Student Theses and Dissertations

Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)

RU Laboratory

Tomasz Laboratory


While the primary biochemical targets of β-lactams have been identified, the mechanisms by which inhibition of these targets (the membrane bound penicillin-binding proteins, or PBPs) leads to the irreversible antibacterial effects of these drugs have remained elusive. Treatment of Streptococcus pneumoniae with penicillin (or other cell wall inhibitors) results in a rapid sequence of events beginning with a slow-down and brief halt in bacterial growth, followed by rapid and irreversible reduction in the reproductive capacity (killing) of the cells, plus massive degradation (depolymerization and release) of the cell wall sacculus and release of cytoplasmic contents (bacterial lysis). By the end of the 1970s, a large body of observations clearly indicated that all these pathological events observable in penicillin-treated bacteria were secondary, indirect consequences of the inhibition of the primary (enzymatic) targets of penicillin (84). Exactly how and why inhibition of PBPs leads to slow-down and inhibition of growth followed by viability loss and lysis (in some bacteria and under some experimental conditions) has remained poorly understood. However, in pneumococci, the phenomenon of penicillin-induced cell wall degradation and lysis has been shown to result from the (drug-induced) deregulated activity of the pneumococcal major autolytic enzyme, a Nacetylmuramyl- L-alanine amidase (referred to as amidase)(80). The concurrence of culture lysis and loss of viability in these organisms has led to the widely held belief that the cause of V cell death was the unregulated (suicidal) activity of bacterial autolysins. On the other hand, in certain bacteria, penicillin can have powerful bactericidal activity without accompanying cell lysis (e.g., pneumococcal mutants with inactivated amidase (lytA) gene (80) and group A streptococci), suggesting that mechanisms other than bacterial autolysis may exist for the cidal effect of this antibiotic. We have investigated the existence of such autolysis-independent lethality in pneumococci. The results of these experiments are presented in four Chapters to be summarized next. Chapter I describes three types of experiments which suggest that only part of the penicillin-induced lethality is due to autolysis by amidase. (i) Suppression of penicillin-induced lysis by specific inhibitors of amidase protects pneumococci only marginally from killing, (ii) Mutants from which the amidase was completely eliminated by insertion-inactivation or deletion of the lytA gene are still killed, albeit at a slower rate than the wild-type Lyt+ strains, (iii) A new mutation (cid-) , not related to the amidase gene (lytA), causing massive reduction in killing was identified and characterized. "Triggering" of the amidase activity by penicillin in situ in growing bacteria is significantly reduced in the presence of the novel cid' mutation, indicating that there is a regulatory interaction between the cid gene product and the amidase Chapter II, describes several attempts to characterize the cid' mutation at the molecular level. Shotgun mutagenesis by insertion-inactivation (using either insertion-duplication or transposon mutagenesis) failed to yield Cid-mutants suggesting that null-mutants in the cid determinant(s) might not be viable. The possibility that cid is an essential gene is supported by the observation that cid- mutants had an extra nutritional requirement, inseparable in genetic crosses from the cid- mutation itself (i.e., growth of cid- mutants in a chemically defined medium required supplementation of the medium with a 500- 1000 kD molecular weight molecule present in yeast extract). Chapter III presents a step-by-step biochemical comparison of the principal penicillin targets of wild-type Cid+ parents and Cid- pneumococcal mutants, including the penicillin-binding proteins, the major autolysin amidase and the cell wall peptidoglycan. Physiological experiments suggest that cidr mutation is likely to operate at the level of the plasma membrane. Genetic blocks in the lytA (amidase) and cid genes can prevent both of the irreversible antibacterial effects of penicillin (lysis and killing) in pneumococci. Yet, these bacteria remain exquisitely sensitive to the drug, but addition of the antibiotic only causes inhibition of growth. The availability of the Cid- Lyt- double mutants allowed us to look closer at the mechanism of this reversible growth inhibitory effect of penicillin and Chapter IV describes studies in this direction. These studies show that penicillin treatment of both Cid+ and Cid" pneumococci results in the shutoff of protein and RNA synthesis, through some as yet unidentified regulatory circuit in the cells. In addition, antibiotic-treated pneumococci induce and or continue the specific radioactive labeling of a unique 72 kD protein with [35S]-cysteine, even when incorporation of radioactive amino acids in all other protein species has come to a virtual halt. The mechanism of radioactive labeling of the 72 kD protein in response to inhibition of synthesis of bacterial biopolymers is likely to involve addition of [35S]-cysteine to a preexisting protein through peptide bonds. The 72 kD species is also labeled in response to treatment with some other antibiotics, mechanistically unrelated to penicillin, but not during heat-shock. The 72 kD protein might be an "stress factor" implicated in the regulatory circuit mediating metabolic idling during antibiotic treatment. Taken together, the results presented in this dissertation underline the complexity of the pneumococcal response to treatment with penicillin. Analysis of Lyt- and Cid- mutants and double mutants has allowed a glimpse into several distinct pathological processes that are related to antibiotic-induced disintegration, viability loss, while the discovery of the 72 kD "stress factor" may serve as a signal for halt in growth and turning off cellular biopolymer synthesis. The major conclusion of the thesis work is that the primary mechanism of penicillin induced lethality is not bacterial lysis. It appears that during penicillin treatment of wild-type (Cid+ Lyt+) pneumococci, the bacteria are first killed and then lysed. Antibiotic-induced killing appears to be mediated by an alternative - membrane related - mechanism, which can be influenced by mutation(s) (cid-) independent of the determinant (lytA) of the major autolysin amidase. The interaction between the bacterial autolysis and the cid product may be analogous to the model proposed for the interactions between the membrane-active "channel" proteins (holins) and endolysins in bacteriophage induced lysis of bacteria (57,96).


A thesis submitted to the Faculty of the Rockefeller University in partial fulfillment of the requirements for the degree of Doctorate of Philosophy

Included in

Life Sciences Commons