Hepatitis C Virus Transmission Bottlenecks Analyzed by Deep Sequencing

Hepatitis C virus (HCV) replication in infected patients produces large and diverse viral populations, which give rise to drug-resistant and immune escape variants. Here, we analyzed HCV populations during transmission and diversification in longitudinal and cross-sectional samples using 454/Roche pyrosequencing, in total analyzing 174,185 sequence reads. To sample diversity, four locations in the HCV genome were analyzed, ranging from high diversity (the envelope hypervariable region 1 [HVR1]) to almost no diversity (the 5′ untranslated region [UTR]). For three longitudinal samples for which early time points were available, we found that only 1 to 4 viral variants were present, suggesting that productive infection was initiated by a very small number of HCV particles. Sequence diversity accumulated subsequently, with the 5′ UTR showing almost no diversification while the envelope HVR1 showed >100 variants in some subjects. Calculation of the transmission probability for only a single variant, taking into account the measured population structure within patients, confirmed initial infection by one or a few viral particles. These findings provide the most detailed sequence-based analysis of HCV transmission bottlenecks to date. The analytical methods described here are broadly applicable to studies of viral diversity using deep sequencing.Hepatitis C virus (HCV) is a positive-strand enveloped RNA virus of the flavivirus family. HCV infects ∼170 million people worldwide with a high rate of persistence (1, 2) and is a major etiological agent of chronic hepatitis, liver cirrhosis, and hepatocellular carcinoma. The current standard of therapy is the combined use of pegylated alpha interferon (IFN-α) and ribavirin (9), although there are substantial limitations due to toxicity and resistance profiles (47). Recent development of various small-molecule inhibitors that specifically target HCV offer some promise (13), but challenges still remain because the size and diversity of viral populations promote rapid development of drug resistance (28, 42). In an infected individual, serum HCV RNA levels can reach 10 to 100 million IU/ml (40). The viral RNA polymerase is estimated to make 1 error per 10,000 to 100,000 bp copied (22), but the viral genome is only 9,600 bases, resulting in diversification of the viral population, so that most viral genomes differ in sequence from the population consensus (16, 20, 21). Thus, when antiviral pressure is exerted on a viral population, sequence variants with reduced sensitivity may expand in the presence of the selective pressure (30, 41) and cause resistance (37). Consistent with this, differential sequence diversity in HCV populations has been linked to clinical outcome (7, 8).The size and complexity of HCV populations has made their analysis challenging. However, new deep-sequencing and bioinformatics methods are well suited to analyzing this problem. Using the 454/Roche technology, it is possible to generate more than 10⁸ bases of DNA sequence in a single 1-day run, albeit in fragments 200 to 500 bases in length (24). In addition, many samples can be multiplexed in single experiments using DNA barcodes introduced in amplification primers to tag each sample (3, 12, 45, 46), allowing many viral sequences to be characterized in a single experiment.Here, we analyze HCV diversity by pyrosequencing a series of representative viral regions contained within PCR amplicons, and we use methods from the environmental microbiology field for data processing and analysis. In both virology and environmental microbiology, populations of interest commonly consist of many related but nonidentical sequences (e.g., viral lineages with related sequences or bacteria harboring related 16S rRNA gene sequences). Assembly of short pyrosequence reads into longer scaffolds is quite difficult in such a setting, because the related sequences present in the population can be assembled in many different ways. Complicated data-processing methods yield at best complex probabilistic models of variants likely to be present in the population (6). For this reason, in studies of bacterial 16S DNA from uncultured communities, many groups have used simplified analysis of single 16S amplicons that query short regions of the 16S rRNA gene (5, 11, 14, 23, 25, 38, 44). Extensive simulations and practical applications show that analysis of such “sequence tags” can disclose biologically meaningful clusters and gradients in collections of samples. Here, we apply a similar approach, using sequence tags for several regions of the HCV genome. This approach has the disadvantage of losing linkage information between amplicons, but it does allow the efficient analysis of large numbers of viral variants over many samples.A major challenge, however, is distinguishing variations authentically present in viral populations from artifactual mutations introduced as a result of the isolation procedure or sequencing error. Sequence recovery involves PCR steps that can result in base pair substitutions or artifactual chimera formation. The 454/Roche method, like any sequencing method, has a characteristic error rate and particularly elevated error rates at homopolymer runs (24). In this study, we took advantage of improved methods for error control using the PyroNoise program of Quince and colleagues, which was first used for analysis of 16S rRNA gene sequences (29). The PyroNoise program preclusters the raw light intensity data generated during pyrosequencing by the 454/Roche method, which removes most homopolymer errors. In reconstruction experiments, Quince and colleagues showed that 454/Roche sequence analysis of artificially constructed mock 16S rRNA gene communities yielded greatly inflated numbers of sequence types due to error, but preclustering using PyroNoise reduced the diversity to values much closer to the correct value. Here, we used a two-stage clustering method to remove noise. In the first stage, raw light intensity data were preclustered with PyroNoise (29); then, in the second stage, after interpretation of the sequence as base calls, sequences were clustered at 98.5% identity. The second step allowed us to take advantage of redundancy in the reads to improve sequence quality, though distinguishing genuine low-level variations in the viral populations from error is a challenge.We determined 174,185 high-quality HCV sequence reads to characterize (i) longitudinal variation in HCV populations following transmission, (ii) differences in HCV variation between HCV-monoinfected and HIV-HCV-coinfected subjects, and (iii) variation in a control HCV genome cloned in a bacterial plasmid to quantify variation arising during the isolation and analytical procedure. We developed amplicons to characterize four regions of the HCV genome (Fig. ​(Fig.1A)1A) and found that HCV sequence diversity ranged from almost nonexistent to extreme depending on the region of the viral genome studied. Using the deep-sequencing data, we estimate that only one or a few viral variants seeded initial infection, but after that, viral variants could expand to >100 in a single individual. Thus, these data specify the numbers of particles seeding productive infection and provide a general framework for the use of deep-sequencing data to characterize the structures of viral populations.Open in a separate windowFIG. 1.HCV genome and characteristics of three subjects studied longitudinally during acute HCV infection. (A) The HCV genome and the positions of amplicons studied. The amplicons are numbered 1 to 4 from left to right, and a letter is used to indicate the direction of sequence determination. For the E1E2 HVR1 (3) and E2 (4) amplicons, two slightly different primers were used in each direction in an effort to maximize the diversity of recovered sequence variants, and these are indicated by the two bars. (B) HCV load and ALT levels for patients 1 to 3 during acute HCV infection. The x axis shows the number of weeks after clinical presentation, which for these patients was close in time to initial infection. Further patient characteristics were as follows: patient 1, injecting drug user, anti-HCV negative on 18 June 2001, first ALT flare (ALT, 677) on 6 July 2001, anti-HCV positive on 11 October 2001; patient 2, possible medical exposure, anti-HCV negative on 16 May 2001, initial ALT flare (ALT, 467) on 9 January 2004, anti-HCV positive on 22 April 2004; patient 3, injecting drug user, anti-HCV negative on 31 January 2006 (slightly abnormal ALT, 73), initial ALT flare (ALT, 640) on 10 April 2006, anti-HCV positive on 11 April 2006.     

Structural Properties of HIV Integrase��Lens Epithelium-derived Growth Factor Oligomers

Integrase (IN) is the catalytic component of the preintegration complex, a large nucleoprotein assembly critical for the integration of the retroviral genome into a host chromosome. Although partial crystal structures of human immunodeficiency virus IN alone and its complex with the integrase binding domain of the host factor PSIP1/lens epithelium-derived growth factor (LEDGF)/p75 are available, many questions remain regarding the properties and structures of LEDGF-bound IN oligomers. Using analytical ultracentrifugation, multiangle light scattering, and small angle x-ray scattering, we have established the oligomeric state, stoichiometry, and molecular shapes of IN·LEDGF complexes in solution. Analyses of intact IN tetramers bound to two different LEDGF truncations allow for placement of the integrase binding domain by difference analysis. Modeling of the small angle x-ray scattering envelopes using existing structural data suggests domain arrangements in the IN oligomers that support and extend existing biochemical data for IN·LEDGF complexes and lend new insights into the quaternary structure of LEDGF-bound IN tetramers. These IN oligomers may be involved in stages of the viral life cycle other than integration, including assembly, budding, and early replication.     

Regulation of O-glycosylation through Golgi-to-ER relocation of initiation enzymes

After growth factor stimulation, kinases are activated to regulate multiple aspects of cell physiology. Activated Src is present on Golgi membranes, but its function here remains unclear. We find that Src regulates mucin-type protein O-glycosylation through redistribution of the initiating enzymes, polypeptide N-acetylgalactosaminyl transferases (GalNac-Ts), from the Golgi to the ER. Redistribution occurs after stimulation with EGF or PDGF in a Src-dependent manner and in cells with constitutively elevated Src activity. All GalNac-T family enzymes tested are affected, whereas multiple other glycosylation enzymes are not displaced from the Golgi. Upon Src activation, the COP-I coat is also redistributed in punctate structures that colocalize with GalNac-Ts and a dominant-negative Arf1 isoform, Arf1(Q71L), efficiently blocks GalNac-T redistribution, indicating that Src activates a COP-I–dependent trafficking event. Finally, Src activation increases O-glycosylation initiation as seen by lectin staining and metabolic labeling. We propose that growth factor stimulation regulates O-glycosylation initiation in a Src-dependent fashion by GalNac-T redistribution to the ER.     

Multiple glycerol shocks increase the calcium phosphate transfection of non-synchronized CHO cells

The exposure of CHO DG44 cells to an osmotic shock, after DNA uptake, results in a cellular volume decrease of approx. 55%. Repetitive osmotic shocks targeted different sub-populations of cells as was demonstrated using two different fluorescent reporter genes. Also the exposure of a calcium phosphate–DNA coprecipitate to high osmolarity in vitro caused the release of the DNA from the precipitate. The results demonstrate the importance of the osmotic shock on the efficient delivery of plasmid DNA to the nucleus of CHO cells following calcium phosphate-mediated transfection.