Melarsoprol sensitivity profile of Trypanosoma brucei gambiense isolates from cured and relapsed sleeping sickness patients from the Democratic Republic of the Congo

Abstract

Background: Sleeping sickness caused by Trypanosoma brucei (T.b.) gambiense constitutes a serious health problem in sub-Sahara Africa. In some foci, alarmingly high relapse rates were observed in patients treated with melarsoprol, which used to be the first line treatment for patients in the neurological disease stage. Particularly problematic was the situation in Mbuji-Mayi, East Kasai Province in the Democratic Republic of the Congo with a 57% relapse rate compared to a 5% relapse rate in Masi-Manimba, Bandundu Province. The present study aimed at investigating the mechanisms underlying the high relapse rate in Mbuji-Mayi using an extended collection of recently isolated T.b. gambiense strains from Mbuji-Mayi and from Masi-Manimba.

Methodology/principal findings: Forty five T.b. gambiense strains were used. Forty one were isolated from patients that were cured or relapsed after melarsoprol treatment in Mbuji-Mayi. In vivo drug sensitivity tests provide evidence of reduced melarsoprol sensitivity in these strains. This reduced melarsoprol sensitivity was not attributable to mutations in TbAT1. However, in all these strains, irrespective of the patient treatment outcome, the two aquaglyceroporin (AQP) 2 and 3 genes are replaced by chimeric AQP2/3 genes that may be associated with resistance to pentamidine and melarsoprol. The 4 T.b. gambiense strains isolated in Masi-Manimba contain both wild-type AQP2 and a different chimeric AQP2/3. These findings suggest that the reduced in vivo melarsoprol sensitivity of the Mbuji-Mayi strains and the high relapse rates in that sleeping sickness focus are caused by mutations in the AQP2/AQP3 locus and not by mutations in TbAT1.

Conclusions/significance: We conclude that mutations in the TbAQP2/3 locus of the local T.b. gambiense strains may explain the high melarsoprol relapse rates in the Mbuji-Mayi focus but other factors must also be involved in the treatment outcome of individual patients.

Figure 1. Restriction digest profile generated with SfaNI PCR-RFLP on DNA of the T.b. gambiense strains as listed in Table 1 and of the two T.b. brucei control strains.

Lanes 1, 26, 27 and 52 = GeneRuler 100 bp Plus DNA Ladder (Fermentas), lanes 2 to 44 = T.b. gambiense strains isolated from Mbuji-Mayi, lanes 45–48 = T.b. gambiense strains isolated from Masi-Manimba, lane 49 = T.b. brucei 427 WT, lane 50 = T.b. brucei 427 AT1/P2 KO, lane 51 = negative PCR control.

Figure 2. Amplicons generated with TbAT1-PCR on DNA of the T.b. gambiense strains as listed in Table 1 and of the two T.b. brucei control strains.

Lanes 1, 26, 27 and 52 = GeneRuler 100 bp Plus DNA Ladder (Fermentas), lanes 2 to 44 = T.b. gambiense strains isolated from Mbuji-Mayi, lanes 45 to 48 = T.b. gambiense strains isolated from Masi-Manimba, lane 49 = T.b. brucei 427 WT, lane 50 = T.b. brucei 427 AT1/P2 KO, lane 51 = negative PCR control.

Figure 3. Amplicons generated with AQP2/3-PCR on DNA of the T.b. gambiense strains as listed in Table 1 and of the two T.b. brucei control strains.

Lanes 1, 26, 27 and 52 = GeneRuler 100 bp Plus DNA Ladder (Fermentas), lanes 2 to 44 = T.b. gambiense strains isolated from Mbuji-Mayi, lanes 45–48 = T.b. gambiense strains isolated from Masi-Manimba, lane 49 = T.b. brucei 427 WT, lane 50 = T.b. brucei 427 AT1/P2 KO, lane 51 = negative PCR control.

Figure 4. Schematic view of the AQP2/3 variants identified in this study (adapted from Graf et al [20]).

Sequence of AQP3 was not verified. Positions of primer: black box = AQP2/3_F, green box = AQP2_R, red box = AQP2/3_R. A) Reference locus of AQP2 and AQP3, with wild-type AQP2 found in the melarsoprol and pentamidine sensitive strain T.b. gambiense LiTat 1.3 and in all strains from Masi-Manimba. B) Chimera of AQP2 and AQP3 occurring in a melarsoprol and pentamidine resistant T.b. brucei strain as described by Baker et al. . C) Chimera of AQP2 and AQP3 plus loss of AQP3 in all T.b. gambiense strains from Mbuji-Mayi as described in this article and by Graf et al. . D) New chimera of AQP2 and AQP3, possibly outside the known locus, found in all T.b. gambiense strains from Mbuji-Mayi. E) New chimera of AQP2 and AQP3 plus loss of AQP3 found in two old Congolese T.b. gambiense strains, MBA and KEMLO. F) New chimera of AQP2 and AQP3, without loss of AQP3, found in all four T.b. gambiense strains isolated in Masi-Manimba.

Figure 5. Restriction digest profile generated with AvaI PCR-RFLP on DNA of the T.b. gambiense strains as listed in Table 1, including the four strains isolated from relapsed mice.

Lanes 1, 26, 27 and 52 = GeneRuler 100 bp Plus DNA Ladder (Fermentas), lanes 2 to 44 = T.b. gambiense strains isolated from Mbuji-Mayi, lane 45 = T.b. gambiense 15BT relapse 10 mg/kg BW, lane 46 = T.b. gambiense 163AT relapse 10 mg/kg BW, lane 47 = T.b. gambiense 346AT relapse 10 mg/kg BW, lane 48 = T.b. gambiense 346AT relapse 12 mg/kg BW, lane 49 = T.b. gambiense MBA, lane 50 = T.b. gambiense MM01, lane 51 = negative PCR control.

Figure 6. Restriction digest profile generated with SduI PCR-RFLP on DNA of the T.b. gambiense strains as listed in Table 1, including the four strains isolated from relapsed mice.

Lanes 1, 26, 27 and 52 = GeneRuler 100 bp Plus DNA Ladder (Fermentas), lanes 2 to 44 = T.b. gambiense strains isolated from Mbuji-Mayi, lane 45 = T.b. gambiense 15BT relapse 10 mg/kg BW, lane 46 = T.b. gambiense 163AT relapse 10 mg/kg BW, lane 47 = T.b. gambiense 346AT relapse 10 mg/kg BW, lane 48 = T.b. gambiense 346AT relapse 12 mg/kg BW, lane 49 = T.b. gambiense MBA, lane 50 = T.b. gambiense MM01, lane 51 = negative PCR control.