Medium and large alleles of the PGC gene are risk factors for gastric cancer

1 Introduction

Pepsinogen C or Pepsinogen II (PGC or PG-II) is a 42-kDa protein that belongs to the aspartic protease family and is synthesized by chief cells of the gastric mucosa and secreted to the stomach under acidic conditions. PGC is activated to pepsin C to digest polypeptides and amino acids. This protein is encoded by the progastricsin (PGC) gene, which is located on chromosome 6p21.1 and consists of 9 exons [1].

In situ expression of PGC occurs mainly in the normal gastric mucosa [2]. PGC expression has been detected in different gastric diseases by immunohistochemistry; positive PGC expression (100%) is observed in normal gastric mucosa, whereas its positivity declines considerably in GC [3].

Several variants in the PGC gene can influence PGC expression and regulation, for instance, rs6912200, rs6941539, rs9471643, and rs6458238 [4–6]; however, the most studied polymorphism is the 100-bp insertion/deletion located between exons 7 and 8 of the PGC gene. Based on fragment lengths, following agarose gel electrophoresis of PCR products, four alleles with different molecular weights were initially obtained: 310 bp, 400 bp, 450 bp and 480 bp [7].

A 100-bp insertion/deletion polymorphism is associated with the risk of peptic ulcer and GC, and individuals homozygous for the 310-bp allele have downregulated PGC expression [8, 9]; more recently, Kumar used similar methods and further concluded that being homozygous for the 310-bp allele of PGC may lead to elevated serum PGC levels in cases of GC, especially in patients with Helicobacter pylori infection and intestinal metaplasia [10]. Moreover, Pinto-Correia reported that being homozygous for the 310-bp allele was related to the upregulation of PGC expression, which served as a protective factor in the development of gastric disease [11].

In this work, we analyzed the 100-bp insertion/deletion polymorphism in subjects with GC, atrophic gastritis and intestinal metaplasia, controls, and the Mexican general population by means of PCR-capillary electrophoresis and GeneScan, which represents an easy, quick and inexpensive method for more precisely identifying the size of alleles for this polymorphism.

2 Methods

We studied the genomic DNA of Mexican subjects with gastric cancer (GC group, n = 80), atrophic gastritis and intestinal metaplasia (AG/IM group, n = 60), and no atrophic gastritis (control group NAG, n = 110) and the Mexican general population (MGP group, n = 97).

Diagnoses of GC, AG, IM and NAG were made by a pathologist through analysis of gastric biopsies obtained by endoscopy. NAG (controls) were subjects with chronic gastritis without athrophy, none of them had healthy gastric mucosa. MGP individuals were healthy donors and they did not undergo upper endoscopy (this group represent the western region of Mexico). All subjects were recruited from the gastroenterology department from two hospitals, the Pediatric Hospital and Specialty Hospital, as well blood bank, from the Western National Medical Center, Mexican Institute of Social Security, in the city of Guadalajara, Jalisco, Mexico.

The study was approved by the National Committee of Health Research, IMSS (CNIC-2017-785-087). All subjects provided written informed consent. The study conforms with the Code of Ethics of the World Medical Association (Declaration of Helsinki), printed in the British Medical Journal (18 July 1964).

DNA was extracted from leucocytes of peripheral blood by the Miller method [12]. The PGC gene insertion/deletion polymorphism was identified by means of PCR and capillary electrophoresis using the following primers: forward 5’-FAM/GGCCAGATCTGCGTGTTTTA-3’ and reverse 5’-AGCCCTAAGCCTGTTTTTGG-3’ [13]. PCR was performed in a 15-μL reaction mixture comprising 200 ng genomic DNA, each primer at 5 pM, 0.5 U Taq polymerase (Invitrogen, USA), 1X PCR buffer, 1.5 mM MgCl2, 2.0 mM deoxynucleotide triphosphate mix (dNTP Set; Vivantis, Malaysia), and 5% DMSO. PCR conditions were 94°C for 5 min, followed by 28 cycles of 94°C for 30 s, 57°C for 30 s, and 72°C for 30 s and a final step at 72°C for 5 min; PCR was carried out in a 2720 Thermal Cycler (Applied Biosystems, USA). PCR fluorescent products were resolved by capillary electrophoresis in a 310 Genetic Analyzer (Applied Biosystems, USA) with Hi-Di Formamide (Applied Biosystems, USA) and GeneScan 600 LIZ Size Standard v2.0 (Applied Biosystems, USA); the fragments were analyzed by GeneScan software (ThermoFisher Scientific).

3 Results

The median age and sex of the Mexican subjects here studied are shown in Table 1. We found differences in terms of sex between GC group and controls (p = 0.000), Odds Ratio analysis showed protection against GC for females with OR 0.320 (0.17–0.58), and an increased risk for GC for males with OR 3.125 (1.71–5.69). For AG/IM vs controls was not significant (p = NS). Also, we observed differences regarding age, between GC and controls (p = 0.000), as well AG/IM and controls (p = 0.001), adults > 60 years had an increased risk for GC with OR of 1.98 (1.06–3.68) p = 0.030 as well for AG/IM with OR of 2.03 (1.3–3.99) p = 0.038 (Table 1).

Different sizes of alleles of PGC polymorphism were observed in the studied groups, from 266 bp to 499 bp, which were grouped for the analysis as follows: short alleles of 266–399 bp, medium alleles of 400–433 bp and large alleles of 434–499 bp (Fig. 1). The control group had the highest diversity of alleles, with 34 different alleles (in a range of sizes from 267 to 470 bp), followed by the GC group with 28 alleles (305–499 bp), the AG/IM group with 24 alleles (305–471 bp) and the MGP group with 22 alleles (266–496 bp).

OPEN IN VIEWER

Fig. 1

Structure of human Progastricsin (PGC) gene. The 100-bp insertion/deletion polymorphism located between exons 7 and 8 is shown, different size of alleles/repeats in TATA box are grouped as short, medium and large. [Modified from 13].

The three most frequent alleles observed in each group were as follows: in the GC group, 305 bp 29.4%, 431 bp 28.1%, and 406 bp 5.0%; in the AG/IM group, 305 bp 40.0%, 432 bp 15.0%, and 431 bp 11.7%; in controls, 305 bp 42.3%, 431 bp 12.3%, and 430 bp 5.5%; and in the MGP, 433 bp 32.0%, 305 bp 30.9%, and 411 bp 4.6%.

Short alleles were more frequently observed in the control group, medium alleles were more frequent in the GC group, and large alleles were more frequent in the MGP group. Significant differences were observed between the GC and control groups (p = 0.021) as well as the MGP and control groups (p = 0.0055) (Table 1).

The odds ratio analysis performed between the GC group and the control group showed that short alleles had an OR of 0.55 (CI95% 0.36–0.84, p = 0.006), meaning that short alleles (266–399 bp) confer protection against GC; conversely, the medium and large alleles showed an OR of 1.81 (CI95% 1.18–2.77, p = 0.006), thus conferring an increased risk for GC. With Cochrane-Armitage trend test [14], we found that under the recessive inheritance model, carriers of two short alleles (homozygotes) had protection against GC, with an OR of 0.42 (CI95% 0.22–0.812, p = 0.009), compared to those with only one or no short alleles. Under the dominant model, carriers of one or two medium alleles, had an increased risk of GC, with OR of 1.99 (CI95% 1.08–3.67) p = 0.026 compared to homozygotes (no medium/no medium) (Table 2).

The PGC gene coding for the pepsinogen II protein consists of nine exons, and different lengths of the TATA box are represented by short (308–378 bp), medium (400–413 bp) and large (434–479 bp) repeats, which are located between exons 7 and 8 [13]. This polymorphic length of the TATA box is better known as the 100-bp insertion/deletion polymorphism and has been related to peptic ulcers and gastric cancer. In this work, we analyzed the relationships of the 100-bp insertion/deletion polymorphism with GC, AG, and IM in the Mexican population. We found different allele sizes of PGC polymorphism which were grouped as short alleles of 266–399 bp, medium alleles of 400–433 bp and large alleles of 434-499 bp. Carriers of one or two medium alleles, had an increased risk of GC, with OR of 1.99 (CI95% 1.08–3.67 p = 0.026). To our knowledge, up to now there is no consensus on the cut-off point for the size of the alleles of the PGC gene, only one study has reported the cut off for size of alleles in PGC gene, group 1 or short 308–378 bp, group 2 or medium 400–413 bp, and group 3 or large 434–479 bp [13], therefore we adapted a classification system for PGC alleles very similar to that of those authors, though we found more diversity in the size of PGC alleles than these authors.

Some authors have reported that individuals homozygous for the 310-bp allele have downregulated PGC expression, and this status has been related to gastric cancer in China [8]. More recently, other authors concluded that being homozygous for the 310-bp allele led to elevated serum PGC levels and was associated with gastric cancer in Indian patients [10]. Pinto-Correia et al. 2006 hypothesized that the 100-bp insertion/deletion polymorphism could interfere directly with the number of TATA-box repeats accessible for the activation of PGC gene expression. A large number of TATA-boxes together in one sequence might function as confounders for transcriptional activation factors. Thus, short alleles would stabilize the activation of gene expression, increasing its levels [11].

Discordance is observed among results for the frequency 310-bp alleles between Caucasian and Asian populations, conferring protection or risk, respectively, which can be explained by the distinct distributions of frequencies of these alleles between these populations. Specifically, in Caucasian populations, the frequency of the 310-bp allele is lower in gastric cancer patients (0.509 compared to 0.724 in controls), and this allele can even serve as a protective factor against the development of gastric cancer in the Portuguese population (OR 0.39, CI95% 0.21–0.75, p = 0.004) [11]. In contrast, in Asian populations, the frequency of the 310-bp allele is higher in gastric cancer cases than in controls (0.516 vs. 0.357), while homozygotes for the 310-bp allele have an increased risk for gastric cancer with an OR of 2.93 (CI95% 1.08–7.93, p = 0.033) [8].

In our population, short alleles (266–399 bp) conferred protection against gastric cancer (OR 0.55, CI95% 0.36–0.84, p = 0.006), which is in concordance with the results of Pinto-Correia et al. 2006 [11], and also we observed in our study that carriers of one or two medium alleles had an increased risk of gastric cancer with an OR of 1.99 (CI95% 1.08–3.67, p = 0.026). As the methodology used here (capillary electrophoresis) is more precise than agarose or polyacrylamide electrophoresis, we think that the 310-bp allele is represented in our study by the 305-bp allele, which had a lower frequency in the gastric cancer group than in the controls (29.4% vs. 42.3%), as was observed by Pinto-Correia et al. in the Caucasian population [11]. It should be noted that the genetic ancestry of the Mexican Mestizo population is 56% Native American, 37% Caucasian and 5% African [15], which may explain the similarity between our results and those observed for the Caucasian population [11].

Although our results are interesting, the study has some limitations. It is important to note that the observed differences in sex and age between GC and controls, could be attributed to the fact that most of the cases of GC are men (65%) older than 50 years of age (74.7%), while the control group included mainly women (62.7%) under 60 years (72.7%). These differences can be attributed to some hormonal issue in which women have protection against gastric cancer while men do not, or to the fact that more women seek medical attention than men when they present gastric symptoms. The fact that there are differences in terms of age and sex between the GC and control groups suggests that the results could have a bias, which could be corrected by conducting the study with samples matched by age and sex; on the other hand, there is a lack of information about the relationships between PGC alleles and the expression or regulation of PGC levels, which would allow a better approach to the role of this variant in gastric cancer in our population.

5 Conclusions

Previous studies have related short alleles as a risk for or protection against gastric cancer. In our population, short alleles were related to protection against gastric cancer, but we also analyzed and emphasize the importance of medium alleles, which were related to risk for gastric cancer in our population. Further studies are required to establish the importance of this polymorphism in the origin of gastric neoplasia.