In a medical microbiology textbook I'm reading (Murray et al, 1994), the authors state the following:

The retrovirus genome has a 5' cap and is polyadenylated at the 3' end. Although the genome resembles a messenger RNA (mRNA), it is not infectious because it does not encode for a polymerase that can directly generate more mRNA. The genome consists of at least three major genes that encode for the enzymatic and structural proteins of the virus: gag (group-specific antigen), pol (polymerase), and env (envelope). At the end of the genome are long terminal repeat (LTR) sequences

From this, I interpreted that because viral RNA doesn't encode an RNA polymerase, it is not itself infectious. However, if a cell were to be injected with retroviral single-stranded RNA alone, wouldn't this RNA be translated into proteins (including reverse transcriptase) as if it were normal mRNA into a protein? If this were correct, then the single-stranded RNA has all the information in itself to propagate the virus. Single-stranded RNA needs to be reverse-transcribed into the host's DNA before being transcribed into mRNA that can be translated.

Is there something different between viral RNA and the mRNA transcript that makes one translatable and the other not?

Source: Murray, Patrick R et al, Medical Microbiology (1994), Second Edition.


I contacted the author of this chapter (Dr. Ken Rosenthal) and gained some more insight into how this hypothetical (injecting plain retroviral RNA into a cell) would play out. His answer is that yes producing new viruses possible but not probable. The reasoning for this is as follows:

  • The main reason that a new virus would likely not be created is that for the genome to make a new virus, reverse transcriptase is needed. Without reverse transcriptase, cDNA can't be made. Suppose that a ribosome was able to make reverse transcriptase with naked retroviral RNA and this newly made reverse transcriptase stayed near the ribosome. It would have difficulties making new cDNA from the original RNA template. One of those challenges is that cDNA synthesis needs a specific primer on reverse transcriptase.
  • The translated retroviral RNA would likely be degraded due to their finite lifespan in the cytoplasm.
  • There would need to be enough viral proteins translated to form a capsid. This would be unlikely considering the lifetime of the RNA in
  • Even if there were enough proteins in the cytoplasm, these proteins would then have to remove the RNA from the ribosome to encapsidate it. This is also unlikely.

Considering this, the microinjection of naked retroviral RNA would be an inefficient process to cause an infection.

  • $\begingroup$ @F16Falcon — Could you indicate what relevance the article you cite has to this question? $\endgroup$ – David Jan 26 at 14:46
  • $\begingroup$ @F16Falcon — But retroviruses do not possess self-replicating RNA and depend on DNA for their replication. So as far as I can see the article — which is on "The RNA World" — however interesting, would seem irrelevant to the poster's concerns. $\endgroup$ – David Jan 27 at 10:16
  • $\begingroup$ So now compare the author's response to my answer. Despite my current score of –1, I do believe I was right. Perhaps you might consider giving credit where it is due and accepting it. $\endgroup$ – David Feb 2 at 19:29
  • 1
    $\begingroup$ Thanks. It’s the principle not the points (which I have little need of.) $\endgroup$ – David Feb 3 at 16:55
  • $\begingroup$ Sure thing, though I disagree with you on one particular point (namely that I shouldn't focus on hypotheticals and "injecting" RNA). I find hypotheticals a useful way to find interesting questions. All the best David. $\endgroup$ – Howsikan Feb 3 at 17:43

As far as I can understand, the poster is concerned about the statement:

Although the genome resembles a messenger RNA (mRNA), it is not infectious because it does not encode for a polymerase that can directly generate more mRNA.

I find this a somewhat unfortunate aside as the infectivity or not of viral RNA “injected into a cell” has no physiological relevance, and would advise the student to focus on the essential details of the actual viral life cycle rather than on this sort of thing.

However, the point it would seem that the poster and the other answer misses is that numbers count. A virus (or viral RNA) needs to generate enough progeny for infection to overwhelm antiviral defences. This entails there being sufficient mRNA generated from the genomic RNA (or DNA) for translation into protein.

In the case of ‘normal’ RNA viruses (i.e. non-retroviruses), this involves an RNA-dependent RNA polymerase, which has to be virally coded as the host cell does not have such an enzyme. Retroviruses, however, produce multiple mRNA molecules from an integrated DNA genome (integrated through the RNA-dependant DNA polymerase — reverse transcriptase) by using the cell’s DNA-dependant RNA polymerase, not from replicating the transcribed mRNA.

All viruses rely on host protein synthesizing systems to translate their mRNAs. But a single retroviral RNA that can act as both mRNA or (part of) the genome can only generate enough coat proteins etc. to encapsulate it (or enough reverse transcriptase to reverse copy it). A situation unlikely to cause retroviral RNA to be infectious in this hypothetical situation. I imagine it is this that the authors of the text have in mind.

The student of this subject would do well to distinguish the different transcriptional enzymes involved in viral replication and learn to refer to them by their full names.

Reference: Coffin et al. Retroviruses (1997), available on NCBI Bookshelf

  • $\begingroup$ RdRP is virally encoded, but is not necessarily carried in a virion. Thus, the replication of many positive-strand RNA viruses literally starts with translating their genome on a ribosome and thus generating RdRP. In other words, such an RNA introduced to a cell is literally infectious - which seems to be the premise of the question. $\endgroup$ – Vadim Jan 27 at 3:11
  • 2
    $\begingroup$ @Vadim — "many positive-strand viruses". But not retroviruses which do not encode an RNA-dependent RNA polymerase, which was the premise of the question. $\endgroup$ – David Jan 27 at 10:03
  • $\begingroup$ In other words, what counts is not the numbers, but how virus replicates itself and what genes it carries. Note that RdRP is not a precondition here: if retroviral RNA encoded the reverse transcriptase, the integrase, and the other proteins necessary for initiating the retroviral replication cycle, it would be still infectious. $\endgroup$ – Vadim Jan 27 at 10:27
  • 1
    $\begingroup$ @Vadim — There may be other reasons, but the authors of the book that the poster is trying to understand say that it is because of lack of a replicase or replication, from which one draws the inescapable conclusion that they regard the generation of many copies of a viral RNA as important for infection. So they think numbers are important, and it needed someone to explain that to the poster and those other six people who share his confusion. $\endgroup$ – David Jan 27 at 18:30
  • 1
    $\begingroup$ The lack of replication means no copies at all rather than few (vs. many) copies. In other words, it is qualitative rather than quantitative difference. In fact, in your earlier comments you acknowledge that much - it is the total absence of RdRP rather than its low efficiency. But the formulation in your answer is misleading. $\endgroup$ – Vadim Jan 27 at 19:57

In RNA viruses with a single-stranded genome, this RNA can be positive or negative sense. Positive sense RNA is directly translatable by a ribosome, while a negative strand RNA cannot be directly translated and therefore non-infectious. It either has to be converted to a positive strand RNA with the help of an RdRP (RNA-dependent RNA polymerase, carried in the virion) or reversely transcribed into a DNA which then serves as the template for generation of translatable mRNAs. In either case, this RNA is non-infectious in absence of the necessary proteins.

Note that positive strand RNA has to be propertly capped in order to be translatable. This is also not always the case - some viruses generate or steal the cap from the cellular mRNAs once uncoated in the cell.

Another aspect worth mentioning is that positive-sense retroviruses usually code for their genes in multiple reading frames and make heavy use of splicing to produce their proteins. Thus, directly translating their positive sense RNA is not sufficient for infection.

From Chapter 3 in Viral ecology:

For plus-strand RNA viruses except retroviruses, translation of the viral RNA 
follows immediately after uncoating. The viral RNA extruded from the capsid is then 
used by the host translation machinery for directing protein synthesis (Fig. 9). 
For all other viruses, whether of DNA or RNA genome, the step immediately following 
uncoating is either transcription of the genome yielding functional mRNAs or 
reverse transcription of vRNA yielding proviral DNA (retroviruses).


  1. Viral Ecology
  2. Principles of Virology

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.