| BackgroundChronic hepatitis B virus(HBV)infection remains a public health issue of great concern worldwide.Approximately 240 million people with persistent HBV infection are at high risk of developing HBV-related liver cirrhosis and hepatocellular carcinoma(HCC).HBsAg loss is a desirable clinical target in patients with CHB and serves as a surrogate indicator for clinical outcomes.However,both spontaneous and antiviral treatment-induced HBsAg seroclearance rarely occur.Hence,it is necessary to gain further insights into the underlying factors involved in the inability to achieve HBsAg seroclearance during chronic HBV infection.Recently,cumulative data from experimental and clinical studies have highlighted the critical role of the metabolic milieu in regulating host immunity.Immune response requires distinct metabolic programs to support immune cell survival,development,differentiation,fate,and behavior.Metabolic reprogramming of tumors or infections was confirmed to be associated with down-regulation of effector T cell function.L-carnitine(L-Cn,3-hydroxy-4-N-trimethylaminobutyrate)plays an essential role in transporting activated long-chain fatty acids across mitochondrial membranes for β oxidation,requiring enzymes and transporters that accumulate L-Cn within the cell(OCTN2 L-Cn transporter),conjugate it with long chain fatty acids(carnitine palmitoyl transferase 1,CPT1),transfer the acylcarnitine across the inner plasma membrane(camitine-acylcarnitine translocase,CACT),and conjugate the fatty acid back to Coenzyme A for subsequent β oxidation(carnitine palmitoyl transferase 2,CPT2).The main factor regulating L-Cn body pools is OCTN2 which is a high affinity plasma-membrane sodium-dependent L-Cn transporter,containing 12 transmembrane segments that transfer L-Cn across cell membranes.Hepatocytes in the liver have a different low-affinity,high-capacity OCTN2.In individuals consuming a regular diet about 75%of L-Cn comes from diet and only 25%of it comes from endogenous synthesis from the amino acids lysine and methionine in liver,kidney and brain.One month’s red meat diet consumption had been shown a significant increase in plasma carnitine levels,relative to either non-meat or white meat diets.Previous studies have shown that lipid metabolism plays a key role in T cell differentiation and function.It has been reported that the serum L-Cn levels were stepwise altered in HBV-cirrhosis and HCC,and a previous study has documented an association between lower baseline levels of plasma L-Cn and HBsAg loss in chronic hepatitis B(CHB)patients receiving Peg-IFN-α based treatment.Previous studies have found that L-Cn displays an immunosuppressive effect on immune cells,but its effect on the immune system in the setting of HBV infection is poorly understood.The pathogenesis and outcome of HBV infection mainly depend on the interaction between the virus(or antigen)and host immunity.Germinal centers(GC)are histologically distinct structures that develop within B cell zones(follicles)of secondary lymphoid tissues,and it is within GC that the interrelated and multifaceted processes of B cell affinity maturation,class switch recombination,plasma cell differentiation,and memory B cell differentiation predominantly occur.Publications from our group have documented the favorable effect of circulating C XCR5+CD4+T cells and CXCR5+CD8+T cells on achieving hepatitis B e antigen(HBeAg)seroconversion and HBsAg clearance in chronic HBV infection.Based on the available data,we hypothesized that excess L-Cn might potently disturb effective antiviral GC-related immune responses and hamper the achievement of HBsAg clearance in chronic HBV infection.ObjectiveCross sectional cohort of chronic HBV infection in natural history and longitudinal cohort of CHB patients with antiviral treatment were enrolled in this study.We aimed to evaluate L-Cn status in patients with chronic HBV infection,investigate the relationship of L-Cn with HBsAg clearance,and elucidate the immunomodulatory effect of L-Cn on GC-related immune cells as well as the GC formation by phenotypic and functional analyses.In addition,C57BL/6 male mice were injected hydrodynamically with the pAAV-HBV 1.2 plasmids via the tail vein to further investigate the influence of L-Cn in vivo.Factors that affect plasma L-Cn levels were also primarily explored.This study opens novel avenues to broaden our understanding of immunometabolic reprogramming of the host immune system via L-Cn manipulation and thus elicit an effective immune response against HBV.MethodsNinety-eight treatment-na?ve patients with chronic HBV infection,including 34 with HBeAg-negative chronic infection,9 with HBeAg-negative chronic hepatitis,25 with HBeAg-positive chronic infection,and 30 with HBeAg-positive chronic hepatitis,were recruited from real-life clinical practice.Twenty-four patients who achieved HBsAg loss following antiviral treatment,30 patients who suffered from HBV-related HCC,38 patients who underwent liver biopsy,and 8 patients who had elevated serum ALT levels(more than 2 times the upper limit of normal)after cessation of antiviral treatment were enrolled.Another 22 HBsAg-negative healthy individuals with normal ALT levels were recruited as controls.Additionally,HBeAg-positive CHB patients who participated in a clinical trial of telbivudine(trial number:CLDT600ACN07T)or Peg-IFNα-2a(trial number:NCT01086085)were longitudinally studied.Eight telbivudine-treated subjects were defined as the complete response(CR)group(with normal ALT,HBeAg seroconversion,and an HBV DNA level<1000 copies/mL)at week 48,and the other subjects were classified as the noncomplete response(NCR,n=10)group,Nine Peg-IFNα-2a-treated subjects were defined as the CR group according to the treatment outcomes at week 72,while the other 23 patients were defined as the NCR group.Additionally,648 HBeAg-positive CHB patients who treated with Peg-IFN-αat the dose of 180 μg/week for 48 weeks and a treatment-free follow-up for 24 weeks(Peg-IFN-α-2a,n=324;Peg-IFN-α-2b,n=324)were used to evaluate the association of single-nucleotide polymorphism(SNP)rs1799821 genotype with complete response(defined as HBeAg seroconversion and an HBV DNA level<2000 IU/mL at week 72).To evaluate the influence of diet intervention on plasma L-Cn levels,eight healthy volunteers were recruited and the dynamic L-Cn levels were recorded.Plasma L-Cn levels were measured by enzyme-linked immunosorbent assay.In addition,HepG2.2.15 cells were lysed and mouse model with acute liver injury was established by intraperitoneal injection of acetaminophen(APAP)to investigate the origin of L-Cn,and the levels of L-Cn in supernatant or serum were detected.RNA sequencing was conducted to define the transcriptional profiles of peripheral blood mononuclear cells after L-Cn stimulation and we analyzed the microarray data from the Gene Expression Omnibus series GSE62595.In vitro assays were performed to assess the effect of L-Cn on distinct lymphocytes;the frequencies and function of lymphocytes were analyzed by flow cytometry.C57BL/6 male mice were injected hydrodynamically with the pAAV-HBV1.2 plasmids via the tail vein,and L-Cn was sequentially administered.Serum HBsAg,HBeAg,and HBV DNA were measured at the indicated time points.Paraffin-embedded sections of spleen tissues were used to examine the effect of L-Cn on GC formation.Receiver-operating characteristic(ROC)curve was performed to evaluate the ability of plasma L-Cn levels at baseline to predict favorable treatment response to telbivudine or Peg-IFNα-2a therapy.Factors that affect plasma L-Cn levels were also explored by diet intervention and genotyping some SNPs that were reported to be associated with L-Cn levels.Results1.The characteristics of L-Cn in HBV infection1.1 A higher L-Cn level was observed in patients with chronic HBV infection or HCC,relative to healthy controls;moreover,L-Cn levels of patients with chronic HBV infection or HCC were higher than those who achieved HBsAg loss.The plasma level of L-Cn in the HBeAg+CHep group who had flared ALT was significantly elevated than the HBeAg-CInf,HCs,and HBsAg loss groups;1.2 Compared with patients with HBsAg loss or those with HBsAg levels<100 IU/mL,patients with HBsAg levels≥100 IU/mL have significantly elevated plasma levels of L-Cn;1.3 A positive correlation was observed between plasma L-Cn levels and HBsAg levels as well as HBeAg and HBV DNA levels.2.Damaged hepatocyte-derived L-Cn is the main cause of plasma L-Cn elevation during chronic HBV infection2.1 A positive correlation between plasma levels of L-Cn and serum levels of ALT and AST was found;2.2 CHB patients with moderate-to-severe liver necroinflammation showed significantly higher levels of plasma L-Cn than those without or with only mild necroinflammation;2.3 The levels of L-Cn in supernatant from physically lysed HepG2.2.15 cells showed an increasing trend;2.4 The results from a mouse model of acetaminophen(APAP)-induced acute liver injury showed that a transient sharp increase of serum L-Cn was detected 2 hours post-injection(hpi)of APAP,while the expression of L-Cn decreased subsequently and reached a level comparable to that measured in control mice at 4,8,12,and 24 hpi,accompanied by an ALT boost.3.The effect of L-Cn on GC-related immune cells in chronic HBV infection3.1 RNA sequencing was performed to define the transcriptional profiles of peripheral blood mononuclear cells(PBMCs)after L-Cn stimulation.GSEA confirmed immune response dysfunction as a significant abnormality,especially the B cell and T cell receptor signaling;3.2 L-Cn markedly inhibited the proliferation of CXCR5+CD4+T cells and CXCR5+CD8+T cells,and L-Cn-mediated inhibition of CD19+B cells showed a dose-dependent trend.Besides,exposure to L-Cn led to decreased production of IFN-γ by CXCR5+CD4+T cells and CXCR5+CD8+ T cells,while a remarkable decrease in IL-21 was only observed in CXCR5+CD4+T cells.Strikingly,ELISpot assay showed that L-Cn treatment led to significantly decreased proportions of HBsAb-,HBeAb-,and HBcAb-secreting B cells.In contrast,L-Cn upregulated the proportion of FoxP3+CXCR5+CD4+regulatory T cells and the production of IL-10 by such cells;3.3 We found the elevated expression of PD1 and CD160 in CXCR5+CD4+T cells;meanwhile,CXCR5+CD8+T cells showed elevated expression of canonical inhibitory receptors,including PD1,CTLA4,Tim3,CD160,and LAG3,and reduced expression of the transcription factors TCF1 and Eomes after L-Cn treatment.Although L-Cn reduced expression of CD22,CD32,FCRL5,and PD1 on CD19+B cells,an enhanced expression of BTLA was observed(Figure 3A,top right).FoxP3+CXCR5+CD4+regulatory T cells showed elevated expression of TIGIT but decreased expression of CD25,CTLA4,ICOS,PD1,and Tim3 after L-Cn stimulation.4.L-Cn hinders HBsAg clearance and GC formation in a mouse model with HBV expression4.1 We found that the serum HBsAg levels at 7,14,21,and 28 days post-injection(dpi)were significantly higher in L-Cn-treated mice than in control mice and L-Cn-treated mice showed significantly lower levels of serum HBsAb than control mice at 28 dpi;4.2 Paraffin-embedded sections of spleen tissues were stained with hematoxylin-eosin,CD3,and B220 respectively to examine the effect of L-Cn on GC formation.Data showed that L-Cn-treated mice had fewer GC than control mice at 35 dpi.5.Diet intervention increases plasma L-Cn levels and L-Cn levels could predict favorable antiviral treatment response5.1 Longitudinal analysis in patients receiving telbivudine or Peg-IFNα-2a treatment showed that decreased plasma L-Cn levels were detected after 48 weeks of therapy.Intriguingly,plasma L-Cn levels of the complete response(CR)patients at baseline were significantly lower than those with the noncomplete response(NCR)in the two cohorts.Furthermore,ROC analysis showed the ability of plasma L-Cn levels at baseline to predict favorable treatment response to telbivudine or Peg-IFNα-2a therapy;5.2 After challenged with 240 g cooked-beef(a major source of dietary L-Cn,equivalent to an estimated 180 mg L-Cn),two HCs showed gentle ripples of plasma L-Cn levels over time.Intriguingly,we observed an increase in plasma L-Cn levels following a one-week intervention with beef-rich diets(equivalent to an estimated 330 mg L-Cn per day);5.3 Factors that affect plasma L-Cn levels:for SNP rs 179982(gene CPT2),patients with GG genotype had lower plasma L-Cn levels than those with AA genotype;plasma L-Cn levels in HCs with high BMI(>25 kg/m2)were lower than those with low BMI(<18 kg/m2).Conclusions1.L-Cn levels of patients with chronic HBV infection or HBV-HCC were higher than those who achieved HBsAg loss;moreover,the plasma level of L-Cn in the HBeAg+CHep group who had flared ALT was significantly elevated than the HBeAg-CInf,healthy controls,and HBsAg loss groups;2.The levels of L-Cn in supernatant from physically lysed HepG2.2.15 cells showed an increasing trend.The results from a mouse model of APAP-induced acute liver injury showed that a transient sharp increase of serum L-Cn,accompanied by an ALT boost,suggesting that L-Cn may be derived from damaged hepatocytes during liver injury;3.RNA sequencing analysis demonstrated that L-Cn altered the transcriptional profiles related to immune response.L-Cn markedly inhibited the proliferation of CXCR5+CD4+T cells,CXCR5+CD8+T cells and B cells.Besides,exposure to L-Cn led to decreased production of IFN-γ by CXCR5+CD4+T cells and CXCR5+CD8+T cells,while a remarkable decrease in IL-21 was only observed in CXCR5+CD4+T cells.Strikingly,L-Cn treatment led to significantly decreased proportions of HBsAb-,HBeAb-,and HBcAb-secreting B cells.In contrast,L-Cn upregulated the proportion of FoxP3+CXCR5+CD4+regulatory T cells and the production of IL-10 by such cells.these data show that L-Cn impedes HBV-specific immune response of GC-related immune cells,promotes the expression of inhibitory receptors on GC-related immune cells,and enhances immunosuppressive profiles;4.L-Cn hinders HBsAg clearance and GC formation in a mouse model with HBV expression,indicating that L-Cn might prevent GC formation,thus impeding HBsAg clearance in the setting of HBV infection;5.ROC analysis showed the ability of plasma L-Cn levels at baseline to predict favorable treatment response to telbivudine or Peg-IFNα-2a therapy.Short-term L-Cn challenge test by beef provocation did not cause plasma L-Cn levels fluctuation,which suggested L-Cn as a potential stable plasma biochemical marker.An increase of plasma L-Cn levels was observed following one week-long L-Cn challenge test.It is reasonable to speculate that shifting diet habits to create a low plasma L-Cn condition may be beneficial for the clinical outcome of chronic HBV infection,improving the antivirus immune response. |