In our last few liver articles, we highlighted the urgent need for new models and treatments for non-alcoholic steatohepatitis (NASH), and looked at the potential for human liver organoids to not only model disease, but also to test new drugs for safety and efficacy.
Here, we dive into a recent study that identified novel potential mediators of liver fibrosis using a genome-wide CRISPR screen in primary human liver cells. Studies like this are a starting point towards understanding the mechanisms underlying liver fibrosis and will hopefully help to reveal new drug targets for NASH and other forms of liver fibrosis.
What causes liver fibrosis? What we know so far…
Liver fibrosis is an advanced stage of chronic liver disease that results from the progressive build-up of extracellular matrix (ECM) proteins in the liver. This eventually alters the physiological architecture of the liver, rendering it dysfunctional.
Underlying conditions such as chronic hepatitis B/C virus infection, rare liver diseases, untreated or advanced diabetes, alcohol abuse, and NASH can all lead to damaged hepatocytes and infiltration of immune cells to the liver. Hepatic stellate cells (HSCs) are quiescent resident-liver fibroblasts by nature, but activation in response to damage and immune cell infiltration to the liver causes them to express an ECM-producing phenotype and secrete collagen.
When activated, e.g., upon liver injury, HSCs play an important role in generating a temporary scar at the injury site to protect the organ from further damage. However, unless counteracted by naturally-occurring anti-fibrotic mechanisms, e.g., HSC inactivation and apoptosis of ECM-secreting cells, activated HSCs can enter “overdrive” mode, producing excess ECM proteins. This is exactly what happens during chronic liver disease, whereby prolonged and repeated HSC activation causes liver fibrosis.
As of September 2023, there is not a single approved anti-fibrotic therapy that can directly treat hepatic fibrosis, as the standard of care is primarily to try and target the underlying cause. For advanced liver fibrosis, liver transplantation is currently the only broadly accepted lifesaving option, but a shortage of donors, the long-term requirement for immunosuppressants to prevent transplant rejection and the high cost of transplantation limit the feasibility of this approach for most patients.
TGF-β is an important driver of fibrosis, but maybe not the best drug target
On a molecular level, the major regulatory cytokine TGF-β1 is central to the ECM elaboration seen in liver and other types of fibrosis. TGF-β is produced in an inactive precursor state by non-parenchymal liver cells, e.g., the HSCs, liver sinusoidal endothelial cells, Kupffer cells, and others. Upon cleavage, TGF-β matures and becomes functional by certain furin-like proteases. Otherwise, it remains latent through its association with a certain protein complex until it becomes activated by specific proteases or the matricellular protein thrombospondin 1 (reviewed in 1). When active, TGF-β induces the HSC activation mentioned above
Given its increasingly apparent role in fibrogenesis, TGF-β has gained much attention as a therapeutic target for fibrosis, as well as several cancer types.
More than 20 candidate therapies targeting TGF-β have entered clinical trials in recent years, although these were primarily for cancers (reviewed in 2, and references therein). However, and despite its attractiveness as a drug target, clinical data for TGF-β1 inhibition strategies has been disappointing overall, and further clinical development has been slowed down by toxicities arising from the systemic effects and functions of TGF-β, as well as the complexity of fibrosis itself.
A deeper understanding of the systemic effects of TGF-β and exactly how it contributes to HSC regulation may reveal better therapeutics targets for liver fibrosis than TGF-β itself.
Using a CRISPR-Cas9 screen to look for new regulators of fibrosis in human liver cells
To try and unravel the multi-faceted role of TGF-β in HSC-mediated liver fibrogenesis and to find new therapeutic targets for liver fibrosis, researchers at Takeda Pharmaceutical and HemoShear Therapeutics performed a genome-wide CRISPR-Cas9 knockout screen in primary human HSCs (3). The screen was arrayed, i.e., each gene disruption was introduced separately, in this case in single wells of a 384-well plate.
When induced by TGF-β, activated HCSs display a number of altered phenotypes, including a contractile and migratory phenotype via expression of actin alpha 2 (ACTA2, also known as alpha-smooth muscle actin). Based on this, the team used ACTA2 protein expression as a surrogate readout for HSC activation.
The CRISPR screen setup
Primary HSCs were added as single-cell suspensions to a 384-well plate already containing CRISPR-Cas9 ribonucleprotein complexes and a gRNA library designed to target the entire annotated human genome covering 19,027 genes.
The cells were electroporated to take up the CRISPR reagents, and incubated for 48 hours to allow for CRISPR editing.
The cells were then stimulated with or without exogenously added TGF-β for a further 48 hours. This was done to ensure a robust TGF-β response, as verified by upregulated ACTA expression in preparatory experiments.The cells were then fixed and stained with an anti-ACTA2 antibody. A secondary fluorophore-labeled antibody was added to allow for imaging of the ACTA2-expressing cells.
Genes that, upon CRISPR disruption, led to 75% or more inhibition of ACTA2 protein expression were considered as hits in the primary screen.
Any hits identified in the primary screen were tested in a follow-up confirmatory screen using the setup described above, as well as a counter screening viability assay. Any confirmed hits at this point were tested in further validation assays.
ACTA2 expression was determined by immunofluorescence staining and high content imaging. Results were expressed as % inhibition of each gene disruption on ACTA2 expression, whereby the negative and positive controls were treated with a non-functional gRNA (0 % inhibition) and a pool of four ACTA2-targeting gRNAs (100 % inhibition), respectively.
CRISPR screen reveals proteins with druggable properties
The primary screen revealed 372 genes hits. Some of those genes are known to be involved in TGF-β signaling, e.g., TGFBR1, SMAD3, and SMAD4, which meant that the screen could pick up biologically-relevant hits.
Following an ACTA2 downregulation assay to confirm the phenotypes of disrupting the primary hits, as well as cell viability assays to assess toxic effects of disrupting any of those genes, the team shortlisted the hits to 52 genes, whose disruption inhibited ACTA2 expression by more than 50 %, without reducing cell viability by more than 30 %. When further categorized based on potential modes of action, these genes fell into 11 groups, including proteasome and other enzyme subunits, transcriptional and translational and post-transcriptional and translational modifiers, ribosome protein-like proteins, proteins in the TGF-β pathway, and others.
Importantly, several of the shortlisted hits are described in the literature as playing a role in the pathobiology of liver fibrosis, e.g., tropomyosin-1 (TPM1) is reported to be correlated with elevated α-SMA levels during liver injury in animal models of fibrosis as well as human cirrhotic livers. Furthermore, the proteins encoded by some of the hits exhibited druggable properties.
To understand the clinical potential of the shortlisted hits, the researchers knocked out approx. 20 of the hits individually in HSCs, and assessed the impact on mRNA levels of 14 biomarkers associated with liver fibrosis. Small molecules that have been explored experimentally or clinically for the treatment of fibrosis were included as positive controls. The biomarker panel included ECM proteins, enzymes involved in fibrogenesis, and TGF-β signaling mediators. They found that knocking out any of the 20 genes individually had a significant impact on ACTA2 expression, but a varied impact on the mRNA levels of the biomarker panel. Of the panel genes, TRAPPC11-, PAFAH1B1-, TGFBR1- and PSMC2-knockout had the most widespread effect on the biomarker panel, with significant reduction in the expression of 11, 13, 12 and 11 genes, respectively.
Probing the hits in a liver-like environment – the main findings
To probe if and how the hits contribute to liver fibrogenesis, the researchers set up a co-culture system to mimic the liver environment. Briefly, they cultured primary human HSCs and hepatocytes, as well as macrophages, in an organotypic liver model. Conditioned culture media was added to induce healthy, fibrotic or NASH states and expression of all hit genes was assessed by RNA sequencing.
Here, we briefly summarize the main findings from those experiments:
Several of the genes were upregulated in HSCs and macrophages in the NASH model, and at least two of those genes (TPM1 and TGFBR1) were also significantly upregulated in public human transcriptome data from primary TGF-β-activated HSCs. These findings suggest a possible link between these hits and NASH pathogenesis and fibrosis, although further studies are needed to understand exactly how these genes might be implicated.
Although not approved to treat liver fibrosis, statins have been linked with reduced liver fibrosis prevalence in type 2 diabetics. The researchers identified the HMGCR gene (which encodes an enzyme target of statins) as a hit in their study, and suggest it might be a relevant new therapeutic target for liver fibrosis.
Several genes reported in the literature as therapeutic targets for liver fibrosis were not picked up in the current study. The researchers present possible explanations for this, including possible essential functions for those genes, or that those genes influence fibrosis independently of TGF-β signaling. These discrepancies highlight the importance of including multiple readouts to reduce screening bias, and the need for thorough follow-up and validation of any identified hits.
In summary, the researchers have developed a novel cell-based CRISPR screening method in a culture environment that aims to mimic human liver-relevant physiological conditions. The screen was capable of identifying targets previously described in the context of liver fibrogenesis, as well as showcasing some new targets.
We would like to note that a number of CRISPR screens to identify new disease targets have been undertaken to date. While these appear to be a reasonable starting point for target discovery, we welcome follow-up studies that rigorously test and validate hits in disease-relevant human cell models that recapitulate liver disease phenotypes, e.g., reproducible and scalable hiPSC-derived cells that provide a uniform human baseline.
hiPSC cells allow for meaningful interlaboratory comparisons and paired evaluations of “healthy vs. patient” cells during drug development, and are thus critical for pre-clinical drug development. For instance, with Tempo’s collection of iPSC-derived human liver cells, including sinusoidal endothelial cells, Kupffer cells, hepatic stellate cells and hepatocytes, it is possible to culture 3D liver organoids that are validated to model liver NASH, NAFLD, liver fibrosis, cirrhosis, and hepatocellular carcinoma.
Murphy-Ullrich JE, Suto MJ. Thrombospondin-1 regulation of latent TGF-β activation: A therapeutic target for fibrotic disease. Matrix Biol. 2018 Aug;68-69:28-43. doi: 10.1016/j.matbio.2017.12.009. Epub 2017 Dec 27. PMID: 29288716; PMCID: PMC6015530.
Peng D, Fu M, Wang M, Wei Y, Wei X. Targeting TGF-β signal transduction for fibrosis and cancer therapy. Mol Cancer. 2022 Apr 23;21(1):104. doi: 10.1186/s12943-022-01569-x. PMID: 35461253; PMCID: PMC9033932.
Yu S, Ericson M, Fanjul A, Erion DM, Paraskevopoulou M, Smith EN, Cole B, Feaver R, Holub C, Gavva N, Horman SR, Huang J. Genome-wide CRISPR Screening to Identify Drivers of TGF-β-Induced Liver Fibrosis in Human Hepatic Stellate Cells. ACS Chem Biol. 2022 Apr 15;17(4):918-929. doi: 10.1021/acschembio.2c00006.
Karen O’Hanlon Cohrt is an independent Science Writer with a PhD in biotechnology from Maynooth University, Ireland (2011). After her PhD, Karen relocated to Denmark where she held postdoctoral positions in mycology and later in human cell cycle regulation, before moving to the world of drug discovery. Karen has been a full-time science writer since 2017, and has since then held numerous contract roles in science communication and editing spanning diverse topics including diagnostics, molecular biology, and gene therapy. Her broad research background provides the technical know-how to support scientists in diverse areas, and this in combination with her passion for learning helps her to keep abreast of exciting research developments as they unfold. Karen is currently based in Ireland, and you can follow her on Linkedin here.