Lin28 promoting the protective effect of PMSCs on hepatic ischaemia–reperfusion injury by regulating glucose metabolism

Abstract Human placental mesenchymal stem cells (PMSCs) can prevent liver ischaemia–reperfusion injury (LIRI). However, their therapeutic effects are limited. Therefore, additional research is required to elucidate the mechanisms of PMSC‐mediated LIRI prevention and enhance the related therapeutic effects. This study aimed to examine the role of the Lin28 protein in the regulation of glucose metabolism in PMSCs. Further, it explored whether Lin28 could enhance the protective effects of PMSCs against LIRI and investigated the underlying mechanisms. Western blotting was performed to examine Lin28 expression in PMSCs under hypoxic conditions. A Lin28 overexpression construct was introduced into PMSCs, and the effect on glucose metabolism was examined using a glucose metabolism kit. Further, the expression of some proteins involved in glucose metabolism and the PI3K‐AKT pathway and the levels of microRNA Let‐7a–g were examined using western blots and real‐time quantitative PCR, respectively. To examine the relationship between Lin28 and the PI3K‐Akt pathway, the effects of AKT inhibitor treatment on the changes induced by Lin28 overexpression were examined. Subsequently, AML12 cells were co‐cultured with PMSCs to elucidate the mechanisms via which PMSCs prevent hypoxic injury in liver cells in vitro. Finally, C57BL/6J mice were used to establish a partial warm ischaemia–reperfusion model. The mice received intravenous injections containing PMSCs (control and Lin28‐overexpressing PMSCs). Finally, their serum transaminase levels and degree of liver injury were assessed using biochemical and histopathological methods, respectively. Lin28 was upregulated under hypoxic conditions in PMSCs. Lin28 exerted protective effects against hypoxia‐induced cell proliferation. Moreover, it increased the glycolytic capacity of PMSCs, allowing PMSCs to produce more energy under hypoxic conditions. Lin28 also activated the PI3K‐Akt signalling pathway under hypoxic conditions, and its effects were attenuated by AKT inhibition. Lin28 overexpression could protect cells against LIRI‐induced liver damage, inflammation and apoptosis and could also attenuate hypoxia‐induced hepatocyte injury. Lin28 enhances glucose metabolism under hypoxic conditions in PMSCs, thereby exerting protective effects against LIRI by activating the PI3K‐Akt signalling pathway. Our study is the first to report the potential of genetically modified PMSCs for LIRI treatment.


| BACKG ROU N D
Liver ischaemia-reperfusion injury (LIRI) is an important clinical complication occurring in patients undergoing liver surgery (e.g. transplantation and hepatectomy) and in cases of shock. 1 LIRI can cause primary graft dysfunction after transplantation, liver dysfunction after hepatectomy and biliary tract injury. 2,3 During liver surgery, hepatic blood vessels are often blocked, leading to ischaemia and hepatocyte death. 4 In addition, once blood supply is restored, free radicals attack healthy hepatocytes. Together, these events lead to necrosis and apoptosis, causing inflammatory cell infiltration and the aggregation of inflammatory mediators, resulting in further liver damage. 5,6 Studies have shown that the damage caused by reperfusion is far greater than that caused by ischaemia. 7 Given their clinical impact, the mechanisms underlying LIRI have been studied extensively.
Surgical strategies that reduce the duration of liver ischaemia, clinical approaches that provide intraoperative and postoperative liver protection, and immunosuppressive drugs that prevent liver damage have helped in decreasing the incidence of LIRI. [8][9][10] Mesenchymal stem cells (MSCs) are present at multiple sites in the body, including the bone marrow, cord blood and umbilical cord and placental tissue.
In adults, these cells represent non-haematopoietic stem cells with multidirectional potential. MSCs are mainly used during bone marrow transplantation to prevent graft versus host disease, improve tissue repair and provide anti-inflammatory and immunomodulatory effects. [11][12][13][14] Placental MSCs (PMSCs) show better immunogenicity and immune regulation than other types of MSCs, because they act as negative immune regulators and can inhibit the body's immune response 15 . These cells can restore the immune balance and inhibit the inflammation associated with injury in several organs. Therefore, these cells are used for kidney, brain, heart and liver treatment. [16][17][18] However, the widespread clinical application of MSCs is hindered due to some challenges. These include the short survival duration of MSCs, number of organs to be protected, and injury to MSCs during ischaemia and reperfusion. 19 Further, the molecular regulation of MSC survival under hypoxic conditions is unclear.
Cancerous cells typically gain energy from glycolysis instead of aerobic respiration. Owing to this switch, called the Warburg effect, cancer cells show unique mitochondrial utilization. 20 Lin28 is an RNA-binding protein required for early embryonic development, stem cell differentiation/reprogramming, tumorigenesis and metabolism. [21][22][23] Studies have shown that Lin28 protein can protect various organs from ischaemia-reperfusion injury. 24,25 Studies have shown that the Lin28/Let-7 pathway can regulate mammalian glucose metabolism and is itself intricately controlled. 26 Another important intracellular pathway is the PI3K-Akt signalling pathway, which regulates processes such as energy metabolism, growth and development and autophagy. 27,28 Studies have shown that this pathway is activated under anoxic conditions and enables the switch from aerobic respiration to glycolysis, allowing cells to produce the required energy. 29 However, whether Lin28 can enhance metabolism in human PMSCs during hypoxia and protect hepatocytes against LIRI is unclear. Moreover, the role of the PI3-Akt pathway in this process remains to be elucidated.
In the present study, a partial warm ischaemia-reperfusion injury mouse model and hypoxia-treated hepatocyte model were used to examine whether Lin28 can increase the glycolytic capacity of PMSCs under hypoxic conditions and protect liver cells against LIRI.
Further, the involvement of PI3-Akt signalling in the effects of Lin28 on PMSC glycolysis was also explored.

| Partial LIRI model
Male C57BL/6J mice (20-22 g) were randomly assigned to one of four groups: Sham; LIRI; LIRI + PMSC; and LIRI + PMSC-Lin28. A partial LIRI model was developed using methods described in an earlier study. 30 The mice in the sham group underwent the same operative procedure, without any blood vessel blocking. The LIRI + PMSC and LIRI + PMSC-Lin28 groups received injections with PMSCs or PMSCs overexpressing Lin28 (PMSCs-Lin28). The cells were injected into the portal vein (100 μL, 10 7 cells/mL) 1 h prior to the induction of hepatic ischaemia. After 1 h of ischaemia and 6 h of reperfusion, mice were sacrificed, and serum and liver tissue samples were collected. effects against LIRI by activating the PI3K-Akt signalling pathway. Our study is the first to report the potential of genetically modified PMSCs for LIRI treatment.

K E Y W O R D S
glucose metabolism, human placental mesenchymal stem cells, lactate dehydrogenase a, liver ischaemia-reperfusion injury, PI3K-Akt pathway

| Serum biochemistry assay
Blood collected from mice was first chilled on ice (30 min) and then centrifuged at 8000 RPM (15 min). The supernatant was diluted as appropriate and a standard automatic analyser (Mindray BS-200) was used to detect serum alanine transaminase (ALT) and aspartate transaminase (AST) levels. containing 5% CO 2 , 1% O 2 and 94% N 2 was used for maintaining an anoxic environment. To establish the ischemic and hypoxic cellular models, cells were cultured to a density of approximately 70%.

| Cell culture and hepatocyte hypoxia model
Subsequently, the old medium was removed, and serum-free medium was added; cells were incubated in anoxic conditions for 24 h.

| Overexpression of Lin28
When cells reached a density of approximately 70%, they were cultured in OPTI-MEM Reduced Serum Medium (Gibco) with a Lin28 overexpression lentivirus (10 μL/mL; with green fluorescence; obtained from Hanbio Biotechnology Co. Ltd.) for 72 h.
Subsequently, fluorescence microscopy was used to detect transfection efficiency.

| Cell viability assay
The CCK-8 kit (Dojindo) was used to detect cell viability. Before hypoxia treatment, 10,000 cells were planted in each well of a 96-well plate, and three holes were made in each well. After 24 h of culture, the medium was replaced and 10 μL of the CCK-8 reagent was added to each well (incubation, 37°C, 2 h). Finally, a microplate reader was used to measure the absorbance at 450 nm.

| Flow cytometry analysis
An Annexin V-FITC and propidium iodide staining kit (MULTISCIENCES) was used to examine cell apoptosis using manufacturer's instructions. Using a Flow Cytometer (BD FACSCelesta) and FlowJo software for data analysis, we measured the percentages of apoptotic cells.

| qRT-PCR
The Total RNA Rapid Extraction Kit (Fastagen) was used to isolate the total mRNA from PMSCs. The mRNA was then reverse transcribed using a high-capacity cDNA reverse transcription kit (TAKARA) based on manufacturer's instructions. Target gene expression was evaluated with qRT-PCR using a SYBR Green kit (TAKARA). StepOne software (Thermo Fisher Scientific) was used to analyse the data. For examining microRNA levels, the high-capacity microRNA reverse transcription kit (TAKARA) was used; the other steps remained the same as those for mRNA analysis. GADPH and U6 were chosen as internal controls for mRNA and microRNA, respectively. Each experiment was performed at least in triplicate. Primer sequences are shown in the table below (Table 1). Relative gene expression was calculated using the 2 −ΔΔC T method.

| Statistical analysis
All statistical analyses were performed using Prism software (GraphPad, v8.0.2). All data were expressed as the mean ± standard error of the mean. One-way anova was used to compare multiples groups, whereas unpaired t-tests were used to compare two groups.
p-values < 0.05 were considered statistically significant.

| Lin28 affects glucose metabolism levels in PMSCs
A PMSC hypoxia model was used to examine the expression of Lin28 in PMSCs under hypoxic conditions. We found that hypoxia obviously increased Lin28 expression in PMSCs ( Figure 1A). The Lin28 overexpression lentivirus showed a 76.36 ± 2.002% transfection efficiency in PMSCs ( Figure 1B,C). On examining several biochemical indicators of glycolysis, we found that PMSCs transfected with Lin28 had a higher LA level, NADPH/NADP+ ratio and ATP content than non-transfected cells. These cells also had lower intracellular glucose levels. Although these differences were also observed in normoxic environments, they were more obvious in anoxic environments ( Figure 1D). Consistent with these findings, we observed an increase in the basal glucose metabolism and glycolytic capacity of PMSCs after Lin28 overexpression, indicating that Lin28 promoted glycolysis in PMSCs ( Figure 1E,F). A CCK-8 assay demonstrated that the proliferation of PMSCs was significantly reduced under hypoxia.
Moreover, the proliferation of Lin28-PMSCs was also significantly reduced ( Figure 1G). These results indicated that the overexpression of Lin28 under hypoxic conditions could promote anaerobic glycolysis in PMSCs, allowing them to produce more energy under hypoxia and maintain cellular function.

| Lin28 can regulate proteins related to glucose metabolism in PMSCs
In order to understand the pathways through which Lin28 in- showed no changes in expression ( Figure 2B). Therefore, we speculated that Lin28 may regulate glucose metabolism in PMSCs via LDHA.

| Lin28 regulates LDHA via the Let7-PI3K-Akt pathway
Lin28 can downregulate microRNAs from the Let7 family. Using qRT-PCR, we found that the levels of microRNA Let7a, Let7b and Let7c were dramatically lower in the PMSC-Lin28 group than in the PMSC group. However, microRNA Let7d, Let7e, Let7f and Let7g showed no significant difference in expression ( Figure 3A).
Previous studies have confirmed that a reduction in Let7c levels leads to the activation of PI3K and the PI3K-Akt pathway, altering the expression of LDHA. Thus, the mRNA levels of PI3K and LDHA were quantified in PMSCs and PMSCs-Lin28 under normoxic and hypoxic conditions using qRT-PCR. Moreover, the protein expression of AKT, pAKT (ser473), Lin28, and LDHA was detected using western blots. The transcriptional levels of PI3K and LDHA were higher in the PMSCs-Lin28 group than in the PMSCs group. These levels were also higher under hypoxic conditions than under normoxic conditions ( Figure 3B). The protein levels of pAKT (ser473) and LDHA showed similar trends. However, the protein expression of AKT was comparable among all groups ( Figure 3C). Based on these results, we speculated that Lin28 downregulates Let7c by interacting with the microRNA. This results in PI3K upregulation, activating the PI3K-Akt pathway and promoting AKT phosphorylation. Such changes lead to increased LDHA expression, promoting the switch from aerobic oxidation to anaerobic glycolysis in PMSCs.

| Inhibitors of the PI3K-Akt pathway weaken the effect of Lin28 on glucose metabolism
In order to verify that Lin28 regulates glucose metabolism via the PI3K-Akt pathway, we treated PMSCs with the AKT phosphorylation inhibitor MK2206 in vitro. After treatment with MK2206, there was a decrease in intracellular LA levels, the NADPH/ NADP+ ratio, and the ATP content in Lin28-overexpressing PMSCs. Moreover, the intracellular glucose content increased significantly. Interestingly, these effects were observed under both normoxic and anoxic conditions ( Figure 4A,B). On examining protein expression, we found that cells from the PMSC + MK2206 and PMSC-Lin28 + MK2206 groups had lower levels of pAKT (ser473) and LDHA than those from the PMSC and PMSC-Lin28 groups. In contrast, the levels of AKT and Lin28 were similar between the PMSC-Lin28 + MK2206 and PMSC-Lin28 groups and the PMSC + MK2206 and PMSC groups ( Figure 4C). Furthermore, ECAR analysis showed that MK2206 treatment reduced basal glucose metabolism and glycolysis in PMSCs overexpressing Lin28 ( Figure 4D,E). Hence, we concluded that Lin28 enhances the viability of PMSCs in anoxic environments by activating the PI3K-Akt pathway.

| Lin28 overexpression in PMSCs enhances their protective effects against ischaemia-reperfusion-induced apoptosis in AML12 cells
After 24 h of anoxic culture, as mentioned above, the cells were cultured in a normoxic incubator containing 5% CO 2 for 1 h. The cells were then harvested and analysed using flow cytometry. The apoptosis rate was found to be markedly lower in the AML12 + PMSC and AML12 + PMSC-Lin28 groups than in the AML12 group.
Furthermore, this value was much lower in the AML12 + PMSC-Lin28 group than in the AML12 + PMSC group ( Figure 5A,B). This indicated that Lin28 enhances the protective effects of PMSCs against apoptosis in the AML12 hypoxia model.

| Lin28 overexpression in PMSCs enhances their protective effects against LIRI in mice
We examined the effects of treatment with Lin28-overexpressing PMSCs in a mouse model of LIRI using biochemical and histological

| DISCUSS ION
Orthotopic liver transplantation is now emerging as the frontrunner for the treatment of end-stage liver disease. However, LIRI remains a key and common cause of postoperative hepatic impairments, (E) Basal glucose metabolism levels and the glycolysis ability of PMSCs and PMSCs-Lin28 were calculated using ECAR, and significant differences were observed. ns, p > 0.05; *p < 0.01; **p < 0.01;***p < 0.001; ****p < 0.0001.
inflammatory infiltration, and apoptosis. More importantly, its severity is tightly linked to postoperative recovery. 32,33 Many studies have focused on the treatment of LIRI. 34,35 MSCs are considered to be effective in reducing LIRI and consequently protecting liver function. [36][37][38] The findings of the present study also prove their effectiveness. However, the mechanism by which MSCs protect the body against LIRI is unclear. Currently, it is hypothesized that these effects are related to the inhibition of mitochondrial reactive oxygen species overproduction. The improvement of mitochondrial function, 39 regulation of miRNA, 40 and regulation of immune cells 41 by extracellular vesicles are also thought to play a role. Changes in macrophage activation can prevent inflammation. 42 PMSCs, which express higher levels of several immunomodulatory and pro-angiogenic cytokines than other MSCs, are widely used for the prevention of ischaemia-reperfusion injury. 43 However, during liver ischaemia, MSCs are themselves exposed to ischaemia and hypoxia. Currently, the effect of these conditions on the cellular functions of MSCs and the mechanisms underlying these effects are unclear.
In the present study, we first constructed a hypoxia model in PMSCs and observed an increase in Lin28 protein levels. Although Lin28 is known to activate the PI3K-Akt pathway, 53 the mechanism underlying this activation is not well-understood ( Figure 6). We found that the promoter region of the PIK3CA gene (which encodes the PI3K α subunit) has a region that can bind to microRNAs from the Let7 family. We speculated that members of the microRNA Let7 family bind to the promoter region of PIK3CA and inhibit its transcription. This decreases the expression of PI3K, consistent with our previous results. We aim to explore these mechanisms in subsequent studies.

| CON CLUS IONS
In conclusion, our study shows that Lin28 can improve the glyco- formal analysis (equal); funding acquisition (equal); investigation (equal); writing -review and editing (equal).

ACK N O WLE D G E M ENTS
Thanks to Professor Lan Peixiang for his constructive suggestions on experimental design, and thanks to MJEditor (www.mjedi tor.com) for its linguistic assistance during the preparation of this manuscript.
A preprint has previously been published. 54

FU N D I N G I N FO R M ATI O N
This work was funded by the National Natural Science Foundation of China (81800580 and 81770652).

CO N FLI C T O F I NTER E S T S TATEM ENT
The authors have declared that no conflict of interest exists.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data included in this study are available upon request by contact with the corresponding author.