【作者】 蒋小滔；

【导师】 侯金林；

【作者基本信息】 南方医科大学 ， 内科学， 2012， 博士

【摘要】 背景：乙型肝炎病毒(hepatitis B virus,HBV)感染可引起一系列肝脏疾病,包括无症状携带、慢性乙型肝炎(chronic hepatitis B,CHB)、肝硬化和肝癌。全世界有超过3亿人为慢性HBV感染,其中约三分之一在中国。在我国,每年因HBV感染相关肝病死亡的患者达30万人。目前尚无能彻底清除HBV的理想药物,因此HBV感染已经成为严重影响我国人民健康的公共卫生问题。HBV感染后肝细胞的损伤并不是由HBV的入侵、复制直接造成的,而主要是由宿主的免疫系统识别病毒感染细胞后引起的免疫杀伤。不仅如此,机体的免疫功能在很大程度上还决定了抗病毒治疗的效果。对抗病毒治疗无应答者,机体抗HBV免疫应答能力持续低下；而完全应答者其抗病毒免疫应答能力显著升高,即机体的免疫系统在HBV慢性感染的病程和转归中发挥重要作用。机体的免疫可分为天然免疫(亦称固有免疫或非特异性免疫)和适应性免疫(亦称获得性免疫或特异性免疫)两类。天然免疫是机体抵御感染的第一道防线,由物理、化学屏障(如皮肤、粘膜、胃酸等),吞噬细胞(巨噬细胞和中性粒细胞),体液因子(补体、溶菌酶、细胞因子等),NK细胞和NKT细胞等组成。天然免疫能非特异性识别并清除入侵的病原体,并在适应性免疫的启动、效应阶段发挥重要作用,还参与免疫应答的调节。在生理性免疫应答和某些病理性免疫过程中,天然免疫都处于关键环节。Invariant NKT (invariant NKT)细胞是天然免疫的重要组成细胞,最早于1987年在小鼠体内发现。invariant NKT细胞与传统T细胞的主要区别在于：①invariant NKT细胞的T细胞受体(TCR)比较保守,多样性高度受限。在小鼠主要为Vα14-Ja18/Vp8.2, Vβ7和Vβ2,在人主要为Va24-Jα18/Vp11。②与传统T细胞主要识别MHC分子递呈的抗原肽不同,invariant NKT细胞识别的是MHC I类样分子CD1d递呈的脂类或糖脂类抗原。③nvariant NKT细胞活化后能迅速(数小时内)分泌大量的Th1和Th2类细胞因子。invariant NKT细胞在外周血、淋巴结中比例很低,约占淋巴细胞的0.5%,但其在肝内却异常丰富,占肝内淋巴细胞的比例在小鼠高达40%,在人体也在12%左右。结构决定功能,肝内如此丰富的invariant NKT细胞在HBV慢性感染过程中发挥何种作用?HBV慢性感染是否对invariant NKT细胞的功能有影响?这种影响有何临床意义?在抗病毒治疗过程中,invariant NKT细胞的功能有何变化?对优化治疗方案有何意义?对于这些问题,目前的研究还未能给出清晰的答案。鉴于此,本研究比较了慢性HBV感染后不同疾病状态下患者外周血invariant NKT细胞的比例、表型及功能的变化；分析其与病毒复制和肝脏炎症的相关性；观察抗病毒治疗后这群细胞的变化规律；通过体外实验揭示invariant NKT细胞参与慢性HBV感染的潜在机制。方法：1.研究对象本研究共纳入35例CHB患者,其中一部分来自南方医院感染内科住院患者,一部分来自参加替比夫定抗病毒治疗的临床试验患者；15例免疫耐受期的HBV携带者(Immunotolerant Carrier,IT)和25例非活动性HBsAg携带者(Inactive Carrier, IC),主要为2009年12月至2011年4月南方医院感染内科门诊病人以及义诊患者；以及36例健康成人(Healthy Control, HC),主要为南方医院肝病中心实验室工作人员、在读研究生和南方医科大学学生2.血清学和病毒学检测用商品化的AxSYM MEI试剂盒(Abbott Laboratories, North Chicago, IL)检测乙肝两对半、抗-HCV和抗HDV。用罗氏Cobas Taqman48(Meylan, France)定量检测血浆中的HBV DNA,其最低检测下限为300拷贝/ml。此部分检测由南方医院肝病中心临床检测室完成。3.外周血单个核细胞(PBMC)的分离每例研究对象在入组时,用肝素钠抗凝管采集静脉血10-30ml。19例接受替比夫定抗病毒治疗的CHB患者在治疗后的第12周、24周和52周再另外抽取10-30ml静脉血。用Ficoll-Hypaque密度梯度离心法分离PBMC。4.流式细胞术(Flow Cytometry,FCM)用无菌PBS调整PBMC到合适浓度,转移至无菌流式管。加入荧光标记抗体,用FCM检测invariant NKT细胞的频数和相关受体的表达。在某些情况下,先用PMA对PBMC进行刺激,然后收集细胞,用胞内细胞因子染色检测invariantNKT细胞及其他细胞分泌细胞因子的情况。5.细胞迁移实验用Trans well系统检测趋化因子5(CCL5)和趋化因子20(CCL20)对invariant NKT细胞的趋化作用。将新鲜分离的PBMC置于上室,下室加入一定浓度的CCL5或CCL20,37℃培养10小时。收集上、下室的细胞,用FCM检测invariant NKT细胞的比。6.invariant NKT细胞体外增殖实验将新鲜分离的PBMC用RPMI1640完全培养基调整浓度至1×106个/ml,加入到96孔细胞培养板,200μl/孔。加入a-GalCer (invariant NKT细胞的特异性刺激剂)和重组人IL-21(rhIL-21),37℃细胞培养箱培养7天。收集细胞,用FCM检测invariant NKT细胞的增值情况。7. ELISPOT将新鲜分离的PBMC置于预先包被有鼠抗人IFN-γ的ELISPOT板,加入a-GalCer口rhIL-21,37℃细胞培养箱培养24小时。弃细胞,然后加入生物素标记的另一鼠抗人IFN-γ抗体,反应完成后再加入碱性磷酸酶标记的亲和素。加入底物显色,观察、记录结果。在某些情况下,用B细胞ELISPOT检测invariant NKT细胞活化对HBV特异性抗体产生的影响。具体步骤如下：ELISPOT板预包被重组HBeAg或HBcAg,然后加入PBMC,用a-GalCer刺激60小时。弃细胞,加入生物素标记的羊抗人IgG,37℃孵育1小时。洗涤后加入亲和素-碱性磷酸酶,加底物显色。8.肝内invariant NKT细胞的检测用无菌手术剪将新鲜切除的肝组织(来源于肝脏手术病人)剪成小块,加入组织消化液室温消化1小时。滤膜过滤,然后用人淋巴细胞分离液分离肝内淋巴细胞,FCM检测肝内invariant NKT细胞的比例。9.免疫组织化学染色将肝穿组织或手术切除肝组织经洗涤、脱水、透明、浸蜡后制成石蜡切片。用免疫组织化学染色检测肝内invariant NKT细胞的频数及分布。10.统计分析用GraphPad Prism5软件对实验结果进行统计分析。不同组间的结果比较用Mann-Whitney U检验。同一研究对象不同时间或剂量的比较用重复测量的方差分析。计数资料的比较用Fisher’s exact test。相关分析采用Spearman秩相关。用受试者工作特征曲线(ROC曲线)比较不同参数对替比夫定抗病毒治疗后HBeAg血清学转换的预测价值。双侧检验、P<0.05为统计学差异有显著性。结果：1.CHB患者外周血invariant NKT细胞比例显著下降我们首先比较、分析了invariant NKT细胞在CHB患者、HBV携带者以及健康人外周血中占CD3+T细胞的比例。结果显示,在健康成人外周血中,invariant NKT细胞占CD3+T细胞的比例中位数(最小值,最大值)为0.24%(0.064%-0.81%)。HBV感染后,在HBV特异性免疫还没起来的IT期,invariantNKT细胞比例升高至0.39%(0.203%-0.96%),显著高于健康对照(P<0.01)。在CHB患者,其HBV特异性免疫已被激发,外周invariant NKT细胞的比例下降至0.129%(0.027%-1.071%),显著低于IT(P<0.001),甚至比健康对照还要低(P=0.0143)。而在HBV受到免疫控制的IC期,invariant NKT细胞的比例得到一定恢复,为0.19%(0.06%-1.38%),显著高于CHB患者(P=0.0235),虽然比健康人略低,但没有统计学意义,也要显著低于IT(P<0.001)。2.IT期CD4-invariant NKT细胞的比例与病毒载量呈正相关我们对30例CHB患者进行Spearman秩相关分析,结果显示,invariant NKT细胞占CD3+T细胞的比例与血浆中的病毒滴度和ALT水平并没有明显的相关性。进一步将invariant NKT细胞分为CD4-和CD4+两个亚群分析,在IT期,HBV DNA滴度与CD4-NKT细胞的比例呈明显的正相关(r=0.928,P=0.008),而与CD4±invariant NKT细胞比例和总的invariant NKT细胞比例无明显相关性。3.替比夫定抗病毒治疗后外周血invariant NKT细胞的比例升高19例CHB患者在抗病毒治疗前(Baseline)其外周血invariant NKT细胞占CD3+T细胞的比例为0.18%(0.03%-1.07%)；接受抗病毒治疗后的12W,比例为0.21%(0.04%-1.43%),显著高于Baseline(P=0.0231)；治疗24w后,比例为0.20%(0.04%-1.49%),高于Baseline,但没有统计学意义(P=0.08)；治疗52w后,invariant NKT细胞占CD3+T细胞的比例为0.21%(0.05%-1.22%),也是显著高于Baseline的(P<0.01)。治疗后12W、24W和52W之间互相是没有显著性差异的。4.CHB患者基线invariant NKT细胞亚群比例能预测替比夫定治疗52W后e抗原血清学转换在19例接受替比夫定抗病毒治疗的CHB患者中,有7例(36.84%)在治疗52W后发生了e抗原血清学转换。在治疗过程中,无论转换组还是非转换组,患者血浆中的HBV DNA滴度都是下降的,但两者下降的趋势不一样。非转换组在治疗后的前12W,病毒滴度快速下降至4-5log,然后在这个较高的病毒水平上维持相对的稳定。而转换组在治疗后的前12W,病毒滴度快速下降至3-4log,然后随着治疗时间的延长,病毒滴度继续下降至3个log以下。统计结果显示,在替比夫定抗病毒治疗后的24W和52W,转换组的HBV DNA水平是显著低于非转换组的(12W,P<0.05,52W,P<0.01)。在基线,转换组的ALT水平(U/L)为188.9±78.17,虽然高于非转换组(128.2±98.84),但由于例数较少,并没有统计学意义(P=0.1082)。抗病毒治疗后,无论是转换组还是非转换组,其血清ALT水平都是持续下降的。把19例接受替比夫定抗病毒治疗的患者按基线CD4-/CD4±invariant NKT细胞的比值分成两组：高比例组(≧1)和低比例组(<1)。然后比较两组抗病毒治疗52W后e抗原血清学转换情况。结果显示,高比例组(n=12)中有7例(58.33%)发生了血清学转换,显著高于低比例组。ROC曲线下面积为0.75(95%CI,0.502-0.917),能用于预测治疗52W后的e抗原血清学转换(P=0.04)。CD4-/CD4±invariant NKT细胞比值在0.94时具有最好的预测效果(敏感性100%,特异性58.6%)。5.活化的invariant NKT细胞能促进e抗体分泌细胞的产生用a-GalCer刺激新鲜分离的PBMC(来自于IC携带者),用ELISPOT检测其对e抗体分泌细胞产生的影响。结果显示,不加a-GalCer刺激时,有2例(40%)IC可以用ELIPOT检测出e抗体分泌细胞；加入100ng/ul a-GalCer刺激60h后,有4例(80%)可检测出e抗体分泌细胞,且前面俩例不加a-GalCer刺激也能检测出的在加了a-GalCer刺激后,e抗体分泌细胞的数量也是增多的。统计结果也显示a-GalCer刺激能显著增加IC携带者PBMC中e抗体分泌细胞的产生(P<0.05)。我们还用雅培AxSYM MEI kits半定量检测了a-GalCer刺激60h后的上清中e抗体的水平,发现5份样本中只有一份检测结果为阳性。6.IL-21能抑制a-GalCer对invariant NKT细胞的增值我们前期的研究结果显示,CHB患者外周血中invariant NKT细胞的比例是降低的,而血清中IL-21浓度是升高的。考虑到invariant NKT细胞是IL-21重要的分泌细胞,我们推测IL-21浓度的升高可能会反馈抑制invariant NKT细胞的增值。我们用a-GalCer在体外刺激新鲜分离的CHB患者的PBMC,7天后收集细胞,流式检测invariant NKT细胞的比例。结果显示其比例为0.59±0.28%,明显高于未刺激组(0.28±0.18%,P<0.05)。如果在加入a-GalCer刺激的同时,加入终浓度为100ng/mlrhIL-21,刺激7天后,invariant NKT细胞的比例为0.25%±0.17%,显著低于a-GalCer单独刺激组(P<0.05),但与未刺激组相比无明显差异。收集培养上清,用FlowCytomix检测Th1和Th2细胞因子的浓度。不加IL-21刺激,上清中IL-10的浓度为23.87±14.96pg/ml,IL-6的浓度为577.2±892.9pg/ml,IL-1β的浓度为18.92±4.16pg/ml o10ng/ml的IL-21刺激后,上清中IL-10的浓度为109.54±21.31pg/ml,显著高于未刺激组(P<0.05)；IL-6的浓度为1467.5±680pg/ml,显著高于未刺激组(P<0.05)；IL-1β的浓度为50.13±28.444pg/ml,显著高于未刺激组(P<0.05)。100ng/ml IL-21刺激后,上清中IL-10的浓度为378±266.9pg/ml,显著高于未刺激组(P=0.0022)和lOng/ml IL-21刺激组(P<0.05)；IL-6的浓度为5640±1165pg/ml,显著高于未刺激组(P=0.0087)和10ng/ml IL-21刺激组(P<0.05)；IL-1β的浓度为315.8±142.4pg/ml,显著高于未刺激组(P=0.0095)和10ng/ml IL-21刺激组(P<0.05)。IL-21刺激后,TNF-a的浓度也有升高,但无统计学差异。我们用FlowCytomix还检测了培养上清中IL-12p70、IFN-γ、IL-2、IL-8、IL-4、IL-5和TNF-β的浓度,发现IL-21刺激组和未刺激组间无明显差异。结论：1.我们的横向研究结果显示外周血invariant NKT细胞的比例在慢性HBV感染中有一个升高、降低、恢复的过程,表明invariant NKT细胞参与了慢性HBV感染的免疫学发病机制,且与病毒复制和肝内炎症相关。2.CHB患者外周血invariant NKT细胞比例的降低可能与其高表达CCR5和CCR6相关,即因肝内炎症而产生的大量趋化因子招募invariant NKT细胞进入肝内。3invariant NKT细胞具有异质性,而CD4-和CD4+invariant NKT细胞的比值可作为预测CHB患者替比夫定治疗52周后e抗原血清学转换的一个潜在指标。更多还原

【Abstract】 Background and aim:Worldwide, more than300million people suffer from chromic hepatitis B virus (HBV) infection, leading to a wide spectrum of liver diseases including asymptomatic carrier (AsC), chronic hepatitis B (CHB), cirrhosis and hepatocellular carcinoma. In china, over7%of the population are infected with HBV, out of which30million people are developed to CHB. Moreover, there are about300thousand patients die from HBV infection related diseases in our country every year. Given that there is no ideal drug to eliminate HBV completely from human body currently, chronic HBV infection has become a heavy health burdern at present. The pathogenesis of CHB and cirrhosis is thought to be mediated by the immune response to the HBV rather than the HBV itself. Multiple types of immune cells and molecules are involed in HBV associated liver damage. However, the precise roles of these cells and molecules are still incompletely understood.Invariant NKT (invariant NKT) cells are a unique group innate T lymphocytes that express an identical T cell antigen receptor (TCR) a chain, Val4-Jα18in mice and Va24-Jα18in humans.invariant NKT cells differ from conventional T lymphocytes in that they recognize lipid or glycolipid antigens presented by the MHC class I-like molecule CD1d. When activated with CD1d tetramer or anti-CD3antibody, invariant NKT cells rapidly secrete a variety of Thl and Th2cytokines within a few hours. Although invariant NKT cells comprise a very small proportion of peripheral T cells, about1%in mice and0.2%in humans, they seem to play important roles in regulating a number of immune responses, including transplant rejection, cancer, autoimmunity, allergy, and infection.The liver contains a large number of invariant NKT cells relative to blood and other lymphoid organs. Increasing evidence suggests that invariant NKT cells contribute to a variety of liver disorders, including drug-induced liver injury, primary biliary cirrhosis, alcoholic liver injury, autoimmune hepatitis, hepatocellular carcinoma, non-alcoholic fatty liver disease, and viral hepatitis. In CHB, alpha-galactosylceramide (a-GalCer) activated invariant NKT cells are able to inhibit HBV replication in vivo and are implicated in the pathogenesis of cirrhosis by producing profibrotic cytokines. Activation of invariant NKT cells also promotes the loss of tolerance to HBV-specific CD8+T cell antigens. However, most of these reports are based on mouse models, and the role of invariant NKT cells in CHB patients is largely unknown. There are few reports about the changes in invariant NKT cell frequency or activity in CHB patients during antiviral therapy.In the present study, we compared the frequency of circulating invariant NKT cells in chronic HBV infection subjects and healthy individuals by flow cytometry. We also compared the expression of chemokine receptors on invariant NKT cells, the ability of invariant NKT cells to migrate toward different chemokines and the ability to secret cytokines between CHB patients and healthy controls. Afterward we analyzed the longitudinal changes of invariant NKT cells frequency in CHB patients who received antiviral therapy with telbivudine. Finally, we detected intrahepatic invariant NKT cells by immunohistochemistry and try to understand the mechanisms of the decrease of circulating invariant NKT cells in CHB patients and subsequent changes by expriments in vitro. Methods1. SubjectsThirty-five CHB patients,15immunetolerant carriers (IT),25inactive HBV carriers (IC) and36healthy controls (HC) were enrolled in the present study. CHB patients, IT and IC were diagnosed according to the AASLD criteria. The subjects with previous antiviral therapy, with co-infection by the HIV, other hepatitis virus, and with diabetes, severe systemic illness, regular alcohol consumption and hepatocellular carcinoma were excluded.2. Flow Cytometry AnalysisTen to thirty milliliter (ml) of heparinized blood was collected from all individuals at the time of recruitment. Of the CHB patients,19received anti-viral therapy with telbivudine and were followed up for at least52weeks, and additional blood samples were obtained at week12, week24and week52during the course of therapy. Peripheral blood mononuclear cells (PBMC) were isolated from whole blood by Ficoll density gradient centrifugation. One million fresh isolated PBMC were washed and stained with fluorochrome-conjugated antibodies. The frequencies and phenotype of circulating invariant NKT cells were analysed.In some cases, the cytokine production within peripheral invariant NKT cells was determined by Intracellular cytokine staining.3. Migration Assay The ability of recombinant CC chemokine ligand5(CCL5) and CCL20to attract invariant NKT cells was examined using a transwell system with24-well and5μm pore size. One and half million fresh isolated PBMC were added to the upper chamber,500ng/ml CCL5or CCL20were added to the lower chamber. After10hours incubation at37℃, the cells in the lower chamber were harvested and analysed using flow cytometry. 4. Elispot AssayIFN-γ Elispot assay was performed to detect the effect of recombination human IL-21on the IFN-y production of invariant NKT cells stimulated with a-GalCer. Multiscreen-iP96-well plates were coated with capture mouse antihuman IFN-y monoclonal antibody, then blocked with RPMI1640-10%FCS. PBMC were seeded in the wells and a-GalCer was added to stimulate invariant NKT cells. Plates were incubated overnight in the presence or absence of IL-21. After washing, biotinylated secondary mouse antihuman IFN-y monoclonal antibody was added. After3h of incubation, goat alkaline phosphatase antibiotin antibody was added and the plates were incubated for a further2h. Substrate was added and the colorimetric reaction was stopped by washing with distilled water. Plates were air dried, and spots were photographed and counted. In some cases, plates were coated with recombinate HBeAg or HBcAg to evaluate the function of invariant NKT cells on the production of HBV specific antibody.5. ImmunohistochemistryParaffin-embedded, formalin-fixed liver tissue from patients with liver surgery was cut into5μm sections and placed on polylysine-coated slides. Antigen retrieval was achieved via pressure cooking for10minutes in citrate buffer. Mouse anti-human invariant NKT cells monoclonal was added and incubated for45minutes. EnVisionTM Detection Kit was used to stain invariant NKT cells followed by counterstaining with hematoxylin.6. Statistical AnalysisComparisons between different groups were performed using the Mann-Whitney U test. Within-subject data were compared with the repeated measures ANOVA followed by Tukey’s or Bonferroni post-tests. The Spearman rank order correlation coefficient was used for correlation analyses. Categorical variables were compared by Fisher’s exact test. The area under receiver operating characteristics (ROC) curves were calculated to assess the use of peripheral invariant NKT cells frequency at baseline to predict HBeAg seroconversion. For all tests, a two-sided P<0.05was considered significant.Results1. We compared the frequencies of peripheral invariant NKT cells in subjects from the HC, IT, CHB and IC groups.invariant NKT cells were detected by flow cytometry using a monoclonal antibody6B11, which specifically reacts with the complementarity determining region3(CDR3) of the Va24-Ja18T cell receptor of human invariant NKT cells. We found that the frequency (medium, range) of invariant NKT cells in CHB patients (0.13%,0.027%-1.071%) was significantly lower than that in HC (0.24%,0.064%-0.81%, P=0.0143) and IC (0.19%,0.06%-0.38%, P=0.0235). By contrast, there was no significant difference between HC and IC (P=0.703). Invariant NKT cells are heterogeneous and can be categorized into two subsets according to the expression of CD4, namely CD4+and CD4-invariant NKT cells. A comparison of the frequency of CD4+and CD4-invariant NKT cells was performed in the three groups. We found that the frequency of CD4-invariant NKT cells in CHB patients (0.071%,0.013%-0.98%) was decreased compared to HC (0.175%,0.027%-0.7%, P<0.01), and the frequency of CD4-invariant NKT cells in IC (0.072%,0.019%-0.73%) was also significantly lower compared to HC (P=0.023). There was no significant difference (P=0.458) between CHB and IC in the frequency of CD4-invariant NKT cells. By contrast, the frequency of CD4+invariant NKT cells in CHB patients (0.053%,0.007%-0.83%) was significantly lower than IC (0.1%,0.027%-0.65%, P=0.019), but was similar to HC (0.055%,0.01%-0.2%,P=0.7).2. We next compared the ratio of CD4-/CD4+invariant NKT cells in the different groups, and found that the ratio was decreased in CHB patients (1.21,0.11-10.77) relative to the HC (2.61,0.53-9.55, P<0.01), and even further decreased in the IC (0.78,0.11-8.5) relative to CHB (P=0.018). We also analyzed the correlation between peripheral invariant NKT cells and plasma HBV DNA load and serum alanine aminotransferase (ALT) level in the CHB patients (n=30). The results indicated that there was no correlation between the frequency of total invariant NKT cells and plasma HBV DNA load or serum ALT level. By contrast, CD4-invariant NKT cells and the ratio of CD4-/CD4+invariant NKT cells were negatively correlated with ALT levels.3. We detected the expression of CCR5and CCR6on peripheral invariant NKT cells in CHB patients (n=6) by flow cytometry and found that the proportion (mean±SD) of invariant NKT cells expressing CCR5(82.83%±9.87%) or CCR6(67.67%±16.83%) was dramatically higher than that of other T cells (CCR5,30.5%±5.65%, CCR6,14.02%±5.92%, both P<0.001). Further analysis revealed that the percentage of CD4-invariant NKT cells expressing CCR5or CCR6was also significantly higher than that of CD4+invariant NKT cells (P<0.05and P<0.01respectively). Interestingly, the expression of CCR5and CCR6on invariant NKT cells was comparable between CHB patients and HC subjects.4. Our data above indicated that invariant NKT cells expressed higher levels of CCR5and CCR6than other T cells. Therefore, we performed a chemotaxis assay to test the hypothesis that CCL5and CCL20, the ligands of CCR5and CCR6respectively, could induce the migration of invariant NKT cells in vitro. We found that the frequency of invariant NKT cells in the lower chamber in response to CCL5(0.87%±0.24%) was significantly higher compared to medium alone (0.65%±0.23%, P<0.001). Surprisingly, the frequency of invariant NKT cells in the lower chamber in response to CCL20was similar to medium alone. 5. We also assessed the ability of invariant NKT cells to produce IFN-y and IL-4in response to the specific antigen a-GalCer and non-specific mitogen PMA by flow cytometry. The IFN-γ production of circulating invariant NKT cells in response to a-GalCer was dramatically higher compared to IL-4, irrespective of whether they were obtained from CHB patients or HC (both P<0.001). Interestingly, there were no differences between the two groups in both IFN-γ(P=0.673) and IL-4expression (P=0.8884). When stimulated with PMA, there were also no significant differences between CHB patients and HC in IFN-γ and IL-4production (P=0.12, P=0.75respectively). Surprisingly, the IFN-γ response to PMA was lower compared with the response to a-GalCer, regardless of CHB patients or HC (both P<0.001).6. We analyzed the changes in peripheral invariant NKT cell frequency in CHB patients on antiviral therapy. PBMC were obtained at baseline and at week12,24and52after initiating treatment with telbivudine (n=19). We found that the frequency of peripheral invariant NKT cells was significantly increased at week52(0.21%,0.05%-1.22%) compared with baseline (0.18%,0.03%-1.07%, P=0.0176). At week12and week24, the frequency of circulating invariant NKT cells was also increased compared to the baseline, but the difference did not reach statistical significance. Further analysis indicated that the frequency of CD4-but not CD4+invariant NKT cells was dramatically increased by week52(P<0.01, P=0.2898respectively).7. Of the19CHB patients treated with telbivudine,7(37%) achieved HBeAg seroconversion by week52. The plasma HBV DNA load in patients with seroconversion decreased more rapidly than patients not achieving seroconversion during the course of therapy, especially after week12. By contrast, both the seroconversion and non-seroconversion patients had the similar variation mode in serum ALT levels and circulating invariant NKT cells. The precise mechanisms contributing to HBeAg seroconversion during antiviral therapy are not fully understood. The patients who achieved HBeAg seroconversion tended to have a slightly higher ratio of CD4-/CD4+invariant NKT cells (1.47,1-3.56) compared to those who failed to achieve HBeAg seroconversion (0.915,0.11-3.4, P=0.0826). We also analyzed this data by dividing the patients into high (≥1) and low (<1) ratio groups and found that the patients with a high ratio of CD4-/CD4+invariant NKT cells at baseline were more likely to achieve HBeAg serocon-version than those with a low ratio (P=0.02). Next, a receiver operating characteristics (ROC) curve was constructed to determine whether the ratio of CD4-/CD4+invariant NKT cells could be used as a predictor of HBeAg seroconversion within52weeks of starting telbivudine therapy. The area under the ROC curve (AUC) indicated the existence of a significant association between the CD4-/CD4+invariant NKT ratio and HBeAg seroconversion (AUC=0.75,95%confidence interval0.507-0.993, P=0.04). A CD4-/CD4+invariant NKT cell ratio cutoff at baseline of0.94gave the maximum combination of sensitivity (100%) and specificity (58.33%) to predict HBeAg seroconversion.Conclusions1. Circulating invariant NKT cells are decreased in CHB patients and recovered to normal level after viral control. There is no correlation between the frequency of total invariant NKT cells and plasma HBV DNA load or serum ALT level. By contrast, CD4-invariant NKT cells and the ratio of CD4-/CD4+invariant NKT cells were negatively correlated with ALT levels.2. Invariant NKT cells express high levels of CCR5and CCR6, and migrate more likely toward CCL5than other conventional T cells, which may contribute to the decreasing of circulating invariant NKT cells. 3. The ability of peripheral invariant NKT cells to produce IFN-γ and IL-4is not impaired with chronic HBV infection and IFN-y is the dominant cytokine secreted by circulating invariant NKT cells in response to a-GalCer.4. The ratio of CD4-/CD4+invariant NKT cells in CHB patients at baseline is associated with HBeAg seroconversion by week52after telbivudine therapy, and has the potential to serve as a predictor for HBeAg seroconversion.更多还原