In a healthy liver, most cholangiocytes remain dormant mitotically, and proliferation occasionally occurs to achieve cellular equilibrium (11). In this review, we mainly focus on cholangiocyte proliferation in pathological conditions. The cell size of cholangiocytes gradually increases with morphological and functional changes along with the canal of hering to the large bile duct (12). This cell heterogeneity along with the biliary tree endows them with different responsiveness to insults (13). Major signaling pathways driving proliferation in both small and large cholangiocytes are Ca²⁺ and cyclic adenosine monophosphate (cAMP)-dependent (14). Secretin can bind to secretin receptors, and only large cholangiocytes express and elevate intracellular cAMP, initiating cell proliferation through the protein kinase A (PKA)/mitogen-activated protein kinase (MEK)/extracellular signal-regulated protein kinase1/2 (ERK1/2) signaling pathway (15–17). For small cholangiocytes, agents such as forskolin and follicle-stimulating hormone can also trigger cAMP increase and lead to cholangiocyte proliferation (17, 18). In addition, activation of the inositol 1,4,5-trisphosphate (IP₃)/Ca²⁺ signaling pathway can also stimulate the proliferation of small cholangiocytes (14, 19).

Usually, large bile ducts are more susceptible to damage, resulting in compromised proliferative capability (20). Under this condition, small cholangiocytes can acquire the phenotype of large cholangiocytes and help to restore biliary epithelium as a compensatory method (21, 22). However, the mechanisms of such differentiation remain elusive. Heterogeneity of regenerating capacity was also found independent of cell size in cholangiocytes. It was indicated that cholangiocyte proliferation predominantly contributes to biliary repair. Proliferation is chiefly generated by a subgroup of cholangiocytes with persistent regenerating capability, which is not predestined but rather “stochastically” regulated during injury (23). In general, the biliary system does not uniformly proliferate as a whole due to significant cellular heterogeneity among the biliary tract, and this highly dynamic and adaptive replenishment process plays a fundamental role in biliary repair.

Although cholangiocytes possess immense regenerative capacity, severe acute injury and prolonged chronic injury could exhaust their proliferative potential (24). The liver could also activate alternative reparation methods, such as the LPC expansion, to avoid abnormal repair, which eventually leads to ductular fibrosis and cirrhosis. LPCs, also known as oval cells, are considered bipotent progenitor cells that reside adjacent to the canal of hering and can differentiate into cholangiocytes or hepatocytes (25, 26). However, whether LPCs can give rise to cholangiocytes or hepatocytes under physiological conditions has obtained controversial results (27–30), so as in the biliary repair of cholangiopathies. While LPCs expand in response to insults, the cell fate of LPCs is chiefly determined by the type of liver injury (11, 31). When biliary damage causes regenerative capacity exhaustion of cholangiocytes, LPCs can differentiate toward cholangiocytes and impede hepatocyte differentiation simultaneously. Such lineage specification is modulated by an intricate signaling network, including the Notch (31), Wnt-β-catenin (31–34), and Hippo-Yes-associated protein (YAP) signaling pathways (35, 36).

LPC differentiation is also, to a large extent, modulated by their surrounding cellular milieu, called the “LPC niche”. The LPC niche comprises hepatic stellate cells (HSCs), macrophages and other immune cells, endothelial cells, and extracellular matrix (ECM) (37–39). Based on their adjacent distributions, HSCs (31, 40, 41) and macrophages (31, 42–44) can regulate parenchymal infiltration and differentiation orientation of LPC by intimate cellular crosstalk. In addition, ECM composition and remodeling during liver injury were also found to be crucial to LPC expansion and activation (45, 46). Though the contribution of LPCs for liver repair has been confirmed in several disease models, the cellular source of LPCs during liver repair is still elusive. LPCs originating from pre-existing hepatic stem cells is one of the hypotheses; however, compelling evidence, i.e., exclusive molecular markers is lacking (47). Moreover, the contribution of LPCs to the regenerating biliary pool differs according to the liver injury model and lineage tracing methods utilized ( Table 1 ).

Another theory with nonnegligible evidence is that hepatocytes and cholangiocytes actually serve as facultative progenitor cells with reprogramming capacity and can transdifferentiate into each other. The intermediate phase of cells undergoing this process expresses both hepatocyte and cholangiocyte phenotypes and might be interpreted as LPCs, which seem inappropriate as they are not independent cell clusters and harbor considerable heterogeneity within the cluster (27, 47, 64). The more moderate theory is that hepatocytes or cholangiocytes first de-differentiate to LPCs and then differentiate to each other (65, 66). Each postulation obtains proponents and opponents, and in this review, we focus on those relevant to biliary repair.

Several studies failed to observe cell plasticity of lineage-tagged hepatocytes in the condition of biliary injury caused by bile duct ligation (28) or 3,5-diethoxycarbonyl-1,4-dihydrocollidine (DDC) diet model (23). Conversely, other researchers have confirmed HCT in vivo in rodent biliary injury models by utilizing the lineage-tracing technique. This HCT process seems to occur without LPC involvement. However, the proportion of HCT-derived cholangiocytes varied extensively in different models ( Table 1 ). Transplanting labeled hepatocytes into the liver of recipients can also repopulate the biliary pool through HCT after biliary damage, whereas its contributions to the biliary pool remain varied ( Table 1 ). During acute biliary injury, almost all hepatocytes are capable of HCT (67), while a subpopulation of Sry HMG box protein 9 (SOX9)-positive hepatocytes was identified as most potent in biliary reprogramming because they comprise 3% of all hepatocytes but contribute to nearly one-third of HCT (60, 68). These results correspond with clinical findings that cells coexpressing hepatocyte and cholangiocyte markers exist in liver samples from patients with cholangiopathies (69).

Although HCT derived cholangiocytes can acquire a genetic phenotype similar to that of bona fide cholangiocytes, it remains inconclusive whether these cells have sufficient biliary function. It was reported that cholangiocytes of HCT origin could obtain featured cellular polarity (positive staining of apical and basolateral markers of cholangiocytes, including primary cilia) (27) and even form mature bile ducts capable of bile drainage when the biliary system develops abnormally (70). In contrast, some studies found that HCT-derived cholangiocytes are functionally immature, and unable to coalesce with the pre-existing bile duct. These HCT-derived cholangiocytes might be a transient form that will revert to hepatocytes after injury subsides (23, 56, 59, 71). The possibility that these maladaptive cells contribute to CCA formation was also raised (9, 72).

In general, the role of HCT in biliary repair is not yet fully understood. It seems to be a double-edged sword that can respond to biliary damage and rescue the biliary system by replenishing the biliary pool. However, it might not be sufficient to form a mature bile duct to counterbalance compromised biliary proliferative capacity and function, even giving rise to carcinogenesis.

Except for the processes mentioned above, there are some less common sources of biliary repair. Some progenitor cells reside in the peribiliary gland in the extrahepatic biliary tree (73). They resemble LPCs in many ways, such as similar genetic phenotypes and are likewise capable of bipotent differentiation to hepatocytes and cholangiocytes. In comparison, they are primarily activated when large intrahepatic or extrahepatic bile ducts are affected (74–76). Moreover, it was reported that bone marrow stem cells can be recruited to the liver during several liver diseases and can differentiate into LPCs and cholangiocytes (77–79). Compared to the abovementioned process, these approaches may play a less important role in biliary repair. However, they remain worthy of further study as they, especially bone marrow stem cells, serve as a potential cellular therapy for liver disease (80).

Macrophages are the most abundant immune cells in the liver and comprise liver-resident Kupffer cells located in the space of Disse and macrophages derived from monocytes recruited from peripheral circulation (97). Macrophages are functionally heterogenous and are traditionally categorized into the proinflammatory “M1” subtype and the anti-inflammatory “M2” subtype (98). However, accumulating evidence indicates that macrophages have a more complicated phenotype spectrum, as they perform manifold or even opposing functions and express both M1 and M2 markers concurrently during liver injury (99, 100). Macrophages generally play a dual role in the pathogenesis of cholangiopathies; for example, during cholestasis, macrophages can either be activated to release proinflammatory factors and promote liver fibrosis (101), or perform an enhanced anti-inflammatory response mediated by the bile acid signaling pathway (102, 103). The balance between the proinflammatory and anti-inflammatory effects of macrophages was shown to be associated with the severity of cholestasis diseases and played a crucial role in biliary repair (104).

In cholangiopathies, CC-chemokine receptor 2 (CCR2) positive macrophages can be recruited to the periductular niche by CC-chemokine ligand 2 (CCL2) and other chemokines released from injured cholangiocytes (90). These peribiliary macrophages can provoke cholangiocyte proliferation by inducing the expression of Integrin αvβ6 (ITGB6) on cholangiocytes (105). Additionally, macrophages can eliminate senescent cholangiocytes and cell debris by phagocytosis favoring cholangiocyte proliferation (106). In addition, surrounding macrophages can incite expansion of LPC by secreting TNF-like weak inducer of apoptosis (TWEAK) even in the absence of liver injury. Depletion of CD11b⁺ macrophages can ameliorate this effect (107). After LPC expansion, macrophages can further regulate the lineage specification of LPC by paracrine signaling (31).

Using single-cell RNA sequencing (scRNA-seq), functionally distinct subpopulations of macrophages were discovered, and the composition of the macrophage population changes dynamically during liver diseases (108). For example, TREM2⁺CD9⁺ profibrogenic macrophages expand in liver fibrosis and induce HSC proliferation and activation (109). MACRO⁺ and MACRO^- macrophages were also found responding differently to inflammatory stimulations (110). Immune-related genes expression of macrophages are significantly altered in the DDC mouse model, and macrophage interactions with other liver cells are largely inhibited (111). However, contributions of heterogenous macrophages to biliary repair during cholangiopathies are largely unknown at the single-cell level. Studies further dissecting the specific role of different subsets of macrophages in biliary repair are warranted.

Neutrophils respond swiftly to liver injury and are the first immune cells recruited to the injury sites. Neutrophils are considered proinflammatory cells and are responsible for exacerbating tissue damage (112). The infiltration of neutrophils is ubiquitous in liver injury and is involved in many cholangiopathies (97). For example, neutrophils can directly interact with cholangiocytes via integrin β1 and elicit cholestasis (113). Neutrophils were also indicated as central immune cells in the pathogenesis of immune-mediated cholangiopathies such as primary biliary cholangitis (PBC) and primary sclerosing cholangitis (PSC) (114). Recently their contribution to tissue repair was also discovered. Neutrophils can clear the injury site by phagocytosis, degranulation, and neutrophil extracellular traps to create an appropriate milieu for tissue reconstruction (115). Moreover, monocytes can be recruited by neutrophil-released IL-37 and help tissue repair (116). Nevertheless, their specific role in biliary repair has scarcely been studied, and further research is needed.

Mast cells, a kind of inflammatory cell capable of releasing histamine and other particles, are emerging as novel participants in cholangiopathies (117). The infiltration of mast cells was detected in close proximity to cholangiocytes during cholestasis, PSC, PBC, and CCA (118–120). In vivo studies using antagonists and gene knockout mouse models further revealed a mast cell-mediated histamine-tissue growth factor-β (TGF-β)/stem cell factor (SCF) axis in the pathogenesis of PSC and CCA (117, 121). Concerning biliary repair, mast cells mainly regulate cholangiocyte proliferation. Coculture of cholangiocytes with mast cells isolated from bile duct ligated (BDL) rats significantly increased cholangiocyte proliferation via mast cell-derived histamine (118). Moreover, in vivo H1/H2 histamine receptor blockage dampened the proliferation of small cholangiocytes and large cholangiocytes, respectively (119). Whether other mechanisms of mast cells also contribute to biliary repair remains unclear and needs more studies.

Lymphocytes encompass many cell types, including T lymphocytes, B lymphocytes, natural killer cells, etc. ScRNA-seq of the human liver revealed distinct transcriptomic profiles of intrahepatic lymphocytes compared to peripheral lymphocytes (109). In the condition of liver cirrhosis, phenotype shift was observed in T cells and natural killer cells, such as expansion of a SELL⁺CD4⁺ memory T cell population and loss of CD56⁺CD16⁻ natural killer cells population (109, 122). Moreover, significant T cell heterogeneity was observed in liver of CCA and PSC patients and associated with disease pathogenesis and prognosis (123, 124). In contrast, there was no significant phenotypic changes of intrahepatic B cells in cirrhotic livers (109).

Nevertheless, the effect of these lymphocytes on biliary repair remains largely unknown. Innate lymphoid cells (ILCs), a newly defined, less enriched lymphocyte population located on the mucosal surface, are involved in regulating cholangiocyte proliferation. IL-33 can trigger cholangiocyte proliferation in a mouse model of biliary atresia, and type 2 ILCs (ILC2s) and ILC2-secreted IL-13 were found downstream of IL-33 in this mitogenic effect, which was abolished in the absence of ILC2s (95). IL-13 was also reported to be potent to drive LPC expansion and differentiation toward biliary fate (125). ScRNA-sec further revealed that two CD45⁺ ILC2 subsets, canonical ILC2, and newly defined “biliary immature myeloid cells”, interaction played an important role in IL-33 induced cholangiocyte proliferation (126). However, 10 weeks of IL-33 injection resulted in biliary tumorigenesis in mice genetically predisposed to CCA (95). It seems that the effect of lymphocytes on biliary repair is less well-defined.

HSCs, the primary producer of hepatic ECM, are also closely related to biliary repair during cholangiopathies (127). Therefore, we included HSCs and introduced their role in biliary repair. HSCs and infiltrated immune cells together constitute the LPC niche (39). Likewise, HSCs can be activated and transformed to collagen-depositing myofibroblasts through immune cells-mediated signaling pathways, such as those involving macrophage-derived TGF-β and platelet-derived growth factor (PDGF) (127). Myofibroblast activation and ECM deposition are essential for LPC expansion. Myofibroblast-secreted Jagged1 could promote LPC differentiation toward the biliary lineage in the condition of biliary injury (31, 41, 128). Moreover, myofibroblasts can also regulate HCT by interacting with laminin and ITGB6 (61). Except for the abovementioned functions, HSCs were even reported to gain progenitor features and function as bipotent LPCs to repopulate the liver parenchyma after partial hepatectomy (PHx) (129, 130). By utilizing scRNA-seq, functional zonation patterns and heterogeneous subpopulations of HSCs were discovered (131). However, its critical role in biliary repair remains to be elucidated.

The role of different immune cells in biliary repair is summarized in Figure 2 . Notably, immune cells are the key participants in the inflammatory response and have a double-sided effect on tissue repair. RDCs are highly proliferative to refurbish damaged biliary trees. Recruited immune cells could perpetuate the proliferation of RDCs, whereas protracted inflammation and excessive immune cell infiltration disturb biliary function and induce cell death (132, 133), ultimately leading to ductopenia caused by overbalanced apoptosis (134, 135). Similarly, cholangiocyte senescence, a mechanism to avoid neoplasia by irreversible cell cycle arrest, also has a secretory phenotype comparable to RDCs (5). Although senescence means decreased proliferation, mild senescence can provoke tissue repair by other mechanisms, including secretion of growth factors and proteins facilitating the expansion of LPCs and recruitment of immune cells regulating tissue repair (136, 137). Nonetheless, preserving senescence leads to tissue damage and impaired tissue repair (137). The immune cells show a double-edged sword role in biliary repair, and further studies that focus on the crosstalk of immune cells and cholangiocytes are urgently needed. As the heterogeneity of immune cells exists during biliary repair, it is also crucial to clarify the specific properties and factors secreted by these immune cells.

PBC and PSC are two major chronic cholestatic diseases caused by disordered immune conditions (82). Putatively, the pathogenesis of PSC is associated with the detrimental effect caused by the crosstalk of active cholangiocytes with innate and adaptive immune systems, which are activated by proinflammatory mediators originating from gut microbiota (138). Toxic bile acid was also proposed to contribute to PSC development (138). For PBC, a breach of biliary immune tolerance of the E2 subunit of the pyruvate dehydrogenase complex and an aberrantly activated immune response were postulated to play a central role in PBC pathogenesis (139). Despite these hypotheses, the pathogenesis of the two diseases remains incompletely understood, and effective treatment is lacking. Studies concerning biliary repair in these diseases are warranted to facilitate understanding of disease development and provide potential therapeutic targets.

Inherited cholangiopathies are a subset of cholangiopathies caused by hereditary or genetic factors, including Alagille syndrome (ALGS), polycystic liver diseases (PLD), biliary atresia (BA), etc. Inherited cholangiopathies are rare but can lead to chronic and progressive damage to cholangiocytes (10). The pathogenesis of inherited cholangiopathies remains elusive and effective treatment is limited. Thus, understanding the mechanism and regulation of biliary repair during inherited cholangiopathies is of great significance in expanding our knowledge of these diseases and searching for potential therapeutic targets.

Cholestasis induced by obstructive cholangiopathies is one of the important causes of chronic liver disease and cirrhosis. A significant role of LPC differentiation and HCT in biliary repair has been observed in animal cholestatic models.

CCA is a biliary epithelial tumor categorized according to anatomical location as intrahepatic cholangiocarcinoma (iCCA) and extrahepatic cholangiocarcinoma (ECC). CCA is the second most frequent type of hepatic malignancy after hepatocellular carcinoma (HCC), accounting for 10% to 20% of newly diagnosed primary liver cancer cases (177). Abnormal biliary repair and chronic inflammation are known to be involved in the development of CCA.

In many studies, conclusions have been drawn utilizing the lineage tracing technique by labeling specific cell subsets, which enables researchers to distinguish the hepatocyte or biliary origin of regenerating cholangiocytes. The lineage tracing technique is a powerful tool to trace the progeny of certain cell types and reveal cell plasticity. However, one common drawback exists: the lack of exclusive cell biomarkers. For example, the most frequently adopted biomarkers of LPC, such as SOX9, EpCAM, and osteopontin, are shared by cholangiocytes or other liver cell types. Therefore, they are insufficient to detect LPC identity and might bias the conclusions, particularly in research regarding biliary repair (38). The identification of more specific and sensitive makers will benefit the understanding of liver regeneration. A feasible resolution is the application of a dual-recombinase system. Combining two separate recombinase systems, Dre-rox and Cre-lox, could effectively avoid non-specific labeling in lineage tracing (194). One study intercrossed CK19-Dre mice with Alb-CreER mice and yielded CK19-Dre/Alb-CreER/IR1 mice which showed significantly increased labeling accuracy of hepatocytes and cholangiocytes and further confirmed the appearance of HCT in DDC and BDL models (55). Another study concerning biliary repair designed a different labeling strategy using a dual-recombinase system. They labeled SOX9⁺ hepatocytes by SOX9-CreER; HNF4a-DreER mice and found that hepatocytes highly expressing SOX9 were the most vigorous origin of HCT during BDL and DDC induced biliary injury (60).

Due to varied biliary injuries, pathological changes, and profound cell heterogeneity, cholangiocytes respond diversely to mitogenic mediators, and different repair processes are present in cholangiopathies. Additionally, LPC, hepatocytes, and liver resident or infiltrated immune cells also present varied subtypes and contribute differentially to biliary repair. However, traditional experimental tools such as immunostaining or RNA-sequencing are insufficient to reveal cell heterogeneity. In this case, scRNA-seq is a powerful technique to unmask transcriptional differences at the single-cell level. Utilizing scRNA-seq, researchers further confirmed heterogeneity within cholangiocytes and hepatocytes during cholangiocyte proliferation and HCT induced by biliary injury. It is identified that YAP enriched in a subset of liver epithelial cells dynamically drives liver regeneration (195). It is also suggested that some broadly-used biliary markers such as CK19 were inconsistently even low expressed in cholangiocytes. This might account for the contradictory results of studies applying these markers as biliary labels in lineage tracing ( Table 1 ). Hepatic immune cells, as crucial regulators of biliary repair, also present significant heterogeneity (196, 197). scRNA-seq and more advanced techniques such as single-cell lineage tracing combined with single-cell transcriptomics (198), single-cell chromosome immunoprecipitation-sequencing, and single-cell assays for transposase-accessible chromatin sequencing are waiting for further application to dissect the role of different subsets of cells involved in biliary repair and their specific regulatory mechanisms.

The liver organoid model might also serve as a novel therapeutic method and experimental tool for liver injury and repair. Organoids are typically defined as in vitro three-dimensional organ-like architectures derived from isolated stem/progenitor cells and in many aspects superior to traditional two-dimensional culture systems (199). Liver organoids were reported to contain hepatocytes, cholangiocytes and bile ducts of normal function, which enable better recapitulation of liver disease and liver regeneration (200). Ductal organoids and hepatocyte organoids could be generated using a culture medium supplemented with different growth factors and could help to enhance the understanding of cholangiocyte proliferation and HCT (201). Moreover, liver organoids are currently deemed potential candidates for transplantation therapy for liver disease. Transplantation of ductal cell-derived and primary hepatocyte-derived liver organoids yielded promising results as they perform considerable repopulation capacity of liver parenchyma, including biliary tree, and physiological functions, such as albumin secretion (202–205). Notwithstanding, the possibility that transplantation therapy exacerbates liver disease by activating the LPC niche and raising cancer risk has not been fully ruled out (206). Thus, there is still a long way to go before the clinical application of cell transplantation therapy for liver diseases.