Introduction

Interactions between etiological factors of gastric cancer (GC) induce alterations in the expression of proto-oncogenes and tumor suppressor genes [1]. In GC and in Helicobacter pylori infection, the expression of genes that regulate gastric homeostasis, such as Gastrokine 1 (GKN1) is altered [2]. GKN1 is a tumor suppressor that inhibits inflammation [2]. The transcript and protein of GKN1 decrease in the mucosa of patients with intestinal and diffuse-type GC [3,4,5] as well as in GC cell lines [6]. It is proposed that GKN1 may be a highly specific diagnostic biomarker for early GC detection [7].

Research on the transcriptome of various organisms indicates that the gene expression profile differs between sexes [8]. Sex is a factor that determines the incidence and mortality of individuals in different types of cancer [9]. In women with GC, serum levels of cathepsin B (CTSB), gastrokine 2 (GKN2), gastric lipase (LIPF), endothelial lipase (LIPFG), zinc alpha glycoprotein 2 (AZGP1), DNA topoisomerase 2-alpha (TOP2A) and pepsin A (PGA4) integrate a profile of promising predictive markers [10]. Currently, it is unknown if GKN1 expression varies with the sex of patients with chronic gastritis or GC. In this study, we evaluated GKN1 expression in biopsies of patients with chronic gastritis and with H. pylori infection and analyzed the relationship between the level of GKN1 and the sex of the patients.

Material and methods

Patients

Thirty-six biopsies of patients with histopathological diagnosis of chronic gastritis were studied. The patients underwent endoscopy at the Instituto Estatal de Cancerología “Dr. Arturo Beltrán Ortega”, Acapulco, Guerrero, Mexico. The patients or their relatives signed a letter of consent. This project was approved by the bioethics committee of the Universidad Autónoma de Guerrero and by the research department of the Instituto Estatal de Cancerología “Dr. Arturo Beltrán Ortega”.

Detection of H. pylori

Helicobacter pylori DNA was detected using the oligonucleotides HP16SF and HPGR16SR, which amplify a 522 bp fragment of the 16S rRNA gene, following the methodology described by Román-Román et al. [11].

GKN1 expression

RNA and total proteins were obtained from the biopsies with the TRIzol reagent (Invitrogen, Carlsbad, CA, USA), following the manufacturer's instructions. GKN1 mRNA expression (assay no. Hs00219734) was determined using the TaqMan RNA-to-CT 1 step kit (Thermo Fisher Scientific, Waltham, MA, USA). The relative expression of GKN1 was calculated by the 2−Δct method. GAPDH (assay no. Hs99999905_m1) was used as endogenous control. GKN1 protein was detected by western blot with an anti-GKN1 monoclonal antibody at a 1:15,000 dilution (Abnova, Walnut, CA, USA). GAPDH was used as a loading control, and was detected with a specific monoclonal antibody (Santa Cruz, Dallas, TX, USA) at a 1:5000 dilution. Immunocomplexes were revealed by chemiluminescence using Immobilon (Millipore, Burlington, MA, USA) in the ChemiDoc Digital Imaging System (Bio-Rad, Hercules, CA, USA). GKN1 expression was calculated by densitometric analysis of the bands using the Image J program.

Bioinformatic and statistical analyses

Statistical analysis of the data was performed using the GraphPad Prism v5.0 (GraphPad Software, San Diego, CA, USA) and STATA v14.0 (StataCorp, College Station, TX, USA) software. Absolute and relative frequencies of the qualitative variables were obtained, and the differences were calculated by the X2 test. Means ± standard deviations (SD) or medians were calculated for quantitative variables. The Student’s t test or the Mann–Whitney U test were used to compare the differences between the groups, respectively. A p value of < 0.05 was considered statistically significant. The bioinformatic analysis tosearch the potential binding sites of the estrogen and androgen receptors within the GKN1 promoter was done using the ExPASy program [12]. The binding sites ofthe estrogen and androgen receptors to the GKN1 promoter were selected considering a p value of < 0.001.

Results

Thirty-six patients with chronic gastritis were studied. The average age was 55 years (range 16–84). 47.2% (17/36) of patients were H. pylori-positive (Table 1). Based on the Sydney System, 75% (24/32) of the patients had moderate chronic gastritis and of these 84.6% were women and 68.4% were men.

Table 1 Clinical features of chronic gastritis patients stratified by sex
Full size table

Chronic chemical gastritis was present in 44.4% (16/36) of the patients whereas 55.6% (20/36) had chronic follicular gastritis. GKN1 mRNA expression was significantly lower in patients with chronic follicular gastritis than in those with chronic chemical gastritis (p = 0.00071, Fig. 1a). Likewise, GKN1 protein expression was lower in patients with chronic follicular gastritis (Fig. 2a, b), however this difference was not significant.

Fig. 1
figure 1

GKN1 mRNA expression a by the type of gastritis; b in chronic chemical gastritis (CCG) and chronic follicular gastritis (CFG) grouped by sex; c with and without H. pylori in all cases; d in H. pylori-positive chronic follicular gastritis grouped by sex. The horizontal line represents the median of GKN1 expression (*p < 0.05, **p < 0.001)

Full size image
Fig. 2
figure 2

Western blot of GKN1 a, b by the type of gastritis; c, d in chronic chemical gastritis (CCG) and chronic follicular gastritis (CFG) grouped by sex; e, f with and without H. pylori in all cases; g, h in H. pylori-positive follicular gastritis cases grouped by sex. The GKN1 level was normalized with GAPDH

Full size image

The levels of mRNA and protein of GKN1 were different between the types of gastritis and regarding the sex of the patients. GKN1 expression was significantly lower in women with chronic follicular gastritis compared tomales with the same condition (p = 0.0279, Fig. 1b; p = 0.0014, Fig. 2c, d). Irrespective of the sex, lower levels of GKN1 were detected in H. pylori-positive patients (p = 0.0275, Fig. 1c; p = 0.0112, Fig. 2e, f). Women with H. pylori-positive follicular gastritis expressed the lowest levels (p = 0.0175, Fig. 1d; p = 0.0111, Fig. 2g, h).

By means of a bioinformatic analysis, it was found that the GKN1 promoter contains four estrogen response elements (ESR1, ESR2, ESRRα and ESRRβ) and an androgen response element (AR), (Fig. 3).

Fig. 3
figure 3

Estrogen and androgen response elements in the GKN1 promoter. The colored boxes show the estrogen or androgen binding sites in the GKN1 promoter from −1 to −1000 bp. TSS transcription start site, EREs estrogen response elements, AREs Androgen Response Elements, ESR1 estrogen receptor 1, ESR2 estrogen receptor 2, ESRRα estrogen-related receptor alpha, ESRRβ estrogen-related receptor beta, AR androgen receptor

Full size image

Discussion

GKN1 decreases in mucosa infected with H. pylori and is absent in uninfected gastric cancer [5, 13,14,15]. In this study, the expression of GKN1 was evaluated in men and women with chronic gastritis with and without H. pylori infection. In patients with chronic follicular gastritis, the expression of GKN1 mRNA was significantly lower than in patients with chronic chemical gastritis (p < 0.05). Patients with H. pylori-positive chronic gastritis express significantly lower levels of the GKN1 transcript than the H. pylori-negative cases (p < 0.05). These data are consistent with those reported by Nardone et al., and Mao et al. [3, 6], who observed that in H. pylori-positive gastric mucosa the expression of GKN1 is decreased. The low expression of GKN1 may be due to H. pylori promoting the expression of AUF1, a protein that induces degradation of the GKN1 mRNA [16]. On the other hand, there may be a negative regulation of the GKN1 transcript translation by non-coding RNAs [17] or degradation of the protein by ubiquitination [18]. These and other mechanisms can act synergistically and cause decrease or silencing of GKN1 expression in the gastric tissue [19].

The mRNA and protein levels of GKN1 were lower in women with chronic follicular gastritis than in men with the same condition (p = 0.0279 and p = 0.0014, respectively). Women with H. pylori-positive chronic follicular gastritis expressed significantly lower levels of transcript and protein than infected men (p = 0.0175 and p = 0.0111, respectively; Figs. 1d, 2d, h). These findings suggest that, in women, in addition to H. pylori status and the type of gastritis, other factors modify the expression of GKN1. Taking the above point into account, it is likely that female hormones are involved in the regulation of GKN1 expression. This hypothesis is strengthened by the results of the bioinformatic analysis that in the GKN1 promoter, four estrogen response elements and an androgen response element are located (Fig. 3). The influence of sex hormones on the expression of GKN1 has not been described, however, it has been proposed that estrogens may have a protective effect on gastric cancer [20,21,22]. In breast cancer, estrogen was reported to positively regulate the expression of Trefoil 1 protein (TFF1) [23], a protein that contributes to the protection of gastric mucosa [24]. GKN1 expression in women could be influenced by hormonal status. Considering that the age of the women studied varied from 16 to 84 years and that menopause can start at 45, the fluctuation in estrogen production may explain, at least partially, the variation in GKN1 levels. In menopausal women (> 45 years) the minimum estrogen production could be related to the lower levels of GKN1 found. To strengthen these results, it is necessary to increase the sample size and conduct further studies on the differences in the expression of GKN1 in relation to the sex and age of the patients. It is imperative to elucidate the role of estrogens in GKN1 expression regulation.