Antibiotics

J. Worth Estes, M.A.,M.D.
Professor of Pharmacology
Boston University School of Medicine

Questions or comments should be mailed to Carol Walsh

Experimental data for a new antibacterial agent, Drug A, are presented below.

Toxicity studies in uninfected mice indicate that the LD50 of Drug A is about the same as that for penicillin G. Drug A is distributed, in man, in the extracellular fluid compartment, and has a half-life in the plasma of 1.5 hours. Approximately 30% of a rapidly injected intravenous dose of Drug A is found unchanged in the urine within 12 hours of its administration. At pH 2.5, in vitro, Drug A irreversibly loses all of its antibacterial potency.

The in vitro antibacterial effectiveness of Drug A and penicillin G was studied using microorganisms isolated from patients ill with bacterial infections. The minimally effective concentration (MEC) for Drug A and penicillin G are given in Table I. Suppression of growth for 24 hours after inoculation was the criterion of effectiveness.

Table I
Organism MEC (µg/ml)
Drug A Penicillin G
Streptococcus pneumoniae
Streptococcus hemolyticus
(Patient 1)
Streptococcus hemolyticus
(Patient 2)
Streptococcus hemolyticus
(Patient 3)
Streptococcus hemolyticus
(Patient 4)
Staphylococcus aureus
(Patient 5)
Staphylococcus aureus
(Patient 6)
Staphylococcus aureus
(Patient 7)
Staphylococcus aureus
(Patient 8)
0.15
0.28
0.17
0.10
0.35
2.00
4.00
2.50
3.10
0.040
0.025
0.019
0.008
0.037
0.190
0.250
100.000
70.00

The effects of penicillin G, vancomycin, Drug A, and bacitracin on bacterial growth and viability were measured using a bacterial strain (called no. 28) isolated from a patient. The data obtained are given in Table II. The control culture medium contained no antibacterial agent. All the culture media contained p-aminobenzoic acid in excess of the normal requirements for bacterial growth. All cultures contained 10^7 viable colonies per ml of broth at the start of incubation. After 8 hours of incubation each culture contained the number of colonies listed in Table II.

Table II
Antibiotic Added To Broth Number of Viable Bacterial Colonies
per ml of broth after 8 hours of incubation
None
Penicillin G (10mg/ml)
Vancomycin (10mg/ml)
Drug A (10mg/ml)
Bacitracin (10mg/ml)
10^12.0
10^11.5
10^5.0
10^4.5
10^3.5

I. It can be inferred from the data in Table I that drug A is:

a. Effective against penicillin G-resistant streptococci

No. The data really provide no evidence on this point, inasmuch as none of the Streptococcal strains appear to have been particularly resistant to penicillin G. The rank orders of sensitivities to the two drugs are identical, in fact. Go back to Item I.

b. Effective against penicillin G-resistant streptococcus pneumoniae

Come now. With only one patient, you have no convincing evidence that his Streptococcus pneumoniae were, in fact, resistant to penicillin G. Go back to Item I.

c. Effective against penicillin G-resistant staphylococci

Yes, because the two bacterial isolates which required massive amounts of penicillin G to achieve efficacy were effectively suppressed by relatively small concentrations of Drug A. Go on to Item II.

d. Ineffective against penicillin G-resistant staphylococci

No, to the contrary, as you’ll find when you look at Table I again.

Table I
Organism MEC (µg/ml)
Drug A Penicillin G
Streptococcus pneumoniae
Streptococcus hemolyticus
(Patient 1)

Streptococcus hemolyticus (Patient 2)
Streptococcus hemolyticus
(Patient 3)
Streptococcus hemolyticus
(Patient 4)
Staphylococcus aureus
(Patient 5)
Staphylococcus aureus
(Patient 6)

Staphylococcus aureus (Patient 7)
Staphylococcus aureus
(Patient 8)

0.15
0.28
0.17
0.10
0.35

2.00
4.00
2.50
3.10

0.040
0.025
0.019
0.008
0.037

0.190
0.250
100.000
70.00

e. Two of the above

You may be half-right, but that’s not enough. Go back to the table and restudy the data there.

II. Vancomycin, bacitracin, Drug A, and penicillin G were administered parenterally to mice experimentally infected with the bacterial strain isolated from Patient No. 7. The available data suggest that, with optimum treatment schedules:

a. The median curative dose of vancomycin would be less than the median curative dose of Drug A.

No, if anything the data available in the two Tables suggest that Drug A and vancomycin are nearly equipotent. Go back to Item II.

b. The median curative dose of Drug A would be less than the median curative dose of penicillin G.

Yes, of course, because equivalent concentrations of Drug A and penicillin G permitted very different bacterial growth rates. Go on to Item III.

c. The median curative dose of Drug A would be greater than the median curative dose of penicillin G.

No. Go back and restudy the data in Table II.

d. The therapeutic index of bacitracin would be greater than that of Drug A.

No; we have absolutely no evidence that permits us to make this inference. In fact your knowledge of bacitracin’s toxicity should have prevented you from making this inference in the first place. Try again.

e. None of the above.

Well, you might be 75% correct, but you still missed. Study the data again.

III. From the data available, you can conclude that:

a. Less than 50% of Drug A is metabolized in the body after its parenteral administration, as is true for sulfisoxazole and other commonly used sulfonamides

No. You’re right about the sulfonamides, but the data suggest that, after eight half-lives, as much as 70% of the dose of Drug A was metabolized. Try again.

b. The administration of probenecid would be expected to decrease the fraction of Drug A that is metabolized, as is true for penicillin G

Come now, probenecid inhibits the renal tubular secretion of organic acids, not their metabolism, which might well be increased if they remain in the body longer in the presence of probenecid. However, probenecid does prolong the duration of penicillin G action. Go back to Item III.

c. When, administered orally, Drug A would be ineffective as an antibacterial agent, as is true for penicillin V or ampicillin

No. We don’t really know whether all of an orally administered dose of Drug A would be destroyed in the stomach or not, do we? Only 80% of penicillin G is destroyed by this mechanism. At any rate, the virtues of penicillin V and ampicillin include their stability at gastric pH. Try again.

d. It would require much greater doses of Drug A for effective oral administration than for effective parenteral administration, as is true for penicillin G

Yes. This would seem to be a perfectly reasonable inference. Go on to Item IV.

e. Drug A will be effective in the treatment of brain abscesses caused by penicillin G-resistant staphylococci, when administered intravenously

No. We really don’t have any evidence that Drug A can cross the blood-brain barrier, although we cannot be sure that it cannot. Try again.

IV. From the data available in the Tables above, and from your knowledge of other bacterial agents, it can be concluded that:

a. Drug A, like methicillin and oxacillin, is hydrolyzed by beta-lactamase

No. There is good reason, from the data in Table I, to infer that Drug A is NOT susceptible to beta-lactamase; moreover, methicillin and oxacillin are known not to be susceptible to beta-lactamase. Go back to Item IV.

b. Drug A, like ampicillin, tetracycline, and ciprofloxacin, is a broad-spectrum antibiotic

No. Well, you’re right that ampicillin and the others have wide spectra of antibacterial efficacies, but we have no evidence for Drug A beyond a few Gram-positives, do we? Try again.

c. Drug A, like erythromycin, is effective against Gram-positive organisms

Well, you’re half-right. Now identify the other correct answer. Go back to Item IV.

d. Drug A, like chloramphenicol, inhibits protein synthesis at the transcriptional level

No. You’re right about chloramphenicol , but we can’t make any such inference about Drug A yet. Go back to Item IV.

e. Drug A, like streptomycin, is bacteriocidal

No. You are right about streptomycin and Drug A being bacteriocidal (for Drug A because the number of organisms following incubation was less than at the beginning of the experiment). However, you are only half-way to answering the question. Go back to Item IV.

f. Two of the above are equally plausible

Good, if you identified answer choices c and e as correct, you can now go on to Item V.

V. You can infer, from the available data, that the plasma half-life of Drug A in a patient with renal insufficiency (after parenteral administration) would be:

a. Shorter than that in the normal subject, as would be expected for chloramphenicol

No. Come now, renal insufficiency is not known to shorten the half-life of any substance. At any rate, renal insufficiency does not affect the persistence of microbiologically active chloramphenicol in the body, because it is primarily cleared by glucuronide conjugation in the liver. Go back to Item V.

b. Longer than that in the normal subject, as would be expected for tetracycline

Very likely; in fact, renal disease is a contraindication to the use of tetracycline (but not of demeclocycline or doxycycline). However, you still haven’t completely answered this question. Go back to Item V.

c. Shorter than that in the normal subject, as would be expected for the naturally occurring penicillins

No. You should have recognized this as a nonsense answer. The half-lives of penicillins can be prolonged (not shortened!) in anuria, because penicillins are primarily cleared by renal tubular secretion. Adjustments in dose are usually unnecessary in lesser degrees of renal impairment.

d. Longer than that in the normal subject, as would be expected for erythromycin

No, little or no change in dose or schedule is required for erythromycin in the presence of renal disease, because the antibiotic is metabolized in the liver. Go back to Item V.

e. Longer than that in the normal subject, as would be expected for streptomycin and the sulfonamides

Yes, both drug groups are excreted chiefly through the kidney, However, you haven’t really completely answered the question, so go back to Item V.

f. Two of the above

Yes, you’re entirely correct. Tetracycline, streptomycin and sulfonamides are all excreted principally through the kidney. Go on to Item VI.

VI. Agents which provide effective treatment for tuberculosis include:

<”h4″ class="bu_collapsible">”a.

No. Well, both drugs are effective in tuberculosis, but resistance to both can impair their effectiveness. Go back to Item VI.

b. Isoniazid and rifampin, because both are relatively lipid-soluble

Quite true, which means that both can be expected to be effective against intracellular organisms like M. tuberculosis. But is this all? Go back to Item VI.

c. Streptomycin, because the tuberculosis organism is unable to develop resistance to aminoglycosides

No, M. tuberculosis strains, which can overcome the inability to read the genetic code induced by streptomycin, often develop during therapy with this antibiotic. Go back to Item VI.

d. Streptomycin, because the vestibular nerve is more sensitive to the antibiotic's toxic effect than is the acoustic nerve

Quite true. Vestibular dysfunction may well develop during streptomycin therapy, but it can be better tolerated than can deafness (which more often developed during therapy with the now unavailable dihydrostreptomycin). However, you’re only half correct on this question. Go back to Item VI.

e. Pyrazinamide and ethambutol, because both are potent inhibitors of DNA gyrase

No, although both are effective in tuberculosis, pyrazinamide is an antimetabolite, and we have no good clues to ethambutol’s mode of action. Go back to Item VI.

f. Two of the above

Yes, if you’ve perceived that isoniazid and rifampin, being lipid-soluble, are especially useful for treating intracellular organisms, and that streptomycin’s side effects on the vestibular nerve can be better tolerated than its effects on the acoustic nerve. Go on to Item VII.

VII. The acetylated biotransformation products of sulfonamides:

a. Non-competitively inhibit dihydrofolate reductase

No, no! On three counts. One, the acetylated biotransformation products are microbiologically inactive. Two, the parent compounds are competitive inhibitors, and three, they are competitive inhibitors of dihydrofolate synthetase. It makes a difference. Go back to Item VII.

b. Cannot oxidize hemoglobin

No. It’s a fine point, but the metabolites are just as effective in producing methemoglobinemia as the parent drugs. Try again.

c. Are more effective than their parent drugs in the treatment of lower urinary tract infections

No, the metabolites are microbiologically ineffective. That is why, for instance, sulfisoxazole is useful in treating lower urinary tract infections, because it is not extensively metabolized, as well as the fact that it has a low pKa, and so is very soluble in the urine. Back to Item VII.

d. Are responsible for the renal side effects of these agents

True. The acetylated metabolites are, by and large, less soluble in urine, especially when its pH is about 7.5, so they tend to crystallize out in the renal tubules. Go on to Item VIII.

e. None of the above

No, you’ve missed something. Try again.

VIII. Which of the following would you expect to have the smallest therapeutic index in man:

a. Agents that cross-link murein-peptide fragments, like the cephalosporins

Come now. Which cells in man contain murein? Go Directly to Jail. Do not Collect $200. Then, try again.

b. Agents that interfere with the incorporation of p-aminobenzoic acid into dihydrofolic acid

Come now. To what extent are human cells dependent on PABA for their vital processes? What are some examples of drugs which act in this fashion? Do you still persist in this answer? If so, go back to Item VIII and reconsider.

c. Agents that inactivate DNA, like bleomycin or actinomycin D.

Well, of course, because DNA is common to all organisms. That is why these two true antibiotics – bleomycin and actinomycin D – are more appropriately used in cancer chemotherapy than for treating infectious diseases. Go on to Item IX.

d. Agents that inhibit the synthesis of RNA, like the aminoglycosides

No. Well, drugs that inhibit protein synthesis at the transcriptional level ARE more likely to be toxic than most others, but the aminoglycosides (can you name at least three?) do not exert their effects in this way. Try again.

e. Agents that prevent the repair of bacterial DNA, like ciprofloxacin

No. Because mammalian cells lack the enzyme DNA gyrase, the fluorinated quinolones will not affect your patient’s DNA. Indeed, they are relatively free of side effects in man. Go back to Item VIII.

IX. Which of the following is (are) contraindicated in the treatment of infections caused by appropriately sensitive organisms in patients who also have elevated plasma bilirubin concentrations, prolonged prothrombin times, and low serum albumin concentrations:

a. Penicillins, the most frequent cause of drug-induced allergies

No. Yes, penicillins ARE the most frequent cause of drug-induced allergy, but they are excreted via renal tubular secretion, and need not be withheld from patients with liver disease. Go back to Item IX.

b. Tetracycline, which inhibits peptidyl transferase

No, you’ve confused tetracycline with drugs that bind to the 50 S ribosomal subunit, like chloramphenicol and erythromycin. Now, try again.

c. Chlortetracycline and oxytetracycline, which can cause liver damage

You’re exactly right. Both of these tetracycline derivatives are metabolized in the liver, and their toxicity is increased markedly if your patient is unable to metabolize them completely. Go on to Item X.

d. Aminoglycosides, some of which permit the development of malabsorption and/or superinfection by organisms not usually pathogenic in humans

No, streptomycin and its relatives are excreted chiefly in the urine. Go back to Item IX.

e. Erythromycin base, which can produce jaundice

No. Fortunately, only erythromycin esters, such as the estolate, produce cholestatic hepatitis. Erythromycin base, even if it cannot be concentrated in the liver for emptying in to the bile, is not sufficiently toxic to necessitate withholding it from even a jaundiced patient who is infected with bacteria that are sensitive to it. Go back to Item IX

X. Delayed hypersensitivity following the administration of penicillin is attributable to :

a. The drug's binding to red cell membranes

No, this is a cause of the hemolytic anemia that may, rarely, be caused by peniclillin, but not of skin rash. Try again.

b. The products of acid hydrolysis of the beta-lactam ring

No, no! This occurs only in the stomach. Try again.

c. The products of enzymatic hydrolysis of the beta-lactam ring

No, opening the beta-lactam ring results from the action of bacterial B-lactamase, and, unless the resulting penicilloic acid becomes a haptene, it will not cause hypersensitivity. Go back to Item X.

d. The products of enzymatic hydrolysis to penicillenic acid

No, penicillenic acid itself does not function as an antigen or as a haptene. Go back to Item X.

e. The conjugates of penicilloic acid with tissue proteins

Yes, you’ve remembered that it is penicilloic acid, a metabolite of penicillenic acid, which conjugates with terminal lysine groups to form the essential antigen complex in penicillin allergy.

This concludes this Workshop Program on Antibiotics. Obviously, it could have included hundreds of questions. Those you have answered exemplify a few of the problems you will have to answer in choosing a drug for your patients’ infectious diseases.