Molecular study on Pasteurella multocida and Mannheimia granulomatis from Kenyan Camels (Camelus dromedarius)

Background Outbreaks of a Haemorrhagic Septicaemia (HS) like disease causing large mortalities in camels (Camelus dromedarius) in Asia and in Africa have been reported since 1890. Yet the aetiology of this condition remains elusive. This study is the first to apply state of the art molecular methods to shed light on the nasopharyngeal carrier state of Pasteurellaceae in camels. The study focused on HS causing Pasteurella multocida capsular types B and E. Other Pasteurellaceae, implicated in common respiratory infections of animals, were also investigated. Methods In 2007 and 2008, 388 nasopharyngeal swabs were collected at 12 locations in North Kenya from 246 clinically healthy camels in 81 herds that had been affected by HS-like disease. Swabs were used to cultivate bacteria on blood agar and to extract DNA for subsequent PCR analysis targeting P. multocida and Mannheimia-specific gene sequences. Results Forty-five samples were positive for P. multocida genes kmt and psl and for the P. multocida Haemorrhagic Septicaemia (HS) specific sequences KTSP61/KTT72 but lacked HS-associated capsular type B and E genes capB and capE. This indicates circulation of HS strains in camels that lack established capsular types. Sequence analysis of the partial 16S rRNA gene identified 17 nasal swab isolates as 99% identical with Mannheimia granulomatis, demonstrating a hitherto unrecognised active carrier state for M. granulomatis or a closely related Mannheimia sp. in camels. Conclusions The findings of this study provide evidence for the presence of acapsular P. multocida or of hitherto unknown capsular types of P. multocida in camels, closely related to P. multocida strains causing HS in bovines. Further isolations and molecular studies of camelid P. multocida from healthy carriers and from HS-like disease in camels are necessary to provide conclusive answers. This paper is the first report on the isolation of M. granulomatis or a closely related new Mannheimia species from camelids.


Background
Pasteurella multocida capsular types B and E are the specific cause of seasonal outbreaks of Haemorrhagic Septicaemia (HS) in tropical cattle and buffaloes [1][2][3]. The veterinary literature provides a long record of an HS-like disease causing significant mortality in camels [4][5][6][7][8]. Yet the aetiology of HS in camels remains elusive [1,3,9]. Previous reports on HS-like disease in camels often failed to isolate the pathogen or provided only limited information on phenotypical characteristics of the isolates [9]. Attempts to infect camels with highly virulent Pasteurella multocida capsular type B strains produced only very mild clinical symptoms that resolved completely within 3 days [9].
A respiratory disease in Ethiopian camels caused by Pasteurella (Mannheimia) haemolytica has been described by Bekele [10]. This pilot study is the first of its' kind to use state of the art molecular methods for investigating Pasteurellaceae in camels.

Methods
In 2007 and 2008, 388 nasopharyngeal swabs were collected at 12 locations in North Kenya from 246 clinically healthy camels in 81 herds that had reportedly been affected by outbreaks of HS-like disease. Flocculated swabs (FLOQSwabs®) were used for DNA extraction while swabs in Amies transport medium (Sterilin®) were used for standard bacteriological investigation. Bacteriological examination and extraction of DNA was carried out at a laboratory in Nairobi (Analabs Ltd.). DNA eluates from 341 swabs and 19 isolated Pasteurella-like cultures were transferred to the Institute for Microbiology and Epizootics at Freie Universität Berlin (IMT/ FUB) in Germany. At the IMT/FUB, DNA was extracted from 19 culture isolates, two of which did not belong to the Pasteurellaceae based on PCR results. Seventeen isolates were subsequently selected for biochemical testing to differentiate M. granulomatis from other Mannheimia spp. according to Ewers et al. (2004) [11]. Three hundred five samples with positive reaction in the 16S rRNA gene PCR, as well as, heat denatured DNA from bacterial cultures were investigated for the presence of P. multocidaand M. haemolyticaspecific DNA sequences. For P. multocida the specific sequences tested were kmt, psl, and KTSP61/KTT72. Samples that proofed positive in one of these PCRs were further tested for capsular genes capA, capB, capD, capE, and capF, and for virulenceassociated genes toxA, ptfA, pfhA oma87, ompH, hgbA, hgbB, exbB/tonB, tbpA, nanB, and nanH. To detect pathogenic Mannheimia spp., we screened for the leukotoxin (lktA) and the outer membrane protein (pomA) genes. All PCRs, except for the P. multocida capsular gene PCR, which was performed as a multiplex PCR, were conducted as single PCRs according to previously published protocols [12][13][14][15][16][17][18].

Results
The species P. multocida could not be isolated from 312 nasal swabs cultured on Blood Agar. Of the 305 DNA eluates containing sufficient quantities of DNA to undergo molecular characterization, 60 were positive for at least one of the P. multocida species-specific sequences tested by PCR. Forty-five samples gave a positive result for the HS-associated sequence KTSP61/ KTT72, known to be present in P. multocida capsular type B strains. Neither capB nor capE gene, encoding HS-associated capsular types B and E, were detected in the sample material. Blast search of 18 KTSP61/KTT72 sequences [19] matched with high scores (98.8-99%) to nucleotide sequences submitted under GenBank accession numbers AF016260 (P. multocida unknown protein 1 gene, partial cds and unknown protein 2 gene, complete cds), AY948545.1 (P. multocida HS-B specific genomic sequence) and AJ421513.1 (P. multocida DNA fragment specific for HS). CapA and capD genes encoding for P. multocida capsular types A and D were identified in five and in one sample, respectively. Screening for adhesionrelated genes ptfA and pfhA was positive in 15 and 2 samples, respectively. Outer membrane protein gene ompH was detected in 10, oma87 in 14 samples. While toxA, encoding the dermonecrotoxin, was not found in any sample, iron acquisition-related genes were present as follows: hgbA (n = 12), hgbB (n = 11), exbB/tonB (n = 1). Neuraminidase encoding genes nanB and nanH were present in 9 and 11 samples, respectively.

Discussion
This study investigated the active (nasopharyngeal) carrier state for Pasteurellaceae in Kenyan camels. Importance of the latent carrier state in the epidemiology of Pasteurellosis was reiterated by Dziva et al. (2008) [23].
The fact that no P. multocida capsular type B or E specific DNA sequences were identified in this study may indicate the presence of non-capsulated strains or the emergence of a hitherto unknown capsular type of P. multocida in camels. According to the OIE Terrestrial Manual [24] vaccines against Haemorrhagic Septicaemia in cattle and buffaloes must be based on local isolates that represent the prevalent serotype; seed cultures for the production of HS vaccines should contain capsulated organisms. Based on the results of this study the indiscriminate use of vaccines based on HS causing P. multocida isolates from cattle and buffaloes cannot be recommended for the prevention of HS-like disease in camels. -This is the first molecular study to confirm the presence of P. multocida capsular types A and D specific DNA sequences in the nasopharynx of healthy Kenyan carrier camels, albeit at low frequency. Both capsular types have been reported in camels previously [25] based on phenotypic characterisation. Failure to culture P. multocida in this study is possibly related to the fastidiousness of the species which does not withstand cold chain transport of several days [26].
The lktA or pomA sequences do not occur regularly in all Mannheimia species or strains [27,28], hence negative findings do not rule out presence of Mannheimia spp., but lktA-negative M. haemolytica strains are reported to be less virulent. In this study the most common Pasteurellaceae species cultured from the nasopharynx of healthy carrier camels in North Kenya and identified by 16S rRNA sequence analysis was a Mannheimia sp. with 98% to 99% sequence identity to M. granulomatis. Only more recently has M. granulomatis been recognised as a significant pathogen in domestic and wild ruminants [29][30][31]. Phenotypic characterization of Pasteurellaceae species is of limited value [32] and published phenotypic characteristics for M. granulomatis are based on a limited number of bovine, leprine and deer strains [2,21,22,30]. Hence it is to be expected that biochemical reactions of Kenyan camelid M. granulomatis strains differ from those described for M. granulomatis under http://www.bacterio.net/mannhei mia.html/. Comparison of our results with a previous communication on involvement of P. (M.) haemolytica in respiratory disease in Ethiopian camels [10] is also limited, because the methodology used there would not have permitted a differentiation between M. granulomatis, M.
(Pasteurella) haemolytica and other Mannheima spp. The possibility that this respiratory pathogen isolated from Ethiopian camels [10] may in fact have been M. granulomatis or a new Mannheimia species very closely related to M. granulomatis cannot entirely be ruled out.

Conclusions
This study has documented the carrier state for acapsular P. multocida or unknown capsular types of P. multocida, closely related to the P. multocida strains causing HS in cattle and buffaloes, in healthy camels. At the same time the study found no evidence for the presence of P. multocida capsular types B and E or their specific DNA sequences in healthy camels in North Kenya. Further isolations and molecular studies of camelid P. multocida from healthy carriers and from HS-like disease in camels are necessary to provide conclusive answers. To our best knowledge this is the first report on the isolation of M. granulomatis or a closely related new Mannheimia species from camelids.