Calvin M. Kunin

Evidence for a Common Hapten Associated with Endotoxin Fractions of E. coli and other Enterobacteriaceae.

Summary Sera from rabbits immunized at 3 different laboratories with E. coli 014 antigens prepared from boiled whole cultures or with Boivin type extracts agglutinate human and sheep erythrocytes coated with lipopoly-saccharides or crude 0 antigen type preparations from a wide variety of E. coli and other Enterobacteriaceae but do not produce heterologous bacterial agglutination or precipitin formation. Inhibition and absorption studies suggest that the observed phenomena are due to a hapten associated, but not necessarily identical, with the crude endotoxin moiety of Enterobacteriaceae which is so structured in E. coli 014 as to be able to induce antibody formation.

CLINICAL SIGNIFICANCE OF PROTEIN BINDING OF THE PENICILLINS

The clinical significance of serum protein binding of penicillins has not been clearly established by trial in patients. Furthermore, it is difficult to conceive how a properly conducted comparative dose-response curve study can be constructed in view of ethical considerations inherent in the management of seriously ill patients. For this reason, indirect means of obtaining evidence of antimicrobial activity and distribution of active drug in tissues must be relied on for guidance. Useful information required in comparing drugs with similar mode of action include 1 ) relative antimicrobial potency, 2 ) extent of serum binding, 3 ) relative rate of gastrointestinal absorption, 4) tissue distribution, 5 ) rate of removal by liver, kidneys and other organs, and 6) rate of inactivation. No one factor, considered alone, is sufficient to judge potential value in infection (Kunin, 1964). The phenomenon of serum binding has been amply reviewed in the general pharmacologic literature (Goldstein, 1949; Thorp, 1964; Gillette, 1965) and for the penicillins by Verwey (1966) and Rolinson (1967). This literature may be summarized as follows: Binding of small molecules to serum proteins is a normal physiologic process which solubilizes and sequesters metabolites and hormones and permits their slow release to meet tissue requirements. It may also be considered as a defense mechanism which retains potentially toxic substances in an inactive form until their disposal by the liver or kidneys. Albumin is the best known of the binding proteins, because of the wide variety of substances that bind to it. Normal metabolites that bind to albumin include such diverse substances as fatty acids, bilirubin, calcium and various hormones. Transcortin, transferrin, thyroxinbinding globulins and ceruloplasmin are more specialized serum proteins that selectively bind cortisol, iron, thyroxin and copper, respectively. This reversible form of binding is not adequate for induction of antibody formation. In penicillin allergy, for example, a rearrangement product of the penicillin molecule binds irreversibly to a serum protein to form an antigen. Binding by albumin should be viqwed as a dynamic process in which a constant proportion of drug exists in the unbound or free state. This is shown in the following equation:

Enhancement of Antimicrobial Activity of Penicillins and Other Antibiotics in Human Serum by Competitive Serum Binding Inhibitors.

Summary A series of compounds, many of which had been shown in dialysis studies to be effective in displacing penicillins from binding to human serum, have been tested for ability to enhance the antimicrobial activity of various penicillins and other antibiotics in the presence of 50% human serum. Enhanced antimicrobial activity in serum was shown in the case of sulfonamides to be only partially reversible by para-amino ben-zoic acid, suggesting that augmentation of penicillin activity was due to serum displacing effects as well as to the intrinsic antibacterial activity of sulfonamides. Evidence is presented that the “R” groups of penicillins are the most important determinants of binding to serum. The specificity of the penicillin binding sites was shown by the failure of displacing agents shown to be effective against the penicillin site(s), to effect binding of novobiocin and tetracyclines, and by the lack of effect of probenecid, known to be highly bound, to displace penicillins from serum. The author is indebted to Mrs. Louise Moore and Mrs. Ann Bugg for technical assistance, and to Dr. William S. Jordan, Jr., for review of the manuscript.

Comparative Serum Binding, Distribution and Excretion of Tetracycline and a New Analogue, Methacycline

Summary The absorption, distribution, excretion, and serum binding of tetracycline and methacycline, a new analogue, have been compared. The higher and more sustained levels of the latter drug noted after oral and intravenous administration in man appear to be due to delayed renal excretion and more limited distribution rather than to enhanced gastrointestinal absorption. These differences are thought to be a reflection of greater serum binding of methacycline.

Inhibitor of antibiotics in bacteriologic agar.

Summary A series of antibiotics have been tested for in vitro activity in various media. A component in routine bacteriologic agar, not present in agarose and which is blocked by addition of protamine or toluidine blue has been found to interfere with diffusion of the polymyxins, aminoglycoside and some other antibiotics. This is thought to be due to acidic sulfate groups present in bacteriologic agar. The relation of this effect to possible in vivo binding to tissue components and their significance for antimicrobial sensitivity tests has been briefly discussed. Addendum An enhancement of infectivity of respiratory syncytial virus by diethylaminoethyl dextran was recently described by Nomura (Nomura, S., Proc. Soc. Exptl. Biol. Med. 128, 163, 1968). This agent (mol. wt. 2 × 106, Pharmacia, Uppsala, Sweden) was tested in bacteriologic agar to determine whether or not it would increase the size of zones of inhibition of polymyxin B. A marked enhancing effect was found with as little as 0.03 mg per ml. Concentrations in medium 199 and 2% agar of 1 mg per ml of DEAE dextran increased the zone size of polymyxin disks against E. coli from 10.7 mm to 16.3 mm, corresponding to the zone size found with agarose gel using medium 199 and the same test organism. Thus, DEAE dextran has been shown to have an effect in agar as well as on virus infectivity.

Serum binding, distribution and excretion of four penicillin analogues following intravenous injection in man.

Summary The in vitro binding to human serum proteins, and the half-life in serum, relative volume distribution and urinary recovery of 4 penicillin analogues, G, V, phenethicillin, and mercaptomethyl penicillin following single intravenous injections in normal male subjects were investigated to aid interpretation of oral absorption studies. Penicillin G was found to be significantly less bound than the others (48.7% compared to 74.4 − 78.8%). Significant differences in relative distribution volumes of the 4 analogues could not be demonstrated due in large part to considerable subject and laboratory variation. The shortest half-life in serum was obtained with penicillin V (0.52 hour), the longest with penicillin G and phenethicillin (0.75-0.78 hour), while phenylmercaptomethyl penicillin was intermediate (0.64 hour). The 24-hour urinary recoveries were penicillin G 46%, V 57%, phenethicillin 60%, and phenylmercaptomethyl penicillin 21%. It is concluded from comparison of these results with those of others that differences noted in protein binding may be outweighted by other considerations such as cost, optimum therapeutic dose, relative antimicrobial activity and extent of gastro-intestinal absorption when oral preparations are used.

Significance of borderline counts in screening programs for bacteriuria.

The conservative approach to bacterial counis appears to be reasonable in thesymptomatic patients. The question remains whether or not borderline counts are of similar significance in routine screening of asymptomatic patients under field conditions. The current report attempts to provide information on this point by analysis of the emergence of persistent significant bacteriuria in relation to the initial bacterial count obtained among school girls routinely screened at yearly intervals for 5 years.

Preventing Catheter-Induced Urinary Tract Infections

Must the urinary catheter be outlawed or do methods exist to forestall the bacteriuria and related, often dangerous, complications that almost invariably accompany its use? This symposium discusses the new approaches that are proving effective in minimizing the hazards of urinary catheterization for the great majority of patients.Must the urinary catheter be outlawed or do methods exist to forestall the bacteriuria and related, often dangerous, complications that almost invariably accompany its use? This symposium discusses the new approaches that are proving effective in minimizing the hazards of urinary catheterization for the great majority of patients.

Urinary tract disease in school girls with bacteriuria.

This paper will summarize some of the results obtained during the first four years of a long-term study of the epidemiology and natural history of urinary-tract infections in school children. Large populations of school children can be screened readily for bacteriuria by mass procurement of clean voided specimens utilizing the quantitative bacterial count. Careful preparation of the community has permitted us to achieve repeated voluntary participation in these studies of 90 percent of all children enrolled in schools. School children are particularly suited to such investigations, because they are a “captive” group representative of the total population, because parents are keenly interested in their health, and because children provide an opportunity to apply preventive measures at an early stage in the natural history of their disease. The epidemiologic approach, combined with long-term study of cases, appears to offer the most fruitful means of evaluating the relation of bacteriuria to pyelonephritis in view of the insidious nature of the disease and the difficulties inherent in histologic diagnosis of pyelonephritis as employed in retrospective studies. This report is based on a study of almost 20,000 urine cultures obtained from more than 16,000 persons and close follow-ups of 122 cases discovered in studies conducted in Waynesboro and Charlottesville, Va. Surveillance of a cohort of this population is being continued by yearly urine cultures to detect occurrence of silent infection and by repeated interviews to determine the frequency of overt infection treated by physicians in the community. Reports of the various phases of these studies have already been

Non-Antigen-Antibody Precipitin Reactions Observed with Dextran Sulfate, DEAE-Dextran, Antibiotics, Proteins, and Phospholipids

Discussion and Summary A series of precipitin reactions in agarose gel are described which resemble, but are unrelated to antigen-antibody reactions commonly studied by similar methods. This appears to be due to the lack of interference in agarose by sulfate groups present in ordinary agar. The reactions appear to be due to complexes formed between highly negatively charged compounds such as dextran sulfate, heparin, and inositol phosphatide and the highly positively charged polymyxin and aminoglycoside antibiotics, lysozyme, protamine, DEAE-dextran, and presumably beta lipoproteins as well. Unfortunately, the formation of a precipitin band between human serum and dextran sulfate and sensitivity limited to antibiotic concentrations several-fold higher than achieved during antimicrobial therapy, prevents direct adaptation of these methods to a rapid procedure for detecting basic antibiotics in serum. The approach could conceivably be adapted to measurement of basic drugs in urine or to ultrafiltrates of serum. Evidence is presented that the precipitin band formed between human serum and dextran sulfate is due to complexes formed with beta lipoproteins. A simple method for screening for hypo-beta lipoproteinemia is described. It is unlikely that protamine is present in sufficient concentrations in human serum to account for this band. Human serum contains small amounts of lysozyme, but in somewhat lower concentrations than can be measured by the precipitin reactions reported here with dextran sulfate. Osserrnan (5) reported lysozyme levels of about 7 μ/g ml in human serum, while dextran sulfate detects about 64 μ/g/ml. It is possible that dextran sulfate could be useful in detecting lysozyme in urine of patients with monocytic leukemia and other diseases. A non-antigen–antibody precipitin reaction in agarose has recently been described by Gardner and Rosenberg (7) between a lipoidal tissue extract and an IgM serum component. Several other isolated reports of similar phenomina are cited by these authors.