Fatty liver disease protective MTARC1 p.A165T variant reduces the protein stability of MTARC1

Abstract

Non-alcoholic fatty liver disease (NAFLD) is one of the most common causes of liver disease worldwide. MTARC1, encoded by the MTARC1 gene, is a mitochondrial outer membrane-anchored enzyme. Interestingly, the MTARC1 p.A165T (rs2642438) variant is associated with a decreased risk of NAFLD, indicating that MTARC1 might be an effective target. It has been reported that the rs2642438 variant does not have altered enzymatic activity so we reasoned that this variation may affect MTARC1 stability. In this study, MTARC1 mutants were generated and stability was assessed using a protein stability reporter system both in vitro and in vivo. We found that the MTARC1 p.A165T variant has dramatically reduced the stability of MTARC1, as assessed in several cell lines. In mice, the MTARC1 A168T mutant, the equivalent of human MTARC1 A165T, had diminished stability in mouse liver. Additionally, several MTARC1 A165 mutants, including A165S, A165 N, A165V, A165G, and A165D, had dramatically decreased stability as well, suggesting that the alanine residue of MTARC1 165 site is essential for MTARC1 protein stability. Collectively, our data indicates that the MTARC1 p.A165T variant (rs2642438) leads to reduced stability of MTARC1. Given that carriers of rs2642438 show a decreased risk of NAFLD, the findings herein support the notion that MTARC1 inhibition may be a therapeutic target to combat NAFLD.

1. Introduction

Non-alcoholic fatty liver disease (NAFLD) is one of the most prominent causes of liver disease worldwide, affecting both adults and children. NAFLD will likely emerge as the leading cause of end-stage liver disease in the coming decades [1], but there are currently no pharmacological treatments available. To discover new therapeutic options, there is an urgent need to understand the molecular mechanisms driving the progression of NAFLD. The mitochondrial amidoxime-reducing component (MTARC1), a Mo (molybdenum)-containing enzyme encoded by the MTARC1 gene, is involved in the activation of N-hydroxylated prodrugs [2–4]. However, the physiological role of MTARC1 remains unknown.

Interestingly, MTARC1 p.A165T (rs2642438), a variant with 29 % of minor allele frequency (MAF), is associated with a decreased risk of NAFLD [5–9]. Since MTARC1 p.A165T (rs2642438) does not alter the enzymatic activity of MTARC1 [10], we reasoned that this variant may affect the stability of this protein. Herein, we developed a sensitive reporter system to explore the effect of several different mutations on MTARC1 stability both in vitro and in vivo. Using this reporter, we discovered that MTARC1 A165T has reduced stability. Furthermore, substituting alanine 165 for other amino acids with different features also led to a decrease in MTARC1 stability as well, suggesting that the alanine residue of this site is critical for its protein stability. Given that the MTARC1 p.A165T (rs2642438) variant is highly associated with a decreased risk of NAFLD, our findings support the notion that the inhibition of MTARC1 could be targeted therapeutically to combat NAFLD.

2. Materials and methods

2.1. Cell culture

HEK293T, HepG2, Li-7, LO2, and Huh-7 cells were maintained in DMEM supplemented with 10 % fetal bovine serum. To obtain stable cell lines, cells of interest were infected with lentivirus and selected with puromycin at the dosage of 2 μg/mL for 72 h.

2.2. Virus packaging

Adeno-associated virus serotype 8 packaging and production were performed by OBiO Technology (Shanghai). Lentivirus was packaged as described previously [11]. Briefly, HEK293T cells were transfected with pMD2.G, psPAX2, and transfer plasmid, and the medium containing the virus was harvested at 72 h post-transfection.

2.3. Plasmids

Plasmid pLJM1-GFP (Addgene #19319) was digested with AgeI and EcoRI. Then the fragment of HA-P2A-GFP prepared by PCR was cloned into the position of GFP by using an in-fusion cloning strategy following the instruction provided in the kit (Vazyme, C115) to generate a backbone plasmid pLJM1-EcoRI-HA-P2A-GFP. Wild-type (WT) MTARC1 ORF was prepared by PCR using cDNA from human cell line HepG2 and cloned into the site of EcoRI of pLJM1-EcoRI-HA-P2A-GFP using in-fusion cloning to generate plasmid pLJM1-MTARC1-HA-P2A-GFP. For each mutant of MTARC1, two fragments were prepared by PCR using the WT MTARC1 as a template and cloned into pLJM1-EcoRI-HA-P2A-GFP with the same cloning strategy as described above. Cre of AAV.TBG. PI.Cre.rBG plasmid (Addgene #107787) was replaced by HA-P2A-GFP to make a backbone plasmid AAV.TBG.HindIII.HA.P2A.GFP using the strategy as reported previously [12]. Mouse Mtarc1 was amplified by PCR using cDNA from the mouse liver as the template and then cloned to the HindIII site of AAV.TBG.HindIII.HA.P2A.GFP to make a plasmid AAV.TBG.MTARC1.HA.P2A.GFP. For mouse Mtarc1 p.A168T, two fragments were prepared by PCR using the WT Mtarc1 as a template and cloned into pLJM1-EcoRI-HA-P2A-GFP with the same cloning strategy as described above.

2.4. Animal experiments

Animal experiments were performed following the guidelines provided by the Ethics Committees of Anhui Medical University. C57BL/6J male mice were purchased from GemPharmatech (Nanjing, China). All mice had free access to water and food (normal chow diet) and were housed in a specific pathogen-free environment with a 12-h light/12-h dark cycle. Adeno-associated virus serotype 8 (AAV8), suspended in 200 μL phosphate-buffered saline (PBS), was injected via the tail vein at the dosage of 2e¹¹ viral genome/mouse. Mice were sacrificed for liver collection at two weeks post-virus administration.

2.5. Immunoblotting and immunoprecipitation

Cells and liver tissues were subjected to protein extraction as described previously [12,13]. Briefly, cells and liver tissues were lysed with RIPA buffer supplemented with a proteinase inhibitor cocktail, homogenized using a homogenizer, and then centrifuged at 12,000 rpm for 10 min. The supernatant was collected and subjected to bicinchoninic acid (BCA) assay. Forty μg of total proteins for each sample were loaded for SDS-PAGE with a standard protocol. Immunoprecipitation was performed as described previously [13]. Briefly, harvested cells were lysed with a lysis buffer (Tris-HCl 20 mM, NaCl 100 mM, EDTA 1 mM, Glycerol 10 %, SDS 0.2 %, TritonX-100 0.5 %, NP-40 0.1 %). The lysate was centrifuged at 12,000 rpm for 10 min, and then the supernatant was transferred to a new tube, and subjected to the preclear operation using anti-mouse nanobody agarose beads and immunoprecipitation using anti-HA nanobody agarose beads. Antibodies are listed below: HA (51064–2-AP), GFP (50430–2-AP), Vinculin (66305–1-Ig), Ubiquitin (10201–2-AP), Tubulin (11224–1-AP), anti-mouse HRP (SA00001–1), and anti-Rabbit HRP (SA00001–2) from Proteintech (China); anti-mouse nanobody agarose beads (KTSM1341), and anti-HA nanobody agarose beads (KTSM1305) from AlpaLifeBio (China). The quantification of blotting bands was performed using ImageJ.

3. Results

3.1. MTARC1 p.A165T promotes the degradation of MTARC1

Given that MTARC1 p.A165T (rs2642438) does not alter the enzymatic activity of MTARC1 [10], we reasoned that MTARC1 p.A165T may affect the stability of MTARC1. To test this hypothesis, we linked MTARC1 and GFP with the P2A element—a “self-cleaving” peptide [14], which makes MTARC1 and GFP have a fixed ratio of protein synthesis. They are “cleaved” once they are synthesized so that GFP can act as an internal reference to compare the stability of MTARC1 variants (Fig. 1A). We named this method the Protein Stability Reporter System (PSRS). Utilizing PSRS, we found that MTARC1 p.A165T had dramatically reduced the stability of MTARC1 as assessed in multiple cell lines (Fig. 1B, C, 1D, 1E, 1F, and 1G; Figs. S1A and S1B). To confirm the conclusion derived from PSRS, we chased HepG2 cells stably expressing MTARC1 variants with cycloheximide—an inhibitor of mRNA translation [15] and found that the half-life of MTARC1 A165T is much shorter than the wild-type protein (3.5 vs 11.5 h) (Fig. 1H and I). In line with this observation, MARC1 A165T is highly ubiquitinated (Fig. 1J). Interestingly, MTARC1 p.M187K, a raw variant associated with a decreased risk of NAFLD as well [16], does not affect protein stability (Figs. S1C, S1D, S1E, and S1F), indicating MTARC1 p.M187K might affect the enzymatic activity of MTARC1. Taken together, MTARC1 p. A165T (rs2642438) promotes the degradation of MTARC1.

Fig. 1.

MTARC1 p.A165T variant reduces the stability of MTARC1. (A) The graphic view of the Protein Stability Reporter System. HA: HA tag; P2A: P2A element—a “self-cleaving” peptide. (B–C) HEK293T, (D–E) HepG2, and (F–G) Huh-7 cells stably expressing MTARC1-HA-P2A-GFP were subjected to immunoblotting and quantification (n = 2/group). (H–I) HepG2 cells stably expressing MTARC1-HA wild-type (WT) or A165T variant were subjected to cycloheximide chase (300 μM), and harvested for immunoblotting and quantification at the indicated time. The half-life of MTARC1 WT and A165T are 11.5 and 3.5 h, respectively. (J) HEK293T cells stably expressing MTARC1-HA WT and A165T were subjected to immunoprecipitation (IP) using an anti-HA antibody followed by immunoblotting. Data were expressed as mean ± SEM and analyzed by Student’s t-test. ***p < 0.001.

3.2. Mouse Mtarc1 p.A168T, the counterpart of human MTARC1 p. A165T, promotes the degradation of MTARC1

Mouse MTARC1 A168T is the counterpart of human MTARC1 A165T (Fig. 2A). To confirm the findings above in vivo, we coupled the PSRS with adeno-associated virus serotype 8 (AAV8) to compare the protein stability of different variants in mouse liver (Fig. 2B). In line with our in vitro findings, mouse MTARC1 A168T shows remarkably reduced stability compared to the wild type protein as assessed in mouse liver (Fig. 2C and D), indicating that MTARC1 p.A165T variant modulates the stability of MTARC1 in an evolutionarily conserved manner.

Fig. 2.

Mouse Mtarc1 p.A168T variant, the counterpart of human MTARC1 p.A165T, reduces the stability of MTARC1. (A) The graphic view showing the alignment of human and mouse MTARC1 protein. (B) The workflow of the experiment comparing protein stability of mouse MTARC1 WT and A168T variant in vivo. (C–D) Liver samples from (B) were subjected to immunoblotting and quantification (n = 3–5/group). Data were expressed as mean ± SEM and analyzed by Student’s t-test. ***p < 0.001.

3.3. Alanine residue of MTARC1 site 165 is essential for MTARC1 protein stability

Next, we explored how MTARC1 p.A165T increases MTARC1 degradation. The β-position of threonine is a hydroxyl group. Compared to the methyl group of alanine, the hydroxyl group of threonine is hydrophilic. To mimic this feature, we generated an MTARC1 A165S mutant that also showed reduced stability as assessed in HEK293T cell (Fig. 3A and 3B). Of note, both threonine (T) and serine (S) can be phosphorylated. To test whether the capability of phosphorylation of this site could manipulate its stability, MTARC1 A165D, a phosphorylation mimetic mutant, was generated and showed reduced stability as well (Fig. 3A and 3B). Surprisingly, MTARC1 A165V, a dephosphorylation mimetic mutant, showed reduced stability also (Fig. 3A and 3B), suggesting that the capability of phosphorylation of MTARC1 A165T does not affect MTARC1 stability. Additionally, both MTARC1 A165G and A165 N exhibited reduced stability as well (Fig. 3A and 3B). The findings above were further confirmed in HepG2 cells (Fig. 3C and D). Taken together, our data suggests that the alanine residue of MTARC1 site 165 is essential for MTARC1 protein stability.

Fig. 3.

Alanine residue of MTARC1 site 165 is essential for MTARC1 protein stability. (A–B) HEK293T and (C–D) HepG2 cells stably expressing the indicated MTARC1 variants were subjected to immunoblotting and quantification.

4. Discussion

MTARC1 p.A165T (rs2642438), a newly identified variant with 29 % of MAF, is significantly associated with lower ALT levels, lower hepatic fat, and lower all-cause liver cirrhosis [6,8,16]. These findings suggest that MTARC1 could be a potential therapeutic target for liver disease. However, the mechanism underlying this seemingly protective variant remains unknown. Interestingly, Ott et al. found that both the wild-type (WT) and p.A165T variant of MTARC1 expressed in E.coli present comparable binding ability of its cofactor molybdenum [10]. Furthermore, MTARC1 A165T showed no distinguished kinetic parameters in the N-reduction of benzamidoxime [10], suggesting that MTARC1 p.A165T might not alter the enzymatic activity of MTARC1.

In this study, we found that MTARC1 p.A165T robustly promotes the degradation of MTARC1, implying that the loss/inhibition of MTARC1 could prevent liver disease progression. Of note, Hudert et al. found that MTARC1 is expressed at similar levels across rs2642438G > A genotypes in a relatively small cohort, as assessed by proteomics, although they predicted that the variant MTARC1 p.A165T could exert a destabilizing effect [7]. Since the information on MTARC1 messenger is absent, it’s not feasible to assess the stability of MTARC1 variants in that work [7]. Interestingly, it seems that the alanine residue of the MTARC1 165 site is critical for protein stability since replacing alanine with several other amino acids promotes the degradation of MTARC1 as well, implying that the substitution of this amino acid might lead to the misfolding of the MTARC1 protein followed by degradation. However, how the p.165 mutant of MTARC1 is degraded requires further study.

Notably, no cases of liver cirrhosis are observed in a small cohort of carriers of MTARC1 p.R200Ter—a rare truncated variant with 0.009 % of MAF [6]. Since this observation derives from a small cohort and the carriers are heterozygous, it might not be appropriate to conclude that the loss of MTARC1 could lead to the protective effect of liver disease. Of note, MTARC1 p.M187K, with 1.1 % of MAF, is also associated with a reduced risk of liver disease, similar to MTARC1 p.A165T [16]. Our data suggests that MTARC1 p.A165T, but not p.M187K, promotes the degradation of MTARC1 protein, implying these two variants cause the same phenotype, most likely via two different mechanisms. Given that the former promotes the degradation of MTARC1, the latter might inhibit the enzymatic activity of MTARC1. While this work was in progress, Lewis et al. found that hepatocyte-specific MTARC1 knockdown alleviated diet-induced hepatic fat accumulation, but not fibrosis [17]. Given the findings of Lewis [17], our results here suggest that the deficiency of MTARC1 in non-hepatocytes might contribute to the decreased risk of all-cause fibrosis/cirrhosis associated with MTARC1 p. A165T.

In summary, our data suggests that MTARC1 p.A165T but not p. M187K promotes the degradation of MTARC1 protein. Given that MTARC1 p.A165T with 29 % of MAF is highly associated with a reduced risk of liver disease, our data support the notion that the inhibition of MTARC1 could be targeted therapeutically to treat liver diseases such as fibrosis.

Abbreviations

NAFLD

Non-alcoholic fatty liver disease

MTARC1

Mitochondrial Amidoxime Reducing Component 1

AAV8

Adeno-associated virus serotype 8

IP

immunoprecipitation

MAF

minor allele frequency

PSRS

Protein Stability Reporter System

Data availability statement

Further information and requests for resources, and reagents should be directed to, and will be fulfilled by, the lead contact, AH (heanyuan85@foxmail.com).