Molecular Detection and Typing of Pathogenic Leptospira in Febrile Patients and Phylogenetic Comparison with Leptospira Detected among Animals in Tanzania

Abstract

Molecular data are required to improve our understanding of the epidemiology of leptospirosis in Africa and to identify sources of human infection. We applied molecular methods to identify the infecting Leptospira species and genotypes among patients hospitalized with fever in Tanzania and compared these with Leptospira genotypes detected among animals in Tanzania to infer potential sources of human infection. We performed lipL32 real-time PCR to detect the presence of pathogenic Leptospira in acute-phase plasma, serum, and urine samples obtained from study participants with serologically confirmed leptospirosis and participants who had died with febrile illness. Leptospira blood culture was also performed. In positive specimens, we performed species-specific PCR and compared participant Leptospira secY sequences with Leptospira reference sequences and sequences previously obtained from animals in Tanzania. We detected Leptospira DNA in four (3.6%) of 111 participant blood samples. We detected Leptospira borgpetersenii (one participant, 25.0%), Leptospira interrogans (one participant, 25.0%), and Leptospira kirschneri (one participant, 25.0%) (one [25%] undetermined). Phylogenetic comparison of secY sequence from the L. borgpetersenii and L. kirschneri genotypes detected from participants was closely related to but distinct from genotypes detected among local livestock species. Our results indicate that a diverse range of Leptospira species is causing human infection. Although our analysis suggests a close relationship between Leptospira genotypes found in people and livestock, continued efforts are needed to obtain more Leptospira genetic material from human leptospirosis cases to help prioritize Leptospira species and genotypes for control.

Figure 1.

Phylogenetic tree showing the relationship between Leptospira secY gene (434-bp fragment) derived from human infections in Tanzania with sequence from reference serovars and previously published sequences from human and animal infection in Tanzania^, The phylogenetic tree was constructed using the maximum likelihood method using the Tamura 3-parameter (T92) nucleotide substitution model with a discrete gamma distribution. The tree with the highest log likelihood is shown. Sequences from our study are labeled with unique identifiers (FEVERXXXX) and GenBank accession numbers and highlighted in red. Sequences generated from reference Leptospira serovars and from other studies in Tanzania are shown and labeled with GenBank accession numbers and ST types, respectively. Expanded clades show reference serovars most closely related to the study genotypes. Clades of more distantly related species are collapsed and labeled with species names only. Country locations and host are shown in parenthesis for East African studies. DRC = Democratic Republic of Congo; sv = serovar; ST = sequence types. This figure appears in color at www.ajtmh.org.