|Home | About | Journals | Submit | Contact Us | Français|
After partial hepatectomy, hepatocytes proliferate to restore mass and function of the liver. Macrophages, natural killer (NK) cells, natural killer T (NKT) cells, dendritic cells (DC), eosinophils, gamma delta T (?dT) cells, and conventional T cells, as well as other subsets of the immune cells residing in the liver control liver regeneration, either through direct interactions with hepatocytes or indirectly by releasing inflammatory cytokines. Here, we review recent progress regarding the immune cells in the liver and their functions during liver regeneration.
The liver is the largest solid organ within the body. It receives about 20% of its blood supply from the hepatic artery and 80% from the portal vein [1–3]. Owing to its anatomical location and blood supply, the liver is the first line of defense against various blood-borne pathogens and is very important to host immunity and survival [3–6]. To accomplish its immunological roles, the liver is enriched in several subsets of innate (such as macrophages, NK cells, NKT cells, neutrophils, γδT cells, dendritic cells, innate lymphoid cells) [6–8] and adaptive immune cells (such as T cells and B cells) [3, 9], which are finely tuned to affect the status of immuno-tolerance, pathogen clearance, tumor progression, and acute injury of the liver (refer to Table Table11 for the information of non-parenchymal cell subsets in the liver) [3, 6, 7, 10].
Normally quiescent hepatocytes will undergo proliferation in response to various stimulations, such as toxic injury, viral infection and surgery. Most studies concerning liver regeneration take the advantage of the two-thirds partial hepatectomy model in mice or rats. In this model, two-thirds of the liver, usually the median and left lateral lobes, is surgically removed. In response to this, the remnant liver enlarges until it restores normal mass and functions [11–13]. This process usually takes about 10 days, after which the regeneration process stops. Unlike the conventional meaning of ‘regeneration’, which usually means the complete re-growth of an excised tissue , liver regeneration is a very different process, which does not lead to the restoration of the excised lobules, but rather the compensatory hyperplasia of the remnant lobules.
There have been different groups of researchers attempting to explain the mechanisms of liver regeneration. Accumulating evidence demonstrates that partial hepatectomy can lead to an acute phase response in the liver, during which the immune system will be robustly activated, and inflammatory mediators, including cytokines, chemokines, and complements will be released, stimulating quiescent hepatocytes to enter the G1 phase of cell cycle. Thereafter, various growth factors are secreted to further enhance the proliferation of the primed hepatocytes. At last, inhibiting signals are activated to avoid excessive regeneration, until the liver restores its normal mass, architecture, and function (Figure (Figure1)1) [11, 12]. The effects of these mediators are complicated and finely tuned to ensure an efficient and effective regeneration process. Here, we mainly summarize the recent literatures concerning the immune system in the liver and their functions during the process of liver regeneration.
It was formerly believed that all macrophages were differentiated from blood monocytes [15, 16]. However, only recently did researchers find that there were in fact two distinct populations of macrophages in various tissues according to their progenitors and development process, namely yolk-sac-derived tissue-resident macrophages and bone marrow-derived circulating macrophages. The former were F4/80highCD11blow and the latter were CD11bhighF4/80low in various tissues [17, 18]. In fact, tissue-resident macrophages and bone marrow-derived macrophages have been demonstrated to play distinct and non-redundant roles in models of injury, repair, and regeneration [19–24].
In response to inflammatory signals, macrophages could be polarized into two functionally distinct subsets, namely M1 and M2 macrophages. Interferon-γ (IFN-γ) and LPS lead to the M1 activation of macrophages (classical activation), whereas IL-4 and IL-13 induce the M2 activation of macrophages (alternative activation) [25–28]. The M1 phenotype is exemplified by high levels of pro-inflammatory cytokines, high secretion of reactive oxygen and nitrogen intermediates, which promote strong tumoricidal and microbicidal activities. On the other hand, M2 activation is characterized by strong phagocytic activity, high production of ornithine and polyamines and expression of mannose, scavenging, and galactose receptors. M2 macrophages mainly exert protumoral and immunoregulatory functions [29–31].
In the liver, macrophages represent about 20% of the non-parenchymal cells. They serve as the immune sentinel of the liver, sensing various stimulants and alerting other immune cells through delicate cell-cell interaction and secreted cytokines [3, 32]. Among the innate immune system in the liver, macrophages are the most extensively studied cells during liver regeneration. Several lines of evidence demonstrated that macrophage activation is beneficial to liver regeneration and provide the initial priming force for hepatocyte proliferation. The most convincing evidence would be experiments involved in macrophage depletion. A number of groups found that macrophage depletion would greatly compromise the regeneration rate of the liver [33–36]. It is believed that macrophage-derived cytokines such as tumor necrosis factor-α (TNF-α) and IL-6 were the most important forces for liver regeneration after partial hepatectomy [12, 37–39]. The results of several different groups corroborated this notion. For example, macrophage-depleted mice failed to exert a similar cytokine response thus resulted in retarded liver regeneration after partial hepatectomy [33, 40–42]. And macrophage colony stimulating factor (M-CSF) deficiency was reported to impair liver regeneration . To elucidate the mechanism in this, it was shown that the complement components C3a and C5a were critical for activating macrophages through complement receptors, and stimulated the production of IL-6 and TNF-α, thus primed liver regeneration after partial hepatectomy . Besides, it was indicated that TLRs/MyD88 signaling was responsible for the production of TNF-α and IL-6 from Kupffer cells . The upregulation of these cytokines in response to partial hepatectomy was suggested to be due to the stimulation of liver macrophages by enteric-derived bacterial products [12, 42, 46]. In addition, monocyte chemoattractant protein-1 (MCP-1), intercellular adhesion molecule 1 (ICAM-1), and osteopontin were demonstrated to be the factors responsible for recruiting macrophages to the liver after partial hepatectomy (Figure (Figure2)2) [34, 40, 47].
Besides these directly mitogenic functions, liver macrophages could also interact with other cell subsets thus indirectly promote the process of liver regeneration. Recently, it was reported that gut commensal microbiota, especially ampicillin sensitive bacteria were essential to keep Kupffer cells in a tolerant state, thus preventing NKT cell overactivation after partial hepatectomy . Moreover, circulating macrophages could also interact with liver endothelial cells, and this interaction was crucial for vascular growth and liver regeneration . In addition, Kupffer cells were also required for the proliferation of liver progenitor cells that were able to differentiate into hepatocytes in a CDE diet-mediated liver injury and regeneration model .
Interestingly, in addition to hepatic macrophages, macrophages from other tissues could also be functional during liver regeneration. Recently, it is reported that a subset of F4/80hiGATA6+ macrophages in the peritoneal cavity could be rapidly recruited into the liver via the mesothelium and have pivotal reparative ability during liver regeneration .
NK cells are another subset of innate immune cells that can kill invaded pathogens and transformed cells in different tissues. Through a sophisticated repertoire of activating and inhibitory receptors, their cytotoxicity activities are controlled delicately [51, 52]. It was demonstrated that bone marrow was the main development site for NK cells, providing various stimuli to guarantee its differentiation . In mice, expressions of CD27 and CD11b define four sequential developmental stages of NK cells: CD11blowCD27low represents the most immature NK cells, then comes CD11blowCD27high, and then CD11bhighCD27high, and at last comes the most mature CD11bhighCD27low NK cells . Among these NK cells, CD11bhighCD27high NK cells produce the largest amount of cytokines and are most cytotoxic , whereas the final step CD11bhighCD27low NK cells express the inhibitory receptor KLRG1 and show decreased effector functions and proliferation ability .
Apart from macrophages, NK cells represent another important component of the innate immune system in the liver . NK cells constitute 30%~50% of the intrahepatic lymphocytes in humans, and 10%~15% in mice . They are one of the most important cells in the control of bacterial and viral infection in the liver, such as HBV, HCV, murine cytomegalovirus, and Listeria monocytogenes infection [57, 58]. Liver NK cells are unique, because there is a subset of CD49a+DX5- liver-resident NK cells exerting memory responses in a contact hypersensitivity model . Accumulating evidence suggests that NK cells are increased and activated in the liver after partial hepatectomy. In contrast to macrophages, liver NK cells inhibit rather than promote liver regeneration, and this inhibitory effect is likely mediated through the secretion of IFN-γ. The most convincing evidence for this is that liver regeneration is impaired when NK cells were activated by Poly I:C, and NK cell depletion enhanced liver regeneration rate. Moreover, liver regeneration was enhanced in IFN-γ deficient mice [60, 61]. The retarding effect of IFN-γ on liver regeneration was thought to be mediated by anti-proliferative proteins such as STAT1, IRF-1, and p21cip1/waf1 in the hepatocytes . This function of NK cells in liver regeneration was recently supported by a study that TIGIT, which was a co-inhibitory receptor on NK cells, was involved in liver regeneration. TIGIT expression level was selectively up-regulated on liver NK cells after partial hepatectomy, and its deficiency promoted the activation of NK cells, ultimately hampering liver regeneration through TIGIT-PVR interaction .
NKT cells are a unique subset of T cells which are CD1d-restricted and lipid-antigen reactive . In the liver, NKT cells have been shown to play a pathogenic role in ischemia reperfusion injury, nonalcoholic fatty liver disease, con-A induced hepatitis, primary biliary cholangitis, and so on [64–67]. Interestingly, NKT cells could produce both type I and type II cytokines, making them both pathogenic and protective. For instance, they are beneficial in acetaminophen-mediated acute liver injury, and they can also limit inflammatory cytokine secretion from macrophages in a CCl4 mediated acute liver injury model [68, 69]. Similar with NK cells, NKT cell number also increases in the regenerating liver. However, NKT cells may not be so potent as NK cells during liver regeneration, which was indicated by the fact that CD1d-/- or Jα281-/- mice showed similar regenerating rate compared with WT mice after partial hepatectomy [60, 70]. Nevertheless, NKT activation clearly impeded liver regeneration, which was also involved in IFN-γ mediated STAT1 signaling. This was strengthened by a finding that ampicillin-sensitive commensal bacteria depletion would activate liver NKT cells in an IL-12 dependent way and impair liver regeneration after partial hepatectomy . Thus, it is likely that normal NKT cell biology was dispensable for liver regeneration, but activation of NKT cell by context-dependent stimuli would obviously impede liver regeneration after partial hepatectomy.
γδT cells in liver regeneration
Apart from macrophages, NK cells, and NKT cells, the liver is also selectively enriched in a special subset of T cells, namely γδT cells, which constitute about 15%~25% of the liver T cells. Most γδT cells are developed from the fetal thymus, from a common precursor of both αβT cells and γδT cells. Unlike αβT cells, a small proportion of γδT cells also generated in intestinal epithelium during early weeks of life . To recognize antigens, αβT cells use T cell receptor (TCR) bound to CD3 to interact with the major histocompatibility complex on antigen-presenting cells, whereas γδT cells do not require such interaction . Thus, γδT cells mainly sense early environmental signals to initiate local immno-surveilance, whereas the activation of αβT cells is relatively late . In addition, even though γδT cells and αβT cells are both CD3 positive, most of γδT cells do not express CD4 and CD8 . Like other cell subsets, γδT cells may be protective or pathogenic, and are also involved in various kinds of liver diseases, ranging from acute liver injury to liver cancer to chronic liver infection [6, 74]. It is observed that γδT cells are also activated and increased in number after partial hepatectomy . Even though it had been reported earlier that IL-17A is crucial for normal liver regeneration , only in 2014 did Rao et al. found that it was γδT cell-derived IL-17A that exerted the regeneration-promoting function. They found that these IL-17A-secreting γδT cells were able to induce the production of IL-6 from antigen-presenting cells, and at the same time inhibited the secretion of IFN-γ from NKT cells, which meant that they could be both directly mitogenic for hepatocytes and promoted a regenerative-contributing phenotype in hepatic lymphocytes .
DCs are sparsely distributed within the liver. Usually, DCs in multiple tissues can be identified as CD45+ cells with high expression of MHCII and CD11c. However, to exclude other hematopoietic cell types when identifying DCs is very important, since DCs could also express some markers that are frequently expressed by other cells, such as macrophages and B cells . In the liver, DCs can be divided into two subsets under steady state, namely plasmacytoid DCs (pDCs) and classical DCs (cDCs). pDCs express relatively lower levels of MHCII and thus have a limited ability to capture and present antigens, whereas cDCs express very high levels of MHCII and are very professional antigen-presenting cells . Even though they are relatively rare in number, DCs are of great importance of liver immune regulation, such as liver fibrosis, alcoholic liver injury, as well as liver cancer [77–81]. In response to partial hepatectomy, liver DC number increased dramatically, indicating these cells were also functional during liver regeneration process . Moreover, it was found that the partial hepatectomy-stimulated DC facilitated IL-10 production while inhibited IFN-γ production of T cells, and they could also enhanced estrogen receptor expression, thus promoting liver regeneration rate . In a later study, researchers found that DCs were also capable of secreting TNF-α, and the reduced TNF-α production level would compromise liver regeneration rate .
Eosinophils are myeloid derived cells. Upon activation, these cells are high granulated and will secrete cytokines, cytotoxic granule proteins, as well as enzymes and lipid mediators to kill pathogens or host cells . These cells could be involved in a variety of pathogenic processes, such as asthma, allergy, and helminthic infection [84–88]. In the liver, these cells could be associated with transplantation rejection [89, 90], drug induced liver injury , liver fibrosis , and viral induced hepatitis [93, 94]. Of note, eosinophils also actively participate in the process of liver regeneration. In models of partial hepatectomy and toxin treatment, the number of eosinophils increased significantly. Also, eosinophil-lacking mice showed compromised regeneration rate of the liver. Later, it was found that eosinophil-derived IL-4 was the central factor promoting the proliferation of quiescent hepatocytes .
The innate lymphoid cells (ILCs) are the most recently found members of the innate immune system and have been investigated intensely over the past six years. According to their specific surface markers, transcriptional factors, and the effector cytokines they secrete, the ILCs could be divided into three distinct subsets, namely group 1 ILCs (ILC1s), ILC2s, and ILC3s. All ILCs are developed from a common lymphoid progenitor, but lack the specific markers of other immune cell subsets. Moreover, unlike T cells and B cells, ILCs do not have antigen specificity because of their lack of antigen receptors [96, 97]. In multiple tissues of the body, ILCs can orchestrate homeostasis, inflammation and immunity through communicating with multiple cell types [98, 99]. In the liver, it was reported that CD49a+ ILC1s express high levels of NKG2A, thus limited the recruitment of peripheral NK cells, making the liver as a tolerogenic site during various kinds of viral infections [100, 101]. Moreover, it was reported that in response to IL-33, liver ILC2s were activated and secreted considerable amount of IL-13, leading to hepatic stellate cell activation and ultimately liver fibrosis ; and this circuit of signaling could also improve biliary repair to promote biliary carcinogenesis . However, the functions of ILCs during the process of liver regeneration are largely unexplored. It was reported that liver specific IL-22 overexpression accelerated liver regeneration, yet the underlying mechanism was elusive . Recently, it has been reported that ILC1s are indispensable for efficient liver regeneration. In this process, IL-22 is secreted by ILC1s in response to extracellular ATP signaling, and is identified as the critical mediator of liver regeneration after partial hepatectomy .
The adaptive immune responses are consisted of two forms: the humoral immunity, mediated by B cell-produced antibodies, and the cellular immunity, which is mediated by T cells . Within the liver, the adaptive immune system is indispensable in a number of physiological and pathological processes, ranging from tolerance maintenance [106, 107], autoimmune diseases [107–110], tumors [111, 112], bacterial infection , viral infection [113–115], transplant rejection [116–118], obesity [119, 120], fibrosis [121–123], acute injuries [124–128], and so on. Nevertheless, the function of the adaptive immune system in the process of liver regeneration is rarely explored. However, this is not to say that the function of the adaptive immune system can be neglected. It was reported that T cell deficiency would greatly compromise the ability of liver regeneration and increase the mortality rate in mice after partial hepatectomy, indicating that T cells were critically indispensable for normal liver regeneration. The mechanism in this process was that T cell-secreted lymphotoxin could stimulate IL-6 production and signal transducer and activator of transcription 3 (STAT3) activation in the liver .
Liver regeneration is a complicate process that involves the cooperation of various immune cells (see Figure Figure3).3). Despite a great number of studies have been focusing on elucidating the mechanisms of the immune system in liver regeneration, many questions still need to be addressed. First, most studies concentrate on one specific cell subset, but further studies would be needed to investigate the cell-cell interactions in the liver cell network. Second, from a clinical perspective, investigators need to distinguish the differences between animal models and human cases, and more relevant and appropriate studies are needed to explore the regeneration process of clinical cases, such as toxins, tumors, ischemia reperfusion, or liver transplantation cases. In addition, considering the potential application of regenerative medicine, investigations concerning the transplantation or activation of intrahepatic stem cells are potentially promising for novel treatment of liver diseases.
Conflicts of interest
The authors declare no conflict of interest.
This study was supported by Program of National Natural Science Foundation of China, 31272518 and 31572399; the China National Basic Research Program (JFYS 2016YFA0100203), and the Agriculture science and technology innovation and research in Shaanxi Province (Grant No. S2015YFNY0014).