For many years we have been studying how these polyamines are synthesized, how their biosynthesis and degradation are regulated, their physiologic functions and how they act in vivo. For this purpose we have constructed null mutants in each of the biosynthetic steps in both Escherichia coli and Saccharomyces cerevisiae, and have prepared over-expression systems for the biosynthetic enzymes. Previously, we have reported the construction of the complete deletion of all polyamine biosynthetic genes in E. coli. These E. coli cells grow at 40-50% of the normal growth rate in purified amine deficient medium in air. However, they are highly sensitive to oxidative and anaerobic stress. Our overall studies have aimed at the use of these mutants to elucidate the physiological functions of the polyamines and have involved studying the response of cultures of the polyamine-requiring mutants to the addition of polyamines. For our current studies we have developed the use of a chemostat where the growth is limited by factors other than the polyamines. As opposed to a number of previous studies reported from our own and other laboratories, we feel that it is very important not to have the results complicated by changes in the growth rates after polyamine addition since any changes in growth rate will necessarily have multiple effects other than those specifically resulting from the polyamine additions. Our most recent studies involve the use of this chemostat technique for microarray analyses on the effect of adding polyamines to our polyamine-deficient mutant. The results showed that 80 genes were up-regulated more than 2-fold and 25 genes were up-regulated more than 3-fold within a short time after the polyamine addition. Polyamine addition also caused more than 2-fold down-regulation of 51 genes. Particularly striking was the 2- fold to 5-fold increase in the expression of many genes in the E. coli acid resistance (AR) pathway (gadA, gadB, gadC, gadX, gadY, slp, ybaS, hdeD). This finding was of particular interest because to colonize the mammalian gastrointestinal tract, E. coli must survive passage through the acidic environment of the stomach. The most effective component of the E. coli AR system involves the two glutamate decarboxylase (GAD) proteins GadA, GadB and Glutamate-GABA antiporter GadC and their regulators. We then directly assayed the glutamic decarboxylase activity in response to acid stress in our polyamine mutant, and found that these cells had no glutamic decarboxylase activity unless polyamines were added. A dose response study showed that at least 100 micromolar putrescine or 10 microolar spermidine are needed to induce GAD activity in E. coli;cadavarine had no effect. To unravel the molecular mechanism, we looked for two important regulators of this pathway, rpoS and cAMP. This polyamine-requiring mutant was in the standard K12 strain that also has an amber mutation in the rpoS gene. Therefore since rpoS is known to be essential for the GAD pathway, we constructed a new polyamine-deficient strain that did not have the amber mutation in the rpoS gene. Although there is a report in the literature that inhibition of cyclic AMP synthesis is the primary effect of polyamine addition on gad gene expression, we have found, in contrast to this report, that polyamine addition to the polyamine deficient mutants causes an increase in the cyclic AMP levels. To extend our findings further, we repeated the above studies with the new polyamine-deficient strain that we constructed (lacking rpoS amber mutation) that also had deletions in either cyclic AMP synthesis or the cAMP receptor protein, and found that these strains even when acid stressed had no GAD activity in the absence of polyamines, but had very high activity in the presence of polyamines, indicating that the polyamine effect can occur in the absence of any cyclic AMP.. We plan to study the other regulatory genes in this acid island that are induced in our microarray study, and to investigate the molecular mechanisms of polyamine induced acid resistance in E. coli.
