Bacterial membrane lipids: where do we stand?

* Abstract Phospholipids play multiple roles in bacterial cells. These are the establishment of the permeability barrier, provision of the environment for many enzyme and transporter proteins, and they influence membrane-related processes such as protein export and DNA replication. The lipid synthetic pathway also provides precursors for protein modification and for the synthesis of other molecules. This review concentrates on the phospholipid synthetic pathway and discusses recent data on the synthesis and function of phospholipids mainly in the bacterium Escherichia coli.

Key Words fatty acids, phospholipids, DNA replication, protein modification, phospholipid function

 
CONTENTS 
 
PERSPECTIVE 
BACTERIAL FATTY ACID METABOLISM 
  Recent Highlights and Remaining Gaps 
PHOSPHOLIPID SYNTHESIS 
  Phospholipid Synthetic Pathways 
  Topological Aspects of Membrane Phospholipid Synthesis 
PHOSPHOLIPID FUNCTION 
  Physiological Functions of Specific Phospholipid Species 
CONCLUSIONS 

PERSPECTIVE

Our insight into the metabolism and functions of bacterial membrane lipids has increased dramatically in recent years. Much of the recent activity has been concerned with bacterial fatty acid synthesis and this has engendered a number of reviews both from this laboratory (11, 20) and from others (44, 64, 81). Postsynthetic modifications of phospholipid acyl chains (17) and the biosynthesis of lipid A have also been reviewed (84). (Note that the intermediates of the lipid A pathway are also phospholipids, although in this review phospholipid denotes lipids derived from glycerol 3-phosphate.) In marked contrast to these areas, the bacterial phospholipid synthetic pathway has not been reviewed for a number of years, and most of those reviews predate the (ongoing) torrent of bacterial genome sequence data. Indeed, this explosion of DNA sequence data has engendered a curious dichotomy in that we have complete genome sequences for bacteria that lack published lipid compositional data. Given the restricted space available and the lack of recent reviews on membrane phospholipid synthesis and function, this review concentrates on these areas. Fatty acid synthesis is given short shrift and only some recent and noteworthy aspects of fatty acid metabolism are discussed.

BACTERIAL FATTY ACID METABOLISM

Recent Highlights and Remaining Gaps

Knowledge of the Escherichia coli fatty acid biosynthesis pathways has been extremely useful in understanding the pathways of other organisms. Although this information has been widely applicable to other bacteria and plants, the pathways of lower eukaryotes and mammals have also profited. Moreover, recent work has shown that some eukaryotic organisms that cause serious human parasitic diseases have a fatty acid synthetic pathway that uses bacterial enzymes. These successes aside, our knowledge of E. coli fatty acid metabolism has some serious shortcomings as a paradigm and a number of notable gaps in understanding. Also, bacterial fatty acid metabolism must be viewed in a context larger than membrane biosynthesis. Some highlights, shortcomings, and gaps of our knowledge are as follows:

* The classical pathway of anaerobic unsaturated fatty acid synthesis in E. coli is not widely distributed in bacteria (11, 45). Genomic analyses indicate that only the alpha and gamma proteobacteria encode the proteins of this pathway. Most other bacteria, including many pathogens, synthesize unsaturated fatty acids under anaerobic conditions but lack recognizable homologs of the key enzymes (FabA and FabB) of the E. coli pathway. A new pathway (or pathways) remains to be discovered. It may be that several pathways exist because Streptococcus pneumoniae, which lacks FabA and FabB homologs, has an enzyme called FabM that performs the key FabA isomerization reaction in vitro (the pathway has not yet been confirmed in vivo) (63). However, FabM (which lacks homology to FabA) seems specific for streptococci and thus does not appear to provide a general answer for other organisms that make unsaturates anaerobically but lack the FabA-FabB pathway.

* Not only are newly synthesized fatty acids used to make bacterial membrane lipids, but they also are the precursors of the acyl groups of acylated proteins such as Bordetella pertussis adenylate cyclase and E. coil hemolysin (96), the acyl groups of the acyl-homoserine lactone quorum sensing signals (autoinducers) (47, 74, 92, 101), and acylated polysaccharides [such as the rhamnolipids of pseudomonads and polyhydroxyalkonates (85)].

* Bacterial fatty acid synthesis is thought to be an excellent target for new antimicrobials (11,45, 67, 81), and this premise has led to a marked increase in the number of laboratories engaged in this research area. Two potent antibacterial compounds, cerulenin and thiolactomycin, are natural products that specifically inhibit fatty acid synthetic enzymes. Second, several compounds empirically selected as inhibitors of bacterial growth such as isoniazid, ethionamide, diazaborines and triclosan owe their antibacterial activities to inhibition of fatty acid synthetic enzymes (the first two compounds are widely used antituberculosis drugs). Although the chemical mechanisms of the pathway of fatty acid synthesis in mammals are virtually identical to those in bacteria (with the exception of unsaturated fatty acid synthesis), the protein sequences and the enzyme active sites are sufficiently different between mammals and bacteria such that specific inhibitors can be made. Indeed, of the known inhibitors of bacterial fatty acid synthesis, only one (cerulenin) also inhibits mammalian fatty acid synthesis. Driven largely by searches for new antimicrobials, new enzymes have been found (for example there now are three distinct families of enoyl reductases) and high-resolution structures of many enzymes (e.g., FabA, -B, -D, -F, -G, -H, -I, -R, an acyl carrier protein (ACP)-AcpS complex, butryl-ACP, the AccB biotin domain, and AccC) are available, whereas 10 years ago we had only a low-resolution NMR structure of ACP (11, 44, 45). The advent of available genome sequences has greatly accelerated progress by indicating which proteins are encoded by a given genome as well as (equally important) the proteins that are absent (such as homologs of the E. coli unsaturated fatty acid synthesis genes). The challenge for the future is to gain both a deeper and a broader picture of bacterial fatty acid metabolism to include both organisms and metabolic transformations that are thus far uncharacterized.

* Bacterial fatty acid synthetic enzymes are not found solely in bacteria. Most of the enzymes of plant fatty acid synthesis are of bacterial origin and reside in the plastids such as chloroplasts (although most of the encoding genes are now located in the plant nuclear genome). Antibiotics specific to bacterial type II fatty acid synthesis enzymes inhibit the growth of several protozoan parasites including the malarial parasite Plasmodium falciparum and the opportunistic toxoplasmosis pathogen, Toxoplasma gondii (97, 107). Recently it has been shown that the malaria FabH and AcpP are functional in vitro and the P. falciparum ACP is recognized by the E. coli fatty acid synthetic pathway (105). Identification of nuclear encoded fabH, fabl, acpP, and fabZ homologs that appear to be targeted to the apicoplasts of these organisms suggests that the site of thiolactomycin inhibition is a type II fatty acid synthesis pathway specific to the apicoplast (97, 107). The current model for the origin of the apicoplast is that it arose by engulfment of an algal cell by a nonphotosynthetic cell (107). Because the algal cell is thought to have arisen by engulfment of a cyanobacterium, the apicomplexan parasites seem to result from a series of engulfments (or invasions) that resulted in a secondary endosymbiosis such that a plant organelle provides fatty acids for a nonphotosynthetic cell using bacterial enzymes (107). The organelle is also thought be the site of lipoic acid synthesis in the organism. One of these antibiotics has been reported to kill the African sleeping sickness trypanosome, Trypanosoma brucei, by inhibiting the production of myristic acid (76, 80). The blood-borne form of the parasite uses myristate to anchor the variable surface glycoproteins that mask the parasite from host immune surveillance. Therefore, although trypanosomes lack the plant organelle, they seem to have retained bacterial fatty acid synthetic enzymes acquired from a temporary endosymbiont.

* There are no E. coli mutants reported in the literature for several key fatty acid synthetic genes; therefore, we have no in vivo evidence of the metabolic roles of the encoded enzymes (11). We lack mutations in acpP, the structural gene encoding ACP, and in fabH, the structural gene encoding 3-ketoacyl-ACP synthase III, an enzyme thought to play an essential role in the initiation of fatty acid synthesis. The only fabZ mutant has a weak phenotype (72) and thus cannot be used for genetic complementation experiments. One might argue that such mutations are not needed, but because so many of the annotations of the genomes of other organisms rests on that of E. coli, it seems that our knowledge of E. coli metabolism must be as solid and as thoroughly tested as possible in vivo. Moreover the genome sequences of all of the pathogenic E. coli and Shigella strains thus far examined contain the same fatty acid synthetic genes found in E. coli K-12 but also have what seems to be additional copies of many of these genes encoded within large conserved "islands" of sequence not present in E. coli K-12. What is the purpose of retaining these "extra" genes?

* In addition to the roles described above, fatty acids are good carbon sources in addition to being cellular building blocks. Fatty acids are degraded by the classical [beta]-oxidation pathway in aerobically growing E. coli, but the major fatty acid--rich environment for this and many other bacteria is an anaerobic environment such as the mammalian gut. Recent work has demonstrated that fatty acids are utilized as anaerobic carbon sources and that this utilization involves regulatory mechanisms …