FAs are characterized by the lengths of their carbon chains and their degrees of unsaturation. FAs with carbon-to-carbon double bonds are classified as unsaturated and those without double bonds as saturated. FAs can be further classified by the length of their carbon chains: short-chain FAs (< 6 carbons), medium-chain FAs (6 to 12 carbons), long-chain FAs (13 to 21 carbons), and very-long-chain FAs (> 22 carbons). The biological activities of FAs vary depending on their degree of unsaturation and length [10].

In 1989, Hazell et al. [11] reported that bovine serum albumin and catalase might reduce the toxic effects of fatty acids by adsorption to short-chain fatty acids and prevent the formation of toxic products from long-chain unsaturated FAs, thereby promoting the growth of H. pylori. To determine the toxic effect of long-chain polyunsaturated fatty acids (PUFAs), that group evaluated the effect of arachidonic acid (with 20 carbon atoms and 4 double bonds, or C20:4) and confirmed its bactericidal activity against H. pylori following 1 hour incubation in 0.1 mM arachidonic acid [12]. In an in vitro study by Thompson et al. [13], other unsaturated FAs also exhibited inhibitory effects on H. pylori growth; the inhibitory effects were greater for the FAs with a higher degree of unsaturation (oleic acid [C18:1] < linoleic acid [C18:2] < linolenic acid [C18:3] = arachidonic acid [C20:4] = eicosapentaenoic acid [C20:5]). That study also examined the intracellular distribution of ¹⁴C-labeled linolenic acid and found that the majority (82%) was in the membrane of H. pylori, suggesting that FAs could be associated with or incorporated into the bacterial membrane and may increase its permeability. These results were confirmed by Khulusi et al. [14]. The levels of incorporation into the membrane and growth inhibition of linoleic acid were greater than those of oleic acid, indicating that the antibacterial mechanism of FAs is associated with its incorporation into the bacterial membrane. Electron microscopy revealed a distortion of the protoplasmic cylinder, as well as disruption and fragmentation of the bacterial cell membrane. Later studies showed that lauric acid (C12:0), a fully saturated FA, has bactericidal effects similar to those of PUFAs [15,16]. However, other short- or medium-chain saturated FAs exhibited little or no inhibitory activity against H. pylori. Interestingly, the development of spontaneous resistance to lauric acid was much lower than that to metronidazole or tetracycline. This result suggests that lauric acid also acts on the membrane of bacterial cells [15].

In a study by Sun et al. [17], the potency of FAs appeared to be related to the equivalent carbon number (ECN, or the number of carbon atoms –2 × the number of double bonds), which is used for the chromatographic analysis of lipids, in which similar ECNs imply similar retention times caused by molecules of similar size, shape, charge, or polarity. FAs with an ECN of 12, including lauric acid (C12:0), myristoleic acid (C14:1), and linolenic acid (C18:3), had the most potent antibacterial effects. This hypothesis can be applied to the previous example for arachidonic acid (C20:4) [12]. This study also reported that the bactericidal potency of unsaturated FAs increased with the degree of unsaturation, that the potency of lauric acid increased at lower pH values, and that urea and endogenous urease did not protect H. pylori from the bactericidal action of FAs. Docosahexaenoic acid (C22:6) decreased H. pylori growth in a dose-dependent manner and induced conversion from the bacillary to coccoid form, which is associated with decreased cell viability [18]. The results of the above studies on antibacterial activity are summarized in Table 1.

PUFAs have been investigated for their protective effects against various inflammatory diseases, such as inflammatory bowel disease and autoimmune diseases [19,20]. n-3 PUFAs can be converted into bioactive mediators and have anti-inflammatory properties via the counter-regulation of lipid mediators, including pro-inflammatory leukotrienes and prostaglandins [21,22]. Therefore, PUFAs may also have anti-inflammatory effects on H. pylori-infected gastric mucosa, in addition to their antibacterial activity. Correia et al. [18] reported that treatment with docosahexaenoic acid reduced inflammatory responses and the production of prostaglandin E2, which is associated with inflammation and tissue injury in the mouse gastric mucosa. That group also found that docosahexaenoic acid reduced bacterial adhesion to the gastric epithelium, reduced the metabolic activity of H. pylori, and reduced the production of interleukin-8 (IL-8), cyclooxygenase 2, and inducible nitric oxide synthase from gastric epithelial cells [23]. The inhibitory effects of PUFAs on IL-8 mRNA and protein expression in H. pylori-infected cells were confirmed by Lee et al. [24]. Interestingly, the inhibitory potency of individual FAs on IL-8 expression differed. Among PUFAs, longer and more unsaturated FAs had greater anti-inflammatory effects (linoleic acid [C18:2] < arachidonic acid [C20:4] < docosahexaenoic acid [C22:6]). However, palmitic acid (C16:0), a saturated FA, did not inhibit IL-8 production.

Although many in vitro studies have confirmed the antibacterial effects of FAs, the efficacy of FAs against H. pylori infection in vivo remains controversial. Based on the finding that in vitro PUFAs inhibited H. pylori growth and modified gastric mucosal injury [13,25], Duggan et al. [26] investigated the effects of orally ingested PUFAs in patients with H. pylori-associated duodenal ulcers. Disappointingly, dietary interventions failed to result in significant changes in either gastric colonization by H. pylori or prostaglandin levels after 6 weeks, despite a significant difference in linoleic and linolenic acid consumption. Contrary to this result, another study investigating the effects of dietary PUFAs in 15 patients with functional dyspepsia and H. pylori infection demonstrated the possibility of PUFAs as an adjuvant treatment in the eradication of H. pylori [27]. The 8-week supplementation of mixed oil containing PUFAs induced bacterial clearance in 53% of patients by the end of treatment and bacterial eradication in 20% of patients 6 months later. Based on an analysis of bacterial 16s rRNA from mouse feces, fish oil with high PUFA content suppressed the growth of H. pylori [28]. In addition, Correia et al. [18] reported that docosahexaenoic acid reduced the ability of H. pylori to colonize the gastric mucosa in 50% of mice, and that the combination of docosahexaenoic acid with standard triple therapy decreased the recurrence of H. pylori infection in an in vivo mouse model. This suggests that docosahexaenoic acid should not be regarded as a replacement for conventional antibiotic regimens, but it can be useful in combination therapy to reduce recurrence [18].

In another interesting study, the average levels of eicosapentaenoic and docosahexaenoic acids were higher in the abdominal and buttock adipose tissues of patients without H. pylori compared with those of H. pylori-positive patients, indicating that dietary FAs could inhibit the growth of H. pylori [29]. However, a randomized double-blind trial reported that an eradication regimen containing fish oil (pantoprazole, clarithromycin, and eicosapen) exhibited a significantly inferior H. pylori eradication rate compared with a conventional eradication regimen (pantoprazole, clarithromycin, and metronidazole), but it improved symptoms in patients with non-ulcer dyspepsia regardless of H. pylori status [30]. Recently, Khandouzi et al. [31] evaluated the effects of adding PUFAs (eicosapentaenoic and docosahexaenoic acids) to a bismuth-based quadruple therapy on both H. pylori eradication and inflammatory markers. PUFAs as a supportive therapy had no additive effects on H. pylori eradication, IL-6 levels, or total antioxidant capacity.

There is a considerable discrepancy between the reported in vitro and in vivo antibacterial effects of FAs. Overall, dietary FA regimens currently are not effective enough to serve as a primary or adjuvant eradication therapy against H. pylori infection. However, linolenic acid, a PUFA, exerted bactericidal effects on Staphylococcus aureus in vitro and also reduced the numbers of this bacterium on human skin [32]. Therefore, it can be inferred that poor delivery of FAs to H. pylori may result in reduced or absent antibacterial effects in vivo. The antibacterial activity of FAs in the stomach may be reduced, because some FAs, especially long-chain PUFAs, effective against H. pylori have poor solubility, which is further decreased following oral administration, are sensitive to oxidation and esterification, form lipid-protein complexes, and bind to proteins or other compounds [10,33]. Previous studies have reported that esterification of oleic acid resulted in the complete loss of its inhibitory effects, and that the antibacterial activity of FAs was markedly reduced in the presence of proteins [14,34].

Liposomes are considered novel drug delivery vehicles and are widely used to deliver therapeutic agents. Because of the phospholipid bilayer structure of liposomes, liposomes can improve the solubility of FAs, protect FAs from degradation, and easily fuse with bacterial membranes, thereby delivering their entrapped FAs into the bacterium [35-37]. Obonyo et al. [38] developed a liposomal nanoformulation of linolenic acid (LipoLLA) and evaluated its bactericidal activity against H. pylori in vitro. LipoLLA had a bactericidal effect on both the spiral and coccoid forms of the bacterium and eradicated various clinical strains, regardless of their antibiotic-resistance status. This group also investigated the antibacterial mechanism of LipoLLA against H. pylori [39]. LipoLLA rapidly killed H. pylori within 5 minutes, and the antibacterial activity of LipoLLA was considerably more potent than that of liposomal oleic acid (C18:1), which was consistent with previous studies using FAs. They also found that LipoLLA increased the permeability of both the outer and inner plasma membranes of H. pylori by measuring the uptake of 1-N-phenylnaphthylamine and detecting the release of adenosine triphosphate from bacterial cells, respectively. Structural changes to the bacterial membrane upon LipoLLA treatment were observed by both transmission and scanning electron microscopies.

Recently, the anti-H. pylori efficacy of LipoLLA was evaluated in a mouse model [40]. After the oral administration of LipoLLA, linolenic acid accumulated within the gastric mucus layer and a significant portion was retained for up to 24 hours. To evaluate its in vivo therapeutic efficacy, H. pylori-infected mice were treated with linolenic acid, LipoLLA, or standard triple therapy (omeprazole, clarithromycin, and amoxicillin). While linolenic acid failed to exhibit antibacterial activity, LipoLLA had in vivo efficacy superior to that of the standard triple therapy and resulted in reduced levels of H. pylori-induced proinflammatory cytokines, including IL-1β, IL-6, and TNF-a. In a toxicity test, LipoLLA had no effect on body weight, gastric histopathology, or gastric mucosal integrity in mice, indicating that LipoLLA has excellent biocompatibility.