Quick viewing(Text Mode)

Commentary a Brief History of Hemoglobins: Plant, Animal, Protist, and Bacteria Ross C

Commentary a Brief History of Hemoglobins: Plant, Animal, Protist, and Bacteria Ross C

Proc. Natl. Acad. Sci. USA Vol. 93, pp. 5675-5679, June 1996 Commentary A brief history of : , animal, protist, and Ross C. Hardison Department of and Molecular Biology, The Center for Gene Regulation, The Pennsylvania State University, University Park, PA 16802 Porphyrins, , , and Electrons fold (2). Further studies have found hemoglobins in jawless vertebrates and in diverse invertebrates ranging from flies The utility of metal-bound porphyrin rings for the transfer of () to earthworms () to nematodes (3-5). electrons was established early in , as witnessed by the These invertebrate hemoglobins are clearly related to those of ubiquitous that pass electrons down an electro- the vertebrates in primary structure, although in some inver- chemical gradient in the respiratory chain, eventually reducing tebrates, the large extracellular hemoglobins are fusion pro- 02 to H20 and yielding energy in the form of ATP in the teins composed of multiple copies of the familiar monomeric process. Other catalyze a variety of oxidations, . As hemoglobins are found in more and more distantly with roles as diverse as protecting cells from peroxides and related , the estimated time for the last common breaking down the very stable wood polymer, lignin. Even in ancestral gene moves further back to at least 670 photosynthesis, where solar energy is used to take electrons million years ago in the case of the invertebrate/vertebrate from a "low-energy" donor such as H20 (evolving 02) and pass divergence (6) (Fig. 1). them down an ATP-yielding electron transfer chain, the porphyrin rings in chlorophylls serve to harvest the quanta of Plant Hemoglobins: Symbiotic and Nonsymbiotic light energy and eventually feed them to the photosynthetic reaction center. Protoporphyrin IX with an iron ion coordi- not only make oxygen during photosynthesis, but they nately bound in the middle of its flat, planar structure (known also use it for respiration through the electron transfer chain as a molecule when the iron is in the 2+ oxidation state) in mitochondria. Recent studies show that they use hemoglo- is used in as diverse as cytochromes and ligninases for bins to bind and transfer that oxygen. The first plant hemo- just such electron transfers and oxidations, with the iron globins were discovered in the root nodules of (for switching cyclically between oxidation states. Other hemopro- review, see ref. 9). These nodules are a between teins have been adapted to allow the reversible binding of rhizobial bacteria and the plant to allow fixation (reduction) of oxygen to the heme, with the iron staying in its 2+ state. The atmospheric nitrogen into a usable form, eventually appearing proteins that carry the heme and facilitate the reversible in amino acids and other building blocks for the cells. Reduc- binding of oxygen are called hemoglobins. Hemoglobins are tion of nitrogen consumes large amounts of energy, and the usually thought of as the major proteins in erythrocytes nodules have an abundant, plant-encoded hemoglobin, called circulating in the blood of vertebrates, carrying that oxygen , that facilitates the of oxygen to the generated by photosynthesis to the far reaches of respiring respiring bacteriods in the (9). In addition, the tissues in the body. Indeed, hemoglobins were first found in binding of oxygen to leghemoglobin may help sequester the blood simply because they are so abundant, with a concentra- oxygen away from the nitrogen-fixing machinery, which is tion in normal blood of 15 g per 100 ml. However, it has readily poisoned by oxygen (2). Although the become clear that hemoglobins are very widespread in the sequences of differ from those of vertebrate biosphere and are found in all groups of including genes at about 80% of the positions, leghemoglobin prokaryotes, fungi, plants, and animals. Recent papers have folds into the same three-dimensional structure as the animal clarified some issues about the evolution and possible func- globins (3). Thus, it was clear that some plants had hemoglo- tions of these hemoglobins, although many issues remain bins, but initially they appeared to be limited to legumes. unresolved. So how could a selected subset of plants have a homolog to the vertebrate hemoglobins? Was a hemoglobin gene planted Animal Hemoglobins in a genome by horizontal gene transfer (10), possibly through an insect or nematode vector? Or are the leghemo- The hemoglobin that can be readily isolated from the blood of globins a specialized product of divergence from an ancient any vertebrate is a heterotetramer of two a-globin and two plant hemoglobin gene-a gene that is itself descended from 13-globin polypeptides, with a heme tightly bound to a pocket a hemoglobin gene in the last common ancestor to plants and in each globin monomer. The movements and interactions animals, and hence is still widespread in plants? between the a- and p-globin subunits lead to the cooperative The discovery of hemoglobins in a large variety of plants binding of oxygen to this heterotetramer, allowing it to pick up strongly supports the latter model. A hemoglobin distinct from oxygen readily in the lungs and to unload it efficiently in the leghemoglobin was initially discovered in root nodules of the peripheral respiring tissues. The amino acid sequences of the nonleguminous plant Parasponia andersonii (11). This is a a-globins and 3-globins are about 50% identical, regardless of relative of the elm tree that is nodulated by strains of Rhizo- which vertebrate species is the source, arguing that these two bium, and it is likely that the Parasponia hemoglobin also plays genes are descended from a common ancestor about 450 a role in oxygen transport during symbiotic . million years ago, in the ancestral jawed vertebrate (1). Both [Subsequent work indicates that this single has both a a-globins and P-globins are about equally divergent from the symbiotic and a nonsymbiotic role (12)]. But the discovery of monomeric , an oxygen storage and delivery protein a hemoglobin in a non-nodulating relative of Parasponia, found in many tissues. Myoglobin lacks the exquisite cooper- Trema tomentosa, suggested that hemoglobins are in fact ativity of the blood hemoglobins, but its relationship to them widespread in plants and can carry out more generalized roles is clear both from the primary sequence and from the virtually besides those in nodulation (13). Indeed, hemoglobins have identical three-dimensional structures-the classical globin now been found in many plants, not only in the dicots just 5675 Downloaded by guest on September 28, 2021 5676 Commentary: Hardison Proc. Natl. Acad. Sci. USA 93 (1996)

X b562 cytochrome b5 ? other hemoproteins K Ancestral Hb gene I - *or or - Bacterial Hb genes .- 1800 Myr ,-5FAD_ Fungal flavohemoglobin gene Alcaligenes FAD Yeast Saccharomyces 35 E14 F9 Hb E Algal gene I z Chlamydomonas F2 Protozoan Hb gene B12 E? G6 Paramecium 1500 Myr ^ Plants Animals B12 E14 G6 B12 E8 G6 * E * UEI- Nematode central intron 3.^J 670 Myr k ^650 Myr nonsvmbiotic symbiotic Hb gene /6 MVertebratey Hb gene lose all / |-L introns/ < 450 Myr AP 150 Myr Monocot 0'0 -globin gene -gobin I LI + Dicot * ** ^^H ca-globin gene P-globin I j gene Moncot ULCOtI Insect Mammal or Earthworm nonsymbioticI symbiotic Chironorr1US Homo Lumbricus Hordeum Glycine Nematode Mammal P barley Hb Dicot Hb Pseudoterranova Homo nonsymbiotic Glycine soybean Lb

FIG. 1. Some key events in the evolution of hemoglobin genes. Three different phases are shown at increasing levels of resolution. The top three lines indicate that several contemporary genes may be derived from a common ancestral gene, and some may be descended from others, such as the suggested derivation of hemoglobin genes from a cytochrome b5 gene. The next set of lines illustrates the descent of hemoglobin genes in bacteria, fungi, and protists from a common ancestral globin gene, following the cladogram in ref. 7. Globin exons are darkly shaded boxes, introns are open boxes, and the FAD-binding domain is a lightly shaded box. The predicted or known position in the a-helical globin fold of the amino acid whose codon is interrupted or followed by an intron in the gene is shown above the intron, e.g., B5 is the amino acid at the fifth position of helix B. Although current data do not address whether the ancestral globin gene contained introns, it is clear that introns were present in the hemoglobin gene in the last common ancestor to plants and animals. The events depicted in the "Plants" section summarize the data and cladogram in ref. 8. Selected events in the evolution of invertebrate and vertebrate hemoglobin genes are shown in the "Animals" section. The dates and branching patterns are taken from refs. 6 and 8. The lightly shaded diamonds represent gene duplications, the striped circles represent speciation events. Representative contemporary examples of the hemoglobin genes in plants, invertebrates, and vertebrates are given at the bottom of the figure. Myr, millions of years ago; Hb, hemoglobin. mentioned but also in the monocot cereals Hordeum (barley), This would suggest that all plants should have the nonsym- Triticum (wheat), and Zea (corn) (14). Thus, two different biotic hemoglobin, but previous studies with legumes have types of hemoglobin have been discovered in plants: (i) a revealed only the abundant leghemoglobins. In this issue ofthe nonsymbiotic type that is widely distributed among species and Proceedings, Andersson et al. (8) report the isolation of the (ii) a symbiotic type that is induced on nodulation of at least "missing" nonsymbiotic hemoglobin gene from legumes. two families of plants. Armed with sequence data for hemoglobins from nonlegumi- Downloaded by guest on September 28, 2021 Commentary: Hardison Proc. Natl. Acad. Sci. USA 93 (1996) 5677 nous plants, they successfully searched for a hemoglobin gene The presence of the central exon should be diagnostic of in the legume soybean (Glycine max) that is distinct from the relatives of the plant hemoglobin genes. Examination of he- well-known leghemoglobin genes found in the same plant. It is moglobin genes in invertebrates reveals a variety of structures. clearly not associated with symbiotic nodulation, being ex- The gene in the earthworm Lumbricus (an pressed in a wide range of tissues, including stems and young annelid) has two introns in homologous positions to those in leaves of mature plants, seed cotyledons, and young shoots. vertebrate hemoglobin genes (19). Of particular interest are Although mRNA from this gene is also present in root nodules, the reports of hemoglobin genes in the nematodes Pseudoter- it is much less abundant than that for the leghemoglobins. The ranova andAscaris that have not only the first and third introns expression pattern indicates that this gene has a more gener- homologous to those in vertebrate hemoglobin genes, but also alized function in plants, such as facilitating oxygen diffusion a central exon that interrupts the region coding for the E helix, to rapidly respiring cells. similar (but not identical) to the position of the plant hemo- The newly discovered soybean hemoglobin gene is much globin central exon (4, 5). more closely related to the hemoglobin genes found in non- Thus, one can propose that the ancestor to plants and leguminous plants than it is to the leghemoglobin genes, even animals had a hemoglobin gene with three introns (see Fig. 1). from the same species. This is seen in comparisons of both the This arrangement has been retained in all the plant hemoglo- encoded amino acid sequences of the proteins and in features bin genes, both symbiotic and nonsymbiotic, and also in certain of the promoters for the genes. In particular, the promoters for nematodes. The central intron was lost before the divergence leghemoglobin genes have a motif that is necessary for the of annelids and arthropods, and hence is absent in all verte- nodule-specific action of the gene, but the promoters of genes brate hemoglobin and myoglobin genes. The phenomenon of for the nonlegume plant hemoglobins and the new soybean intron loss was carried to an extreme in several hemoglobin hemoglobin lack this motif, having instead their own common genes in the midge Chironomus (an ), which lack conserved motifs. A cladogram based on the pairwise com- introns (20). Other nematode hemoglobin genes have lost one parisons of the amino acid sequences shows two distinct or more introns from the ancestral three intron structure (for branches: one with the symbiotic hemoglobins (characterized review, see ref. 21). by the leghemoglobins) and the other with the nonsymbiotic Although this model is attractive in its simplicity, the hemoglobins (including both the hemoglobins found in non- assignment of the central intron as homologous between plant leguminous plants and the newly discovered hemoglobin from and nematode hemoglobins is not definitive. Alignment of ). plant and nematode hemoglobin sequences shows several These observations of Andersson et al. (8) strongly support matches on both sides of the E helix, but the region interrupted the notion that a gene encoding the nonsymbiotic hemoglobin by the central intron is completely different. Simply starting was present in the ancestor to plants (see Fig. 1). The expres- from the predicted beginning of the E helix in the published sion data are consistent with a role in delivery of oxygen to alignments (5), the central intron in nematodes interrupts the respiring tissues, much like the familiar functions for animal eighth codon, whereas the central intron in plants falls between hemoglobins. It is likely that the symbiotic hemoglobins arose the fourteenth and fifteenth codons encoding this helix. Thus, through duplication of an ancestral gene followed by diver- the introns appear to be in slightly different places and in gence to fulfill more specialized functions in root nodules. The different phases. Whether this is the result of extensive diver- fact that nonsymbiotic hemoglobins from both dicots and gence from a common ancestor, resulting in intron "sliding," monocots are more closely related to each other than to or independent insertions of introns (22) requires further study symbiotic leghemoglobins (8) may suggest that the gene du- (discussed in ref. 21). Regardless ofwhether the central intron plication leading to symbiotic versus nonsymbiotic hemoglobin was present early or inserted later, the sequence relationships genes preceded the separation of monocots from dicots, and overall gene structures argue strongly for an ancestral approximately 150 million years ago (see Fig. 1). To date the hemoglobin gene being present more than 1500 million years symbiotic hemoglobins have only been seen in dicots, not as ago, before the divergence of plants and animals. This model widely distributed as the model in the Fig. 1 predicts. However, should be subjected to a rigorous statistical test. If verified, it is clear that the presence of the nonsymbiotic hemoglobin then it follows that the first and third introns are at least as old gene preceded the separation of dicots and monocots by quite as the last common ancestor to plants and animals. some time. Indeed, evidence based on number and positions of introns argues for a common ancestral gene for both plant and Hemoglobins in Protists, Fungi, and Bacteria animal hemoglobins. Recent studies show that the hemoglobin gene is truly ancient, A Common Ancestral Gene for Plant and preceding the divergence of prokaryotes and , and Animal Hemoglobins it now appears in a variety of exon/intron arrangements (see Fig. 1). The subfamily of hemoglobins found in protists forms The soybean nonsymbiotic hemoglobin gene is separated into a distinct branch on cladograms (7). The hemoglobin gene in four exons by three introns (8), like the genes for leghemo- the protozoan Tetrahymena has no introns, the homologous globins (15) and all other plant hemoglobins (see Fig. 1). The gene in Paramecium has a single intron that does not corre- first and third introns are in positions homologous to those of spond to any other globin gene intron, and one hemoglobin the two introns found in vertebrate a-globin and P-globin and gene in the algae Chlamydomonas has three introns, at least myoglobin genes. The second plant intron interrupts the region two of which are in unique locations (23). Because these coding for the E helix of hemoglobin. This central intron, in appear to be homologous genes, one must propose either many addition to the two introns in vertebrate globin genes, had been introns in the ancestral gene followed by differential loss, predicted by G6 (16) based on the domains or "modules" of substantial intron sliding, or repeated insertions of introns to hemoglobins, derived from the analysis of distances between obtain the' contemporary structures. The hemoglobins in the a-carbon atoms in the folded polypeptide chain. Its subse- Chlamydomonas are found in the chloroplast, are light- quent discovery in the genes for leghemoglobins supported the inducible, and may serve to trap oxygen and perhaps deliver it notion that exons encode discrete structural domains of proteins to the cytochrome oxidase of the respiratory chain (23). (17). However, more recent evidence of diversity of intron A flavohemoglobin from the yeast Saccharomyces is a fusion positions in other hemoglobin genes (see below) has led to a of a heme-binding domain and an FAD-binding domain (7); robust challenge to the exon theory of genes (18). like most genes in Saccharomyces, it has no introns. The yeast Downloaded by guest on September 28, 2021 5678 Commentary: Hardison Proc. Natl. Acad. Sci. USA 93 (1996) flavohemoglobin is induced by high oxygen conditions and has proteins could now serve as electron-transfer agents (leading been implicated in regulation or signaling (7, 24). to contemporary cytochromes) and possibly to scavenge scarce Hemoglobins have been found in many bacteria as well. Like oxygen to provide to the respiratory chain (leading to con- the yeast protein, a flavohemoprotein from Escherichia coli has temporary myoglobins and nonsymbiotic hemoglobins). Fur- two domains, one for binding each cofactor. The three- ther gene duplications and divergence would allow the capacity dimensional structure has been determined for a similar to catalyze other redox reactions to evolve. In multicellular protein from the strictly respiratory Gram-negative bacterium organisms, the oxygen-scavenging hemoglobins could evolve Alcaligenes eutrophus. Even though the amino acid sequence of into the abundant hemoglobins now used to transport oxygen. the heme-binding domain is quite different from that of plant Primary sequence relationships may not be particularly or animal hemoglobins, the structure corresponds to the useful in testing these proposed connections, because the classical globin fold, strongly arguing for homology between ancestral amino acid sequence may have diverged beyond the bacterial and eukaryotic hemoglobins (25).TheAlcaligenes recognition after billions of years of evolution. A more useful flavohemoglobin has been implicated in catalyzing a redox guide will be determination of three-dimensional structures by reaction, transferring a hydride ion from NADH to FAD and x-ray crystallography. In this regard, it is notable that the then the two electrons, through the heme, to a still-unknown light-harvesting biliprotein, C-, from the cya- substrate (25). The hemoglobin in the sliding bacterium Vit- nobacterium Mastigocladus laminosus has a three-dimensional reoscilla is not fused with a flavoprotein domain. In contrast to structure very similar to that of a globin (31), suggesting a the regulation of the yeast hemoglobin, it is induced 50-fold common ancestry. Although this is not a heme binding protein when cells are grown in hypoxic conditions. Its ability to per se, it does bind a linear tetrapyrrole pigment derived from complement deficiencies of terminal cytochrome oxidases in heme. Heme binds between two a-helices, coordinated to E. coli suggests that this hemoglobin can receive electrons during histidine, in proteins as diverse as lignin peroxidase (32) and respiration (26). A hemoglobin has also been found within a nif cytochrome b562 (33), but the topology of these helices differs operon in the cyanobacterium Nostoc commune (27). from the globin fold. Is this divergent or convergent evolution? Thus, hemoglobins are found in virtually all kingdoms of As more structures are determined, it will be highly instructive organisms, including eubacteria, unicellular eukaryotes, to see which distant relationships among hemoproteins will be plants, and animals. No hemoglobins have been reported in the confirmed, and determining how far the superfamily of he- archaebacteria to date, but it would be surprising if they were moglobin genes reaches. truly absent. There is a consistent function associated with many of the hemoglobins-the effective transport of oxygen. I thank Drs. J. Peacock, D. Goldberg, D. Bryant, G. Farber, A. This can occur by making very large amounts of hemoglobin Clark, and the members of my laboratory for discussion and helpful in particular cells (e.g., erythrocytes) for transport through the comments on this manuscript. body or in specialized locations requiring intense respiration (e.g., nitrogen-fixing root nodules). Alternatively, smaller 1. Goodman, M., Czelusniak, J., Koop, B., Tagle, D. & Slightom, J. amounts of proteins such as myoglobin or the nonsymbiotic (1987) Cold Spring Harbor Symp. Quant. Biol. 52, 875-890. in can be made in all 2. Dickerson, R. E. & Geis, I. (1983) Hemoglobin: Structure, Func- hemoglobins plants virtually cells, perhaps tion, Evolution and Pathology (Benjamin/Cummings, Menlo providing an efficient intracellular oxygen-delivery system to Park, CA). the respiring mitochondria and chloroplasts (28). This latter 3. Riggs, A. F. (1991) Am Zool. 31, 535-545. function appears to be very widespread in the biosphere and 4. Dixon, B., Walker, B., Kimmins, W. & Pohajdak, B. (1992)J. Mol. is found in plants, animals, and algae. When the respiratory Evol. 35, 131-136. electron transport system is not in a separate intracellular 5. Sherman, D. R., Kloek, A. P., Krishnan, B. R., Guinn, B. & compartment, as in bacteria, some species still use hemoglobin Goldberg, D. E. (1992) Proc. Natl. Acad. Sci. USA 89, 11696- under hypoxic conditions, perhaps for delivery ofoxygen as the 11700. terminal electron acceptor. Indeed, the flavohemoglobins 6. Goodman, M., Pedwaydon, J., Czelusniak, J., Suzuki, T., Gotoh, appear to catalyze redox reactions, with the heme playing a T., Moens, L., Shishikura, F., Walz, D. & Vinogradov, S. (1988) direct role in electron transfers as it does with the J. Mol. Evol. 27, 236-249. cytochromes. 7. Zhu, H. & Riggs, A. F. (1992) Proc. Natl. Acad. Sci. USA 89, Perhaps these latter proteins provide clues about the function 5015-5019. of the ancestral hemoglobins. 8. Andersson, C. R., Jensen, E. O., Llewellyn, D. J., Dennis, E. S. & Peacock, W. J. (1996) Proc. Natl. Acad. Sci. USA 93, 5682-5687. Common Origins for Many Hemoproteins 9. Appleby, C. A. (1984) Annu. Rev. Plant Physiol. 35, 443-478. 10. Jeffreys, A. J. (1982) in Genome Evolution, eds. Dover, G. & Is there an evolutionary connection between the hemoglobins, Flavell, R. (Academic, London), pp. 157-176. the cytochromes that pass electrons down an energy gradient 11. Appleby, C. A., Tjepkema, J. D. & Trinick, M. J. (1983) Science in respiration and in photosynthesis, and other hemoproteins 220, 951-953. that oxidations? Keilin some time 12. Landsmann, J., Dennis, E. S., Higgins, T. J. V., Appleby, C. A., catalyze (29) suggested ago Kortt, A. A. & Peacock, W.J. (1986) Nature (London) 324, that hemoglobins may have evolved from heme enzymes that 166-168. use oxygen (discussed in ref. 3). In support of the general idea 13. Bogusz, D., Appleby, C. A., Landsmann, J., Dennis, E. S., Trin- that some of these ancient hemoproteins may be related by ick, M. J. & Peacock, W. J. (1988) Nature (London) 331,178-180. divergent evolution, Runnegar (30) has proposed that globins 14. Taylor, E. R., Nie, X. Z., MacGregor, A. W. & Hill, R. D. (1994) may be derived, at least in part, from the bS cytochromes based Plant Mol. Biol. 24, 853-862. on similarities in the amino acid sequences (see Fig. 1). 15. Jensen, E. O., Paludan, K., Hyldig-Nielsen, J. J., Jorgensen, P. & Regardless of the relationships among the proteins that bind Marcker, K. A. (1981) Nature (London) 291, 677-679. them, it is likely that metal-bound porphyrin rings or related 16. G6, M. (1981) Nature (London) 291, 90-92. compounds were present at the time that photosynthesis 17. Gilbert, W. (1978) Nature (London) 271, 501. evolved; indeed, they may have been utilized then as now in 18. Stoltzfus, A., Spencer, D. F., Zuker, M., Logsdon, J. M., Jr., & One can on Doolittle, W. F. (1994) Science 265, 202-207. capturing light energy. speculate interesting 19. Jhiang, S. M., Garey, J. R. & Riggs, A. F. (1988) Science 240, scenarios for the use of hemoproteins once oxygen appeared 334-336. through photosynthesis. Perhaps they were initially used to 20. Antoine, M. & Niessing, J. (1984) Nature (London) 310,795-798. protect cells from the oxygen that now confronted them in the 21. Goldberg, D. E. (1995) BioEssays 17, 177-182. environment. Once the utility of oxygen as an electron accep- 22. Dixon, B. & Pohajdak, B. (1992) Trends Biochem. Sci. 17, tor was realized in the evolution of respiratory chains, hemo- 486-488. Downloaded by guest on September 28, 2021 Commentary: Hardison Proc. Natl. Acad. Sci. USA 93 (1996) 5679 23. Couture, M., Chamberland, H., St.-Pierre, B., Lafontaine, J. & 28. Wittenberg, B. A. & Wittenberg, J. B. (1987) Proc. Natl. Acad. Guertin, M. (1994) Mol. Gen. Genet. 243, 185-197. Sci. USA 84, 7503-7507. 24. Crawford, M. J., Sherman, D. R. & Goldberg, D. E. (1995) J. 29. Keilin, D. (1966) The History of Respiration and Cytochrome Biol. Chem. 270, 6991-6996. (Cambridge Univ. Press, Cambridge, U. K.). 30. Runnegar, B. (1984) J. Mol. Evol. 21, 33-41. 25. Ermler, U., Siddiqui, R. A., Cramm, R. & Friedrich, B. (1995) 31. Schirmer, T., Bode, W., Huber, R., Sidler, W. & Zuber, H. (1985) EMBO J. 14, 6067-6077. J. Mol. Biol. 184, 257-277. 26. Dikshit, R. P., Dikshit, K. L., Liu, Y. & Webster, D. A. (1992) 32. Edwards, S. L., Raag, R., Wariishi, H., Gold, M. H. & Poulos, Arch. Biochem. Biophys. 293, 241-245. T. L. (1993) Proc. Natl. Acad. Sci. USA 90, 750-754. 27. Potts, M., Angeloni, S. V., Ebel, R. E. & Bassam, D. (1992) 33. Mathews, F. S., Bethge, P. H. & Czerwinski, E. W. (1979)J. Biol. Science 256, 1690-1691. Chem. 254, 1699-1706. Downloaded by guest on September 28, 2021

Top View