Synthetic Biology: Synthesis and Modification of a Chemical Called Poliovirus

S. Mueller, J. R. Coleman, J. Cello, A. Paul, and E. Wimmer

School of Medicine, SUNY–Stony Brook, Stony Brook, NY 11974, U.S.A.

D. Papamichail and S. Skiena

Dept. of Computer Science, SUNY–Stony Brook, Stony Brook, NY 11974, U.S.A.

1.   Introduction

Synthetic biology is a newly emerging scientific discipline, encompassing knowledge of different disciplines such as engineering, physics, chemistry, computer sciences, mathematics, and biology.¹ Synthetic biology aims to create novel biological systems with functions that do not exist in nature. There seem infinite possibilities of constructing unique derivatives of existing organisms (bacteria, yeast, viruses). Apart from designing novel building blocks for engineering biological systems, a fundamental requirement in synthetic biology is the ability of large-scale DNA synthesis and DNA sequencing.

Viruses can be described in chemical terms; the empirical formula of the organic matter of poliovirus being²

C_332,652H_492,388N_98,245O_131,196P_7,501S_2,340.

There might be little practical use in describing poliovirus by an empirical formula, but it persuasively portrays the virus as a chemical. Placing these atoms into order, particles of high symmetry emerge,³ see Fig. 1. These particles can be purified and crystallized.

Figure 1.  Computer model of poliovirus, generated from x-ray crystallographic data.³

The structure of the particle and the strategy for proliferation are encoded in the viral genome, which is a single stranded nucleic acid (RNA) of about 7,500 nucleotides – see Fig. 2. We have determined the sequence of this viral genome in 1981,⁴ which was a key to understanding aspects of the molecular biology of poliovirus proliferation.

Whereas the poliovirion is a beautiful molecular entity, it has frightening properties as a human pathogen. Infecting by the oral-fecal route, the virus proliferates very efficiently in the gastrointestinal tract. Rarely, the virus finds itsway into the central nervous system (CNS), where it destroys with cunning specificity motor neurons that control muscle movement. This results in irreversible paralysis and sometimes death, a disease called poliomyelitis.⁵ The virus has caused horrific epidemics in the first part of last century until two excellent vaccines were developed. The killed poliovirus vaccine (IPV) by Jonas Salk, and the live oral vaccine (OPV) by Albert Sabin now effectively control the disease.⁵ The fact that humans are the only reservoir for the poliovirus and the efficacy of the vaccines has prompted the World Health Organization to attempt eradicating the agent from the globe. Although great progress has been made over a period of eighteen years, poliovirus has shown great resilience and a successful end to the eradication campaign is not yet in sight.

Figure 2. Nucleotide sequence of the poliovirus genome.⁴

In the early 1990’s, the intriguing dual nature of poliovirus as an inanimate entity, as well as a replicating organism, has led us to ask the question whether the virus can be synthesized in the test tube. To be sure, nucleic acid synthesis by then was already developed to a stage where the DNA of individual genes could be assembled from their known sequence. But the synthesis of a replicating entity, requiring the assembly of nucleic acid much larger than had been reported before, was unprecedented. It provided the proof of principle that viruses can be considered to be chemicals, which can be regenerated in the absence of a natural template outside living cells.⁶ Moreover, we were probably amongst the first to realize that the advances in nucleic acid synthesis could pose a threat if misused for the synthesis of bioterrorist agents. The Defense Advanced Research Project Agency (DARPA) had come to the same conclusion and supported the synthesis. It was meant as a “wake up call” to pinpoint the dual nature of advances in biotechnology.⁶

We are now studying the possibility of generating via whole genome synthesis novel polioviruses whose ability to proliferate in the CNS is debilitated, whereas its efficiency to replicate in tissue culture cells remains largely unchanged. The basis of engineering new polioviruses is altering codon usage of the viral mRNA.⁷

2.   The synthesis of poliovirus

At the present time, RNA of the size of the poliovirus genome cannot be synthesized. The DNA-complement of the RNA, however, can be assembled from small oligonucleotides, available from biotech companies. As shown in Fig. 3, the assembly of DNA fragments proceeds via the double-stranded form, allowing subsequent transcription of the DNA with a RNA transcriptase⁸ into synthetic, infectious poliovirus genomic RNA.⁸ Incubation of the RNA in a cell-free extract leads to the translation and replication of the genome until sufficient quantities of components are synthesized for the spontaneous assembly of virions to occur.²

The identity of the virus was proven by a variety of biological and serological methods. For example, the synthetic virus was shown to produce plaques on monolayers of human cells that could be prevented by anti sera specific to the virus.⁶ Moreover, the virus produced disease symptoms in experimental animals (transgenic mice) indistinguishable from those of wild type poliovirus.⁶ However, because the synthetic virus was genetically marked through the intentional introduction of 27 nucleotide changes, the virus proved “attenuated” (less virulent), which we later showed to be due to the exchange of a single nucleotide within the 7500 nucleotide long genome.⁹

Figure 3. Synthesis of poliovirus in the absence of a natural template.⁶ The sequence derived from the Internet⁴ was converted to a DNA sequence and divided into small oligonucleotides. These were obtained from a biotech company and assembled to double stranded DNA. The DNA was enzymatically transcribed into infectious synthetic poliovirus RNA⁸ that, in turn, was used to generate poliovirus in an extract of human cells.² The newly synthesized virus grew on a layer of human cells, forming plaques. Reproduced with permission from Mueller et al.¹⁰

3.   The response to the poliovirus synthesis

The synthesis of a replicating biological entity in the absence of the natural template was without precedence at the time of publication and provoked unusually strong and conflicting responses. As summarized in Table 1, it was praised as landmark in biology, perceived as a challenge to divine power, scorned as a stunt, considered to wreck poliovirus eradication, and condemned as dangerous to national security.

A major reason for the intense discussions surrounding the publication⁶ by Cello et al. in 2002 was the timing of publication. Following the terrorist assault of September 11, 2001, and the anthrax bioterrorist attack in 2001–2002, the publication reached a highly sensitized public, particularly in the U.S.A. Ethical considerations, e.g. whether viruses are chemicals amenable to synthesis and modification or living entities, were less of an issue. Instead, the synthesis was considered almost exclusively in the context of bioterrorism. It was concluded, correctly, that bioterrorist agents, such as smallpox virus, could in the future be synthesized from information available on the Internet. Indeed, a discussion of this possible threat was included into the original draft of our Cello et al. manuscript submitted to Science in 2002.⁶ Regrettably, the editors of Science eliminated a passage referring to possible dual use of modern technology from our original manuscript. We emphasize that the threat of synthesizing dangerous bioterrorist agents for malicious intent did not arise from the publication of the poliovirus synthesis. This threat is intrinsic to the recent advances in biotechnology.^6,11 However, we believe that it is important to bring these issues to the attention of the public, as we disagree with suggestions that the poliovirus synthesis should have been kept a secret. A detailed discussion of the responses to the poliovirus synthesis listed in Table 1 can be found in Ref. 11.

Table 1. Responses to the poliovirus synthesis.

4.   Application of the strategy of whole genome synthesis to viruses

Synthetic biology strives to generate novel biological systems that do not exist in nature, mainly with medical or commercial applications in mind. The poliovirus synthesis as described above does not fall into this category since it resulted in poliovirus with nearly identical phenotypes as the model wild-type virus (as mentioned, only the pathogenic phenotype of the synthetic virus in transgenic mice was different because of nucleotide changes introduced intentionally as genetic markers^6,9).

Recently, we have generated novel polioviruses with vastly different phenotypes when compared to wild-type poliovirus.⁷ This was done by chemical/biochemical synthesis of very large segments of the viral genome. In these segments (~3,000 nucleotides), synonymous codon usage was either deoptimized or the position of synonymous codons was randomized. This strategy is based on the fact that all amino acids but two are encoded by more than one codon (there are 20 amino acids and 64 codons). However, the frequency of synonymous codon use for each amino acid is unequal and, overall, the frequencies have co-evolved with the cell’s translation machinery to avoid excessive use of suboptimal codons. Changing synonymous codons altogether or randomizing them can unbalance the synthesis of proteins without changing the amino acid sequence of the protein. Hence, the product of the translational machinery remains the same while the efficiency of protein synthesis may be vastly changed. Thus, regardless of the changes introduced into the genome, a virus synthesized in the infected cell will have the same structure and will encode the same replication proteins as the virus with the wild-type nucleotide sequence. This means that the virus variant with the altered genome will infect and replicate in the same cells as the wild-type virus, but it may be significantly disadvantaged when it comes to proliferation. Such a variant of a human pathogenic virus may enter the host by its normal route, replicate poorly, but still allow the host to mount a sufficiently strong immune response to clear the infection and develop lasting immunity. In other words, a human virus with altered codon usage or codon pairs could possibly serve as vaccine.

We have made use of the two possibilities to change codons in the (messenger RNA) template for translation without changing the sequence of the translation product.⁷ One is to shuffle existing codons to maximize their positional changes while codon usage (the overall frequency of synonymous codons in the viral mRNA) is maintained. The extent of codon “shuffling”, which we have applied, did not exert any influence on poliovirus proliferation, regardless of whether the assays were conducted in tissue culture cells or in poliovirus-sensitive transgenic mice.⁷ On the other hand, changes in the nature of codons (exchanges of commonly used codons to codons that are rarely used) had a profound effect on viral infectivity:⁷ the ratio of particles to infectious units (the measure of how many particles are necessary to achieve productive infection of one cell) was changed from 100:1 to 100,000:1. However, once a successful infection was achieved, the virus managed to produce per cell as many particles as wild-type virus.⁷ Experimental evidence clearly showed that the reason for the altered particle to infectious units ratio was due to a translational handicap early in the infectious cycle.⁷

By fine-tuning the relationship of translation to pathogenesis, a virus variant may be developed with properties useful for vaccine application. Viral vaccines generated by large-scale genome alteration have the advantage over conventional live virus vaccines in that, due to the large-scale codon alterations (changes in the nucleotide sequence of the viral mRNA), “reversion” of the genetic information from attenuation to virulence in vaccine recipients is all but impossible.

5.   Conclusion

The synthesis of poliovirus in the absence of natural template was a landmark experiment without precedence. Although it had been predicted prior to its publication, the experiment has caused unusual responses ranging from ethical to political issues. It was largely overlooked that a virus was regenerated (“recreated”) outside living cells from written information in the public domain, a new reality that reduced viruses to chemicals. Since viruses assume properties of living entities upon entering the cell – heredity, genetic variation, selection towards fitness, evolution into different species, and even some form of sex (exchange of genetic material with closely related viruses through recombination) – we consider viruses as “chemicals with a life cycle”.

Whole scale genome synthesis of viruses can now be easily realized with rapidly developing new technologies of DNA synthesis.¹² We believe that there is great promise in using this strategy to developing novel viral derivatives useful in medicine. Fully exploiting the potential of genome-level synthesis will require a new generation of sequence design tools. We have developed algorithms for designing optimal synthetic coding sequences under a variety of criteria, including avoiding undesired subsequences¹³ and RNA secondary structure¹⁴ or interleaving in overlapping genes on alternate reading frames (B. Wang, D. Papamichail, S. Mueller, and S. Skiena, unpublished results). Undoubtedly, the new technologies could be misused for malicious intent from which only powerful and open research can protect us.¹¹ The experiments that we have described in this chapter⁷ are but the beginning of exploring the vast space of genetic information for the generation of novel kinds of vaccines. This, indeed, connects our work with Synthetic Biology.

Acknowledgements

This work was supported in part by grants from the National Institutes of Health and the Defence Advanced Research Project Agency (to E. W.) and from the National Science Foundation (to S. S.).

References

1. D. Endy, “Foundations for engineering biology,” Nature 438, 449 (2005).
2. A. Molla, A. V. Paul, and E. Wimmer, “Cell-free, de novo synthesis of poliovirus,” Science 254, 1647 (1991).
3. J. M. Hogle, M. Chow, and D. J. Filman, “Three-dimensional structure of poliovirus at 2.9 Å resolution,” Science 229, 1358 (1985).
4. N. Kitamura, B. L. Semler, P. G. Rothberg, et al., “Primary structure, gene organization and polypeptide expression of poliovirus RNA,” Nature 291, 547 (1981).
5. S. Mueller, E. Wimmer, and J. Cello, “Poliovirus and poliomyelitis: A tale of guts, brains, and an accidental event,” Virus Research 111, 175 (2005).
6. J. Cello, A. Paul, and E. Wimmer, “Chemical synthesis of poliovirus cDNA: Generation of infectious virus in the absence of natural template,” Science 297, 1016 (2002).
7. S. Mueller, D. Papamichail, R. J. Coleman, S. Skiena, and E. Wimmer, “Reduction of the rate of poliovirus protein synthesis through large scale codon deoptimization causes attenuation of viral virulence by lowering specific infectivity,” J. Virology 80, 9687 (2006).
8. S. van der Werf, J. Bradley, E. Wimmer, F. W. Studier, and J. J. Dunn, “Synthesis of infectious poliovirus RNA by purified T7 RNA polymerase,” Proc. Natl. Acad. Sci. U.S.A. 78, 2330 (1986).
9. N. DeJesus, N., D. Franco, A. Paul, E. Wimmer, and J. Cello, “Mutation of a single conserved nucleotide between the cloverleaf and internal ribosome entry site attenuates poliovirus neurovirulence,” J. Virology 79, 14235 (2005).
10. S. Mueller, P. Jiang, E. Rieder, et al., “Pathogenesis and prevention of poliomyelitis and the chemical synthesis of poliovirus,” Nova Acta Leopoldina NF 92, Nr. 344 (2005), pp. 35–43.
11. E. Wimmer, “The test–tube synthesis of a chemical called poliovirus. The simple synthesis of a virus has far reaching societal implications,” EMBO Reports 7:S3-S9 (2006).
12. J. Tian, H. Gong, N. Sheng, X. Zhou, E. Gulari, X. Gao, and G. Church, “Accurate multiplex gene synthesis from programmable DNA microchips,” Nature 432, 1050 (2004).
13. S. Skiena, “Designing better phages,” Bioinformatics 17, 253 (2001).
14. B. Cohen and S. Skiena, “Natural selection and algorithmic design of mRNA,” J. Comput. Biol. 10, 419 (2003).