The adaptive responses of the bacterium Acinetobacter calcoaceticus to different aliphatic alcohols on the level of the membrane fatty acids were studied in detail. The toxicity of the aliphatic alcohols increased with an increasing hydrophobicity. As alcohols are known to increase the fluidity of the membrane they consequently should cause the same adaptive effect on membrane level. Yet, cells of A. calcoaceticus react completely different to the alcohols: in the presence of long-chained alcohols they increase their degree of saturation, while in the presence of short-chained alcohols they decrease the degree of saturation. So, there are no observable differences in the adaptive responses of bacteria with the so-called anaerobic pathway, like Escherichia coli and Pseudomonas putida, and the bacterium carrying the so-called aerobic pathway like A. calcoaceticus. These results strongly indicate a physico-chemical difference in the membrane effect of both the partitioning and localisation of the different alcohols into the membrane and the membrane adaptive responses of the bacteria to these effects.
Micro-organisms are able to adapt to the presence of toxic organic solvents, including aliphatic alcohols, using a whole cascade of adaptive mechanisms (for review: see [1–3]). Those leading to a modification of the membrane to keep it in the same fluidity condition have been described most intensively. Changes in the fatty acid composition of membrane lipids are the most important reactions of bacteria against membrane active substances. This mechanism is called ‘homeoviscous adaptation’ [4,5].
Also aliphatic alcohols increase the fluidity of the membrane and consequently should have the same effect on membranes [4,6,7]. However, it has been observed that bacteria like Escherichia coli and Pseudomonas putida react to the presence of long-chained alcohols with an increase of their degree of saturation [8,9], while in the presence of short-chained alcohols, like i.e. ethanol, the degree of saturation decreases . This behaviour towards alcohols is very puzzling and difficult to explain as the bacteria react to an increase in their membrane fluidity by decreasing the degree of saturation of their membrane fatty acids which causes an additional increase in the membrane fluidity.
For this phenomenon two different theories occur: one explanation is related to the physico-chemical properties of short-chained alcohols, which can only penetrate slightly into the hydrophobic centre of the phospholipid bilayer and therefore only cause a swelling effect on the hydrophilic headgroups . To counteract this effect, the insertion of unsaturated fatty seems to be the better reaction against those short-chained alcohols. In contrast, long-chained alcohols that penetrate deeply into the membrane and behave like i.e. aromatic solvents and therefore cause an increase in the degree of saturation [10,11].
Another explanation of this phenomenon is a more biochemical one. As short-chained alcohols are also present in high concentrations in the cytoplasm, they are known to inhibit the key enzyme of the regulation of the biosynthesis of saturated fatty acids in the so-called anaerobic biosynthesis of fatty acids, the cytoplasmic β-hydroxydecanoyl-ACP-dehydratase (HDD) that is present in E. coli and P. putida[12,13]. Due to this inhibition, the cells might be hindered to synthesise saturated fatty acids in the presence of short-chained alcohols [14,15]. Yet, up to now it could not be verified which of those two theories is the right one.
Both theories can now be tested in an elegant way by investigating the adaptive responses of different aliphatic alcohols in a bacterium that synthesises fatty acids by the so-called aerobic pathway, using desaturases . As this pathway does not include the HDD, that is known to be effected by short-chained alcohols, it can easily be tested which of the two theories is the correct one.
In the present paper, we studied the effects of a range of aliphatic alcohols on growth and fatty acid composition of Acinetobacter calcoaceticus, a bacterium only containing the aerobic pathway of fatty acid biosynthesis in which very recently the desaturase activity was proven . Thereby, we could find an adaptive behaviour pattern very familiar to that previously described for bacteria containing the anaerobic pathway and therefore strongly indicating that the adaptive responses towards aliphatic alcohols are caused by the physico-chemical properties of these chemicals.
Materials and methods
Strain and chemicals
A. calcoaceticus was taken from the type culture collection of Instituto Superior Técnico, Lisbon, Portugal. All chemicals were reagent grade and obtained from commercial sources.
A. calcoaceticus was cultivated in a mineral medium as described by Hartmans et al.  with 4 g l−1 Na2-succinate as sole carbon source. Cells were grown in 50 ml shake cultures in a horizontally shaking water bath at 30°C. Cell growth was monitored by the turbidity (optical density) at 560 nm (OD560). A 2-ml inoculum from an overnight culture was transferred to 50 ml fresh medium and cells were grown exponentially during 3–4 h (until an OD560 of 0.6).
Incubation with aliphatic alcohols
Aliphatic alcohols with a chain length from C1 to C10 were added to the cells in the exponential growth phase. The concentration range of each alcohol leading to moderate and complete inhibition of growth, respectively, was estimated according the correlation between toxicity and hydrophobicity (logP value) described by Heipieper et al.  and Hage et al. . The cultures were then incubated in the presence of octanol for another 3 h in a shaking water bath at 30°C before the cells were harvested, washed twice with potassium phosphate buffer (50 mM, pH 7.0) and stored at −20°C.
Lipid extraction, transesterification, and fatty acid analysis
The lipids were extracted with chloroform/methanol/water as described by Bligh and Dyer . Fatty acid methyl esters (FAME) were prepared by a 15-min incubation at 95°C in boron trifluoride/methanol using the method of Morrison and Smith . FAME were extracted with hexane.
Analysis of fatty acid composition by gas chromatography (GC)
Analysis of FAME in hexane was performed using quadropole GC-MS System (HP6890, HP5973, Hewlett-Packard, Palo Alto, CA, USA) equipped with a split/splitless injector. A CP-Sil 88 capillary column (Chrompack, Middelburg, The Netherlands; ID: 0.32, 30 m; 0.25 µm film) was used for the separation of the FAME. GC conditions were: injector temperature was held at 250°C. The split flow was 1:10 and carrier gas He. The temperature programme was: 80°C, 1 min isotherm, 15°C min−1 to 140°C; 4°C min−1 to 280°C. The mass spectrometry (MS) conditions were: ionisation mode EI; ionisation energy 70 eV. The peak areas of the carboxylic acids in total ion chromatograms were used to determine their relative amounts. The fatty acids were identified by GC-MS and co-injection with authentic reference compounds obtained from Suppelco (Bellefonte, PA, USA). The degree of saturation of the membrane fatty acids was defined as the ratio between the two saturated fatty acids (16:0, 18:0) and the unsaturated fatty acids (16:1Δ9cis, 18:1Δ9cis) of the extracts.
In all cases the average results of three identical experiments are shown. Thereby, the standard derivation was less than 5%.
Effects of aliphatic alcohols on growth of A. calcoaceticus
To get an overview over the toxic potential of the investigated aliphatic alcohols their hydrophobicity was determined. This is usually expressed in terms of the logP value which gives the partitioning of a compound over an octanol/water two-phase standard system [22,23]. The logP values were calculated according to the hydrophobic fragmental constants given by Rekker and de Kort . From these calculations, presented in Table 1, the possible range of toxicity was estimated according to the correlation between toxicity (50% growth inhibition) and logP received for 10 organic solvents by Heipieper et al. .
|Organic compound||logP||LD50 (mM)||Δ degree of saturation|
|Organic compound||logP||LD50 (mM)||Δ degree of saturation|
Cells were grown in a mineral medium with succinate as energy and carbon source. The growth rate µ of the cells was about 0.55 h−1, which corresponds to a doubling time (tD) of about 1 h 20 min. Toxins were added at different concentrations during the exponential growth phase. The organisms continued to grow exponentially, but at reduced growth rates. The results for eight investigated alcohols are listed in Table 1 in order of their increasing logP values.
The toxicity of the aliphatic alcohols increased with an increasing logP value. This caused a direct correlation between toxicity and hydrophobicity as shown in Fig. 1. Yet, this relation is only correct up to a logP value of about 4.5 (1-dodecanol). Alcohols with a higher logP value and an increased chain length, respectively, are no longer toxic to the cells (data not shown).
Changes in fatty acid composition of A. calcoaceticus
The main fatty acids of this strain were palmitic acid (16:0), palmitoleic acid (16:1cisΔ9), and oleic acid (18:1cisΔ9), while myristic acid (14:0), stearic acid (18:0) and the cyclopropane fatty acids C17cyc and C19cyc were present in trace amounts of between 0.5 and 2%. The occurrence of cis-vaccenic acid or oleic acid in bacterial membranes gives an indication of the biosynthesis route of unsaturated fatty acids which is used by the organisms. Therefore, the presence of oleic acid (18:1Δ9cis) in this bacterium indicates that it only contains of the aerobic pathway of fatty acid biosynthesis [12,16] whereas the absence of cis-vaccenic acid (18:1Δ11cis) proves the absence of the anaerobic pathway and HDD, respectively.
As the main adaptive mechanism on the level of membrane lipids are changes in the degree of saturation of the fatty acids, this value was calculated and presented in the figures. Fig. 2 shows detailed curves of growth inhibition and changes in the degree of saturation of membrane fatty acids in the cells caused by ethanol, 1-butanol, 1-octanol, and 1-decanol, respectively. As can also be seen from Table 1, the concentration ranges applied of these alcohols differ extremely to cause the same growth inhibiting effects. After 3 h in the presence of the different concentrations of several compounds, respectively, the cells were harvested and fatty acid composition was investigated.
Thereby, a direct correlation between non-lethal alcohol concentrations and the decrease in degree of saturation was observed. For every alcohol, the highest response occurred at a concentration that inhibited cell growth around 50%. At higher concentrations the reaction was less intensive, whereas at concentrations which totally inhibit cell growth no reaction occurred.
Cells of A. calcoaceticus react completely different to the alcohols: in the presence of long-chained alcohols like 1-octanol (Fig. 2C), and 1-decanol (Fig. 2D), they increase their degree of saturation, while in the presence of short-chained alcohols, like ethanol (Fig. 2A), and 1-butanol (Fig. 2B) they decrease the degree of saturation. So, there are no observable differences in the adaptive responses of bacteria with the anaerobic pathway, like E. coli and P. putida, and the bacterium carrying the aerobic pathway like A. calcoaceticus.
Fig. 3 shows the highest differences in the degree of saturation caused by the investigated aliphatic alcohols. Obviously, the different alcohols also caused a qualitatively different reaction in the degree of saturation of the membrane lipids. The more hydrophilic or hydrophobic, respectively, an alcohol differs from a certain chain length the more the cells react with a decrease or increase, respectively, of the degree of saturation of the membrane fatty acids.
There are no observable differences in the adaptive responses of bacteria with the anaerobic pathway, like E. coli and P. putida, and the bacterium carrying the aerobic pathway like A. calcoaceticus. Both groups of bacteria react to the presence of long-chained alcohols with an increase of their degree of saturation, while in the presence of short-chained alcohols the degree of saturation decreases [8,9]. Therefore, a biochemical explanation, namely an inhibition of HDD by short-chained alcohols, can be excluded. On the other hand, these results strongly indicate a physico-chemical difference in the membrane effect of either the partitioning and localisation of the different alcohols into the membrane and the membrane-adaptive responses of the bacteria to these effects. The increase in the content of unsaturated fatty acids, which spread the membrane with their angle in the acyl chain seems to be a counterbalance to the same effect small alcohols have on the phospholipid headgroups. In this way, the bacteria conserve the bilayer structure of their membranes in the presence of the alcohols.
Yet, the bacteria also increase the fluidity of their membrane in the presence of compounds that already cause the same effect. Phospholipids containing 18:1cis fatty acids have a transition temperature which is about 64°C lower than those containing 16:0 fatty acids .
Therefore, other mechanisms clearly must be in operation to overcome the increased membrane fluidity not only due to the presence of short-chained alcohols but also due to the resulting shift to the unsaturated fatty acids. Dombek and Ingram  found that liposomes, consisting only of phospholipids, of cells grown in presence of ethanol and therefore with a high content of unsaturated fatty acid showed an increased fluidity. However, the authors also found that intact membranes of ethanol-grown cells had a reduced fluidity . This observation would hint at an important role for the protein content of membranes in regulating fluidity. Proteins appear to compensate the fluidising effect of the increased unsaturated fatty acid content. Changes in the lipid-to-protein ratio of the membrane have also been found in E. coli cells adapted to phenol .
The exact identification and quantification of these adaptive mechanisms must be the subject of further research.
This work was partially supported by contracts No. QLK3-CT-1999-00041 and QLRT-2001-00435 of the European Commission within its Fifth Framework Programme. P.M.S. was supported by the fellowship SFRH/BPD/5693/2001 (FCT, Portugal).