Topic 17.3

The Origins of Phytochrome as a Bacterial Two-Component Receptor

Bacteria often use two-component signaling systems to sense and respond to environmental signals such as nutrients or light. The first component is a sensor protein, consisting of an input domain that receives the input signal, linked to a transmitter domain that transmits the signal to the second component. The second component is the response regulator protein, composed of a receiver domain that accepts the signal from the transmitter of the sensor protein, linked to an output domain that initiates the cellular response (see textbook Chapter 14).

These two components interact by a series of phosphorylation and dephosphorylation reactions. When it receives a signal, the sensor autophosphorylates a specific histidine residue in the transmitter domain. This phosphate is then transferred to a specific aspartate residue on the receiver domain of the response regulator (see textbook Chapter 14). But what has all this to do with phytochrome?

Although they lack the specific histidine residue that becomes autophosphorylated, the C termini of phytochromes have some sequence similarity to the transmitter domains of bacterial sensor proteins (Web Figure 17.3.A) (Schneider-Poetsch 1992). This sharing of sequences raises the interesting possibility that phytochrome evolved from the sensor protein of a bacterial two-component system.

Web Figure 17.3.A Homologous regions of phytochrome E, cyanobacterial photoreceptor, and sensor protein from Bacillus subtilis. The percentages indicate the proportion of identical amino acids between pairs of sequences. (After Yeh et al. 1997.)

Provocative new evidence for the idea that phytochrome evolved from the histidine kinase of a bacterial two-component system has recently emerged from studies of cyanobacteria. The cyanobacterium Fremyella uses light to regulate the composition of its photosynthetic pigments contained within the light-harvesting phycobilisomes (see textbook Figure 7.6, part C), a response termed chromatic adaptation.

Like phytochromes, phycobiliproteins bind linear tetrapyrrole chromophores that are closely related in structure to the phytochrome chromophore, but there is no evidence for red/far-red reversible responses of the type seen in higher plants. Instead, chromatic adaptation occurs in response to growth in red light and green light.

When the gene for the photoreceptor for this response (RcaE) was cloned, it was found to encode a 74 kDa polypeptide, the N-terminal end of which is related to the N terminus of higher-plant phytochromes, while the C-terminal end is related to histidine kinases (Kehoe and Grossman 1996). However, the lack of the conserved chromophore-binding site and of red/far-red reversibility suggests that this protein is a distant relative of higher-plant phytochromes.

Publication of the complete sequence of the genome of the cyanobacterium Synechocystis in 1996 led to great excitement because this organism contains a gene that has between 50 and 60% amino acid identity with Arabidopsis phyE along its length, including a bona fide chromophore attachment site. In addition, the amino acid sequence of the C terminus of the Synechocystis pigment protein is related to the KinA sensor protein of Bacillus subtilis (see Web Figure 17.3.A). This Synechocystis gene has been named cph1 (cyanobacterial phytochrome 1) (Yeh et al. 1997).

Recently it was shown that the purified Cph1 apoprotein can assemble with phycocyanobilin, a cyanobacterial linear tetrapyrrole, in vitro, and the resultant holoprotein exhibits red/far-red reversibility and absorption spectra similar (but not identical) to higher-plant phytochrome assembled in vitro with the same chromophore (Web Figure 17.3.B) (Hughes et al. 1997).

Web Figure 17.3.B Absorption spectra of cyanobacterial phytochrome after irradiation with red (R) or far-red (FR) light. Note the similarities to and differences from the plant phytochrome spectra shown in textbook Figure 17.3. (After Hughes et al. 1997.)

Perhaps the most exciting finding of all was the identification, 10 base pairs downstream from the cyanobacterial phytochrome gene, of another gene having the features of a bacterial response regulator (see Web Figure 17.3.A). Since the C-terminal end of the cyanobacterial phytochrome contains all the conserved features of a histidine kinase transmitter module, and the downstream response regulator resembles aspartate kinase receiver modules (see textbook Chapter 14), it was now possible to test whether this phytochrome acts as part of a light-regulated transmitter–receiver pair, as had been proposed earlier (Schneider-Poetsch 1992).

Using the purified proteins, the Cph1 protein was shown to phosphorylate itself in a manner consistent with its having a histidine kinase activity, and further, this phosphate could be transferred to the response regulator (Yeh et al. 1997). Surprisingly, Cph1 is far more active as a kinase in the Pr form than in the Pfr form, suggesting that the light signal is transduced through regulation of Pr abundance, rather than through Pfr abundance as in higher plants.

Taken together, the results from these studies suggest that plant phytochromes originally evolved from a class of photoreversible transmitter histidine kinases in cyanobacteria. Are plant phytochromes also kinases? In support of this idea, a phytochrome gene isolated from the moss Ceratodon contains an additional C-terminal sequence that has kinase homology (Thümmler et al. 1992).

In addition, highly purified preparations of higher-plant phytochrome have been found to contain a polycation-stimulated protein kinase activity that is also greater in the Pr form than in the Pfr form (Wong et al. 1989). However, the kinase activity associated with these phytochromes is of the serine–threonine type, and consistent with this observation, these proteins lack the conserved histidines of the bacterial sensor proteins. More recently, it was shown that recombinant oat phyA is a serine-threonine kinase with the same biochemical properties as the plant-derived photoreceptor (Yeh and Lagarias, 1998). Moreover, R-light stimulated its autophosphorylation, supporting the hypothesis that Pfr is the active form. Polycations were shown to stimulate Pr phosphorylation more than Pfr, a result consistent with previous studies.

Thus, while higher-plant phytochromes share a homology with bacterial two-component signaling systems, the reaction mechanism and components may have diverged considerably. Nevertheless, the discovery of cyanobacterial phytochromes has opened up many exciting new areas of research that have already contributed much to our understanding of phytochrome evolution and function.