Article Text

Download PDFPDF

In sero veritas: what serum markers teach us about HCV infection of primary human hepatocytes
  1. Thomas Pietschmann
  1. Correspondence to Professor Thomas Pietschmann, Institute of Experimental Virology, TWINCORE, Centre for Experimental and Clinical Infection Research, a Joint Venture of Hannover Medical School and the Helmholtz-Centre for Infection Research, Feodor-Lynen-Str. 7, Hannover 30625, Germany; thomas.pietschmann{at}

Statistics from

HCV has chronically infected an estimated number of 160 million individuals worldwide.1 In the course of 10–20 years of persistent infection, approximately 20% of these patients are at risk of developing severe liver disease, including fibrosis, cirrhosis and hepatocellular carcinoma.2 Besides the ability to establish chronic infection, the pronounced genetic variability of HCV and the diverse course of chronic infection ranging from relatively benign to exacerbating liver disease are hallmarks of hepatitis C. Moreover, response to classical antiviral treatment based on pegylated interferon alpha (PEG-IFN-α) and ribavirin is substantially influenced by viral and host factors, since viral genotypes 1 and 4 are more resilient to therapy compared with genotypes 2 and 3 and as polymorphisms in the vicinity of the IFN28B locus predict the chances to naturally clear HCV infection and of response to IFN-based therapies.3 ,4 The high degree of virus variability is also a formidable challenge in the emerging era of directly acting antivirals for treatment of hepatitis C by mediating drug escape through resistance mutations. Moreover, many molecules, most prominently the already licensed protease inhibitors telaprevir and boceprevir, display a genotype-dependent antiviral activity. Therefore, both viral and host genetic variability complicate provision of efficacious, safe, well-tolerated and cost-effective therapies to the large number of HCV patients.

At the same time, HCV has been resistant to disclosing its secrets as the road for development of suitable cell culture and animal models to dissect mechanism of virus replication and liver disease has been paved with many roadblocks. For instance, infection of cultured cells by primary patient-derived virus from sera of infected individuals has been notoriously difficult, thus precluding detailed assessment of viral strain-specific features. Although development of HCV subgenomic replicons that efficiently propagate in human hepatocarcinoma cells like Huh-7 and its subclones has been a major breakthrough, this system is only suitable to monitor HCV RNA translation and genome replication, because of the lack of viral structural genes needed for virus assembly and production of infectious progeny.5 While this gap has been closed more recently by the development of fully infectious systems based on the genotype 2a clone JFH1—and in the meantime, a few additional viral genomes—it is important to realise that these few viral constructs only incompletely reflect the large spectrum of genetic—and likely functional—diversity of HCV.6 Specifically, HCV strains are grouped into seven distinct viral genotypes with more than 30% variability between distinct genotypes illustrating the dramatic range of viral diversity. Not only this, currently available cell systems to dissect replication mechanisms are dominated by Huh-7-derived cell clones. However, these cells are poorly differentiated; they do not polarise and certainly do not mimic all features of primary human hepatocytes (PHHs). This is maybe most strikingly illustrated by the inability to produce highly infectious HCV from Huh-7 cells: likely due to the aberrant production of lipoproteins in Huh-7 cells, HCV particles, which tightly associate with human lipoproteins, originating from these cells are much less infectious than viruses derived from humanised mice repopulated with PHHs or from cell cultures of PHHs.7 ,8 Therefore, development of robust cell-based model systems for assessment of viral features of serum-derived HCV (HCVser) and involving PHHs is an important endeavour which holds great promise for dissecting individualised aspects of the virus–host interaction, including its implications for customised therapies. Unfortunately, infection efficiency of PHHs by HCV from human sera is low and the outcome of infection is highly variable. Therefore, and due to the difficulty in obtaining these scarce and precious primary cells, relatively limited work has been done in this most authentic cell-based HCV infection model.810

The report by Gondeau et al exactly this pressing need for more robust and predictable systems to investigate infection of primary liver cells by serum-resident HCV.11 To this end, they first systematically established the time point and conditions for optimal infection of PHHs by HCVser. In line with previous observations in HepG2 cells,12 they report that hepatocyte polarisation decreases PHH permissiveness to HCVser, to cell culture-derived HCV (HCVcc) as well as HCV pseudo-particles (HCVpp). Since susceptibility for HCVpp, that is, retroviral particles carrying HCV envelope proteins and thus monitoring solely the cell entry step of HCV, was also reduced these findings argue that time-dependent polarisation of PHHs after seeding reduces cell entry and in turn virus infection. As both claudin-1 and occludin, two of the four absolutely crucial HCV entry cofactors, reside in cellular tight junctions, it is reasonable to assume that access to these cofactors may be limited under conditions of cell polarisation, thus reducing cell entry. Strikingly, Gondeau et al were able to recover permissiveness of PHH cultures for HCV after trypsinisation and re-seeding corroborating the notion that polarisation, which is broken by this process, reduces HCV entry efficiency.

Having established optimal infection conditions and using elaborate controls to firmly establish that their cells sustain productive HCV entry, and de novo RNA replication and virus production, the authors challenged PHHs from different donors with about 120 distinct HCV sera from many different donors infected with varying HCV genotypes and showing diverse clinical profiles. Infection efficiency was quantified by determination of HCV genomes in the inoculated cells. In parallel, the abundance of 52 serum makers was determined by multiplex serology. Using this approach, Gondeau et al make a number of remarkable observations: first, as expected, infectivity of HCVser was dramatically different between samples with peak virus load in the cell lysates differing by more than four orders of magnitude. Not surprisingly, there was a correlation between the virus load of the inoculum and the infectivity. Importantly, however, the virus load of the serum per se was not sufficient to predict if a sample would be highly infectious, intermediate or poorly infectious. This is strikingly illustrated by the observation that four different genotype 3a samples with similar virus load displayed up to 35-fold different infectiousness. Second, they observed that the HCV genotype does not predict infectiousness. Third, infectivity was not related to the source of the PHHs or the well-known IL28B nucleotide polymorphism. In fact, the authors show a very high reproducibility in the infectivity of a given serum across PHHs from different donors. This not only shows how robust their infection system is, but also highlights that serum (and/or) viral factor are dominantly determining infectiousness of HCV in PHHs. To explore which serum factor may predict HCV infectiousness, virus load in the cell lysate of inoculated PHHs was correlated with the abundance of 52 serum markers, including various cytokines, growth factors and inflammatory molecules. While no unifying signature directly predicting HCV infectiousness was evident, clearly a lower abundance of inflammatory mediators correlated with increased infectiousness. Notably, the serum donor’s clinical status could also be important, since eight of 14 samples collected after liver transplantation were highly infectious. This latter finding is in line with the notion that excessive inflammatory mediators may suppress HCV infectiousness. It remains open, however, which combination of factors is responsible, and how they suppress HCV infectiousness. Alternatively, the analytes investigated in this study may correlate with another serum factor—that may not have been determined—but which could influence HCVser infectiousness. Exploring this further is certainly an important avenue of future research. Nevertheless, the work and findings of Gondeau et al provide unique insights into the parameters that determine infection of primary HCV in PHHs, which is the most authentic HCV cell culture system. This should help to make this important model more accessible for HCV research. In the long run, this investment should pay off by disclosing unique features of the diverse mechanisms of the HCV–host interaction.

Competing Interest: None declared.


View Abstract


  • Funding TP is supported by grants from the Deutsche Forschungsgemeinschaft (SFB 900, project A6) and (PI 734/2–1) by grants from the Helmholtz Association (SO-024, HAI-IDR) and by a grant from the European Research Council (ERC-2011-StG_281473-VIRAFRONT).

  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles