3 4 202 CellulaGr rluecsopnoenosgentoesgislu(GcoOs:e00s0t6a0rv9a4)t0ion (G1O:0042149) 3 4 40 32 ThNeromno-agelcnoehsoisl0ic fatty1liver d2isease3(NAFL4D)202 144 -Log10 (P adj) -Log10 (P adj) Gluconeogenesis (GO-L:0o0g060(0P9a4d)j) 1 2 3 4 0 1 2 3 4 0 1 102 3 4 40 0 1-Log102(TPhaedrjm) 3ogene4sis 202 -Log10 (P adj) F -Log10 (P adj) G -Log10 (P ad0j) 1 2 3 4 -Log10 (P adj) 0 1 2 3 4

-Log10 (P adj) -Log10 (P adj) Fig. 3. Upregulation of physiological pathways in visceral adipose tissue with high expression of Lon protease homolog, mitochondrial (LONP1) in the Genotype-Tissue Expression Database. (A) LONP1 expression levels in 541 human visceral adipose tissue samples in the UCSC database. (B) Numbers of differentially expressed genes in a comparison of the samples in the highest quartile for LONP1 expression (n=134) and those in the lowest quartile (n=134). (C) Association of differential LONP1 expression with Gene Ontology (GO) biological processes (GO-BP) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. (D, E, F, G) Upregulated biological processes in the GO annotation or KEGG pathway analysis. TPM, transcripts per million.

high group at 16 weeks old, and blood glucose was significantly consistent with the results of the human VAT GTEx gene analy- lower in the Lonp1-high group during OGTT at 17 weeks old, sis. On the other hands, the lipid profiles including free fatty ac- although insulin levels during OGTT did not differ significantly ids, high-density lipoprotein, low-density lipoprotein, triglycer- (Fig. 4B-E). In comparisons between high-fat-diet-fed mice in ides and cholesterol, were not significantly different according the highest quartile for Lonp1 expression in sWAT (Lonp1-high- to Lonp1-expression level in the both chow and high fat diet HFD) and those in the lowest quartile (Lonp1-low-HFD), body BXD RI mouse strain (data not shown). weight and blood glucose did not differ between the groups (Fig. 4F-H), but the insulin level during OGTT was more than DISCUSSION twofold higher in the Lonp1-high-HFD group than in Lonp1- low-HFD mice (Fig. 4I). Therefore, Lonp1 expression in murine Mitochondria play an essential role in energy homeostasis and sWAT was related to systemic glucose metabolism, which was have evolved to respond to nutritional status [17]. Therefore, Copyright © 2021 Korean Endocrine Society www.e-enm.org 667 Lee JH, et al.

CD Lonp1-low a b Lonp1-high 10.8 10.6 50

10.6 b 10.4 40 10.4 10.2 30

level in sWAT 10.2 level in sWAT

10.0 Body weight (g) 20 10.0 Lonp1 Lonp1 9.8 9.8 10 CD HFD A Lonp1- Lonp1- B 8 16 28 C low high Age (wk)

HFD Lonp1-low Lonp1-low b Lonp1-high Lonp1-high 400 3 10.8 b b b b 300 10.6 b 2 200 10.4

1 level in sWAT

100 (μg/mL) Insulin 10.2 Lonp1 Blood glucouse (mg/mL) 0 0 10.0 0 15 30 45 60 90 120 150 180 0 15 30 Lonp1- Lonp1- Time (min) D Time (min) E low high F

Lonp1-low-HFD Lonp1-low-HFD Lonp1-low-HFD Lonp1-high-HFD Lonp1-high-HFD Lonp1-high-HFD 60 500 8

400 b 50 6 b 300 40 4 200 30 Insulin (μg/mL) Insulin Body weight (g) 2 100

20 Blood glucouse (mg/mL) 0 0 8 16 28 0 15 30 45 60 90 120 150 180 0 15 30 Age (wk) G Time (min) H Time (min) I

Fig. 4. Metabolic phenotypes in BXD mouse strains in relation to Lon protease homolog, mitochondrial (Lonp1) expression. (A) Lonp1 expression level in the subcutaneous white adipose tissue (sWAT) according to diet. (B) Comparison of lowest quartile (Lonp1-low) and highest quartile (Lonp1-high) of Lonp1 expression levels in the sWAT of BXD mouse strains with chow diet (CD). (C) Body weight change in BXD mice fed CD. Blood glucose (D) and insulin (E) were measured during the oral glucose-tolerance test (OGTT) in male mice at 17 weeks of age receiving CD. (F) Comparison of lowest quartile (Lonp1-low high-fat diet [HFD]) and highest quartile (Lonp1-high-HFD) of Lonp1 expression levels in the sWAT of BXD mouse strains with HFD. (G) Body weight change in BXD mice fed a HFD. Blood glucose (H) and insulin (I) measured during OGTT in male mice at 17 weeks of age receiving HFD. P values were calculated by t test. aP<0.001; bP<0.05. mitochondrial homeostatic responses are expected in conditions with upregulation of lipid and glucose metabolism in human such as obesity, chronic overnutrition, and starvation. In this VAT. Similarly, mice with high levels of Lonp1 gene expression study, we found that, among human genes that are involved in in sWAT had better glucose-tolerance profiles than those with UPRmt (which is a central part of the mitohormetic response) in low Lonp1 expression. VAT, LONP1 expression is associated with obesity. Further- Mitochondrial quality control is a process that operates in re- more, bioinformatics analysis of the GTEx database demon- sponse to mitochondrial stress [4]. Different quality control strated that high levels of expression of LONP1 were associated mechanisms are activated depending on the type and duration 668 www.e-enm.org Copyright © 2021 Korean Endocrine Society Mitochondrial LON Protease Expression in Obesity

of the stress, thereby maintaining mitochondrial function. Char- of type 2 diabetes mellitus, and the methods of gene-expression acterization of the quality control network is important to fur- analysis used [24-26]. In one study, global downregulation of ther our understanding of complex biological processes such as mitochondrial oxidative pathways, mtDNA levels, and expres- obesity and related metabolic diseases. Previous results have sion of proteins of the OXPHOS machinery was identified in demonstrated the effect of UPRmt signaling components on met- the subcutaneous adipose tissue of individuals with high BMI abolic phenotype [18]. OXPHOS-complex inhibition in skeletal relative to their lower BMI monozygotic twins [24]. However, muscle and adipose tissue activates UPRmt in those tissues, and in another study, with Affymetrix gene profiling, none of the is associated with protection against obesity and insulin resis- genes of the electron-transport chain were downregulated in the tance in a mouse model [19,20]. Here, we investigated the ex- visceral fat of healthy obese women compared with healthy pression of both mitochondrial chaperones and proteases in hu- non-obese women [25]. In a mitochondrial functional analysis, man VAT, which is the primary responsive organ in obesity. Our oxygen-consumption rates and citrate-synthase activity were results showed that LONP1 was significantly upregulated in the significantly lower in adipocytes from obese individuals than VAT of individuals with high BMI, and that it was the only one from those who were not obese, although adipocyte mitochon- of the UPRmt genes that we studied that was significantly asso- drial content did not significantly differ [27]. Our results con- ciated with BMI. tribute to the existing body of evidence, identifying elevation of The LONP1, which is a member of the highly conserved LONP1 as a response to metabolic stress in obesity. AAA+ superfamily, is involved in protein quality control in the We found that LONP1 expression in VAT did not correlate mitochondrial matrix, which it achieves by degrading misfolded with expression of OXPHOS-complex genes or with mtDNA or oxidized polypeptides, and it is one of the principal compo- count. However, LONP1 expression was closely associated with nents of the UPRmt [21,22]. An association between LONP1 and systemic energy metabolism in the GTEx analysis. We also metabolic disorders has been shown by the presence of low lev- found that Lonp1 expression was higher in the sWAT of BXD els of LONP1 expression in the livers of diabetic db/db mice RI mice on high-fat diets than in those on chow diets, which [23], and by the fact that a reduction in LONP1 expression was consistent with our findings in human VAT. Furthermore, causes impairment of insulin signaling and elevation of expres- mouse strains with high Lonp1 expression in subcutaneous tis- sion of gluconeogenic enzymes in human liver cells [23]. In the sue had better systemic glucose metabolic profiles than those present study, we showed that LONP1 expression was signifi- with low Lonp1 expression. Therefore, elevation of LONP1 ex- cantly higher in the VAT of individuals with high BMI than in pression in the human VAT in obesity could be a homeostatic those with low BMI. Furthermore, high LONP1 expression was mechanism for preservation of mitochondrial function and sys- related to upregulation of lipid and glucose metabolism in the temic metabolism as well as the glucose metabolism. GTEx database. These results suggested that the mitochondrial Some limitation should be noted in this study. First, we did matrix protease LONP1 of the VAT, in response to a high-nutri- not analyze the correlation of Lonp1 expression in VAT with the ent state in obesity, modulates systemic glucose metabolism. BMI as well as gender and age in the GTEx dataset, because We also showed that glucose tolerance was associated with phenotypic data are not freely available. Second, we analyzed Lonp1 expression in the sWAT in BXD RI mice. However, we murine metabolic phenotypes in relation to Lonp1 expression in could not show correlation between Lonp1 expression of VAT the sWAT of BXD RI mouse strains, although sWAT is some- and serum glucose, lipid profiles and HOMA-IR, because those what different from VAT in the metabolic profiles [28]. The data were not much different among the participants who have nor- for VAT were not available in the BXD RI mouse strain data- mal glucose tolerance in this study. base. In human VAT, we found that mtDNA count and expression In conclusion, we found that mitochondrial LONP1, which is of genes related to the OXPHOS complex and mitochondrial involved in mitochondrial quality control in response to stress, biogenesis were generally not significantly associated with showed differential expression in VAT that was dependent upon BMI, although UQCRC2 mRNA level and NDUFB8 protein BMI. Moreover, high LONP1 expression in VAT was associated level were lower in the VAT of individuals with high BMI than with enhancement of glucose and lipid metabolism in a bioin- in those with low BMI. In previous studies, the relation between formatics analysis. mitochondrial gene expression of human WAT with obesity has varied according to the depot studied, the presence or absence Copyright © 2021 Korean Endocrine Society www.e-enm.org 669 Lee JH, et al.

CONFLICTS OF INTEREST 7. Mottis A, Jovaisaite V, Auwerx J. The mitochondrial unfolded protein response in mammalian physiology. Mamm Ge- No potential conflict of interest relevant to this article was re- nome 2014;25:424-33. ported. 8. Jovaisaite V, Mouchiroud L, Auwerx J. The mitochondrial unfolded protein response, a conserved stress response path- ACKNOWLEDGMENTS way with implications in health and disease. J Exp Biol 2014;217(Pt 1):137-43. This work was supported by the 2017S-Hyangseol grant (Ju 9. Lee JH, Kim JM, Choi MJ, Kang YE, Joung KH, Yi HS, et Hee Lee) from Korean Diabetes Association, Korea. Ju Hee Lee al. Clinical implications of UCP1 mRNA expression in hu- was supported by the Basic Science Research Program, through man cervical adipose tissue under physiological conditions. the National Research Foundation of Korea (NRF) by the Min- Obesity (Silver Spring) 2018;26:1008-16. istry of Science, ICT (NRF-2020R1C1C1003269), Korea. 10. Kang YE, Kim JM, Joung KH, Lee JH, You BR, Choi MJ, et al. The roles of adipokines, proinflammatory cytokines, AUTHOR CONTRIBUTIONS and adipose tissue macrophages in obesity-associated insulin resistance in modest obesity and early metabolic dys- Conception or design: J.H.L., S.B.J. Acquisition, analysis, or in- function. PLoS One 2016;11:e0154003. terpretation of data: J.H.L., S.B.J., S.E.L., J.E.K., J.T.K., Y.E.K., 11. Venegas V, Halberg MC. Measurement of mitochondrial S.G.K., H.S,Y., Y.B.K., K.H.L., B.J.K., M.S., H.J.K. Drafting DNA copy number. Methods Mol Biol 2012;837:327-35. the work or revising: J.H.L., S.B.J., M.S., H.J.K. Final approval 12. Love MI, Huber W, Anders S. Moderated estimation of fold of the manuscript: J.H.L., S.B.J., H.J.K. change and dispersion for RNA-seq data with DESeq2. Ge- nome Biol 2014;15:550. ORCID 13. Andreux PA, Williams EG, Koutnikova H, Houtkooper RH, Champy MF, Henry H, et al. Systems genetics of metabo- Ju Hee Lee https://orcid.org/0000-0001-5976-7175 lism: the use of the BXD murine reference panel for multis- Saet-Byel Jung https://orcid.org/0000-0003-2471-8108 calar integration of traits. Cell 2012;150:1287-99. Hyun Jin Kim https://orcid.org/0000-0002-6760-4963 14. Williams EG, Wu Y, Jha P, Dubuis S, Blattmann P, Argmann CA, et al. Systems proteomics of liver mitochondria func- REFERENCES tion. Science 2016;352:aad0189. 15. Wu Y, Williams EG, Dubuis S, Mottis A, Jovaisaite V, 1. Sun K, Kusminski CM, Scherer PE. Adipose tissue remod- Houten SM, et al. Multilayered genetic and omics dissection eling and obesity. J Clin Invest 2011;121:2094-101. of mitochondrial activity in a mouse reference population. 2. Kusminski CM, Scherer PE. Mitochondrial dysfunction in Cell 2014;158:1415-30. white adipose tissue. Trends Endocrinol Metab 2012;23:435- 16. World Health Organization. The Asia-Pacific perspective: 43. redefining obesity and its treatment. Geneva: World Health 3. Ozcan U, Cao Q, Yilmaz E, Lee AH, Iwakoshi NN, Ozdelen Organization Western Pacific Regional Office; 2000. E, et al. Endoplasmic reticulum stress links obesity, insulin 17. Liesa M, Shirihai OS. Mitochondrial dynamics in the regu- action, and type 2 diabetes. Science 2004;306:457-61. lation of nutrient utilization and energy expenditure. Cell 4. Held NM, Houtkooper RH. Mitochondrial quality control Metab 2013;17:491-506. pathways as determinants of metabolic health. Bioessays 18. Yi HS, Chang JY, Shong M. The mitochondrial unfolded 2015;37:867-76. protein response and mitohormesis: a perspective on meta- 5. Moehle EA, Shen K, Dillin A. Mitochondrial proteostasis in bolic diseases. J Mol Endocrinol 2018;61:R91-105. the context of cellular and organismal health and aging. J 19. Choi MJ, Jung SB, Lee SE, Kang SG, Lee JH, Ryu MJ, et Biol Chem 2019;294:5396-407. al. An adipocyte-specific defect in oxidative phosphoryla- 6. Melber A, Haynes CM. UPRmt regulation and output: a tion increases systemic energy expenditure and protects stress response mediated by mitochondrial-nuclear commu- against diet-induced obesity in mouse models. Diabetologia nication. Cell Res 2018;28:281-95. 2020;63:837-52. 670 www.e-enm.org Copyright © 2021 Korean Endocrine Society Mitochondrial LON Protease Expression in Obesity

20. Chung HK, Ryu D, Kim KS, Chang JY, Kim YK, Yi HS, et 3135-45. al. Growth differentiation factor 15 is a myomitokine gov- 25. Dahlman I, Forsgren M, Sjogren A, Nordstrom EA, Kaaman erning systemic energy homeostasis. J Cell Biol 2017;216: M, Naslund E, et al. Downregulation of electron transport 149-65. chain genes in visceral adipose tissue in type 2 diabetes in- 21. Nolden M, Ehses S, Koppen M, Bernacchia A, Rugarli EI, dependent of obesity and possibly involving tumor necrosis Langer T. The m-AAA protease defective in hereditary spas- factor-alpha. Diabetes 2006;55:1792-9. tic paraplegia controls ribosome assembly in mitochondria. 26. Qatanani M, Tan Y, Dobrin R, Greenawalt DM, Hu G, Zhao Cell 2005;123:277-89. W, et al. Inverse regulation of inflammation and mitochon- 22. Zurita Rendon O, Shoubridge EA. LONP1 is required for drial function in adipose tissue defines extreme insulin sen- maturation of a subset of mitochondrial proteins, and its loss sitivity in morbidly obese patients. Diabetes 2013;62:855- elicits an integrated stress response. Mol Cell Biol 2018;38: 63. e00412-17. 27. Yin X, Lanza IR, Swain JM, Sarr MG, Nair KS, Jensen MD. 23. Lee HJ, Chung K, Lee H, Lee K, Lim JH, Song J. Down- Adipocyte mitochondrial function is reduced in human obe- regulation of mitochondrial lon protease impairs mitochon- sity independent of fat cell size. J Clin Endocrinol Metab drial function and causes hepatic insulin resistance in human 2014;99:E209-16. liver SK-HEP-1 cells. Diabetologia 2011;54:1437-46. 28. Wajchenberg BL. Subcutaneous and visceral adipose tissue: 24. Heinonen S, Buzkova J, Muniandy M, Kaksonen R, Ol- their relation to the metabolic syndrome. Endocr Rev 2000; likainen M, Ismail K, et al. Impaired mitochondrial biogen- 21:697-738. esis in adipose tissue in acquired obesity. Diabetes 2015;64: