OUP user menu

Antibacterial actions of fatty acids and monoglycerides against Helicobacter pylori

Cynthia Q. Sun, Charmian J. O'Connor, Anthony M. Roberton
DOI: http://dx.doi.org/10.1016/S0928-8244(03)00008-7 9-17 First published online: 1 May 2003


The bactericidal potencies of saturated and unsaturated fatty acids (FAs) and monoglycerides (MGs) against Helicobacter pylori were determined following short incubations with freshly harvested cells over a range of pHs. FAs and their derivatives with an equivalent-carbon number of 12 were the most potent: lauric acid had a minimum bactericidal concentration (MBC) at pH 7.4 of 1 mM, myristoleic and linolenic acid were the most potent unsaturated FAs (MBCs of 0.5 mM, pH 7.4), and monolaurin was the most potent MG (MBC 0.5 mM). Potencies of saturated FAs were increased sharply by lowering pH, and a decrease of only 0.5 pH units can cause a change from non-lethal to lethal conditions. Conversely, the bactericidal action of monolaurin was not pH-dependent. The bactericidal potencies of unsaturated FAs increased with degree of unsaturation. When more than one FA or FA plus MGs were present, their combined action was additive. Urea and endogenous urease did not protect H. pylori from the bactericidal action of FAs. These results suggest that H. pylori present in the stomach contents (but not necessarily within the mucus barrier) should be rapidly killed by the millimolar concentrations of FAs and MGs that are produced by pre-intestinal lipase(s) acting on suitable triglycerides such as milkfat.

  • Helicobacter pylori
  • Fatty acid
  • Monoglyceride
  • Bactericidal agent

1 Introduction

Helicobacter pylori colonises the stomach or duodenum of about half the human population. The presence of these bacteria is causally linked to gastritis, dyspepsia, stomach and duodenal ulcers, and gastric cancer [1,2]. The best available treatments with combinations of drugs give between 70 and 100% eradication of the bacterium, but patient compliance, side-effects of the drugs, and the developing resistance of the bacterium to the available antibiotics (present in 2–40% of patients depending on the antibiotic) [37] are continuing problems. Currently, many new methods for eradication of H. pylori colonisation are being explored.

Fatty acids (FAs) and monoglycerides (MGs) possess both bacteriostatic and bactericidal properties against a diverse range of bacteria [8], and these compounds are the key ingredients that make hydrolysed milkfat antimicrobial [9], at least under in vitro conditions. It may be no coincidence that relatively recently, in the 1950s, when hyperacidity and stress were thought to be the main cause of stomach ulcers, their treatment was a bland milk diet with or without antisecretory compounds [10]. The function of the milk in alleviating the symptoms of stomach ulcers was believed to be its buffering effect, which suppressed acidity in the stomach.

In studies on hydrolysis of bovine milkfat by pregastric lipase, O'Connor and co-workers found that short-chain and medium-chain FAs are preferentially released and such hydrolysed milkfat products exert antimicrobial effects in vitro [1114]. Recent research has revealed that H. pylori is susceptible to medium-chain MGs and FAs in vitro [15]. From a combination of these observations we can speculate that the FAs and MGs released from milkfat in the stomach might have been agents which affected the survival of H. pylori in patients given the milk-diet treatment.

In the present study, selected FAs and MGs have been tested for their bactericidal effects against H. pylori. Concentration profiles of the most potent FAs and MGs have been investigated in short-term (40 min) incubations, to determine the concentrations required to kill 4 log or more of bacteria. The bactericidal activities of combinations of FAs and MGs have also been explored. As the pH of the stomach may vary from acidic to just below neutral, especially in the presence of urea and urease [16], the effect of pH was examined and measured before and after experiments. The effect of urea on the bactericidal effects of FAs was also examined.

2 Materials and methods

2.1 Bacterial strains and culture conditions

A widely studied isolate of H. pylori (CCUG 17874, NCTC 11637) was mainly used to test the antibacterial properties of FAs and MGs in this research. The Sydney strain of H. pylori (strain SS1), a pathogenic strain often used in in vivo mouse experiments [17], was used in some experiments to check for possible strain variation in susceptibility to FAs and MGs. Stock cultures were stored in 50% (v/v) L-broth (1% tryptone, 0.5% yeast extract, 0.5% NaCl) and 50% (v/v) glycerol at −75°C.

H. pylori was grown routinely on Columbia agar (Oxoid, Basingstoke, UK) containing 5% horse serum (Life Technologies, Auckland, New Zealand) at 37°C in an anaerobic jar gassed with 85% N2, 5% O2 and 10% CO2. Organisms were transferred to new plates every 2 days, but were discarded after subculturing for 2 weeks. Cells from the frozen source were then used. Iso-Sensitest broth (Oxoid) containing 5% horse serum was used for broth culture of H. pylori.

2.2 Chemicals

The saturated FAs screened for antibacterial activity against H. pylori were butyric acid (C4:0), caproic acid (C6:0), caprylic acid (C8:0), capric acid (C10:0), lauric acid (C12:0), myristic acid (C14:0), and palmitic acid (C16:0). The unsaturated FAs used were myristoleic acid (C14:1, n-9), palmitoleic acid (C16:1, n-9), oleic acid (C18:1, n-9), linoleic acid (C18:2, n-9,12), and linolenic acid (C18:3, n-9,12,15), all in the cis form. The MGs used were 1-monocaprylin (MG C8:0), 1-monocaprin (MG C10:0), 1-monolaurin (MG C12:0), 1-monomyristin (MG C14:0), 1-monopalmitin (MG C16:0), and 1-monostearin (MG C18:0). They were purchased from Sigma Chemical Co. (St. Louis, MO, USA).

The stock solutions of FAs and MGs were prepared in Milli-Q water, and autoclaved (except for the stock solutions of unsaturated FAs). The stock solutions of C4:0, C6:0 and C8:0 were 300 mM, while stocks for the other saturated FAs and MGs (which have low solubility in water) were 50 mM. Stock solutions of unsaturated FAs were 100 mM. Unsaturated FAs are not stable at high temperature or in aqueous solution, so their stock solutions were freshly prepared in sterile Milli-Q water, using sterile glassware, immediately before use.

Before aliquots of the stock lipids were added to broth, the stocks were left in a waterbath at 70°C for 10 min (except for the stocks of unsaturated FAs), then sonicated for 1 min to ensure homogeneous suspension. The sonicator probe was pre-sterilised by washing with alcohol and flaming between sonications.

Urea (Ultra Pure™ enzyme grade from Gibco BRL, Life Technologies) was freshly dissolved in water just before use (2 M stock solution) and sterilised by filtration through a 0.2-µm membrane, because its breakdown product cyanate is toxic to bacteria.

2.3 Measurement of bactericidal action of FAs and MGs

H. pylori, grown 2 days on Columbia agar, was suspended in Iso-Sensitest broth at approximately 109 cells ml−1 using a glass spreader. Loosely capped glass tubes, containing 3.96 ml Iso-Sensitest broth, 5% horse serum and FAs and/or MGs, were inoculated with 0.04 ml of this suspension, giving an initial inoculum of 106–107 cells ml−1. Tubes were placed in an anaerobic jar, rapidly gassed with 85% N2, 5% O2 and 10% CO2, and rotary shaken (250 rpm) at 37°C for 40 min. Bacterial numbers (both initial and final) were determined after dilution in Iso-Sensitest broth by plate counting in triplicate on agar plates, after growth for 4 days. In experiments examining the effect of pH on the bactericidal potency of FAs and MGs, the pH of the Iso-Sensitest broth containing horse serum was carefully adjusted and the pHs before and after incubation recorded.

2.4 Effect of urea on the bactericidal action of FAs

A defence mechanism employed by H. pylori to protect itself from stomach acidity is hydrolysis of endogenous urea [18]. The urea level in normal human plasma is 1–5 mM [16]. Experiments to examine the protection afforded H. pylori by urea were carried out in Iso-Sensitest broth containing 5% (v/v) horse serum, 5 mM urea and FA. The bacteria were incubated in this medium for 40 min as above.

3 Results

3.1 Effect of FA and MG concentration on H. pylori survival

When H. pylori was incubated for 40 min in broths containing a range of FA concentrations, the relationship between bacterial survival and FA concentration was not linear. There is an initial low concentration range within which FA has no effect on the bacterium. Then over a small increase in concentration range, there is an abrupt change from no effect to complete killing. Fig. 1A,B,C and D show such bactericidal profiles for FAs C10:0, C12:0, C14:0 and C18:3 over selected concentration ranges. Fig. 1E,F show similarly shaped profiles for MGs C12:0 and C14:0. The concentration range over which different FAs and MGs change from being ineffective to completely bactericidal is not the same for different acyl chain lengths.

Figure 1

The effect of concentration of FAs and MGs on survival of H. pylori. Incubations were carried out in Iso-Sensitest broth containing 5% horse serum, for 40 min at 37°C, pH 7.4, in the presence of varying concentrations of (A) FA C10:0, (B) FA C12:0, (C) FA C14:0, (D) FA C18:3, (E) MG C12:0 and (F) MG C14:0. Initial cell concentrations were between 106 and 107 cfu ml−1.

3.2 Minimal bactericidal concentrations (MBCs) of FAs and MGs against H. pylori

To compare the bactericidal potencies of a range of FAs and MGs, a standard endpoint for measurement of killing is needed. The MBC will be defined as the minimum concentration of the bactericidal agent required to kill 4 log (99.99%) of the bacteria during a standard 40-min incubation. Since the majority of H. pylori in the stomach occupy a neutral pH niche within the mucus layer, initial experiments were carried out at pH 7.4. Table 1 details the bactericidal concentrations of FAs and MGs needed to kill 50% (the BC50%) and 99.99% (the MBC) of H. pylori cells in 40 min at pH 7.4. The short-chain FAs, C4:0, C6:0 and C8:0, had no effect on H. pylori survival at concentrations ≤20 mM. The longer FAs, C10:0, C12:0, C14:0, C14:1, C16:1, C18:3, and MGs, MG C12:0, MG C14:0, and MG C16:0, were tested over the concentration range of 0.1–10 mM. The most potent FAs were C12:0 and C16:1 which had MBCs of <1 mM, and C18:3 with a MBC of <0.5 mM. The most potent MGs were MG C12:0 and MG C14:0, with MBCs of <0.5 mM and 0.9 mM, respectively. It was striking that the difference between the concentrations of an agent needed to kill 50% as compared with 99.99% of cells was quite small. For example, 0.6 mM FA C12:0 killed 50% of cells, while 1 mM killed 6 log of cells.

View this table:
Table 1

Bactericidal potencies of FAs and MGs against H. pylori

FA/MGBC50% (mM)Change in cell numbers (log10 cfu ml−1)MBC (mM)Change in cell numbers (log10 cfu ml−1)
Butyric acid (FA C4:0)>300NDND
Caproic acid (FA C6:0)>300NDND
Caprylic acid (FA C8:0)>200NDND
Capric acid (FA C10:0)2.5−1.21±0.024−6.05±0.11
Lauric acid (FA C12:0)0.6−0.60±0.031−6.26±0.10
Myristic acid(FA C14:0)3.0−1.85±0.115−5.17±0.10
Palmitic acid (FA C16:0)>100NDND
Myristoleic acid (FA C14:1)0.25−2.00±0.120.5−6.23±0.04
Palmitoleic acid (FA C16:1)0.25−1.20±0.121−6.40±0.05
Oleic acid (FA C18:1)1.0−0.68±0.27NDND
Linoleic acid (FA C18:2)0.5−1.52±0.13>1−2.55±0.22
Linolenic acid (FA C18:3)0.2−1.55±0.220.5−6.26±0.05
Monolaurin (MG C12:0)0.3−0.80±0.160.5−5.12±0.15
Monomyristin (MG C14:0)0.3−0.57±0.050.9−5.86±0.32
Monopalmitin (MG C16:0)1.5−0.72±0.10>2−1.95±0.18
  • ND, not determined.

  • Incubations in Iso-Sensitest broth containing 5% horse serum for 40 min at 37°C.

  • BC50% and MBC are defined in the text.

  • BC50% or MBC are not detected at the highest concentration tested, the changes in cell numbers are the data at the highest concentration of FA or MG tested.

The non-pathogenic H. pylori strain NCTC 11637 was used in the experiments described in Table 1. We anticipated that other strains of H. pylori would be equally susceptible to FAs and MGs. The Sydney strain of H. pylori was therefore tested for its susceptibility to a range of concentrations of FA C12:0 and MG C12:0. The MBC values obtained were identical to those obtained with the strain NCTC 11637 (data not shown).

3.3 Effect of pH on the bactericidal potency of caprylic acid

The stomach/duodenum will contain a range of pHs from acidic to neutral, depending on the presence or absence of food and the location within the pH gradient associated with the mucus barrier covering the mucosal cells. The solubility of FAs and their distribution between lipophilic and hydrophilic locations will be affected by their ionisation which is pH-dependent. The effect of pH on the bactericidal potency of FAs and their MG derivatives was therefore investigated.

FA C8:0 is a good model for study of pH effects because it has little effect at neutral pH (see Table 1). Experiments were carried out in Iso-Sensitest broth with 5% horse serum, adjusted to selected pHs. Survival of H. pylori (initial cell numbers 106–107 ml−1) was determined after 40 min incubation at 37°C. Fig. 2 shows the bactericidal potencies of FA C8:0 at 1 mM (pH 2.5–5.0), 10 mM (pH 4.0–5.5); and 20 mM (pH 5.0–6.5), compared with controls without FA. In the control experiments, H. pylori survived well at pHs down to 3.5, though below this pH, numbers began to decrease. The presence of 1, 10, or 20 mM FA C8:0 was bactericidal to H. pylori at pH ≤3.5, ≤5.0, and ≤5.5, respectively. Comparison of the three survival profiles shows that higher FA concentrations are needed at higher pHs for killing H. pylori. There appears to be a logarithmic relationship between the concentration of FA C8:0 and the pH range over which survival changes rapidly. Doubling the FA concentration shifts the steep part of the profile approximately 0.5 pH units higher. In addition, the profiles become steeper with increasing pH. FA C8:0 at 1 mM caused a 6 log decrease in cell survival over a 1.5-unit pH range (from pH 5.0 to 3.5), while in the presence of 10 mM or 20 mM FA C8:0, the same 6 log decrease covered only 0.5 pH units (from 5.5 to 5.0, and from 6.0 to 5.5).

Figure 2

The effect of pH on survival of H. pylori. Incubations were carried out in Iso-Sensitest broth containing 5% horse serum and FA C8:0 for 40 min at 37°C. Initial pH values are shown. ■, 1 mM FA C8:0, initial cell numbers 1.2×107 cfu ml−1. □, Control experiments without 1 mM FA C8:0, initial cell numbers 1.2×107 cfu ml−1. •, 10 mM FA C8:0, initial cell numbers 3.3×106 cfu ml−1. ◯, Control experiments without 10 mM FA C8:0, initial cell numbers 3.3×106 cfu ml−1. ▲, 20 mM FA C8:0, initial cell numbers 7.5×106 cfu ml−1. △, control experiments without 20 mM FA C8:0, initial cell numbers 7.5×106 cfu ml−1.

3.4 Effect of pH on the bactericidal potency of lauric acid and monolaurin acid

The pH profiles for survival of H. pylori after exposure to 0.5 mM and 1 mM FA C12:0 and to 0.5 mM MG C12:0 for 40 min were also studied, together with controls in the absence of FA or MG (Fig. 3). Analogous profiles of pH dependence were found for the two FA C12:0 concentrations as previously for FA C8:0, the difference being much greater FA potency as predicted from Table 1. In the presence of 1 mM FA C12:0, the pH range over which there is a steep increase in H. pylori survival is pH 8–9, while for 0.5 mM FA C12:0 the pH range is 4.5–5.0.

Figure 3

The effect of pH on survival of H. pylori in Iso-Sensitest broth with 5% horse serum containing FA C12:0 and MG C12:0. Initial pH values are shown. Incubations were for 40 min at 37°C. ▲, 1 mM FA C12:0, initial cell numbers 1.9×107 cfu ml−1. △, 0.5 mM FA C12:0, initial cell numbers 3.2×106 cfu ml−1. •, 0.5 mM MG C12:0, initial cell numbers 1.9×107 cfu ml−1. ◯, Control experiments without FA or MG, initial cell numbers 1.9×107 cfu ml−1.

In complete contrast, the bactericidal potency of 0.5 mM MG C12:0 was unaltered over the measured pH range from pH 4 to 10, and >5 log of cells were killed irrespective of the medium pH.

3.5 Effect of urea upon the bactericidal action of FAs

Urea is always present in the stomach. As the bactericidal action of FAs is pH-dependent, it is critical to determine whether the bactericidal action of FAs is suppressed by the presence of urea and endogenous urease. An experiment to test this is shown in Fig. 4. H. pylori was grown overnight at pH 5.0 in the presence of 5 mM urea. A 1 log increase in cell numbers per ml was noted, and medium pH increased from 5.0 to 6.3. In a control incubation at pH 5.0 without urea, cells numbers decreased 4 log. A 200-µl inoculum of the cell suspension that had been exposed overnight to the urea (the cells clearly possessed urease activity) was transferred to tubes containing fresh medium, pH 5.0, containing either 5 mM urea without FA or 5 mM urea plus 10 mM FA C8:0. After a 40-min incubation, cell numbers of H. pylori were maintained in the former but decreased by >5 log in the latter. Thus, the bactericidal action of FAs against H. pylori is not suppressed by urea. This result may be of significance in an in vivo situation when FAs are present, since H. pylori urease activity is understood to be an important protective factor against stomach acidity.

Figure 4

Survival of H. pylori exposed to urea and FA. Cells (7×106 cfu ml−1) were incubated overnight in broth, pH 5.0, (1) with no added urea or FA, and (2) with added 5 mM urea and no FA. A subculture of the latter incubation, initially containing 2.2×106 cfu ml−1, was incubated for 40 min at 37°C in broth containing (3) 5 mM urea, or (4) 5 mM urea plus 10 mM FA C8:0. Bars show surviving cells at the end of each incubation. ND*, no colonies detected, indicating <10 cfu ml−1.

3.6 Additive bactericidal effects between FAs and between FAs and MGs

During partial hydrolysis of dairy products and certain other food triglycerides by pre-intestinal lipase(s) in the stomach, both FAs and MGs are produced. To assess the possible bactericidal interactions of such natural products, H. pylori survival was studied in the presence of mixtures of FAs and mixtures of FA plus MG.

Table 2 shows the effect on H. pylori numbers of combinations of FAs, and combinations containing FA and MG. In experiment A, two saturated FAs, FA C10:0 (2 mM) and FA C12:0 (0.5 mM), were first tested alone at a concentration at the end of the plateau region at which they had no effect on survival (see Fig. 1). When the FAs were combined at these concentrations the H. pylori numbers decreased by 5 log. The MBC values of FA C10:0 and FA C12:0 are 4 mM and 1 mM (Table 1), respectively, which means that the bactericidal potency of FA C12:0 is four times greater than that of FA C10:0. Broth containing 2 mM FA C10:0 plus 0.5 mM FA C12:0 should therefore have a bactericidal action equivalent to either 4 mM FA C10:0 or 1 mM FA C12:0, if the actions are additive, i.e. the mixture should be completely bactericidal to H. pylori. The result shown in Table 2, experiment A, clearly shows this effect. The same analysis can be applied to the bactericidal activity of combinations of FA plus MG (Table 2, experiments B and C), in which the combined bactericidal activities (based on the fraction of MBC values) are additive but not synergistic.

View this table:
Table 2

Survival of H. pylori in broth containing different combinations of FAs and MGs

ExperimentAdditionsCell number (cfu ml−1)
0.5 mM FA C12:0(3.16±0.09)×107
2 mM FA C10:0(2.48±0.17)×107
3 mM FA C10:0(7.35±0.49)×104
0.5 mM FA C12:0+2 mM FA C10:0<100
0.5 mM FA C12:0(1.36±0.16)×107
0.3 mM MG C12:0(6.65±0.09)×106
0.1 mM MG C12:0+0.5 mM FA C12:0(7.15±0.93)×105
0.2 mM MG C12:0+0.5 mM FA C12:0<10
0.3 mM MG C12:0+0.5 mM FA C12:0<10
0.3 mM MG C14:0(9.28±0.88)×106
2 mM FA C14:0(9.30±0.49)×106
0.1 mM MG C14:0+2 mM FA C14:0(9.95±1.18)×106
0.2 mM MG C14:0+2 mM FA C14:0(2.34±0.17)×105
0.3 mM MG C14:0+2 mM FA C14:0(9.25±4.40)×103
  • Incubations for 40 min at 37°C. Initial pH of broth, pH 7.4.

  • No added FA or MG in the broth.

3.7 Test for acquired resistance to the bactericidal activity of lauric acid

Acquired resistance to bactericidal molecules, especially to antibiotics, is a major reason why H. pylori infections are difficult to eradicate. An experiment was carried out to test for evidence of resistance to the antibacterial effect of FAs (Fig. 5). FA C12:0 (1 mM) is lethal to H. pylori after 40 min incubation, while 0.7 mM results in 1 log decrease in cell numbers. Therefore, 0.7 mM FA C12:0 was chosen for testing whether resistance can develop in H. pylori to FA C12:0. Cultures grown in the absence and presence of FA C12:0 (1 mM) were tested, in parallel, as positive and negative controls.

Figure 5

Test of resistance of H. pylori to the bactericidal effect of FA C12:0 after 40 min incubation at 37°C for six cycles. The initial pH was always 7.4. White bars, control containing no added fatty acid in the broth. Light grey bars, 0.7 mM FA C12:0 in the broth. Dark grey bars, 1.0 mM FA C12:0 in the broth. Experimental details are in the text.

Two-day-old H. pylori cells (ca. 107 cfu ml−1) were inoculated into three tubes containing Iso-Sensitest broth and 5% horse serum, with addition of 0, 0.7, or 1.0 mM FA C12:0, and shaken at 250 rpm for 40 min at 37°C. Surviving cells (40 µl) from the broth containing 0.7 mM FA C12:0 were plated onto Columbia agar and grown for 2 days in the absence of FA (cycle 1). Cells were harvested from the agar, inoculated again (ca. 107 cfu ml−1) into tubes containing Iso-Sensitest broth with or without added 0.7 mM FA C12:0, and incubated for 40 min. Surviving cells from the tube containing 0.7 mM FA C12:0 were again regrown on plates (cycle 2). Cycle 2 was then repeated three more times. During a sixth cycle, cells were incubated for 40 min in broths without FA C12:0 (control), 0.7 mM FA C12:0, and 1.0 mM FA C12:0. After each cycle, cell survival was measured by plate counts. The incubations containing 0.7 mM FA C12:0 in each cycle always showed approximately 1 log cell reduction compared with the control. After six cycles, 1.0 mM FA C12:0 was still lethal to H. pylori. Thus, there was no evidence that cells resistant to FA were being selected.

4 Discussion

The bactericidal effect of saturated FAs and MGs against H. pylori has been previously documented [15]. FA C12:0 (1 mM) was found to be the most potent FA, and MG C10:0–C14:0 (1 mM) were found to decrease cell survival at pH 7 by >4 log after a 1-h incubation. The present work confirms these studies, and extends the findings.

In the present work we have studied the bactericidal potency of a series of saturated FAs (C4:0–C16:0) and unsaturated FAs (C14:1–C18:1, C18:2 and C18:3). Using a standard incubation as a measure of potency at pH 7.4, the MBC for FAs and MGs was determined. FA C12:0 was the most bactericidal saturated FA, and potency decreased rapidly as the chain length increased or decreased (Table 1). Somewhat surprisingly, certain longer unsaturated FAs were even more potent than FA C12:0, and this trend is further discussed below.

The potency of FA effect was pH-dependent. In contrast to the findings of Petschow et al. [15], who used 1 mM FA C12:0 and 1 mM MG C12:0 in their pH studies between pH 3 and 7, we found a dramatic effect of pH on potency of FAs. A sharp transition in potency, from no effect to complete killing, occurred during a small decrease in pH while the FA concentration was unchanged. We have previously suggested this change in potency is related to the availability of un-ionised fatty acid, based on an analysis of bactericidal potency against ionisation state [14]. In contrast to FAs, bactericidal potency of MGs did not show any variation over a broad pH range, and these molecules remain un-ionised.

The profile of bactericidal potency of FAs and MGs versus concentration showed an initial range over which there was no effect on cell survival. Then, at a concentration that was related to acyl chain length, each FA or MG became very potent within a small further increase in concentration. We theorise this is due to the sudden overwhelming of a metabolic process which is critical for viability. MBCs were used as a measure of the potency of each FA or MG at a defined pH. It was found that the fractions of the MBC concentration of each FA and MG were additive when more than one antibacterial compound was present in an incubation, i.e. two sub-bactericidal concentrations of a mixture of FAs and MGs could be added, neither having an effect on its own, and the potency of the mixture could be calculated as the sum of the fractional MBCs.

It was somewhat surprising to find that unsaturated FAs also formed a series with predictable MBCs, but different from the saturated FA series. The potency of FAs (saturated and unsaturated) appeared to be related to the calculated equivalent carbon number (ECN). This property is often used during chromatographic analysis of lipids, and the triacylglycerol molecular species are identified by matching the calculated ECN to those of known reference triacylglycerols. Similar ECNs imply similar retention times caused by similar molecule size, shape, charge, or polarity. According to Firestone [19], ECN is defined as: Embedded Image where ECN is the equivalent carbon number, CN is the number of carbon atoms in the fatty acids, and db is the number of double bonds. Thus the ECNs for FA C12:0, MG C12:0, FA C14:1 and FA C18:3, are all equal to 12. As documented in Table 1, FA C12:0 was the most potent saturated FA, and FA C14:1 and FA C18:3 the most potent unsaturated FAs, and MG C12:0 was the most potent MG. As in chromatography, it seems that the size or shape of the FAs and MGs may also be critical in determining bactericidal properties. It can be predicted that arachidonic acid, FA C20:4, with an ECN of 12 should have a similar bactericidal potency to FA C18:3. Research by Knapp and Melly [20] on the antibacterial activity of FA C20:4 against Gram-positive bacteria, and by Hazell and Graham [21] on FA C20:4 toxicity to H. pylori supports this hypothesis. ECN analysis might also explain why MG C12:1 (ECN=10) is much less potent than MG C12:0 against H. pylori[15]. This same principle might also be used to predict antibacterial applications of other lipid agents.

One of the primary survival mechanisms used by H. pylori against stomach acidity is protection by urease [22]. If FA is affecting the bacterium by a mechanism involving acidification of the cell cytoplasm, then it might be expected that the presence of urea plus endogenous urease would reverse the effect of FA. We obtained no evidence that urea addition was able to prevent the acute bactericidal effect of FA. This observation may prove useful if FAs are, in future, to be useful in vivo, in prevention of H. pylori colonisation or eradication.

Petschow et al. [15] have looked for evidence of resistance to FAs in H. pylori, but found no resistant colonies. In the present work a different approach was adopted, by exposing a culture to six cycles of exposure to concentrations of FA that killed approximately 90% of cells. No evidence for selection of FA-resistant cells was found, suggesting that the FA target was not dependent on a single protein or metabolic step that could be mutated to prevent the effect, followed by selection. The genome of H. pylori lacks genes encoding FA oxidative catabolic pathways [23], and it is unclear how or if H. pylori could remove FAs if they enter the cytoplasm and form anions.

The mechanism by which FAs and MGs kill H. pylori and other bacteria is unclear. The similar potencies between FAs and MGs with comparable ECNs, the additive bactericidal effects, and close values of the MBCs suggest that there may be a common denominator between the effects of FAs and MGs. However, their effects may all relate to physical properties involving rates of permeability. The observation that FA effects are pH-dependent, while those of MG are not, is consistent with permeability being dependent on physical properties.

Cell membranes have long been regarded as a primary target for the antimicrobial effect of FAs [24]. Greenway et al. [25] suggested that FA C18:2 may have a surfactant action, increasing membrane permeability. Others have suggested autooxidation products of unsaturated FAs could be toxic to bacteria [26,27]. Catalase protects against the bactericidal effect of unsaturated FAs, suggesting a peroxidative process involving H2O2[20].

Sun et al. [14] suggested that short- and medium-chain FAs diffuse into bacterial cells in their acidic form, resulting in decreased intracellular pH, and if there is an energy-linked transport system re-exporting the FA anions, this will reach a critical point where it is overwhelmed. This hypothesis is consistent with the observations of Viegas et al. [28], who showed that the plasma membrane ATPase of Saccharomyces cerevisiae at pH 4 becomes more active in the presence of FA C8:0, but there was little difference at pH 6.5. An alternative hypothesis, related to the effect of a disturbed intracellular pH, might be an effect on pH-sensitive intracellular enzymes [28] or inhibition of amino acid transport into cells [24].

The mechanism by which MGs are bactericidal is less studied. MGs are good non-ionic surfactants and can form micelles in the medium, a process which may make them more dynamic and able to penetrate the cell membrane. Once MGs penetrate, they might be incorporated into the lipid membrane, thereby altering the permeability, as suggested above, as one possible mechanism for killing by FAs.

The results in this research provide a possible strategy for a new treatment for H. pylori infection that would be characterised by minimal side-effects and low bacterial resistance. FAs and MGs can be generated from triglycerides in certain foods by pre-intestinal lipases contained in salivary and stomach secretions. Bovine milk is an example of a triglyceride source that is partially broken down in the stomach lumen, giving a mixture of FAs and MGs that will have bactericidal activity, and composition and concentrations that should be in excess of those needed to kill H. pylori in our 40 min duration in vitro assays (Sun, unpublished data). The pH of the contents will be also critical.

Clearly the mucus layer, in which H. pylori are mainly located, is another niche that needs to be considered separately from the lumen. There is a steep pH gradient across the mucus barrier from acidic at the lumen interface to neutral at the mucosal membrane, due to the bicarbonate secreted by epithelial cells. Little information is available on rates of FA and MG penetration through mucus, though it is known that little fat is absorbed in the stomach. A small fraction of H. pylori are also located in the gastric pits, and whether the outward flux of secretions would prevent free FAs and MGs from reaching bacteria in this location is not easy to assess.


We would like to acknowledge financial support for this work from the New Zealand Dairy Board and the University of Auckland Research Committee. Also, we wish to thank Dr A.K.H. MacGibbon for his help and discussions.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
View Abstract