Background
Helicobacter pylori is a Gram-negative, spiral-shaped bacterium that colonizes the gastric mucosa in humans. In many cases, it can persist without causing symptoms, although in some cases it can produce chronic gastritis. The persistent colonization of the gastric mucosa by
H. pylori can lead to the development of gastroduodenal diseases like peptic ulcer, gastric lymphoma, and gastric cancer [
1].
H. pylori has two main virulence factors, the
cag pathogenicity island (
cagPAI) and the pore forming toxin
vacA, whose presence or absence identifies highly virulent (type-1) from less virulent (type-2) strains, respectively, thus being strong predictors of severe disease outcome [
2,
3]. The
cagPAI is a 40-kb chromosomal region that contains 27–31 genes that encode the type IV secretion system (T4SS) and an effector protein [
4]. The T4SS forms a needle-like surface appendage called the T4SS pilus, which is induced upon contact with the host cell membrane [
5]. The effector protein CagA is encoded by the
cagA gene. This protein is translocated into the cytoplasm of the gastric epithelial cells through the T4SS, where it is phosphorylated at the tyrosine residues of the EPIYA motifs, causing multiple cellular alterations [
6,
7]. One of the genes that encode the T4SS is
cagL, which encodes for the 26-kDa protein CagL. It has been shown that this protein is localized on the surface of the T4SS of
H. pylori and in intracellular pools, is necessary for CagA translocation, and interacts with CagI [
5,
8]. It also contributes to transient hypochlorhydria by disrupting the interaction between the integrin and metalloprotease ADAM17 and the integrin α
5β
1, thus activating NF-κB-mediated repression of the gastric H, K-adenosine triphosphatase α-subunit (HKα) [
9].
The role of some regions or motifs of CagL in different phenotypes has been assessed. For example, its C-terminal coiled-coil region is involved in IL-8 secretion, host cell elongation, and binding of the T4SS to the host cell [
10]. CagL also contains an arginine–glycine–aspartate (RGD) motif at residues 76–78, located within the second large α-helix [
11], which is essential for binding to the human integrin α5β1 receptor [
5], as well as integrins α
vβ
5 and α
Vβ
3 [
12,
13], although it can also bind to human fibronectin in a RGD independent manner [
14]. This motif is also responsible of CagL fibronectin-like effect on host cells, since it is involved in cell spreading triggering, formation of focal adhesions, and activation of several tyrosine kinases [
15]. The RGD motif is right next to the sequence LXXL, of which, the two-leucine residues (L79 and L82) contribute to the adhesion to different cell lines through interaction with integrin α
Vβ
6 [
16]. CagL also contains a second motif named RHS (RGD helper sequence), which contains residues phenylalanine–glutamic acid–alanine–asparagine–glutamic acid (FEANE). This motif also contributes to CagL binding to integrins [
17]. The third important region in CagL is the CagL hypervariable motif (CagLHM), which spans residues 58 to 62 [
18]. It is located in an unresolved flexible region between helices α1 and α2 [
19]. Although its function is still controversial, it has been shown that certain polymorphisms correlate with diseases in a geographical-dependent manner. For example, polymorphism Y58/E59 is associated with gastric cancer in patients from Taiwan, while in India the polymorphism associated with this disease is D58/K59 [
20,
21]. In Mexico, only the polymorphism N58 has been associated with a higher risk of gastric cancer [
22]. So far, 33 combinations of polymorphisms in the hypervariable motif have been identified worldwide, the most common being DKMGE, NEIGQ, NKIGQ, and DKIGK. Of these, all are found in North and South American strains, except for DKIGK, which is more prevalent in East/Southeast Asia/Australasia [
23]. The aim of this work was to analyze the sequence of the hypervariable motif (residues 58 to 62) of clinical isolates from Mexican patients with chronic gastritis, and to correlate these polymorphisms with the
vacA genotype of
H. pylori.
Discussion
Gastric diseases are of diverse etiology, and although chronic gastritis and more severe gastroduodenal disorders are attributed to
H. pylori, Epstein–Barr virus and human Cytomegalovirus can also cause these diseases [
25]. In this study we found that the prevalence of
H. pylori, determined by bacterial isolation, was 30.5%. This percentage is similar to that reported by Román-Román et al. (47.8%) [
26], but lower than that reported by other authors [Paniagua et al. (60.1%); Martínez-Carrillo et al. (77%)] [
27,
28]. This variation in frequencies may be due to bias in the information provided by the patients; also, it is impossible to rule out that they have received antimicrobial treatment against
H. pylori or any other infection in the recent past. Among the Mexican population self-medication is common and, although the sale of antibiotics without medical prescription is prohibited, in the state of Guerrero the metronidazole, an antiparasitic that is part of the treatment scheme for
H. pylori, is available over the counter. Additionally, the probability of finding
H. pylori in biopsies decreases as structural and functional changes appear in the infected mucosa. Another factor that may have contributed to the low frequency found in this work, is that the prevalence of
H. pylori infection was determined by culture, which is a very specific method (95–100%) but not very sensitive (80–90%), since the results depend on the number of viable bacilli in the biopsy [
29].
Among infected patients, the clinical outcome is determined by the genetic background and lifestyle of the host, as well as by
H. pylori virulence factors. It has been proposed that the
H. pylori virulence factor CagL has an important role in the pathogenesis of gastroduodenal diseases, and that its function can be determined by the amino acid sequence in specific positions [
11,
12]. We analyzed the prevalence of
cagL in
H. pylori clinical isolates from Mexican patients with chronic gastritis, which was 72%. This is a lower percentage than those found in Asian patients, > 85% in Chinese, Indian, Iranian, Taiwanese and Malay patients [
20,
21,
30,
31]; however, these patients had more severe diseases, like gastric cancer, peptic ulcer disease, and duodenal ulcer. Nevertheless, our result is higher to that found in Iranian patients with gastritis (64.2%) [
32]. Apparently, the prevalence of
cagL in clinical isolates from patients with gastritis is lower than that found in isolates from patients with peptic ulcer disease or gastric cancer. All our
cagA+ isolates were
cagL+, which is in accordance with previous reports that have shown that the prevalence of
cagL is higher than, or very close to that of
cagA [
20,
30,
33].
We also analyzed the polymorphic region (residues 58 to 62) of CagL. We found that the most common polymorphism in position 58 corresponds to aspartic acid (D) (84.8%), followed by asparagine (N) (15.2%). It has been proposed that strains carrying aspartic acid at this position lead to a lower risk of gastric cancer in comparison with the asparagine carrying strains. In agreement with this, the polymorphism N58 was significantly associated with gastric cancer in Mexican patients from Mexico City [
22,
32]. Regarding residue combinations in positions 58 and 59, polymorphism D58/K59 has been associated with gastric cancer in India, where patients carrying bacteria with this polymorphism are 3.8 times more likely to develop this disease [
20]. However, our results showed that D58/K59 was the most common polymorphism in Mexican patients with chronic gastritis and it is associated with the
vacA genotype s1/m1. We also found a high frequency of the polymorphisms M60 (85%), which has been previously shown to increase the chance of developing gastritis [
32], and E62 (81.8%), which has been shown to be the most common polymorphism in strains from Iranian patients with chronic gastritis [
32]. It is worth noting that in this region the only invariable residue is G61.
Besides the polymorphisms in the polymorphic region, Cherati et al. [
32], reported other polymorphisms outside this region associated with different disease outcomes. They found that the polymorphism K122 had frequency of 88.8% in strains from patients with gastritis. Similarly, we found that the frequency of K122 in our isolates was 82%. In fact, 24/33 strains (73%) had the combination M60/K122. They also found that the frequency of polymorphism V134 is significantly higher in patients with gastric cancer than gastritis. In our strains, V134 had a prevalence of only 18.2%, being the most common the polymorphism I134 (81.8%). This result shows that in Mexican patients, polymorphism I134 is more common in chronic gastritis. We also identified two more residues with high variability, I73 and T175 (in
H. pylori ATCC 26695). In strains isolated from Mexican patients with chronic gastritis, polymorphisms M73 and I175 have a frequency of 60.6% and 42.4%, respectively, however, in the literature there is no information regarding these residues or their association with clinical outcome, hence, the clinical significance of these results as well as the biological function of these polymorphisms need to be addressed in more detail. With respect to the I73 M mutation, isoleucine (I) and methionine (M) are nonpolar amino acids without charge, therefore, it is likely that this change has no important effects on the structure and function of CagL. The I175 mutation consists of the change of a polar threonine residue by one of nonpolar isoleucine, next to the TASLI motif and it is probable that this change influences the functions of CagL. It has been proposed that the TASLI region binds to integrin in an RGD-independent manner and it is probably that I175 residue contributes also to this function and to other functions of CagL. The TASLI motif is involved in binding to the cell, although only in the absence of another host cell ligands. Deletion of TASLI is related with a moderate reduction in IL-8 secretion and CagA translocation. In addition, it has been shown that the deletion of TASLI reduces de CagL binding to integrin [
34].
Interestingly, the most common CagLHM sequence found in our study population was DKMGE, which is the most frequent worldwide, mainly in Africa and the American Continent. The second most frequent sequences were NEIGQ and NKMGQ. NEIGQ is also the second most common worldwide mainly in Europe, Asia, and North America, while NKMGQ has been found in Asia/Australasia but not in the American Continent. Additionally, we also found three sequences that are prevalent in East/Southeast Asia/Australasia (DKIGK), West/Central/South Asia (DEIGQ), and Central/South America (NKMGE) [
23], as well as a new sequence not informed before (DKIGE). This sequence was found in only 1 clinical isolate (HG-193), and contains the sequence DKI, which is found mainly in strains from Asia/Australasia [
23,
24,
35]. Despite of this, this isolate seems to be more related with those carrying sequence DKMGE, and in a lesser degree to strains carrying sequence NKMGQ (Fig.
5). These results show that the CagLHM sequences found in Mexican strains are diverse, with sequences not only common in our continent, but also in Asia, although all the CagA sequences are Western according to the EPIYA motifs [
36].
The sequence diversity of CagL in the locally circulating strains may reflect the mobility and complex human interactions with the bacteria, but it may also be related to differences that modify the pathogenic function of the protein. More studies are needed on the diversity of CagL and the relationship of its polymorphisms with gastric diseases, as well as on the role of variations in the CagL hypervariable motif on the structure and function of the protein, and on the inflammatory process.
In the population of the state of Guerrero, Mexico, a high frequency of gastric diseases related to H. pylori persists, therefore, it is important to identify the virulence factors that confer a higher risk of developing serious diseases. The knowledge about these virulence factors will allow a better understanding of this process, facilitating the identification and characterization of biomarkers helpful in the detection of patients with greater risk of developing gastric cancer. It will also help to determine schemes and priorities of eradication treatments with anti–microbial compounds, as a measure to prevent gastric cancer.
Methods
Patients
We performed a cross-sectional study among 164 patients that attended to the Gastroenterology Service at the General Hospital Dr. Raymundo Abarca Alarcón, and to the Specialized Unit in Gastroenterology Endoscopy, both in Chilpancingo, Guerrero, Mexico. Patients who attended for an endoscopic study due to dyspepsia symptoms, and that have had no H. pylori eradication treatment during 1 month prior to the endoscopic procedure were selected. None of the patients included in this study were under treatment with proton pump inhibitors or with gastric pH neutralizing agents within 15 days prior to biopsy. Patients receiving non-steroidal anti-inflammatory therapy were excluded from the study. All patients signed a letter of consent. This project was approved by the Bioethics Committee of the Autonomous University of Guerrero, by the Department of Education and Research of the General Hospital Dr. Raymundo Abarca Alarcón, and by the authorized personnel of the Specialized Unit in Gastroenterology Endoscopy.
Biopsies
The endoscopic study was performed after a fasting night with a video processor and video gastroscope (Fujinon, Wayne, NJ USA). Two biopsies were taken from the antrum of patients with chronic gastritis, one was immediately fixed in 10% formalin for histological examination, and the other one was placed in Brain Heart Infusion Broth (BHI) (Becton–Dickinson, North Carolina, USA) with 10% glycerol for the isolation of H. pylori. The biopsies were transported at 4 °C, and those intended for isolation of H. pylori were processed immediately.
Isolation and identification of H. pylori
Each biopsy transported in BHI broth with 10% glycerol was macerated with a sterile wood applicator. Fifty microliters of the homogenates were cultivated on Columbia Agar plates (Becton–Dickinson, North Carolina, USA) supplemented with 10% ram blood, IsoVitaleX Enrichment and Helicobacter pylori selective supplement Dent (10 mg/L of vancomycin, 5 mg/L of trimethoprim, 5 mg/L of cefsulodin, 5 mg/L of amphotericin B) (Oxoid, Basingstoke, UK) at pH 6.8 to 7.0. The homogenates were distributed on the culture medium by isolation strip. The inoculated plates were incubated under microaerophilic conditions with 5% O2 and 5% CO2 at 37 °C in GasPak jars for 3–7 days. H. pylori was identified by colony morphology (small, transparent colonies, 1 mm in diameter), Gram staining and biochemical tests (urease, catalase and oxidase positive). H. pylori strain ATCC 43504 was used as positive control.
Isolates identified as
H. pylori were subcultured and incubated for 72 h. A pool of colonies from each isolate was resuspended in extraction solution (10 mM Tris pH 8, 10 mM EDTA, 0.5% SDS) for digestion with proteinase K. Total DNA was obtained by phenol: chloroform: isoamylic alcohol technique [
37]. Total DNA concentration was determined using a NanoDrop 2000. DNA samples were stored at − 20 °C until use.
Molecular confirmation and genotyping of vacA and cagA of H. pylori strains
Confirmation of
H. pylori strains was done using oligonucleotides HP16SF and HPGR16SR (Table
4), which amplify a fragment of the 16S rRNA gene, according to the method described by Román-Román et al. [
38].
vacA and
cagA genotyping was assessed by multiple PCR with specific oligonucleotides VAIF and VAIR, VAGF and VAGR, and F1 and B1, respectively (Table
4). The reaction mixture contained 1.5 mM MgCl
2; 0.2 mM dNTPs; 2.5 pmol of oligonucleotides F1 and B1, or 5 pmol of VAGF and VAGR, or 2.5 pmol of VAIF and VAIR; 1.5 U of Taq DNA polymerase (Invitrogen, Carlsbad, CA, USA) and 200 ng of DNA, in a total volume of 25 μL. Amplification conditions were: 1 cycle at 94 °C for 10 min; 35 cycles at 94 °C for 1 min, 57 °C for 1 min, 72 °C for 1 min; and a final extension cycle at 72 °C for 10 min. The PCR products were subjected to 2.5% agarose gel electrophoresis, stained with ethidium bromide and visualized with ultraviolet light (UV). In each PCR, DNA from strain ATCC 43504 (
vacA s1m1/
cagA+) was used as positive control, and as negative control DNA was replaced with sterile deionized water. All reactions were done in a Mastercycler Ep gradient thermal cycler (Eppendorf, Hamburg, Germany).
Table 4
Oligonucleotides used in this work
16S rRNA
| HP16SF | GCTAAGAGATCAGCCTATGTCC | 522 | |
HPGR16SR | CAATCAGCGTCAGTAATGTTC |
vacAs1
| VAIF | ATGGAAATACAACAAACACAC | 259 | |
vacAs2
| VAIR | CTGCTTGAATGCGCCAAAC | 286 |
vacAm1
| VAGF | CAATCTGTCCAATCAAGCGAG | 570 | |
vacAm2
| VAGR | GCGTCTAAATAATTCCAAGG | 645 |
cagA
| F1 | GATAACAGGCAAGCTTTTGAGG | 349 | |
B1 | CTGCAAAAGATTGTTTGGCAGA |
Empty site | ESf | ACATTTTGGCTAAATAAACGCTG | 360 | |
ESr | TCATGCGAGCGGCGATGTG |
cagL
| cagL sense | GAAGATATAACAAGCGGTTT | 651 | |
cagL antisense | TTTAACAATGATCTTACTTGA |
cagL-Fwd-2 | ACVAAGAGACCAACCARCAAG | 165 | This work |
cagL-16 | TCGCTTCAAAATTGGCTTTC | |
| cagL-Fwd-2 | ACVAAGAGACCAACCARCAAG | 611 | This work |
| cagL-16 | TTTAACAATGATCTTACTTGA | | |
Confirmation of cagPAI empty site
The absence of
cagA and the pathogenicity island
cagPAI in the
cagA− strains was confirmed by the empty-site assay by conventional PCR, using the ESf and ESr oligonucleotides (Table
4), which bind upstream and downstream, respectively, of the region where the
cagPAI is inserted in the genome of the reference strain NCTC 12455 (NCTC: National Collection of Type Culture) [
45]. The PCR mixture contained 50 ng of DNA, 0.08 mM dNTPs (Invitrogen, Carlsbad, CA, USA), 1 mM MgCl
2, 5 pmol of each oligonucleotide and 1 U of Platinium Taq DNA Polymerase (Invitrogen, Carlsbad, USA), in a final volume of 15 μL. The amplification conditions were: 1 cycle at 94 °C for 5 min; 35 cycles at 94 °C for 30 s, 61 °C for 30 s and 72 °C for 45 s; and one final extension cycle at 72 °C for 7 min. PCR products were subjected to 2% agarose gel electrophoresis, followed by ethidium bromide staining and UV light observation. As negative control, DNA was replaced with sterile deionized water. DNA from strain ATCC 43504 (
cagA+,
cagPAI
+) was used as a second negative control, and strain UEGE-644 (
cagA−,
cagPAI
−) as positive control. The presence of a 360 bp product was considered indicative of the absence of
cagA and
cagPAI [
43,
45].
CagL amplification by PCR
A fragment of 651 bp from gene
cagL (
hp0539) was amplified by PCR using primers cagL sense and cagL antisense (Table
4). The reaction mixture had a final volume of 15 μL, and contained 2.5 mM MgCl
2, 0.25 mM dNTP’s, 5 pmol of each oligonucleotide, 1 U of Taq recombinant DNA polymerase (Invitrogen, Massachusetts, USA) and 50 ng of DNA. The conditions used were: 1 cycle at 94 °C for 5 min; 45 cycles at 94 °C for 30 s, 55 °C for 30 s, and 72 °C for 1 min; and 1 cycle at 72 °C for 7 min. Each reaction included a positive (DNA from strain 26695) and a negative (DNA was substituted with water) control. All reactions were performed in a Mastercycler Ep gradient thermocycler (Eppendorf, Hamburg, Germany). PCR products were analyzed by agarose gel electrophoresis at 1.5% and stained with ethidium bromide. A second PCR was performed with DNA from those strains that were negative in the first reaction, using primers cagL-Fwd-2 and cagL-16, which amplified a 165 bp product (Table
4). The reaction mixture had a final volume of 15 μL, and contained 1.5 mM MgCl
2, 0.25 mM dNTP’s, 5 pmol of each oligonucleotide, 1 U of Taq recombinant DNA polymerase (Invitrogen, Massachusetts, USA) and 50 ng of DNA. The conditions used were: 1 cycle at 94 °C for 5 min; 35 cycles at 94 °C for 45 s, 54 °C for 30 s, and 72 °C for 1 min; and 1 cycle at 72 °C for 7 min. Each reaction included a positive (DNA from strain 26695) and a negative (DNA was substituted with water) control. PCR products were analyzed by agarose gel electrophoresis at 1.8% and stained with ethidium bromide. A third PCR was performed with DNA from those strains that were negative in the first reaction, using primers cagL-Fwd-2 and cagL antisense (Table
4), in order to amplify a fragment of 611 bp suitable to upload in the GeneBank. The reaction mixture had a final volume of 25 μL and contained 1.5 mM MgCl
2, 0.25 mM dNTP’s, 5 pmol of each oligonucleotide, 1 U of Taq recombinant DNA polymerase (Invitrogen, Massachusetts, USA) and 50 ng of DNA. The conditions used were: 1 cycle at 94 °C for 5 min; 35 cycles at 94 °C for 45 s, 56 °C for 30 s, and 72 °C for 45 s; and 1 cycle at 72 °C for 3 min. PCR products were analyzed by agarose gel electrophoresis at 1.8% and stained with ethidium bromide.
Purification and sequencing of PCR products
PCR products were purified by the isopropanol method and sequenced in the Unidad de Síntesis y Secuenciación of the Instituto de Biotecnología, UNAM, using primers cagL sense and primer cagL-Fwd-2 (Table
4). All sequences were deposited in GenBank with accession numbers MG051618-MG051641 and MG214979-MG214987.
Nucleotide sequences were translated to aminoacidic sequences using the ExPASy Translate Tool (
http://web.expasy.org/translate/). Multiple sequence alignment of translated sequences and phylogenetic analysis were performed using Molecular Evolutionary Genetics Analysis version 7.0 (MEGA7) software [
46].
The evolutionary history was inferred using the Neighbor-Joining method [
47]. The tree was drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The analysis involved 34 amino acid sequences of CagL.
Authors’ contributions
Conception and design: GF-T, AR-R. Experiments: AJA-M, PG-M, CAC-S, DGS-F. Data analysis: VIM-S, GF-T, AR-R, DNM-C. Drafting the manuscript: VIM-S, GF-T, AR-R, DNM-C. Sample collection: AJA-M, PG-M. All authors read and approved the final manuscript.