Introduction

Helicobacter pylori infection affects more than 50% of the world population and is associated with various disorders of the proximal gastro‒intestinal tract such as chronic gastritis and duodenal ulcers, gastric adenocarcinoma and gastric lymphoma [1]. The infection is usually acquired in childhood [2, 3], and risk factors are related to socio˗economic status and living conditions [2–5]. Gastric colonization by the microorganism is also influenced by host defences including reactivity of immune cells [6], and efficiency of mechanisms against oxidative challenge [7, 8]. Although some of these conditions have been investigated in detail, a number of them remained largely unexplored.

D-glucose-6-phosphate dehydrogenase (G6PD, D-glucose-6-phosphate: NADP oxydoreductase; EC 1.1.1.49) is a cytoplasmic enzyme able to catalyze the first step of the pentose phosphate pathway, which plays a key role in producing the extra‒mitochondrial coenzyme nicotinamide-adenine dinucleotide phosphate (NADPH) as well as the ribose necessary to synthesize DNA [9]. G6PD deficiency is the most common inherited enzymatic disorder of red blood cells in humans. The G6PD gene maps on the long arm of X chromosome (Xq28 locus), hence deficiency is transmitted by a X-linked pattern [10]. Therefore, total enzyme deficiency is observed in hemizygote men and homozygote women, whereas variable levels of enzyme activity are detected in heterozygote women, resulting from partial compensation by the normal allele and random inactivation of one X chromosome [11]. Carriers of the G6PD deficient allele appear to have a selective advantage against malaria caused by Plasmodium falciparum [12]. For this reason, in areas of the world where malaria is endemic, the frequency of G6PD deficiency could be considerably increased.

Among hundreds of mutant alleles associated with G6PD deficiency, the Mediterranean mutation (C→T transition at nucleotide 563 of G6PD gene) has a frequency of about 12–24% in Sardinia, the highest in Italy [13, 14].

The majority of G6PD‒deficient individuals are entirely asymptomatic. However, subjects carrying the most common G6PD variants are subjected to recurrent acute hemolytic anemia. Hemolysis is triggered by infections, ingestion of Vicia faba (favism), and treatment with several pro‒oxidant drugs.

Deficient enzyme activity may impair protection from oxidative injury in cells other than erythrocytes [15]. More specifically, leukocytes from G6PD deficient subjects show a reduced response to bacterial challenge [16–21]. G6PD deficient granulocytes in G6PD Mediterranean variant displayed reduced function by 25% to 33% in most studies [22, 23], suggesting an increased susceptibility to bacterial infection. However, population studies on the incidence and prevalence of bacterial infections, and more specifically on H. pylori infection in G6PD deficiency patients are lacking.

Therefore, in this study we investigate a possible association between H. pylori infection and G6PD deficiency in a large patient cohort from Northern Sardinia.

Materials and Methods

Results and Discussion

A total of 6565 clinical records were available for the analysis (Table 1). The proportion of women was 65.3% (4287). Mean age was comparable between G6PD-deficient patients and controls (50.4 ± 16.9 vs 50.6 ± 17.3, p = 0.789). The entire study cohort was white Caucasian. Prevalence of H. pylori infection according to age decade and G6PD status is shown in Fig 1. Enzyme deficiency was recorded in 12% (789/6565) of the entire cohort, and more specifically in 8.3% of men and in 14.0% of women. Overall, the proportion of patients positive for H. pylori (active, past and long standing infection) was 50.6% and of 51.5% among G6PD deficient and non-deficient patients respectively, with no statistical difference (χ² = 0.271; p = 0.315). The statistical power given by our sample size was 84%, large enough to avoid the rejection of a false zero hypothesis of a 5% difference around a frequency of H. pylori of about 50%. Table 2 displays the unadjusted ORs and their 95% CIs for acquiring H. pylori infection among patients with and without G6PD deficiency.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Prevalence of H. pylori infection according to age decade and D-glucose-6-phosphate dehydrogenase status.

https://doi.org/10.1371/journal.pone.0160032.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic features of study population according to D-glucose-6-phosphate dehydrogenase status.

https://doi.org/10.1371/journal.pone.0160032.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Unadjusted odds ratio and 95% confidence interval for H. pylori infection among patients with and without G6PD deficiency undergoing upper endoscopy.

https://doi.org/10.1371/journal.pone.0160032.t002

Interestingly, the G6PD deficiency seems to entail a risk for active or past H. pylori infection similar to patients without enzyme deficiency. Patients with G6PD deficiency had a risk of long‒standing infection slightly increased, (OR = 1.21 [95% CI 0.80‒1.83]) although not significantly. Multiple logistic regression analysis is reported in Table 3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Multiple logistic regression analysis for G6PD status and age and gender associated with H. pylori infection.

https://doi.org/10.1371/journal.pone.0160032.t003

After adjustment for all covariates G6PD deficiency was not associated with a significantly increased risk for acquisition of H. pylori infection (OR = 1.02, 95% CI 0.827‒1.252). This lack of association was further confirmed also in the subgroups of patients with past infection (G6PD deficient 104/494; 21.1% vs G6PD normal 794/3598; 22.1%) (p = 0.643) suggesting that the clearance of infection was not affected by the G6PD status. Similarly, no difference was detected in subgroups of patients with long‒standing infection according to the G6PD status (G6PD deficient 28/418; 6.7% vs G6PD normal 166/2964; 5.6%) (p = 0.368). In the regression model the only strong significant predictor of H. pylori infection was age (OR = 1.81, 95% CI 1.572‒2.090).

Several loss‒of‒function mutations in the G6PD gene resulting in decreased erythrocyte enzymatic activity have been described. The distribution of the G6PD Mediterranean mutation in Sardinia varies according to historical and geographical pattern of malaria and altitude [25, 26] and the allele occurs in about 12% of the general population [13].

Since G6PD is a housekeeping enzyme, a number of studies focused on its role in different blood cells and tissues [27–29]. Cultured neutrophils, fibroblasts and lymphocytes displayed low levels of enzyme activity (8–15% of normal) and a marked reduction in the NADPH/NADP⁺ ratio [28, 29]. Several studies indicated that G6PD deficiency in leukocytes can result in chronic granulomatous disease with impaired host defense mechanisms against bacterial or fungal infection [30, 31]. For example G6PD-deficient epithelial cells in vitro showed a reduced tolerance to Staphylococcus aureus [18].

In one of the few studies in humans, Clark et al. reported a higher frequency of G6PD deficiency in patients hospitalized for infection in Iran significantly higher compared to non-infected control groups (p < 0.05) [32].

Previous studies reported a high prevalence of H. pylori infection in Sardinia [3, 33], making the island the ideal scenario to investigate the association between G6PD deficiency and H. pylori infection.

However, our study failed to detect a significant difference in the risk for acquiring H. pylori infection across subgroups of patients with normal or deficient G6PD enzyme activity. More specifically, no difference was observed among patients who acquired H. pylori infection at any time in their life (i.e. no matter if still active or past infection) and patients who never had contact with the pathogen, as demonstrated by the biopsy samples negative for the bacteria. The logistic regression model clearly showed that the major predictor of H. pylori infection was age, confirming the birth cohort effect [3]. The lack of association persisted after adjustment for age, and gender. Our results are consistent with the observation that G6PD-deficient mice are not more susceptible to infection than animals with normal enzyme activity [34]. It has been postulated that in cells other than erythrocytes, the intra‒mitochondrial NADPH generation may be sufficient to compensate the lack of its production in the pentose phosphate pathway making G6PD deficiency in these cells of poor clinical significance [35].

More interestingly, subgroups analysis showed that the prevalence of G6PD deficiency and G6PD normal activity was similar in patients with a past H. pylori infection indicating a similar behaviour for clearing infection.

Several limitations in our study should be taken into consideration. First, the patients' cohort investigated might pose some problems of sample representativeness as they were referred to a GI section for dyspeptic symptoms. However, the genetic and sociocultural homogeneity of the study population is great enough to minimize the effect of residual confounders. In addition, the chance to have gastric histological samples gave us the opportunity to accurately classify the type of H. pylori associated gastritis.

Conclusion

Our results demonstrate for the first time that G6PD deficiency does not enhance patients’ susceptibility to acquire H. pylori infection and does not affect its eradication in Sardinia.