Rickettsia parkeri and Candidatus Rickettsia andeanae in Gulf Coast Ticks, Mississippi, USA

To the Editor: Rickettsia parkeri, a spotted fever group Rickettsia (SFGR) bacterium, is transmitted by Amblyomma maculatum, the Gulf Coast tick (1). The prevalence of R. parkeri in Gulf Coast ticks has been reported as <42% in the United States, which is higher than reported rates of R. rickettsii (the cause of Rocky Mountain spotted fever) in Dermacentor species ticks. Misdiagnosis among SFGR infections is not uncommon, and R. parkeri rickettsiosis can cause symptoms similar to those for mild Rocky Mountain spotted fever (1). We evaluated infection rates of R. parkeri and Candidatus Rickettsia andeanae, a recently identified but incompletely characterized SFGR, in Gulf Coast ticks in Mississippi, USA. 
 
During May–September 2008–2010, we collected adult Gulf Coast ticks from vegetation at 10 sites in Mississippi. We extracted genomic DNA from the ticks using the illustra tissue and cells genomicPrep Mini Spin Kit (GE Healthcare Life Sciences, Piscataway, NJ, USA). We tested amplifiable tick DNA by PCR of the tick mitochondrial 16S rRNA gene (2). We tested for molecular evidence of any SFGR species by nested PCR of rompA (rickettsial outer membrane protein A gene) (1). Samples positive for SFGR were subsequently tested by using species-specific rompA PCR for R. parkeri (3) and Candidatus R. andeanae (4). All PCRs included 1) a positive control of DNA from cultured R. parkeri– (Tate’s Hell strain) or Candidatus R. andeanae–infected Gulf Coast ticks and 2) a negative control of water (nontemplate). PCR products were purified by using Montage PCR Centrifugal Filter Devices (Millipore, Bedford, MA, USA) and sequenced by using Eurofins MWG Operon (Huntsville, AL, USA). We generated consensus sequences using ClustalW2 (http://www.ebi.ac.uk/Tools/msa/clustalw2/) alignment and identified the sequences using GenBank BLAST searches (www.ebi.ac.uk/Tools/clustalw2/). 
 
Proportions of ticks infected with SFGR, by region and year, were compared separately by using Fisher exact test followed by pairwise comparisons with a Bonferroni adjustment (PROC FREQ, SAS for Windows, V9.2; SAS Institute, Cary, NC, USA). For all analyses, p<0.05 was considered significant. An index of co-infection was calculated by using the formula IC = ([O – E]/N) × 100, in which IC is index of co-infection, O is number of co-infections, E is expected occurrence of co-infection caused by chance alone, and N is total number of ticks infected by either or both Rickettsia species. A χ2 test was used to determine statistical significance (5). 
 
A total of 707 adult Gulf Coast ticks were collected during the 3 years (350 in 2008, 194 in 2009, and 163 in 2010). Tick mitochondrial 16S rRNA gene was detected in 698 (98.7%), of which 128 (18.3%) were positive for SFGR DNA, comprising 106 (15.2%) positive only for R. parkeri, 10 (1.4%) positive only for Candidatus R. andeanae, and 12 (1.7%) co-infected with R. parkeri and Candidatus R. andeanae (Table). Positive test results from 22 ticks singly or co-infected with Candidatus R. andeanae were confirmed by sequencing. 
 
 
 
Table 
 
PCR results for adult Rickettsia parkeri– and Candidatus Rickettsia andeanae–infected Gulf Coast ticks (Amblyomma maculatum) collected from 10 sites in Mississippi, USA, 2008–2010* 
 
 
 
Most (94.6%) ticks were from northern (n = 260) and southern (n = 409) Mississippi (Technical Appendix Figure). No significant difference in the number of R. parkeri–infected ticks between northern and southern Mississippi was observed (p = 0.13) (Table). However, significantly more ticks were singly infected with Candidatus R. andeanae in southern sites than in northern sites (p = 0.03). The infection rate for co-infected ticks in southern sites was higher than that in northern sites (p = 0.06). Among the 3 collection years for northern and southern sites, only the prevalence of R. parkeri in singly infected ticks differed significantly (p = 0.01) (data not shown); the infection rate was significantly greater during 2010 than during 2009 (p = 0.003, α/3 = 0.02). The overall index of co-infection with R. parkeri and Candidatus R. andeanae was 6.5, statistically higher than expected by chance alone (Table) (p<0.0001). 
 
The overall prevalence of infection with SFGR species in Gulf Coast ticks sampled was 18.3%; 15.2% of ticks were singly infected with R. parkeri, and 1.7% were infected with R. parkeri and Candidatus R. andeanae. As reported, the frequency of R. parkeri in Gulf Coast ticks is generally high, ranging from ≈10% to 40% (3,4,6–8). We found approximately 1 R. parkeri-infected Gulf Coast tick for every 6 ticks tested, suggesting that infected Gulf Coast ticks are commonly encountered in Mississippi. Because Gulf Coast ticks are among the most common human-biting ticks in Mississippi (9), awareness of R. parkeri rickettsiosis should be increased in this state. We identified Candidatus R. andeanae in ≈3% of Gulf Coast ticks in Mississippi; this frequency is similar to those reported in other studies of Gulf Coast ticks in the southern United States (4,6). Our finding of co-infected Gulf Coast ticks is at a frequency significantly higher than expected from chance alone. The biologic role of co-infections of Gulf Coast ticks with R. parkeri and Candidatus R. andeanae remains to be determined. 
 
Technical Appendix: 
Distribution of Gulf Coast ticks and cases of rickettsiosis. 
 
Click here to view.(117K, pdf)

. We tested for molecular evidence of any SFGR species by nested PCR of rompA (rickettsial outer membrane protein A gene) (1). Samples positive for SFGR were subsequently tested by using species-specifi c rompA PCR for R. parkeri (3) and Candidatus R. andeanae (4). All PCRs included 1) a positive control of DNA from cultured R. parkeri-(Tate's Hell strain) or Candidatus R. andeanae-infected Gulf Coast ticks and 2) a negative control of water (nontemplate). PCR products were purifi ed by using Montage PCR Centrifugal Filter Devices (Millipore, Bedford, MA, USA) and sequenced by using Eurofi ns MWG Operon (Huntsville, AL, USA). We generated consensus sequences using ClustalW2 (www.ebi.ac.uk/Tools/ msa/clustalw2/) alignment and identifi ed the sequences using GenBank BLAST searches (www.ebi. ac.uk/Tools/clustalw2/).
Proportions of ticks infected with SFGR, by region and year, were compared separately by using Fisher exact test followed by pairwise comparisons with a Bonferroni  (Table). Positive test results from 22 ticks singly or co-infected with Candidatus R. andeanae were confi rmed by sequencing.
Most (94.6%) ticks were from northern (n = 260) and southern (n = 409) Mississippi (online Technical Appendix, Figure, wwwnc.cdc.gov/ EID/article/18/10/12-0250-F1.htm). No signifi cant difference in the number of R. parkeri-infected ticks between northern and southern Mississippi was observed (p = 0.13) (Table). However, signifi cantly more ticks were singly infected with Candidatus R. andeanae in southern sites than in northern sites (p = 0.03). The infection rate for coinfected ticks in southern sites was higher than that in northern sites (p = 0.06). Among the 3 collection years for northern and southern sites, only the prevalence of R. parkeri in singly infected ticks differed signifi cantly (p = 0.01) (data not shown); the infection rate was signifi cantly greater during 2010 than during 2009 (p = 0.003, α/3 = 0.02). The overall index of co-infection with R. parkeri and Candidatus R. andeanae was 6.5, statistically higher than expected by chance alone (Table) (p<0.0001).
The overall prevalence of infection with SFGR species in Gulf Coast ticks sampled was 18.3%; 15.2% of ticks were singly infected with R. parkeri, and 1.7% were infected with R. parkeri and Candidatus R. andeanae. As reported, the frequency of R. parkeri in Gulf Coast ticks is generally high, ranging from ≈10% to 40% (3,4,6-8).
We found approximately 1 R. parkeriinfected Gulf Coast tick for every 6 ticks tested, suggesting that infected Gulf Coast ticks are commonly encountered in Mississippi. Because Gulf Coast ticks are among the most common human-biting ticks in Mississippi (9), awareness of R. parkeri rickettsiosis should be increased in this state. We identifi ed Candidatus R. andeanae in ≈3% of Gulf Coast ticks in Mississippi; this frequency is similar to those reported in other studies of Gulf Coast ticks in the southern United States (4,6). Our fi nding of co-infected Gulf Coast ticks is at a frequency signifi cantly higher than expected from chance alone. The biologic role of coinfections of Gulf Coast ticks with R. parkeri and Candidatus R. andeanae remains to be determined. Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 18, No. 10, October 2012  (1). Our data support the diffi culty of attributing cause of death for patients with CDI.
In 2004 in Quebec, Canada, a mandatory CDI surveillance program was implemented. Deaths that occurred within 30 days after CDI diagnosis were classifi ed as 1) directly attributable to CDI (e.g., toxic megacolon, septic shock), 2) having a CDI contribution (e.g., acute decompensation of chronic heart failure), or 3) unrelated to CDI (e.g., terminal cancer) (2  . Bars indicate the number of reviewers who assigned each category. Gray bars indicate that CDI was unrelated to death, white bars indicate that CDI contributed to death, and black bars indicate that death was directly attributable to CDI.
All material published in Emerging Infectious Diseases is in the public domain and may be used and reprinted without special permission; proper citation, however, is required.