Diagnostic of H. pylori infections in humans is based on examination of gastric tissue specimens collected during gastroscopy. The gold standard of diagnostic tests contains a rapid urease test (RUT) for screening the activity of urease produced by H. pylori, histological staining of gastric tissue for detection of Helicobacter-like organisms (HLO), assessment of the inflammatory response, and microbiological culture of bacteria^35,36. Non-invasive diagnostic methods include the ¹³C urea breath test (UBT), which is sufficiently specific and sensitive for primary diagnosis and for confirming the effectiveness of eradication therapy. Serological assays, such as enzyme-linked immunosorbent assay (ELISA), detect particular anti-H. pylori antibodies in serum samples or H. pylori antigens in stool samples^36,37. However, a comprehensive method is needed to quantify selected markers to track the systemic effects of H. pylori local infection in gastric tissue. Potentially, such markers may help predict the course of infection and its regional and systemic consequences, including coronary heart disease, anemia, and type 2 diabetes^{38,39,40,41,42}. Knowledge about the mechanisms driving different courses of H. pylori gastric infections and the potential role of these bacteria in developing systemic diseases is insufficient. Experimental animal models allow deepening this knowledge. We used the experimental model of H. pylori infection in guinea pig, which is accepted for studying the pathogenesis of H. pylori infection and comparing to some extent data derived from this model to H. pylori infection in human due to structural and functional similarity of the immune system and physiology of the gastroduodenal tract in both species.

Although serum antibody titer assay, urease activity assay, and stool antigen assay in conjunction with histological examination of gastric tissue may be sufficient for diagnosing H. pylori infection in humans, there are no reliable markers to differentiate the course of infection and its possible systemic consequences. Metabolomic analyses offer such opportunities. They have been applied to search for markers of infectious diseases and monitor treatment effectiveness. For instance, in plasma samples from subjects infected with human immunodeficiency virus (HIV), the LC–MS analysis showed the presence of acetate, citrate, creatine, dicarboxylicacylcarnitines, dopamine, glucose, glycerophospholipids, glycolysis, L-aspartate, plasmalogen/plasminogen, lysophospholipids, methylglutarylcarnitine, phosphatidylcholines, sphingomyelin, sphingosine-1-phosphate as potential biomarkers⁴³. In serum samples from patients infected with influenza virus Banoei et al., using ¹H-NMR, detected citrate; fumarate; 3-Methyl,2-Isovalerate; alanine; tyrosine; methionine; histidine; 4-hydroxybutyrate⁴⁴, while in patients infected with COVID-19 LC–MS analysis showed the presence of bile acids, bilirubin, diacylglycerols, free fatty acid, glucose, glucuronate, glycerol 3-phosphate, kynurenine, lysophosphotidylcholines, malic acid, monosialodihexosylganglioside, phosphatidylcholines, sphingomyelin, triglycerides and tryptophan⁴³. In urinary tract infections caused by Escherichia coli by using ¹H-NMR and LC–MS, the following metabolites characteristic for this type of infection were selected by examination of urine: acetate, amines, aspartic acid, cadaverine, citrate, glutamic acid, glycine, hippurate, trimethylamine, and trimethylamine n-oxide⁴⁵. In the patients with tuberculosis, the potential biomarkers selected by assessment of serum/plasma samples using LC–MS, FIA-MS, and GC-M methods were amino-acyl tRNA, asparagine, aspartate, citrulline, cysteine, gamma-glutamylglutamine, glutamate, glutamine, histidine, inosine, kynurenine, lysophosphatidylcholines, mannose methionine, sphingolipid, sphingosine-1-phosphate, sulfoxymethionine, tryptophan and urea⁴⁵. In the previous study, we used Fourier-transform infrared spectroscopy (FTIR) to combine hierarchical cluster analysis (HCA) to determine the fragments of infrared spectroscopy (IR) spectra and their corresponding biomolecules of guinea pig sera characteristic for H. pylori infection. Specific molecules, which were identified in the composition of the IR spectra of guinea pig sera, included glucose, α2-globulins, IgM, IgG1, transferrin, IgG4, C-reactive protein (CRP), and tumor necrosis factor alfa (TNF-alfa), which facilitated the differentiation between H. pylori infected and H. pylori uninfected animals⁴⁶.

Data obtained in this study using the UPLC-QTOF/MS in conjunction with Fold change analysis and T-test results facilitated selection of 22 metabolites with a signal intensity significantly lower and 48 metabolites with a signal intensity markedly higher in H. pylori infected vs. H. pylori uninfected animals, and 12 metabolites, which delivered significantly higher signals were selected as well differentiating biomarkers associated with H. pylori infection. Database searching and literature analysis indicated that these 12 selected metabolites well differentiating animals infected with H. pylori from animals uninfected with these bacteria were primarily associated with immune regulation, energy metabolism, lipid/fatty acid metabolism, lipid peroxidation, oxidative stress, and cell signaling, and correlate with the pathogenesis of H. pylori infection in humans. The remaining 36 metabolites can potentially be treated as additional markers related to H. pylori infection, however their role in the course of H. pylori pathogenesis has not been described.

Among selected 12 differentiating metabolites leukotriene B4 is a pro-inflammatory dihydroxy fatty acid derived from arachidonic acid produced by leukocytes involved in the inflammatory response. It promotes the adhesion of leukocytes to the vascular endothelium and activates them to produce pro-inflammatory cytokines and mediators⁴⁷ during chronic inflammation and cancer⁴⁸. Hüseyinov et al. showed an elevated level of leukotriene B4 in gastric juice in H. pylori-positive children⁴⁹.

Lysophospholipid (LPC) exerts biological effects on other cells by stimulating calcium flux, cell proliferation, differentiation, and activation. The LPC-driven migration of glioma cells and mouse fibroblasts but not macrophages was shown in several studies^50,51,52. Higher levels of LPC induce cyclooxygenase-2 and endothelial nitric oxide synthase (eNOS) expression in endothelial cells, which can have vasoprotective effects via prostacyclin or nitric oxide. LPCs have been shown to elicit several effects on the innate immune system and effectively serve as dual-activity ligand molecules. In particular, LPCs directly activate toll-like receptor (TLR) 4 and TLR-2 without classical TLR ligands. However, LPCs can also inhibit TLR-mediated signaling in the presence of classical TLR ligands, thereby acting as anti-inflammatory molecules. Low levels of LPC during a bacterial or viral infection with TLR-mediated signaling can lead to opposing (inflammatory vs. anti-inflammatory) effects and immune dysregulation³¹.

Phospholipids (PS) bind to various proteins and activate enzymes, apoptosis, neurotransmission, and synaptic refinement⁵³. Phosphatidylglycerol (PG) is distributed in the animal cell mitochondria and, when located in lung surfactant, is critical for regulating innate immunity^54,55. PG competes with bacterial lipopolysaccharides to disrupt the signaling pathways dependent on toll-like receptors (TLRs)^56,57. Phospholipids, mainly phosphatidylethanolamine (PE), comprise most H. pylori lipids in developing membrane domains^58,59. H. pylori also produces outer membrane vesicles (OMVs) that contain various virulence factors that promote the survival of these bacteria in the gastric mucosa⁶⁰. H. pylori OMVs contain PG, PE, lyso PE (LPE), phosphatidylcholine (PC), lyso PC (LPC), and cardiolipin⁶⁰. PS externalization is accompanied by increased permeability of eukaryotic cell membranes⁶¹. H. pylori OMVs, due to the induction of gastric barrier instability and dysbiosis, can be translocated to deeper layers of gastric mucosa and in the gut. They may also be translocated to the circulation where, due to interaction with leukocytes, OMVs may affect the activity of these immune cells. Also, direct contact of H. pylori with the host cell membranes induces rapid and transient externalization of PS, independently of the apoptotic process at the bacterial attachment site⁶². Moreover, H. pylori exploits host membrane phosphatidylserine for delivery, localization, and pathophysiological effects of the CagA oncoprotein⁶². Therefore, increased levels of lipids, including PG and PS, in the serum samples of guinea pigs infected with H. pylori may be potential markers of systemic effects of H. pylori infection related to chronic inflammation. Some of the metabolomic biomarkers increased in H. pylori infected guinea pigs, such as 20-dihydroxyleukotriene B4 and dihomo-gamma-linolenoylethanolamide is engaged in the peroxidation of lipids, which in such form are involved in the development of vascular disorders. It is worth mentioning that H. pylori infection is considered a non-classical risk factor in the development of coronary heart disease^{14,15,36,63,64,65}. It has been shown that H. pylori LPS increases oxidative stress, which results in lipid peroxidation assessed as an increased level of 4-Hydroxynonenal (4HNE) both in gastric epithelial cells and vascular endothelium³⁶. Furthermore, H. pylori components induced the transformation of macrophages into foam cells in vitro. They increased the severity of the metabolic syndrome and hepatic manifestations caused by a high-fat diet in a model of Cavia porcellus. The infiltration of inflammatory cells into the vascular endothelium of animals infected with H. pylori and exposed to a high-fat diet was observed in conjunction with an increased level of inflammatory markers systemically^39,41.

In this study, the leucine metabolites (Iso leucyl-phenylalanine, N-Lactoylleucine) differentiated the animals infected with H. pylori from those uninfected with these bacteria. Felig et al., using the rat model, showed an increased leucine plasma level in these rodents, which was related to insulin resistance⁶⁶ and cardiovascular dysfunction in obese rats⁶⁷. Interestingly, increased concentration of plasma leucine was associated with impaired endothelium-dependent relaxation in Zucker diabetic fatty rats⁶⁸ and with reduced energy usage in rats with diet-induced obesity^69,70. The effect of affected endothelium relaxation was also shown in humans with hyperglycemia⁷¹. Our previous study showed diminished vascular relaxation in guinea pigs receiving a high-fat diet infected with H. pylori^39,41.

In the present study, deoxycholic acid (DCA) differentiated guinea pigs infected with H. pylori from those not infected. Frazier et al. showed that elevated serum level of DCA, a secondary bile acid, was associated with vascular calcification in patients with chronic kidney disease (CKD)⁷². DCA causes numerous harmful effects, including decreased insulin sensitivity and immune dysregulation⁷³ and potential cardiovascular toxicity by promoting vascular calcification⁷³. Therefore, increased levels of leucine and/or DCA in the serum may be potential markers of vascular endothelium dysfunction.

5-Methoxytryptophol is an indoleamine synthesized from the pineal gland hormone serotonin and, together with melatonin, is involved in regulating daily rhythms. 5-Methoxytryptophol is highest during the day, while melatonin is highest at night. 5-methoxytryptophol-mediated receptors play a role in regulating cerebral artery contractility, intra arrhythmia, and the regulation of renal function. 5-Methoxytryptophol also shows immunomodulatory, antioxidant, and anxiolytic properties with different mechanisms. Environmental factors other than light, including insulin-induced hypoglycemia, can override the inhibitory effects of light and accelerate melatonin synthesis^45,46. In this study, 5-Methoxytryptophol was a differentiating metabolite in the sera of H. pylori-infected guinea pigs. A decrease in the serum levels of melatonin synthesizing enzymes is seen in patients with symptomatic H. pylori infection, and potentially, the processes regulated by 5-methoxytryptophol/melatonin might be affected in these patients. Luo et al., showed that melatonin used to immunize mice infected with H. pylori could eradicate these bacteria. However, the exact molecular mechanism is unknown. It can be that melatonin treatment of mice infected with H. pylori resulted in a decreased count of regulatory lymphocytes (CD4 + CD25 + Foxp3 + Treg) in the spleen and diminished secretion of transforming growth factor-beta (TGF-beta), regulatory cytokine, on TLR-4 and TLR-2 dependent manner⁷⁴. It has been suggested that different neurological disorders, such as Parkinson’s disease, Alzheimer’s disease, Guillain-Barré syndrome, and multiple sclerosis, are linked to H. pylori infection. In some of them, sleep disruption, nightly restlessness, sundowning, and other circadian disturbances are frequently seen⁷⁵. In animal models of Alzheimer’s disease, H. pylori infection localized in the stomach induced neuroinflammation, and the potential role of outer membrane vesicles has been suggested in developing this phenomenon⁷⁶.

GABA is the principal inhibitory neurotransmitter in the central nervous system. Both the downregulation of GABA production and excessive release of this neurotransmitter can drive neurological disorders⁷⁷. However, little is known about the potential link between H. pylori infections and GABA modulation. It has been shown that H. pylori infection releases several neurotransmitters, such as acetylcholine, adrenaline, noradrenaline, serotonin, and dopamine. Moreover, H. pylori infection might lead to axonal/neuronal damage, production of free radicals, and changes in neuropeptide expression, such as vasoactive intestinal peptides⁷⁸. In this study, elevated levels of GABA were observed in serum samples of H. pylori-infected guinea pigs. It has been reported that GABA has antioxidant and anti-inflammatory action by regulating major inflammatory events and immune cell activities⁷⁹. Considering those GABA activities, excessive concentration of GABA in our model of H. pylori infection in guinea pigs may be a part of the host response against exposure to H. pylori.

This study shows that the above selected serum metabolites potentially may be valuable biomarkers for predicting and monitoring the systemic consequences of H. pylori infection.