Genetic diversity and genotype multiplicity of Plasmodium falciparum infections in symptomatic individuals in the maritime region of Togo


Simplice D. Karou, Ecole Supérieure des Techniques Biologiques et Alimentaires, Université de Lomé (ETSBA-UL), BP 1515 Lomé, Togo. Tel.: +228 22 25 64 35; Fax: +228 22 21 85 95, E-mail:


Objective  To assess the genotype prevalence and the multiplicity of Plasmodium falciparum infections in the maritime region of Togo.

Methods  We enrolled 309 symptomatic individuals aged from 6 months to 15 years from Bè/Lomé and Tsévié, two malaria endemic zones. The number and the proportions of merozoite surface proteins 1, 2 and 3 genotypes in patients were determined using capillary electrophoresis genotyping. We further investigated the possible association between transaminases and homocysteine, and the severity of the disease.

Results  Of the 309 samples genotyped, 210 tested positive to msp-1, 227 to msp-2 and 193 to msp-3. The nested PCR revealed 22 different alleles for the allelic family msp-1, 33 for msp-2 and 13 for msp-3. At each locus, the family distribution was 54.58% of K1, 25% of MAD20 and 20.42% of RO33 for msp-1, and 51.71% and 48.29% of FC27 and 3D7, respectively, for msp-2. For all these allelic variants, the distribution was associated with neither the severity of malaria nor the zone of habitation. Pearson correlation coefficients between either the levels of homocysteine or the transaminase and the severity of the disease were very low.

Conclusion  The severity of malaria was not associated with higher multiplicity of infections and did not appear restricted to particular genotypes. More comprehensive explorations including immunity, genetic factors, nutritional and sociologic status of the population could clarify the situation.


Objectif:  Evaluer la prévalence génotypique et la multiplicité des infections àPlasmodium falciparum dans la région maritime du Togo.

Méthodes:  Nous avons recruté 309 enfants symptomatiques âgés de 6 mois à 15 ans à Bè/Lomé et à Tsévié, deux zones endémiques pour le paludisme. Le nombre et les proportions des génotypes des protéines de surface du mérozoïte (msp) -1, 2 et 3 chez les patients ont été déterminés en utilisant le génotypage par électrophorèse en capillaire. Nous avons également étudié l’association possible entre les transaminases et l’homocystéine avec la sévérité de la maladie.

Résultats:  Sur les 309 échantillons génotypés, 210 se sont révélés positifs pour msp-1, 227 positifs pour msp-2 et 193 positifs pour msp-3. La PCR nichée a révélé 22 allèles différents pour la famille allélique msp-1, 39 pour msp-2 et 13 pour msp-3. À chaque locus, la distribution de la famille était de 54,58% de K1, 25% de MAD20 et 20,42% de RO33 pour msp-1; 51,71% de FC27 et 48,29% de 3D7 pour msp-2. Pour toutes ces variantes alléliques, la distribution n’a été associée ni à la sévérité du paludisme ni à la zone d’habitation. Les coefficients de corrélation de Pearson entre soit les taux d’homocystéine ou de transaminase et la gravité de la maladie étaient très faibles.

Conclusion:  La sévérité du paludisme n’était pas été associée à des infections multiples et ne semblait pas limitée à des génotypes particuliers. Des investigations plus complètes, y compris de l’immunité, des facteurs génétiques, de l’état nutritionnel et sociologique de la population pourraient clarifier la situation.


Objetivo:  Evaluar la prevalencia de genotipos y la multiplicidad de infecciones por Plasmodium falciparum en la región marítima de Togo.

Métodos:  Se incluyeron 309 niños sintomáticos con edades comprendidas entre los 6 meses y los 15 años en Bè/Lomé y Tsévié, dos áreas endémicas para malaria. El número y las proporciones de los genotipos 1, 2 y 3 de la proteína de superficie del merozoito se determinaron mediante genotipado por electroforesis capilar. Se investigó también la posible asociación entre las transaminasas y la homocisteína y la severidad de la enfermedad.

Resultados:  De las 309 muestras genotipadas, 210 dieron positivas para msp-1, 227 para msp-2 y 193 para msp-3. La PCR anidada reveló 22 alelos diferentes para la familia alélica msp-1, 39 para msp-2 y 13 para msp-3. En cada locus, la distribución de la familia era un 54.58% de K1, 25% de MAD20 y 20.42% de RO33 para msp-1, y 51.71% y 48.29% de FC27 y 3D7, respectivamente, para msp-2. Debido a todos estos variantes alélicos la distribución no estaba ni asociada con la severidad de la malaria ni con la zona de habitación. Los coeficientes de correlación de Pearson entre los niveles de homocisteína o de la transaminasa y la severidad de la enfermedad eran muy bajos.

Conclusión:  La severidad de la malaria no estaba asociada con una mayor multiplicidad de infección y no parecía estar restringida a genotipos particulares. Se requieren estudios más integrales que incluyan inmunidad, factores genéticos, estatus nutricional y sociológico de la población para clarificar la situación.


Malaria caused by Plasmodium falciparum remains the main cause of death in African young children (Konate et al. 1999; Paul et al. 1999; Montoya et al. 2003). In these patients, the clinical state deteriorates quickly towards neurological malaria or severe anaemia. The virulence of P. falciparum is because of the ability of infected red blood cells to clump in the capillaries of vital organs, in addition to the adhesion of parasites to the host endothelial cells and blood cells, which leads to blockage of the microvascular circulation. Another virulence factor of this parasite is its ability to generate variability within genetic families. This diversity relies on the extensive allelic polymorphism, the antigenic variation and the sexual reproduction that ensures genetic mixing. The genetic diversity of P. falciparum is also responsible for the parasite’s resistance to antimalarial drugs. The parasite polymorphism markers most commonly used in molecular epidemiology are the merozoite surface proteins namely msp-1 and ms-2, and other antigens associated with the surface of the merozoite such as msp-3, GLURP, SERP and S-antigen (Takala et al. 2006).

The identification of genes coding for methylenetetrahydrofolate reductase (MTHFR) and methionine synthase (MS) in the Plasmodium has shown that the parasite has an enzymatic complex that permits metabolizing folate taken from infected cells. Thus the level of plasma homocysteine (tHcy), whose metabolism is related to the status of folate, is positively correlated with the severity of the disease (Chillemi et al. 2004; Abdel Gader et al. 2009). Hypertrasaminasemia cases have also been reported in severe malaria (Mabiala-Babela et al. 2002).

Few studies on the prevalence of malaria in Togo are descriptive and have focused on malaria and its various complications in children (Assimadi et al. 1998). No current molecular epidemiology data on malaria parasites are available. This study therefore aimed (i) to identify the genetic diversity of populations of malaria parasites in Togo; (ii) to evaluate the polyallelism frequency; (iii) to evaluate polyclonal infection in relation with the severity of the disease, (iv) to determine whether homocysteine and transaminases levels could be used to apprehend the severity of patients with P. falciparum malaria.

Materials and methods

Study area

The study was conducted in the maritime region of Togo, between the Republic of Ghana, the Republic of Benin, the Atlantic Ocean and the Plateau region of Togo. The climate is sub-equatorial. There are a valley and the flood plains of rivers Haho, Mono and Zio. These environmental conditions allow the persistence of mosquito vectors, mainly Anopheles gambiae, year-round. Therefore, malaria is endemic in the region. Plasmodium falciparum is the predominant species. The two districts selected for sample collection were Bè in Lomé, the capital, and Tsévie, 35 km north of Lomé.


Patients were children aged between 6 months and 15 years, suffering from malaria with clinical symptoms meeting the criteria of malaria severity defined by the World Health Organization (WHO 2000). All patients were attending the two hospitals for the diagnosis and the treatment. Verbal consent was obtained by the paediatrician after a detailed explanation of the study for illiterate mothers. Guardians who could write were asked to give written formal consent. Ethical approval was obtained from the Ethical Committee of the districts.

Blood samples

From each patient, a digital puncture was made to deposit blood on Whatman no.°3 papers for parasite DNA extraction. Afterwards, oil-immersed, Giemsa-stained thick film smears were examined microscopically for Plasmodium species identification and parasite density. Parasite density was assessed by counting the number of asexual parasites per 200 leucocytes. A sample was considered negative after 200 fields were examined without parasites. Parasite counts were converted to parasites per μl of blood assuming an average of 8000 leucocytes per μl blood. Patients included in the study were those who presented a monoinfection caused by P. falciparum with a parasite count of at least 1000 parasites per μl of blood. Venous blood was collected from these patients for haematology parameters, glycaemia, creatinine, transaminases and homocysteine quantification.

Parasite DNA extraction and nested PCR

The DNA template for the nested PCR for each sample was obtained using rapid boiling methods as described by Henning et al. (1999) with slide modifications. Briefly, a piece of Whatman no.°3 paper (5 mm/2 mm) with blood was placed in a tube containing 100 μl of methanol for 15 min. Methanol was discarded and the paper dried at room temperature. Then the paper was reintroduced in a tube with 100 μl of sterile water and heated at 90 °C for 15 min under permanent agitation. Afterwards, the mixture was centrifuged and the supernatant was used for PCR.

PCR genotyping was performed as described by Snounou et al. (1999), using repetitive regions found in three polymorphic genetic markers, namely msp-1, msp-2 and msp-3. Allelic variants of msp-1 (MAD20, K1 and RO33), msp-2 (FC27 and 3D7) and msp-3 (non-repetitive C-terminal) were detected by allelic family-specific nested PCR. The PCR conditions and the primer sequences were fully described by Snounou et al. (1993a,b, 1999) and included in the database UNDP/World Bank/WHO-TDR.PCR. The amplification was performed with Biometra T3 thermocycler. Amplification products were separated by electrophoresis at 150 V on 1.5% agarose gel, and the fragments were visualized under UV light with ethidium bromide. The sizes were determined using a molecular weight marker (Biolabs).

The prevalence of each family was calculated as the percentage of samples containing at least one allele from that family. Multiclonal infections were defined as infections with more than one genotype.

Biological analysis

Blood cells’ counts were performed from whole blood collected in the tubes coated with EDTA with haematology semi-automate type Sysmex KX-21N. Then the blood was centrifuged at 2000 g for 5 min to discard plasma. This was used for transaminase, homocysteine, creatinine and blood glucose quantification with biochemistry automate AxSYM (Abbott Technologies) using Cypress diagnostics kits.

Statistical analysis

The results were presented as averages followed by standard deviations. The data were analysed using Epi Info version 6.04 software (Centers for Disease Control & Prevention, USA). Differences between frequencies and means were estimated using chi-square and Student’s t-test, respectively. Pearson regression analysis was used for the correlation between severity of the disease and biochemical data. Statistical significance was set at P < 0.05.



We enrolled 309 patients, of whom 177 were recruited from Bè/Lomé (80 severe and 97 uncomplicated malaria cases) and 132 from Tsévié (56 severe and 76 uncomplicated malaria cases). The mean age of patients was significantly younger in severe malaria cases in both two places (57.78 vs. 68.62 months and 37.29 vs. 52.24 months, from Bè Lomé and Tsévié, respectively) (Table 1). The mean parasite density was higher in the patients with severe malaria, in both places (44 993.04 vs. 27 966.09 parasites/μl and 27 966.09 vs. 18 823.16 parasites/μl, respectively), although temperatures did not differ significantly. Haemoglobin (8.04 vs. 9.32 and 8.63 vs. 10.42 g/dl), haematocrit (25.05 vs. 28.62% and 25.10 vs. 28.55%), and red blood cells (3.37 vs. 3.77 and 2.93 vs. 3.29) dropped significantly in patients with severe malaria. Similarly, the means of corpuscular volume (MCV) (75.13 vs. 76.00 and 86.53 vs. 88.92), corpuscular haemoglobin (MCH) (24.17 vs. 24.83 and 29.98 vs. 31.94) and corpuscular haemoglobin concentration (MCHC) (32.16 vs. 32.60 and 34.69 vs. 35.49) were lower in patients with severe malaria. On the other hand, mean white blood cells increased in patients with severe malaria (11.51 vs. 8.88 and 9.35 vs. 8.39 for the white blood cells (WBC), and 44.12 vs. 45.54 and 55.38 vs. 48.82 for lymphocytes). Neutrophils increased in patients with severe malaria from Bè/Lomé (46.75 vs. 44.28), whereas the opposite was observed in patients from Tsévié (42.33 vs. 49.91). Patients from Bè/Lomé with severe malaria had the highest level of glutamic-oxaloacetic transaminase (GOT) (167.28 vs. 114.07) while in Tsévié this was so in patients with uncomplicated malaria (112.59 vs.135.64). Glutamic-pyruvic transaminase (GPT) did not vary significantly between patients. Serum glucose level fell in patients with severe malaria (0.71 vs. 0.93 and 0.87 vs. 1.14) in both provinces; however, this was not significant in Tsévié patients. There was no significant variation of creatinine level in Bè/Lomé patients, while this level significantly decreased in patients with uncomplicated malaria from Tsévié. The homocysteine decreased significantly in severe malaria cases in Bè/Lomé (9.43 vs. 10.46) while no significant variation was noticed in Tsévié patients.

Table 1.   Main clinical data and characteristics of patients with uncomplicated or severe malaria according to areas of collection
Severe malaria (= 80)Uncomplicated malaria (= 97)Severe malaria (= 56)Uncomplicated malaria (= 76)P
  1. Data in the table are geometric means followed by standard deviations.

  2. RBC, red blood cells; WBC, white blood cells; MCV, mean corpuscular volume; MCH, mean corpuscular haemoglobin; MCHC, mean corpuscular haemoglobin concentration; GOT, glutamic-oxaloacetic transaminase; GPT, glutamic-pyruvic transaminas.

  3. *P for chi-square test by comparison of percentages, other P-values are for Student’s t-test by comparison of means.

Sex (male %)47.37%45.83%NS*56.36%52.05%NS*
Age (months)57.78 ± 33.2168.62 ± 42.90<0.0137.29 ± 28.4352.24 ± 44.85<0.01
Temperature (°C)38.52 ± 1.2238.45 ± 1.25NS38.35 ± 1.1638.59 ± 1.29NS
Parasitemia (Parasites/μl)44 993.04 ± 65 639.1027 966.09 ± 46 193.51<0.0127 966.09 ± 37 336.5618 823.16 ± 29 393.51<0.01
Haemoglobin (g/dl)8.04 ± 2.409.32 ± 1.60<0.018.63 ± 2.4410.42 ± 2.04<0.01
Haematocrit (%)25.05 ± 7.4428.62 ± 4.64<0.0125.10 ± 6.6828.56 ± 4.80<0.01
RBC (106)3.37 ± 1.033.77 ± 0.59<0.012.93 ± 0.723.29 ± 0.590.01
WBC (103)11.51 ± 10.368.88 ± 4.36<0.019.35 ± 6.748.39 ± 3.810.01
MCV (fl)75.13 ± 8.6176.00 ± 6.720.0386.53 ± 11.2488.92 ± 8.83<0.01
MCH (pg)24.17 ± 3.1824.83 ± 2.890.0129.98 ± 4.2531.94 ± 4.73<0.01
MCHC (g/dl)32.16 ± 2.0932.60 ± 1.530.0334.69 ± 4.8535.49 ± 3.600.02
Neutrophils (%)46.75 ± 15.3244.28 ± 14.82<0.0142.33 ± 12.0049.91 ± 14.92<0.01
Lymphocytes (%)44.12 ± 15.2645.54 ± 14.380.0155.38 ± 14.2948.82 ± 14.88<0.01
Glycaemia (g/l)0.71 ± 0.350.93 ± 0.310.010.87 ± 0.811.14 ± 0.65NS
Creatinine (mg/l)6.33 ± 5.356.34 ± 6.30NS5.22 ± 1.866.10 ± 4.48<0.01
GOT (U/l)167.28 ± 164.95114.07 ± 108.67<0.01112.59 ± 109.39135.64 ± 147.67<0.01
GPT (U/l)32.60 ± 24.1631.00 ± 31.61NS32.34 ± 24.9633.43 ± 23.30NS
tHcy (μm/l)9.43 ± 4.9310.46 ± 6.63<0.019.19 ± 2.449.25 ± 3.44NS

We used Pearson linear regression to analyse the possible correlation between the parasite counts and the tHcy levels (Figure 1), the parasite counts and GPT and GOT levels (Figures 2 and 3), and the tHcy levels and the haemoglobin amounts (Figure 4). In both cases, there was a very low correlation coefficient (R2 < 0.1), suggesting no relation between the above mentioned data by linear regression analysis.

Figure 1.

 Association between parasitemia and homocysteine levels in malaria patients from Bè/Lomé (a) and Tsévié (b).

Figure 2.

 Association between parasitemia and GPT in malaria patients from Bè/Lomé (a) and Tsévié (b).

Figure 3.

 Association between parasitemia and GOT in malaria patients from Bè/Lomé (a) and Tsévié (b).

Figure 4.

 Association between homocysteine levels and the amount of haemoglobin in malaria patients from Bè/Lomé (a) and Tsévié (b).

Allelic variant distribution among patients with malaria

The first round PCR of the 309 samples yielded 210 msp-1 fragments, 227 msp-2 fragments and 193 msp-3 fragments corresponding to a detection efficiency of 67.96%, 73.46% and 62.45%, respectively. The second round PCR yielded 890 fragments distributed as follows: 284 fragments for msp-1, 379 for msp-2 and 227 for msp-3. As shown in Table 2, for genotypes msp-1, 22 different alleles were recorded: 9, 9 and 4 for K1, MAD20 and RO33 varying in size from 80 to 320, 80 to 300 and 100 to 180 bp, respectively. For the block msp-2, 33 different alleles were detected, 21 for FC27 and 12 for 3D7. The sizes varied from 80 to 850 and 380 to 950 bp, respectively. Thirteen alleles with sizes ranging from 200 to 600 bp were recorded for msp-3. At each locus, the family distribution was 54.58% K1, 25% MAD20 20.42% RO33 for msp-1, and 51.71% and 48.29% FC27 and 3D7, respectively, for msp-2.

Table 2.   Allelic families’ distribution
Allelic families
msp-1 (= 210)msp-2 (= 309)msp-3 (= 227)
K1 (= 155)MAD20 (= 71)RO33 (= 58)FC27 (= 196)3D7 (= 183)
  1. Number of samples for a given allele size is indicated in parenthesis.

80 (1)80 (1)100 (6)80 (3)380 (1)200 (5)
150 (7)100 (6)150 (38)120 (1)400 (31)280 (1)
180 (9)120 (3)160 (12)150 (1)420 (12)300 (4)
200 (67)150 (6)180 (2)180 (9)450 (21)350 (4)
220 (21)180 (8) 200 (31)480 (25)380 (4)
250 (34)200 (32) 220 (3)500 (47)400 (29)
280 (9)220 (8) 240 (1)520 (18)420 (6)
300 (6)250 (6) 250 (13)550 (17)450 (91)
320 (1)300 (1) 280 (16)580 (5)480 (3)
   300 (52)600 (4)500 (71)
   320 (6)650 (1)520 (2)
   350 (27)950 (1)550 (6)
   380 (7) 600 (1)
   400 (15)  
   420 (1)  
   450 (3)  
   520 (1)  
   550 (2)  
   580 (1)  
   600 (2)  
   850 (1)  

Prevalence of multiclonal infections

Among the PCR positive samples, a maximum of nine genotypes was detected in one patient suffering from severe malaria. The maximum of eight genotypes was detected in two patients with uncomplicated malaria. Patients with severe malaria and uncomplicated malaria were mainly infected with at least three clones. Indeed, 28.67% and 26.16% were infected with three clones in severe malaria and uncomplicated malaria, respectively. The detailed distribution of the genotype number per isolate is displayed in Figure 5.

Figure 5.

 Distribution of the number of genotypes per sample of patients with severe malaria and uncomplicated malaria.

Influence of age and parasite density

The patients were divided into three age groups: 6–60, 61–120 and 121–180 months (Table 3). The children in the age group of 61–120 months had the highest average number of different genotypes in their samples: 1.50 with msp-1 and 1.88 with msp-2. On the other hand, the average number of msp-1 and msp-2 bands per isolate was lower in 121–180 months age group (Table 3). The frequencies of multiclonal infections for theses alleles were significantly different (P = 0.002 for msp-1 and P = 0.007 for msp-2). However, there was no significant influence of age on the distribution of the allelic families of msp-3 (P = 0.310). The parasite density was also stratified into three groups: 1000–50 000 parasites/μl of blood, 50 001–100 000 parasites/μl of blood and >100 001 parasites/μl of blood. Chi-square analysis of proportions revealed that the parasitemia did not affect the distribution of allelic families (P = 0.062 for msp-1, P = 0.151 for msp-2 and P = 0.141 for msp-3).

Table 3.   Mean number of msp-1, msp-2 and msp-3 alleles by age groups and parasite density
  1. P-value is for chi-square test by comparison of the number of allele within stratified groups.

Stratified parasite density (parasites/μl of blood)
1000–50 000 (= 247)2331.360.0623211.680.1511881.210.141
50 001–100 000 (= 22)291.47261.41211.31
>100 001 (= 40)221.37 321.87 181.11 
Stratified age groups (months)
6–60 (= 203)1991.320.0022571.620.0071571.210.310
61–120 (= 69)731.501011.88521.24
121–180 (= 37)121.20211.42181.20

Influence of malaria type and area of collection

Data in Table 4 reveal that for msp-1, the allelic variants K1 and MAD20 seemed to be most prevalent in patients with uncomplicated malaria, while RO33 was most frequent in patients with severe malaria. For msp-2, both allelic variants FC27 and 3D7 seemed to be frequent in patients with uncomplicated malaria. Moreover, msp-3 also seemed to be most frequent in patients with uncomplicated malaria. For all these allelic variants, the distribution was not associated with the type of malaria (P > 0.05). In relation with the zone of habitation, all allelic variants for msp-1 and msp-2 seemed to be most frequent in Bè/Lomé patients, while msp-3 seemed to be most prevalent in Tsévié patients, although no significant P-value was recorded for these data by chi-square test (P > 0.05).

Table 4.   Allelic distribution in relation with the severity of malaria and the area of collection
Allelic familyNumber of alleles
Type of malariaPAreaP
SM (= 136)UM (173)Bè/Lomé (= 177)Tsévié (= 132)
  1. SM, severe malaria; UM, uncomplicated malaria.

  2. P-value is for chi-square test by comparison of the number of alleles in relation with the type of malaria and in relation with the area of collection.



We investigated, for the first time, the genetic diversity and genotype multiplicity of P. falciparum infections in symptomatic individuals living in the maritime region of Togo. The high level of allelic diversity recorded could be underestimated because microscopy positive samples may be negative to the nested PCR. In this study of 309 samples, 99 were negative to msp-1, 82 to msp-2 and 116 to msp-3.

At the msp-1 locus, K1 and MAD20 were the most polymorphic allelic families with nine alleles each. Mlambo et al. (2007) in their study also found same number of alleles for K1 and MAD20. However, K1 was the allelic family most frequently detected in isolates in our study. These findings are in contrast with the study of Dolmazon et al. (2008). At the locus msp-2, FC27 was the most polymorphic with 19 different alleles. This is consistent with previous reports (Aubouy et al. 2003; Basco et al. 2004; Dolmazon et al. 2008). Our results indicate that msp-2 is more polymorphic than msp-1 and msp-3. This genetic marker is the best for estimating the multiplicity of P. falciparum infection in symptomatic patients because it is significantly related to parasite density (Dolmazon et al. 2008). However, in this study we did not find significant relation between allelic distribution and the parasite count. Consequently, there was no significant relation between the allelic distribution and the severity of the disease. Indeed, the mean number of circulating genotypes in severe malaria cases was similar to that of the uncomplicated ones. Our results are consistent with those of Durand et al. (2008) who did not find an association between severity of disease and allelic distribution in Madagascar. Similar findings were reported by Soulama et al. (2005) in Burkina Faso. Globally, the allelic distribution was significantly associated with the age of patients. Indeed, multiplicity of infection (MOI) was highest in the age group of 6–120 months with allelic families msp-1, msp-2 and msp-3 but this MOI decreased significantly in patients aged from 121 to 180 months. Similarly, several authors found the highest MOI in 3–5-year-old patients, and this MOI seemed to decrease with the age of patients (Konate et al. 1999; Dolmazon et al. 2008). In fact some authors suggested that MOI could be stable throughout the first year of life and increase only at its end. Paralleled by the increasing ability to control parasitaemia, this may indicate a change of the predominant type of host response at this age (Buchholz et al. 2010). Conversely, others did not find significant association between MOI and the age of patients (Felger et al. 1999; Smith et al. 1999).

The sites were urban (Bè/Lomé) and rural (Tsévié), separated by 35 km. Our results indicate that there was no significant difference in allelic distribution between the two. However, Dolmazon et al. (2008) found significant difference in allelic distribution in two districts of Bangui, confirming that the epidemiological features of P. falciparum may vary within a given city (Trape et al. 1992).

The homocysteine level significantly decreased in the patients with severe malaria from Bè/Lomé, and there was a very low Pearson correlation either with the parasite counts or with the amounts of haemoglobin. These findings contrast several reports that described high positive correlation between the tHcy levels with the severity of the disease (Chillemi et al. 2004; Abdel Gader et al. 2009). The accumulation of tHcy in patients with uncomplicated malaria is consistent with the study of Chabi (2009). The moderate hyperhomocysteinemia may have multifactorial causes, including nutritional and environmental factors. Indeed, in a study on nutritional and genetic determinants of tHcy in the population in the same area, Amouzou et al. (2004) revealed a high prevalence of moderate hyperhomocysteinemia, associated with a deficiency of folate and a negative involvement of polymorphisms of the enzymes of the homocysteine metabolism.

The GPT level significantly increased in patients with severe malaria but there was also a very low Pearson correlation with the parasite count. Our results may suggest a drug-induced hypertransaminasemia (Godeau et al. 2002). In fact some parents rely on self-medication and only attend hospital in case of persistent fever or other complications. Therefore, one factor that might potentially compromise a comparison of genotypic characters of isolates collected from patients experiencing a clinical episode is the uncontrolled drug intake before presentation to the hospital. Another limitation of this work was that we studied only circulating parasites and not those that are sequestered in the tissues. Some authors suggested that the dominant clones sequestered in deep organs are usually the same as those in peripheral circulation (Durand et al. 2008).

The capillary electrophoresis genotyping method allowed us to observe a significant allelic diversity in clinical isolates of the endemic area in southern Togo. The severity of malaria was not associated with higher MOI and did not appear either restricted to some particular genotypes. More comprehensive explorations including immunity and genetic factors as well as the nutritional and sociologic status of the population could lead to highlight the current situation.


We are grateful to the patients and the accompaniers included in this study and the staff of Bè, Tsévié hospitals and Clinic Biassa. We also thank Mr Lawson from Prolabo Diagnosis who kindly provided reagents for the quantification of biochemical parameters.