This week’s BEACON Researchers at Work blog post is by MSU graduate student Evin Hildebrandt.
While people are often all too familiar with those nasty virulent viruses that cause disease, attenuated viruses do not seem to be as well known as their nasty brethren. You may be wondering “So just what actually is an attenuated virus anyways?” In short, an attenuated virus is a virus which is no longer virulent and is unable to cause disease. Virulent viruses can become attenuated via serial passage by repeated passage through tissue culture to yield a virus which generally replicates very well in vitro, but has become attenuated, or “weakened,” in vivo and no longer causes disease. While the process of attenuation occurs in a variety of viruses, my project involves exploring what is the genetic basis for attenuation in Marek’s disease virus.
Marek’s disease virus (MDV) is an oncogenic herpesvirus that causes Marek’s disease (MD) in chickens, which costs over 1 billion dollars in losses a year, so controlling this disease is critical. Fortunately, there are multiple vaccines that have been introduced to prevent tumors and control MD over the years. Unfortunately, the reason that new vaccines have been required is that virulent field strains of MDV are evolving to greater virulence. The most effective vaccine available today is an attenuated virus that was created by serially passing a virulent strain repeatedly in vitro. While the process of in vitro attenuation has a well-known history in generating vaccines, the genetic basis behind what causes this change from virulence to avirulence remains unclear. Understanding what genes play a role in attenuation may allow for the design of more efficient vaccines, instead of relying on chance during serial passage to generate new vaccine candidates.
To identify what genes play a role in attenuation, three attenuated replicates were created by serially passing a virulent MDV BAC for 100 passages. MDV strains are often described as quasi-species, meaning that viral strains are not just one particular genotype, but are actually a mixture of genotypes that compose the viral population. Conversely, an MDV BAC clone contains only a single viral genotype in the entire viral population. Using an MDV BAC clone allowed us to determine if attenuation changes a strain’s phenotype through selection or mutation. If selection was the driving factor behind attenuation then avirulent genotypes already present in the quasi-species that replicate faster in tissue culture would dominate the viral population compared to de novo mutation generating new, avirulent genotypes. We expected that an MDV BAC clone would be unable to attenuate if selection primarily drove attenuation but de novo mutation would allow attenuation of even a 100% virulent BAC clone.
Initially, I was serially passing the three replicates every 5 days but progressively the time between passages decreased, eventually to passing the replicates every other day! While this dramatic change suggested that the viruses had become well adapted to in vitro growth and likely attenuated, I could not classify the serially passed replicates as attenuated without testing to determine virulence in vivo. Results of bird trials testing serially passed replicates showed that the MDV BAC clones did attenuate between passages 60-80, depending on the replicate. Since all replicates experienced a loss in virulence, this showed that de novo mutations are responsible for generating new, avirulent viruses and to attenuate the virus.
Since my goal is to identify the genetic basis of attenuation, I next needed to track down what mutations occurred in attenuated viruses compared to the virulent parental virus. Since the MDV genome contains only approximately only 100 genes and that the process of attenuation is such a reliable outcome following serial passage, we predicted there may be candidate genes responsible for attenuation shared in common among the attenuated replicates. Following Illumina sequencing of the attenuated MDV strains, we found 41-95 SNPs (depending on the replicate) in attenuated replicates compared to the virulent parental strain. While no identical, high frequency SNPs were shared among attenuated replicates at a nucleotide level, a much more promising picture emerged when considering the mutations at a gene level. Comparing nonsynonymous mutations that occurred in the same gene between the attenuated replicates revealed a very interesting candidate gene that contained multiple nonsynonymous mutations, many of which occurred at high frequencies, in all attenuated replicates. This gene, known as ICP4, contained between 3-8 nonsynonymous mutations (depending on the replicate) at various frequencies in ICP4. Some mutations occurred at frequencies as high as 40%, 60%, 80% in the viral population, and one SNP which was even completely fixed 100% in a particular replicate!
Looking at only this sequencing data from the attenuated replicates, ICP4 appears to be a very attractive candidate gene involved in attenuation, but considering the biological role of ICP4 helps to strengthen the case for the importance of ICP4 even more. Herpesvirus genes are classified as immediate early, early or late genes depending on when they are expressed. As an immediate early gene, ICP4 negatively regulates itself and other immediate early genes while activating early and late genes. One thought as to why ICP4 may be important in attenuation of MDV is perhaps mutations within ICP4 alter regulation of downstream genes, causing a widespread cascade effect in which mutations within ICP4 have a much wider impact than solely ICP4 itself.
Currently I am working on the next exciting step of finally making recombinant viruses in order to test how much of an affect candidate SNPs, especially ones within ICP4, have on virulence. Recombinant viruses will contain the attenuated version of a candidate SNP into an otherwise virulent MDV BAC background. These recombinant viruses will be tested for in vitro replication and disease incidence. This will allow us to identify what candidate mutations and genes affect MDV virulence, and possibly provide a starting point for developing and engineering new, beneficial vaccine candidates.
For more information about Evin’s work, you can contact her at hildeb35 at msu dot edu.