SHMT2 Promotes Liver Regeneration Through Glycine-activated Akt/mTOR Pathway

Abstract

Background. 

The development of liver transplantation (LT) is increasingly being limited by the unavailability of liver grafts. Unique regenerative capacity of liver in response to injuries makes living-donor liver transplantation (LDLT) a feasible strategy to meet clinical demands. Serine hydroxymethyl-transferase 2 (SHMT2) serves as the key enzyme in the biosynthesis of glycine. Glycine affects the activity of mammalian target of rapamycin (mTOR), which is important for cellular growth and proliferation. In this study, the effects of SHMT2 on mouse liver regeneration were investigated using a classical partial hepatectomy (PH) model.

Methods. 

In vivo, PH was performed on mice with or without knockdown of SHMT2. In vitro, SHMT2 was overexpressed in primary hepatocytes, which were cultured in customized Dulbecco’s modified eagle media and LY294002 (an Akt inhibitor). Relevant indexes of liver regeneration, cell proliferation, and Akt/mTOR signal pathways were analyzed.

Results. 

After PH, the expression levels of SHMT2 fluctuated with time and knockdown of SHMT2 in vivo lowered the regenerative ability of liver, with reduced glycine levels compared to the scramble group. In addition, overexpression of SHMT2 in hepatocytes boosted glycine production while enhancing Akt/mTOR pathway activity. These results were validated by the application of LY294002 in vitro.

Conclusions. 

SHMT2 can contribute to liver regeneration after PH, and this is likely related to the activation of Akt/mTOR signaling pathway by its metabolic product, glycine, in hepatocytes. These results might have therapeutic implications for the prognosis of patients undergoing hepatic resection or transplantation.

Liver diseases are among the primary causes of illness and death worldwide. In China, liver diseases have always been regarded as one of the most challenging public health issues, causing a huge burden on families and societies.¹ Ever since the first implementation of deceased-donor liver transplantation (DDLT) in humans,² and owing to a rapid progress in surgical techniques and immunosuppressive therapies, it has been well accepted as the most effective method for the treatment of end-stage diseases and certain metabolic defects in livers. Organ demand, however, exceeds supply by tens of thousands. In recent years, an increase in the application of living donor liver transplantation (LDLT) and advancements in auxiliary partial liver transplantation (APLT) have been effective in overcoming the shortage of liver grafts.^3,4 Unlike in the case of deceased donors, in LDLT, the procurement of grafts from live donors, which involves major hepatectomy, diligent efforts are needed on the part of surgeons to minimize injury and risk to the donor.

Mature hepatocytes have a long life cycle and generally do not undergo cell division. However, in mammals, liver is the only major organ with unique regenerative capacity in response to injuries.^5,6 The key role of liver shifts from maintaining metabolic homeostasis to tissue repair and regeneration after LDLT; these new roles are essential for restoring the function of liver, both in the donor and recipient. Therefore, extensive studies on liver regeneration are necessary not only for ensuring positive clinical outcomes in LDLT, but also for obtaining insights that would be helpful in the development of other hepatic surgeries.⁶

Glycine, a simple but crucial nonessential amino acid (AA), plays key roles in metabolism and nutrition of humans and other mammals.⁷ It has long been demonstrated that glycine has cytoprotective properties in a variety of liver injuries, including those induced by liver transplantation (LT).^8-11 However, the pivotal roles of glycine (and serine) in the regulation of cellular homeostasis were first demonstrated in a variety of cancer cells only recently.¹² Cell growth and proliferation requires synthesis of proteins, nucleic acids, and lipids, which are used in the formation of new cellular components and for maintenance of redox equilibrium. The pathway for serine/glycine biosynthesis is one of the most important pathways of AA metabolism required for the growth and proliferation of cells.^12-14 Serine hydroxymethyl-transferase 2 (SHMT2) occupies a crucial position in serine/glycine metabolism because it catalyzes the conversion of serine to glycine in mitochondria,⁷ and has been identified as an important factor for redox homeostasis during hypoxia.¹⁵ Emerging evidences reveal a positive correlation between the expression of SHMT2 and carcinogenesis in different types of tumors.^16-19

As an important downstream protein kinase of the phosphatidylinositol 3-kinase/protein kinase B (PI3K/Akt) pathway, mammalian target of rapamycin (mTOR), a serine/threonine kinase, serves as a regulator of cell growth, proliferation, survival, and protein synthesis.²⁰ Interestingly, the ability of mTOR to interact with its substrates is affected by AA availability and activation of mTOR could promote hepatocellular proliferation in mice subjected to partial hepatectomy (PH).^21,22

Despite lower ischemic injury in LDLT compared to that in DDLT, considerable regeneration of liver is required, because only 50% to 60% of the expected liver volume (in adults) is implanted and both the donors and recipients must depend on vigorous regeneration of an insufficient liver to regain fundamental homeostasis. In this context, regenerative therapies intend to enhance liver tissue repair and regeneration on all accounts are of particular importance. Here, we speculated that SHMT2 was likely to protect the partial liver by facilitating liver regeneration through glycine-activated Akt/mTOR pathway. To test our hypothesis, in the present study, we analyzed a number of relevant parameters in a mouse model of PH and primary hepatocytes.

MATERIALS AND METHODS

Animals

C57BL/6 mice (male, 19 to 22 g) were obtained from the Experimental Animal Center of Chongqing Medical University (Chongqing, China) and were bred in standard specific pathogen-free (SPF) environment with unlimited water and food. The animal experiments were approved by the ethics committee of the Second Affiliated Hospital of Chongqing Medical University and the animal care and experimental protocols conformed to the Animal Management Rules of the Ministry of Health of the People’s Republic of China.

Adeno-associated Virus-8

Adeno-associated virus-8 (AAV8) with green fluorescent protein (GFP) targeting SHMT2 were constructed by GenePharma (Shanghai, China), as previously described.²³ The AAV8 that merely expressed GFP was used as a scramble control and the sequences of shRNA were showed in Appendix 1 (SDC, https://links.lww.com/TP/B728). Two weeks before the surgeries, mice were administered with 5 × 10¹¹ genome-equivalents of AAV8 by tail vein injection.

Experimental Groups and PH

Mice were randomly divided into three groups as follows: (1) Normal control group (NC): mice underwent sham and PH surgery (sham, n = 9 per time point; PH, n = 9 per time point); (2) Scrambled group: after tail vein injection of AAV8-scramble, mice underwent sham and PH surgery (sham, n = 9 per time point; PH, n = 9 per time point); (3) shSHMT2 group: after tail vein injection of AAV8-shSHMT2, mice underwent sham and PH surgery (sham, n = 9 per time point; PH, n = 9 per time point).

Mice underwent classical PH, as originally described by Higgins and Anderson.²⁴ Briefly, they were first adequately anesthetized by intraperitoneal injection of sodium pentobarbital (50 mg/kg). Thereafter, the left and median lobes of the liver were ligated and resected. The same surgical procedure was performed on the sham groups, without conducting PH. Animals were euthanized after administration of an inhalable anesthetic (2% to 3% isoflurane) at designated times (0, 2, 3, 5, 7, 10 days) after operation and whole blood was acquired via portal vein puncture. Liver tissues were quickly perfused with ice-cold phosphate buffered saline (1×) before they were excised and weighed. Liver/body weight ratio was calculated using the following equation:²⁵ Liver/body weight ratio = (remnant liver weight [g]/body weight [g]) × 100%.

Isolation of Primary Hepatocytes

Primary hepatocytes were isolated according to the collagenase perfusion method introduced by Edwards et al.²⁶ Briefly, mice were fully anesthetized, decontaminated, and liver was immediately infused with buffer 1 and 2 (Appendix 2, SDC, https://links.lww.com/TP/B728) before removing and transferring to a 100 mm cell culture plate for mechanical dissociation. The suspension (1 mL; 5 × 10⁶ hepatocytes/mL) was dispensed into each well of 12-well culture plates and incubated at 37°C in an atmosphere of 5% CO₂ for later use. The viability and purity of hepatocytes were assessed by light microscopy to ensure the presence of at least 90% hepatocytes in the suspension.

Cell Culture and Treatment

The primary hepatocytes were cultured in Dulbecco’s modified eagle medium (DMEM) with 7% fetal bovine serum and a custom-made, glycine-free DMEM (mimicking the physiological concentrations of AAs in the plasma of healthy mice) was also used to culture hepatocytes, as described in a previous study.²⁷ In other treatments, after the cells were starved for 6 hours in serum-free DMEM, 10 μmol/L of LY294002 (an Akt inhibitor) was added to the medium in the cells overexpressed with SHMT2. In SHMT2-Ori group, the original culture medium was added after washing cells with PBS thrice so as to maintain glycine level after the starvation; while in SHMT2-Gly^- group, a glycine- and serum-free medium was used for the starvation to reduce intracellular glycine as indicated by Wang et al,²⁸ and hepatocytes in these groups were cultured for another 2 hours before relative tests.

Assessment of the Viability of Hepatocytes

The cells were first seeded in 96-well plates (10 000 cells/well) for adherence and were then starved by culturing in glycine-free DMEM for 6 hours to reduce intracellular concentrations of glycine, as described by Sun et al.²⁷ Finally, the cells were incubated for 24 hours in the presence of different concentrations of glycine (0, 0.25, 0.5, and 1.0 mmol/L). Cell Counting Kit-8 (MCE, HY-K0301) was used to evaluate the viability, according to the manufacturer’s protocol. The absorbance was measured using xMark Microplate Absorbance Spectrophotometer (BIO-RAD, 1681150) and values were taken from 6 independent experiments.

Cell Transfection

Hepatocytes were transfected for expression of SHMT2 using Lipofectamine MessengerMAX (Invitrogen, LMRNA015), according to the manufacturer’s instructions. For control, the cells were transfected with GFP, in a manner identical to that used for transfection of SHMT2. The transfection efficiency was confirmed by fluorescence detection and western blot analysis (Figure S1, SDC, https://links.lww.com/TP/B728) and cells were collected for further investigations.

Protein Preparation and Western Blot Analysis

Total protein from liver tissues and cells was extracted using RIPA Buffer (Abcam, ab156034) containing protease inhibitor cocktail (Beyotime, P1005) and quantified with BCA Protein Assay Kit (CST, 7780). The proteins were resolved by electrophoresis on SDS-polyacrylamide gels and transferred onto a PVDF membrane. The membranes were incubated overnight with proper primary antibodies (Table S1, SDC, https://links.lww.com/TP/B728) at 4°C and subsequently incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (Table S1, SDC, https://links.lww.com/TP/B728) in room temperature for 1 hour, the next day. The target bands were detected with ChemiDoc Touch Imaging System (BIO-RAD, 1708370) using Western-Bright ECL kit (Advansta, K-12045-D20) and analyzed with Image-Pro Plus software (Media Cybernetics, USA).

Serological Examination

Serum was isolated by centrifugation (1000g, 20 min, 24°C) of coagulated blood. The levels of alanine aminotransferase (ALT) and aspartate transaminase (AST) were determined with a fully automatic biochemical meter (Beckman Coulter, CX7).

Immunohistochemistry

After dewaxing and rehydration of paraffin-embedded sections, a heat mediated antigen-retrieval step was performed in citrate buffer. The samples were then blocked (by treating with 3% H₂O₂) and incubated overnight at 4°C in primary antibodies against SHMT2 (1:50; Proteintech Group, 1099-1-AP) and proliferating cell nuclear antigen (PCNA) (1:200, CST, 13110S). The samples were then incubated in HRP-conjugated antibody and chromogenic detection was performed with DAB, the next day. The stained areas were analyzed with Image-Pro Plus software (Media Cybernetics, USA) and the number of PCNA-positive nuclei was counted in 10 randomly selected fields per section under ×200 magnification.

5-Ethynyl-2′-deoxyuridine Staining

A commercial Cell-Light 5-ethynyl-2′-deoxyuridine (EdU) In Vitro Imaging Kit (RiboBio Co., C10310) was used to assess cell proliferation, according to the manufacturer’s instructions. In brief, EdU was added to the cell culture medium (as indicated earlier) at a final concentration of 100 nmol/L for labeling and standardized permeabilization and fixation was done before incubating the cells with the reaction cocktail (30 min). After fluorescence microscopy, results were expressed as the percentage of EdU-positive cells in 6 independent experiments.

Measurement of Glycine Content

Dried extracts of liver tissue and hepatocytes were first chemically derivatized, as described in a previous published protocol.²⁹ Briefly, the samples were first suspended in sodium hydroxide (1 M) and then resuspended in methanol and pyridine. Thereafter, MCF was added and mixed by vortexing to initiate derivatization; chloroform and bicarbonate (50 mM) were then added sequentially and mixed; finally, the organic layers of samples were extracted by centrifugation (12 000g, 5 min, 24°C). The derivatized samples were transferred into glass inserts inside gas chromatography (GC) vials after adding anhydrous sodium sulfate and empty tubes subjected to the same process were used as negative controls. The derivatized extracts were analyzed with a GC-mass spectrometry (GC-MS) system (Agilent, 7890B-5977A) to quantify glycine. Data extraction, normalization, metabolite identification, relative quantification, and statistical analysis were done as reported by Delplancke et al.³⁰ The retention times for all the metabolites detected by GC-MS were shown in Figure S2 (SDC, https://links.lww.com/TP/B728).

Statistical Analyses

Statistical analyses were performed by using SPSS 25.0 software (IBM, New York). All biological samples were included in the quantitation of each experiments and results were expressed as means ± SD. Comparisons between two groups were performed by two-tailed Student test, whereas comparisons among more than 2 groups were performed by one-way analysis of variance followed by Bonferroni post hoc test; correlations were analyzed by Spearman’s rank test. A P < 0.05 was considered to be statistically significant.

RESULTS

Expression of SHMT2 in Liver Tissues After PH

We first investigated the expression pattern of SHMT2 after PH in the NC group via western bolt and immunohistochemistry (IHC) analysis. The expression levels of SHMT2 in liver tissue after 2, 3, 5, 7, and 10 days of PH were shown in Figure 1A. The expression level of SHMT2 was significantly elevated at 2, 3, and 7 days after PH. (P < 0.001, P < 0.01, P < 0.05, respectively) Two peaks were shown in the early (2 days) and late (7 days) stage of PH in the whole process of liver regeneration. Meanwhile, a similar situation was also detected by IHC (Figure 1B), which was suggested by the significant elevation of SHMT2-positive cells at 2 and 7 days of PH (P < 0.001, P < 0.01, respectively), when compared to the sham group. To determine the function of SHMT2 in liver regeneration, mice were subjected to AAV8 injection to selectively knockdown the SHMT2 gene in hepatocytes in vivo and the design of animal experiment was illustrated in Figure 1C.

FIGURE 1.: Expression of serine hydroxymethyl-transferase 2 (SHMT2) in the mouse model of liver regeneration. A, Expression of SHMT2 measured by western blot analysis at designated time points after partial hepatectomy (PH) and densitometric analysis of the bands obtained in the western blot. B, Representative pictures of SHMT2-positive cells in tissue sections from mice in the normal group as detected by immunohistochemistry (magnification: ×400) and quantification of positively stained areas. C, A brief diagram of animal experiments (values are means ± SD, animals in each group ≥ 4, ***P < 0.001). SD, standard deviation.

Knockdown of SHMT2 Inhibits Liver Regeneration

Two weeks after AAV8 injection, mice were subjected to PH surgeries and the knockdown effect of transfection was verified by the GFP expression and western blot analysis (Figure S3, SDC, https://links.lww.com/TP/B728). To evaluate the effect of SHMT2 on liver function during liver regeneration, we first investigated the serum level of AST and ALT. The ALT levels in shSHMT2 group at 2, 3, and 5 days of PH were significantly higher than those in the Scrambled group (P < 0.05) (Figure 2A), whereas the levels of AST in shSHMT2 group at 3, 5, and 7 days of PH were significantly higher than those in the Scrambled group (P < 0.05) (Figure 2B). It is important to note that no such differences were observed between NC and Scrambled group (Figure 2A and B) indicating that the liver function was not damaged by AAV8 transfection. The expression of PCNA is well accepted as a marker of cell proliferation, so the regeneration status of liver tissues was determined by PCNA staining via IHC (Figure 2C). Knockdown of SHMT2 significantly reduced the PCNA-positive cells in shSHMT2 group in the early stage (2 days and 3 days after PH) of liver regeneration compared to the Scrambled group (P < 0.05) (Figure 2D); however, the PCNA-positive cells remained unchanged between mice in NC and Scrambled group (Figure 2D). Liver/body weight ratios were also calculated to reflect the regenerative ability of the liver (Figure 2E) and representative liver samples from NC, scramble, and shSHMT2 group were demonstrated in Figure S4 (SDC, https://links.lww.com/TP/B728). Compared with the Scrambled group, mice in shSHMT2 group had a significantly lower liver/body weight ratios (P < 0.05, P < 0.01, P < 0.001, respectively) at 5, 7, and 10 days after PH. These results suggested that downregulation of SHMT2 in vivo dampened the function of liver thereby inhibiting the proliferation process of hepatocytes in the early stage of liver regeneration.

FIGURE 2.: Knockdown of serine hydroxymethyl-transferase 2 (SHMT2) inhibits liver regeneration in mice. A and B, Levels of alanine amino transferase (ALT) and aspartate transaminase (AST), respectively, in the serum. C, Representative immunohistochemical staining of proliferating cell nuclear antigen (PCNA) in each group after partial hepatectomy (PH; magnification: ×200). D, Quantification of PCNA-positive cells in each group after PH (*P < 0.05; n ≥ 4). E, Liver/body weight ratio of mice in each group after PH (values are means ± SD, animals in each group ≥ 4, *P < 0.05, **P < 0.01, ***P < 0.001). NC, normal control; SD, standard deviation.

SHMT2 Activates Akt/mTOR Signaling and Glycine Production During Liver Regeneration

mTOR is composed of two multiprotein complexes, namely, mTOR complex 1 (mTORC1) and mTOR complex 2 (mTORC2). mTORC1 regulates cellular growth, autophagy, and apoptosis; therefore, the activation of mTORC1 is curial for the cell growth and proliferation. We detected the Akt/mTOR signaling in the most active period (2 days of PH) of liver regeneration via western blot analysis (Figure 3A). After PH treatment, the activity of Akt/mTOR signaling was significantly upregulated in the NC group compared to the Sham group, which was suggested by the phosphorylation of Akt, mTOR, and P70^S6K (Figure 3B–E). Downregulation of SHMT2 in shSHMT2 group inhibited the expression of SHMT2 as well as the expression levels of p-Akt, p-mTOR, and p-P70^S6K (Figure 3B–E), when compared to the Scrambled group. The relative level of glycine in liver tissues increased remarkably in the NC group, 2 days after PH (Figure 3F); however, when the key enzyme of glycine biosynthesis was inhibited, the relative level of glycine decreased in the shSHMT2 group, 2 days after PH (Figure 3F).

FIGURE 3.: Knockdown of serine hydroxymethyl-transferase 2 (SHMT2) inhibits Akt/mTOR signaling and reduces glycine production. A, Expression of key proteins of Akt/mTOR signaling pathway as detected by western blot analysis of liver tissue samples after partial hepatectomy (PH). B–E, Densitometric analysis of the western blot results. F, Glycine levels, as detected by GC-MS of samples, in each group (values are means ± SD, *P < 0.05, **P < 0.01, ***P < 0.001). GC-MS, gas chromatography-mass spectrometry; mTOR, mammalian target of rapamycin; NC, normal control; SD, standard deviation.

Glycine Activates Akt/mTOR Signaling in Primary Mouse Hepatocytes

Primary mouse hepatocytes were isolated to validate the effects of glycine on proliferation in vitro. Hepatocytes were first starved to reduce the concentration of intracellular glycine before the supplementation of exogeneous glycine. The EdU assay suggested that glycine promoted proliferation of hepatocytes in a dose-dependent manner (0.25 to 1 mmol/L) and its positive effect peaked at a concentration of 0.5 mmol glycine/L (Figure 4A and B). The cell viability demonstrated a similar result for it increased significantly upon supplementation of 0.25 to 1 mmol/L glycine (Figure 4C). The upregulation of Akt/mTOR signaling pathway in hepatocytes by exogeneous glycine also suggested a dose-dependent effect compared to the case where no glycine was supplemented, as indicated by the western blot analysis (Figure 4D–G).

FIGURE 4.: Glycine promotes proliferation of hepatocytes by activating Akt/mTOR signaling in vitro. A, EdU staining was used to assess the proliferation of primary mouse hepatocytes in the presence of different concentrations of glycine (magnification: ×400). B, Quantification of EdU-positive cells in each group. C, Growth of primary mouse hepatocytes cultured for 24 h in the presence of different concentrations of glycine. D, Expression of key proteins of Akt/mTOR signaling pathway as detected by western blot analysis of mice primary hepatocytes cultured in the presence of different concentration of glycine (values are means ± SD, *P < 0.05). EdU, 5-Ethynyl-2’-deoxyuridine; mTOR, mammalian target of rapamycin; SD, standard deviation.

SHMT2 Overexpression Activates Akt/mTOR Signaling and Glycine Production in Primary Mouse Hepatocytes

Primary hepatocytes transfected with SHMT2 were applied to examine the positive effects of SHMT2 on cell proliferation. EdU and GC-MS assay indicated that overexpression of SHMT2 enhanced the proliferation of hepatocytes and intracellular glycine compared to the GFP group (Figure 5A–C). Upregulation of SHMT2 also promoted the activity of Akt/mTOR signaling pathway, for the phosphorylation levels of Akt, mTOR, and P70^S6K were found to be elevated via western blot analysis in SHMT2 group (Figure 5D–H). However, the positive role of SHMT2 on both cell proliferation and Akt/mTOR signaling pathway was eliminated markedly up on the administration of LY294002, a PI3K/Akt inhibitor (Figure 5D–H), while the relative level of intracellular glycine remained unchanged compared to the SHMT2 group (Figure 5C).

FIGURE 5.: Overexpression of serine hydroxymethyl-transferase 2 (SHMT2) activates Akt/mTOR signaling and enhances glycine production in vitro. A and B, Primary hepatocytes overexpressing SHMT2 were incubated with or without LY294002 and cell proliferation was detected by EdU staining (magnification: ×400) and quantification of the EdU staining in each group. C, Primary hepatocytes overexpressing SHMT2 were incubated with or without LY294002 and glycine levels were detected by GC-MS. D, Primary hepatocytes overexpressing SHMT2 were incubated with or without LY294002, and the abundance of SHMT2 and key proteins of Akt/mTOR signaling pathway were evaluated by western blot analysis. E–G, Densitometric analysis of the western blot results (values are means ± SD, *P < 0.05, **P < 0.01, ***P < 0.001). EdU, 5-Ethynyl-2’-deoxyuridine; GC-MS, gas chromatography-mass spectrometry; mTOR, mammalian target of rapamycin; SD, standard deviation.

SHMT2-Activated Akt/mTOR Signaling is Inhibited Through Depletion of Glycine

To determine whether the activation of the Akt/mTOR signaling by SHMT2 is depended on the glycine, a glycine-free medium was used. The percentage of EdU-positive hepatocytes was decreased remarkably upon depletion of glycine via the glycine-free medium (P < 0.05) (Figure 6A and B). Western blot analysis showed that overexpression of SHMT2 could induce phosphorylation of Akt, mTOR, and P70^S6K; however, its effect was restrained upon depletion of intracellular glycine in the SHMT2-Gly⁻ group when compared to the SHMT2-Ori group (Figure 6B–E).

FIGURE 6.: Depletion of glycine blocks Akt/mTOR signaling. A, Primary hepatocytes overexpressing serine hydroxymethyl-transferase 2 (SHMT2) were incubated in custom-made DMEM to deplete glycine and EdU staining was performed to evaluate their proliferation (magnification: ×400). B, Quantification of the EdU staining. C, Key proteins of Akt/mTOR signaling pathway were evaluated by western blot analysis of cells cultured with or without glycine. D–F, Densitometric analysis of the western blot results (values are means ± SD, *P < 0.05, **P < 0.01, ***P < 0.001). EdU, 5-Ethynyl-2′-deoxyuridine; GFP, green fluorescent protein; mTOR, mammalian target of rapamycin; SD, standard deviation.

Expression of HIF-1α During Liver Regeneration

Hypoxia is a common situation during the rapid process of liver regeneration, which is induced by the reduction in blood flow after PH.³¹ Previous studies demonstrate the induction of SHMT2 upon hypoxia is a hypoxia-inducible factor-1 (HIF-1)-dependent manner in cancer samples,¹⁵ and considering its important function of transcriptional regulation under hypoxia, we tested the expression levels of both SHMT2 and HIF-1α in the liver tissue after PH via western blot analysis (Figure 7A). The Spearman’s rank test indicated a positive correlation (P < 0.01) between the expression of HIF1-α and SHMT2.

FIGURE 7.: Expression of hypoxia-inducible factor-1α (HIF1-α) during liver regeneration in mice. A, Expression levels of HIF1-α and serine hydroxymethyl-transferase 2 (SHMT2) in the normal group were determined by western blot analysis at the indicated time points after partial hepatectomy (PH). B, Correlation analysis between the expression of HIF1-α and SHMT2. C, A model for the regulation of liver regeneration by SHMT2 (*P < 0.05). mTOR, mammalian target of rapamycin.

DISCUSSION

DDLT is a proven curative therapy for end-stage liver diseases. However, barriers of culture, religion, and society make it a less likely treatment for a majority of patients in Asia.³² Several advantages, such as lower mortality of patients in the waiting list and lesser suffering, have made LDLT an effective therapeutic choice compared to DDLT.³³ Restoration of liver architecture and function in both the donor and recipient is the key to a successful transplantation. When the liver fails to regenerate, causing dysfunction and failure, patients undergo small-for-size syndrome (SFSS) after transplantation or extensive liver resection.^34-38

In the present study, we targeted the intracellular synthesis of glycine, which is induced by SHMT2, using serine to generate tetrahydrofolate-activated 1-C units and glycine. The expression level of SHMT2 fluctuated with the postoperation time and reached a relatively high level at 2 days of PH, which is consistent with the time for the peak in DNA synthesis in rodent models of PH, as previously reported.³⁹ The downregulation of SHMT2 gene in hepatocytes in vivo resulted in deteriorated liver function, delayed cell proliferation, and lower glycine levels at an early stage of liver regeneration compared to that in the scramble group.

Interestingly, in recent studies, it has been suggested that SHMT2 might be an underlying therapeutic target in many cancers. Patients with intrahepatic cholangiocarcinoma and breast cancers showing higher SHMT2 levels have a lower survival rate than those with lower expression of SHMT2.^18,40 Woo et al¹⁶ confirmed the upregulation of SHMT2 in various liver cancer cell lines, and knockdown of SHMT2 was reported to suppress cell proliferation and tumorigenicity in Huh-7 and HepG2 cells. In contrast, overexpression of this gene promoted growth, instead of malignant transformation, in an immortalized hepatic cell line. A study in Drosophila also indicated that SHMT-dependent metabolites would suffice for the development of embryos at specific times.⁴¹ By a similar mechanism, these findings might explain the delay in regenerative process in shSHMT2 mice after PH, because both normal and malignant cells need to reprogram their metabolic pathways to sustain rapid growth and proliferation. However, in the late stage (10 days of PH) of liver regeneration, most parameters were normalized apart from the liver/body weight ratio, suggesting that although SHMT2 could facilitate liver regeneration, it was still not a decisive factor. Liver regeneration is a highly organized biological process that includes many growth factor- and cytokine-mediated signal pathways (eg, TNF-α, IL-6, and TGF-β) in both parenchymal and nonparenchymal cells.⁵ As a limitation of this study, the results obtained only indicate that SHMT2 promotes regeneration via the Akt/mTOR-dependent pathway in hepatocytes at an early stage; the effects of SHMT2 in different cell types, including hepatic stellate cells (HSCs), Kupffer cells, and hepatic sinusoidal endothelial cells, were not specified. A recent study by Langiewicz et al⁴² has demonstrated that JNK1-mediated induction of Indian hedgehog (IHH) paracrine signaling from HSCs can accelerate liver regeneration in the mouse model of associating liver partition and portal vein ligation for staged hepatectomy (ALPPS). Transgenic models may provide stronger evidences, but Tani and coworkers⁴³ observed that mice deficient in SHMT2 were embryonically lethal because of respiration defects and retardation of cell growth. Therefore, AAV8 transfection was applied, considering its higher affinity to hepatocytes as well as because of its proven safety in clinical applications.^44,45

Mechanisms underlying typical signaling pathways, including PI3K/Akt, have been defined in detail. In contrast, metabolic regulation of liver regeneration has started to garner attention only recently.^46,47 Environmental conditions, such as nutrients and energy status, can activate mTOR either directly or indirectly (via the upstream PI3K/Akt), and activated mTOR in turn promotes translation by phosphorylating P70^S6K.⁴⁷ The present data indicate that overexpression of SHMT2 promotes cell proliferation by increasing the glycine level and by activating mTOR through an Akt-dependent manner, because such effects could be blocked by the administration of LY294002 in vitro. This is consistent with studies in mouse myoblasts and pig intestinal epithelial cells, which suggested that glycine could stimulate protein synthesis via mTOR signaling pathway, thereby, regulating cell proliferation.^27,28

During regeneration, hypoxia constantly progresses when the rapid growth of hepatocytes exceeds the ability of available vasculature to provide oxygen and nutrients to liver tissues.^15,31 Interestingly, parallel expression of HIF-1α could be found in rat model of both PH and portal vein ligation, whereas HIF-1α was also reported to act as upstream of SHMT2.^15,31,48 We showed a positive correlation between the expression profiles of HIF-1α and SHMT2 in the mouse model of PH. It is, thus, reasonable to speculate that the elevated expression of SHMT2 is likely related to the increasing expression of HIF-1α under hypoxia; however, no direct evidences have been provided in the present study. Therefore, further efforts should be made to reveal if there is any direct interaction between HIF-1α and SHMT2 during liver regeneration. Although SHMT2 appears to be responsible for the bulk of glycine synthesis in most cell types,⁴⁹ and glycine metabolism is a complex process involving multiple metabolic pathway, isotopic labelling studies might better reveal the essence of SHMT2 in the metabolism of glycine.

In conclusion, the data present here suggest that the level of SHMT2 is elevated in the mouse model of liver regeneration, which might be related to the elevated expression of HIF-1α. In addition, downregulation of SHMT2 in vivo would delay the process of liver regeneration. Besides, SHMT2-activated Akt/mTOR signaling pathway could promote proliferation of mouse hepatocytes through its metabolic product, glycine (Figure 7B). Collectively, SHMT2 may benefit the prognosis of patients undergoing hepatic resection or transplantation.