Association of Progranulin Gene Expression from Dyspeptic Patients with Virulent Helicobacter pylori Strains; In Vivo Model

(1) Background: Gastric cancer, the fourth most common cause of death from tumors in the world, is closely associated with Helicobacter pylori. Timely diagnosis, therefore, is essential to achieve a higher survival rate. In Chile, deaths from gastric cancer are high, mainly due to late diagnosis. Progranulin has reflected the evolution of some cancers, but has been poorly studied in gastric lesions. Aiming to understand the role of progranulin in H. pylori infection and its evolution in development of gastric lesions, we evaluated the genic expression of progranulin in gastric tissue from infected and non-infected patients, comparing it according to the epithelial status and virulence of H. pylori strains. (2) Methods: The genic expression of progranulin by q-PCR was quantified in gastric biopsies from Chilean dyspeptic patients (n = 75) and individuals who were uninfected (n = 75) by H. pylori, after receiving prior informed consent. Bacteria were grown on a medium Columbia agar with equine-blood 7%, antibiotics (Dent 2%, OxoidTM), in a microaerophilic environment, and genetically characterized for the ureC, vacA, cagA, and iceA genes by PCR. The status of the tissue was determined by endoscopic observation. (3) Results: Minor progranulin expression was detected in atrophic tissue, with a sharp drop in the tissue colonized by H. pylori that carried greater virulence, VacAs1m1+CagA+IceA1+. (4) Conclusions: Progranulin shows a differential behavior according to the lesions and virulence of H. pylori, affecting the response of progranulin against gastric inflammation.


Introduction
H. pylori is the principal bacteria associated with several gastric pathologies. Over 50% of the world's population is colonized by H. pylori [1][2][3]. H. pylori can colonize gastric epithelial cells through various mechanisms, causing several gastro-duodenal diseases, participants responded to a survey for sociodemographic information (age, sex, educational level, Mapuche ethnicity, household, place of residence, and H. pylori infection history), in addition to morbidity characteristics, metabolic and cardiovascular diseases, and family history of gastric cancer. Patients were excluded from this research if they had received any treatment with non-steroidal or anti-inflammatory drugs, antibiotics, or proton pump inhibitors three weeks before the sampling.
Through an endoscopy procedure, three antral gastric biopsies were obtained from patients, and simultaneously, esophagus, stomach, and duodenum abnormality (gastritis, ulceration, erosion, and others) were recorded.
To generate case and control groups with similar characteristics, and thus eliminate variability caused by factors external to the study, both infected and non-infected groups were selected by matching people based on similarity of age, sex, and ethnicity. The infected group included urease-positive patients (n = 75), with or without H. pylori eradication treatment, and bacteria isolation through microbiological culture, which was confirmed by the ureC gene genotype. The non-infected group included urease-negative patients (n = 75) without a history of H. pylori eradication treatment, and negative microbiological culture for H. pylori.

Recollection of Specimens
From the three biopsies available, one biopsy was used for the Rapid Urease test in the respective Health Centre using the Rapid Urease Test (RUT) CLOtest TM , (Halyard, Alpharetta, GA, USA). For the microbiology analyses, the second sample was stored in a sterile environment in an Eppendorf tube with 200 µL of Brucella Broth (BD-DIFCO TM , Berkshire, UK). The third sample was recollected in cryotube with 500 µL of RNA later (QUIAGEN TM , Hilder, Germany), and conserved at −80 • C prior to GRN expression analyses. Finally, samples were transported in containers at 4 • C to the Applied Biology Molecular Laboratory of the Centre of Translational Medicine at the Universidad de La Frontera.

Categorization of the Gastric Epithelial Status
According to the gastric or duodenal mucosa status, there were four categories of the epithelial gastric status that were assigned by endoscopic observation, which are as follows: (1) Non-lesion (NL) for tissues without inflammatory or injuries signals, (2) Erosive lesion (EL) for tissues with wounds or ulcered, (3) Non-erosive lesion (NEL) for the simple inflammation or non-ulcerated epithelial injuries, and (4) Atrophic lesion (AtL), in the case of the atrophic epithelium [21,23].

Extraction of Total RNA and Relative Expression of Progranulin by q-PCR
To obtain the ARN from biopsies we employed the miRNA Isolation Kit (mirVANA TM , Thermo Fisher Scientific, Pleasanton, CA, USA). The protocol was adjusted to the manufacturer's recommendations by considering the size of the gastric samples. First, the tissues were manually crushed with a mortar, and the reaction volumes were adjusted to the mass of the samples, with approximate mass reaching an average of 5.0 mg. Total RNA quantity and quality were evaluated using the absorbance index 260/280 in Nanoquant M200 PRO, TECAN (Tecan Trading AG, Männedorf, Switzerland). Total RNA was reversetranscribed to cDNA with oligo (dT), Random Primers assay Invitrogen TM , and SuperScript II Invitrogen TM . Finally, 20 µL of the reaction mix was exposed at 25 • C for 10 min, 42 • C for 50 min, and 70 • C for 15 min in the Labnet Digital Thermocycler Equip, Multigene Optimax model (LabNet International. Inc., Darmstadt, Germany), to obtain cDNA.
The real-time PCR reaction involved the TaqMan Assay (TaqMan Gene Expression Assay (FAM) Biosystems TM , Foster City, CA, USA). Progranulin mRNA expression was analyzed in a duplicate GRN assay (Hs00963703_g1-Applied Biosystems TM , Foster City, CA, USA), the Glyceraldehyde 3-phosphate dehydrogenase assay (GAPDH) expression (Hs00266705_g1-Applied Biosystems TM , Foster City, CA, USA) as a housekeeping gene, and Master Mix PCR Universal TaqMan2x (Applied Biosystems TM , Foster City, CA, USA). The initial denaturation step at 95 • C for 10 min was followed by 37 cycles with a denaturation duration of 15 s at 95 • C, annealing 30 s at 58 • C and elongation at 45 s at 75 • C using 7500 Fast Dx Real-Time PCR Instrument, (Applied Biosystems™, Waltham, MA, USA). The fluorescence intensity reflected the amount of actually formed PCR-product. The genic expression was calculated with the comparative threshold cycle method, 2 −∆Ct .

Isolation and Confirmation of Helicobacter pylori Species
The gastric tissue for culture was processed within a maximum of two hours. Initially, the tissues were manually grounded with a sterile polypropylene baguette until detached from the mucus tissue. The macerated tissues (100 µL) were inoculated on Columbia agar (Oxoid TM CMO 331, Basingstoke, Hampshire, UK) plates enriched with 7% (v/v) equine blood and supplemented with the antibiotics Trimethoprim (5 µg mL −1 ), Amphotericin B (5 µg mL −1 ), Cefsulodin (5 µg mL −1 ), and Vancomycin (10 µg mL −1 )-(Supplement DENT 2%, Oxoid TM , Basingstoke, Hampshire, UK)-to favor H. pylori growth. Tissues were incubated in a microaerophilia atmosphere (5% O 2 , 10% CO 2 , 85% N 2 , and 90% humidity) using the CampyGen closed system (Oxoid TM , Basingstoke, Hampshire, UK) for three to seven days at 37 • C. Visible bacterial colonies were separated according to their phenotype, morphology characteristics, Gram stain, Catalase, and Oxidase test. All pure isolates were stored at −80 • C prior to molecular identification.
The Polymerase Chain Reaction (PCR) was applied for confirmation of the H. pylori species. First, the DNA of isolated bacteria with the suggestive H. pylori phenotype was extracted using a microbial DNA extraction kit (DNeasy UltraClean Microbial Kit ® Qiagen, Hilden, Germany) following the manufacturer's recommendations. Subsequently, PCR amplification of the ureC gene and 16SrRNA conserved region [24,25], (Table S1, Supplementary Material) was performed. The PCR amplification assays were made using a reaction mix prepared with 2.5 µL of the PCR Buffer 10X, 2.5 µL of dNTPs 200 µM, 0.125 µL Taq polymerase 1.25U (BioLabs, Inc. NEB, Ipswich, MA, USA), 1 µL of the specific forward and reverse primer of each gene at 2.5 µM mL −1 , 16.875 µL of molecular grade water, and 1 µL of DNA. Finally, the 25 µL reaction mixture was subjected to one denaturation cycle at 94 • C for 5 min, followed by 35 additional cycles, including denaturation at 94 • C for 30 s, hybridization at 53 • C for 30 s, extension at 72 • C for 30 s, and the last extension cycle at 72 • C for 5 min in a digital thermocycler, Multigene Optimax model (LabNet International. Inc. Merck KGaA, Darmstadt, Germany). Subsequently, the amplified DNA was visualized in 1% agarose gel/TBE 0.5X Buffer (Sigma-Aldrich, St. Louis, MO, USA), stained with Red Gel Ladder (Bio Labs Inc. NEB, Ipswich, MA, USA) and DNA Ladder molecular weight marker of 50 pb, by electrophoresis methods, and observed under a UV transilluminator MyEcl Image (Thermo Fisher, Pleasanton, CA, USA). The H. pylori ATCC 26695 was used as a positive control of the PCR assay.
A patient was considered H. pylori infected when the rapid urease test was positive, isolates spiral-shaped bacillus bacteria and negative Gram stain were present, the Catalase and Oxidase test was positive, and when the presence of 16SrRNA and ureC was detected.
2.6. Detection of Virulence Gene vacA and s/m Alleles, iceA More iceA1, iceA2 Alleles, and cagA, of the Helicobacter pylori Species by PCR and PCR-RFLP The characterization of the virulence of H. pylori strains was carried out through the amplification of the vacA, cagA, and iceA genes by the conventional PCR technique, including the s and m alleles of vacA and the A1 and A2 of iceA alleles, using specific primers according to the respective virulence genes, detailed in Table S1 (Supplementary Material). The reaction mixture and amplification conditions previously described in item 2.7, were adjusted to the specific melting temperature according to each primer used (Table S1). Finally, the PCR products were visualized on agarose gel by electrophoresis.
Additionally, vacA s/m alleles were confirmed by RFLP (Restriction Fragment Length Polymorphism) using DNA restriction enzyme, BstUI (Thermo Scientific™, Carlsbad, CA, USA). The reaction mixture used in this assay was made up of 10 ul of amplified segment of vacA (alleles s and m, respectively), 2 uL of Buffer NE 10x (Merck TM , Darmstadt, Germany), 0.4 µL of the BstUI enzyme (Thermo scientific TM . Carlsbad, CA, USA) and 7.6 µL of molecular-grade water (BioLabs, Inc. NEB, Ipswich, MA, USA) [25]. The reaction mix was subjected to 60 • C for 5-10 min in a thermocycler. The DNA fragments were separated by agarose gel electrophoresis 1.5% (Cleaver Scientific, Rugby, UK), in Buffer 0.5 X TBE, for 90 min, 90 V, and 300 A, and visualized in transilluminator.

Statistical Analyses
The sociodemographic records according to the infected (case) and uninfected groups (control), as well as the virulence characteristics of the H. pylori strains and the levels of expression of PGRN present in the different status of gastric tissue were analyzed with the statistical GraphPad Prisma 9.0 program. The data were expressed as average ± SD, percentages and mean and range. The continuous variables were analyzed by t-test. The categorical variables according to the sociodemographic characteristics between the infected and non-infected groups, and for virulence comparisons and gastric tissue status, were analyzed using Chi square and Fisher's exact test. In addition, the GRN expression levels were analyzed by the non-parametric Mann-Whitney test to compare two groups, and the Kruskal-Wallis test for multivariable analyses, according to the gastric tissue status of infected and uninfected patients, and H. pylori strain characteristics. Statistical significance was established at p < 0.05, applying a 95% confidence interval.

Gastric Epithelium Status
According to the endoscopic report, normal-appearing epithelium was predominantly observed, denominated as epithelia non-lesion (n = 94/62.67%), of which 46 samples (61.33%) corresponded to tissues infected with H. pylori (Table 2). Concerning the altered status of the epithelium, 56 (37.33%) biopsies presented some level of lesions, 29 (38.67%) belonging to the case group (infected by H. pylori). Within this group, 11 (14.67%) presented non-erosive lesions, with a significant difference in the non-infected (n = 2, 2.67%), of p = 0.010. Of the 35 participants with erosive lesions (EL) (23.33%), only 14 (18.67%) had H. pylori. The remaining were non-colonized by the bacteria, with significant differences (p = 0.029). In the atrophic lesions, the distribution of lesions was similar in the infected and non-infected groups, LAt n = 6, 4.00%, (n = 3, 4.00% belonging to infected group) with non-significant differences. It is important to mention that in advanced lesions (AdL), specifically metaplastic lesions, participants both in the infected group (cases) and in the non-infected group (controls), presented only one gastric tissue sample. For this reason, they were not analyzed.

Progranulin and Host Response to H. pylori Colonization and Virsulence
The GRN expression levels observed in Figure 1 show similitude in both the infected and non-infected groups, with non-significant differences.
When observing the levels of PGRN gene expression in the gastric tissue, according to the presence or absence of lesions, these values remain similar between the infected and non-infected groups ( Figure 2). Furthermore, non-significant differences of the GRN expression between infected and non-infected, as was the case between the non-lesion (NL) or lesion tissue (L) status of the cases (infected) or controls (non-infected) were observed. A significant reduction in GRN expression for individuals infected with H. pylori, according to the progress of gastric lesions and based on the precancerous cascade by Correa [26], was observed ( Figure 3). In non-erosive lesions (NEL) the GRN expression began to decrease in comparison with an NL or EL status. Furthermore, the decrease in GRN expression was more significative as it progressed towards precancerous lesions, with PGRN reaching the lowest expression in tissues with atrophic lesions (LAt) (p < 0.05).

Progranulin and Host Response to H. pylori Colonization and Virsulence
The GRN expression levels observed in Figure 1 show similitude in both the infected and non-infected groups, with non-significant differences.

Infected
Non  Furthermore, non-significant differences of the GRN expression between infected and non-infected, as was the case between the non-lesion (NL) or lesion tissue (L) status of the cases (infected) or controls (non-infected) were observed. A significant reduction in GRN expression for individuals infected with H. pylori, according to the progress of gastric lesions and based on the precancerous cascade by Correa [26], was observed ( Figure 3). In non-erosive lesions (NEL) the GRN expression began to decrease in comparison with an NL or EL status. Furthermore, the decrease in GRN expression was more significative as it progressed towards precancerous lesions, with PGRN reaching the lowest expression in tissues with atrophic lesions (LAt) (p < 0.05). The behavior of the PGRN was different in non-infected tissue by H. pylori, showing a significant difference between AtL and SL tissue (p = 0.049). However, we could observe a more acute decrease in the GRN expression of infected group (cases) than in the noninfected (controls) groups, as well as progress toward serious injuries in the Correa Cascade.

Genetic Identification of Virulence Factors of Helicobacter pylori
The virulence characteristics of colonizing H. pylori strains in the gastric tissue were studied with the virulence factors VacA, CagA and IceA in the 75 strains belonging to the infected group (cases) (Figure 4). All the isolates presented VacA with a predominance of alleles of the m2 (41/75, 54.67%) and s2 (40/75, 53.33%) strains, respectively. The iceA, present in 38 (50.67%) strains, was the second-most frequent. The allelic A2 profile was the most frequent (n = 20/26.67%), and four (n = 5.33%) strains presented both A1/A2 alleles. When observing the distribution of the vacA and iceA alleles, the most predominant were s2m2 (n = 29, 38.67%), which was distributed mainly in the tissue without lesions (n = 19, 41.30%). For iceA, the A2 allele was the most predominant, as it was detected in 20 (52.63%) isolates, and 12 (n = 63.15%) colonizing strains in tissues with lesions. Regarding the vacA s1m1 alleles, which were more closely linked to gastric lesions, 11 (37.93%) strains were observed in tissues with lesions that were mainly erosive (LE, n = 5, 35.72%). The A1 alleles were characterized by an association with superficial gastric lesions, with six (31.58%) isolates taken from tissues with lesions, 33.3% (n = 1) of in which colonized the AtL. The cagA was present in 45.33% of isolates with distribution in all epithelial conditions (Table S2, supplementary material). The behavior of the PGRN was different in non-infected tissue by H. pylori, showing a significant difference between AtL and SL tissue (p = 0.049). However, we could observe a more acute decrease in the GRN expression of infected group (cases) than in the non-infected (controls) groups, as well as progress toward serious injuries in the Correa Cascade.

Genetic Identification of Virulence Factors of Helicobacter pylori
The virulence characteristics of colonizing H. pylori strains in the gastric tissue were studied with the virulence factors VacA, CagA and IceA in the 75 strains belonging to the infected group (cases) (Figure 4). All the isolates presented VacA with a predominance of alleles of the m2 (41/75, 54.67%) and s2 (40/75, 53.33%) strains, respectively. The iceA, present in 38 (50.67%) strains, was the second-most frequent. The allelic A2 profile was the most frequent (n = 20/26.67%), and four (n = 5.33%) strains presented both A1/A2 alleles. When observing the distribution of the vacA and iceA alleles, the most predominant were s2m2 (n = 29, 38.67%), which was distributed mainly in the tissue without lesions (n = 19, 41.30%). For iceA, the A2 allele was the most predominant, as it was detected in 20 (52.63%) isolates, and 12 (n = 63.15%) colonizing strains in tissues with lesions. Regarding the vacA s1m1 alleles, which were more closely linked to gastric lesions, 11 (37.93%) strains were observed in tissues with lesions that were mainly erosive (LE, n = 5, 35.72%). The A1 alleles were characterized by an association with superficial gastric lesions, with six (31.58%) isolates taken from tissues with lesions, 33.3% (n = 1) of in which colonized the AtL. The cagA was present in 45.33% of isolates with distribution in all epithelial conditions (Table S2, Supplementary Material).

Expression Levels of GRN mRNA in the Gastric Tissue, According to the Virulence Characteristics of H. pylori Strains
The gene expression of PGRN showed no changes according to the virulence of H. pylori ( Figure 5).  When we evaluated the gene expression of PGRN by the presence or absence of CagA, the levels of GRN expression decreased in the tissues colonized with the CagA + strains, although with significant non-differences ( Figure 6). an amplification of the vacA, cagA, and iceA genes with the PCR method. The Kruskal-Wallis and Mann-Whitney tests were used for the analysis in or between the groups. Statistical significance was established at <0.05.
When we evaluated the gene expression of PGRN by the presence or absence of CagA, the levels of GRN expression decreased in the tissues colonized with the CagA + strains, although with significant non-differences ( Figure 6).

CagA+
CagA The expression of GRN in the epithelium infected with H. pylori strains carrying cagA+ tends to decrease more strongly than epithelium without H. pylori CagA + , with lower values in preneoplasic lesions (AtL) (Figure 7). Although the AtL the expression of GRN fell in the presence and absence of cagA, the differences were not significant. The expression of GRN in the epithelium infected with H. pylori strains carrying cagA+ tends to decrease more strongly than epithelium without H. pylori CagA + , with lower values in preneoplasic lesions (AtL) (Figure 7). Although the AtL the expression of GRN fell in the presence and absence of cagA, the differences were not significant. When we analyzed the gene expression levels of PGRN according to the presence of allelic profiles of vacA in the s1m1 allele, GRN was found to reach lower levels of gene expression than in the absence of this allele, presenting significant differences between both groups (p = 0.030), and in only the NL epithelium (p = 0.043) ( Figure 8B). Regarding the vacA s2m2 + alleles, we did not observe important differences when comparing the levels of GRN in tissues colonized by strains that express, or do not express, this allele. When observing the tissue status of s2m2 + strains, the GRN remains equally expressed in erosive epithelia (EL) and non-erosive epithelia (NEL). Furthermore, the GRN expressions were even higher than those expressed in NL tissues, although they were non-significant.  Finally, in the presence of iceA + (Figure 9), GRN expression was lower than observed by IceA − . In epithelia infected with strain carrying IceA + , the GRN expression presented a significant difference between NL and AtL epithelium (p = 0.021). In NEL and AtL, the expression of GRN was low, a trend previously described according to the other virulence genes analyzed. Finally, in the presence of iceA + (Figure 9), GRN expression was lower than observed by IceA − . In epithelia infected with strain carrying IceA + , the GRN expression presented a significant difference between NL and AtL epithelium (p = 0.021). In NEL and AtL, the expression of GRN was low, a trend previously described according to the other virulence genes analyzed.

Discussion
H. pylori infection is characterized by its chronicity [7]. In general, the infection is acquired in childhood. It may remain and asymptomatically colonize the gastric epithelium for decades or generate inflammatory manifestations in a small group of infected people, mainly in adults [27]. It is important to mention that the prevalence of lesions observed in our study group correlates with that expected for H. pylori infection. Of the infected dyspeptic patients (case), 61.33% (n = 46) did not present lesions observable by endoscopy, 18.67% presented EL and 14.67% presented NEL, while pre-neoplasia lesions reached a prevalence of 5.3% (n = 4), all in agreement with previous epidemiological reports [28][29][30].
The environment, characteristics of the population, and the virulence of the bacteria are critical factors for the development of the gastric disease [2]. Clinical and family history may impact the development of advanced pathologies, including cancer [31], and interestingly, the participants had an average age of 48.18 years. It is believed that H. pylori infection is usually acquired at an early age (before 10 years old), and is characterized by chronicity. In this case, a large proportion of this group would already be affected by a chronic process for more than three decades, a sufficient period for colonization and to generate changes in the gastric mucosa [5]. However, it is crucial to mention that the gastric epithelium is exposed to multiple factors that can mediate between health and disease.

Discussion
H. pylori infection is characterized by its chronicity [7]. In general, the infection is acquired in childhood. It may remain and asymptomatically colonize the gastric epithelium for decades or generate inflammatory manifestations in a small group of infected people, mainly in adults [27]. It is important to mention that the prevalence of lesions observed in our study group correlates with that expected for H. pylori infection. Of the infected dyspeptic patients (case), 61.33% (n = 46) did not present lesions observable by endoscopy, 18.67% presented EL and 14.67% presented NEL, while pre-neoplasia lesions reached a prevalence of 5.3% (n = 4), all in agreement with previous epidemiological reports [28][29][30].
The environment, characteristics of the population, and the virulence of the bacteria are critical factors for the development of the gastric disease [2]. Clinical and family history may impact the development of advanced pathologies, including cancer [31], and interestingly, the participants had an average age of 48.18 years. It is believed that H. pylori infection is usually acquired at an early age (before 10 years old), and is characterized by chronicity. In this case, a large proportion of this group would already be affected by a chronic process for more than three decades, a sufficient period for colonization and to generate changes in the gastric mucosa [5]. However, it is crucial to mention that the gastric epithelium is exposed to multiple factors that can mediate between health and disease. Transmissible factors, such as bacteria, protozoa, fungi and viruses, commensals, and pathogens, may induce inflammatory responses and cause epithelial damage [32]. Additionally, non-transmissible factors may affect these responses. Biliary or pancreatic secretions, diet, consumption of alcohol, drugs (e.g., non-steroidal anti-inflammatory drugs, NSAIDs), and some immunological disorders, may cause inflammation and could be reflected in gastric tissue alterations [33]. This is observable in tissues uninfected by H. pylori, such as our control group (controls).
Regarding the pathogenic process of H. pylori, the mechanisms that lead to the development of precancerous lesions and cancer are not fully understood [34]. However, diverse regulatory mechanisms, including progranulin, could occur to regulate the inflammatory process involved in lesion development [14,16]. Different functions are attributed to progranulin, depending on complete or cleaved status. In its complete form, the antiinflammatory capacity of progranulin is mainly associated with the inhibition of TNFα through binding TNF receptors [35,36]. On the other hand, TNF-α is a proinflammatory cytokine that plays an important role in tumorigenesis by the gene-related expression of cytokines, adhesion molecules, and proangiogenic molecules [37]. Therefore, its presence indicates a poor prognosis for increased cancer development, including gastric cancer [38]. In the gene expression of PGRN in the antral tissue, no significant differences were observed, regardless of H. pylori infection or the presence of lesions; however, this behavior is different when evaluating specific types of gastric lesions. In the tissue colonized by H. pylori (case), the levels of GRN expression tend to decrease as the lesions progress by the Correa precancerous cascade [39]. In these infected tissues, the NL or EL expressed similar levels of GRN, which decreased in the tissue with micronodular gastropathy (NEL), with a significant drop in AtL where the glandular epithelial was lost (p < 0.05).
PGRN participates in the repair of damaged tissue in gastric ulcers. In fact, PGRN administered experimentally in fresh skin wounds of rats stimulates specific pathways of inflammation, and favors the accumulation of fibroblasts and vascular regeneration, which are necessary conditions for the repair of damaged tissue [21]. Studies of PGRN and the inflammatory response in gastric ulcers in murine models, attribute a role in the healing of gastric ulcers to PGRN by macrophages dependent on the macrophage colony-stimulating factor (M-CSF), promoting angiogenesis through the upregulation of Cyclooxygenase/Prostaglandin E (COX-2/PGE) production and expression of VEGF (Vascular Endothelial Growth Factor), [40]. Furthermore, studies of muscle injury regeneration suggest that PGRN is involved in the regulation of macrophage kinetics for muscle regeneration [41]. Analyses in human gastric tissue demonstrated the presence of PGRN both in the gastric epithelium, as well as in immune cells, except in lymphoid follicles [22]. In our study, EL reached levels of GRN expression similar to tissues without lesions (NL), which does not occur in NEL, where GRN levels begin to decrease.
A study by Wex et al. (2011) recognized higher levels of protein in PGRN, by immunohistochemistry, in the glandular tissue and on the basis of foveola of the stomach that present dense inflammatory infiltrates and its expression is weak or practically null between the gastric pits [22]. The transition of the epithelium in response to H. pylori infection in the progress of the Correa cascade may modify the receptor to TNFα, on which progranulin has an antagonistic effect [16] or may alter mitogenic pathways MEK1/2 and p38, as proposed by Wang 2011 [14]. Thus, epithelial and immune response changes would affect the expression of GRN in LNE-type lesions or deeper lesions such as AtL, which could explain GRN downregulation in cases where the precancerous cascade progresses, as observed in our results.
On the other hand, Wex previously described a lower expression of GRN in infected tissues when compared with uninfected tissues [22]. This downregulation of GNR could be attributed to translational regulation exerted by H. pylori infection, but further investigation is needed to elucidate our understanding. These changes in the cell cycle caused by H. pylori can push on transcriptional factors and mitogenic pathways [42,43], and therefore impact the GRN expression and the integrity of the gastric epithelium, although not all cases will progress to disease and malignancy [43].
Bacterial populations of H. pylori are extremely variable, and the effect of virulence factors can impact the cellular response [44]. VacA is a virulence factor which the pathogenic effect depends on genotypes conformed by alleles s and m, i.e., s1m1 and s1m2 alleles, which are recognized due to their ability to induce vacuolization of colonized cells, leading to a more accentuated cell damage [8]. In our study, the presence of vacA was observed in all the strains analyzed and at all levels of epithelial damage. Additionally, the expression levels of GRN in epithelium with and without the VacA s1m1 allele were statistically different (p < 0.05). Coincidentally, this allele (VacAs1m1 + ) predominates in atrophic epithelia that also show a significant drop in GRN expression. Furthermore, the vacA s1m1 alleles trigger damage, intracellular modifications, induce apoptosis, and inhibit the activation and proliferation of T lymphocytes, as well as the modulation of the inflammatory cytokine and immunosuppression [8] [14]. VacA, by a mechanism not yet elucidated, usurps the lysosomal and autophagy pathways that are favorable for the survival, colonization, and chronicity of the bacteria [18], and interrupts endolysosomal traffic, inducing autophagosomal survival [45].
Other virulence factors of H. pylori strains also influence the epithelium transformation process. Among them, iceA was present in 50.67% (n = 38), whereas cagA was present in 45.33% (n = 34). Coincidentally, IceA1 has been linked to ulcerative lesions when combined with CagA. CagA + IceA2 + is more dominant in active chronic inflammation, gastric ulcer, and carcinoma than when combined with the A1 allele, according to studies in dyspeptic patients from Pakistan [46]. In our reports, cagA is present in atrophic lesions similarly to vacA s1m1 alleles ( Figure S1, Supplementary Material). Studies have shown that infection by CagA + strains is associated with cell cycle alterations, apoptosis, pronounced levels of inflammation, and greater intensity of gastric atrophy and metaplasia [47,48]. Regarding CagA and its relationship with epithelium status, Tserentogtokh et al. (2019) reported a prevalence of CagA of at least 30% in peptic ulcers, 41.6% in gastritis, and 64.7% in gastric cancer for western variant strains, and of up to 100% in isolates from Asian strains through sequencing [49]. In our study, CagA strains were isolated from all epithelia studied with predominance of the erosive type (n = 8, 57.14%).
CagA has been closely linked to malignant lesions, and is related to other virulence factors, with a tendency to migrate towards the healthy epithelium [50]. However, Wang et al., 2011, showed that bacterial virulence, and CagA overexpression did not induce GRN mRNA levels from the cDNA amplifier in cell lines immortalized by q-PCR, suggesting that CagA is not sufficient to induce PGRN upregulation [14]. In fact, according to our results, CagA does not influence PGRN gene expression.
CagA transcription factors increased cell proliferation through the Ras/MERK/ERK mitogenic pathway, as with PGRN [14,51]. Additionally, Cag A can induce the mitogenic response by the JAK2/STAT3 signaling pathway when it binds to the membrane protein GP130 and to SHP-2 [52,53]. Nevertheless, we did not find a link to PGRN. Other researchers also point out that CagA regulates inflammatory responses through the activation of IL-8, [54], however, no reports were found linking PGRN with the activation of this cytokine. This may explain the lack of association of CagA with the gene expression of PGRN. On the other hand, CagA induces inflammation through the activation of NF-κB by way of tumor necrosis factor alpha (TNF-α) [55,56], a pathway also controlled by PGRN [16]. This common utilization of inflammatory pathways is possibly due to a competition effect generated between TNF-α and PGRN molecules. Further studies are needed to corroborate this interaction in future studies, considering that the samples analyzed using these epithelial characteristics are scarce compared to the status of other epithelial lesions.
Additionally, IceA + is also found in atrophic lesions, presenting significant differences between the levels of GRN detected in the epithelium without lesions and the atrophic one (p < 0.05). However, the pathogenesis of molecular pathways of IceA still need to be elucidated.
Finally, downward deregulation in GRN levels is observed in the most severe lesions, coincidentally colonized by more aggressive strains of H. pylori. Although the data are interesting, it is essential to consider the limitations of our results, particularly the bias from the lesion identification criteria and the few participants with atrophic and advanced lesions. These factors contributed to a wide dispersion of the data, making it necessary to increase the sample size. Further studies are needed to focus on the regulatory mechanisms of PGRN and the GRN expression as a factor in the inflammatory process and cellular response associated with the infectious process.

Conclusions
Although the levels of GRN expression detected in H. pylori-infected tissues are similar to those obtained in uninfected tissue, the GRN expression progressively decreases according to lesion progression in the precancerous Correa's cascade, mainly in the tissues colonized by H. pylori with a lower GRN expression in the more advanced stages of the lesion. With H. pylori infection, GRN expression is affected by the virulence of the infecting strains, particularly vacA s1m1 and IceA1. However, cagA did not show significant effects. Additional research should consider molecular processes occurring at the immune response level and regulatory pathways of the cell cycle associated with VacA and CagA strains.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/microorganisms10050998/s1, Figure S1: Distribution of vacA, cagA, iceA genes in H. pylori isolates obtained from gastric biopsies; Table S1: Primers used for confirmation of species and study of virulence of Helicobacter pylori by PCR; Table S2: Characterization of the virulence of Helicobacter pylori strains, including s and m alleles of vacA, and A1 and A2 of iceA alleles, colonizing in the different gastric epithelium status. Document S1: Informed consent and protocol code no. 028/18, approval date 27 November 2018. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Written informed consent has been obtained from the patient(s) to publish this paper. Data Availability Statement: Not applicable.