Intracellular signalling: The chloroplast talks!

Dispatch

R307

Intracellular signalling: The chloroplast talks! Paul Jarvis

Plant cells have a unique problem: the coordination of three different genomes. While the dominance of the nuclear genome is indisputable, it is now clear that organellar signals can have profound effects, not just on nuclear gene expression but, as the Arabidopsis laf6 mutant reveals, also on whole plant morphology. Address: Department of Biology, University of Leicester, University Road, Leicester LE1 7RH, UK. Current Biology 2001, 11:R307–R310 0960-9822/01/$ – see front matter © 2001 Elsevier Science Ltd. All rights reserved.

It is now widely accepted that the ancestors of mitochondria and chloroplasts were free-living prokaryotic organisms capable of aerobic respiration and photosynthesis, respectively [1]. These organisms were engulfed by a plant cell progenitor in serial endosymbiotic events that each occurred over a billion years ago. Since then, endosymbiont genes have gradually been transferred to the nuclear genome so that, now, most genes encoding mitochondrial and chloroplast proteins reside in the nucleus. Organellar gene loss is incomplete, however, so both organelles retain a fully functional, endogenous genetic system and, as a consequence, a certain degree of autonomy. The existence within the same cell of three different genomes — in the nucleus, mitochondria and chloroplasts — presents a unique set of problems for plants and algae: the coordination of these genomes so that complex processes requiring input from multiple compartments, such as photosynthesis, can be elaborated efficiently and effectively in response to changing developmental and environmental cues. The coordination of events in the nucleus and the chloroplast — a type of plastid specialized for photosynthesis — necessarily involves the exchange of information, or intercompartmental signalling. Most information exchange between these organelles flows from nucleus to chloroplast, rather than vice versa, not least because >90% of chloroplast genes are in the nucleus. The import of nucleus-encoded proteins into chloroplasts — from the cytosol where they are made — constitutes a huge flow of information, and elaborate machinery has evolved to ensure that it occurs efficiently [2]. Although many of the proteins imported into chloroplasts are structural components — particularly of the photosynthetic apparatus and the chloroplast’s own genetic system — a large number perform regulatory functions, and serve to enforce nuclear control over the organellar compartment. Post-transcriptional regulation is prevalent in chloroplasts, and, in many cases, a single nucleus-encoded factor is specifically required for a

particular step in the expression of a single chloroplastencoded gene — a kind of ‘gene-for-gene’ interaction [3]. Despite all of the above, a large body of evidence indicates that the flow of information in a plant cell is not entirely unidirectional, and it is clear that signals from the chloroplast wield significant influence in the nucleus and the rest of the cell [4,5]. Early studies of chloroplast-tonucleus signalling used carotenoid-deficient mutant plants, or plants grown on carotenoid biosynthesis inhibitors such as the herbicide Norfluorazon [4]. Carotenoids prevent the formation of reactive oxygen species by quenching excited, triplet-state chlorophyll, and so chloroplasts deficient in these photoprotective compounds suffer massive internal photooxidative damage under intense light. The damage appears to be restricted to chloroplasts, and yet a subset of nuclear genes encoding chloroplast proteins is severely and specifically repressed as a result. This indicates that chloroplasts are able to communicate their functional status to the nucleus — the elusive ‘plastid signal’ — and that the nucleus responds by making appropriate changes in gene expression [4]. Møller and colleagues [6] have now demonstrated that the influence of the chloroplast extends far beyond the regulation of nuclear genes encoding products for its own direct use. In a genetic screen for Arabidopsis mutants with defects in developmental responses to light — photomorphogenesis — they isolated a novel mutant with a Ds transposon inserted into the nuclear gene for a chloroplast protein. The mutant is called long after far-red 6 (laf6) and, as the name suggests, its main phenotype is impaired hypocotyl growth inhibition in response to far-red light. Such long hypocotyl mutants usually have defects in photoreceptors — such as the phytochromes, which sense red and far-red light — or associated signal transduction components, and these all localize to the nuclear and/or cytosolic compartments [7]. Although this is not the first time that chloroplasts have been implicated in the transduction of light-signals controlling gene expression [8], it is the first time that a chloroplast protein has been shown to be required for the transduction of photomorphogenic signals. The LAF6 gene encodes a previously uncharacterized ATP-binding cassette (ABC) protein, called atABC1, with strong homology to ABC proteins from prokaryotes and lower eukaryotes, and an amino-terminal transit peptide for targeting to the chloroplast compartment. ABC proteins constitute a superfamily of solute transporters with representatives in all kingdoms [9]. These transporters have four domains — two ABC domains and two

R308

Current Biology Vol 11 No 8

Figure 1

Glutamate

5-Aminolevulinate

Protoporphyrinogen IX Protoporphyrinogen oxidase Protoporphyrin IX Fe-Chetalase

Haem Haem oxygenase

laf6

Mg-Chetalase

Mg-Protoporphyrin IX

hy1 Mg-Protoporphyrin IX monomethyl ester

Biliverdin IXα Phytochromobilin synthase 3Z-Phytochromobilin

3E-Phytochromobilin

gun5 brs-1, brc-1

hy2

Protochlorophyllide

Chlorophyllide

Chlorophyll Current Biology

The tetrapyrrole biosynthetic pathway of plants and algae. Chlorophyll and haem are synthesized in chloroplasts from glutamate by different branches of the same pathway. The branch point is defined by the enzymes Mg-chelatase (for the chlorophyll branch) and Fe-chelatase (for the haem branch). The phytochrome chromophore, 3E-phytochromobilin, is synthesized from haem. Steps in the pathway disrupted by the Arabidopsis hy1, hy2 and gun5 mutations, and the Chlamydomonas brs-1 and brc-1 mutations, are indicated. The Arabidopsis laf6 mutation is proposed to disrupt the indicated step indirectly.

membrane-spanning domains — that can exist within single or multiple polypeptides. The atABC1 protein has a single ABC domain and no membrane-spanning domain, suggesting that it functions as an importer rather than as an exporter — import systems tend to have ABC and membrane domains in separate subunits, whereas the opposite is true for export systems [9]. Given the homology between atABC1 and a cyanobacterial protein, it appears that import in this case means transport into chloroplasts. So what might atABC1 be importing into chloroplasts? The pale-green phenotype of laf6, the involvement of ABC proteins in the vacuolar transport of chlorophyll catabolites [10], and the implication of chlorophyll

precursors in the signalling of light responses [11–13], prompted Møller et al. [6] to investigate the levels of chlorophyll precursors in laf6 mutant plants. Interestingly, protoporphyrin IX (Figure 1) levels were found to be elevated two-fold in the mutant [6], and it was therefore proposed that atABC1 functions to import protoporphyrin IX from the envelope — where it is synthesized — so that chlorophyll biosynthesis can be completed inside the chloroplast (Figure 2). In laf6 mutant plants, protoporphyrin IX accumulates — as it is not accessible to downstream enzymes of the chlorophyll biosynthetic pathway — and leaks out across the outer envelope membrane. How does the perturbation of chlorophyll biosynthesis in laf6 mutant plants impair photomorphogenic responses to far-red light? One possibility is that protoporphyrin IX acts as a negative ‘plastid signal’ to attenuate the expression of light-regulated nuclear genes. This hypothesis is supported by the work of Johanningmeier and Howell [11] who, working with the unicellular alga Chlamydomonas, observed a negative correlation between factors presumed to cause chlorophyll precursor accumulation and chlorophyll a/bbinding (Cab) protein transcript levels. More recent advances in the plastid signalling field paint a more complex picture, however. For example, Beck and co-workers [12,13] have provided compelling evidence that chlorophyll precursors actually act as positive signals in the light-induction of nuclear gene expression. They showed that the Chlamydomonas brown mutants brs-1 and brc-1, which are defective in the synthesis of Mg-protoporphyrin IX, were unable to induce nuclear heat shock protein (HSP70) gene expression in response to light. This defect could be rescued by feeding Mg-protoporphyrin IX or its methyl ester to mutant cells, suggesting roles for Mg-protoporphyrin IX and/or its methyl ester as positive ‘plastid signals’. Unfortunately, no plastid signalling role for protoporphyrin IX could be demonstrated, as feeding protoporphyrin IX did not stimulate HSP70 gene expression in mutant cells [12] or downregulate HSP70 gene expression in light-grown wild-type cells [13]. Further work is necessary to resolve these apparent discrepancies — the use of non-photosynthetic genes in the latter study may be significant — but it seems that the interpretation of the laf6 data given above is an oversimplification. Recent work by Chory and colleagues [14] also points towards tetrapyrrole involvement in chloroplast-to-nucleus signalling. A genetic screen for Arabidopsis plastid signalling mutants was conducted back in 1993 [15]. The screen exploited the earlier observation that photooxidative chloroplast damage after growth on Norfluorazon results in the repression of nuclear genes for chloroplast proteins [4]. Chory and colleagues [15] fused the Cab gene promoter to an antibiotic resistance marker, introduced the chimeric gene into Arabidopsis plants, and then

Dispatch

screened for antibiotic-resistant mutants after photooxidation in the presence of Norfluorazon. Several genomes uncoupled (gun) mutants were isolated and, just recently, one of the GUN genes was cloned [14].

R309

Figure 2

Proto

laf6

Another interesting development is the demonstration that hy1 and hy2 mutants of Arabidopsis have gun phenotypes — gun2 and gun3 are in fact alleles of hy1 and hy2, respectively [14,16]. These mutants were originally identified by their long hypocotyl phenotypes, and it is now clear that they are both phytochrome-defective mutants. Phytochromes each comprise two 120 kDa polypeptides — of which there are five types in Arabidopsis — covalently attached to an invariant linear tetrapyrrole chromophore, phytochromobilin. Phytochrome A is the main photoreceptor for far-red responses, and so phyA apoprotein mutants have long hypocotyl phenotypes in far-red light [7]. Phytochrome B, on the other hand, is the dominant photoreceptor in red and white light, and so phyB apoprotein mutants have long hypocotyl phenotypes in red and white light, but not in far-red light [7]. The hy1 and hy2 mutants are defective in phytochromobilin synthesis — HY1 encodes a haem oxygenase [17,18], and HY2 encodes a phytochromobilin synthase [19] (Figure 1) — and so have phenotypes consistent with loss of function of all five phytochromes. Why do mutations in phytochromobilin synthesis disrupt plastid-to-nucleus signalling? The pale-green phenotypes of hy1 and hy2 provide a clue. The tetrapyrrole biosynthetic pathway is subject to extensive feedback regulation, and it seems that disruption of downstream steps in the haem branch can cause repression of very early steps in the pathway, so that flux through the chlorophyll branch is also reduced [20]. Indeed, double mutant studies indicate that hy1 affects the same plastid signalling pathway as gun5 [16], and so the effects of these mutations are probably mediated through the proposed Mg-chelatase ChlH sensor (Figure 3) [14]. So why does laf6, unlike gun5, affect nuclear gene expression and photomorphogenesis? Does the laf6 mutation affect phytochrome A apoprotein or, like hy1 and hy2,

oto

Pr

Cytosol

M PPO

atABC1

gen

ATP Protogen

Pro to

Remarkably, GUN5 turns out to encode the ChlH subunit of Mg-chelatase, one of the enzymes at the branch point of the tetrapyrrole biosynthetic pathway (Figure 1). Interpretation is once again not straightforward, however, as Arabidopsis Mg-chelatase ChlI mutants surprisingly do not have gun phenotypes, and Mg-protoporphyrin IX levels are reduced in the gun5 mutant — rather than elevated, as one might have expected given the results of Beck and colleagues [12,13]. These data led to the formulation of a model in which ChlH — located at the inner envelope membrane — measures flux through the chlorophyll biosynthetic pathway, and either sends a positive signal to the nucleus or inhibits a negative signal (Figure 3).

Proto ADP Chlorophyll Chloroplast

OE

IE Current Biology

A model illustrating the role proposed for atABC1 in protoporphyrin IX transport. Protoporphyrinogen IX (Protogen) is exported to the chloroplast envelope where it is oxidized to protoporphyrin IX (Proto) by protoporphyrinogen oxidase (PPO). Then, atABC1 — presumably acting in conjunction with some membrane spanning protein (M) — mediates the import of protoporphyrin IX into the stroma, where chlorophyll biosynthesis proceeds. In the Arabidopsis laf6 mutant, which lacks atABC1, protoporphyrin IX accumulates in the envelope and then leaks out into the cytosol. OE denotes outer envelope membrane, and IE denotes inner envelope membrane. (Adapted from [6].)

phytochromobilin biosynthesis? Apparently not. Phytochrome A apoprotein expression is normal in laf6 mutant plants and, in any case, the laf6 mutation clearly lies within the atABC1 gene [6]. What about the chromophore? Any perturbation of tetrapyrrole biosynthesis might affect phytochromobilin biosynthesis. But the laf6 phenotype is specific to far-red light and, what is more, cannot be rescued by feeding a chromophore substitute to mutant plants [6]. The only logical conclusion is that laf6 disrupts a plastid signalling pathway required for both photomorphogenic and gene-expression responses to far-red light. An obvious question, then, is: does the laf6 mutant have a gun phenotype? If it does, the relationship between the LAF6 and GUN5 plastid signalling pathways can be tested in double mutants studies. Clearly, plastid-to-nucleus signalling is a far more complex process than was previously assumed. Chlorophyll precursors are obviously involved in some way, but quite what their role is remains to be determined. Moreover, there appear to be multiple ways in which chlorophyll precursor ‘signals’ can be transduced to the nucleus: the laf6 mutation disrupts a pathway that controls gene expression and photomorphogenesis, whereas gun5 affects another pathway — or a branch pathway — that controls gene expression only. In addition, double-mutant studies involving the gun and hy1 mutants suggest that there are even multiple signalling

R310

Current Biology Vol 11 No 8

Figure 3

References

Chloroplast

Chl I D

ALA

MPM Proto

MP

H

Y

X

?

MP(M) ?

Cab Nucleus

Current Biology

A model illustrating the role proposed for the ChlH subunit of Mg-chelatase in plastid signalling. Mg-chelatase comprises three subunits — ChlH, ChlI and ChlD (H,I and D) — and catalyses the formation of Mg-protoporphyrin IX (MP) from protoporphyrin IX (Proto). Mutation of the ChlH subunit — as in the Arabidopsis gun5 mutant — results in a plastid signalling defect. ChlI mutations do not affect plastid signalling, indicating that the catalytic and plastid signalling roles of Mg-chelatase can be uncoupled. ChlH may act as a sensor to monitor flux through the chlorophyll biosynthetic pathway, relaying this information to the nucleus via positive and/or negative plastid signalling pathway(s). X and Y represent putative repressor and activator components controlling Cab transcription. Chlamydomonas feeding experiments suggest that Mg-protoporphyrin IX and its methyl ester derivative (MPM) may act directly — perhaps in conjunction with an unknown factor — as positive plastid signals. ALA denotes 5-aminolevulinic acid, and Chl denotes chlorophyll. (Adapted from [14].)

pathways for controlling gene expression [14,16], and other — perhaps entirely separate — signalling pathways relay information on the redox status of the chloroplast to the nucleus [21–23]. One clear message emerging from this increasingly complex picture, however, is that chloroplasts are not at all inert in their dealings with the nucleus — they have a fair say in the running of the cell. Acknowledgements I am grateful to Joanne Chory, Enrique López-Juez, Simon Geir Møller and Garry Whitelam for their helpful comments on the manuscript, and to the Royal Society for support from the Rosenheim Research Fellowship. Work in my laboratory on chloroplast biogenesis is supported by grants from the Biotechnology and Biological Sciences Research Council (C12976 and P12928) and the Nuffield Foundation (NAL/00026/G).

1. McFadden GI: Chloroplast origin and integration. Plant Physiol 2001, 125:50-53. 2. Keegstra K, Cline K: Protein import and routing systems of chloroplasts. Plant Cell 1999, 11:557-570. 3. Rochaix J-D: Posttranscriptional control of chloroplast gene expression. From RNA to photosynthetic complex. Plant Physiol 2001, 125:142-144. 4. Oelmüller R: Photooxidative destruction of chloroplasts and its effect on nuclear gene expression and extraplastidic enzyme levels. Photochem Photobiol 1989, 49:229-239. 5. Hedtke B, Wagner I, Borner T, Hess WR: Inter-organellar crosstalk in higher plants: impaired chloroplast development affects mitochondrial gene and transcript levels. Plant J 1999, 19:635-643. 6. Møller SG, Kunkel T, Chua N-H: A plastidic ABC protein involved in intercompartmental communication of light signaling. Genes Dev 2001, 15:90-103. 7. Neff MM, Fankhauser C, Chory JC: Light: an indicator of time and place. Genes Dev 2000, 14:257-271. 8. López-Juez E, Jarvis RP, Takeuchi A, Page AM, Chory J: New Arabidopsis cue mutants suggest a close connection between plastid- and phytochrome regulation of nuclear gene expression. Plant Phys 1998, 118:803-815. 9. Holland BI, Blight MA: ABC-ATPases, adaptable energy generators fuelling transmembrane movement of a variety of molecules in organisms from bacteria to humans. J Mol Biol 1999, 293:381-399. 10. Lu YP, Li ZS, Drozdowicz YM, Hortensteiner S, Martinoia E, Rea PA: AtMRP2, an Arabidopsis ATP binding cassette transporter able to transport glutathione S-conjugates and chlorophyll catabolites: functional comparisons with Atmrp1. Plant Cell 1998, 10:267-282. 11. Johanningmeier U, Howell SH: Regulation of light-harvesting chlorophyll-binding protein mRNA accumulation in Chlamydomonas reinhardtii. J Biol Chem 1984, 259:13541-13549. 12. Kropat J, Oster U, Rüdiger W, Beck CF: Chlorophyll precursors are signals of chloroplast origin involved in light induction of nuclear heat-shock genes. Proc Natl Acad Sci USA 1997, 94:14168-14172. 13. Kropat J, Oster U, Rüdiger W, Beck CF: Chloroplast signalling in the light induction of nuclear HSP70 genes requires the accumulation of chlorophyll precursors and their accessibility to cytoplasm/nucleus. Plant J 2000, 24:523-531. 14. Mochizuki N, Brusslan JA, Larkin R, Nagatani A, Chory J: Arabidopsis genomes uncoupled 5 (GUN5) mutant reveals the involvement of Mg-chelatase H subunit in plastid-to-nucleus signal transduction. Proc Natl Acad Sci USA 2001, 98:2053-2058. 15. Susek RE, Ausubel FM, Chory J: Signal transduction mutants of Arabidopsis uncouple nuclear CAB and RBCS gene expression from chloroplast development. Cell 1993, 74:787-799. 16. Vinti G, Hills A, Campbell S, Bowyer JR, Mochizuki N, Chory J, López-Juez E: Interactions between hy1 and gun mutants of Arabidopsis, and their implications for plastid/nuclear signalling. Plant J 2000, 24:883-894. 17. Muramoto T, Kohchi T, Yokota A, Hwang I, Goodman HM: The Arabidopsis photomorphogenic mutant hy1 is deficient in phytochrome chromophore biosynthesis as a result of a mutation in a plastid heme oxygenase. Plant Cell 1999, 11:335-348. 18. Davis SJ, Kurepa J, Vierstra RD: The Arabidopsis thaliana HY1 locus, required for phytochrome-chromophore biosynthesis, encodes a protein related to heme oxygenases. Proc Natl Acad Sci USA 1999, 96:6541-6546. 19. Kohchi T, Mukougawa K, Frankenberg N, Masuda M, Yokota A, Lagarias JC: The Arabidopsis HY2 gene encodes phytochromobilin synthase, a ferredoxin-dependent biliverdin reductase. Plant Cell 2001, 13:425-436. 20. Terry MJ, Kendrick RE: Feedback inhibition of chlorophyll synthesis in the phytochrome chromophore-deficient aurea and yellowgreen-2 mutants of tomato. Plant Physiol 1999, 119:143-152. 21. Escoubas JM, Lomas M, LaRoche J, Falkowski PG: Light intensity regulation of cab gene transcription is signaled by the redox state of the plastoquinone pool. Proc Natl Acad Sci USA 1995, 92:10237-10241. 22. Karpinski S, Escobar C, Karpinska B, Creissen G, Mullineaux PM: Photosynthetic electron transport regulates the expression of cytosolic ascorbate peroxidase genes in Arabidopsis during excess light stress. Plant Cell 1997, 9:627-640. 23. Oswald O, Martin T, Dominy PJ, Graham IA: Plastid redox state and sugars: interactive regulators of nuclear-encoded photosynthetic gene expression. Proc Natl Acad Sci USA 2001, 98:2047-2052.