In Vitro Comparison of Rosamicin and Erythromycin Against Urinary Tract Pathogens

The in vitro activity of rosamicin and erythromycin was compared at various pH values against 311 strains of bacteria representing common urinary tract pathogens. Alkalinization of the media consistently and significantly increased the antibacterial activity of rosamicin against all of the organisms tested. This was also true for erythromycin except when tested against strains of Proteus. At pH 8, rosamicin was two- to sixfold more active than erythromycin against Enterobacteriaceae. The activity of both antibiotics against Pseudomonas aeruginosa was very similar when tested at pH 8. Erythromycin was twice as active as rosamicin at pH 8 against group D streptococci. The activity of both antibiotics was bacteriostatic and inoculum size dependent, regardless of the organism tested or the pH of the test media. The greater activity of rosamicin against Enterobacteriaceae warrants clinical investigation.

Rosamicin, a relatively new macrolide antibiotic produced by Micromonospora rosaria, shares a number of chemical characteristics with erythromycin and has been shown to have a similar spectrum of antibacterial activity (10,11). Like erythromycin, it is most active against gram-positive organisms; however, it has been shown to be more active than erythromycin against Enterobacteriaceae and Pseudomonas (1, 11).
It has been well demonstrated, first by Haight and Finland in 1952 (3) and later by others (5, 7,12), that alkalinization of culture media enhanced the antibacterial activity of erythromycin. Subsequent clinical studies suggested that erythromycin plus alkalinization of the urine could be effective treatment for many urinary tract infections (4, 6,13,14). The increased activity of rosamicin against gram-negative bacilli at neutral or slightly alkaline pH suggested the possibility that it could be more active than erythromycin when used with alkalinization (11). The purpose of this investigation was to assess the relative in vitro activity of rosamicin and erythromycin against common urinary tract pathogens, both gram positive and gram negative, over a range of pH levels readily attainable in urine. MATERIALS  Susceptibility testing method. Isolates were simultaneously tested against rosamicin and erythromycin at pH values of 5, 6, 7, and 8 by means of a microtiter broth dilution technique. All of the tests were performed in Mueller-Hinton broth (Difco Laboratories, Detroit, Mich.) which had been adjusted to the appropriate pH by the addition of 1 N hydrochloric acid or 1 N sodium hydroxide. A Corning Research pH meter model 12 was used to determine pH values.
Serial twofold dilutions of freshly prepared antibiotics were made in Mueller-Hinton broth at each pH level, and 0.05 ml of each dilution was dropped into the wells of a microtiter plate (Cooke Engineering Co., Alexandria, Va.). Overnight broth cultures of the organisms containing approximately 3.1 x 109 organisms per ml were diluted loin Mueller-Hinton broth at each pH, and 0.05 ml was then added to the diluted antibiotic. The plates were placed on a shaker for 5 min and then incubated overnight at 35°C in ambient air. The minimum inhibitory concentration (MIC) was defined as the lowest concentration of antibiotic in which there was no visible growth. The minimum bactericidal concentration (MBC) was determined by using an adaptation of the Steers replicator (8) which delivered approximately 0.003 ml from each well of the microtiter plate to Mueller-Hinton agar (Difco). Plates were incubated overnight at 35°C in ambient air.
Inoculum size effect. The effect of inoculum size on both the MIC and the MBC at each pH level was evaluated by testing 10 strains from each group of organisms with a 102 as well as a 10-4 dilution of an overnight broth culture. Figure 1 graphically illustrates the effect of pH on the susceptibility of E. coli. At pH 5, only 2% of the 42 strains tested were inhibited by 16 ,tg of rosamicin per ml; at pH 8 this concentration inhibited 100% of the strains. Similarly, at pH 5, a concentration of 500 ,tg of erythromycin per ml was not inhibitory for any of the E. coli tested; however, at pH 8, 93% were inhibited by 16 ,g/ml.

RESULTS
Klebsiella MICs were extremely high at a pH of 5 ( Alkalinization of the media did not significantly enhance the activity of erythromycin against either P. mirabilis or indole-positive strains of Proteus; however, rosamicin MICs were appreciably decreased when a pH of 8 was used (Fig. 2). Only 10% of the P. mirabilis were inhibited by 250 ,g of rosamicin per ml at a pH of 5, but 83% were inhibited by 31 JLgg/ml when tested at pH 8. Very similar results were obtained with indole-positive Proteus. At a pH of 5, only 10% were susceptible to 250 ,tg of rosamicin per ml; at a pH of 8, 71% of the strains were inhibited by 31 ,ug/ml.
Although it would appear that rosamicin is more active at pH 5 than pH 6 ( Fig. 2), it is difficult to interpret data obtained when testing P. aeruginosa at that level of acidity. Organisms whose normal habitat is the gastrointestinal tract (which includes all other groups tested in this study) tolerate a more acid pH (i.e., 4.4) (2). The minimum pH for growth of P. aeruginosa tends to be slightly higher, ranging from 5.6 to 4.4 (2, 9), and indeed a number of strains tested failed to grow at pH 5. Although only strains that visibly grew in the pH 5 broth control were included in these results, it is probable that the effect of acidity on the population was a significant factor. At more alkaline pH values, the pattern of decreasing MICs was again observed, with rosamicin and erythromycin being approximately equal in activity at pH 8. Ninety-five percent of the strains were inhibited by 125 ,ug of either antibiotic per ml at that pH, and, at a concentration of 31 Ag/ml, 50% of the strains were inhibited by rosamicin and 43% were inhibited by erythromycin.
Group D streptococci (Fig. 2) were the most susceptible organisms tested and the only group which demonstrated lower MICs to erythromycin than to rosamicin at pH 8. At a concentration of 0.5 ug/ml, only 56% of the strains were inhibited by rosamicin whereas 85% were inhibited by erythromycin. Again, increasing alkalinity MBC data are not shown, but MBC values remained consistently high for both antibiotics, regardless of the organism or pH used. Table 1 demonstrates the dramatic increase in geometric mean MICs obtained when the inoculum size was increased 100-fold. This was evident for both antibiotics, all organisms tested, and at each pH when MIC values were such as to permit comparison. DISCUSSION These results indicate that, like other macrolide antibiotics, the antibacterial activity of rosamicin was consistently and significantly increased by increasing the pH of the media, even when tested against Proteus. This is an advantage over erythromycin which, as demonstrated by the data in this study and earlier by Sabath et  bDilution of overnight culture in Mueller-Hinton broth.
tested at pH 8. Although erythromycin did demonstrate enhancement of antibacterial activity by alkalinization when tested against the other Enterobacteriaceae in this study, rosamicin at pH 8 was the more active antibiotic, with MICs two-to sixfold lower than those of erythromycin. The activity of both antibiotics against P. aeruginosa at pH 8 was very similar and, again, significantly improved over activity at lower pH values. MICs against group D streptococci were also decreased by alkalinization, but unlike the gram-negative bacilli studied, erythromycin appeared to be more active than rosamicin at an alkaline pH. Previous studies by Sabath and his colleagues (6) have demonstrated that alkalinization of urine has no effect on the amount of erythromycin excreted in urine in normal volunteers and that the urine concentration of erythromycin varied from 33 to 368 ug/ml after four oral doses of 1.0 g of erythromycin estolate at 8-h intervals. In a subsequent clinical study, Zinner et al. found (14) that erythromycin plus alkalinization eradicated the infecting organism during therapy in 73% of 37 patients with chronic bac-teriuria and that urine cultures remained negative 3 weeks after treatment in 17 patients.
The in vitro data provided here suggest that, if pharmacokinetic studies prove it to be similar to other macrolides in terms of urinary concentration and toxicity, rosamicin could be a useful agent in the treatment of urinary tract infections, particularly those caused by members of the Enterobacteriaceae. Clinical investigation of rosamicin would then be warranted.