Search

Colistin, a fatty acyl peptide antibiotic, was attacked by proteolytic enzymes such as papain, ficin and bromelain, and as degradation product, a peptide portion retaining the ring structure of colistin was liberated. In contrast, an analogous antibiotic polymyxin B showed a characteristic resistance to the catalytic activity of papain.Colistin nonapeptide and α-N-fatty acyl α,γ-diaminobutyric acid were obtained as products from the above enzymatic hydrolyzates of colistin and their chemical and physicochemical properties were investigated.Contrary to colistin, this colistin nonapeptide was inactive to Escherichia coli. NIHJ and to many other strains even at a concentration of 800 mcg/ml by the agar dilution method. As α-N-fatty acyl α,γ-diaminobutyric acid which is rest part of colistin was added to colistin nonapeptide, antimicrobial activity of colistin nonapeptide did not increase.

A suitable technique was developed to determine the antibiotic sensitivity of Propionibacterium. Then, inhibition patterns of 30 strains of 11 different Propionibacterium species against 39 antibiotic or antimicrobial agents were obtained. Propionibacteria were resistant to penicillinase-resistant penicillins such as cloxacillin, nafcillin, oxacillin, and a naphtheridine derivative nalidixic acid. Moderate sensitivity to kanamycin and colistin was shown. Only a few strains of propionibacteria were inhibited by gantrisin ( sulfisoxazole), one of nine sulfonamides tested. Determination of sensitivity to antibiotic and antimicrobial agents would not assist in speciation of this genus.

In summation, the effect of three mycin antibiotics on pancuronium bromide-induced depression of neuromuscular transmission was studied, using a sciatic nerve tibialis anterior muscle preparation of anaesthetized dogs. Muscle twitch responses were recorded before and after administration of non-paralysing doses of the muscle relaxant and the test antibiotic. It was inferred from the results that colymycin, gentamycin and kanamycin potentiated the neuromuscular blockade induced by pancuronium. Colymycin showed the least potency while kanamycin induced the most effective blockade. Calcium and neostigmine had limited effect in reversing the neuromuscular blockade due to gentamycin and kanamycin. The antibiotics are believed to depress neuromuscular transmission by depleting acetylcholine at the motor endplate, through inhibition of its release. With decreased calcium levels in the motor nerve terminal membranes, non-depolarizing muscle relaxants exert a similar effect. At the post-synaptic membrane the antibiotic desensitizes the receptors to acetylcholine, while the muscle relaxant acts by occupying the receptor. The mode of action of the antibiotic and muscle relaxant at the presynaptic membrane is similar to that of magnesium ions. The resistance to reversal of the neuromuscular blockade by both calcium and neostigmine finds explanation in its multifocal nature.

All of 37 recent human isolates of Edwardsiella tarda were resistant to colistin. All strains were resistant to 6.3 μg of sodium colistimethate per ml; 82% of them were resistant to 100 μg/ml. With 34 Salmonella strains, the average minimal inhibitory concentration (MIC) was 1.6 μg of sodium colistimethate per ml. There was virtually no overlapping of MIC values between both groups of bacteria. Using the Kirby-Bauer method and 10-μg colistin disks, inhibition zone diameters were less than 11 mm for all Edwardsiella strains, and larger for all Salmonella strains. An atypical, mannitol-fermenting strain of E. tarda was resistant to colistin. It is concluded that E. tarda is naturally resistant to colistin, that this property may be helpful in differentiation, and that colistin should be tried in enrichment and selective media for the isolation of E. tarda . It was also observed that, under conditions as described, all strains of E. tarda were susceptible to streptomycin and to novobiocin, whereas all strains of Salmonella were resistant.

SUMMARY A total of 880 samples of faeces from 584 neonates in four hospitals were examined for the presence of antibiotic-resistant Gram-negative bacilli. The proportion of strains of Escherichia coli and Klebsiella spp. carrying R factors was also determined. Some resistant strains of E. coli and Klebsiella spp. were rapidly acquired after birth and the number of isolates increased in relation to the length of hospital stay. The proportion of isolates of Klebsiella spp. increased to over 60% in babies in special care units but increased to a lesser extent (less than 30%) in other wards. The four hospitals showed differences in the proportion of babies carrying antibiotic-resistant strains and this was partly attributable to the inclusion of samples from special care units in three of the hospitals. Antibiotic-resistant Gram-negative bacilli other than E. coli and Klebsiella spp. were isolated much less frequently. Transferable resistance was demonstrated in 49% of 208 resistant strains of E. coli and in only 6·5% of 240 strains of Klebsiella spp. An increased frequency of resistant strains was found in babies associated with abnormal deliveries and in babies treated with antibiotics. Resistant strains were probably acquired initially from the mothers, but although there was evidence of cross-infection with E. coli and Klebsiella spp. particularly in special care units, no common source was identified.

Summary The leucocyte migration test (LMT) was used in thirty patients with skin reactions to drugs (except contact dermatitis) together with other diagnostic tests: lymphocyte transformation test (LTT), specific IgE, passive haemagglutination and patch-tests. The optimal concentrations of drugs for the use of LMT were determined. In all cases in this series, sensitization to the drug considered responsible was confirmed by a positive LTT or by specific IgE. On the contrary, LMT was positive only four times. The value of LMT in detecting allergic reactions to drugs is therefore limited. Out of four positive LMT, three were obtained in patients with bulla formation, showing a definite mononuclear cell infiltration of the basal membrane, suggestive of a cellular mediated mechanism in such drug reactions.

Three hundred and thirty-eight strains of Escherichia coli from clinical sources were examined for their drug resistances and R factors in 1971. With reference to four drugs, i.e., tetracycline(TC), chloramphenicol(CM), streptomycin(SM) and sulfanilamide(SA), the SA-resistant strains were isolated most frequently followed by SM-, TC- and CM-resistant ones in decreasing order. Among those resistant to the aforementioned four drugs, 20.3, 19.0, 7.7 and 18.3% were single-, double-, triple- and quadruple-resistant, respectively. The R factors were demonstrated at a frequency of 29.2% among these resistant strains. From the strains possessing quadruple-resistance, the conjugationally transferable R factors were demonstrated most frequently (60.5%) followed by triple- (59.1%), double- (21.7%) and single-resistant strains (4.5%). The isolation frequencies of strains resistant to kanamycin (KM) and ampicillin (APC) were 1.5 and 4.7%, respectively; being fewer in numbers compared with those resistant to the four drugs. Strains resistant to cephaloridine, nalidixic acid, Colistin, gentamicin C complex and furatridine were not isolated. Strains resistant to KM and APC were demonstrated mostly from those possessing multiple resistance with reference to the aforementioned four drugs. Among the strains carrying either APC- or KM-resistance, 76.9 and 100% respectively, were found to carry R factors, indicating that most of the APC- and KM-resistances were conferred by the R factors. Twenty percent of the R+ strains with reference to resistance to the four drugs carried two types of R factors in a cell, being in a hetero-R state, and 66.7% of the strains carrying R factors with APC- or KM-resistance were in a hetero-R state.

terized as to their patterns of resistance to antibiotics. Identical antibiograms that were found in E. coli of members of the farm family and associated groups of animals were counted as matches, using three types of counting techniques. Five hundred random assortments were made between the set of 14 farm families and the set of 14 groups of livestock. Matches were also counted when farm families were matched with their own livestock. With randomization, three of the 500 assortments yielded a value as large as the 44 matches observed when farm families were matched with their own livestock. Statistical tests indicate a strong association between the resistance patterns of the E. coli isolated from people and those isolated from their livestock.

Gale, Glen R. (Veterans Administration Hospital, Durham, N.C.). Urease activity and antibiotic sensitivity of bacteria. J. Bacteriol. 91: 499–506. 1966.—An investigation was made of the responses of certain urease-positive bacteria to various antibacterial drugs in the presence of highly specific urease inhibitors, in a test of the hypothesis proposed by other workers that inhibition of bacterial urease enhances the sensitivity of the cells to antimicrobial agents. Urease inhibitors employed were seven hydroxamic acids (HA). Six of the seven HA reduced the sensitivity of nine Proteus strains to ampicillin and methenamine mandelate. Two HA increased the sensitivity to colistin, and six HA increased the sensitivity to kanamycin. Investigation of the mechanism of action of the synergistic effect between kanamycin and HA led to the tentative conclusion that potentiation was mediated through an initial alteration of cell permeability by the aminoglycoside antibiotic which permitted accumulation of each of the six HA into the cell, at which point each interacted with pyridoxal phosphate. The single HA which failed to yield synergism with kanamycin failed to interact with pyridoxal phosphate in a nonenzymatic system; the other six HA produced alterations of the normal ultraviolet absorption spectrum of the coenzyme.

The one-stage prothrombin time of human plasma was prolonged by the addition to the thromboplastic mixture of either colistin sulfate or polymyxin B. Colistimethate sodium, a derivative of colistin in which the amino groups of the cyclic polypeptide antibiotic are substituted, did not show this effect. The inhibition was greater if the antibiotics were incubated with thromboplastin in the absence of Ca++. In the Hicks–Pitney modification of the thromboplastin generation time test, the two polyamine antibiotics reduced the amount of thromboplastic activity generated. Generation, however, appeared to be initiated sooner. It is postulated that these actions by the antibiotics are the results of modifications in the zeta potentials of the thromboplastic micelles.

GRAM-negative bacilli, in particular pseudomonas species, are of increasing concern to the clinician as infectious and therapeutic problems.1 , 2 Accordingly, polymyxin B has grown to singular importance as its utility to the treatment of systemic infections caused by pseudomonas species has been appreciated and exploited. In 1961 colistin was introduced as an agent that supposedly retained the antibacterial merit of polymyxin B in a polypeptide compound of lesser inherent nephrotoxic liability. From the outset, the striking pharmacologic and chemical similarities of colistin to the entire polymyxin group of antibiotics, including polymyxin B, were recognized.3 4 5 6 However, critical comparison of these agents, taking . . .

Anti-bacterial effect of various agents for pyocyaneus bacilli was evaluated by test tube and disc method. The bacilli used were obtained from the cases with post-surgical infections folio-- wing to radical operation of the middle ear and also those which had been preserved in the- Department Department of Bacteriology. The bacilli were relatively sensitive to 3 to 5% solution of mer- curochrome and hydrogen peroxide. Trypaflanin, rivanol and monaf, uracin were not effective, even with high concentration. Among sulfas, homosulfamin was most effective and sulfa thiazol, sulfaisomidin and sulfaisoxazol were much less effective. In, antibiotics colistin and polymyxin B were most effective followed by streptomycin and chloramphenicol. The organisms were, much more resistant to terramycin, aureomycin, penicillin and erythromycin.

Five cases of Pseudomonas keratitis have been presented. Suggestions relative to the use of subpalpebral lavage in the management of this infection have been offered.

Faecal volatile fatty acids represent the end products of the metabolism of the anaerobic flora of the large bowel. The excretion of these volatile acids has been investigated in five leukaemic patients maintained in plastic `germ-free' isolators. Under `isolator' conditions there is a pronounced fall in volatile fatty acid excretion. The possibility that the measurement of these acids may be used to monitor anaerobic overgrowth and recolonization in these patients is discussed.

CONTENTS I. Introduction 108 II. The polymyxins 108 III. Circulin 111 IV. Colistin (colimycin) 111 V. Polypeptin 112

Summary and ConclusionsPolymyxin B and colistin. 2 closely related polypeptide antibiotics, had similar antibacterial activity against a large number of pathogenic bacteria. Polymyxin was somewhat more active than colistin against 55% of susceptible strains. on the average 3.4 and 4 times as active against most strains of N. gonorrhoeae and H. influenzae, respectively. Highly resistant variants developed on repeated subculture of a strain of A. aerogenes and one of Ps. aeruginosa on agar containing either of these antibiotics; complete crossresistance to the other antibiotic developed in each instance. Single-step mutants of the same strain of A. aerogenes, highly resistant to each antibiotic, were readily obtained by seeding a large inoculum on the surface of agar containing 100 or 200 μg/ ml of that antibiotic; these mutants were completely cross-resistant to the heterologous antibiotic. Antibiotic-dependent variants were not encountered.

Patterns of susceptibility of gram-negative anaerobic bacilli to antibiotics have been found to be distinctive and of significant help in classification and identification. Five major groups of gram-negative anaerobic bacilli have been defined on the basis of morphological and biochemical criteria. Antibiotic sensitivity patterns conform to these groupings and provide additional taxonomic criteria. The Bacteroides fragilis group is resistant to penicillin G, whereas the other groups are generally sensitive. B. fragilis strains are relatively sensitive to erythromycin, whereas the Sphaerophorus necrophorus group is resistant. B. melaninogenicus strains, B. oralis , and Fusobacterium are all more sensitive to kanamycin and neomycin than the other two groups. Kanamycin is more active against Fusobacterium strains than neomycin, but less active against all other groups. Colistin or polymyxin B is useful for distinguishing between the resistant B. fragilis and the sensitive S. necrophorus . Antibiotic susceptibility determinations may be more readily performed in clinical laboratories than certain biochemical tests recommended for differentiation of the gram-negative anaerobic bacilli and may serve as helpful adjuncts to morphological and biochemical observations in classifying and characterizing these organisms. The use of standardized procedures for antibiotic susceptibility tests is essential if comparable results are to be obtained in different laboratories.

The incidence of Escherichia coli and Salmonellae on eggs from 20 chicken and 20 turkey flocks in Ohio was determined. Cultures isolated from 734 chicken and 412 turkey eggs were tested for drug resistance to nine antimicrobial agents by a standardized single disc method. The isolates resistant to chlortetracycline and/or dihydrostreptomycin were tested for the ability to transfer drug resistance to E. coli K-12NA in vitro. E. coli was isolated from 294 (25.5%) of the chicken and turkey eggs examined during this study. Only four cultures of Salmonella (three species) were isolated and only one of these was drug-resistant (to dihydrostreptomycin only). This single resistant culture did have R factor. Of the 294 isolates of E. coli, 180 (61.2%) were found to be resistant to one or more antimicrobial agents. Among the resistant isolates, 82.8 percent were multiple resistant and 17.2 percent were resistant to only one drug. Over half (55.4%) of the “resistant” E. coli isolates transferred all or a part of their resistance patterns to the drug-sensitive recipient strain. A higher percentage of E. coli strains were resistant to dihydrostreptomycin (49.6%) and to the tetracyclines (46.6%) than to any other antimicrobial used in this study. The drug resistance patterns of the R factor-carrying E. coli cultures isolated from the four flocks from which Salmonellae were also isolated were not found in the Salmonella species.

Jolipeptin causes a release of 260 mμ, absorbing materials from the jolipeptin-sensitive strains, Escherichia coli B and Bacillus subtilis PCI 219. Moreover, Jolipeptin lyses not only protoplasts of B. subtilis PCI 219 and Micrococcus lysodeikticus, but also spheroplasts of E. coli B and Pseudomonas aeruginosa IFO 3901. In contrast, colistin, produced simultaneously with Jolipeptin, lysed spheroplasts of E. coli B and P. aeruginosa IFO 3901, but not protoplasts of gram-positive bacteria, the latter being colistin-insensitive strains. On the basis of this study, it was concluded that Jolipeptin acts primarily on the cellular membrane.

There are currently available a bewildering array of antimicrobial agents marketed in hundreds of dosage forms. Each physician has the responsibility of choosing the most effective, safest and least expensive agent for treating a given infection. To accomplish this requires some knowledge of the significant differences in these drugs. In some cases, the major difference lies in spectrum of antimicrobial activity; in others, the difference lies in some pharmacologic property such as reliability of absorption from the gastrointestinal tract or rapidity of excretion by the kidneys. Certain new antibiotics such as the semi-synthetic penicillins and cephalosporins have significantly added to the therapeutic armamentarium. Other drugs such as lincomycin and polymyxin E (colistin) are very similar to older and well-studied antimicrobial agents and do not have any significant advantages which warrant their use in preference to the older drugs.

The activity of methacycline against microbial L-forms and their parent bacteria was compared with that of oxytetracycline, chlortetracycline, tetracycline, and demethylchlortetracycline, as well as with that of 22 other antibiotics which included examples of major groups of antibiotics. The L-forms and bacteria used were Streptococcus faecalis, S. faecium, S. faecalis var. zymogenes, Staphylococcus aureus (three strains), Proteus mirabilis (two strains), Pseudomonas aeruginosa, Escherichia coli (two strains), Sarcina flava, Serratia marcescens , and Klebsiella pneumoniae . The five tetracyclines had similar activities and were more active against L-forms than bacterial forms, except that the bacterial form of S. flava was more susceptible than the L-form. In general, other antibiotics (except the penicillins) were more active against L-forms than bacterial forms. There were certain exceptions where the bacterial form was more susceptible than the L-form. These included the effect of polymyxin B and colistin on P. aeruginosa, E. coli , and P. mirabilis , and the effect of gentamicin on P. aeruginosa, E. coli, S. flava , and S. marcescens .

The effects of colistin or ampicillin, in combination with furosemide-induced diuresis, on the renal titers of Escherichia coli after unilateral intramedullary inoculation of the bacteria into rats were determined. Therapy was started four days after inoculation and continued for seven days. Colistin therapy in either diuresing or nondiuresing rats significantly reduced renal titers of E. coli in both kidneys when compared with controls (rats receiving saline injections) or with those animals receiving only furosemide. There was no difference between titers in diuresing and nondiuresing rats receiving colistin. Renal titers of E. coli in rats receiving furosemide without colistin showed no significant differences in either kidney from those in control rats. Similar results were obtained with ampicillin therapy.

1. Therapeutic doses of colistin sulphomethate sodium B.P. (Colomycin), pento-N sulphomethylpolymyxin B sodium (Thiosporin) and ampicillin B.P. (Penbritin) as well as normal saline have been given to volunteers with normal kidney function. 2. A blind crossover technique was used and the effects of the compounds on blood levels and clearances of urea and creatinine studied. 3. All four treatments resulted in some changes in the parameters measured but were reversible. The most marked changes were found with pento-N sulphomethylpolymyxin B sodium and were also associated with severe side-effects. 4. In contrast to penicillins it is established that the process of tubular secretion sensitive to probenecid block plays no part in the renal excretion of polymyxins.

Shigella strains isolated in Japan from 1968 through 1970 were surveyed for drug resistance and distribution of R factors. Of the 2688 strains, 93.4% were resistant to either one or various combinations of four drugs, tetracycline (TC), chloramphenicol (CM), streptomycin (SM) and sulfanilamide (SA). Among these resistant strains, 74.2, 10.7, 1.48, and 13.6% were quadruply, triply, doubly, and singly resistant, respectively. Fifty-eight per cent of these resistant strains were found to carry R factors when judged by transferability of the resistance. The isolation frequencies of R (TC. CM. SM. SA), R (CM. SM. SA), R (SM. SA), and R (TC. SM. SA) factors were 73.2, 13.0, 11.5, and 1.3%, respectively. The strains resistant to drugs other than the aforementioned four were very few; 4.3, 3.4, and 0.7% being resistant to ampicillin (APC), nalidixic acid (NA), and kanamycin (KM), respectively. Among 117 APC-resistant strains, 97.4% could transfer their APC resistance together with other resistance markers. Seventeen out of 18 KM-resistant strains could transfer KM resistance by mixed culture. But none of the NA-resistant strains could transfer their NA resistance. The authors could demonstrate strains carrying two different R factors in a cell and one of them was consistently an R (SM. SA) factor. These results were very similar to those obtained in surveys of strains isolated from 1965 through 1967.

The polymyxin antibiotics are inactivated in vitro by tissues because they are bound to phospholipids of cell membranes. A method that liberates bound drug in an active form from tissues permitted study of distribution and persistence in vivo. Studies using single injections in the rabbit showed that bound drug persists in liver, kidney, brain, heart, muscle, and lung for as long as 72 hr. Accumulation in tissue but not in serum was noted on repeated injection, with persistence for at least five days after seven daily injections. Free polymyxin B was detectable in liver, kidney, muscle, and brain and persisted for many days in muscle and brain. Free colistimethate was detectable in all tissues other than brain and persisted for many days in liver, kidney, and muscle. Colistimethate appears to be incompletely converted in vivo to the parent compound, colistin. These observations may account for differences in toxicity and chemotherapeutic efficacy of polymyxin B and colistimethate.

The biosynthetic mechanism of 6-methyloctanoic and isooctanoic acids, which are present in the amide linkage with the α-amino group of the terminal α, γ-diaminobutyric acid residue of colistin A and B, respectively was investigated. From the isotopic experiments using isoleucine-U-14C, valine-U-14C and acetic acid-2-14C, it was concluded that 6-methyloctanoic and isooctanoic acids were derived from isoleucine and valine, respectively.Amino acids pooled in colistin-producing cells grown in the synthetic medium were abundant in isoleucine, valine and leucine, which were probable precursors of the abovedescribed fatty acid components of colistin and cellular fatty acids. On the other hand, 6-methyloctanoic and isooctanoic acids were not found in the cellular fatty acids, while C-15 and C-16 branched chain fatty acids usually found in Bacillus sp. were abundantly contained in the cells, indifferently of an improved capacity of colistin formation.

SUMMARY: Mechanically disrupted cells of polymyxin E sensitive strains of Pseudomonas denitrificans and Escherichia coli and a resistant strain of Staphylococcus aureus were studied in the electron microscope. The occurrence of a secondary layer or membrane was observed within the outer cell walls derived from Ps. denitrificans and E. coli; a similar structure was not observed in cell walls prepared from Staph, aureus. Treatment of the disrupted cells of Ps. denitrificans and E. coli with polymyxin E resulted in marked morphological changes, whilst similar exposure of Staph. aureus cell walls to the antibiotic produced no significant changes in their morphology.

A patient with acute renal failure secondary to an overdose of colistin is presented. Peritoneal dialysis was used in an attempt to reduce the serum colistin level. Most previous reports have indicated that dialysis is not an effective way of removing colistin from the serum; there have been no published reports of the use of dialysis specifically for colistin intoxication. In our patient serum colistin was reduced from 16 to 4 μg per milliliter during the initial dialysis, with a further reduction to 2 μg per milliliter during a second dialysis. Renal function was markedly impaired until after the second dialysis, but thereafter gradually improved. Subsequent measurements of renal function have been normal.

Variations in the treatment of polymyxin B and polymyxin E (colistin) with formaldehyde and sodium bisulphite produce sulphomethyl derivatives which differ quantitatively in acute toxicity and in antibacterial activities in vitro and in vivo. The acute intravenous LD50 values of some sixty samples of these derivatives range from six- to more than eighty-fold those of the parent antibiotics; the in vitro antibacterial activities range from 2 to 12% and the in vivo activities from 20 to 50% of those of the parent antibiotics, with the most toxic derivatives showing the highest activities. When short and prolonged incubation methods are used, assays of the derivatives in solutions of different ages and of blood collected from man and dog after intramuscular injection, show that the antibacterial activities of these sulphomethyl derivatives depend on reversion to the unsubstituted form, and that the differences in the activities are due to variations in stability. These conclusions are supported by comparison of these sulphomethyl derivatives with stable acetyl derivatives. The lower in vivo activity is due, at least partly, to the high renal excretion of the substituted form. Electrophoresis shows that the derivatives are composite, the components corresponding to mono- to pentasulphomethyl polymyxin. Pain at the injection site is the most troublesome side-effect of polymyxin therapy, and this is avoided with these derivatives. In rats injected with quantities some twenty-times the usual human dose, the derivatives cause transitory decrease in urinary output and transitory proteinuria. After intramuscular injection of these derivatives into dogs, no antibiotic is detectable in the cerebrospinal fluid and concentrations present in the bile are not significantly different from those after injection of the parent antibiotic. When injected intracisternally into these animals, derivatives are less toxic than the parent compounds. These studies show that acute intravenous toxicity is a useful index of therapeutic efficiency and that derivatives with intravenous LD50 values of about 100 mg/kg are the most satisfactory ones. Because activity depends on reversion to the parent antibiotic, the use of these derivatives for topical application is contraindicated.

A case of colistln nephrotoxicity without history of previous renal disease was studied by light and electron microscopy of a renal biopsy specimen obtained 11 days after discontinuation of the drug. The histologic picture showed variation from early acute necrosis to regeneration of the damaged renal epithelium. There appeared to be difference of susceptibility among individual tubular epithelial cells to sodium colistinmethanesulfonate in the development of nephrotoxicity. Morphological changes were essentially similar in both the proximal and distal tubules. Indentation of the nuclear membranes was conspicuous and appeared to be demonstrable earlier than the cytoplasmic alterations. ACTA PATH. JAP. 19: 55˜67, 1969.