By all accounts, there has been a lot of PRRS activity in U.S. swine farms this fall and early winter. Our molecular genetics laboratory has been busy with frequent requests for sequencing PRRS (porcine reproductive and respiratory syndrome) virus from diagnostic cases we have received. There have also been frequent requests to quantify the amount of virus present in diagnostic samples, which has led to some interesting observations that will be discussed later.
PRRS virus sequences are compared against other PRRS strains of interest, based on a history of a particular virus in the farm or region, or for other reasons to make a comparison. In order to put some sort of name on a virus, laboratories provide an RFLP cut pattern number based on the sequence of a portion of the virus genome. RFLP stands for “restriction fragment length polymorphism,” which is a method developed in 1996 to distinguish field viruses from vaccine viruses. The procedure involves using three enzymes to cut the virus gene into segments and then comparing viruses based on the size of the fragments produced, depending on where the virus gene was snipped apart by the enzymes.
Three numbers are generated based on the fragments produced by the three enzymes. Each time a new cut pattern is identified, a new number is assigned for that enzyme. Currently, four patterns have been identified for the first enzyme, 55 for the second, and eight for the third. Thus, it’s theoretically possible to have a PRRS virus with the RFLP pattern ranging anywhere from 1-1-1 to 4-55-8. Table 1 lists the cut sites and assigned numbers for all of the RFLP designations. We generate predicted RFLP patterns based on the sequencing information, without actually having to perform the RFLP procedure.
RFLP Trends – 4’s and 8’s
In Figure 1, a trendline of the distribution of major RFLP cut patterns for viruses sequenced at the University of Minnesota Veterinary Diagnostic Laboratory from 2001 to 2010 is shown. The emergence of the 1-8-4 strain in 2002 is evident, along with the 1-18-2 in 2008, and the 1-26-2 in 2009. In general, the figure illustrates the ability of the strains that have a “4” or an “8” as the middle RFLP number to dominate in swine herds in this region over time.
The RFLP patterns do not provide information on the degree of relatedness among the different RFLP patterns. The relatedness is determined using dendrograms, tables that compare an individual sequence to a group of other PRRS virus sequences. This approach (not shown) provides evidence that the 1-26-2 isolates appear to be closely related to, and are likely descended from, earlier 1-18-2 strains. In Figure 2, the frequency of a PRRS sequence falling in the 1-26-2 RFLP pattern is shown.
Why are certain PRRS strains more competitive?
We know that some PRRS strains are more virulent than others, causing more severe or different types of clinical signs than other strains. Thus far, we have not been able to figure out the genetic basis for this. Based on the predominance of certain PRRS virus strains over time, it appears that some are also more successful at competing for pigs within a farm or a region. This has been demonstrated anecdotally in farms where multiple strains are present at a given point in time, with some fading out over time, while others dominate and persist within the population.
A recent observation among isolates in the 1-26-2 group is that the concentration of virus in serum samples tends to be higher than other strains based on quantitative PCR (polymerase chain reaction) results. Where typical RNA copies are in the range of 105–108/ml of serum, recent results have consistently ranged from 1011–1012 copies/ml for viruses with a 1-26-2 cut pattern – a thousand to a million-fold higher. If pigs infected with these PRRS strains shed that much more virus, it stands to reason that this viral strain may be more effective in being transmitted to other pigs or herds. This observation of possible strain-based difference in virus production needs to be validated experimentally.
As an RNA virus, PRRS has continued to mutate over time. It was pointed out earlier that the RFLP cut pattern is not reliably predictive of relatedness among different PRRS viruses. The relative variations among three major RFLP cut pattern groups sequenced over the past two years are shown in Figure 3 and Table 2. The basic point is that PRRS viruses continue to change at a rapid pace.
Click to view graphs.
Jerry Torrison, DVM
University of Minnesota Veterinary Diagnostic Laboratory