Suppressed AR formation contributes to failure of small-for-size liver grafts
Insufficient liver volume is the major cause of small-for-size graft syndrome which decreases survival after PLT (5,31,32). Therefore, major transplant centers reject potential donors if estimated graft weight/recipient body weight is less than 0.8–1.0 or graft weight/recipient standard liver weight <30%–40% (32). Dual graft LT (two liver grafts from two donors into one recipient) was developed to increase the graft mass (33); however, it puts two donors at risk. A recent attempt was made to avoid small-for-size syndrome in living donor LT by increasing the donor's liver mass before liver donation by eating a high-protein and high-carbohydrate diet for 3 months to increase calorie intake (34). Whether this approach increases the risk of steatosis and diabetes remains to be explored.
The mechanisms of small-for-size graft failure remain unclear. Alterations in hemodynamics, sheer stress, microcirculatory disturbances and increased reactive oxygen species (ROS) formation in small-for-size grafts may contribute to the grafts injury (11,35,36). Mitochondrial dysfunction occurs in small-for-size grafts, compromising energy supply (6,10). STY720, which internalizes and downregulates sphingosine-1-phosphate receptors, decreased injury after PLT (37).
The liver has a robust capacity to regenerate under most conditions which allows rapid recovery of liver mass and function after liver injury or loss of liver mass (e.g. liver resection). However, liver regeneration is suppressed in small-for-size grafts (6,9,38) which may contribute to graft dysfunction. Consistent with previous findings, we observed a markedly decreased liver regeneration in TSG which was associated poorer liver function and survival (Figures 3 and 4).
Liver regeneration is tightly regulated by variety of transcription factors, cytokines and growth factors (13,15). In the early stage after PHX, a transcriptional shift occurs, which consists of activation of genes and transcriptional factors, and these massive changes in immediate early gene expression prime quiescent hepatocytes, making them responsive to growth factors, such as HGF and ligands of EGFR (13,15,39). Growth factors stimulate cell cycle progression by two major growth factor signaling systems: EGFR ligands/receptors/coreceptors and HGF/c-Met (12–15). Cytokines (e.g. TNFα and IL-6) activate transcription factors and also provide early signals triggering regeneration (12,40). Inhibition of transcription factor activation or decreased synthesis of promitogenic cytokines and growth factors can decrease regeneration.
Previously we found that HGF, TNFα and IL-6 increased after rat PLT and peaked at about 5 h, an effect that was greater in QSG than in FSG and HSG (6). In this study, we confirmed that HGF, TNFα and IL-6 mRNAs increase to a greater extent in TSG (Figure 2). Therefore, inhibited regeneration in small-for-size grafts was unlikely due to deficiency of pro-regenerative HGF, TNFα and IL-6 formation.
In this study, we further investigated the role of EGFR ligands AR, HB-EGF, EGF and TGFα in suppressed regeneration of small-for-size liver grafts. After transplantation, AR increased markedly in HSG, which regenerated rapidly, but remained at low in TSG which failed to regenerate (Figures 1, 3 and 4). AR neutralizing antibody blunted liver regeneration in HSG, whereas AR supplementation stimulated regeneration of TSG. Together, these findings are consistent with the conclusion that AR plays an important role in liver regeneration after PLT (Figure 4).
Decreased AR was associated with lower AR mRNA in TSG, suggesting suppressed synthesis of AR in TSG (Figure 1C). Signaling pathways that regulate AR formation include transcription factor WT1, protein kinases A and C, prostaglandin E2, Toll-like receptor-4, hypoxia-inducible factor-1α and ROS (41–43). However, we observed that Toll-like receptor-4 expression and ROS increased to a greater extent in TSG than in HSG (data not shown) (44). Therefore, Toll-like receptor-4 and ROS are unlikely the major factors contributing to suppressed AR formation. Future studies will be performed to investigate the mechanisms underlying suppressed AR expression in small-for-size grafts.
We also examined the possibility that AR was not released to act upon the EGFR. AR precursor is activated by a transmembrane metalloproteinase ADAM17 (20). ADAM17 was expressed at high levels in both HSG and TSG (Figure 1). Therefore, decreased liver regeneration is not due to ADAM17 deficiency.
In this model, it does not appear that EGF and HB-EGF play a role in suppressed TSG regeneration since the levels of neither growth factor increased significantly following PLT. It is also unlikely that TGFα plays a limiting role in suppressed TSG regeneration because TGFα levels increased equally following HSG and TSG. Together, these data are consistent with previous studies which demonstrated that removal of AR decreased liver regeneration whereas deletion of some other EGFR ligands did not (14,19), and the findings support the hypothesis that decreased AR synthesis is responsible, at least in part, for the suppressed regeneration of small-for-size liver grafts.
There are other examples of ligands that do not appear to be interchangeable. For example, TGFα is a strong mitogen for hepatocytes in culture and increased modestly after PHX (45,46). However, TGFα knock-out mice have essentially normal liver regeneration and liver embryonic development (47). In HB-EGF-deficient mice, liver regeneration decreased moderately in the early stage but increased in late stage after PHX (48).
EGFR mediates stimulation of small-for-size graft regeneration by AR
Treatment with RNAi against the EGFR decreased cell proliferation after PHX in rats (49), indicating that EGFR plays an essential role in regeneration. In the present study, EGFR activation was increased in HSG but not in TSG (Figure 6), consistent with the regenerative responses in HSG and the lack of one in TSG (Figures 3–4). Inhibition of EGFR by PD153035 significantly decreased liver regeneration in HSG (Figure 6), confirming the importance of EGFR in regeneration of partial liver grafts.
AR treatment increased EGFR phosphorylation in TSG to the levels of HSG, indicating that EGFR retains its responsiveness to AR stimulation (Figure 6). Thus, AR could be used in small-for-size transplantation to stimulate liver regeneration. AR exclusively binds and activates the EGFR but not other members of the EGFR family of receptors like ErbB4 or ErbB3, although it can promote receptor heterodimerization (ErbB-2 with ErbB-1) (43,50). However, ErbB-2 is mainly expressed in tumors. EGFR inhibition largely blocked AR-stimulated liver regeneration in TSG (Figure 6), suggesting that the promitotic effects of AR in small-for-size grafts are primarily mediated by EGFR. Whether other EGFR ligands have similar therapeutic effects for small-for-size grafts is beyond the scope of this study and will be investigated in the future.
Role of EGFR downstream signaling pathways in stimulation of small-for-size graft regeneration by AR
Ligand binding to EGFR activates a number of downstream signaling pathways (21,22). Activation of the PI3K pathway produces phosphatidylinositol (3,4,5)-P3 which activates different kinases such as phosphoinositide-dependent kinase1 (PDK1) and protein kinase B. PDK1-deficiency suppressed regeneration after PHX (51). Akt, an important downstream effector of PI3K, activates mTOR and p70S6 kinase which in turn regulates the 40S ribosomal protein S6 to control protein synthesis and cell proliferation. Inhibition of mTOR decreased DNA synthesis after PHX (52). Deletion of S6 protein led to a profound deficit in DNA replication and alterations in cyclin E induction after PHX (53).
In this study, activation of PI3K, mTOR and its downstream effector p70S6K all increased in HSG, not in TSG, and was restored with AR treatment (Figure 7). The PI3K/mTOR pathway is highly sensitive to nutrient/energy alterations (13,54). Small-for-size liver grafts face significantly higher metabolic challenges and need to meet the demands for significant nucleotide and protein synthesis for cell division. However, ATP production decreases substantially in small-for-size liver grafts (6). Therefore, compromised energy status may also contribute to decreased activity in the PI3K/mTOR pathway.
EGFR also leads to activation of ERK and JNK pathways (13,55–58). Both ERK and JNK mediate AR-induced hepatocyte proliferation (19). Thus, interruption of JNK and ERK signaling might be involved in inhibited regeneration of small partial liver grafts. The ERK kinases are activated by EGFR through G proteins in the Ras/Raf/ERK pathway. Activated ERK kinases regulate a variety of transcription factors thus controlling transcription of important cell cycle genes (59). Blockade of ERK activation suppressed synthesis of CyE and A as well as CDK2 activation (60).
Inhibition of JNK decreases expression of CyD1, a molecule that drives hepatocytes to enter the cell cycle (61), and suppresses hepatocyte mitosis after PHX (28). JNK activation is controlled by TNFα; however, TNFα expression is increased in small-for-size grafts (6). Therefore, inhibition of JNK activation is not due to lack of TNFα. EGFR also regulates JNK activation. Indeed, JNK and ERK phosphorylation and expression of cyclins were inhibited in TSG (Figures 7–8), which was associated with suppressed EGFR phosphorylation and liver regeneration. AR treatment recovered the activation of ERK and JNK as well (Figures 7 and 8).
It appears that multiple, redundant pathways mediate the proliferative effects of AR/EGFR in partial liver grafts. In our previous studies we also observed increased formation of TGFβ, an inhibitor of liver regeneration in small-for-size liver grafts (7). Suppression of the TGFβ signaling or stimulation of EGFR signaling both improved regeneration of small-for-size liver grafts. These results indicate that liver regeneration is controlled by a delicate balance of proregenerative and regeneration-inhibitory factors.
Taken together, we observed decreased AR formation and subsequent inhibited EGFR signaling in small-for-size liver grafts. This decreased AR formation is associated with suppressed liver regeneration and failure of small-for-size liver grafts. However, small-for-size grafts retain their responsiveness to AR stimulation. Therefore, AR supplementation could be a promising therapy for small-for-size syndrome. Of note, the therapeutic effect of AR was observed in a model in which the hepatic artery was not reconstructed. The importance of reconstruction of hepatic artery has been controversial. One study showed that hepatic artery reconstruction improved survival and decreased transaminase release after mouse LT (62). By contrast, other studies showed that at 1 and 100 days after mouse LT, no differences in survival, injury and immunologic responses were observed with or without rearterization (63). Transplantation of HSG without rearterization achieves high survival, as confirmed in the current study (64) (Figure 5). Moreover, regardless the absence of hepatic artery reconstruction, AR improved the outcome of small-for-size LT, indicating that its mitotic effect is independent of hepatic arterial blood supply.