Molecular genotypes of Mycobacterium tuberculosis strains circulating in Dakar , Senegal

1 Mycobacteriology Unit, Bacteriology Virology Laboratory at CHU Aristide Le Dantec, Dakar, Senegal. 2 Faculty of Health Sciences, Gaston Berger University, Saint Louis Senegal. 3 Medical Research Council Unit, Fajara, The Gambia. 4 Pulmonology Clinic at CHU Fann Hospital, Dakar Senegal. 5 Faculty of Medicine and Pharmacy, Department of Biological Sciences at Cheikh Anta Diop University Dakar, Senegal. 6 Institut for Tropical Medicine, Antwerp, Belgium.


INTRODUCTION
Tuberculosis (TB) remains a major cause of illness and death worldwide, especially in Africa and Asia.In 2014, the National Tuberculosis Program (NTP) reported 13.647 new TB cases in Senegal, of which 7% were co-infected with HIV.With a population of about 14 million inhabitants, TB incidence is estimated at 138 cases per 100,000 persons per year (all forms), and prevalence at 205 cases per 100,000 persons (WHO, 2015).Molecular genotyping tools for Mycobacterium tuberculosis have become valuable for TB diagnostics and investigations of disease transmission dynamics, outbreaks and phylogeny.They allow the distinction between "modern" MTBC lineages versus "ancient" lineages and their respective geographical distribution worldwide.Although these DNA typing techniques in conjunction with classical epidemiology approaches greatly enhance the understanding of TB transmission dynamics and epidemiology, no such study has been recently performed in Senegal.The only report that characterized circulating strains in the country was conducted by Niang et al. (1999).The authors presented preliminary work on 69 isolates using spoligotyping as a molecular tool to have an overview of strains, circulating in Dakar.As the sample size was relatively small, we are presenting a more comprehensive analysis on the mycobacterial population structure in Dakar between 2004 to 2006 using, molecular genotyping.

Patients
The study population consisted of consecutive consenting new TB patients older than 18 years, who were part of a cohort of the African Tuberculosis Vaccine (AFTBVAC) study between, 2004 and 2006.Patients were recruited based on TB index case and their potential contact within a family.All subjects were diagnosed with clinically and bacteriologically confirmed tuberculosis.After written informed consent, sputa were collected from patients by clinicians from the Pulmonology clinic at Fann Hospital, Dakar, Senegal.All enrolled patients were resident of Dakar or suburbs surrounding the capital.Demographic and clinical patient data including age, sex, HIV status and severity of disease on chest radiography were recorded.This study was approved by the Ethics Committee of the Ministry of Health, Senegal.

Laboratory analysis
Sputum examination and TB culture were done at the Bacteriology laboratory of Aristide Le Dantec University hospital in Dakar, Senegal.After staining with the Ziehl Neelsen method, sputa were decontaminated with N-acetyl cysteine-NaOH.Two hundred microliters of decontaminated sputa were inoculated on each of two glycerol Lowenstein Jensen (LJ) slopes.Cultures were incubated at 37°C for up to 8 weeks.Positive cultures were confirmed by Ziehl Neelsen smear microscopy and conventional biochemical tests.We used Niacin test, urease, niacin accumulation, p-nitrobenzoic acid (PNB) and p-nitro-alpha-acetylamino-beta-hydroxypropio phenome (NAP) for discrimination of the MTB complex from Mycobacteria other than tuberculosis (MOTT) (Rastogi et al., 1989).Confirmed MTBc were tested for their sensitivity to streptomycin (STR), Isoniazid (INH), Rifampicin (RIF) and Ethambutol (EMB) and isolates were stored in glycerol at -80°C (Canetti et al., 1963).

DNA isolation
Genomic DNA was extracted from TB positive cultures as described elsewhere (van Embden et al., 1993).After spectrophotometry, 10 ng of DNA were used for spoligotyping analysis.

Spoligotyping analysis
Spoligotyping was performed on genomic DNA using commercially available activated Biodine C membranes with the 43 synthetic oligonucleotides covalently bound to the membranes (Ocimum Biosolutions, Huda Techno Enclave Madhapur, India), according to standardized methods as described by Kamerbeek et al. (1997).

Data analysis
The spoligotyping hybridization patterns were converted into binary and octal formats and compared with, previously reported strains in the Spol DB4 database (Brudey et al., 2006).Binary spoligotypes were entered in SITVIT2 database.In 2013, SITVIT2 contained genotyping information of about 75.000 M. tuberculosis clinical isolates from different countries.In this database, SIT (Shared International Type) designates spoligotypes shared by two or more M. tuberculosis isolates, as opposed to "orphan" which, designates patterns reported for a single isolate.

Detection of Rifampicin and isoniazid resistance
Resistance of isolates to rifampicin and isoniazid was determined by the Genotype MTBDRplus Line Probe Assay (LPA) (Hain Lifescience GmbH, Nehren, Germany) according to the manufacturer"s instructions (Godreuil et al., 2007).

Single colony plating
Of two suspected mixed strain obtained after spoligotyping were recultured into 7H9 broth media, further plated on 7H11 solid medium that incorporates an enzyme that, digests casein to enhance growth of fastidious M.tb, species in single discreet colonies.Singles colonies of interest based on the cell morphology on 7H11 were marked with a felt pen and given a new lab ID.
Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution License 4.0 International License (2.2%) tested HIV positive.The age of patients ranged from 18 to 74 years.

MTBDR plus results
Of the 203 strains tested by the MTBDR plus assay, only two strains (1.0%) showed resistance to isoniazid, and none to rifampicin.Out of the two INH resistant strains, one was a Beijing isolate with an inhA mutation, and the other one was, an EAI isolate with a mutation in katG.

Single colony culture
Of two "Manu" pattern isolates that were plated to single colonies, for one isolate (D0242) the "daughter" colonies revealed separate patterns of Lineage 2 (Beijing) and Lineage 4 (Euro-American), while the other "daughter" isolates remained identical to the "mother" isolate (Table 2).

DISCUSSION
An interesting finding of the present study is the relatively high proportion (12%) of non-endemic Beijing strains.Such a finding is of general public health interest in Dakar, as Beijing strains are associated with drug resistance, rapid spread and great virulence when compared to other M. tuberculosis strains (Gomgnimbou et al., 2012).Moreover, it was recently suggested that some Manu 2 spoligotypes (which we frequently found in our study), could be the consequence of a "mixed" pattern due to concomitant Beijing and Euro-American strains (the latter comprising H, LAM, X, and T lineages) (Viegas et al., 2010).Therefore, considering the amount of Manu patterns and the possibility of mixed infections in our study, the above mentioned prevalence of Beijing strains could be an underestimation and potentially even be higher than the estimated 12%.Even at 12% prevalence, Dakar already has the highest proportion of TB due to the Beijing family in West Africa.The only other region in West Africa with comparable (though lower numbers) of Beijing strains (10%) is Cotonou, Benin.This phenomenon was identified during the first molecular epidemiological survey in Benin (Affolabi et al., 2009) and later confirmed to be an outbreak of streptomycin resistant Beijing strains in the city (Affolabi et al., 2009).Therefore the high proportion of Beijing strains in 2004 to 2006 is concern in the monitoring of ongoing transmission amongst the current mycobacterial population and their resistance patterns, is important.So far our study suggests that antibiotic resistance was not a major concern in Dakar, as at the time of the study, only two isolates showed INH mono-resistance.Our molecular resistance testing suggests that drug resistance in new TB cases was low in the period of 2004 to 2006, with only 1% isoniazid resistance and no rifampicin resistance.An ongoing nationwide drug resistance survey, will allow us to compare these numbers with current drug resistance levels based on an unbiased sampling of, the entire population.
Our spoligotype results revealed a great diversity of the M. tuberculosis strains circulating in Dakar where we not only found the most common global spoligotypes (Brudey et al., 2006) but also the M. africanum West Africa 2 lineage, exclusively present in the region.Specifically, these include the Haarlem family followed by T, LAM, Beijing, and MAF2 families.Interestingly, we did not identify any M. bovis isolates, which is in line with a previous observation by Diop et al. from 1970(Diop et al., 1976).The described composition of the MTBc overall confirms the findings from a study, analyzing 69 Senegalese M. tuberculosis isolates collected between September 1994 and September 1995, as comparable to our study and the major identified families were the same (Niang et al., 1999).
A major difference between the present and previous studies is in regards to the prevalence of MAF2 in Dakar.A total of 20% of all pulmonary tuberculosis cases in 1970 were still infected by MAF2 (Diop et al., 1976), indicating a slow decrease of the MAF prevalence in Dakar.The present study showed a 10% prevalence of MAF2 in tuberculosis patients between 2004 to 2006, which is similar to 10% in the study conducted in 1995.Countries such as Guinea Bissau, Benin and Nigeria observe high MAF prevalence up to 40% (Kallenius et al., 1999;Affolabi et al., 2009), while a decrease of M. africanum prevalence has been observed in other West and Central African countries like Cameroon and Burkina Faso (Huet et al., 1971;Ledru et al., 1996;Godreuil et al., 2007;Affolabi et al., 2009;Gehre et al., 2013).However, latest studies described 20% MAF2 prevalence in Burkina and 9% of MAF1 in Cameroon (Niobe-Eyangoh et al., 2003;Gomgnimbou et al., 2012).This reduction could be due to the recent expansion of the so-called M. tuberculosis Cameroon family in the country (Niobe-Eyangoh et al., 2003).Interestingly, the Cameroon family was already found in a proportion of 6% in our study indicating that, this family is successfully spreading across West African countries including Senegal.The reason for the relative decline of M. africanum in West Africa could be due to a lower transmission capacity or lower virulence.Although the virulence capacity of MAF2 strains is not well known, it was already demonstrated that these bacteria are overall attenuated and less immunogenic than M. tuberculosis sensu stricto isolates (de Jong et al., 2006;Bold et al., 2012).Another reason for the decline of M. africanum in West Africa could be that, certain M. tuberculosis lineages possess advantages in their ability to generate secondary cases within a community, and therefore outgrow M. africanum in the community (Gehre et al., 2013).The main limitation of our study is that, the geographical sampling area was limited to the Greater Dakar area and therefore is not a representative of the entire country.It is recommended that the sampling should be extended to remote areas of Senegal that would allow the establishment of the population structure of M. tuberculosis in the whole country and follow the prevalence of Beijing, Lam 10 Cam and MAF2 one decade after isolation of the strains described here.Moreover, the used spoligotyping technique provides a broad overview of genotype variability and circulate M. tuberculosis strains in a certain area, yet has a fairly low discriminatory power (Brudey et al., 2006) and needs to be associated with other high resolution techniques such as, the MIRU-VNTR typing or sequenced-based SNP analysis if, transmission dynamics of a potential Beijing outbreak are the focus of a future study.

Conclusion
The genetic variability of the M. tuberculosis complex in Dakar reflects a predominance of Euro-American M. tuberculosis (Haarlem, T and LAM) lineage with lower prevalence of M. africanum West African 2 than previously described.Further investigations focusing on a larger collection of strains covering Senegal nationwide are necessary to describe the population structure of these M. tuberculosis strains more accurately and monitor the development of the increased spread of Beijing strains in the country.
A to J distinct colonies that differ morphologically by size and pigmentation.a: The black and white boxes indicate the presence and absence, respectively, of the specific spacer at positions 1-43 in the DR locusb: Major lineage according to SpolDB4 database.

Table 1 .
Shows the different Spoligotypes pattern identified by a SIT in the SITVIT database.

Table 1 .
Contd.The black and white boxes indicate the presence and absence, respectively, of the specific spacer at positions 1-43 in the DR locus b: Major lineage according to SpolDB4 database c : Family designation according to SpolDB4 database. a:

Table 2 .
Spacer analysis of mixed infection test from single colonies.