The oral polio vaccine (OPV) has been the recommended prevention of wild-type polio in the United States. The administration of this live, attenuated vaccine given to infants at 2,4, and 16-18 months of age is associated with 5-10 cases per year of vaccine-associated paralytic poliomyelitis (VAPP). Paralytic poliomyelitis still exists in the US, but it is solely the result of the administration of OPV (31). The estimated risks of VAPP in recipients of OPV is approximately 1 in 1.2 million for the f irst dose and 1 in 6.2 million for the second dose (31).

Efforts to rid the world of poliomyelitis began when Jonas Salk in 1955 developed a killed injectable vaccine (IPV) (2). Six years later a live, attenuated, orally administered vaccine (OPV) was introduced by Albert Sabin (1). IPV is expensive and provides only humoral immunity without the risk of VAPP (2, 20)(Fig. 2). OPV is easy to administer, inexpensive, and provides humoral and mucosal immunity to recipients and unvaccinated contacts through fecal shedding (1, 5, 6, 7)(Fig 2). OPV also induces intestinal immunity similar to that induced by wild-type infection, but its administration is associated with the risk of VAPP (4, 11, 22).

Fig 2 (32)

The use of OPV and IPV has vastly reduced the incidence of poliomyelitis world wide, and the last reported case of paralytic poliomyelitis in the Western Hemisphere was reported in 1990 (10). Poliomyelitis was declared eradicated from the Western Hemis phere in 1994 (8). While substantial achievements have been made with eradication of poliovirus, developing countries still face widespread poliomyelitis outbreaks.

The nature of poliovirus transmission has required continued vaccination even in areas where poliomyelitis has been declared eradicated (5, 6, 7, 8, 12). People who have humoral and intestinal poliovirus immunity can still be infected with wild-type po liovirus and remain asymptotic. These individuals can serve as point sources for wild-type poliovirus transmission, spreading it to those that are unvaccinated. Periodic outbreaks of paralytic poliomyelitis still occur in developed countries among communi ties that refuse vaccination (1). The general population was highly vaccinated, but wild-type poliovirus circulated until it reached the unvaccinated communities.

One person could import wild-type poliovirus to a country without any signs or symptoms. We are stilled faced with the required vaccination and risk of VAPP until poliomyelitis is eradicated from the world. As a result, the United States Public Hea lth Service (USPHS) suggested that two doses of IPV be given before two doses of OPV in the first 18 months of life. This recommendation is based on the theory that two doses of IPV will provide enough humoral immunity to eliminate the risk of VAPP upon s ubsequent doses with OPV. The theory supports a reduction of VAPP in the recipients of OPV, but it does not necessarily protect contacts of the recipients from getting VAPP. Evidence to support whether this change would reduce the number of VAPP cases per year is lacking. The administration of OPV after IPV may be associated with the shedding of more virulent strains, exposing contacts to higher levels of nonattenuated virus. The number of recipient cases may drop, but the contact cases could sore higher than expected. Estimated costs for the change are about $75-85 million, which could be used to eradicate poliomyelitis in many areas of the world (19).

The cause of VAPP is unknown, but it is probably due to the reversion of vaccine viruses. After OPV is administered, the live attenuated viruses replicate in the gastrointestinal tract (GI). Recipients often shed virus in their stool in which multi ple genetic changes have occurred. Changes in genetic material can alter the attenuation of the vaccine, making it more harmful to the host and contacts. Of the three Sabin viruses, current research has isolated type 3 from a majority of the VAPP cases (1 2). There are many theories regarding the attenuation of Sabin 3, and they all involve the modification of the genetic material as it replicates in the GI tract, whether it be mutations, conformational changes, recombination, host factors, or competition. All of these theories overlap in certain ways, but the focus of research has been limited to only a few of these theories. No one knows exactly how OPV causes VAPP.

Mutations in the Sabin 3 genome have incurred most of the attention for studying the mechanism of OPV in VAPP. Studies have identified certain point mutations in Sabin 3 after replicating in VAPP patients, cell lines, monkeys, and a few healthy hum ans that appear to attenuate the vaccine (13-16, 25,26,27,28). The loss of attenuation of the vaccine, theoretically to the reversions, as is replicates in the GI tract of recipients is thought to play a major role in infecting the CNS of recipients or co ntacts and causing VAPP, but previous studies have been too limited to definitely determine the mechanism and relation these mutations have on the pathogenesis of VAPP. The next theory deals with mutations that may cause conformational changes in the inte rnal ribosomal entry site (IRES), resulting in increased replication and virulence (17). Studies on this theory are limited Sabin 1, but the IRES sites of all three strains are similar. The third hypothesis involves recombination between the three Sabin s erotypes and other viruses (18,23,24,33). There are an infinite number of possible recombinations. The main stipulation involves the loss of attenuation of the vaccine strains through recombination. The fourth theory has not been studied, and it suggests that a host factor is involved, not a viral factor. Immunodeficient children are at a higher risk for VAPP, but normal healthy children still get VAPP, suggesting some other factor(s) may be involved. The fifth theory synthesizes the four previous theorie s and provides a possible mechanism that accounts for the genetic changes and pathogenesis of Sabin 3 in VAPP. Competition among the Sabin strains to replicate could provide the selection pressures necessary for Sabin 3 to mutate, increasing replication a nd virulence until the body is overwhelmed.

Almost all research in the Sabin 3 VAPP field has centered around mutations in Sabin 3. The three viral strains of OPV were derived from virulent strains by serial passage through particular cell lines. During serial passage the virus mutat ed at particular areas and lost the virulent phenotype. Unfortunately, when OPV is administered, the viral strains often correct these mutations and revert back to the virulent phenotype. This is the basis for the theory that particular mutations may be r esponsible for the loss of attenuation and pathogenesis of Sabin 3 in VAPP.

Evans et al. (14) and the World Health Organization (WHO) did a collaborative study on 12 strains of type 3 VAPP and found that all stool samples contained a uracil to cytosine mutation at bp472. Percentages of the mutation in the stool were not determined. Six cases were found to be highly neurovirulent by WHO monkey neurovirulence tests. Two vaccine strains were tested as a control, and they found the unreverted form.

Minor et al. (15) tested every fecal sample of his child for the 472 reversion after the child received OPV. Minor did not test the percentage of the reversion in the stool, just the presence of it. WHO monkey neurovirulence tests were also performed on an unspecified number of isolates. The neurovirulence increased with each stool sample. In a review of this study, Evans et al. noted that the progressive neurovirulence of the stool samples may have been due to the increase proportion of the 472 reversion in each sample.

Chumakov et al (28, 30)

Two interesting studies by Chumakov et al. (28,30) compared the neurovirulence rate, reversion content, and particular mutations in OPV passed through cell lines. With as little as a 1% increase in the 472 reversion content, the neurovirulence of t he vaccine increased dramatically to the point of failure designated by WHO standards. The ranges are startling because this shows either how sensitive the reversion is or that there is another factor involved in the attenuation of Sabin 3 besides the mut ation. The passages consistently selected for the 472 mutations.

Stool samples in the first two studies were passed in cell culture to isolate the virus. Chumakov et al (31) demonstrated that passing the virus through cell culture increases reversions in the original sample, negating the results of the first two studies. Passages through the cell lines could have altered the original mutation content in the stool. Another negative factor of all three studies was that the sample sizes were extremely small, hindering statistical significance.

The most recent analysis of the mutations in Sabin 3 was done by Lu et al. (26). The authors analyzed the time course and content of the specific mutations (472 U® C; 2034 U® C; 2493 C® U) of Sabin 3 after replicating in the CNS of monkeys. They performed three different experiments. In the first experiment, two sets of 4 juvenile rhesus monkeys were inoculated in the lumbar region of the spinal cord with t he Pfizer substrain (RSO) of Sabin 3 poliovirus. One set was inoculated with a vaccine lot containing .7% 472-C and the other with a lot containing 3.7%-C. Vaccine lots (.7%-C) passed WHO neurovirulence tests but 3.7%-C lots did not. The monkeys were sacr ificed on days 4, 8, and 17 postinoculation (p.i.). Samples from different areas of the CNS were taken and run on HEp-2 cell cultures. Mutation analysis was done by MAPREC (Mutation analysis by polymerase chain reaction and enzyme cleavage). A second expe riment was done on another set of monkeys to determine if reversions were happening earlier on days 2, 4, 6, 8 and 17 with the RSO vaccine strain containing .4% 472-C, 1% 2493-U, 0% 2034-C. Day 17 monkeys were inoculated with commercial lots of RSO. MAPRE C was also performed on samples from these sacrificed monkeys. In the third study, the group passed a 40% 2034-C CNS isolate from a day 4 sacrifice over VERO cell lines at two different temperatures and analyzed progression of mutations.

The authors found that the bp472 U® C mutation increased over time p.i. and reached 100% by day 17 in both groups. There was extreme variability in the results for both the 472 and 2493 position, but a general trend of a faster reversion rate in the 3.7% 472-C inoculum could be determined. On hundred percent of isolates at day 2 had the 2493-U reversion but no day 17 isolates did. Reversion at 2034 were too random to even determine a general trend. The 2034-C mutations in creased in VERO cells grown at 38.5° C but decreased in temperatures of 37° C.

The authors concluded that neuroreversion occurs rapidly in monkeys but is a complicated process. MAPREC analysis showed that mutations of Sabin 3 differed in time course, indicating that selection pressures changed as the virus replicates. The author suggests that since the neuroreversion occurs rapidly with no signs of pathogenesis in hosts, there must be a protective defense mechanism issued by the immune system. The mutation at 2493 was determined not to be a significant factor in the loss of atten uation in Sabin type 3 poliovirus vaccine.

This study is the most current progress regarding Sabin 3 mutations, however, the author tried to tackle too many hypotheses for one study. No controls were used for these experiments and most data was not statistically significant. The study p ool was too small, and the monkeys were sacrificed in order to get one isolate. The samples are not representative of the mutations occurring in the same monkey, rendering the data random with too many variables. PCR has the capability of detecting virus in CNS samples, avoiding the cell culture techniques to reduce the in vitro effect of producing reversions, a recurrent problem with the initial studies. The administration of OPV in this study is another problem. OPV is given orally in humans, not by direct injection into the CNS.

Internal ribosomal entry site (IRES) Theory (17)

The IRES is the location of the genome that signals the ribosome to begin translation. An IRES generally has significant secondary and tertiary structure that are sensitive to mutations. Alteration of the genome in the IRES can dramatically cha nge the structure and function it serves, increasing or decreasing the replication rate. This field is new and not many studies have addressed the potential significance of these structures in viruses. One study has begun to look at the IRES of poliovirus 1 (17). It is applicable to poliovirus 3 because the IRESes in the 5’ noncoding region are highly conserved between the three strains of polioviruses.

Gromeier et al. (17) replaced the entire IRES of poliovirus 1 with another virus, creating an in vitro recombination. The study is more representative of the IRES theory, rather than the recombination theory, because it strives to demonstrate t he function and role of the IRES in poliovirus. Attenuation of wild-type poliovirus can be achieved by mutating the IRES. The attenuated vaccine strains in OPV provide good immunity, but they often repair the mutant site in the IRES during replication in the GI tract. Animal models show a nueropathogenic potential for these revertants.

Gromeier et al. replaced the IRES of poliovirus 1 with human rhinovirus (designated RIPO) and encephalomyocarditis (ENPO) IRES. The new recombinants were tested in cell lines and mice for growth properties and neuropathic effects. Three other strains w ere produced, each a replica of the wild-type control (M), RIPO, and ENPO strains except for base modifications throughout the genome (designated S, ENPOS, RIPOS) . Tests were performed on all 6 viruses: M, S, ENPO, ENPOS, RIPO, RIPOS. M and RIPO strains demonstrated similar plaque formation. ENPO plaques were smaller in size. S, RIPOS, and ENPOS barely produced any plaques. The growth curve of M was highest in all three cells lines except for HEp-2 cells in which RIPO displayed an earlier response. RIPO demonstrated a smaller growth curve in the other two cell lines compared to the other strains. In the mice assays, RIPO displayed the smallest viral replication compared to M or ENPO. In strains S, ENPOS, and RIPOS no viral replication was detected.

Gromeier et al. concluded that despite viral tropism to cell receptors, virion stability as well as cell-internal restriction of replication play an important role in infection. Human rhinoviral and poliovirus 1 IRESes display similar functions and eff iciency. The neuronal basis of the human rhinovirus IRES infection remains to be determined. The author also stated that the mouse system was not very useful to study poliovirus replication.

Critique

The study was interesting, but it was very difficult to pick out exactly the hypothesis the author was trying to prove. They failed to effectively address their reason for using the particular cell lines and recombinants that contained point mutations. Sabin vaccines should have been used as controls. Their introduction mentioned the safety of the vaccine, but they failed to address it in their experiments. Mice should not have been used for poliovirus replication when it has been demonstrated previous ly that mice are not good models for these viruses. Statistical significance and the numbers of assays and mice infected were not reported. Altering specific base pairs of poliovirus 1, instead of recombination, would be an effective way of proving the se nsitivity or flexibility of the IRES. Future studies regarding the IRES theory should be analyzed by competitive growth analysis of successive mutations before the issue of recombination with unrelated viruses is analyzed.

During viral replication, recombination occurs when partial stretches of sequence are replaced or combined with foreign genomic information. This foreign genetic information is usually from other viruses replicating in the same location, an d the recombinations that can occur are numerous. Only a few studies have focused on recombination among the polioviruses (18, 21, 23, 24, 33), and few have focused on Sabin 3 recombinations.

Li et al (34) studied 4 cases of poliovirus 1 VAPP. They found recombinations that were a mixture of wild-type and poliovirus 1 with many mutations. This study is not applicable because it involves recombination with wild-type poliovirus wh ich is not circulating in the US. Friedrich et al. (18) did a similar study that focused on VAPP cases and healthy contacts in Brazil. Recombinant type 2 viruses with type 1 were found in the VAPP cases and healthy contacts. Recombination with type 3 was not studied, and since recombination is extremely variable, the data is not applicable to Sabin 3 recombinants. Cell line isolation was also used in both studies, introducing the potential variable of an in vitro genetic alteration.

Georgescu et al. (23) did a fascinating study that compared viral isolates from the CNS and stool samples of VAPP patients. Viral isolates from the CNS and stool were passed over cell lines, isolated, and analyzed extensively. In all but on e sample, viral strains in the CNS corresponded with at least one strain in the stool. Recombinant strains were found in the 5 of 9 CNS samples, and some of these recombinants were of the Sabin 3 origin. Transgenic mice neurovirulence tests showed that al l viral strains, including nonrecombinants, lost attenuation, demonstrating that nonrecombinants as well as recombinants may be involved in the pathogenesis of VAPP. The important finding was that the viral stool samples did not always match the viral CNS samples, indicating that the stool sample are not necessarily representative of the etiological agent in the CNS.

This was an excellent study except for a few limitations that resemble previous research. Plaque purification and cell lines were used to isolate the viral strains, so replication and the potential for further genomic alteration was present . PCR has the capability of isolating certain strains in mixed samples without having to grow or purify viral types. Transgenic mice have been shown to be an unreliable animal model for neurovirulence (17). This study did have a striking finding that vira l samples from the CNS and stool were not always the same.

Furione et al. (24)

Furione et al. (24) did an excellent study on recombination in 70 VAPP cases. Isolates were passed over cell lines and plaqued purified. PCR, restriction fragment length polymorphism, and monoclonal antibody tests were performed to identify strains and particular recombinations. Nucleotide sequencing was used to confirm all recombination findings. Twenty two type 3 strains were found, and 12 of those had recombinant genomes and many different mutations. The author concluded that their results indic ate recombination with or without mutations might be a mechanism of attenuated poliovirus reverting back to neurovirulence.

Critique

Cell lines and plaque purification were used to isolate the viral samples, introducing the potential of genetic change in vitro. The testing was extensive and thorough, but it would have been interesting to see the CNS strains compared to stool iso lates from these VAPP patients, demonstrated in Georgescu et al. (23). OPV should have been passed over cell lines to control for recombination in vitro. Healthy vaccinees or contacts should have been tested to control for normal vaccine replication.

Immunodeficiency is a possible host factor that increases the risk of VAPP, but normal healthy children are still at risk for VAPP. VAPP cases in healthy children comprise a majority of the cases per year, suggesting the involvement of a host f actor other than immunodeficiency. This factor could be beyond our ability to detect right now, but most studies have neglected virus in the blood and CNS of both healthy and VAPP children. It is possible that different viral strains and mutations may be found in the blood compared to those found in the stool, as demonstrated in the studies above comparing the stool and CNS samples (23).

Another potential host factor involved in VAPP may be a large ratio of the patient’s mucosal to humoral immunity. For example, a patient may express excellent mucosal but poor humoral immunity. If the ratio is large, the mucosal immunity could caus e the virus to revert to a more virulent strain. Once these more harmful strains enter the blood, the ability to fight off viremia and the neuropathicity is lost, leaving the CNS vulnerable.

The competitions theory has not been addressed in VAPP cases, and I believe it could hold the foundation of many theories described in this paper. Viruses and other microorganisms provide selection pressures for growth in the GI tract. In the U S, the main concern of competition is among the three Sabin viruses because on average we have fewer microorganisms and viruses replicating in the GI tract than those in developing countries. Maldonado et al (34) found that Sabin 1 and 2 grew well togethe r, but Sabin 2 and 3 did not. It appeared Sabin 2 provided selection pressures for the growth of Sabin 3, and immunogenicity of Sabin 3 was compromised. The exact mechanism of competition is not known, but this finding can be an important hypothesis for S abin 3’s role in VAPP. If Sabin 3 has a hard time growing in the presence of Sabin 2, then it is natural for the viruses to mutate to grow better. If the selection pressures are strong, more mutations, alterations in the IRES site, and potential recombina tions could occur in type 3 as it replicates. Since immunogenicity is compromised for Sabin 3, than the immune system could be more vulnerable to these mutant viruses produced from competition. Those that have a large mucosal/humoral ratio as described ab ove may be the ones that are overwhelmed by the nonattenuated viruses consisting of recombinants and/or mutants.

If I had the chance to design a study, I would not limit myself to the above theories. I would try to encompass as many as I could. Statistical analysis would be performed before the study to see how large of pool would be needed to effectiv ely encompass each of the theories involved. Controls would be designated before the tests were performed. The studies above focused too much on the VAPP children without healthy controls, ignoring the pathogenesis of the vaccine in the healthy. Blood, st ool samples, and general health characteristics would be taken sequentially during a mass or individual vaccination campaign. All tests would employ direct PCR of stool and blood samples to avoid the cell line mutation problem present in most of the previ ous studies. Recombination, mutation analysis, competition, and immunogenicity of the host would be tested, and the results will be compared to find significant correlations regarding the pathogenesis of Sabin 3 in VAPP.

Research on the pathogenesis of Sabin 3 in VAPP is progressive, but there are many areas in all of the studies that need to be improved. Passing viruses through any cell line adds a potential variable of genetic recombination. Results attained from cell line isolates do not always represent the original viral sample. Techniques are capable of detecting virus in original samples without adding this variable. Research should direct more of its attention away from the viral factors and towards hos t factors involved in VAPP. Viral factors should not be ignored, but neither should the host factors. The unifying theory synthesizes many of the previous theories and takes them a step further into host and environmental factors. The larger issue is kept at the forefront without focusing too much or too little on the specifics.

1. Wright PF, Kim-Farley RJ, deQuadros CA, Robertson SE, Scott RN, Ward NA, Henderson RH. Strategies for the global eradication of poliomyelitis by the year 2000. NEJM 1991;325:1774-9.

2. Salk J and Salk D. Control of influenza and poliomyelitis with killed vaccines. Science 1977; 195:834-837.

3. Sabin, AB. Paralytic poliomyletis: Old dogmas and new perspectives. Rev Infect Dis 1981; 3:543-564.

4. Institute of Medicine, National Academy of Sciences. Vaccine supply and innovation. National Academy Press, Washington, D.C. 1985; pp 1-210.

5. Gefland HM, Potash L, Leblanc DR, et al. Intrafamilial and interfamilial spread of living vaccine strains of polioviruses. JAMA 1959; 170:2039-2048.

6. Benyesh-Melnick M, Melnick JL, Rawls WE, Wimberly I, Barrera-Oro J, Benporath E and Rennick V. Studies of the immunogenicity, communicability, and genetic stability of oral polio vaccine administered during the winter. Am J Epidem 1967; 86:11 2-136.

7. Horstmann DM, Emmons J, Gimpel L, Subrahmanyan T and Riordan JT. Enterovirus surveillance following a community-wide oral poliovirus vaccination program: A seven-year study. Am J Epidem 1973; 97:173-186.

8. Center for Disease Control and Prevention. Certification of poliomyelitis eradication - the Americas, 1994. MMWR 1994;43:720-2.

9. Andrus JK, de Quardros CA, Olive JM. The surveillance challenge: final stages of eradication of poliomyelitis in the Americas. MMWR 1992;41:21-6.

10. Prevots RD, Sutter RW, Strebel P, Weibel RE, Cochi SL. Completeness off reporting for paralytic poliomyelitis, United States, 1980 through 1991. Arc Ped Adolesc Med 1994; 148:479-85.

11. Nkowanw BM, Wassilak SGF, Orenstain WA, Bart KJ, Schonberger LB, Hinman AR. Vaccine-associated paralytic poliomyelitis - United States: 1973-1984. JAMA 1987;257:1335-40.

12. Strebel PM, Sutter RW, Cochi SL, Biellik RJ, Brink EW, Kew OM, Pallansch MA, Orenstain WA, Hinman AR. Epidemiology of poliomyelitis in the United States one decade after the last reported case of indigenous wild virus-associated disease. Clin In fect Dis 1992;12:568-79.

13. Almond JW, Cann AJ, Minor PD, Reeve P, Schild GC, Hauptmann R, Stanway G. Nucleotide sequence from neurovirulent and attenuated strains of type 3 poliovirus. Rev Infect Dis 1984;6:S487-93.

14. Evans D, Dunn G, Minor PD, Schold GC, Cann AJ, Stanway G, Almond JW, Currey K, Maizael Jr JV. Increased neurovirulence associated with a single nucleotide change in a noncoding region of the Sabin type 3 polio vaccine genome. Nature 1985;314 :548-50.

15. Minor PD, John A, Ferguson M, Icenogle JP. Antigenic and molecular evolution of the vaccine strain of type 3 poliovirus during the period of excretion by a primary vaccinee. J Gen Virol 1986;67:693-706.

16. Tatem JM, Weeks-Levy C, Mento ST, DiMechele SJ, Georgiu A, Waterfield WF, Sheip B, Costalas C, Davies T, Ritchey MB, Cano FR. Oral poliovirus vaccine in the United States: molecular characterization of Sabin type 3 After replication in the gut of v accinees. Journal of Medical Virology 1991;35: 101-109.

17. Gromeier, Matthias, Louis Alexander, Eckard Wimmer. Internal ribosomal entry site substitution eliminates neurovirulence in intergeneric poliovirus recombinants. Proc. Natl. Acad. Sci. USA. March 19, 1996. (93), pp 2370-2375.

18. Friedrich F; Da-Silva EF; Schatzmayr HG. Type 2 poliovirus recombinants isolated from vaccine-associated cases and from healthy contacts in Brazil. Acta Virologica, 1996 Feb, 40(1):27-33.

19. Miller MA; Sutter RW; Strebel PM; Hadler SC. Cost -effectiveness of incorporating inactivated poliovirus vaccine into the routine childhood immunization schedule. JAMA 1996 Sep 25, 276(12):967-71.

20. Izurieta HS; Sutter RW; Baughman AL; Strebel PM; Stevenson JM; Wharton M. Vaccine-associated paralytic poliomyelitis in the United States: no evidence of elevated risk after simultaneous intramuscular injections of vaccine. Pediatric Infectious Disease Journal, 1995 Oct, 14(10):840-6.

21. Georgescu MM; Delpeyroux F; Crainic R. Tripartite genome organization of a natural type 2 vaccine/nonvaccine recombinant poliovirus. Journal of General Virology, 1995 Sep, 76 ( Pt 9):2343-8.

22. Andrus JK; Strebel PM; de Quadros CA; Olive JM. Risk of vaccine-associated paralytic poliomyelitis in Latin America, 1989-91. Bulletin of the World Health Organization, 1995, 73(1):33-40.

23. Georgescu MM; Delpeyroux F; Tardy-Panit M; Balanant J; Combiescu M; Combiescu

AA; Guillot S; Crainic R. High diversity of poliovirus strains isolated from the central nervous system from patients with vaccine-associated paralytic poliomyelitis. Journal of Virology, 1994 Dec, 68(12):8089-101.

24. Furione M; Guillot S; Otelea D; Balanant J; Candrea A; Crainic R. Polioviruses with natural recombinant genomes isolated from vaccine-associated paralytic poliomyelitis. Virology, 1993 Sep, 196(1):199-208.

25. Chumakov KM. PCR engineering of viral quasispecies: a new method to preserve and manipulate genetic diversity of RNA virus populations. Journal of Virology, 1996 Oct, 70(10):7331-4.

26. Lu Z; Asher DM; Levenbook IS; Chumakov KM. Succession of mutations in the Sabin strain of type 3 poliovirus replicating in the central nervous system of monkeys. Virology, 1996 Jun 15, 220(2):285-9.

27. Rezapkin GV; Norwood LP; Taffs RE; Dragunsky EM; Levenbook IS; Chumakov KM. Microevolution of type 3 Sabin strain of poliovirus in cell cultures and its implications for oral poliovirus vaccine quality control. Virology, 1995 Aug 20, 211(2): 377-84.

28. Chumakov KM; Dragunsky EM; Norwood LP; Douthitt MP; Ran Y; Taffs RE; Ridge J; Levenbook IS. Consistent selection of mutations in the 5'-untranslated region of oral poliovirus vaccine upon passaging in vitro. Journal of Medical Virology, 1994 Jan, 42(1):79-85.

29. Lu Z; Douthitt MP; Taffs RE; Ran Y; Norwood LP; Chumakov KM. Quantitative aspects of the mutant analysis by PCR and restriction enzyme cleavage (MAPREC). PCR Methods and Applications, 1993 Dec, 3(3):176-80.

30. Chumakov KM; Norwood LP; Parker ML; Dragunsky EM; Ran YX; Levenbook IS. RNA sequence variants in live poliovirus vaccine and their relation to neurovirulence. Journal of Virology, 1992 Feb, 66(2):966-70.

31. Paralytic Poliomyelitis - United States, 1980-1994. MMWR. January 31, 1997. 46(4), pp. 79-83.

32. Old M, L Logan, Maldonado Y. A unique nucleic acid probe for screening Sabin 3 poliovirus revertants. Western Regional Conference for Pediatric Research. Feb 7, 1997.

33. Li J, ZhangLB; Yoneyama T; Yoshida H, Shimizu H; Yoshii K, Hara M; Nourvra T; Yoshikura H; Miyamura T; et al. Genetic basis of the neurovirulence of type 1 polioviruses isolated from vaccine- associated paralytic patients. Archives of Virology. 1996, 141 (6): 1047-54.