An Observational Study of the Effect of Hemoglobinopathy, Alpha Thalassemia and Hemoglobin E on P. Vivax Parasitemia.

BACKGROUND: The protective effect of α-thalassemia, a common hematological disorder in Southeast Asia, against Plasmodium falciparum malaria has been well established. However, there is much less understanding of the effect of α-thalassemia against P. vivax. Here, we aimed to investigate the proportion of α-thalassemia including the impact of α-thalassemia and HbE on the parasitemia of P. vivax in Southeast Asian malaria patients in Thailand.
METHODS: A total of 210 malaria patients, admitted to the Hospital for Tropical Diseases, Thailand during 2011-2012, consisting of 159 Myanmeses, 13 Karens, 26 Thais, 3 Mons, 3 Laotians, and 6 Cambodians were recruited. Plasmodium spp. and parasite densities were determined. Group of deletion mutation (--SEA, -α3.7, -α4.2deletion) and substitution mutation (HbCS and HbE) were genotyped using multiplex gap-PCR and PCR-RFLP, respectively.
RESULTS: In our malaria patients, 17/210 homozygous and 74/210 heterozygous -α3.7 deletion were found. Only 3/210 heterozygous -α4.2 and 2/210 heterozygous--SEA deletion were detected. HbE is frequently found with 6/210 homozygotes and 35/210 heterozygotes. The most common thalassemia allele frequencies in Myanmar population were -α3.7 deletion (0.282), followed by HbE (0.101), HbCS (0.013), -α4.2 deletion (0.009), and --SEA deletion (0.003). Only density of P. vivax in α-thalassemia trait patients (-α3.7/-α3.7, --SEA/αα, -α3.7/-α4.2) but not in silent α-thalassemia (-α3.7/αα, -α4.2/αα, ααCS/αα) were significantly higher compared with non-α-thalassemia patients (p=0.027). HbE did not affect P. vivax parasitemia. The density of P. falciparum significantly increased in heterozygous HbE patients (p=0.046).
CONCLUSIONS: Alpha-thalassemia trait is associated with high levels of P. vivax parasitemia in malaria patients in Southeast Asia.

Introduction

Malaria is the most prevalent parasitic disease worldwide where 214 million patients suffer due to Plasmodium vivax and Plasmodium falciparum infection, and more than 400,000 people die annually.1 Both P. vivax and P. falciparum have been the main causes of malaria on the Thailand-Myanmar border for many years. In Thai villagers, P. vivax infection has recently become the largest proportion of cases in patients.2 The result of selective malaria pressure on recent human genome evolution is presented in the form of high frequencies of genetic disorders of hemoglobin including thalassemias and hemoglobinopathies in populations living in historically malarious regions.3–6 Such malaria-protective properties have since been demonstrated in glucose 6-phosphate dehydrogenase (G6PD) deficiency,7 α-thalassemia,4,8–9 hemoglobin C,10–11 hemoglobin S12 and hemoglobin E.13 The protective effect of thalassemia against P. falciparum malaria has been well established.14 However, the impact of thalassemia on P. vivax is not well understood yet. Alpha-thalassemia is caused by the deletion of a number of α-globin genes resulting in an imbalance of α- and β-globin. There are several types of α-thalassemia; silent α-thalassemia, α-thalassemia trait and HbH, which depleted one, two, and three copy of α-globin genes, respectively. The −α3.7 and-α4.2 deletions are most common forms of silent α-thalassemia in Southeast Asians.15–16 Clinical symptoms of α-thalassemia traits are mild anemia with hypochromic erythrocytes, whereas heterozygotes are asymptomatic.17 The meta-analysis demonstrated the protective effect of silent α-thalassemia against P. falciparum.14 A case-control study in Africa and Papua New Guinea (PNG) found that silent α-thalassemia protects against P. falciparum.8–9,18–21 Alpha-thalassemia trait --SEA deletion was commonly found in Thailand and Southeast Asia (SEA).15–16 The --SEA allele has been identified as the recent balancing selected allele triggered by malaria.22 However, several studies failed to detect the association of α-thalassemia traits and parasitemia of P. vivax.23 Hemoglobin E (HbE) is the most common β-hemoglobinopathies in Southeast Asia. Several studies have found that HbE confers protection against P. falciparum.13 However, HbE has been found to be more prone to P. vivax.24 This study aimed to investigate the proportion of α-thalassemia and HbE and to clarify the effect of α-globin gene numbers and HbE genotype on the parasitemia in Southeast Asian malaria patients in Thailand.

Materials and Methods

Study subjects and sample collection

The study protocol was reviewed and approved by the Institutional Review Board of the Faculty of Medicine, Chulalongkorn University (Bangkok, Thailand) (COA No. 040/2013 IRB No. 459/55). Malaria patients in this cohort study were referred from many malaria-endemic provinces including borders of Thailand: Tak (Maesod District), Kanchanaburi (Sangkhlaburi District), Phetchaburi (Kaeng Krachan District), Suphanburi (Dan Chang District), Ranong, Sisaket (Kantharalak District) and Chonburi (Figure 1). Before enrolment in the study, all patients gave written informed consent. Patients who were slide-positive for Plasmodium malaria with no history of antimalarial drug treatment within the preceding 2 weeks, and were admitted to the Hospital for Tropical Diseases in Thailand during 2011–2012, were recruited. G6PD deficiency, an enzymopathy involved in protecting against malaria, which may interfere with interpretation of the effect of α-thalassemia and HbE, was excluded.

Figure 1

Distribution of malaria patients cohort along Thailand and borders during 2011–2012 (Missing geographic data in 19 cases) (Pf., Pv., Pm., and Pf.+Pv. represent P. falciparum, P. vivax, P. malariae, and mixed infection of P. falciparum and P. vivax, respectively.)

Identify Plasmodium spp. infection and parasite density

All blood samples from finger pricks were Giemsa stained for thick and thin blood films. Blood smears were tested every 12 hours from initiation of treatment until they were negative on two consecutive occasions; after that, blood smears were daily tested until patients were discharged. Parasite densities (asexual parasite/microliter of blood) were examined by counting the number per 200 leukocytes (thick film) or per 1,000 erythrocytes (thin film). In interpretation the Plasmodium spp., blood smear films were read under microscope by an independent parasitologist at the Hospital for Tropical Diseases. The species was confirmed by polymerase chain reaction (PCR)-based analysis.

Measurement of G6PD activity

G6PD activity assays were performed prior to treatment and weekly repeated until patients were discharged. Quantitative test for G6PD activity was performed using G6PD kit assay (Trinity Biotech, Bray, County Wicklow, Ireland), which measured NADPH production at wavelength of 340 nm. All samples were run parallel with positive and negative control. Hemoglobin for calculation of G6PD activity was measured using Hb201 (HemoCue, Sweden). G6PD activity <1.5 IU/g Hb classified as G6PD deficient25 was excluded from the study. Leftover blood samples were kept at −20°C for molecular typing.

Detection of α-thalassemia

Genomic DNA was extracted from peripheral blood using phenol-chloroform method.26 Alpha-globin gene variants including α-thalassemia trait (--SEA deletion) and silent α-thalassemia (−α3.7, −α4.2 deletion) were investigated by multiplex gap-polymerase chain reaction (multiplex gap-PCR).27 HbCS and HbE were genotyped using PCR-restriction fragment length polymorphism (PCR-RFLP).28–29

Statistical analysis

All statistical analyses were performed using the SPSS version 22.0. The main outcomes of interest were parasite densities of P. falciparum and P. vivax malaria before treatment. Parasite density that was not normally distributed was log-transformed prior to analysis. Parasitemia of α-thalassemia and HbE patients were compared with that of non-thalassemia (HbA) using unpaired T-test. In all statistical analyses, significance levels were set at the 95% confidence interval (CI) (P<0.05).

Results

Characteristics of the study population

A total of 210 patients (201 males and 9 females) including 159 Myanmeses, 13 Karens, 26 Thais, 3 Mons, 3 Laotians, and 6 Cambodians were recruited for the study. Patients were from Myanmar (N= 159), Tak (Maesod district, N= 127), Kanchanaburi (Sangkhlaburi district, N= 9), Ranong (N= 1), Thailand-Myanmar border (N= 15), Thailand-Cambodia border (N= 1, Figure 1) and missing data (N= 6). The average age of all subjects was 28.0±10.0 (range 14–60) years. Eighty-five had P. falciparum, while 122 had P. vivax infection, two had mixed infection of P. falciparum, and P. vivax and one had P. malariae.

In this study, 17 homozygous and 74 heterozygous −α3.7 deletion were found among 210 patients, while only three heterozygous −α4.2 and two heterozygous --SEA deletion were detected.

HbE was also highly prevalent, with six homozygotes and 35 heterozygotes. For HbCS, five heterozygous were detected (Table 1). The Myanmese was the major ethnic group in this study accounting for 75% of all patients. Among these, the proportion of α-thalassemia was 48.4% (77/159), including 45.9% (73/159) of −α3.7 deletion, 1.8% (3/159) −α4.2 deletion, 0.6% (1/159) --SEA deletion, and 2.5% (4/159) ααCS whereas HbE was 20.8% (29/159). Allele frequencies were calculated for the major population. The most common was −α3.7 deletion (0.282), followed by HbE (0.101), HbCS (0.013), −α4.2 deletion (0.009), and --SEA deletion (0.003) (Table 1). Thalassemia and hemoglobinopathies were not found in 3 patients with P. malariae and mixed infection patients.

Table 1

Proportion of thalassemia and hemoglobinopathies in malaria patients, divided by Plasmodium spp. infection and ethnicity.

Mutation	Genotype	Plasmodium spp.(N)		Ethnic group(N)						Total number
		Plasmodium falciparum	Plasmodium vivax	Myanmese	Karen	Thai	Mon	Laotian	Cambodian
HbAa	αα /αα	45	68	84	9	16	2	0	5	116
−α3.7	−α3.7/αα	31	43	57	4	8	1	3	1	74
	−α3.7/−α3.7	9	8	16	0	1	0	0	0	17
		Allele frequency	0.282	0.154	0.192	0.250	0.500	0.083
−α4.2	−α4.2/αα	1	2	3	0	0	0	0	0	3
		Allele frequency	0.009	0.000	0.000	0.000	0.000	0.000
--SEA	--SEA/αα	0	2	1	0	1	0	0	0	2
		Allele frequency	0.003	0.000	0.019	0.000	0.000	0.000
HbCS	ααCS/αα	3	2	4	0	1	0	0	0	5
		Allele frequency	0.013	0.000	0.019	0.000	0.000	0.000
HbE	βE/β	17	18	26	0	6	0	2	1	35
	βE/βE	1	5	3	0	2	0	0	1	6
		Allele frequency	0.101	0.000	0.192	0.000	0.333	0.250

HbAa 2 cases were mixed infection of P. falciparum and P. vivax and 1 had P. malariae.

Association of α-globin gene dosage, HbE, and parasitemia

To assess the effect of α-globin gene presence and HbE genotype on the parasitemia of P. vivax and P. falciparum, the number of parasites in the blood of α-thalassemia and HbE genotypes were compared with that of non-thalassemia (HbA). The results found that P. vivax density in patients with α-thalassemia trait (−α3.7/−α3.7, --SEA/αα, −α3.7/−α4.2) was 4.21±0.32 log10 value/μl, which was significantly higher than HbA patients (3.89±0.71 log10 value/μl) (p=0.027) (Table 2). Whereas, P. vivax parasitemia was not significantly different in patients who depleted only one α-globin gene or had silent α-thalassemia (−α3.7/αα, −α4.2/αα, ααCS/αα) (3.94±0.66 log10 value/μl) (p=0.707) (Table 2). Nevertheless, HbH patient (−α3.7/--SEA) had low level of P. vivax parasitemia compared with HbA (2.08 log10 value/μl). However, there was no significant effect of the number of alpha globin gene deletions on P. falciparum parasitemia.

Table 2

Association between α-globin gene dosage, HbE genotype and number of Plasmodium vivax and Plasmodium falciparum parasitemia.

Genotype	Plasmodium falciparum			Plasmodium vivax
	Patient (N)	Parasitemia (log10 value/μl)	P-value	Patients (N)	Parasitemia (log10 value/μl)	P-value
α-globin dosage
HbA (αα/αα)	40	4.06±0.89	Reference group for alpha globin gene dosage	62	3.89±0.71	Reference group for alpha globin gene dosage
Silent α-thal (−α3.7/αα, −α4.2/αα,ααCS/αα)	30	4.18±1.06	0.631	42	3.94±0.66	0.707
α-thal trait (−α3.7/−α3.7, −α3.7/ααCS, --SEA/αα, −α3.7/−α4.2)	10	3.96±1.14	0.740	10	4.21±0.32	0.027
HbH (−α3.7/--SEA)	0	-	-	1	2.08	There is not enough statistical evidence.
HbE genotype
β/β	64	4.02±1.03	Reference group for HbE	94	3.94±0.65	Reference group for HbE
βE/β	17	4.45±0.66	0.046	17	3.83±0.94	0.662
βE/βE	1	3.00	There is not enough statistical evidence.	5	3.75±0.55	0.512

However, significant increases of P. falciparum density in heterozygous HbE patients was detected (4.45±0.66 log10 value/μl) (p=0.046) (Table 2). On the other hand, P. falciparum parasitemia was reduced in homozygous HbE patient (3.00 log10 value/μl) (Table 2). Nevertheless, this study could not find the effect of HbE on P. vivax parasitemia.

Discussion

Our study is an association study between α-thalassemia and P. vivax density in Southeast Asia. The proportion of P. vivax infection in this study was higher than P. falciparum infection with a ratio of 1.4:1, which corresponds to the WHO World Malaria Report in 20151 which reported that P. vivax (54%) was detected more frequently than P. falciparum (38%) in Thailand. The distribution of P. vivax in Thailand is predominantly along the western region; Tak Province or the Thailand-Myanmar border (Figure 1), which had the highest malaria incidence.30 Since all patients in the study, who were referred to the Hospital for Tropical Diseases after malaria infection, were immigrant laborers, the ratio of males was much higher than female malaria patients. Since a more numerous population of men had been working outdoors, it was exposed to a higher chance of malaria infection.

The overall frequencies of α-thalassemia and HbE in Myanmar villagers living in malaria-endemic regions of Myanmar were 37.5% (343/916) and 20.3% (186/916), respectively.31 Our study is comparable to a previous study and may reflect real prevalence. From our finding and the report of Than31 support α-thalassemia especially −α3.7 deletion and HbE are highly frequent in both malarial and non-malarial infected Myanmar populations. While it is difficult to demonstrate the protective effect of α-thalassemia and HbE when conducting a study only in malaria patients, our findings of high prevalence of thalassemia traits among malaria patients supports the conclusion that malaria infection risk is not reduced in people with α-thalassemia and HbE. In line with this finding, an increased frequency of uncomplicated malaria was found in people with α+-thalassemia in the Vanuatu study.23 The high prevalence of α-thalassemia and HbE in Southeast Asia remains unexplained.

In contrast to the Haldane hypothesis, where α-thalassemia is expected to protect from malaria, we observed higher levels of P. vivax parasitemia among people with α-thalassemia trait. Similarly, a study in Papua New Guinea also showed higher P. vivax parasitemia (but not P. falciparum) in α+-thalassemia heterozygous and homozygous children.17 In addition, the study in Kenya also showed that α+-thalassaemia neither protected against symptomatic malaria nor reduced parasitemia.9 However, α+-thalassaemia appeared to reduce the rate of severe anemia in falciparum malaria and had lower hospitalization.9 The contrasting effects may be explained by the lack of P. vivax in African population, while both P. vivax and P. falciparum are prevalent in Southeast Asian region.

Despite the dosage effect of P. vivax density where two alpha gene deletions have higher levels of parasitemia than one gene deletion, the single case of HbH (3 genes deletion) had an unexpectedly lower rather than higher level of parasitemia. We could not make a meaningful conclusion from this one case as it could have occurred by chance. It was possible that this patient was referred early, so parasitemia was still low. It is hypothesized that people with α-thalassemia have more baseline erythropoiesis, resulting in a high proportion of reticulocytes which is the susceptible stage for P. vivax infection.23 This hypothesis, however, is unlikely as there is no evidence of reticulocytosis in people with α+-thalassemia heterozygous.32

Our results showed increased parasitemia of P. falciparum in heterozygous HbE, but also a decrease in one single case of homozygote. Our finding is in line with a previous study in Myanmar population.33 In vitro studies reveal conflicting results. Nagel et al. demonstrated impairment of the growth of P. falciparum in homozygous HbE, but an average growth in heterozygous HbE.34 Whereas, Chotivanich et al. found in vitro a reduction in RBC invasion in HbAE heterozygotes, associated with a 4-fold increase in the selectivity index compared the other hemoglobin types studied and in particular the EE homozygotes suggesting that in heterozygote individuals with AE hemoglobin, only a quarter of the RBC population can be invaded by P. falciparum, so parasitemia could remain low.13 Parasitemia of P. vivax in HbE patients had been previously observed but did not reach significant difference.28 The effect of HbE on P. vivax parasitemia was not found in this study. Nevertheless, O’Donnell and colleagues showed that HbE patients might be more susceptible for malaria infection, especially P. vivax because their malarial antibodies were significantly increased than non-thalassemia children, which reflected in their clinical severity.20 Although limited by a small number of patients, one strength of our study is that G6PD deficiency was excluded, which has been well known to confer protection against vivax malaria.7