This work was supported by grants from the European Union (BI04 CT 97-2143), the Ministère de l'Education Nationale, de la Recherche et de la Technologie (Programme de Recherche Fondamentale en Microbiologie et Maladies Infectieuses et Parasitaires), INRA (Programme prioritaire "Microbiologie") and the Ministère des Affaires Etrangères (PROCOPE 97158).


Y.-P. Wang1, Z.-X. Tian1, G. Wu1, M. Buck2, A. Kolb3

'College of Life Sciences, Peking University, Beijing 100871, People's Republic of China department of Biology and Biochemistry, Imperial College of Science, Technology and

Medicine, London SW7 2AZ, UK 3Laboratoire des Regulations Transcriptionnelles. FRE 2364 CNRS, Institute Pasteur, 75724 Paris Cedex 15, France

1. Introduction

The quality and quantity of carbohydrate available to the cell is the major limiting factor for the capacity of energy consuming nitrogen assimilation and nitrogen fixation. In enteric-bacteria, in response to the PTS system, the cAMP receptor protein (CRP) is considered as a proximal activator interacting at the short distances with the major form a70 RNA polymerase (Ea70), at promoters of sugar catabolic operons (reviewed in Kolb et al. 1993; Busby, Ebright 1999). In contrast, the initiation of transcription at nitrogen-responsive, a54 RNA polymerase (Ea54)-dependent promoters is activated by an entirely different mechanism; it requires nitrogen regulator NtrC-phosphate usually bound at distal enhancer sequences and hydrolysis of NTP (reviewed in Magasanik 1996).

Analysis of cr54-dependent dctA promoter reveals a novel negative regulatory function for CRP-cAMP in E. coli (Wang et al. 1993). CRP-cAMP is able to interact in cis from remote sites and in trans with the Ecr54-promoter closed complex. Moreover, such an interaction is kinetically linked to its repression effect (Wang et al. 1998). Due to the fact that CRP-cAMP can exert its effect on a 'core' promoter, which lacks a specific CRP-binding site, it is proposed that the effect might be general (Wang et al. 1998).

2. Results and Discussion

Among the nitrogen utilization regulons, the key promoter, glnApl is a54-dependent. It controls the expression of glutamine synthetase, the most important enzyme of nitrogen assimilation, and of NtrB and NtrC, the two regulatory proteins that controlling expression of the Ntr regulon (Reitzer, Magasanik 1986). The glnApl- and its upstream mutated/deleted derivatives-/acZ fusions (pKUlOO, pKUlOl and pKU102) were constructed and monitored under different conditions. Results show that in E. coli wild-type strain TP2101, the levels of the glnApl expression varied strongly when cells were grown on different carbohydrates. The levels are low when cells were grown on glycerol, and are high when grown on glucose (Table 1). It was previously reported that intracellular cAMP levels in cells growing on glucose are low (Dumay et al. 1996), while intracellular cAMP levels in cells growing on glycerol are high (Inada et al. 1996). It was decided, therefore, to investigate whether CRP-cAMP affected glnApl expression. The expression levels of glnApl (pKUlOl) were assayed in a cya mutant TP2006 and a cya crp double mutant TP2339, in the presence or absence of CRP-cAMP. When cells were grown on glycerol, the expression levels of glnApl were high in the absence of cAMP, and the expression levels were low in the presence of cAMP in the cya mutant TP2006 (Table 2). In contrast, the expression levels of glnApl were high in either presence or absence of cAMP, in cya crp double mutant TP2339, when cells were grown under similar conditions. Only when a plasmid carrying the crp gene (pLG339CRP) was used to complement TP2339's crp genotype, the repression effect of CRP-cAMP on glnApl was observed (Table 2). These results indicate that CRP represses glnApl in a cAMP-dependent manner.

Table 1. Carbon effect on glnApl expression.

Constructs glucose glycerol glucose glycerol pKUlOO CRP NtrC NtrC

V S-V_i=_I—S. 5866±44 885±18 9105±201 9599±122

glnApl glnApl lacZ

pKUlOl CRP NtrC NtrC

glnApl glnApl lacZ

Table 2. CRP-cAMP-mediated repression on glnApl.

Exogenous cAMP

Strain (genotype)



TP2006 (cya )


7558 ±59 354±13

TP2339 (cya crp)


6758±214 6626 ±67

TP2339 (cya crp)


6498+64 142 ±5

The glnApl promoter has CRP-binding site (centered at -186.5 from glnApl) located near, but not overlapping with the enhancer sequences for NtrC-P. It is required for CRP-cAMP-mediated activation of a70-dependent glnApl. When this CRP-binding site is mutated (pKUlOl), or even the entire upstream sequence is deleted (up to -40 of glnApl, pKU102), the CRP-cAMP-mediated repression effect remains (Table 2 for pKUlOl, data not shown for pKU102). These results indicated that specific CRP-binding site(s) are not essential for CRP-cAMP-mediated repression on glnApl, a result similar to that observed on the heterologous dctA promoter (Wang et al. 1998). We rule out the possibility that CRP might interfere with NtrC mediated transcriptional activation via heterodimer/aggregate formation with NtrC, or somehow interfere with NtrC activity by disrupting the histidyl-aspartyl phosphorelay between NtrB and NtrC proteins. This is done by replacing the activator NtrC with an alternative transcriptional activator, NifA, in the cya mutant TP2006 harboring pKUlOl or pKU102, and similar repression effects on glnApl and its derivatives were observed (data not shown).

In order to examine if CRP mediated repression on glnApl is at the transcriptional level, primer extension analysis was carried out. In cya mutant TP2006, when cells were grown without cAMP, it gave clear transcripts at +1 of glnApl. In contrast, these clear transcripts were absent when cells were grown with cAMP (data not shown). To further examine the basis of CRP-cAMP-mediated inhibition of glnApl, we assayed promoter DNA opening using in vivo KMnCU footprinting on the glnApl promoter. The results show that the presence of the CRP-cAMP complex prevents NtrC-dependent open complex formation on glnApl (data not shown). Similar results were obtained when NifA was used as the activator (data not shown). We conclude that the CRP-cAMP-

mediated inhibition on glnAp2 expression is at the transcriptional level and may occur through limiting NtrC and NifA-dependent DNA opening.

Was this article helpful?

0 0

Post a comment