SV40 DNA Signals Particular Host Proteins, then Recruits Its Own Capsid Building
Researchers long have puzzled over how the three capsid proteins of simian virus 40 (SV40) can eventually attach specifically to the SV40 mini-chromosome, a miniscule DNA molecule amid the voluminous cellular DNA surrounding the virus during its replication. Now, Ariella Oppenheim and her graduate student, Ariela Gordon-Shaag, at The Hebrew University-Hadassah Medical School in Jerusalem appear to have solved this mystery. They report that the virus encodes a DNA-packaging signal which includes promoter elements and that a soluble protein binds to these promoter elements and also recruits the three capsid proteins to assemble specifically around the viral DNA. This same signaling apparatus could prove useful when using SV40 as a means to transfer novel genes in gene therapy procedures.
Indeed, these findings about the SV40 DNA packaging apparatus emerge from vector development studies by Oppenheim in the field of gene therapy. "Because of our interest in developing a vector, we have analyzed which parts of the viral genome are required for making the vector," Oppenheim says. "We identified the packaging signal of SV40, which is called ses, for SV40 encapsidation signal." This discovery of a SV40 packaging signal "was a great surprise," says Oppenheim, who points out that other researchers extensively studied this virus without coming across the package-signaling apparatus and thought that any cellular DNA can be packaged in the virus without need for a signal. These new findings also expand upon the general understanding that viruses use host factors for their assembly. "This may provide clues for the development of cures for viral diseases," she says. Oppenheim and Gordon-Shaag describe their findings in the June Journal of Virology (76:5915-5924).
A protein called Sp1 (soluble protein-1) that binds GC boxes, which are promoter elements, is critical to the viral DNA packaging process, according to Oppenheim's genetic analysis. The researchers thus suspected that Sp1 might also be responsible for leading the capsid proteins to the SV40 mini-chromosome. However, Oppenheim says, "We needed to get more direct evidence that this is really recruitment."
A computer search indicates that the arrangement of GC boxes in SV40 is unique to that virus. "The way they are arranged, the DNA double helix has a spacing of 10-11 base pairs between one loop and another," Oppenheim says. "The way they are spaced, they are all facing the same direction." This geometric arrangement determines how the DNA interacts with the protein and what happens once bound. The DNA, says Oppenheim, "is more or less straight, but it becomes curved when Sp1 binds to it."
The specific binding between Sp1 and the capsid proteins is maintained even in the face of a 1,000-fold excess of cellular DNA, "providing strong support for the recruitment hypothesis," Oppenheim and her collaborators note. Apparently, the capsid protein "building blocks are recruited by Sp1 to ses, where they form the nucleation center for capsid assembly. By this mechanism, the virus ensures that capsid formation is initiated at a single site around its mini-chromosome." They also note that Sp1 "enhances the formation of SV40 pseudovirions in vitro, providing additional support for the model."
"What [Oppenheim] has now shown, I think very beautifully, is what's required for encapsulation forming the nidus on which the capsid grows is this Sp1 complex, and so all she needs is the region around the origin where the Sp1 sites are," says Robert Martin, a research molecular biologist at the National Institutes of Health (NIH) in Bethesda, Md. "She can put any piece of cellular DNA and get fairly efficient encapsidation-which is a considerable advance towards using this thing for gene therapy."
However, an important drawback to Oppenheim's system is its limited packaging capacity, says James M. Mason, who is director of the gene therapy vector laboratory at North Shore-Long Island Jewish Research Institute, Manhasset Campus, North Shore University Hospital in New York. The viral genome is 5.2 kb, which sets a low upper limit on the size and number of genes that an SV40-based vector can transfer into target cells. The vector, which is now prepared in vitro, has been modified to hold 15 kb, with no need for an SV40 DNA sequence.
Another drawback is that the SV40-based vector lacks precision delivery, as do most vectors. "There is no way to deliver the gene to the right chromosome, the right segment, the right position," Martin says. The lack of precision delivery is a major problem where precision production is required, he adds. However, "Where all you need is a smidgeon and not every cell has to have it, there is a reasonable chance gene therapy will work."
On the plus side, SV40 vectors packaged both in vivo and in vitro are very efficient transducers of human hematopoietic and liver cells, according to Oppenheim. Moreover, this vector is relatively safe, offering a means for delivering DNA into cells that does not deliver genes that could cause illness or cancer because everything is produced in vitro and because very little viral DNA is involved. For instance, the capsid proteins can be manufactured in insect cells, and the plasmid that carries the therapeutic gene, along with the little bit of viral DNA that is required for binding Sp1 is produced in Escherichia coli.
David Holzman writes from Lexington, Mass.