Characterization of prevailing M. tuberculosis lineages and clones focusing on different geographical levels such as continents, countries, regions or cities is important for locating the origin, evolution and spreading dynamics of a particular M. tuberculosis clone, which is often difficult to be identified by traditional epidemiological investigations alone. Like most of sub-Saharan Africa, Uganda has a high prevalence of TB infection with peri-urban communities of Kampala recording higher rates than those in the rest of the country . In this region where TB is endemic, it is critical to identify predominant strain types in order to study transmission patterns within communities and to understand the epidemiology of the disease in the country as a whole. This report presents the largest amount of molecular epidemiological data on M. tuberculosis isolates from Uganda to date. It is also the first systematic community based study conducted to assess strain diversity, associated HIV sero-status and anti-tuberculosis drug resistance of M. tuberculosis complex in a peri-urban population of Uganda.
Our findings on the predominance of the T2 family of strains in Kampala compare well with previous data from the study at Mulago hospital, Kampala, in which 67% of the isolates were identified as lacking hybridization to either spacer 40 or both 40 and 43 [4, 21]. Else where in East Africa, a previous study in Kenya found only eight (11%) of 73 isolates to be of the T2 family and its variants, while in northern Tanzania four (3%) of 130 strains were T2-Uganda, frequencies much lower than 70% observed in our sample . It is therefore plausible that the TB epidemic in Kampala is local and well established, and that this strain is well adapted to transmit in the local population. Similar trends have been observed in other studies where local genotypes tend to form a greater proportion of the circulating strains. Cases in point are Guinea Bissau where 199 (51%) of 229 isolates belonged to the Guinea Bissau family ; Cameroon where 193 (46.7%) of 413 M. tuberculosis isolates belonged to the Cameroon family, LAM10-CAM ; Harare, Zimbabwe where 68/214 (31.8%) isolates in one study and 116/246 (47.2%) in another were LAM-ZWE variants [25, 26] and in Zambia where 74/114 (65%) isolates were also of the LAM-ZWE family . These are in further agreement with recent findings where it was noted that different strains of M. tuberculosis have adapted to specific human populations, and that such local strains are more likely to transmit compared to others [27, 28].
Other significant spoligotypes in our study were CAS1-Kili (3.5%), LAM9 (2.6%), CAS1-Delhi (2.6%), LAM3/S (1.7%), CAS1 (1.7%), and LAM11-ZWE (1.5%). In comparison to other studies in the region, the CAS, LAM and EAI families were reported at 37%, 22% and 17% respectively of a total of 147 isolates in a study in Dar es Salaam, Tanzania ; while in northern Tanzania, the most predominant families were CAS-Kili (30%), LAM11-ZWE (14.6%), EAI (6.2%), Beijing (5.4%), and CAS1-Delhi, T1 and LAM9 at 3.8% . In Kenya, on the other hand, 35.6% of 73 isolates were of the CAS family, while 11% were LAM . These studies show more success of the CAS, LAM and EAI families in the neighboring East African countries, while in Central Uganda, the T2 family of strains predominates.
A comparison of the prevalence of the Beijing family of strains, highly prevalent in many Asian locations , shows that in East Africa, Tanzania has reported 14 cases [2, 5], Kenya six , while in a recent study of mycobacteria causing human cervical lymphadenitis in pastoral communities in the Karamoja region of Uganda, three isolates with the Beijing spoligotype were identified from 34 biopsies , but the susceptibility pattern and associated HIV sero-status of the individuals is not known. In this study, to the best of our knowledge, we report the first four M. tuberculosis Beijing strains in Uganda with known anti-tuberculosis drug susceptibility pattern to date, with none of the four strains being MDR. Our results further show that the Beijing family at 1.2% is not common. This result differs from that seen in a study in Kenya in which two of the six Beijing strains in the collection were MDR . In Malawi, 44/1,029 (4.3%) of strains in one study were Beijing and susceptible ; in the Dar es Salaam study, the susceptibility pattern of the seven Beijing isolates was not determined , while more recently in northern Tanzania, 5.4% of 130 isolates were Beijing and all susceptible. Generally, the trend and susceptibility pattern of the Beijing genotype in sub-Saharan Africa has been described to be one of three: epidemic and associated with drug resistance (high level in South Africa); epidemic but drug sensitive (Malawi); and very low level or absent (parts of Africa) . Our data suggest that the Beijing strains in Kampala are of the low level and susceptibility pattern.
The emergence of drug resistance in the treatment of TB has complicated its management. In the developing countries, lack of resources hinder regular drug resistance surveys, hence the magnitude of this problem remains largely unknown. The last national anti-tuberculosis drug resistance survey in Uganda (1996–1997) on 586 patients  indicated that primary resistance to isoniazid was 6.7%, that to rifampicin 0.8%, while MDR was 0.5%. In neighboring Rwanda, the most recent survey results  show that of 616 strains from new cases, 6.2% were resistant to isoniazid, 3.9% to rifampicin and 3.9% were multidrug-resistant TB. In northern Tanzania, a study of 111 isolates showed that 9.9% were resistant to isoniazid, 2.7% to rifampicin, with 2.7% being MDR . Our result for resistance to isoniazid at 8.1% is comparable to that in northern Tanzania, while that to rifampicin (4.4%) is comparable to the Rwanda findings.
Regarding cluster analysis, although the T2 family of strains accounted for 13/15 MDR strains, there was no statistical evidence (p = 0.25) to suggest that it might be the driving force of anti-tuberculosis drug resistance in this community. Additionally, as observed in northern Tanzania , our results show that HIV sero-status could not be identified with a particular spoligopattern. Furthermore, import of strains from Asia (CAS1-Kili, 3.5%; CAS1-Delhi, 2.6%; CAS1, 1.7%; and CAS2, 0.6%) has not had a major impact on the M. tuberculosis population in Kampala. However, together with the single M. bovis seen in our sample, there is clearly a diverse set of Mycobacterium tuberculosis complex strains circulating in Kampala, although T2 strains are predominant.