Hidden Power: PAR4’s Role in Liver Damage from Acetaminophen Overdose in Mice

In this issue of ATVB, Rajala and colleagues¹ have identified the roles for both protease activated receptor 1 and 4 (PAR1 and PAR4) in the context of acetaminophen-induced liver injury. They also resolve the long-standing questions of PAR4 distribution, expression level and function in endothelial cells in vivo. These are the first studies to show that PAR4 has a critical role in liver endothelial cells despite its low expression levels.

Vascular endothelial cell (EC) dysfunction is a major contributor to pathological insults in the liver because they regulate tissue structure, permeability, and control inflammation. This is exemplified during acetaminophen (APAP) overdose, the leading cause of acute liver injury in the United States.² APAP overdose is characterized by hepatic necrosis, bleeding, and congestion. During APAP toxicity, there is a significant increase in thrombin generation which drives blood coagulation and inflammation through activation of protease activated receptors (PARs) on the endothelium.^3,4

PARs are G-protein coupled receptors with four family members (PAR1–4). PARs are activated by proteolytic cleavage of the N-terminus exposing a tethered ligand.⁵ PAR1 and PAR4 are both activated by thrombin, but have distinct activation and desensitization mechanisms.^6–8 PAR1 is expressed in a broad range of tissues, whereas PAR4 has more limited tissue distribution. PAR1 has well described roles in regulating endothelial cell permeability depending on the activating protease.⁵ Thrombin activation of PAR1 increases vascular permeability while cleavage by activated protein C (APC) is barrier protective.⁹ In contrast, the relevance of PAR4 signaling to EC function has remained elusive because of its low and varied expression on cultured endothelial cells.¹⁰

Here, Rajala and colleagues used APAP overdose as a model for studying the individual roles of PAR1 and PAR4 on endothelial cells.¹ Previously, the specific contributions of PAR1 and PAR4 on endothelial function have been difficult to to study due the lack of suitable tools. To overcome this obstacle, Rajala et al. generated mice with inducible deletion of PAR1, PAR4, or both in liver endothelial cells. These mice were used to investigate the effect of hepatic vascular permeability, red blood cell accumulation, thrombin generation, and liver function after APAP overdose.

The endothelial cell specific deletion of PAR1 or PAR4 decreased vascular permeability at late time points (24 hr). In contrast, deleting both PAR1 and PAR4 decreased permeability at both 6 hr and 24 hr. These data demonstrate that PAR1 and PAR4 have overlapping roles for vascular permeability in the liver during APAP overdose. These data also show that PAR4 can substitute for PAR1 in this context despite differences in activation and signaling kinetics. The new data presented by Rajala et al. significantly improve our understanding of the impact of PAR1 and PAR4 during APAP overdose. Previous studies using global knockout mouse models clearly demonstrated roles of PAR1 and PAR4. However, it was not possible to determine the specific contribution of hepatic endothelial PAR1 or PAR4 since both receptors are highly expressed on other cells and tissues. The ambiguity of the global knockout mouse models to study PAR4 functionality in different organs has contributed to the ongoing controversy of its function in endothelial cells. We are now beginning to gain clarity on the specific roles of these related proteins in vivo.

Translating Ribosome Affinity Purification and next-generation sequencing (TRAPseq) takes advantage of the RiboTag mice to label actively translating genes. In this model, the ribosomal protein RPL22 is labeled with an HA-tag in a cell-specific manner by selectively crossing the mice with the appropriate Cre-driver. The HA-tag is used to isolate the RNA that is being actively translated. There are two important insights that are gleaned from these experiments. First, it quantified the expression of PAR1 and PAR4 in the liver endothelial cells. PAR4 was expressed at a ratio of 1:153 compared to PAR1. Second, it uncovered genes that are downstream of PAR1 and PAR4 and revealed differences between the two. This is somewhat surprising given the functional redundancy when using vascular permeability as a read out, but highlights the mechanistic complexities this methodology can begin to address. For example, future studies will be able to examine the downstream consequences resulting from activation of PARs by proteases other than thrombin.

The beauty of the approach taken by Rajala and colleagues is the specificity of monitoring the expression of genes in select cell types in vivo using TRAPseq. Following their example, this platform can be used to interrogate the roles of genes with low expression in complex tissues. This approach opens the door for examining molecular mechanisms of disease using in vivo models at a level of detail previously not possible.