Synthesizing a virus from scratch

[jackmell]

I saw Dr. Oz yesterday. His guest suggested that graduate students in five years would be capable of synthesizing smallpox from scratch. I think he was exaggerating right? About 7 years ago it took 2 years to synthesize polio from it's 7000 nucleotide recipe and it was thousands of times less virulent than the wild form. Smallpox has about 170,000 nucleotides. Has molecular biology changed that much in the interim period? I assume it doesn't scale linearly but rather would be much, much more difficult.

How difficult is it to synthesize a single strand of DNA with 170,000 base pairs without so much as a single error?

 

[mgb_phys]

You don't necessarily need to arrange each base pair, even in a simple virus the genome has a lot of well known sub unit.s

If you put dna synthesize into google it actually shows ads for commercial services, 50K pairs for 39c/base

 

[jackmell]

Thanks. I'm just apprehensive to actually e-mail them and ask if one that long could be made since it's a business and I really have no business interest in the matter but rather just curious. However from what you said, seems you can start with large pieces and connect them properly.

Can anyone here explain why the synthetic polio virus was thousands of times less virulent then the wild form? Obviously they didn't get it all right. Sounds to me the made some mistakes with the sequence or some other component that wasn't the same.

 

[Ygggdrasil]

When the scientists synthesized the polio virus, they introduced a handful of changes to the sequence so that these mutations could be used as genetic markers. It turns out that one of these mutations greatly decreased the virulence of the virus. See de Jesus et al. 2005. J. Virol. 79:14235-43 (free full text available here: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1280220)

 

[jackmell]

When the scientists synthesized the polio virus, they introduced a handful of changes to the sequence so that these mutations could be used as genetic markers. It turns out that one of these mutations greatly decreased the virulence of the virus. See de Jesus et al. 2005. J. Virol. 79:14235-43 (free full text available here: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1280220)

Hi. Thank you. I found it interesting that it was found to be due to the substitution of only two base pairs (UA for GG)) and when those changes were removed, the virulence was restored:

"Conversely, the exchange of GG to wild-type (wt) UA at 102/103 in an sPV1(M) background restored wt neurovirulence in CD155 transgenic (tg) mice and suppressed the ts phenotype in SK-N-MC cells. "

I think that's what it means. May spend some more time with the article.

 

[sganesh88]

Sorry if this seems irrelevant. Do you guys say that we have synthesised a living form like a virus by arranging non-living nucleotides?

 

[jackmell]

To be alive it has to be self-contained, metabolizing, reproducing, and evolving. Viruses do not metabolize.

 

[Ygggdrasil]

Scientists at the J. Craig Venter Institute have been able to synthesize and assemble the entire genome of a bacterium (582,970 base pairs). Of course, in the case of bacteria, synthesizing the DNA is not sufficient to create a synthetic life form (for example, you'd need to synthesize the rest of the component of the cell and somehow jump start the metabolic processes). However, this research could allow one to customize microbial genomes for biotechnological purposes or (to get back to the original topic) synthesize viruses with larger genomes such as smallpox.

Reference: Gibson et al. 2008. Complete Chemical Synthesis, Assembly, and Cloning of a Mycoplasma genitalium Genome. Science 319: 1215 - 1220. http://dx.doi.org/10.1126/science.1151721.

 

[SW VandeCarr]

Scientists at the J. Craig Venter Institute have been able to synthesize and assemble the entire genome of a bacterium (582,970 base pairs). Of course, in the case of bacteria, synthesizing the DNA is not sufficient to create a synthetic life form (for example, you'd need to synthesize the rest of the component of the cell and somehow jump start the metabolic processes). However, this research could allow one to customize microbial genomes for biotechnological purposes or (to get back to the original topic) synthesize viruses with larger genomes such as smallpox.

Reference: Gibson et al. 2008. Complete Chemical Synthesis, Assembly, and Cloning of a Mycoplasma genitalium Genome. Science 319: 1215 - 1220. http://dx.doi.org/10.1126/science.1151721.

Interesting. Will synthetic viral nucleic acids self-assemble the normal viral protein structures in the proper cellular medium?

 

[Ygggdrasil]

With polio, the proteins self assemble with the nucleic acids to form infectious viral particles. Polio is a bit more simple than other viruses, however, because it is a non-enveloped virus and can be produced in a cell-free system. I believe that some viruses can be produced simply by injecting the virus's DNA or RNA into cells.

 

[Borek]

I believe that some viruses can be produced simply by injecting the virus's DNA or RNA into cells.

I thought that's the basic mechanism of virus reproduction - inject your DNA/RNA into the cell, everything will take care of itself.

 

[SW VandeCarr]

I thought that's the basic mechanism of virus reproduction - inject your DNA/RNA into the cell, everything will take care of itself.

This is true for natural viruses. However, I was wondering if there were possibly epigenetic factors in natural viruses that would not be present in synthetic varieties. In other words, is it sufficient to just get the nucleotide sequence right?

 

[jackmell]

May I ask how do the ten proteins making up the polio capsid know how to "find" each other as they come off the ribosome and proceed then to encapsulate a copy of the viral RNA? Does the ribosome complex "hold" onto them in the proper configuration so that they can bond? Are they just released into the cytoplasm and find each other randomly? Is the specific sequence in which they are synthesized important for this recombination? Does protein one bond to protein two then these in turn bond to protein three and so forth? If I just drop the ten proteins and viral RNA into a suitable bath, will they self-organize into the capsid? Is the viral RNA used as scaffolding to hold the proteins in place for this assembly?

 

[SW VandeCarr]

May I ask how do the ten proteins making up the polio capsid know how to "find" each other as they come off the ribosome and proceed then to encapsulate a copy of the viral RNA? Does the ribosome complex "hold" onto them in the proper configuration so that they can bond? Are they just released into the cytoplasm and find each other randomly? Is the specific sequence in which they are synthesized important for this recombination? Does protein one bond to protein two then these in turn bond to protein three and so forth?

Good questions, but not my question. I asked if synthetic viruses are in fact identical to complete natural viruses. Are epigenetic factors involved at the level of viruses?

If I just drop the ten proteins and viral RNA into a suitable bath, will they self-organize into the capsid? Is the viral RNA used as scaffolding to hold the proteins in place for this assembly?

So, what's the answer?

EDIT: Sorry. I didn't check to see that you are the OP. Anyway, perhaps someone can shed light on both of our questions.

 

[Ygggdrasil]

I thought that's the basic mechanism of virus reproduction - inject your DNA/RNA into the cell, everything will take care of itself.

It is in some cases, but not all. Various viral proteins are often required to convert the viral genome into a different form so that the next step of viral replication can occur. For example, in order for HIV and other retroviruses to replicate, it must first reverse transcribe its genome into DNA (a process that requires the viral enzyme reverse transcriptase) and then splice that DNA into the host's genome (a process that requires the viral enzyme integrase). The HIV genome itself, however, is an mRNA molecule, so if you inject enough of it into a cell, the cell might be able to read the mRNA and produce the viral proteins required for infection. However, cells do maintain defense mechanisms against viruses, and many viral proteins help to protect the viral genome from these defense mechanisms. Thus injecting just the genome into a cell would be much less efficient at infecting cells than using virions.

Influenza and other (-)stranded RNA viruses have an even bigger problem as their RNA genome is not an mRNA, meaning that its RNA does not contain the coding sequence for its proteins (the sequence is the reverse complement of the coding sequence). Normally, influenza uses a viral RNA-dependent RNA-polymerase to convert its genome into an mRNA that gets read by the cell to produce new viruses. Without such an enzyme present, it is unlikely that the influenza genome alone would be able to replicate itself inside of a cell.

If one wants to get around these problems to produce synthetic viruses, one could first engineer cells to produce these viral factors. Then when the viral genomes are introduced into the cells, the genomes would have all the right factors present to get processed correctly and start producing new virions.


As for polio, I believe the capsid proteins are capable of self-assembling with the viral RNA on their own. Self-assembly of complex structures is certainly possible, just look at all of the complex structures that we have been able to engineer using techniques such as DNA origami. I'll have to look up some references to see what's going on in the case of polio.

 

[Ygggdrasil]

May I ask how do the ten proteins making up the polio capsid know how to "find" each other as they come off the ribosome and proceed then to encapsulate a copy of the viral RNA? Does the ribosome complex "hold" onto them in the proper configuration so that they can bond? Are they just released into the cytoplasm and find each other randomly? Is the specific sequence in which they are synthesized important for this recombination? Does protein one bond to protein two then these in turn bond to protein three and so forth? If I just drop the ten proteins and viral RNA into a suitable bath, will they self-organize into the capsid? Is the viral RNA used as scaffolding to hold the proteins in place for this assembly?

The mechanism for the assembly of polioviruses is not completely understood, but I'll summarize what I've been able to find out about the topic. At the end of the post, I provide a review paper from the scientific literature discussing poliovirus assembly.

Although the poliovirus encodes ten different proteins, the capsid (the protein shell that encases the viral genome) is composed of only four different viral proteins, numbered VP1-4. 60 copies of each of these four proteins form the icosahedral capsid.

One important fact is that all ten polioviral proteins are synthesized together on one single polypeptide chain. This polypeptide chain undergoes a series proteolytic cleavage events that liberate the individual proteins. All of the proteins that form the capsid are located together in the N-terminal portion of this polypeptide chain (called P1). Directly next to P1 is one of the viral proteases, 2A. Although the exact order of cleavage events is not known, it is likely that P1 is rapidly cleaved from the rest of the polypeptide chain by an intramolecular cleavage catalyzed by the 2A protease.

The P1 protein contains all four of the viral capsid proteins (VP1-4) and forms the fundamental subunit for assembly of the viral capsid. Proteolysis of P1 by the viral protease 3CD cuts P1 into its individual subunits (VP0, VP1, and VP3) which form a noncovalent complex. Likely, these subunits are already arranged in the complex prior to cleavage. The cleavage, however, allows five of these VP0-VP1-VP3 complexes to come together to form a pentameric (VP0-VP1-VP3)₅ intermediate. These pentamers likely form via the random collision of VP0-VP1-VP3 complexes in the cytoplasm of infected cells. Twelve pentamers come together to form the full capsid. It is currently unclear whether the genomic RNA is encapsulated before or after the pentamers come together. Finally, in a fully formed viral particle containing viral RNA, a final proteolysis step cleaves the VP0 precursor into VP2 and VP4. The final result is the viral RNA (plus associated protein factors) encapsulated in a mature (VP1-VP2-VP3-VP4)₆₀ capsid.

I do not know if anyone has done the experiment where you separately produce VP1, VP2, VP3, and VP4, then drop them into a test tube to see if they self-assemble correctly. However, from what we know about other self-assembling systems (e.g. the ribosome), the answer is likely no. Many individual subunits in protein complexes are unstable without their binding partners, so it can be difficult to obtain these proteins isolated from each other. Also, it is likely that these proteins would not assemble correctly if you just threw them together. Having VP1-4 linked together in the P1 polyprotein likely helps the proteins fold and assemble correctly prior to their interactions with other subunits. Similarly, that pentamers can form only after P1 has been cleaved enforces more order on the assembly process and again likely helps to ensure that the VP0-VP1-VP3 complexes are properly folded and assembled before they interact with other VP0-VP1-VP3 complexes.

Reference:
Ansardi et al. 1996. Poliovirus Assembly and Encapsidation of Genomic RNA. Adv Virus Res 46: 1-68. http://dx.doi.org/10.1016/S0065-3527(08)60069-X"

 

[jackmell]

Thanks yggdrasil. You've been very helpful. It's a fascinating subject for me. :)

 

[Ygggdrasil]

Moore's law in action.
http://dx.doi.org/10.1126/science.1072266 - chemical synthesis of a 170,000 base pair genome
http://dx.doi.org/10.1126/science.1151721 - chemical synthesis of a 582,970 base pair genome
http://dx.doi.org/10.1126/science.1190719 - chemical synthesis of a 1,080,000 base pair genome.

 

[jackmell]

We're on the same wavelength Ygggdrasil as I was just reading the ABC article about the latest one:

http://abcnews.go.com/Health/Wellness/scientists-create-synthetic-cells/story?id=10708502&page=1

Thanks for your citation:

http://www.sciencemag.org/cgi/rapidpdf/science.1190719v1.pdf

:) Nice progress this field is making I think.

 

[ThomasEdison]

http://www.ted.com/talks/craig_venter_unveils_synthetic_life.html
Craig Venter!

Woohoo!

 

[Ygggdrasil]

May I ask how do the ten proteins making up the polio capsid know how to "find" each other as they come off the ribosome and proceed then to encapsulate a copy of the viral RNA? Does the ribosome complex "hold" onto them in the proper configuration so that they can bond? Are they just released into the cytoplasm and find each other randomly? Is the specific sequence in which they are synthesized important for this recombination? Does protein one bond to protein two then these in turn bond to protein three and so forth? If I just drop the ten proteins and viral RNA into a suitable bath, will they self-organize into the capsid? Is the viral RNA used as scaffolding to hold the proteins in place for this assembly?

I was doing research on an unrelated topic, but found a paper relating to the self assembly of viral capsids that I though others might find interesting. The paper describes various chemical systems that mimic the self-assembly of the icosahedral poliovirus capsid. Of particular interest, the authors constructed a toy capsid that self assembles when shaken in a container:

By appropriate placement of oriented magnets as analogs to the electrostatic complementarity, we produced a model that mimics the self-assembly of the [poliovirus] from twelve pentameric assembly intermediates. Placing 12 of these tiles in a container and shaking with the appropriate strength results in a stable closed shell, usually within 1–2 min. The key aspects of this model were the fivefold symmetric tiles, the appropriate curvature at the tile interfaces, and the geometric and magnetic complementarity of the interfaces. Although intellectually we knew that this type of self-organization occurs spontaneously, watching it happen from random shaking on the macroscopic scale was inspirational.

The videos of the process (freely available http://www.pnas.org/content/suppl/2007/12/05/0709489104.DC1/09489Movie1.mov, and http://www.pnas.org/content/suppl/2007/12/05/0709489104.DC1/09489Movie2.mov) is, as the authors say, inspirational. Movie 2 is especially impressive because they show two different capsids (colored red and green) with the same shapes but reversed magnet polarity can self-assemble in the presence of each other without the formation of misfolded states. You can access the full paper at the PNAS website for free by following the link below.

Olson, Hu, and Keinan. (2007) Chemical mimicry of viral capsid self-assembly. Proc. Natl. Acad. Sci. USA 104: 20731-20736. http://dx.doi.org/10.1073/pnas.0709489104.

 

[nismaratwork]

Thanks for the reading and videos Ygggdrasil, this impressive as hell. I suspect that this will be one of the keys to future medicine more so than the oft lauded nano-machines.

 

[bobze]

I think a couple of times now (as a proof is in the pudding kind of experiment) people have downloaded viral genomes and synthesized them.

Prior to a couple of years ago, many viral genomes were easily accessible on the net (dangerous viral genomes even to boot). Since then, the NIH/CDC/DoD has tried to implement a little tighter controls in the event of well funded, educated and motivated bioterrorists getting a hold of such information.

Really, any positive sense RNA virus would work. Because the positive sense RNA, reads the same as mRNA in a cell and is infectious. So Picornaviridae, like polio, or Togaviridae, like sinbdis or rubella are prime threats. Because of their small genome size and as I said, ability to infectious as naked RNA.

Most viruses (maybe all?) self-assemble. However, since many aren't positive sense RNA viruses, then their genomes are quite harmless without a delivery system (either the capsid/coat or envelope expressing viral coat proteins).

Negative sense RNA viruses must first be converted by RNA dependent RNA polymerase (RDRP) to positive sense RNA (mRNA). While DNA viruses must also be converted to RNA (DNA dependent RNA polyermase, DDRP). And finally the non-functional positive sense RNA viruses, which are the Lentiviridae, retroviruses (like HIV or SIV). Which have to first change their RNA to DNA, then back to RNA before proteins can be made, with RNA dependent DNA polymerase (RDDP).

So assembly of those viral genomes wouldn't necessarily result in viable virus.

The other good thing (at least from a bio-terrorism stand point) is viruses are actually very poor replicators. Many, many of the new "virons" (kinda like baby viruses, but since they don't grow their not really babies), are non-infections because of various assembly or copy errors.

Which means, the infectious dose of viruses is rather high (compared to say some bacteria where 5-10 cells can establish infection). Which also means, production of positive sense viral RNA probably isn't that effective for establishing infections in large populations.