Principles of Virology, Volume 1. Jane Flint
suppression, or pseudoreversion, to distinguish it from reversion at the original site of mutation. Reversion has been studied since the beginnings of classical genetic analysis. In the modern era of genetics, cloning and sequencing techniques can be used to demonstrate suppression and to identify the nature of the suppressor mutation (see below). The identification of suppressor mutations is a powerful tool for studying protein-protein and protein-nucleic acid interactions. Some mutations complement changes made at several sites, whereas allele-specific suppressor mutations complement only a specific change. The allele specificity of second-site mutations provides evidence for physical interactions among proteins and nucleic acids.
Phenotypic revertants can be isolated either by propagating the mutant virus under restrictive conditions or, in the case of mutants exhibiting phenotypes (e.g., small plaques), by searching for wild-type properties. Chemical mutagenesis may be required to produce revertants of DNA viruses but is not necessary for RNA viruses, which spawn mutants at a higher frequency. Nucleotide sequence analysis is then used to determine if the original mutation is still present in the genome of the revertant. The presence of the original mutation indicates that reversion has occurred by second-site mutation. The suppressor mutation is identified by nucleotide sequence analysis. The final step is introduction of the suspected suppressor mutation into the genome of the original mutant virus to confirm its effect. Several specific examples of suppressor analysis are provided below.
TERMINOLOGY
Operations on nucleic acids and proteins
A mutation is a change in DNA or RNA comprising base changes and nucleotide additions, deletions, and rearrangements. When mutations occur in open reading frames, they can be manifested as changes in the synthesized proteins. For example, one or more base changes in a specific codon may produce a single amino acid substitution, a truncated protein, or no protein. The terms “mutation” and “deletion” are often used incorrectly or ambiguously to describe alterations in proteins. In this textbook, these terms are used to describe genetic changes and the terms “amino acid substitution” and “truncation” to describe protein alterations.
BOX 3.11
DISCUSSION
Is the observed phenotype due to the mutation?
In genetic analysis of viruses, mutations are made in vitro by a variety of techniques, all of which can introduce unintended changes. Errors can be introduced during cloning, PCR, or sequencing and when the viral DNA or plasmid DNA is introduced into the cell.
With these potential problems in mind, how can it be concluded that a phenotype arises from the planned mutation? Here are some possible solutions.
Test several independent DNA clones for the phenotype.
Repeat the plasmid construct ion. It is unlikely that an unlinked mutation with the same phenotype would occur twice.
Look for marker rescue. Replace the mutation and all adjacent DNA with parental DNA. If the mutation indeed causes the phenotype, the wild-type phenotype should be restored in the rescued virus.
Allow synthesis of the wild-type protein in the mutant background. If the wild-type phenotype is restored (complemented), then the probability is high that the phenotype arises from the mutation. The merit of this method over marker rescue is that the latter shows only that unlinked mutations are probably not the cause of the phenotype.
Each of these approaches has limitations, and it is therefore prudent to use more than one.
Some mutations within the origin of replication (Ori) of simian virus 40 reduce viral DNA replication and induce the formation of small plaques (see Chapter 9 for more information on the Ori). Pseudorevertants of Ori mutants were isolated by random mutagenesis of mutant viral DNA followed by introduction into cultured cells and screening for viruses that form large plaques. The second-site mutations that suppressed the replication defects were localized to a specific region within the gene for large T antigen. These results indicated that a specific domain of large T antigen interacts with the Ori sequence during viral genome replication.
The 5′ untranslated region of the poliovirus genome contains elaborate RNA secondary-structural features, which are important for RNA replication and translation, as discussed in Chapters 6 and 11, respectively. Disruption of such features by substitution of a short nucleotide sequence produces a virus that replicates poorly and readily gives rise to pseudorevertants that reproduce more efficiently. Nucleotide sequence analysis of the genomes of two pseudorevertants revealed base changes that restore the disrupted secondary structure. These results confirm that the RNA secondary structure is important for the biological activity of this untranslated region.
RNA Interference (RNAi)
RNA interference (Chapter 8) has become a powerful and widely used tool because it enables targeted loss of gene function. In such analyses, duplexes of 21-nucleotide RNA molecules, called small interfering RNAs (siRNAs), which are complementary to small regions of the mRNA, are synthesized chemically or by transcription reactions. siRNAs or plasmids or viral vectors that encode them are then introduced into cultured cells by transformation or infection. The small molecules then block the production of specific proteins by inducing sequence-specific mRNA degradation or inhibition of translation. Duplex siRNAs are unwound from one 5′ end, and one strand becomes tightly associated with a member of the argonaute (Ago) family of proteins in the RNA-induced silencing complex, RISC. The small RNA acts as a “guide,” identifying the target mRNA by base-pairing to specific sequences within it prior to cleavage of the mRNA or inhibition of its translation.
To determine the role of a viral gene in the reproduction cycle, siRNA targeting the mRNA is introduced into cells. Reduced protein levels are verified (e.g., by immunoblot analysis) and the effect on virus reproduction is determined. The same approach is used to evaluate the role of cell proteins such as receptors or antiviral proteins.
In another application of this technology, libraries of thousands of siRNAs directed at all cellular mRNAs or a specific subset can be introduced into cells to identify genes that stimulate or block viral reproduction. The siRNAs are produced from lentiviral vectors as short hairpin RNAs (shRNAs) that are processed into dsRNAs that are then targeted to mRNAs by RISC. In one approach, cells are infected with pools of shRNA-containing lentivirus vectors (Fig. 3.13). The cells are placed under selection and infected with virus to identify changes in reproduction caused by the integrated vector. If necessary, pools of vectors that have an effect on virus reproduction can be further subdivided and rescreened. Enriched shRNAs are detected by high-throughput sequencing and bioinformatic programs that quantitate the number of reads per shRNA compared with the starting population. The likelihood that knockdown of a specific mRNA is a valid result increases as the number of enriched orthologous shRNAs for the targeted gene increases. In other words, a gene targeted by three different shRNAs established by sequencing data is more likely to be a true positive than a gene targeted by only one. Another approach, arrayed RNAi screening, uses transfection of siRNAs into cells grown in a multiwell format (Fig. 3.13). As a record is kept of which siRNAs are added to each well, targeted genes can be readily identified after their effect on virus infection has been ascertained.
No matter which method is used to identify genes that affect viral reproduction, the most convincing confirmation of the result