No Association of the p53 Codon 72 Polymorphism with Malaria in Ghanaian Primiparae and Rwandan Children

In view of the ongoing burden of malaria, improved knowledge of pathophysiological mechanisms may provide novel targets for prevention and treatment.1 Recently, the p53 tumor suppressor protein, a central host cell-signaling factor, has been shown to be critical for successful Plasmodium liver stage infection.2 As a transcription factor, p53 responds to various stimuli to induce apoptosis or cell cycle arrest, or to integrate a variety of other host responses.3,4 Plasmodium liver stage parasites suppress p53 thereby promoting survival of the infected hepatocyte. Conversely, increased levels of p53 counterbalance this suppression and reduce liver stage parasite burden.2

The TP53 gene, encoding p53, shows a common single nucleotide polymorphism at codon 72 (proline to arginine, Pro > Arg, rs1042522), conferring various functional consequences including the Arg allele being more effective at inducing apoptosis.5 This allele has been associated with altered susceptibility to several virus-related and other cancers.4,6 Notably, indirect evidence for a malaria-protective role of the TP53 codon 72 Arg allele is derived from a small study in Sardinia.7

We, therefore, examined the distribution of the TP53 codon 72 alleles in two African populations with or without P. falciparum infection, i.e., placental P. falciparum infection in Ghanaian primiparae, and largely asymptomatic P. falciparum infection among Rwandan children.

Socio-demographic, clinical, and parasitological characteristics of the two cross-sectional studies and the study groups have been reported elsewhere.8,9 In brief, 314 primiparous pregnant women were recruited in southern Ghana of whom two-thirds (by polymerase chain reaction [PCR]) had largely asymptomatic placental P. falciparum infection (i.e., only 6.3% of these were febrile), which nevertheless was associated with maternal anemia, low birth weight, and preterm delivery.8 The second group involved 545 children < 5 years of age randomly selected from 24 rural villages in southern highland Rwanda of whom 16.1% were P. falciparum infected (by PCR; 2.9% categorized as symptomatic malaria defined as a positive blood film plus fever, or a history of fever within the preceding 48 hours).9 All participants (or legal guardians) gave informed consent, and the study protocols were approved by the Committee on Human Research Publications and Ethics, School of Medical Sciences, University of Science and Technology, Kumasi, Ghana, and the National Ethics Committee, Republic of Rwanda, respectively. The DNA was extracted from stabilized blood samples (AS1 and QIAmp DNA Blood Mini Kit; Qiagen, Hilden, Germany). The Arg allele was differentiated from the Pro allele at codon 72 of TP53 by restriction fragment length polymorphism of PCR-generated amplicons using primers and the restriction enzyme BstU1 as published elsewhere.10 Data were analyzed with Statview 5.0 (SAS Institute Inc., Cary, NC). Allele frequencies and genotypes were compared by the χ² test, and continuous variables by the Mann Whitney U test or Student's t test as applicable.

The distribution of TP53 codon 72 genotypes according to P. falciparum infection among Ghanaian primiparous women and Rwandan children is presented in Table 1. Genotypes were in Hardy–Weinberg equilibrium among Rwandan children (P = 0.51) but not among Ghanaian women (P = 0.03). Arg allele frequencies (Ghana, 0.30; Rwanda, 0.31) and genotypes did not differ between P. falciparum infected and non-infected subjects. Adjusting for the age difference between infected and non-infected individuals and for further associated factors,8,9 the lack of association between TP53 genotypes and P. falciparum infection remained (data not shown). Likewise, there was no association between genotypes and peripheral blood geometric mean parasite density (/μL, 95% confidence interval [CI]), in either pregnant women (Pro/Pro, 1,096 [513–2,346]; Pro/Arg, 675 [390–1,168]; Arg/Arg, 1,449 [134–15,690]; P = 0.52) or in children [1,837 (978–3,449]; 968 [404–2,322]; 6,792 [615–75,062]; P = 0.12). In pregnant women, the TP53 codon 72 allele had no influence on maternal anemia, birth weight, or preterm delivery, irrespective of placental malaria (data not shown), and in Rwandan children, the prevalence of symptomatic malaria did not differ significantly with genotype (Pro/Pro, 3.1% [8 of 260]; Pro/Arg, 1.8% [4 of 228]; Arg/Arg, 7.1% [4 of 57]; P = 0.11).

Findings from murine models indicate a critical role of p53 in susceptibility to Plasmodium infection in that increased levels reduce parasite liver stage burden.2 The Arg allele of the TP53 codon 72 variant is pro-apoptotic, potentially protective with respect to virus-related and other cancers,4–6 and suggested to be subject to selection because of protection against malaria.7 In this study, we are unable to show a role of the TP53 codon 72 allele in malaria. As a matter of fact, this study was not a priori designed to assess an association between TP53 codon 72 alleles and P. falciparum infection. Considering the given prevalence, the sub-studies were powered (80%) to detect only substantial reductions of infection caused by the Arg allele (Ghana, OR, ≤ 0.60; Rwanda, OR, ≤ 0.50). Smaller effects may thus be discernible at considerably larger sample sizes. Nevertheless, allele frequencies in our African study populations were considerably lower than in Sardinia or other Caucasian populations,7,11,12 and a declining Arg allele frequency toward the equator argues against selection by malaria.11,12 Notably, respective genetic data from Africans are comparatively scarce. Genotype frequencies deviated from Hardy–Weinberg equilibrium in the Ghanaian subgroup. One reason for this might be a small sample size. Interestingly, the TP53 codon 72 Pro/Pro genotype has been associated with reduced pregnancy rates,13 and in fact, this genotype tended to be underrepresented among the Ghanaian primiparae. The present findings of lacking genetic association do not exclude p53 to be pathophysiologically relevant in malaria,2 but argue against a major respective role of the TP53 codon 72 allele. The p53 is activated by various stress signals and can induce a variety of host responses3,4; possibly, the biological effects conferred by this allele are not those essential in antiplasmodial defense. Alternatively, the variant may not influence P. falciparum infection per se, as examined in this study, but rather specific malaria entities, e.g., severe malaria, or infection risk in other populations.