To the Editor: Plague, caused by Yersinia pestis, is now largely controlled by improved sanitation and the use of antimicrobial drugs. However, before the widespread availability of antimicrobial drugs, an estimated >200 million persons died during pandemics (1). Today, if Y. pestis were to acquire antimicrobial drug resistance determinants, plague could again be a deadly disease.

Antimicrobial drug resistance in Y. pestis has been documented for only a few strains. The best available information is for 2 strains isolated in Madagascar in 1995 (2), in which resistance was conferred by plasmids not typically found in Y. pestis. Strain 16/95 was resistant to streptomycin only; this resistance was mediated by plasmid pIP1203 (3). Strain 17/95 was resistant to 8 antimicrobial drugs, including some commonly used to treat plague, such as streptomycin, tetracyclines, and sulfonamides (2). Multidrug resistance in 17/95 was mediated by plasmid pIP1202 (4). Both plasmids could be transferred by conjugation from the source Y. pestis strains to other Y. pestis strains and Escherichia coli (3,4). pIP1203 could be transferred from E. coli to Y. pestis in the midgut of co-infected rat fleas (Xenopsylla cheopis), common vectors of plague (5).

Comparative sequence analysis has indicated that pIP1202 shares an almost identical IncA/C backbone with multidrug-resistant (MDR) plasmids from Salmonella enterica serotype Newport SL254 and Yersinia ruckeri YR71, suggesting recent acquisition from a common ancestor (6). In this study, this backbone was detected in numerous MDR enterobacterial pathogens (e.g., E. coli, Klebsiella spp., and multiple Salmonella serotypes) isolated from retail meat products. Many of these plasmids transferred at high rates to a plasmid-free Y. ruckeri strain, indicating the ability to efficiently transfer among species. Meat products examined in that study originated from 9 US states, including western plague-endemic states such as California, Colorado, New Mexico, and Oregon.

To determine whether the IncA/C plasmid backbone previously found in MDR Y. pestis and other species exists in Y. pestis isolates from western U.S. states, we screened Y. pestis DNA. The 713 isolates were collected from from humans, small mammals, and fleas in 14 of the 17 western plague-endemic states (Table), including all states that reported human cases during 1970–2002 (7). We used Primer Express software (Applied Biosystems, Foster City, CA, USA) to design a TaqMan-MGB single-probe assay to detect repA, a plasmid replication gene present in the IncA/C plasmid backbone. We based this assay on the repA assay described by Carrattoli et al. (8) and used the same forward primer but a different reverse primer and an additional probe to facilitate screening on a real-time PCR platform.

Real-time PCRs were conducted in 10-μL reaction mixtures that contained 900 nmol/L of forward (5′-GAGAACCAAAGACAAAGACCTGGA-3′) and reverse (5′-TGGCCGGAGATTCAATGATC-3′) primers, 200 nmol/L of the repA-specific probe (5′-6FAM-AGACTCACCGCAAATG-3′), 1× AB TaqMan Universal PCR Master Mix with AmpErase UNG (uracil N-glycosylase) (Applied Biosytems), and 1 μL of template. Thermal cycling was performed on an Applied Biosystems 7900 HT sequence detection system under the following conditions: 50°C for 2 min, 95°C for 10 min, and 40 cycles of 95°C for 15 s and 60°C for 1 min. DNA extracts from Y. pestis strain 17/95 (4) and Salmonella enterica serotype Newport strain SL254 (6) were used as positive controls.

Of the 713 Y. pestis isolates screened, none was positive for the IncA/C plasmid backbone, indicating that MDR as mediated by pIP1202-like MDR plasmids described by Welch et al. (6) was not in these samples. This finding is encouraging with regard to public health. However, we screened only for the plasmid backbone; MDR genes may have been in some of these samples but not carried by pIP1202-like MDR plasmids, especially considering that plasmids can be readily integrated into the Y. pestis chromosome (1).

Could MDR Y. pestis arise in North America by acquisition of an MDR plasmid, such as pIP1202, from food-animal production activities in plague-endemic regions? If so, Salmonella spp. would be a likely MDR plasmid donor for several reasons. First, Y. pestis has several plasmids that are highly similar to those in Salmonella spp., indicating active transfer of plasmids between these 2 bacterial groups (6). Second, fleas that are common vectors of plague have been shown to be naturally co-infected with Salmonella spp. and Y. pestis and capable of transmitting both organisms to rodent hosts (9). Third, MDR plasmids are readily transferred to Y. pestis in the flea gut (5). Fourth, transferable MDR plasmids are common among Salmonella spp. isolates in US food animals (10). Given these linkages, the transfer of an MDR plasmid from Salmonella spp. to Y. pestis seems possible. However, we emphasize that to date no evidence supports this type of event.