Monocyte-derived KCs (MoKCs) contribute to the KC pool in NASH

Nonalcoholic fatty liver disease (NAFLD) is now recognized as the most common liver disease worldwide and has increasingly become a serious hazard to human health. NAFLD consists of a spectrum of disease states ranging from simple steatosis to more severe disease termed nonalcoholic steatohepatitis (NASH), fibrosis, cirrhosis, and even hepatocellular carcinoma.¹ In the United States, the number of NAFLD cases is projected to expand from 83.1 million in 2015 to 100.9 million in 2030, and the proportion of NASH will rise from 20 to 27% of adults with NAFLD during this interval.² Currently, there is no effective treatment for NASH, and liver transplantation is often the only option. Thus, further research is required to understand NASH pathogenesis and to design new therapeutic strategies. Macrophages are believed to play an essential inflammatory role in NASH.³ Kupffer cells (KCs), liver-resident macrophages that self-renew by proliferation independent of monocytes, are essential to preserve tissue homeostasis. Interestingly, embryo-derived KCs (EmKCs) can be replaced by monocyte-derived KCs (MoKCs) under certain circumstances, such as Listeria monocytogenes infection.^4,5 Whether KC maintenance is impaired during pathological processes such as NASH remains ill defined. In their latest publication in Immunity, Tran et al. highlight that EmKC self-renewal is impaired during NASH, leading to MoKC contributions to the KC pool. Moreover, MoKCs are maintained for a long time after NASH regression, exhibiting more inflammatory traits and abnormal responses to lipid overload.⁶

KCs are derived from the yolk sac and are finally located in hepatic sinusoids. KCs are a critical component of the mononuclear phagocytic system and are central to both the hepatic and systemic responses to pathogens. Moreover, KCs are emerging as critical mediators involved in both liver injury and repair.⁷ Depletion of monocytes and macrophages exactly attenuates the progression of NASH in animal models.⁸ Hepatic macrophages, specifically KCs, are believed to play crucial roles in NASH pathogenesis. However, KC biology in NASH remains poorly investigated. The relationship between inflammatory monocyte-derived macrophages (MoDMacs) and KCs is still unknown. An effective strategy to discriminate KCs from MoDMacs needs to be clarified.

Based on the data of published tissue-resident macrophage transcriptome analysis, Tran et al. first found that CLEC2, CD207, VSIG4 and TIMD4 were either very selectively and/or strongly expressed by KCs, among which CLEC2 was expressed highest on KCs, and KCs were defined as CD11b^loCLEC2^hi cells. Then, they generated a model of methionine-choline-deficient (MCD) diet-induced NASH. They found that KC numbers during NASH remained similar to the control condition, while MoDMacs (CLEC2^‒CD11b⁺CD64⁺) drastically increased. Next, they challenged Ccr2^−/− mice with the MCD diet. Surprisingly, KC counts decreased by half in this model. To validate the monocytic origin of KCs during NASH, they developed two models. The first is bone marrow (BM) chimeras in which CD45.2⁺ Ccr2^−/− recipients were transplanted with BM cells from congenic CD45.1⁺ donors followed by a 6-week MCD diet. The second is a parabiosis model in which CD45.2⁺ Ccr2^−/− mice were joined with congenic CD45.1⁺ mice. They observed that 30%~40% of KCs were CD45.1⁺ in the two models above, suggesting that NASH alters KC homeostasis and that circulating monocytes contribute to the KC pool during NASH.

To further explore MoKC generation, characteristics and long-term maintenance, Tran et al. utilized Ccr2^−/−-based chimeras and noticed that the proportion of MoKCs among CLEC2⁺ KCs gradually increased during the progression of NASH. In terms of classical KC markers, MoKCs did not express Clec4f, Cd163, or Vsig4 compared to EmKCs. In addition, they also demonstrated that MoKCs were more hyperproliferative than EmKCs. To address their maintenance, they induced a NASH regression model and found that MoKCs still represented 40%~50% of the KC pool after 6 weeks of regression. Simultaneously, MoKCs were almost all mature VSIG4⁺ cells, as observed in EmKCs. Thus, all of these hints underlined that MoKCs were maintained for a long time during NASH regression and gradually became mature to acquire the potential to self-renew. Additionally, Tran et al. found a marked increase in CLEC4F⁺ EmKC death during NASH compared to healthy conditions. To validate that EmKC death led to MoKC generation, they crossed Cd68-hBCL2 mice with Ccr2^−/− mice, which could specifically modulate EmKC but not monocyte or MoDMac death. A significant decrease in MoKC frequency/absolute numbers was observed in Cd68-hBCL2 × Ccr2^−/− recipients, while EmKCs were increased, demonstrating that EmKC death facilitated the generation of MoKCs.

To deeply study the functional differences between EmKCs and MoKCs, Cd207-DTR mice were introduced to specifically deplete EmKCs after diphtheria toxin (DT) injection compared to Cd207-DTR mice receiving boiled-inactivated DT (bDT). After a 6-week MCD diet, Tran et al. observed a decrease in hepatic triglyceride (TG) storage in DT-treated Cd207-DTR mice, while plasma alanine aminotransfease activity increased in these mice. Furthermore, they found that Scd1, the gene related to toxic saturated fatty acid (SFA) conversion, was markedly decreased in DT-treated Cd207-DTR mice. Meanwhile, several profibrotic mediators were elevated in the livers of DT-treated Cd207-DTR mice. Moreover, RNA-seq was performed to confirm this signature. Overall, they concluded that MoKCs displayed a distinct transcriptional landscape characterized by an increased inflammatory status, which altered the liver response to lipids during NASH. Finally, using Cd207^cre × Bcl2l1^flox/flox mice followed by a two-week MCD diet, Tran et al. confirmed that limiting EmKC survival and favoring the generation of MoKCs ultimately impacted hepatic lipid content and liver damage during NASH.

In conclusion, the present study by Tran et al. identified permanent changes in the KC pool during NASH, which has a potential long-lasting impact on liver homeostasis and function (Fig. 1). CLEC2 was the most useful and one of the earliest markers to selectively discriminate KCs from MoDMacs. There was increased endoplasmic reticulum stress in EmKCs during NASH, implying impaired self-renewal ability. Subsequently, MoKCs seized the opportunity to maintain the KC pool. MoKCs were generated in the course of NASH and maintained for a long time after NASH regression. MoKCs thus acquired EmKC characteristics in this period. In addition to their origin, EmKCs and MoKCs also showed quite different functions. Scd1, encoding an enzyme that inhibits the accumulation of toxic lipids, was decreased when MoKCs dominated. This finding indicated that MoKCs limited hepatic TG storage potential. RNA-seq demonstrated that MoKCs exhibited more proinflammatory and profibrotic traits. Above all, Tran et al. identified two heterogeneous groups of KCs with differential functions during NASH. Steatohepatitis had a major impact on the ability of EmKCs to self-maintain, which in turn generated MoKCs to supply the KC pool and impacted the liver response to lipid overload. This study successfully discriminated KC populations under NASH conditions. Moreover, Tran et al. clarified the KC origin and identified their roles in NASH progression, which would offer great help toward understanding NASH pathogenesis and developing appropriate treatment strategies. However, the key mechanism regulating EmKC death remains unclear. In addition, monocyte tracking should be performed to show the MoKC distribution after NASH regression.

Fig. 1.

Permanent changes in the KC pool during NASH progression. Defined as Clec4f⁺, Cd163⁺, Timd4⁺, and Vsig4⁺ cells, EmKCs take up the KC pool in healthy livers. Owing to their appropriate response to lipid and toxic SFA conversion ability, the liver maintains homeostasis. However, under steatohepatitis conditions, EmKCs might die from lipid-induced stress. Simultaneously, MoKCs contribute to the KC pool. In contrast to EmKCs, MoKCs poorly express Timd4 and CD163 and alter the liver response to lipid overload, exhibiting more hyperproliferative and inflammatory traits. Moreover, during steatohepatitis regression, MoKCs remain for a long time and represent 40–50% of the KC pool. With elevated expression of Vsig4 and Timd4, they acquire a more mature phenotype

Competing interests

The authors declare no competing interests.