Summary of research program
A large fraction of proteins require metals (e.g. Zn, Cu, Fe, Mn) or metal-cofactors (hemes, FeS centers, chlorophylls, cobalamin, molybdopterin) for function. The metal is invariably an important structural constituent of the protein, and in the proteins where it serves a catalytic role, it is essential for function. Metals and metal cofactors are found in every cellular compartment and they function in diverse metabolic pathways. In one genome (of a photosynthetic microorganism), cofactor/prosthetic group metabolism accounts for as much as 12% of its function. The chemical reactivity, exploited in biology to make desirable catalysts, can cause intracellular damage if it is not controlled. Metal metabolism is, therefore, subject to tight homeostatic regulation.
My group is addressing basic questions related to metal and metal cofactor metabolism. How is the abundance of a cofactor controlled by the cell? How is it distributed to various organelles? Is there a hierarchy of distribution when the cofactor might be limiting (as a result of genetic lesion or sub-optimal nutritional supply)? What are the mechanisms that ensure highly selective association between a polypeptide and its cofactor?
We have chosen, as our experimental model, two types of metal cofactors copper and heme, the former representing inorganic cofactors, whose abundance depends on nutritional supply, transport and intracellular storage, the latter representing organic cofactors, whose abundance is determined by the operation of a biosynthetic pathway. Both are redox-active cofactors found in quantity in electron transfer pathways, such as respiration and photosynthesis, which, in eukaryotic cells, are localized to the mitochondria and chloroplasts.
The general questions: In 1979, Paul Wood noted that Chlamydomonas cells could use either plastocyanin (a blue copper protein) or cytochrome c6 (a heme protein) for photosynthesis in vivo; the choice between them is determined by the supply of copper in the growth medium. If copper is provided, Chlamydomonas cells synthesize and accumulate plastocyanin; if the medium is copper-deficient, cytochrome c6 is produced instead. The organism remains photosynthetically competent regardless of the supply of copper in its growth environment. This is one of the most elegant examples of metal-responsive adaptation of a biochemical pathway, and my group set out to understand how it works. What is the signal? Is the cell measuring copper availability directly, or is it responding to a deficiency in photosynthetic electron flow? What are the targets of the signal transduction pathway (besides plastocyanin and cytochrome c6)? How is the response implemented? What is the sensor and how does it communicate with the targets? Can we understand the gene regulation phenomena in a cellular/physiological context? Dissection of this regulatory circuit in a system where it is possible to link molecular events with metabolic physiology is key to the development of fundamental principles underlying trace element homeostasis and metabolism.
Some answers: Plastocyanin and cytochrome c6 are each encoded by a single nuclear gene in Chlamydomonas reinhardtii (Merchant and Bogorad, 1987; Merchant et al., 1990). When copper availability is adequate to allow the synthesis of the full complement of plastocyanin (~ 8 x 106 molecules per cell), the Cyc6 gene encoding cytochrome c6 is transcriptionally silent (Merchant et al., 1991). In copper-depleted medium (< 8 x 106 Cu per cell), the Cyc6 gene is induced to allow cytochrome c6 synthesis to the extent and only to the extent -- required to compensate for the plastocyanin deficiency (Hill and Merchant, 1992). Plastocyanin deficiency in copper-depleted medium results from a block at the level of holoplastocyanin formation in the thylakoid lumen. The pre-protein is synthesized, targeted to the organelle and processed in the usual two-step pathway but lack of copper in the lumen prevents holoplastocyanin formation. Under these conditions, the apoprotein is degraded rapidly. The short half-life (under 20 minutes) is attributed not only to the increased proteolytic susceptibility of the apoprotein, which results probably from the reduced stability of the apopolypeptide, but also to activation of a protease in copper-deficient cells (Li and Merchant, 1995). Induction of the Cyc6 gene occurs by transcriptional activation in direct response to copper-deficiency (Quinn and Merchant, 1995). This is mediated by at least two copper-response element (CuREs) each contains at its core the sequence GTAC and each can function independently in the context of a reporter gene (Ars2). It is possible that the dynamic range of the response in vivo is ensured by the occurrence of multiple elements.
Besides plastocyanin and cytochrome c6 biosynthesis, there are several other adaptations to copper-deficiency (Hill and Merchant, 1995; Hill et al., 1996). Each of these -- transcriptional activation of the Cpx1 gene encoding coproporphyrinogen oxidase, activation of copper transport and a cell surface cupric reductase -- occurs in coordination with Cyc6 induction. Specifically, these responses display the same metal selectivity (Cu > Hg >> Ag) and sensitivity (nanomolar range) as does the Cyc6 gene and are targets of the same copper-sensing signal transduction pathway. The product of the Cpx1 gene is required in copper-supplemented as well as copper-deficient cells; therefore, unlike Cyc6, the gene is transcribed constitutively. In copper-replete cells, two transcripts accumulate; in copper-depleted cells, a third transcript with a distinct 5 end accumulates. Depending on the severity of copper deficiency, the - Cu form of the transcript can accumulate to 35-times the abundance of the two constitutive forms resulting in a large increase in the amount of coproporphyrinogen oxidase in - Cu cells. Transcriptional activation occurs in direct response to copper-sensing rather than to feedback regulation from the tetrapyrrole pathway. Analysis of Cpx1-Ars2 reporter gene reveals only a single CuRE: the sequence GTAC is critical for its activity (Quinn et al., 1999).
With a view to identifying additional targets of the copper-sensing system and also the signal transduction components, we began a genetic screen for copper-conditional growth defects, leading to the discovery of the CRT1 (for copper response target) and CRR1 (for copper response regulator) loci. The crt1 mutants display a chlorophyll deficiency and loss of photosystem I exclusively under copper-deficient conditions. The pattern of plastocyanin, cytochrome c6 and coprogen oxidase accumulation is normal in all crt1 strains indicating that the CRT1 does not encode a regulatory component; rather we propose that it defines a hitherto unknown aspect of photosystem I biochemistry. The crr1 mutants are likely to be regulatory mutants because they do not induce Cyc6 and Cpx1 and are unable to grow in copper-deficient medium.
The reciprocal copper-responsive accumulation of plastocyanin and cytochrome c6 is viewed as an adaptive mechanism, which allows the organism to survive occasional copper-deficiency for instance, during an algal bloom. Since plastocyanin is the major sink for copper in the green algae (which lack a Cu/Zn superoxide dismutase and other abundant copper proteins), its degradation in copper-deficiency would ensure the re-distribution of this essential trace element to cytochrome oxidase in the respiratory pathway. Indeed, loss of plastocyanin precedes loss of cytochrome oxidase as Chlamydomonas reinhardtii cells become copper-deficient.
Specific questions for the immediate future: The immediate objective of our ongoing studies in Chlamydomonas is to understand the molecular basis of copper measurement by the cell through the identification of the copper-sensor and the signal transduction components. Our work suggests that copper utilization and distribution is controlled tightly and by specific regulatory mechanisms so that it is provided where it is most needed for biochemical metabolism. We will identify copper-handling components in Chlamydomonas (such as assimilatory transporters, delivery molecules, and chaperones) through functional complementation of appropriate Saccharomyces mutants with the intent of monitoring intracellular copper flux in response to supply. The repertoire of molecular and genetic tools at our disposal will be applied to build a dynamic picture of homeostatic copper metabolism in a eukaryotic cell. The recent discovery of copper chaperones and delivery mechanisms in Saccharomyces cerevisiae enables us to test the general applicability of our model. Specifically, we will determine the hierarchy of distribution of copper to the respiratory pathway (cytochrome oxidase) vs. the oxidative stress pathway (superoxide dismutase) in Saccharomyces. More recently, we have discovered an oxygen-responsive signal transduction pathway in Chlamydomonas, which may use copper chemistry to sense the intracellular redox level. We are interested in understanding the physiological basis for the response as well as the molecular constituents of what could be a unique oxygen sensor.
The general questions: Heme is found in many different locations including the mitochondrion (respiratory cytochromes), the chloroplast (photosynthetic cytochromes), the endoplasmic reticulum (P450 cytochromes), the nucleus (heme-responsive transcription factors), the cytosol (catalases), the peroxisomes (catalases) and extracellular compartments (peroxidases). Its synthesis is restricted to the mitochondrion in a typical eukaryotic (non-plant) cell. How is heme transported from the mitochondrion? How is it delivered from its site of synthesis to the site where heme proteins are made? What is the mechanism of cofactor protein association? Is heme allocated preferentially to particular pathways in response to metabolic demand? The occurrence of two pathways (plastidic and mitochondrial) adds an extra layer of complexity in a plant cell. Which one provides heme to extra-organellar proteins? Is there a mechanism to control the flux out of the plastid vs. the mitochondrion?
Our discoveries: The genetic dissection of specific assembly pathways was envisioned as a first step towards the discovery of heme chaperones and transporters the expected regulatory targets. We chose the cytochromes of the b6/f complex as our model for the study of heme protein assembly in chloroplasts. The b6/f complex contains two types of cytochromes cytochrome b6, an integral membrane protein with two bis-histidyl liganded b-hemes (bH and bL), and cytochrome f, a membrane-anchored protein with a single covalently attached c-heme.
In the c-cytochromes, heme is covalently attached to the polypeptide through thioether linkages. Thus, the fact that assembly is catalyzed was appreciated decades ago. Both biochemical and genetic approaches have been applied to the study of cytochrome c assembly in bacteria, mitochondria and chloroplasts, and common principles have emerged (reviewed in Howe and Merchant, 1994). In general, the c-cytochromes are synthesized as precursor proteins that are targeted post-translationally to their site of function. Heme attachment is a terminal (or near terminal) step in their maturation, occurring on the p-side of the energy transducing membrane where the cytochromes function. This demands a mechanism for heme delivery from its site of synthesis to its site of utilization for holocytochrome c formation.
The pathway for pre-apocytochrome c6 and f maturation in the chloroplast was deduced by the application of pulse-chase methods (Howe and Merchant, 1994; Howe et al., 1995). A subset of cytochrome-deficient mutants was found to be blocked specifically at the terminal step of heme attachment (Howe and Merchant, 1992; Xie et al., 1998). These mutants display a pleiotropic deficiency in c-type cytochromes and define a minimum of 6 loci, ccsA in the plastid and CCS1 through CCS5 in the nucleus. Ccs1 and ccsA encode multi-spanning proteins in the thylakoid membrane (Xie and Merchant, 1996; Inoue et al., 1997). The proteins are found in a 200 kD complex, probably in association with the products of other Ccs genes. We predict that CcsA functions in a novel heme delivery pathway, while Ccs1 might be an apoprotein chaperone.
In collaboration with the Wollman group (Institut de Biologie Physico-Chimique, Paris), we deduced a pathway for the stepwise association of the bL and bH hemes of holocytochrome b6 in vivo (Kuras et al., 1997). Genetic analysis of a subset of cytochrome b6/f-deficient strains indicates a minimum of 4 CCB loci that function solely at the bH-heme insertion step. These results dispelled the long-standing dogma that b cytochrome assembly occurred without catalysis in vivo. It is our view that specific mechanisms exist for each type of metalloprotein to ensure selective cofactor ligation in vivo; these will undoubtedly be discovered through the more detailed characterization of assembly-defective respiratory and photosynthetic mutants.
Immediate plans: The Ccs factors appear to be novel proteins whose biochemical functions cannot be ascertained through sequence analysis. A key objective is to deduce their specific catalytic roles in heme and apoprotein metabolism in the thylakoid lumen. Eventually, we would like to identify all Ccs and Ccb factors and reconstitute the assembly of cytochromes b and c in vitro. Analysis of the components of the three c-cytochrome assembly pathways occurring in plastids, mitochondria and bacteria (Kranz et al., 1998) predicts the existence of a specific heme export pathway in Saccharomyces and mammalian mitochondria. Nevertheless, one is not evident in the yeast genome. Nor, despite considerable effort, has it been revealed through classical "loss of function" genetic approaches. Yet, heme export from the mitochondrion is a fundamental and important metabolic function. We are presently engaged in a "gain of function" multi-copy suppressor approach for its identification in Saccharomyces, which should open the door to understanding heme flux in a eukaryotic cell.