OUP user menu

The cis–trans isomerase of unsaturated fatty acids in Pseudomonas and Vibrio: biochemistry, molecular biology and physiological function of a unique stress adaptive mechanism

Hermann J. Heipieper, Friedhelm Meinhardt, Ana Segura
DOI: http://dx.doi.org/10.1016/S0378-1097(03)00792-4 1-7 First published online: 1 December 2003


Isomerization of cis to trans unsaturated fatty acids is a mechanism enabling Gram-negative bacteria belonging to the genera Pseudomonas and Vibrio to adapt to several forms of environmental stress. The extent of the isomerization apparently correlates with the fluidity effects caused, i.e. by an increase in temperature or the accumulation of membrane-toxic organic compounds. Trans fatty acids are generated by direct isomerization of the respective cis configuration of the double bond without a shift of its position. The conversion of cis unsaturated fatty acids to trans is apparently instrumental in the adaptation of membrane fluidity to changing chemical or physical parameters of the cellular environment. Such an adaptive mechanism appears to be an alternative way to regulate membrane fluidity when growth is inhibited, e.g. by high concentrations of toxic substances. The cis–trans isomerase (Cti) activity is constitutively present and is located in the periplasma, it requires neither ATP nor any other cofactor such as NAD(P)H or glutathione, and it operates in the absence of de novo synthesis of lipids. Its independence from ATP is in agreement with the negative free energy of the reaction. cti encodes a polypeptide with an N-terminal hydrophobic signal sequence, which is cleaved off during or shortly after the enzyme is transported across the cytoplasmic membrane to the periplasmic space. A functional heme-binding site of the cytochrome c-type was identified in the predicted Cti polypeptide and very recently, direct evidence was obtained that isomerization does not include a transient saturation of the double bond.

  • Cis–trans isomerization
  • Adaptation
  • Membrane fatty acid
  • Toxicity
  • Pseudomonas

1 Introduction — history

In all living cells, stress due to rigorous changes in the environment affects the membranes. As a result disturbance of membrane integrity occurs and, hence, the function as a barrier, as a matrix for enzymes and as an energy transducer is compromised [1]. If countermeasures are not taken growth inhibition or even cell death may occur. The major adaptive response of the cells is to keep the fluidity of their membranes at a constant value irrespective of actual environmental conditions. Such stabilization of membrane fluidity known as ‘homeoviscous adaptation’[2] is brought about by changes in the fatty acid composition of membrane lipids, it constitutes the predominant response of bacteria to membrane-active substances or changing environmental conditions [3]. This fundamental mechanism was investigated and reported in the famous work by Ingram in the late 70s of the past century [4]. However, up to the late 80s, the cis configuration of the double bond was still considered to be the only one occurring naturally in bacterial fatty acids. The improvement of analytical cutting off techniques especially by introducing capillary columns in gas chromatography facilitated clear differentiation of related fatty acid methyl esters and a novel class of fatty acids, i.e. trans configured unsaturated fatty acids, was found in some prokaryotes [5]. The first reports of trans isomers of unsaturated fatty acids were for Vibrio and Pseudomonas [6,7] only 10 years ago. It could be demonstrated then that trans unsaturated fatty acids were synthesized in vivo from acetate in Pseudomonas atlantica [7], although, based on known biosynthetic routes of unsaturated fatty acids, there was no explanation possible how such fatty acids could be formed.

Shortly after it was demonstrated that conversion of cis to trans unsaturated fatty acids constitutes a new adaptive mechanism enabling bacteria to change their membrane fluidity in two species, i.e. in the psychrophilic bacterium Vibrio sp. strain ABE-1 in response to an increase in temperature [8,9] and in Pseudomonas putida P8 as adaptation to toxic organic compounds, such as phenols [10,11].

Our minireview summarizes the present knowledge and progress as to the state of the subject putting emphasis on a rather efficient and elegant mechanism enabling bacteria to adapt to environmental changes which affect membrane fluidity.

2 Physiology and function of the cis–trans isomerase (Cti) of unsaturated fatty acids

Both, in Vibrio sp. strain ABE-1 and in P. putida P8, a clear increase of the normally low amount of trans unsaturated fatty acids is observed when cells are exposed to elevated temperatures or toxic phenol concentrations. Growing cells of P. putida react to phenol in a concentration-dependent manner, i.e. increase in trans and simultaneous decrease in the respective cis unsaturated fatty acids correlates with the amount of phenol accumulated in the membrane [10]. Such conversion is not dependent on growth as it also occurs in non-growing cells in which the ratio between saturated and unsaturated fatty acids and the total amount of unsaturated fatty acids cannot be changed because of the lack of lipid biosynthesis [12,13]. Consistently, the reaction takes place in cells in which fatty acid biosynthesis is inhibited by cerulenin [11,14]. Cis–trans conversion has an enzyme-like kinetics and reaches its final trans to cis ratio 30 min after addition of the membrane-toxic agents. As the rate of conversion is unaffected by chloramphenicol it was concluded that the system is constitutively present and does not require de novo protein biosynthesis [10].

Oleic acid (C18:1Δ9cis), which is normally not synthesized by P. putida P8, is, however, incorporated into membrane lipids in supplemented cultures. After the addition of a toxic 4-chlorophenol concentration oleic acid was converted into its trans isomer, i.e. elaidic acid (C18:1Δ9trans). Such a finding evidenced that trans fatty acids are synthesized by direct isomerization of cis to trans unsaturated fatty acids without shifting the position of the double bond [15]. The increase of the trans unsaturated fatty acids was accompanied by the decrease in the respective cis unsaturated fatty acid, whereas the total amount of both were kept constant at any concentration of added toxins [10]. The system does not require ATP or any other cofactor such as NAD(P)H or glutathione. Its independence from energy providing ATP is in accordance with the negative free energy of the cis to trans reaction [10,11,14].

All these data led to the proposition about cis–trans isomerization being a new adaptive response in bacteria enabling them to deal with increases in temperature or toxic concentrations of membrane disturbing compounds, conditions that otherwise would influence their membrane fluidity [9,10,16,17].

The benefit of the conversion originates from steric differences displayed by cis and trans unsaturated fatty acids. A high content of saturated fatty acids in membranes enables the acyl chains of fatty acids to form an optimal hydrophobic interaction among each other, eventually leading to a tightly packed, rigid membrane. In general, saturated fatty acids have a much higher transition temperature or melting point when compared to cis unsaturated fatty acids. Phospholipids containing 16:0 saturated fatty acids have a transition temperature which is about 63°C higher than those containing 16:1 cis unsaturated fatty acids [5,18]. The phase transition temperature of membranes increases with increasing ratios of saturated to unsaturated fatty acids. The double bond of a cis unsaturated fatty acid provokes an unmovable bend with an angle of 30° in the acyl chain. Accordingly, the highly ordered package of acyl chains in the membranes is disturbed, which in turn results in lower phase transition temperatures of such membranes [19,20]. Thus, unsaturated fatty acids in the cis configuration with bended steric structures (i.e. a kink in the acyl chain) result in a membrane with a relatively high fluidity. In marked contrast, the long extended steric structure of the trans configuration lacks the kink and is able to insert into the membrane similarly to saturated fatty acids [20].

Bacteria adapt to an increase in their membrane fluidity by increasing the degree of saturation of their phospholipid fatty acids and in some cases, changing from cis to trans the configuration of their unsaturated fatty acids. [19,21]. One major disadvantage of changes in the degree of saturation as a stress response originates from its strict dependency on cell growth and fatty acid biosynthesis. Consequently, bacteria using this mechanism are not able to perform post-biosynthetic modifications of their membrane fluidity. Indeed, it has been observed that solvents cause a shift in the ratio of saturated to unsaturated fatty acids only up to concentrations that completely inhibit growth. In the presence of higher, i.e. toxic, concentrations the cells cannot react and are thus not able to adapt to such conditions or they even die [14,22]. The isomerization of cis to trans unsaturated fatty acids found up to now only in strains of the genera Pseudomonas, including the major representatives P. putida and P. aeruginosa [5,23], and Vibrio [23] represents a solution to the problem of growth dependency as it also works in non-growing cells. Though the change from the cis to the trans unsaturated double bond does not have the same decreasing effect on membrane fluidity as a conversion to saturated fatty acids it still causes a substantial effect on the rigidity of the membrane [20,21].

After first observations mainly based on phenolic compounds, a series of organic solvents were tested for their ability to activate Cti, qualitatively and quantitatively. Accordingly, the degree of the isomerization apparently correlates with the toxicity and the concentration of organic compounds in the membrane [16,17,24]. The antimicrobial action of a solvent correlates with its hydrophobicity in a manner expressed by the logarithm of the partition coefficient of the compound in a mixture of n-octanol and water (logPow) [25,26]. Organic solvents with a logPow between 1 and 5 are highly toxic for microorganisms because they partition preferentially in membranes, where they cause an increase in membrane fluidity, finally leading to non-specific permeabilization [1,14,27]. The relationship between the logP value of a compound and its toxicity is shown in Table 1, in which 11 investigated compounds are listed according to their increasing logP values. In Fig. 1 the logP values are plotted against measured estimated concentrations that cause 50% growth inhibition (EC 50) and, simultaneously, the concentrations of the compounds causing a half-maximum increase in the trans/cis (TC 50) ratio of bacteria. Thus, there is a direct relation between the toxicity of organic solvents and their activation effects on Cti, however, this is completely independent of the chemical structures of the compounds.

View this table:
Table 1

Hydrophobicity, toxicity and effect on cis-trans isomerization of several organic compounds

Organic compoundlogPEC 50 (mM)TC 50 (mM)
  • EC 50 concentrations (50% growth inhibition) measured with P. putida cells.

  • Concentrations which caused an increase in the trans/cis ratio of unsaturated fatty acids to 50% of the maximum trans/cis level reached at saturating concentrations of the toxin.

Figure 1

Correlation between the hydrophobicity, given as the logP value of 11 different organic compounds, growth inhibition, and the trans/cis ratio of P. putida cells. Growth inhibition (●, dashed line) is presented as the EC 50 concentration and the TC 50 (◯, continuous line) is given as the concentrations which caused an increase in the trans/cis ratio of unsaturated fatty acids to 50% of the maximum trans/cis level reached at saturating concentrations of the toxin. For the names of the applied organic compounds see Table 1.

Since 1989, when a P. putida strain was discovered which grew in media containing a second phase of the generally highly toxic toluene, styrene, or xylene, several other P. putida strains have been found with similar properties [2729], and many research groups have tried to uncover the mechanisms underlying solvent tolerance. In most of these bacteria, Cti have been involved in solvent tolerance.

Not only organic solvents or increase in temperature but also some other stress elicitors were tested for their effect on Cti. In summary, all membrane affecting stimuli such as organic solvents, osmotic stress (caused by NaCl and sucrose), heavy metals, heat shock, and membrane-active antibiotics were shown to activate the system [11,3032]. However, stress conditions, such as osmotic stress caused by glycerol, cold shock, and high pH, that are known not to be activators of cellular K+-uptake — the first cellular reaction to membrane damage leading to increased permeabilization — did not cause activation of Cti [30,31]. Such findings clearly indicate that the cis/trans ratio is presumably part of a general stress response mechanism of microorganisms [14,25,33,34].

3 Biochemistry and molecular biology of Cti

Following to the physiological description of the overall function of Cti in bacteria to adapt to different stresses, molecular biological and biochemical investigations were performed to characterize this unique adaptive response system.

Based on tests of Cti activity in cell compartments [15,35] the cytoplasmic membrane was considered as the location of the enzyme where also its substrates, the phospholipid fatty acids, are present. Surprisingly, however, Cti was then purified from the periplasmic fraction of Pseudomonas oleovorans [36] and Pseudomonas sp. strain E-3 [37]. Cloning of the enzyme allowed its isolation as a His-tagged P. putida P8 protein heterologously expressed in Escherichia coli [38]. Cti is a neutral protein of 87 kDa and was shown to be monocistronically transcribed and constitutively expressed. The nucleotide sequence of the cti gene from P. putida P8 [38], P. putida DOT-T1E [39] and P. oleovorans Gpo12 [36] finally made evident that the isomerase possesses an N-terminal hydrophobic signal sequence, which is cleaved off after targeting the enzyme to the periplasmic space.

A cti knockout mutant of P. putida DOT-T1E has been constructed that is unable to isomerize cis unsaturated fatty acids. This mutant has a survival rate when shocked with 0.08% (vol/vol) toluene lower than the wild-type strain, and it displays also a longer lag phase than the parental strain when grown with toluene supplied in the gas phase [39], results that clearly implicate Cti in toluene response in this strain. However, the cis–trans isomerization is unlikely to be the only necessary adaptation mechanism to organic solvents because strains are known which can perform the isomerization and are still solvent sensitive [18,40].

Holtwick et al. [41] provided evidence that the enzyme is a cytochrome c-type protein as they could find a heme-binding site in the predicted Cti polypeptide. For an enzyme preparation from Pseudomonas sp. strain E-3, which is presumably homologous to the cti gene product of P. putida P8, it was suggested that iron (probably Fe3+) plays a crucial role in the catalytic reaction [37]. Cis–trans isomerization was found to be independent of the cardiolipin synthase, an enzyme facilitating long-term adaptation of the membrane by enhanced cardiolipin synthesis [42].

Very recently, the molecular mechanism of the isomerization reaction was elucidated. In supplementation experiments with double deuterated oleic acid it was demonstrated that oleic acid was converted exclusively into double deuterated elaidic acid after activation of Cti. A transient saturation of the double bond during isomerization must be excluded as well as a coupled hydration–dehydration reaction [23]. Thus, an enzymatic mechanism is proposed: an enzyme–substrate complex is formed in which the electrophilic iron (probably Fe3+), provided by the heme domain present in the enzyme, removes an electron from the cis double bond, transferring the sp2 linking into a sp3. The double bond is then reconstituted after rotation to the trans configuration. A scheme of this proposed enzymatic mechanism is presented in Fig. 2. Such a mechanism is in accordance with site directed mutagenesis experiments carried out to destroy the heme-binding motif in Cti of P. putida P8 [41]. These mutations result in loss of function of the enzyme and, thus, provide evidence for the presence of cytochrome c and heme in the catalytic center of the enzyme. Since the reaction of the enzyme does not depend on a cofactor Cti activity differs from all other known heme-containing enzymes acting on fatty acids as substrates. There is, however, no need of a cofactor because no net electron power is consumed.

Figure 2

Scheme of a possible enzymatic mechanism of Cti given for double deuterated oleic acid as taken for experiments by von Wallbrunn et al. [23].

Another indication for its uniqueness stems from similarity searches: Cti showed no significant similarities with homologous peptides when the predicted amino acid sequence was compared with other proteins. Not surprisingly, however, comparison of amino acid sequences of the seven up to the present known Cti proteins identified them all as heme-containing polypeptides of the cytochrom c-type [23]. Irrespective of the taxon, a heme group of the cytochrome c-type is present as a highly conserved motif and as a functional domain in all enzymes compared [41], in particular, the heme-binding site in Cyt c proteins is located between heme-vinyl groups and the two cysteines found in the conserved heme-binding motif CXXCH.

In all Cti sequences of the six Pseudomonas strains so far investigated an N-terminal signal sequence is present, indicative of the periplasmic localization of Cti. Such a localization has already been proven for P. oleovorans and P. putida DOT-T1E [36,39]. However, a signal peptide characteristic for Sec-dependent secretion is not present in the Cti protein of V. cholerae. Multiple sequence alignments of the seven known Cti proteins revealed that proteins from Pseudomonas and Vibrio strains form a phylogenetic tree composed of three main branches, suggesting a common ancestor of the enzyme. Interestingly, the predicted polypeptide from V. cholerae obviously does not constitute a separate group but rather emanates from the diverse group of proteins from P. aeruginosa and P. sp. E-3 [23]. Very recently, alignment studies revealed that genes familiar to cti might also be present in the genomes of bacteria belonging to the genera Methylococcus and Nitrosomonas. These organisms are also known to contain trans unsaturated fatty acids [5]. However, direct physiological or biochemical evidence for the presence of Cti in these bacteria is still missing.

4 Regulation of Cti

One of the major open questions regarding the Cti of unsaturated fatty acids is how the activity of this constitutively expressed periplasmic enzyme is regulated. One possibility would be a complex model in which the substrates of the enzyme, the cis unsaturated fatty acids, are cleaved off from the periplasmic phase of the membrane phospholipids. The resulting free unsaturated fatty acid would then be isomerized by Cti action and subsequently reattached to the lysophospholipid, resulting in a phospholipid containing trans unsaturated fatty acids [43]. Yet, such a complex model is not in agreement with data that confirm Cti activity in resting cells and in the complete absence of energy sources [10], as, at least, the reattachment of the modified fatty acids to the membrane would need energy.

Regulation of enzyme activity may, however, be brought about by simply giving the active center of the enzyme the ability to reach its substrate, the double bond, which in turn depends on the fluidity state of the membrane. Accordingly, the observed regio-specificity of the enzyme reflects penetration of the active site of the isomerase to a specific depth in the membrane [44]. The hydrophilic structure of Cti and its periplasmic location support the presumption that the enzyme can only reach its target, i.e. the double bonds of unsaturated fatty acids that are located at a certain depth of the membrane, when the membrane is ‘opened’ by environmental conditions that cause a disintegration of the membrane [45]. It has previously been shown that a decrease in acyl chain order can result in increased penetration and translocation of proteins in membranes [21]. By analogy to certain phospholipases, it is conceivable that Cti displays deeper penetration into the membrane when the ordering of acyl chains is decreased and the spacing of phospholipid head groups is increased. It is also clearly conceivable that decreasing the packing of membranes would allow double bonds to approach membrane surfaces more frequently [45], ultimately facilitating interaction with the isomerase [44]. As acyl chain packing is increased by cis to trans isomerization of the unsaturated fatty acids [19], the penetration of the protein would be counteracted [45] and, concomitantly, cis to trans isomerization inhibited, eventually resulting in tight regulation of the acyl chain packing without involvement of indirect signalling mechanisms or pathways. After removal of the membrane-active compound, recovery of the regularly low trans-cis ratio most likely occurs by normal de novo synthesis of all-cis fatty acids, since the reverse (trans to cis) process would require an energy input.

Such a model for the regulation of Cti activity also sufficiently explains the often reported relation between the degree of cis–trans isomerization and the toxicity caused by a certain concentration of an environmental stress factor [16,30]. As another result of the reaction catalyzed by the enzyme a reduction of membrane fluidity occurs and, as the enzyme cannot reach its target when membrane fluidity has reached its normal level the enzyme is forced out of the bilayer [45].

5 Concluding remarks

Although cis–trans isomerization of unsaturated fatty acids has not been completely understood it became obvious that it is part of a general stress response system in Pseudomonas and Vibrio cells. Another indication for the general function of Cti is also its often described dependency on the induction/activation of other stress response mechanisms [3032].

Evidently, it constitutes an urgent adaptive mechanism allowing rapid modifications of membranes to cope with the emerging environmental stress. Such a rapid response, acting in terms of minutes, provides time for other mechanisms depending on cell growth to facilitate their role in the adaptive response, as the immediate reaction guarantees survival under various stress conditions. Regarding solvent tolerance, a kind of a cascade of fast (urgent), mid-term and long-term mechanisms evidently work together to reach a full adaptation to environmental stress. Cti undoubtedly represents one of the major urgent systems that help the cells to withstand the first toluene shock, eventually allowing activation and induction of further adaptive mechanisms which finally provoke the complete adaptation [27,33].

Because of its easy function and effectiveness and because it works without complex regulations it is amazing that such a cis to trans isomerization mechanism is not ubiquitously present in Gram-negative bacteria. A possible explanation may come from the widespread occurrence of the two genera Pseudomonas and Vibrio. Among non-specialized bacteria, members of the genus Pseudomonas are known to be highly adaptable microorganisms, having conquered all niches of a great number of ecosystems comprising soil, human skin, and seawater. Members of the genus Vibrio likewise conquered a wide range of ecosystems, including soils and deep sea. To be able to colonize all these niches, they need to be extremely flexible and adaptable to changing environmental conditions. Cti provides the cells with an effective mechanism to achieve such an adaptability. This is not required in other Gram-negative bacteria, such as E. coli, which are specialized regarding life in the gastrointestinal tract of mammalians where they can live happily without such an urgent membrane adaptation mechanism.

Membrane lipids offer a promising tool as biomarkers for the analysis of microbial population changes. In fact, Guckert et al. [6] have suggested to use a trans/cis ratio greater than 0.1 (normal index reported for most environmental samples) as an index for starvation or stress. As the measurement of fatty acid profiles has become a routine method in many laboratories this sounds like a promising approach for the assessment of toxic effects. The determination of the trans/cis index may, thus, be a valuable option in studying the toxicity status of natural samples, especially when growth-dependent tests cannot be performed, e.g. in natural habitats. The major field of application of such an indicator appears to be the measurement of toxicity and environmental stress during in situ bioremediation processes where fatty acid profiles have importance as a marker for ecological investigations of the soil microflora. For example, during bioremediation of polluted sites, the level of trans unsaturated fatty acids can be used as a marker for general stress and stress reduction to monitor the biodegradation process [6,46,47]. The application of the cis–trans isomerization as an assessment tool for the general toxicity of organic compounds has already been described for aromatic carbonylic compounds [16,48]. Further studies aiming at and improving the use of the isomerization of cis to trans unsaturated fatty acids as an indicator for stress are vital and may ultimately result in an applicable technique for environmental monitoring.


  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].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
  43. [43].
  44. [44].
  45. [45].
  46. [46].
  47. [47].
  48. [48].
View Abstract