Control of B japonicum Symbiotic Genes by Two Linked Regulatory Cascades

On the basis of their regulation, symbiosis-related genes of B. japonicum can be divided into two groups which are controlled by two largely independent, oxygen-responsive regulatory cascades (Figure 1).

Genes for mfcro-¿iPrnbic life 'Ay'p e.g. fixNOQP

N; fixation genes e.g. nifDK, rufH

RegSR )«M"low02"

Figure 1. The regulatory cascades controlling symbiotic genes and accessory functions in B. japonicum. For details see text.

Genes for mfcro-¿iPrnbic life 'Ay'p e.g. fixNOQP

N; fixation genes e.g. nifDK, rufH

In the left cascade in Figure 1, oxygen sensing and transduction of the "low-oxygen" signal to the target genes is brought about by the FixLJ-FixK2 cascade. FixLJ is a classical two-component regulatory system consisting of the heme-based sensor kinase FixL and the FixJ response regulator (see Tuckerman et al. 2001 and references therein). The only known target of FixJ in B. japonicum is fixKi which encodes an FNR-type transcription activator that acts as a relay between FixLJ and the target genes (Nellen-Anthamatten et al. 1998). Currently, there is no evidence for an additional signal integrated by FixK2. Many target genes belonging to this group are not directly involved in the process of nitrogen fixation but rather are related to the microaerobic lifestyle of bacteroids living in root nodules. Examples include genes required for synthesis and activity of a cbb^-type terminal oxidase (fixNOQP,fixGHIS) which supports respiration in the microaerobic environment of root nodules (Preisig et al. 1993, 1996), heme biosynthetic genes (hemA; Page, Guerinot 1995); (hemB; Chauhan, O'Brian 1997); two hemN genes (Fischer et al. 2001) and genes for a symbiotic uptake hydrogenase (Durmowicz, Maier 1998). An additional target of the FixLJ-FixK2 cascade is one of two rpoN genes found in B. japonicum {rpoN\, Kullik et al. 1991). This gene encodes the specialized a factor, a54, which is required for activation of -24/-12-type promoters associated with the nif and fix genes regulated by the second cascade.

Target genes of the right cascade in Figure 1 comprise those directly involved in nitrogen fixation (e.g. genes required for synthesis and functioning of nitrogenase) but also genes of other (e.g. groESL chaperonins; Fischer et al. 1993) or unknown functions (Nienaber et al. 2001), The major oxygen-responsive component of this cascade is NifA which is active only under low-oxygen conditions. Unfortunately, the sensing mechanism of NifA is only poorly understood. Indirect evidence suggests that it resembles that of FNR which harbors a redox-responsive iron-sulfur cofactor (for review see Kiley, Beinert 1999). In vivo activity of B. japonicum NifA is dependent on four conserved cysteine residues and is inhibited by the presence of metal chelators (see Fischer 1994 and references therein). Moreover, we recently obtained preliminary evidence that activity of B. japonicum NifA (but not that of Klebsiella pneumoniae NifA) is reduced in an Escherichia coli strain lacking the cysteine desulfurase IscS required for in vivo iron-sulfur cluster formation (H.M. Fischer, unpublished; Schwartz et al. 2000).

NifA itself is subject to a dual control mechanism in B. japonicum. It is encoded in the fixR-nifA operon which is preceded by two overlapping promoters, PI and P2 (Barrios et al. 1995, 1998). The o54-dependent promoter PI is autoregulated by NifA under low-oxygen conditions (Thony et al. 1989). The activity of the P2 promoter is dependent on an upstream activator sequence centered at position -64 upstream of the fixR-nifA transcription start site (Bauer et al. 1998). It represents a binding site for the RegR response regulator of the superimposed RegSR two-component regulatory system. Under aerobic conditions, RegR mediates basal expression from the P2 promoter. Under low-oxygen conditions, fixR-nifA expression is increased approx. 5-fold. This is brought about not only by NifA-mediated activation of PI but also by the elevated activity of P2 under decreased oxygen conditions as recently shown with the help of a fixR-lacZ reporter fusion in which PI was inactivated by mutagenesis (M.A. Sciotti et al. this volume). It seems unlikely, however, that the RegSR system is responding directly to oxygen because of the lack of an obvious oxygen sensing moiety in RegS. Rather, we believe that the cellular redox-state is sensed indirectly via an alternative yet unidentified mechanism. Notably, evidence was recently presented that the RegSR-homologous system PrrBA of Rhodobacter sphaeroides is sensing the electron flow through the cbb^-type cytochrome c oxidase (Oh, Kaplan 2000), and the redox state of the quinone pool was shown to act as a signal for the ArcBA and AppA-PpsR redox-responsive systems of E. coli and R. sphaeroides, respectively (Georgellis et al. 2001; Oh, Kaplan 2000).

Was this article helpful?

0 0

Post a comment