In Vitro Activity of Actinospectacin in Human Whole Blood

Transcription

1 524 W. T. SOKOLSKI, C. G. CHIDESTER, AND L. K. SCHADEWALD filter through dialysis membrane would eliminate the low molecular weight components anid retain the polysaccharide inside the cell. Bringing the material inside the cell to volume several times would leave essentially pure polysaccharide inside. As it turned out only part of the type II precipitates would pass through the coarse membrane. All type I precipitates were retained almost quantitatively by the membrane. After only a very small fraction of the solution had passed through the filter, a mat would form over the membrane. No feasible way was found to eliminate formation of this mat. It would appear that a lattice type structure was formed which fit very closely to the filter membrane, probably held securely by electrostatic charges, which reduced the effective pore size of the membrane. LITERATURE CITED ANDERSON, D. A The production of gum by certain species of Rhizobium. Iowa Agri. Expt. Sta. Research Bull. No. 158, BRAY, H. G., E. SCHLUCHTERER, AND M. STACEY Enzyme formation and polysaccharide synthesis by bacteria. 2. Polysaccharide formation by Rhizobium radicicolum strains. Biochem. J. 38: COOPER, E. A., W. D. DAKER, AND M. STACEY Enzyme formation and polysaccharide synthesis by bacteria. 3. Polysaccharide produced by "nitrogen-fixing" organisms. Biochem. J. 32: JONES, A. S The isolation of bacterial nucleic acids using cetyltrimethylammonium bromide (Cetavlon). Biochim. Biophys. Acta 1: NORRIS, D A red strain of Rhizobium from Lotononis bainesii Baker. Australian J. Agr. Research 9: PALMSTIERNA, H., J. E. SCOTT, AND S. GARDELL The precipitation of neutral polysaccharides by cationic detergents. Acta Chem. Scand. 11: SCOTT, J. E The reaction of long-chain quaternary ammonium salts with acidic polysaccharides. Chem. & Ind., p SCOTT, J. E The preparation and fractionation of acidic polysaccharides using long-chain quaternary ammonium compounds. Biochem. J. 62:31 p. SCOTT, J. E Aliphatic ammonium salts in the assay of acidic polysaccharides from tissues. Methods of Biochem. Anal. 8: YAPHE, W The use of agarase from Pseudomonas atlantica in the identification of agar in marine algae. Can. J. Microbiol. 3: YAPHE, W Standards for bacteriological agar. Bacteriol. Proc., p. 26. In Vitro Activity of Actinospectacin in Human Whole Blood W. T. SOKOLSKI, C. G. CHIDESTER, AND L. K. SCHADEWALD ABSTRACT SOKOLSKI, W. T. (The Upjohn Company, Kalamazoo, Mich.), C. G. CHIDESTER, AND L. K. SCHADEWALD. In vitro activity of actinospectacin in human whole blood. Appl. Microbiol. 9: A method for testing antibacterial substances in whole blood is described. The test agent for the method was actinospectacin which reportedly has good in vivo activity, approximately in the range with chloramphenicol, but relatively poor in vitro activity in the common media. In human whole blood, however, the in vitro activity compares favorably with chloramphenicol thus indicating that whole blood may predict in vivo activity better than the usual bacteriological media. Some biological properties of actinospectacin, a new basic antibiotic, were recently reported by Mason, Smith, and Dietz (1961) and by Lewis and Clapp (1961). Although actinospectacin is active in vitro and Biological Control, The Upjohn Company, Kalamazoo, Michigan Received for publication April 7, 1961 in vivo against a variety of gram-positive and gramnegative organisms, it elicited a greater response in vivo than might be expected from its in vitro activities in brain heart infusion broth (Lewis and Clapp, 1961). An in vitro test system for actinospectacin was desired for the purpose of predicting in vivo responses against pathogenic bacteria. For this, a medium was sought which would simulate in vivo environment. Studies with the use of blood serum as a complete antibiotic testing medium have been reported in the past (Sokolski, Vavra, and Hanka, 196; Wolfe and McGuire, 1961). A blood medium which would include the cells may be an improvement since it is a step closer to the in vivo environment than serum. This paper describes a method for testing antibacterial substances in human whole blood and the results of comparative tests with two antibiotics, actinospectacin and chloramphenicol. Actinospectacin with poor in vitro and good in vivo activity is compared to chloramphenicol with good in vitro and in vivo activity (Smith et al., 1948). If whole blood is an

2 19611 ACTINOSPECTACIN IN HUMAN WHOLE BLOOD 525 effective medium for predicting in vivo responses, then both antibiotics should show good activities in this medium. MATERIALS AND METHODS Antibiotic solutions. Actinospectacin sulfate solutions in physiological saline were prepared in concentrations of 1,, 1, and 1,ug base equivalent per ml (1 mg of the sulfate salt was equivalent to 7,ug base). Chloramphenicol was dissolved in 95% ethanol to a concentration of 1 mg per ml and diluted with physiological saline to similar concentrations. All solutions were sterilized by filtration. Test organisms. The test organisms used in these studies were as follows: Staphylococcus aureus UC 76, S. aureus UC 367 (resistant to 1 ;g penicillin per ml), Diplococcus pneumoniae UC 42, Streptococcus hemolyticus UC 754 (S. pyogenes), Salmonella enteritidis UC 114, Proteus vulgaris UC 232, and Escherichia coli UC 527. Cultures to be used for inocula were prepared TABLE 1. Comparison of actinospectacin activity against Staphylococcus aureus UC 76 in brain heart infusion broth vs. human whole blood Actinospectacin base equivalent per ml Human Mwhole blood 4 hr Ag , 1.2X 14 TABLE 2. Brain heart infusion broth 24 hr 4 hr 24 hr 1, 2 4,7 2, 4.5 X X X X 18 6 X X 16 7 X X X X X X 1' 1.7 X X 18 2 X X 18 by growing these organisms in brain heart infusion broth (BHIB) for 18 hr at 32 C. Citrated human blood was added to the D. pneumoniae and S. hemolyticus cultures. D. pneumoniae, S. hemolyticus, and the two strains of S. aureus cultures were diluted 1:1 with physiological saline. The cultures of gramnegative bacteria, S. enteritidis, P. vulgaris, and E. coli, were used undiluted. The undiluted inoculum contained approximately 18 bacteria per drop. Materials for test. The following sterile equipment was prepared for the test: test tubes (13 by 1 mm) with metal caps, wooden applicators, medicine droppers, 1-ml and 1-ml pipettes, 125-ml flasks covered with aluminum foil, and blood collection kits with 17-gauge needles. Procedure for testing. The tests in BHIB, serum, and oxalated blood (.2 % potassium oxalate) were conducted as follows. Antibiotic solutions were added to IOs 19 r 4_ < 3_ 1 resh Serum _j ~~~~~~reated Serum Brain Heart Broth pg. ACTINOSPECTACIN BASE EQUIVALENT PER ML. FIG. 1. Effect of actinospectacin on Staphylococcus aureus strain UC 76 in various media after 24 hr at 82 C. Comparison of actinospectacin against Staphylococcus aureus UC 76 in various media Concn spectacin Whole blood Oxalated blood Fresh serum Treated serum, 56 C 3 min Brain heart infusion broth 4 hr 24 hr 4 hr 24 hr 4 hr 24 hr 4 hr 24 hr 4 hr 24 hr 2 1, ,1 8 1, ,7 2, 5 4 5,4 5 9,2 9 5, X X X X X X X X 16 1, 5.8 X X X 14 1, 6 X X X 16 9, 1.4 X X X X 16 7 X X X X X X X X X X X X X X X X X X X X , 7.5 X X X X X X X X X X 14 8 X X X X X X X 17 2 X X 18

3 526 W. T. SOKOLSKI, C. G. CHIDESTER, AND L. K. SCHADEWALD TABLE 3. Comparison of actinospectacin and chloramphenicol in blood and brain heart infusion broth Antibiotic concn Human whole blood Brain heart infusion broth Actinospectacin base Chloramphenicol Actinospectacin base Chloramphenicol 4 hr 24 hr 4 hr 24 hr 4 hr 24 hr 4 hr 24 hr pg/mi 2 2, 24 1 X , 2, 3,2 1 x 17 1 X 14 2, , 2.5 X 14 2 X 15 1 X 14 2, , 1 X 16 2 X X 14 8, , ,6 7 X 16 3 X 1" 1 X 15 2, 6.2 2,9 9.3 X 14 2, 4.6 X 14 2 X 17 3 X 18 2 X 16 2, X X 15 5,2 1 X 16 3 X 17 8 X X 15 3 X X X 16 3, 1.5 X X 17 1 X 19 8 X 16 3 X X X X 14 2 X X 17 1 X 19 8 X 16 4 X X X X 14 3 X 16 4 X 17 1 X 19 3 X 17 8 X X 13 8 X 16 4 X 17 4 X18 test tubes in volumes which when diluted to 5 ml with media gave the indicated concentrations. Twofold increments in concentrations were used in each test. The volume in each tube was brought to.5 ml with physiological saline followed by 4.5 ml of medium. It should be noted that blood was not added to the medium for the tests with D. pneumoniae and S. hemolyticus. Five-hundredths milliliter (one drop) of culture was added to each tube. All tubes were incubated in a stationary state at 32 C and samples were withdrawn, after thoroughly agitating the cultures by hand, at the indicated periods. Three dilutions of each sample were tested for the number of viable bacteria by mixing.5 or 1 ml of the diluted suspension with 14 or 28 ml of nutrient agar. The lower volumes of suspension or the higher volumes of agar were made for the purpose of diluting the antibiotic and were used with the tests that contained 5 pig or more of antibiotic per ml. In the tests with whole blood as medium, it should be emphasized that the timing in collecting the blood and setting up the tubes was important. The blood was drawn from volunteers with a 17-gauge needle and allowed to flow into sterile flasks by gravity flow. After 5 to 6 ml of blood were collected, the flask was removed and if more blood was desired a new flask was put in its place. The fresh unclotted blood was rapidly pipetted into tubes containing the antibiotics. One drop of the prepared inoculum was added to each tube and the contents of the tube were stirred with a sterile wooden applicator. All tubes were allowed to stand at room temperature for 1 hr or until the blood was clotted. If the clot was stuck to the wall of the tube, it was released using another sterile applicator. Serum samples were removed at the indicated times without w -J cn -J -i wcr- - z LJ -J oli CHLORAMPHENI COL ACT INOSPECTACIN \25 Ag./ml FsL./.mI /ml. -O /Lg./mi HOURS FIG. 2. Effect of actinospectacin and chloramphenicol against Staphylococcus aureus strain 76 in whole blood. disturbing the blood clot. Viable counts were determined in the same manner as indicated above. The minimal effective inhibitory concentration (MIC) of antibiotic was considered to be the lowest concentration in which the numbers of viable organisms did not increase between the 4- and 24-hr sampling with the condition that growth occurred without antibiotic. The MIC end points after 24 hr incubation at 32 C are indicated in all tables by a line in the 24-hr viable count column.

4 1961] ACTINOSPECTACIN IN HUMAN WHOLE BLOOD 527 RESULTS Human whole blood as a medium was compared to BHIB with actinospectacin and S. aureus UC 76 in the test systems. The response measured was the number of viable bacteria remaining after 4 and 24 hr in the antibiotic solution. The results in Table 1 indicate the MIC values at 1,ug per ml in BHIB and 6.2,ug per ml in blood. Actinospectacin activity against S. aureus UC 76 TABLE 4. was compared in whole blood, oxalated blood, fresh serum, treated serum (56 C for 3 min), and BHIB. The results in Table 2 indicate that actinospectacin activity may be greater in whole blood than in fresh serum, treated serum, oxalated blood, or BHIB with MIC end points at 6.2, 12.5, 12.5, 3.1, and 1 Mig base per ml, respectively. Figure 1 illustrates the data in Table 2. Actinospectacin activity was compared to chlor- Comparison of actinospectacin and chloramphenicol activity against several bacteria Viable cells per ml Test organism Antibiotic Actinospectacin base Chloramphenicol 4 hr 24 hr 4 hr 24 hr sg/ml Staphylococcus aureus UC X X X X X X 16 S. aureus UC 367 (penicillin re sistant) X X 16 1 X X 16 Diplococcus pneumoniae UC X X X X X 1, 4.6 X X X X X 17 Streptococcus hemolyticus UC X X X X X X 18 Escherichia coli UC X X X X X X X 1, 2. X 18 Proteus vulgaris UC X X X X 17 Salmonella enteriditis UC X X 15 21Q 2.5 X X X X X X X X X X X X X 1, 4. X X X X 1 -~~~~~~~~~~~

5 528 W. T. SOKOLSKI, C. G. CHIDESTER, AND L. K. SCHADEWALD TABLE 5. Summary of MIC end points from Tables 1 and 4 Test organism Chloramphe- nicol Actinospectacin base Antibiotic jag/mi pg/ml Staphylococcus aureus UC S. aureus UC 367 (penicillin resistant) Diplococcus pneumoniae UC Streptococcus hemolyticus UC Escherichia coli UC Proteus vulgaris UC Salmonella enteriditis UC amphenicol in both human whole blood and BHIB. Table 3 and Fig. 2 indicate that actinospectacin is more active in blood than is chloramphenicol, whereas the reverse is true in BHIB. After 48 hr, each tube with blood was tested for sterility and the results, which were the same as for the 24-hr samples, indicated that 23,ug or higher of actinospectacin base sterilized the culture, whereas viable organisms were still present with 2,ug chloramphenicol per ml. Actinospectacin activity was compared to chloramphenicol against several organisms in human whole blood. The results in Table 4 show that actinospectacin activity is nearly comparable with chloramphenicol. Table 5 is a summary of the comparative activities of actinospectacin and chloramphenicol in whole blood. DISCUSSION The choice of whole blood as a medium rather than serum, plasma, OI defibrinated or oxalated blood was made for several reasons. It was desired to keep the cells in the system because of their possible influence on antibacterial action. For instance, if the agent were adsorbed on or absorbed into the blood cells, it is possible that some antibacterial activity of the agent would be diminished. A medium as natural as possible was preferred without the addition of anticoagulants, such as potassium oxalate, and without hemolysis which usually occurs with mechanical agitation in defibrination. The samples of serum were removed without agitation of the blood clot with the reasoning that in studies in which relative responses are measured, the important thing is to keep all conditions uniform. With agitation there is not only the possibility of hemolysis of the loose red cells, but the number of loose blood cells may vary from tube to tube and thus affect the actual volumes of the serum samples. Since the bacteria were added to the tubes before the blood clotted, undoubtedly some organisms were trapped in the clot as it formed. Approximately 1 X 16 (gram-positive bacteria) or 1 X 18 cells (gramnegative bacteria) were added to 5 ml of media which would make the starting counts 2 X 15 per ml and 2 X 17 per ml. The 4-hr counts vary from 1 X 13 to 1 X 16, indicating a drop in the numbers of bacteria. Because of the possible loss of bacteria in the clot and by the bactericidal action of blood, the 4-hr counts were regarded as initial counts. Figure 1 may be open to criticism since straight lines were plotted for growth curves when in reality numbers of viable bacteria change gradually. However, the purpose of including the figure was to indicate that actinospectacin activity is comparable and perhaps better than chloramphenicol in human whole blood. The comparable in vitro activity in blood reflects the comparable in vivo activity of the two antibiotics. The figure would be markedly different with BHIB as seen in the data in Table 3. Many reports have been made concerning the bactericidal action of blood, two of the more recent ones by Muschel (196) and Colebrook, Lowbury, and Hurst (196). It is reasonable to suggest that antibiotics be tested in the suboptimal medium of whole blood for predicting in vivo responses since the bactericidal forces present will also be operating in vivo. LITERATURE CITED COLEBROOK, L., E. J. L. LOWBURY, AND L. HURST The growth and death of wound bacteria in serum, exudate, and slough. J. Hyg. 58: LEWIS, C., AND H. W. CLAPP Actinospectacin, a new antibiotic: III. In vitro and in vivo evaluation. Antibiotics & Chemotherapy 11: MASON, D. J., A. DIETZ, AND R. M. SMITH Actinospectacin, a new antibiotic. I. Discovery and biological properties. Antibiotics & Chemotherapy 11: MUSCHEL, L. H.: 196. Serum bactericidal actions. Ann. N. Y. Acad. Sci. 88: SMITH, R. M., D. A. JOSIYN,. M. GRUHZIT, I. W. MCLEAN, JR., M. A. PENNER, AND J. EHRLICH Chloromycetin: biological studies. J. Bacteriol. 55: SOKOILSKI, W. T., J. J. VAVRA, AND L. J. HANKA Assay methods and antibacterial studies on streptozotocin. Antibiotics Ann. 1959/196: WOLFE, R. N., AND J. M. McGUIRE The use of blood serum as test medium in the in vitro evaluation of antibiotics, pp Antimicrobial Agents Annual, 196, Plenum Press, New York.