ANALYSIS OF THE PLASMA PROTEOME OF PATIENTS WITH CHRONIC HEPATITIS C INFECTION UNDERGOING TREATMENT WITH INTERFERON ALPHA AND RIBAVIRIN: A PILOT STUDY

Abstract

Background - Chronic hepatitis C virus infection is a massive worldwide healthcare burden with estimated costs in the USA alone of over $5 billon per annum. The only current effective treatment is combination therapy with interferon alpha and ribavirin (IFNα + RBV) which is expensive and has significant side-effects. Unfortunately, it is only effective in up to 50% of those infected with HCV genotype 1. In this study the examination of the patterns for the protein expression followed the peg-IFNα + RBV treatment in patients whom were selected to demonstrate different durations of the disease depending upon their response. The current treatment usually gives an SVR rate of around 55 %. Through the virological and host-related factors, the predictive value of HCV kinetics during treatment seems to exceed that of all virus and host-related baseline factors for predicting the virologic response. The identification of easily measurable biomarkers which predict response to treatment either before or during the early stages would enable better selection of those who would benefit from starting or continuing treatment. The search for novel biomarkers in disease has been the focus of increased attention in recent years due to the advances made in proteomic technology. The plasma proteome is an attractive target for such studies because it is readily available from patients on a regular basis. It is also in contact with all tissues in the body and so may reveal differences in the abundance of proteins expressed in these tissues. Current proteomic technology allows the characterisation of plasma proteins across a wide dynamic range to include those secreted from cells at relatively high concentration to those there due to tissue leakage present at extremely low levels. In this pilot study the plasma from a small number of selected patients was subjected to proteomic analysis to investigate whether there were variations in plasma protein profile which could predict treatment efficacy.
AIMS - The article shows the first stages of an ongoing analysis of the plasma proteome. As the aim is to identifying the potential biomarkers which indicate either responsiveness or resistance to treatment of chronic HCV with interferon alpha and ribavirin.
Methods - Three patients chronically infected with Hepatitis C virus genotype 1b who were to be treated with pegylated Interferon alpha and Ribavirin were selected for the pilot study. The total length of treatment was to be 48 weeks and patients were followed up as out-patients where biochemical and virologic parameters were measured and used as indicators of response.
Peripheral blood samples were taken into EDTA at the start of treatment and again at 14 and 28 days. Blood was immediately cooled on ice and within 15 minutes centrifuged at 3000 rpm for 10 minutes to separate the plasma which was aliquoted and stored at -80°C.
The plasma was depleted of albumin and IgG using the Bio-Rad Aurum Serum protein mini kit. This method removed over 90% of the total albumin and immunoglobin, effectively enriching the other plasma proteins and maximising their resolution by electrophoresis. Protein concentration of the depleted plasma was estimated using the Bradford assay. 2D gel electrophoresis was performed on selected samples using GE Healthcare IPGphor IEF and Ettan units. Briefly, 70 μg depleted plasma protein was mixed with Sample Buffer (8M urea; 4% CHAPS; 2% Ampholytes; 50 mM DTT; 0.001% Pyronin Y) and rehydrated into immobilized pH Gradient (IPG) strips overnight (5-8 pH range).
Isoelectric focusing was carried out. The IPG strips were then equilibrated into Loading Buffer (8M Urea; 50 mM Tris, 2% Lauryl SDS, 0.001% Pyronin Y, pH 6.8) and reduced with 50 mM DTT, prior to alkylation with 20 mM iodoacetamide. The IPG strips were then run onto a 12% Polyacrylamide gel for SDS-PAGE. The gels were then washed briefly and stained with Colloidal Coomassie Blue. The gels were destained in water and imaged using a BioRad GS-800 Scanning Densitometer and PDQuest software.
10 μg samples of plasma for 1D SDS-PAGE were added to four-times the volume of Loading Buffer. Samples were kept cool and then electrophoresed through a 12% SDS-PAGE gel. The gels were then either stained with Colloidal Blue or imaged, or electroblotted onto PVDF membrane. The membrane was blocked with 5% milk powder and then probed with mouse monoclonal antibody to human haptoglobin (Sigma H6395). The membrane was then washed thoroughly with PBS/Tween, incubated with antibody to mouse immunoglobulins-HRP (Dako P0260) and washed again with PBS/Tween. The membrane was developed using ECL plus (GE Healthcare) and the image acquired on a BioRad Fluor-S multi-imager.
Spots and bands were excised from the polyacrylamide gels and dried under vacuum. Proteins were digested using trypsin in 30 mM Ammonium Bicarbonate. The samples were concentrated under vacuum and extracted using a mixture of acetonitrile and 0.1% Formic Acid. Peptides were separated and identified using electrospray mass spectrometry (LC-MS/MS). The peptides of the mixture were separated using reverse-phase chromatography over a C18 column (Dionex) onto a Waters Q-ToF Micro. Tandem mass spectrometry data were analysed using the Mascot programme (Matrix Science).
Results - A number of differences were seen in patients compared with controls when the plasma was analyzed using 2D electrophoresis. Perhaps the most striking being the absence of haptoglobin alpha chain forms which were identified by electrospray mass spectrometry
(Fig 1).
Further analysis of albumin-depleted plasma by 1D SDS-PAGE found that two of the patients showed a considerable increase in an approximately 80 kDa protein band in samples taken at days 14 and 28 on treatment when compared with a pre-treatment sample (Fig 2).
This protein was identified as deriving from the haptoglobin precursor using LC-MS/MS, the patients who exhibited the increase in the ~80 kDa protein were a responder and a transient responder to treatment. The patient who did not respond to treatment showed no temporal increase in this band. The increase in the 80 kDa form of haptoglobin in responders was confirmed using immunoblotting probed with an antibody raised against haptoglobin (Fig 3).
The molecular weight of this band is anomalous. Haptoglobin usually appears as a band of around 20 kDa when strongly reduced and denatured. There have been reports of processed haptoglobin forming dimeric and tetrameric complexes when not fully denatured. The buffer used was not the classic Laemmli buffer and the samples were not boiled, so there may have been interaction between the sub-units. Alternatively, these may be interacting with another protein or ligand or these could be splice-variants of the haptoglobin precursor. Further work is under way to determine whether this is a robust finding that has broader implications for a wider cohort of patients.
Discussion and Conclusions - Haptoglobin has been implicated in liver disease for some time although some contradictory results have been seen. Recently it has been included in a panel of tests, the so-called ‘Fibrotest’, which it has been suggested can estimate the degree of hepatic fibrosis where serum haptoglobin levels inversely correlate with fibrosis.
In this study we have shown that in a small number of untreated HCV patients there is a reduction of haptoglobin compared to normal plasma. However, during treatment some individuals show a dramatic increase in a high molecular weight form of haptoglobin. It is unlikely that this is due to any hemolytic events associated with Ribavirin treatment, as this is usually associated with reductions in serum haptoglobin. Also, most patients had reasonably stable serum hemoglobin during the period of study. In this limited pilot study the patients which exhibited an increase in plasma haptoglobin soon after the commencement of IFNα+R therapy were those that responded to treatment by reduction of viral load to undetectable levels; one albeit transiently. The mechanism of the increase in haptoglobin in response to treatment is unknown, and it remains to be seen whether changes in plasma haptoglobin are consistently associated with responses to treatment in chronic hepatitis C virus infection.
Haptoglobin is a positive acute phase reactant and as such plasma levels may be expected to be disrupted during periods of inflammation in the liver, although no association with the degree of hepatic inflammation has been shown. This of course may have been confounded by the negative correlation seen between serum haptoglobin and liver fibrosis.
Further proteomic analysis of this patient group is ongoing and represents a powerful tool in the search for potential biological markers which may help in the management of therapy.