Using the UTR to determine if a potyvirus is a new species is not an accepted method. According to the International Committee on Taxonomy of Viruses, a new species in the Potyviridae must meet the following criteria :

1) Genome sequence relatedness: different species have CP aa sequence identity less than about 80%; and nt sequence identity less than 76% either in the CP or over the whole genome. There are also differences in polyprotein cleavage sites.

2) Host range and key host reactions; lack of cross protection.

3) Different inclusion body morphology.

4) Antigenic properties: serological relatedness may help in distinguishing species.

(http://www.sciencedirect.com/science/book/9780123846846; http://www.ictvonline.org/).

In order to determine how related two potyviruses are, you need either the amino acid or nucleotide sequence and then BLAST the sequences (http://blast.ncbi.nlm.nih.gov/Blast.cgi). If the protein sequence shares less than 80% homology then it is a new species (for the coat protein) or 76% (nucleotide for the coat protein or entire genome). If higher than 80%, it is a new isolate of a known potyvirus. I published a paper on Watermelon mosaic and Papaya ringspot viruses in Black locust where I use the coat protein sequence (amino acid and nuvleotide) of both viruses to determine how related they are to other isolates of their respective virus:(https://www.researchgate.net/publication/250309615_High_incidence_of_seed_transmission_of_Papaya_ringspot_virus_and_Watermelon_mosaic_virus_two_viruses_newly_identified_in_Robinia_pseudoacacia)

Article High incidence of seed transmission of Papaya ringspot virus...

In Bean there are two viruses BCMV and BCMNV (Poty virus group) and were considered a single virus but serology has shown that these are two different serotypes and now so many reports are there which have used UTR region to identify these 2 strains

Thank you!

Using the UTR to determine if a potyvirus is a new species is not an accepted method. According to the International Committee on Taxonomy of Viruses, a new species in the Potyviridae must meet the following criteria :

1) Genome sequence relatedness: different species have CP aa sequence identity less than about 80%; and nt sequence identity less than 76% either in the CP or over the whole genome. There are also differences in polyprotein cleavage sites.

2) Host range and key host reactions; lack of cross protection.

3) Different inclusion body morphology.

4) Antigenic properties: serological relatedness may help in distinguishing species.

(http://www.sciencedirect.com/science/book/9780123846846; http://www.ictvonline.org/).

In order to determine how related two potyviruses are, you need either the amino acid or nucleotide sequence and then BLAST the sequences (http://blast.ncbi.nlm.nih.gov/Blast.cgi). If the protein sequence shares less than 80% homology then it is a new species (for the coat protein) or 76% (nucleotide for the coat protein or entire genome). If higher than 80%, it is a new isolate of a known potyvirus. I published a paper on Watermelon mosaic and Papaya ringspot viruses in Black locust where I use the coat protein sequence (amino acid and nuvleotide) of both viruses to determine how related they are to other isolates of their respective virus:(https://www.researchgate.net/publication/250309615_High_incidence_of_seed_transmission_of_Papaya_ringspot_virus_and_Watermelon_mosaic_virus_two_viruses_newly_identified_in_Robinia_pseudoacacia)

Article High incidence of seed transmission of Papaya ringspot virus...

Dear Mr. Laney,

I really appreciate your response! Thanks so much for sharing your research article.

I would like to mention here that I have actually sequenced the complete genome of a potyvirus which has not been reported from USA so far. Also, I have sequenced NIb and coat protein regions of the same virus isolated from different cultivars of an ornamental plant. We have collaboration with a grower here in Oklahoma state and we were told that those cultivars were apparently received from Europe and South America. I have annotated the complete genome as well as the partial sequences from other isolates using Biology Workbench (Nucleotide and Protein BLAST). I have done phylogenetic analysis and still working on it.

Here, I know the ICTV criteria for new species demarcation in Potyviridae family. But, my question here was a bit different. I wanted to know if there was any criteria (stringent rule) for strain differentiation based on sequence information? Now I know that symptomatology is one of the criteria for strain differentiation but I was curious to share the information from ResearchGate community and wanted to know if sequence information could help in defining the strain/isolate of a particular virus species?

Again, thanks a lot for your explanatory answer for species differentiation in potyvirus genera. But, do you have any idea if there is any information about how to determine an isolate and strain of a particular virus species in potyvirus genera just based on genome sequence, if possible? Or, serology and symptomatology is the only option?

Dear Ravendra;

You pose a very interesting question. It has been demonstrated that the CP of plant virus (particularly ssRNA (+) viruses) is of capital importance in determining various relevant biological processes. Moreover, several biochemical characteristics of the virion depend on the CP, like ELISA reaction, electrophoretic mobility, cation, binding...). Thus, aminoacidic conservation should be high between strains of a given species and lower between species.

A classical example is given by the classification of potyviruses. Due to the huge number of species into this family, for a long time taxonomy of this group was a confusing matter for specialists. In fact, it was believed that potyviruses formed a "continuum" of few genres, with diffuse borders in inferior taxons.

However, a work focused in the analysis of the CP of several potyviruses species allowed to clarify this point. A homology threshold was established that permitted to separate strains and species: 38-71% homology belong to different viruses but 95% homololgy corresponded to different strains of the same species.

I attach that publication, hope it will help you, best wishes,

Carlos

Article Amino Acid Sequence Homology of Coat Proteins as a Basis for...

Thanks, Carlos! I appreciate your response and this was really helpful. And, thank you for sharing the publication. DD Shukla and CW Ward are the people who laid the basic foundation for studying potyviridae family and especially potyviruses.

Thank you so much,

Ravendra

An isolate is basically the viral population of a given species within a single plant. As far as I know, there isn't a stringent definition for how divergent strains need to be genetically. Differences in the symptoms produced, serology differences, different host ranges or vector transmission efficiencies between different vectors can be (and have been) used to differentiate strains. For Soybean mosaic virus, there are differences in sequence between the strains with the HC-Pro and CP having the greatest variability (the strains were determined using a set of differential hosts and observing symptom phenotypes). I would say the more divergent the amino acid sequence, yet remaining within the species demarcation, the more likely it is that you have different strains since this would be reflected in the serology and potentially the other factors as well. I would suggest looking at both the CP and the HC-Pro for this (though entire genome analysis can be done too) if you have the data. I have found this site useful in determine the ployprotein cleavage sites in different potyviruses: http://www.dpvweb.net/potycleavage/species.html

Thanks Mr. Laney for the useful suggestions!

I appreciate your response.

Dear Ravendra

I came across many views about your question. Our group is working on this aspect for the last three years, sequenced the cp gene of number of BCMV isolates but failed to differentiate the strains or isolates on cp gene seq. basis, hoever, cp gene is very good candidate gene for differentiation of virus species. we have sequenced the whole genome of BCMV strains and compared it with other strains and did get some differences/ groupings. Since biological strain identification involves the pathogenicity factors so inclusion of those genes involved in virus pathogenicity may help us to differentiate the virus strains, which we are looking for very actively.

Thanks Dr. Sharma for your insight!

First, could you please elaborate little bit about the differences you got and used as a basis for strain differentiation among your BCMV isolates?

Second, what approach have you been using to screen the genes responsible for pathogenicity to monitor the expression of those pathogenicity genes to support your findings for different strains?

Interestingly, I just heard a talk on complementation test for strain differentiation today here at ASV meeting in Colorado State University, Fort Collins from Bill Dawson's group (University of Florida). They used three criteria for strain differentiation:

1. 10-20% difference in complete genome sequence

2. Host- range and severity of symptoms

3. Accumulation of virus titer.

Point 2 and 3 I am in complete accord with. But, I am still pretty much confused when Dawson lab says 10-20% difference in complete genome sequence because one of the pioneers in the Taxonomy of Plant Virology, CW Ward published (quite long back) that homology less than 95% would support the claim for strain differentiation among other factors, like symptomatology and serology. So, have the things changed since Ward published his conclusions on strain differentiation?

I think I'm gonna talk to Bill Dawson for his conclusion in detail.