Variation of Branched-Chain Fatty Acids Marks the Normal Physiological Range for Growth in Listeria monocytogenes

The fatty acid composition of Listeria monocytogenes Scott Awas determined by close-interval sampling over the entire biokinetictemperature range. There was a high degree of variation in thepercentage of branched-chain fatty acids at any given temperature.The percentage of branched C₁₇ components increased with growthtemperature in a linear manner. However, the percentages ofiso-C_15:0 (i15:0) and anteiso-C_15:0 (a15:0) were well describedby third-order and second-order polynomial curves, respectively.There were specific temperature regions where the proportionof branched-chain fatty acids deviated significantly from thetrend established over the entire growth range. In the regionfrom 12 to 13°C there were significant deviations in thepercentages of both i15:0 and a15:0 together with a suggesteddeviation in a17:0, resulting in a significant change in thetotal branched-chain fatty acids. In the 31 to 33°C regionthe percentage of total branched-chain components exhibiteda significant deviation. The observed perturbations in fattyacid composition occurred near the estimated boundaries of thenormal physiological range for growth.

INTRODUCTION

Top

Introduction

Listeria monocytogenes is a food-borne pathogen responsiblefor numerous outbreaks of disease in North America, Europe,and Australia linked to the consumption of commercial food products(9, 15). The ability of L. monocytogenes to grow at refrigerationtemperatures has led to concerns about infection from cold-storedfoods, particularly those consumed without subsequent cooking(i.e., ready-to-eat foods) (11). While low levels of L. monocytogenescontamination are not uncommon in such foods, outbreaks of diseasehave occurred in cases where the bacterium has proliferatedin foods stored at low temperatures over time (15).

A major adaptation of L. monocytogenes to growth at low temperatureis the significant proportion of branched-chain fatty acidspresent in the cell membrane. In particular, anteiso-C_15:0 (a15:0)was demonstrated to be essential for growth at low temperatures(3, 10). L. monocytogenes is also unusual in its temperaturerange for growth, which is toward the upper end of the rangeof temperatures reported for bacteria (-2 to 45°C [approximately48°C]) (21).

Nichols et al. (17) reported that they found a correlation betweenbranched-chain fatty acid content and the normal physiologicalrange (NPR) for growth of the psychrophilic bacterium Shewanellagelidimarina by using fine-scale determinations over the entirebiokinetic temperature range. The objective of this study wasto investigate the phenotypic regulation of branched-chain fattyacids in L. monocytogenes by using the same methodology.

MATERIALS AND METHODS

Top

Materials and Methods

Bacterial strains.
L. monocytogenes Scott A was maintained on slopes of tryptonesoya broth (catalog no. CM 129; Oxoid) with 0.6% yeast extract(catalog no. L21; Oxoid).

Temperature gradient incubator experiments.
The effect of temperature on the rate of growth of L. monocytogeneswas determined by using a temperature gradient incubator (ToyoKagaku Sangyo, Tokyo, Japan). Four replicate experiments wereconducted. The growth medium was tryptone soya broth with 0.6%yeast extract. Tubes were inoculated with an actively growingculture, and growth was monitored by measuring the percentageof transmittance at a wavelength of 540 nm until growth wascomplete. Cultures were shaken by the oscillatory motion ofthe incubator through an arc of 60° at a rate of 60 oscillationsmin^-1. The growth rate at each temperature was calculated byfitting a modified Gompertz function to the data (14). The growthtemperature of each incubation tube was determined in triplicatewith a Fluke 51 K/J thermometer and Fluke 80PK-1 thermocouplefollowing cessation of growth. The calculated growth rate datasets were then fitted to the modified four-parameter squareroot model of Ratkowsky et al. (20) by using UltraFit software.Growth rate data were also fitted to the model of Rosso et al.(22) in order to obtain optimal growth temperature (T_OPT) estimates.This model contains T_OPT as a parameter and thereby allows standarderrors to be determined. The two models were found to give similarresults for other parameters.

NPR.
The NPR for growth was estimated by expressing the growth ratedata as an Arrhenius plot. The midpoint of the data set wasdetermined, individual data points were sequentially added inboth directions, and a linear function was fitted. The temperaturepoints whose inclusion resulted in the r² value of the fittedfunction falling below 0.990 were designated the boundariesof the NPR.

Lipid extraction and analysis.
When individual incubation tubes reached a transmittance valueof 27 to 30%, corresponding to the mid-exponential phase, thecultures were immediately harvested by filtration onto glassfiber filters (Whatman) which had been previously heated at400°C for 24 h. Sample filters were then frozen at -20°Cuntil lipid analysis. Filters were extracted by a direct transesterificationprocedure (7) demonstrated to yield fatty acid profiles comparableto those obtained by the Bligh-Dyer solvent extraction method(5, 24). Briefly, filters were added to screw-cap test tubescontaining 3 ml of methanol-chloroform-hydrochloric acid (10:1:1,vol/vol/vol). The tubes were heated at 80°C for 1 h beforethey were cooled to room temperature. After addition of Milli-Qwater (1 ml), the resultant fatty acid methyl esters were extractedthree times with 1.5 ml of hexane-chloroform (4:1, vol/vol).

Fatty acid methyl esters were analyzed by using a Hewlett-Packard5890 II gas chromatograph and 5970A mass selective detectorequipped with a cross-linked methyl silicone (film thickness,0.33 µm) fused-silica capillary column (length, 50 m;internal diameter, 0.22 mm). The operating conditions were similarto those detailed by Nichols et al. (17). Fatty acid methylesters from all samples were identified by comparing the componentspectra to the spectra of known standards. The double bond positionsand geometry in monounsaturated isomers of selected sampleswere determined by producing and analyzing dimethyl disulfideadducts (18). The data presented below are the means and standarddeviations for four complete replicate experiments in whichthe standard deviation of the temperature values was less that0.5°C. Significant differences between adjacent mean valueswere assessed by a Student's t test. The significance of outlyingdata points was assessed by analysis of standardized residualsfor each data point compared to the fitted trends (see Fig.3).

View larger version (37K):
[in this window]
[in a new window]
FIG. 3. Plots of L. monocytogenes fatty acid composition for cultures grown over the entire biokinetic temperature range in tryptone soya broth with yeast extract. (a) Total branched-chain fatty acids; (b) i15:0 fitted with a third-order polynomial curve (r² = 0.93); (c) a15:0 fitted with a second-order polynomial curve (r² = 0.95); (d) i17:0 fitted with a third-order polynomial curve (r² = 0.99); (e) a17:0 fitted with a third-order polynomial curve (r² = 0.96). The 12.5 and 31.3°C data points are highlighted ( ${diamondsuit}$ ). The data points represent the means of four replicates, and the bars indicate standard deviations.

RESULTS

Top

Results

Growth rate.
L. monocytogenes had a theoretical growth range of 46.5°Cunder the experimental conditions employed (Fig. 1). The fittedtheoretical minimum growth temperature and maximum growth temperaturewere 1.5 ± 0.5 and 48.0 ± 0.2°C, respectively,and the estimated T_OPT was 36.6 ± 0.3°C. The variationin the growth rate between data sets (as assessed by standarddeviation) increased immediately preceding and following theT_OPT (Fig. 1).

View larger version (16K):
[in this window]
[in a new window]
FIG. 1. Plot of growth rate (1generation time) versus temperature for L. monocytogenes grown in tryptone soya broth with yeast extract. The data points represent the means of four replicates, and the bars indicate standard errors.

NPR.
An estimate of the NPR was obtained by defining the linear portionof the Arrhenius plot (Fig. 2). Growth rate data, as an additiveseries, in the temperature range from 14.0 to 26.1°C werewell described by a linear function (r² $>=$ 0.996). Inclusion ofdata points outside this range resulted in a decrease in r²(r² < 0.990). The boundaries of the NPR were therefore definedas 12.7 to 14.0 and 26.1 to 27.7°C.

View larger version (13K):
[in this window]
[in a new window]
FIG. 2. Relationship of growth rate and temperature for L. monocytogenes expressed as an Arrhenius plot. The data points represent the means of four replicates and are the same as the data points shown in Fig. 1. The data for the temperature range from 14.0 to 26.1°C (•) are well described by a linear function as an additive series (r² $>=$ 0.996). Inclusion of the other data ( ${diamond}$ ) in the series decreased the goodness of fit (r² < 0.990).

Fatty acid composition.
There was a high degree of variation (as assessed by standarddeviation) in the percentage of branched-chain components betweenreplicate experiments at a given growth temperature (Fig. 3).The variation was particularly marked when the proportion oftotal branched-chain fatty acids was considered. While therewas no trend discernible from the percentage of total branched-chainfatty acids, two significant alterations in composition wereevident for three temperature regions, 11.2 to 12.5°C (P< 0.01), 12.5 to 13.9°C (P < 0.001), and 31.3 to 32.8°C(P < 0.01) (Fig. 3).

These alterations were due primarily to the contributions ofthe two C₁₅ branched-chain components. The proportion of a15:0decreased as the growth temperature increased, and the trendwas well described by a second-order polynomial curve. The percentageof i15:0 was maximal at 17°C and minimal at 6 and 40°Cand was well described by a third-order polynomial curve (Fig.3). One data point (the 12.5°C data point) differed significantlyfrom the values on the fitted curves for both the i15:0 (P <0.05) and a15:0 (P < 0.01) components.

The percentages of both i17:0 and a17:0 increased with growthtemperature in a generally linear manner (Fig. 3), and i17:0exhibited much less variation than the other branched-chaincomponents. There was no significant (P > 0.05) deviationof data points from the values on the fitted curves for thesedata sets. However, for a17:0, the largest standardized residualof the data set occurred at 12.5°C.

DISCUSSION

Top

Discussion

When the fatty acid composition of L. monocytogenes Scott Awas examined, there were specific temperature regions wherethe proportion of C₁₅ branched-chain components of the fattyacids deviated significantly from the trend established overthe entire biokinetic growth range. In the region from 12 to13°C there were significant deviations in both the percentageof i15:0 and the percentage of a15:0 together with a suggesteddeviation in a17:0, resulting in a significant change in thepercentage of total branched-chain fatty acids. In the 31 to33°C region the total percentage of branched-chain componentschanged significantly. These observed perturbations in fattyacid composition correlated with the estimated boundaries ofthe NPR as determined from the linear region of the Arrheniusgrowth rate plot. The goodness of fit of the fitted curves andthe low standard deviation of the i17:0 data (Fig. 3) suggestthat the variation observed in the proportions of other fattyacids was not a function of the experimental procedures.

To date, the most detailed study of L. monocytogenes fatty acidcomposition in relation to temperature was performed by Annouset al. (3). These workers reported fatty acid data for sevengrowth temperatures, two supraoptimal (42 and 45°C), oneoptimal (37°C), and four suboptimal (30, 20, 10, and 5°C).The reported trends in fatty acid composition were similar tothose observed in the present study. In addition, the data ofAnnous et al. (3) illustrated that there were differences inthe proportion of branched-chain fatty acids between strainsand that medium composition had a marked effect on the proportionof a15:0 at low temperature for one strain but not for anotherstrain. However, reliance on single culture determinations didnot allow an assessment of reproducibility to be made. Someof the observations described by Annous et al. (3) may havebeen due to the inherent variation in fatty acid compositionhighlighted by this study. Future examinations of L. monocytogeneslipid composition will be required to consider this phenomenon.

The NPR has been described as the linear portion of the Arrheniusrelationship for bacterial growth, and the boundaries of theNPR are defined at both high and low temperatures by deviationin the bacterial growth rate response from Arrhenius kinetics(16). However, the description of such a linear section maybe subjective (14), and some authors argue that there are multiplelinear regions in the growth rate response of L. monocytogenes(4). In this study we employed an independent method of linearregression to assess the NPR region of the Arrhenius relationshipfor L. monocytogenes. The estimates of the resultant boundaries(12.7 and 27.7°C) may be considered surprising in relationto the broad cardinal temperatures of this bacterium.

Similar perturbations in the proportion of branched-chain fattyacids associated with the NPR boundaries of S. gelidimarinawere revealed by close-interval sampling throughout the entirebiokinetic temperature range for growth (17). It was suggestedthat different primer specificities of ß-ketoacyl-acylcarrier protein (ß-ketoacyl-ACP) synthase enzymesmay be responsible. In the case of Listeria it is also probablethat the provision of branched-chain primer molecules playsa role. Branched-chain fatty acids are biosynthesized by thefatty acid synthase system with specific branched acyl primers.For example, a15:0 is synthesized from 2-methylbutyryl coenzyneA (2-methylbutyryl-CoA) as a chain primer (12). In Listeria2-methylbutyryl-CoA is provided by branched-chain amino acidmetabolism (2, 3) (specifically, by diversion of the 2-keto-3-methylvalerateintermediate). Decarboxylation of 2-keto-3-methylvalerate bya branched-chain keto acid decarboxylase (19) yields the 2-methylbutyryl-CoAprimer for a15:0 production by fatty acid synthase. However,Listeria is deficient in branched-chain amino acid anabolismand relies upon exogenous isoleucine for growth (3). In thiscase the 2-methylbutyryl-CoA used as a chain primer for thebiosynthesis of a15:0 must be provided from the transaminationof exogenous isoleucine by branched-chain amino acid aminotransferase(BcaT) (2, 3). Thus, two additional enzymes (BcaT and branched-chainketo acid decarboxylase) are critical for branched-chain fattyacid biosynthesis in Listeria and represent two other sitesof temperature regulation which may provide a mechanism forthe phenomena observed.

The important role of a15:0 in the low-temperature growth ofL. monocytogenes has been highlighted previously (3, 8). Thisstudy revealed that the low-temperature variation observed inthe percentage of a15:0 at 12.5°C resulted in a proportionequal to that found at a much higher growth temperature (ca.23°C) (Fig. 3). While this phenomenon did not affect thegrowth rate (Fig. 1), it may point to the possibility of increasedsusceptibility to other membrane-perturbing effects at thistemperature. Nisin, a polypeptide bacteriocin, is known to actby disruption of the cytoplasmic membrane (1, 6). While nisinresistance is described as a complex phenotype (6), resistantmutants have been reported to contain a lower proportion ofbranched-chain fatty acids and to differ in the proportion ofphospholipid classes (6, 13, 23). Hence, the observed perturbationsin L. monocytogenes fatty acid composition near the NPR boundariesare similar to those associated with nisin resistance.

Close-interval temperature sampling of L. monocytogenes fattyacid composition revealed significant patterns in the proportionsof branched-chain components over the entire growth temperaturerange. The perturbations associated with the estimated boundariesof the NPR were similar to those described for the Antarcticpsychrophile S. gelidimarina, which also contains an appreciableproportion of branched-chain fatty acids. This phenomenon maybe related to the specific activity profile of enzymes associatedwith branched-chain fatty acid biosynthesis.

ACKNOWLEDGMENTS

This work was supported by the Australian Research Council togetherwith the Meat and Livestock Australia Strategic Food SafetyResearch Scheme.

David Ratkowsky is thanked for advice concerning statisticalanalysis.

FOOTNOTES

* Corresponding author. Mailing address: Centre for Food Safety and Quality, School of Agricultural Science, University of Tasmania, GPO Box 252-54, Hobart, Tasmania 7001, Australia. Phone: 61 3 62 261831. Fax: 61 3 62 262642. E-mail: D.Nichols{at}utas.edu.au.

REFERENCES

Top