Gramicidin isolation

Fi ra 2 Peptide analogs farmed by in vivo and in vitro biosynthesis, (a) Analogs of grarnicidin S isolated from cultures or obtained by in vitro enzymatic synthesis. Note that due to the symmetry of the compound, in principle both mono- and disubstiruted anologs may be found, (b) Analogs of tyrocidine isolated from cultures. Note the similar specificities of positions 3 and 4, the restriction to Phe in position 1, and Tyr acceptance only in position 7. (c) Analogs of linear gramicidin isolated from cultures. Note that only Trp in position 11 is variable, and only Val in position 1. 

Antibiotics. The genes involved in the synthesis of a variety of antibiotics have been isolated (34,35). These include antibiotics such as erythromycin, streptomycin, and also peptide antibiotics such as gramicidin and tyrocidin. Characterization of these gene products facUitates the design of novel antibiotics. In addition, overexpression of some of these gene products is also expected to improve the yield of the antibiotic (34,35). 

In 1939 the isolation of a mixture of microbial products named tyrotbricin from a soil bacillus was described. Further investigation showed this material to be a mixture of gramicidin and tyrocidine. In rapid succession the isolation of actinomycin (1940), streptothricin (1942), streptomycin (1943), and neomycin (1949), produced by Streptomjces were reported and in 1942 the word antibiotic was introduced. Chloramphenicol, the first of the so-called broad spectmm antibiotics having a wide range of antimicrobial activity, was discovered in 1947. Aureomycin, the first member of the commercially important tetracycline antibiotics, was discovered in 1948. 

Gramicidin is a peptide antibiotic first isolated by Dubos in 19391 as a crude complex together with a second peptide antibiotic known as tyrocidin. The mixture of the two antibiotics was called tyrothricin. Although the mixture was discovered about ten years after penicillin2, tyrothricin was the first antibiotic utilized in clinical practice. 

Other workers began to study the structure of gramicidin. Christensen and coworkers12 isolated crystalline tryptophane and leucine from a hydrolysate. They found no evidence for a fatty acid component and established that phenylalanine, proline and hydroxyproline were absent from a hydrolysate. These workers isolated alanine diox-pyridate from a hydrolysate and also established that gramicidin contained a compound with vicinal hydroxy and amino groups. They speculated that this compound might be serine or isoserine and proposed that gramicidin contains two tryptophane, 2 leucine, 2 or 3 alanine and 1 hydroxyamino residues or a multiple of this composition. 

Hotchkiss114 isolated optically and analytically pure d-leucine from the hydrolysate. This was the first non-enzymatic proof that d-amino acids actually occurred in gramicidin. He also noted the presence of an amino-hydroxy compound, but indicated that it was not isoserine. 

Christensen16 isolated the dipeptide valylvaline from completely hydrolyzed gramicidin. This worker later showed that he had isolated a racemic mixture of D(-)-valyl-D(-)-valine and L(+)-valyl-L(4)-valine rather than dipeptides containing one d and one L-residue.17... 

Still using heterogeneous gramicidin, Synge24 isolated the D-leucylglycine, L-alanyl-D-valine and L-alanyl-D-leucine from partial hydrolysates of gramicidin. He also had less conclusive evidence for the tripeptides alanyl-valylleucine or alanylleucylvaline. 

The gramicidin complex was originally isolated by Dubos as a component of the antibiotic mixture called tyrothricin formed by an aerobic sporulating bacillus7. 

Hotchkiss and Dubos have utilized solvent extraction to separate gramicidin and tyrocidine105. Several patents have been issued on procedures for isolating gramicidin from fermentation broth106, 107, 108. 

Gramicidin C (a pentadecapeptite from Bacillus brevis) [9062-61-7]. Same as Gramicidin A since they are isolated together and separated. [Sarges and Witkop Biochemistry 4 2491 1965 as well as references above for Gramicidin A]. 

Our approach has been essentially empirical in nature with less emphasis on the theoretical. We have isolated single substances, proved their purity, and determined their covalent structure by classical methods of organic chemistry we have then used these substances of molecular weight ranging from 1,000 to 14,000 as model solutes for the study of conformation and intermolecular interaction. Solutes of special interest have been gramicidin SA (2), bacitracin A (3), polymyxin B, and the tyrocidines A, B, and C (4). All are cyclic antibiotic polypeptides. The first three behave in aqueous solution as reasonably ideal solutes and do not associate, but the tyrocidines associate strongly and are interesting models for the study of association phenomena. Other model solutes of... 

While gramicidin and other channel formers can show high transport rates, they do not show the high selectivity that characterizes natural channels. There is much interest at present in a class of proteins called porins, which form natural pores in the outer membranes of Gram-negative bacteria. Several different porin proteins have been isolated from Escherichia coli. These form water-filled channels of various sizes in membranes. Thus the proteins OmpC and OmpF seem to be cation-specific channels while other proteins give larger diameter channels that seem to be specific for anions.34,35... 

A non-ribosomal biosynthetic pathway is clearly indicated for cyclosporin A, considering the uncommon structural elements MeBmt, L-a-aminobutyric acid and D-alanine as well as the plethora of isolated congeners [20,21]. Non-ribosomal biosynthesis directed by multienzyme thiotemplates have been reported for other small peptides of microbial origin, for example, gramicidin S [22] and enniatin [23]. Experimental data for cyclosporin A were obtained by feeding appropriate labelled precursors to cultures of T. inflation strains. The distribution profile of the labelled atoms in cyclosporin A was determined by 3H- or 13C-NMR spectroscopy. In preliminary trials with several tritium and carbon-14 labelled precursors, [met/y>/-3H]methionine proved to be the most suitable marker for the biosynthetic preparation of radiolabelled cyclosporin A for pharmacokinetic and metabolic studies [24],... 

To date, the most extensively studied natural ionophore is gramicidin,10 a polypeptide antibiotic isolated from the bacterium Bacillus brevis. Indeed, the idea of a channel-like structure for ion transport was inferred from its study. However, unlike channel proteins, which are exclusively composed of L-amino acids, each alternate amino acid in gramicidin has D-stereochemistry. It is composed of 16 residues, 15 of which are amino acids. The structure is summarized below, in which Xxx has the following identities gramicidin A (gA), Trp gB, Phe gC, Tyr gD, a mixture of gA, gB, gC, -80 5 15. 

Gramicidin contains no acidic or basic groups. Tyrocidine contains strongly basic groups due to ornithine residues. It can be isolated only as the hydrochloride. The crystalline mixture tyrocidine contains 1-omithine, 1-proUne, 1-valine, 1-leucine, d-phenylalanihe, 1-tryptophane, 1-tyrosine, 1-aspartic acid and 1-glutamic acid (11). 

An enzyme has been isolated from the FK520 producer which is believed to be the key one responsible for inserting pipecolic acid into the macrocycle [114]. It is reported to be dimeric and activates pipecohc acid and several structural analogues in an ATP-dependent reaction to give an enzyme-bound amino-acyl adenylate. There is evidence that this then reacts to form a thioester linkage to the enzyme. This mechanism of activation is the same as that found in the non-ribosomal biosynthesis of peptide natural products such as gramicidin [112]. 

Even in the case it should be possible to separate ribozyme activity from the ribosome or to isolate an in vitro selected ribozyme that can catalyze the same type of peptide bond formation as a ribosome, however such a biocatalyst seem does not to be suitable for simple practical use rather than using a chemical coupling reagent. In principle, this conclusion is also valid for the nonribosomal poly- or multienzymes which are involved in the biosynthesis of peptide antibiotics[7Z. Up to now, they have only found application in the synthesis field of cyclosporin, gramicidin S, special P-lactam antibiotics and analogs. 

In 1939 Dubos reported the isolation of the antibiotic tyrothricin from culture filtrates of strains of Bacillus brevis > < He and his co-workers soon succeeded in separating tyrothricin with organic solvents into two crystalline compounds, the neutral gramicidin and the basic tyrocidine (formerly called graminic acid and gramidinic acid). Tyrothricin is thus a term for the partially purified antibiotic obtained from the culture fluids of Bacillus brevis. It contains on the average a mixture of 20 per cent gramicidin and 80 per cent tyrocidine. Besides these, at least three further polypeptides have been found by countercurrent distribution of tyrothricin. 

---