A Self-Replicating RNA Vaccine: More than Just Increased Copy Number
By David A. Corral, MD
The use of vaccines for cancer therapy is not a new idea, and this potential therapeutic approach has been studied from many aspects.1 Studies using tumor-specific proteins as antigens have pointed out the need to enhance immunogenicity considerably if clinical application is to be successful. The addition of cytokines, such as interleukin-12, have been partially successfully at eliciting a more pronounced immune response to a given vaccine; however, the problem of the production of neutralizing antibodies which potentially confound further vaccination has caused scientists to turn to nucleic acids as one possible alternative.2 Preexisting neutralizing antibodies produced by the humoral immune system are more than a theoretical concern, as they can result in diminution of the response by binding future challenges with the vaccine.3 Furthermore, the recombinant viral vectors needed for delivery may themselves be targets for neutralization by the humoral immune system.4,5 In theory, the uptake of "naked" nucleic acid vaccines by cancer cells could elicit a purely cellular immunologic response and avoid the problems of humoral immunity. Specifically, RNA vaccines would have the advantage of not requiring a DNA intermediate, and therefore, would avoid any potential problems caused by genomic integration. This would allow for immunization with constructs that ultimately express proteins such as specific oncogene products. The use of a "naked" RNA vaccine is particularly appealing because it does not contain genes for structural proteins, and therefore, cannot produce reversion mutants. While each of these aspects of RNA vaccines are enticing, their potential therapeutic use has yet to be proven.
Increasing the Copy Number
Rather than increasing immunogenicity by using an adjuvant, a potentially more productive strategy would be to increase the actual copy number of the genes encoding the antigen. Ying and associates have developed a "self-replicating" RNA construct that stands as a model for the development of a vaccine strategy.6 Ying et al take advantage of the activity of the enzyme replicase from the alphavirus Semliki forest virus.7 The replicase polyprotein cleaves itself to form the mature replicase complex, which drives the replication of both viral genomic and subgenomic RNA that normally encodes the structural proteins of the viral coat. Ying et al’s strategy was to use replicase activity to enhance the copy number of the antigen gene downstream from the replicase gene in the RNA construct. The authors built a naked RNA construct consisting of the replicase transcript upstream from the b-galactosidase transcript, which was chosen as the model antigen because it is easily assayable. Control constructs consisted for the same replicase transcript with no antigen encoded downstream, another construct with b-galactosidase alone or a third control with replicase rendered nonfunctional by a large deletion upstream from b-galactosidase.
The authors first proved expression of their construct in transfection experiments in which baby hamster kidney cells (BHK-21) were transfected with the complete construct or controls and stained for b-galactosidase. Transfection with the complete construct resulted in marked expression of b-galactosidase, whereas minimal expression was seen in any of the control samples. Next, Ying et al sought to quantify the expression of the b-galactosidase antigen utilizing reverse transcriptase PCR. In these experiments, RNA levels were quantitated by normalizing the PCR products against 18s ribosomal RNA. The authors noted substantial amplification of the RNA transcript in cells transfected with the complete construct. In fact, 10 hours following transfection, the expression of vaccine mRNA equaled that of 18s RNA. Next, the elicitation of antigen-specific antibody response was tested by serum ELISA for b-galactosidase specific antibodies.
In these experiments, it was demonstrated that a single intramuscular injection of the naked self-replicating RNA vaccine was sufficient to produce a b-galactosidase specific antibody response, whereas negative controls failed to produce this response at much higher doses. Furthermore, the authors went on the demonstrate a CD8+ T-cell response in mice immunized with the self-replicating RNA vaccine by measuring interferon gamma released from splenocytes re-stimulated in vitro with synthetic b-galactosidase peptide. This response was seen with as little as a single intramuscular injection of 0.1 mg of naked self-replicating RNA construct, a feat which had not been previously accomplished with naked DNA or RNA vaccines.
Antitumor Effect of the Vaccine
Antitumor effect of the self-replicating RNA vaccine was determined by challenging mice with intravenous injection of CT26 colon cancer cells 21 days after vaccination. Complete protection from the development of pulmonary metastasis was conferred by only 0.1 mcg of the vaccine (negative controls produced no protection up to 100 mg of RNA injected). When colon tumor cells were injected two days after vaccination, the self-replicating RNA vaccine conferred a 10- to 20-day survival advantage over controls. The combination of the self-replicating vaccine and interleukin 12 significantly prolonged survival, with 40% of mice surviving more than 90 days.
These experiments demonstrated that cells transfected with the self-replicating RNA-construct produce only slightly more than twice the amount of antigen than was produced by a conventional cytomegalovirus promoter driven DNA plasmid. And yet, markedly greater immunogenicity was produced by the self-replicating RNA vaccine in vivo than would be predicted by a two-fold increase in antigen production. Ying et al, therefore, looked for an explanation at the level of the host cells and their interaction with antigen presenting cells.
A direct cytopathic effect on transfected BHK21 cells using an RNA replicase vector expressing enhanced green fluorescence protein was studied. These cells demonstrated morphologic changes, cessation of cell division, and nuclear fragmentation characteristic apoptosis. Furthermore, specific inhibition of caspases (endoproteases known to mediate signals that direct many of the proteolytic events that occur during apoptosis) substantially delayed cell death, indicating that transfection with the self-replicating vaccine leads to caspase-dependent apoptotic cell death.
Apoptotic Transfected Cells
The authors went on to investigate the interactions of apoptotic transfected cells with antigen presenting cells. They generated a population of immature dendritic cells that retain phagocytic activity. After incorporation of a red fluorescent marker into the cell membrane of BHK21 cells, it was demonstrated that transfection of the BHK cells with the self-replicating RNA construct significantly increased uptake by dendritic cells. Putting these findings together, the authors proposed that one of the mechanisms underlying the enhanced immunogenicity of self-replicating RNA vaccines may be the enhanced uptake of antigen by dendritic cells and other "professional" antigen-presenting cells of cells that undergo apoptotic death as a result of the replicase activity.
Conclusion
The study by Ying et al is important in that it not only demonstrates the activity of a naked RNA construct which is able to increase its own copy number through replicase activity, but also demonstrates that its antitumor effect is mediated by more than just increased copy number. A direct cytopathic effect on transfected cells appears to result in apoptosis, allowing dendritic cells to phagocytize these cells and present tumor antigens to the immune system. These properties make replicase driven RNA vaccines appealing in the development of future vaccine strategies. Given these findings using a common model tumor antigen like b-galactosidase, one wonders how necessary it will be to use specific tumor oncogene products as antigens. Future studies using oncogene antigens with and without replicase, or with replicase alone, should shed further light on this area.
References
1. Liu MA, Fu TM, Donnelly JJ, et al. Adv Exp Med Biol 1998;452:187-191.
2. Irvine KR, Rao JB, Rosenberg SA, et al. J Immunol 1996;156:238-245.
3. Rosenberg SA, et al. J Natl Cancer Inst 1998;90: 1894-1900.
4. Restifo NP. Curr Opin Immunol 1996;8:658-663.
5. Pardoll DM. Nat Med 1998;4:525-531.
6. Ying H, Zaks TZ, Wang R, et al. Nat Med 1999;5:823-827.
7. Atkins GJ, Sheahan BJ, Liljestrom P. Mol Biotechnol 1996;5:33-38.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content