Ktsuy Ngok ,Joud Mull ,Kevin Co ,Zhixing Cheng ,Dn Liu ,Willim Mueller ,Amli By ,Grce Hildernd ,Sholei Lu ,Chiung-Kuei Hung ,*

a Liver Research Center,Division of Gastroenterology＆Liver Research Center,Warren Alpert Medical School of Brown University and Rhode Island Hospital,Providence,RI,USA

b Department of Pathology and Laboratory Medicine,Warren Alpert Medical School of Brown University,Rhode Island Hospital,Providence,RI,USA

Keywords:

Peroxisome proliferator-activated receptor

gamma (PPARγ)

Steatosis

Non-alcoholic fatty liver disease (NAFLD)

Non-alcoholic steatohepatitis (NASH)

Alpha-ketoglutarate (α-KG)

Gender difference

C57BL/6 mouse

ABSTRACT

Background:Non-alcoholic liver disease is of increased concern and contributing to economic burdens not only in developing countries but in developed countries as well.Identifying the biomarker of early diagnosis and early intervention approaches for non-alcoholic liver disease is unmet and required further investigation.Although the alpha-ketoglutarate (α-KG) is recently proposed to be a potential biomarker in differentiating patients with obesity from those with non-alcoholic liver disease,how α-ketoglutatate is involved in the fatty liver progression is not clear.

Methods:A high-fat diet(HFD)feeding animal model,liver functional assays,and molecular approaches were adopted to clarify the impact of α-KG in fatty liver progression.

Results:In the current study,it was found that dietary α-KG would inhibit weight gain in male and female mice fed with a normal chew or HFD.HFD feeding caused fatty liver in male mice,but α-KG treatment could substantially inhibit hepatic steatosis progression.Biochemical studies revealed the possible linkage of α-KG protective functions to lipid metabolism.Further analysis identified the important role of peroxisome proliferator-activated receptors in beneficial α-KG-mediated effects on fatty liver progression.

Conclusions:The current study demonstrates the therapeutic potential of α-KG and how it may be used,via dietary supplementation,as a preventive intervention for non-alcoholic liver disease in obese patients.

1.Introduction

The global increase in metabolic diseases and obesity has led to an unexpected medical care expense in non-alcoholic fatty liver disease (NAFLD) which ranges from simple steatosis,steatohepatitis,liver fibrosis/cirrhosis,to advanced hepatocellular carcinoma(HCC).It is estimated that the number of NAFLD patients will grow to 101 million by 2030 with 27% of them expected to have steatohepatitis.1The patients with steatohepatitis or its advanced stages,including liver fibrosis/cirrhosis and HCC,generally require liver transplantation as a therapeutic intervention.2,3However,patients have a low chance of surviving long enough to receive a liver transplant,indicating the unmet need for early diagnosis as well as a therapeutic approach in NAFLD patients.Identifying early intervention strategies may prevent the incidence of NAFLD patients who are in need of liver transplantation.

The metabolite of citric acid cycle,alpha-ketoglutarate (α-KG)has recently been linked to the severity of obesity and steatohepatitis.4It was found that the plasma α-KG levels are significantly elevated in obese patients when compared with lean control ones.More importantly,the α-KG level was used to differentiate patients with NAFLD from those without NAFLD in a cohort of 230 participants,indicating the potential of α-KG as an early diagnosis marker in NAFLD.4However,it was demonstrated that the α-KG could extend the longevity of Caenorhabditis elegans.5In contrast to the study on obesity,the beneficial effect of α-KG on lifespan was linked to dietary restriction since α-KG could deactivate the target of rapamycin(TOR)signaling.In addition,the α-KG levels were found to be significantly increased in starved bacteria,yeast,6and the liver of starved pigeons.7Similarly,α-KG levels were elevated in humans after exercise.8Based on these opposing observations,we determined that the next step was to analyze the involvement of α-KG in NAFLD progression,which should be further characterized,thus,the current study was aimed to investigate this relationship.

2.Materials and methods

2.1.Cell lines and reagents

The human hepatocyte,OUMS29,which we received from Dr.Hironori Koga,was cultured in DMEM medium as previously described.9Pointe Scientific Glucose (Hexokinase) Liquid Reagents(23-666-278),Pointe Scientific HDL Cholesterol(PEG)reagents(23-666-301),Alkaline Phosphatase (ALP) Reagent Kit (23-666-074),Pointe Scientific Aspartate Aminotransferase(ALT)Liquid Reagents(23-666-087),Pointe Scientific Alanine Transaminase (AST) Liquid Reagents (23-666-120),Stanbio™Cholesterol Liquid Reagent for Diagnostic Set (SB-1010-430),Triglyceride Reagent Set (23-666-410),Triglyceride Standard (23-666-422),alpha-Ketoglutarate (α-KG,AAJ1795022),and oil red O (AAA1298922) were purchased from Fisher Scientific.Linoleic Acid-Oleic Acid-Albumin,100 × (L9655-5 ML) and dimethyl-α-KG (DM-α-KG,349631) were from Sigma-Aldrich.

2.2.Animals and animal study

The animal protocol was approved by the Lifespan Animal Welfare Committee(Institutional Animal Care and Use Committee)of Rhode Island Hospital.Eight-week-old male and female C57BL/6 mice (Charles-River) were used in the study.Mice were fed with normal chew or high-fat diet (HFD,60% of total calories from fat,Bio-Serv,S3282)for 12 weeks.For the groups of α-KG treatment,1%α-KG was prepared in drinking water and changed every other day.Mouse body weights were measured bi-weekly with a portable scale.At the end of experiments,mice were euthanized with overdose isoflurane followed with cervical dislocation.Blood samples were collected via cardiac puncture.Liver samples were dissected,stored at a -80°C refrigerator,and prepared as paraffin tissue block as previously described.10Food and water intakes were measured while mice were fed with the HFD for 10-12 weeks.Food and water weights were determined at the beginning (d0) of experiments and 4 days (d4) after the first measurement.The difference (d4-d0) was calculated as food and water intake.

2.3.Cell culture and in vitro lipid assay

The OUMS29 cells were sub-cultured in 12-well plates.Twentyfour hours later,cells were treated with 0,5,10,15 μM DM-α-KG,0.5 mM dimethyloxallyl glycine(DMOG),and 50 μM deferoxamine(DFO) for one day before adding the linoleic acid-oleic acid-albumin solution (1:100 dilution in culture medium).Cells were incubated with the linoleic acid-oleic acid-albumin solution for 2 days and collected for staining.

2.4.Oil red O staining

The cells were fixed with 37% formaldehyde for 1 h and then washed with 1 × phosphate-buffered saline 3 times before incubating with the oil red O staining solution for 30 min.The stock oil red O staining solution was prepared by adding 0.5 g oil red O powder to 100% isopropanol and stirring for 30 min.The working oil red O solution was prepared by adding 3 parts of the stock oil red O solution with 2 parts of double distilled water freshly.The working solution was mixed for 30 min and then used to stain lipid droplets for minutes.Stained cells were washed with double distilled water for 3 times (10 min each) before taking images.

2.5.Haemotoxylin and eosin (H＆E) staining

The H＆E staining was performed with the assistance of molecular pathology core facility at Rhode Island Hospital.Tissue slides were deparaffinized,re-hydrated,and processed for H＆E staining following the previous description.11

2.6.Western blotting

The antibodies for HSP-90(sc-101494),GAPDH(sc-47724),SIRT-1 (sc-74465),peroxisome proliferator-activated receptor (PPAR) α(sc-398394),PPARδ (sc-74517),PPARγ (sc-7273) were from Santa Cruz Biotechnology.Total protein was prepared from mouse liver tissues homogenized with TissueLyser II (Qiagen)or human hepatocytes treated as indicated,by using radioimmunoprecipitation assay (RIPA) buffer.Total protein concentrations of extracted samples were determined using the Pierce™BCA Protein Assay Kit(23225 ThermoFisher Scientific) via a spectrometer ELISA reader.Fifty μg of total protein per sample was resolved in 10% SDS-page gel and transferred to PVDF membrane.The targeted proteins were detected by hybridizing membranes with the specific first antibody following with hybridization of the secondary antibody conjugated with horseradish peroxidase.Protein expression levels were visualized by using Pierce ECL Western Blotting Substrate and photographed with the Fujifilm image system.Semi-quantification of western blotting results was performed using the Image J software(National Institute of Health (NIH),USA).

2.7.Liver functional assays

AST,ALT,and ALP activities were determined by using the AST,ALT,and ALP ELISA kit (Fisher Scientific) following the instruction manuals.Basically,sera collected from the experimental animals were used to incubate with the reagents in a 96-well plate.The Optical Density (O.D.) for absorbance at specific wavelength was determined by a multiplex ELISA reader.The enzymatic activities of AST,ALT,and ALP were calculated following the instruction.

2.8.Cholesterol,triglyceride,and glucose analysis

Cholesterol,triglyceride,and glucose levels were determined in mouse sera following the instruction manuals.The mouse sera were collected at the end of HFD feeding without fasting.

2.9.Statistical analysis

Prism GraphPad software (San Diego,CA,USA) was used to analyze the results of the data.Experimental data were subjected to Student’s t-test when comparing the intervention group to the relevant control.The data were presented as the mean ± standard error of the mean (SEM).P ＜0.05 was considered statistically significant.

3.Results

3.1.The α-KG metabolite inhibits body weight gain in male and female mice challenged with an HFD

To gain insight into how α-KG affects NAFLD progression,1%α-KG was given,through drinking water,to male and female mice fed with normal chew(NC)or HFD(60%of total calories)for 12 weeks.Consistent with previous studies,12,13HFD feeding substantially promoted body weight (BW) gain in male and female mice.α-KG treatment significantly reduced weight gain in the NC group and almost antagonized the impact of HFD on weight gain in the HFD groups (Fig.1A-D).The analysis of visceral fat tissues suggested that α-KG diminished visceral weight gain in the female mice challenged with an HFD and in the male and female mice fed with an NC diet.Surprisingly,α-KG treatment significantly enhanced visceral fat weight gain and fat weight over BW in male mice fed with HFD(Fig.1E-H),indicating the body metabolism in male mice fed with an HFD may be altered by α-KG.

3.2.NAFLD developed in male mice and α-KG treatment alleviated the disease progression

The liver is one of the major organs that controls body’s metabolism.To further characterize how α-KG impacted the progression of fatty liver,we examined the liver weight,liver weight over BW,and liver histology.Although α-KG slightly reduced liver weight in the male and female NC groups,it did not significantly alter liver weight over BW in the NC groups(Fig.2A-D).When we evaluated the impact of HFD on NAFLD development,it was found that the HFD significantly promoted liver weight gain and liver weight over BW only in male mice (Fig.2A-D).Intriguingly,α-KG treatment robustly reduced liver weight gain and liver weight over BW in male mice fed with an HFD (Fig.2A-D).The histological staining results also revealed that female mice fed with NC or HFD have no pathological changes in the liver.In contrast,male mice developed severe NAFLD and α-KG significantly reversed the deleterious effects of HFD challenged on NAFLD progression (Fig.2E and F).As HFD did not elicit NAFLD progression in female mice due to the short feeding period in this study,we only evaluated how α-KG impacts disease progression by monitoring liver functions in male mice.The HFD feeding induced enzymatic activities of ALT,ALP,and to a less extent in AST in male mice.Interestingly,α-KG treatment could improve liver function in male mice fed with an HFD(Fig.3A-C),indicating that α-KG may be a potential early intervention approach in NAFLD patients.

3.3.The α-KG-mediated suppressive effects on NAFLD progression are associated with lipid metabolism

To investigate how α-KG influenced NAFLD,we determined food and water consumption in male mice fed with NC and HFD in combination with or without α-KG treatment.We did not observe any significant difference regarding water and food intakes between controls and α-KG-treated mice fed with the NC diet (data not shown).Unexpectedly,α-KG treatment enhanced food intake and water consumption in the mice challenged with an HFD(Fig.4A and B).Under these circumstances,it was impossible that α-KG treatment suppressed NAFLD progression via repressing food intake.To further gain insight on how α-KG treatment leaded to reduced steatosis development,we determined lipid profiles and glucose levels in mouse serum.Interestingly,treating mice with α-KG diminished cholesterol and blood glucose levels in mice fed with an HFD (Fig.4C-E).Together,these results indicate the possibility that α-KG may inhibit NAFLD by targeting lipid metabolism and insulin signaling.

3.4.PPARs are involved in α-KG-meditated beneficial effects on NAFLD

Fig.1.Phenotype characterization in male and female mice fed with an HFD and treated with α-KG.

Fig.3.Liver functions were improved in male NAFLD mice by α-KG treatment.

The α-KG metabolite is the intermediate of tricarboxylic acid(TCA)metabolic cycle.14Disruption of the TCA cycle would interfere with the production of acetyl-CoA which may impair the enzymatic function of sirtuin 1 (SIRT-1),15a deacetylase highly involving in NAFLD progression.16Thus,we investigated SIRT-1 expression in the liver samples derived from studied animals.We did not observe any difference in SIRT-1 expression among groups (Fig.5A and B).To further clarify how α-KG influenced NAFLD,we decided to examine the levels of PPARs in the mouse liver samples based on the observation that α-KG inhibited NAFLD but stimulated fat deposition,a similar phenotype found NAFLD mice treated with thiazolidinediones which are PPARγ agonists.17-19In line with previous findings,20,21PPARγ was slightly induced in the liver samples derived from HFD-treated mice.The expression levels of PPARα and PPARδ were found to be slightly decreased.17Interestingly,α-KG treatment significantly enhanced the expression levels of PPARα,PPARδ,and PPARγ (Fig.5A and B),suggesting the involvement of PPARs in α-KG-mediated improvement in NAFLD.To replicate these findings in vitro,we examined PPARs in a human hepatocyte cell line,OUMS-29 challenged with DM-α-KG which is a membrane permeable form of α-KG.9,22The DM-α-KG treatment promoted PPARα and PPARδ consistently at different doses.However,different doses of DM-α-KG may impact PPARγ differentially(Fig.5C).Therefore,it would be important to verify the impact of α-KG on fatty acid intake in the human hepatocytes.We found that α-KG (data not shown) and DM-α-KG (Fig.5D) both inhibited fatty acid intake in OUMS-29 hepatocytes.In summary,our in vitro and in vivo data reproducibly demonstrated that α-KG could inhibit NAFLD progression by targeting lipid metabolism.

4.Discussion

Obesity is one of the major underlying causes for the development of NAFLD,ranging from hepatic steatosis,non-alcoholic steatohepatitis,liver fibrosis/cirrhosis,and possibly to HCC.The α-KG metabolite was recently suggested to be a disease progression marker in obesity and NAFLD patients.4,23Yet little is known about how α-KG is involved in NAFLD progression.In the current study,we characterized that α-KG supplementation could reduce BW gain in both sexes of mice fed with an NC or HFD.As NAFLD developed in male mice fed with an HFD for 12 weeks in the current study,we determined how α-KG impacts NAFLD progression in these groups.It was found that α-KG could suppress lipid deposition in the liver samples of the male mice fed with an HFD.Interestingly,α-KG treatment could enhance visceral fat formation in these male mice.Mechanistically,PPARs were suggested to be involved in α-KG-mediated beneficial effects on NAFLD progression.

Fig.4.Treating mice with α-KG enhanced water and food intakes but improved lipid and glucose metabolism.

Fig.5.α-KG treatment inhibited NAFLD possibly by targeting PPARs.

The beneficial effects of α-KG on disease progression have been demonstrated through extending lifespans,promoting bone homeostasis,improving kidney function,and helping with recovery in trauma and sepsis patients.5,14α-KG extends longevity by suppressing the TOR signaling cascade,5which is a protective mechanism for low nutrition found in animals.24Surprisingly,the 2-hydroxyglutarate which is a proposed onco-metabolite functioning through inhibiting α-KG-dependent enzymes has been demonstrated to increase lifespan though targeting TOR signaling as well.25,26No matter how α-KG or 2-hydroxyglutarate regulates α-KG-dependent enzymes,both of them inhibit TOR signaling,which is thought to protect animals from NAFLD development.27In this situation,α-KG-mediated TOR signaling could not explain our findings that α-KG supplementation diminishes NAFLD progression.To answer how α-KG modulates NAFLD,we identified that the α-KG treated male mice challenged with an HFD displayed the similar phenotypes to the thiazolidinediones(a PPARγ agonist) treated NAFLD patients who have improved NAFLD accompanied by fat gain.17Thus,we analyzed how α-KG treatment affects PPARα,PPARδ,and PPARγ in male mice fed with an HFD.Expectedly,α-KG promotes the expression levels of PPARs whose agonists have been suggested to be potential therapies for NAFLD patients due to the reason that they are key regulators in lipid metabolism.17

As α-KG is the important intermediate involved in the TCA cycle,increasing levels of α-KG may,due to balance of biochemical reaction,possibly lead to citrate accumulation resulting in free fatty acid synthesis.Interestingly,we did observe the activation of sterol response element-binding protein 1c,which is the master transcriptional factor for fatty acid synthesis,in the liver tissues derived from male mice fed with HFD and challenged with α-KG (data not shown).Our data did reveal slight elevation of triglyceride in this group of mice.Besides,supplementation of α-KG could increase succinyl-CoA to stimulate mitochondrial electron transport chain and energy metabolism,subsequently leading to fatty acid oxidation.As expected,the expression levels of medium-chain acyl-coenzyme A dehydrogenase,which is an important enzyme involved in fatty acid oxidation,is slightly increased (data not shown).The increased fatty acid synthesis seems to counter to our finding that α-KG inhibits NAFLD progression.But,the unexpected phenotype that visceral fat weight is significantly increased in male mice fed with an HFD and challenged with a-KG may explain where the fatty acids are transported.

Although the current study does not directly answer why α-KG promotes weight gain of fat in male mice fed with an HFD,there are some possible explanations for the observed phenotype.The PPARγ is the master transcriptional factor controlling adipogenesis.28As we revealed that α-KG could promote the expression levels of PPARs,it is possible that α-KG promotes adipogenesis of HFD-challenged male mice leading to the elevated formation of adipocytes,thus resulting in an increased gain of fat weight.Indeed,it was found that dietary α-KG stimulates beige adipogenesis by enhancing gene demethylation.29If this is the case,2-hydroxyglutarate treatment should suppress adipogenesis.Actually,a study suggested that isocitrate dehydrogenase 1 mutation accompanied by elevated 2-hydroxyglutarate would result in the arrest of cell differentiation through increased global and promoter-specific methylation of histone H3 lysine 9 and histone H3 lysine 27.30Although the original study aimed to characterize glial differentiation,the authors did use 3T3-L1 which is a pre-adipocyte as a representative cell differentiation model in their study.30Therefore,this study further corroborates our assumption that α-KG promotes adipocyte differentiation in HFD-fed male mice.Together,it is possible that α-KG promotes fat storage in adipose tissue from hepatocytes because α-KG treatment could suppress NAFLD but increase visceral fat weight in male mice.

5.Conclusions

The α-KG metabolite,generated from iso-citrate,is an intermediate of TCA cycle.14Additionally,α-KG can be generated from glutaminolysis by which glutamine is converted into glutamate that is subsequently transformed into α-KG by ALT and AST enzymes.31Therefore,circulating α-KG levels are expected to be increased in NAFLD and non-alcoholic steatohepatitis patients when compared with control ones.4In previous studies,it was reproducibly proved that α-KG treatment would suppress oxidative stress.14The current study also demonstrated the beneficial effects of α-KG on suppressing NAFLD progression.It is,therefore,rational to speculate that elevated enzymatic activities of ALT and AST in NAFLD may,by raising the α-KG production,exert protective functions in slowing NAFLD development.Although further experiments might be required in order to validate the hypothesis,the current study conclusively demonstrates the therapeutic potential of α-KG in a preclinical NAFLD model.

Authors’contributions

Study concept and design:K.Nagaoka,and C.-K.Huang.Acquisition of data:K.Nagaoka,J.Mulla,K.Cao,Z.Cheng,W.Mueller,A.Bay,G.Hildebrand,D.Liu,and C.-K.Huang.Analysis and interpretation of data:S.Lu,K.Nagaoka,and C.-K.Huang.Drafting of the manuscript:C.-K.Huang.Revision of manuscript:A.Bay,C.-K.Huang,and S.Lu.Material support:S.Lu.Obtained funding:C.-K.Huang.Study supervision:C.-K.Huang.

Declaration of competing interest

The authors declare that they have no conflict of interest.

Acknowledgements

Funding for this work was provided by 2017 AASLDF Pinnacle Research Development Award and Rhode Island Foundation#134279.