Article
Reference
Molecular and Biophysical Analysis of Herbicide-Resistant Mutants of Chlamydomonas reinhardtii : Structure-Function Relationship of
the Photosystem II D1 Polypeptide
ERICKSON, Jeanne Marie, et al .
Abstract
Plants and green algae can develop resistance to herbicides that block photosynthesis by competing with quinones in binding to the chloroplast photosystem II (PSII) D1 polypeptide.
Because numerous herbicide-resistant mutants of Chlamydomonas reinhardtii with different patterns of resistance to such herbicides are readily isolated, this system provides a powerful tool for examining the interactions of herbicides and endogenous quinones with the photosynthetic membrane, and for studying the structure-function relationship of the D1 protein with respect to PSII electron transfer. Here we report the results of DNA sequence analysis of the D1 gene from four mutants not previously characterized at the molecular level, the correlation of changes in specific amino acid residues of the D1 protein with levels of resistance to the herbicides atrizine, diuron, and bromacil, and the kinetics of fluorescence decay for each mutant, which show that changes at two different amino acid residues dramatically slow PSII electron transfer. Our analyses, which identify a region of 57 amino acids of the D1 polypeptide involved in herbicide [...]
ERICKSON, Jeanne Marie, et al . Molecular and Biophysical Analysis of Herbicide-Resistant Mutants of Chlamydomonas reinhardtii : Structure-Function Relationship of the Photosystem II D1 Polypeptide. Plant Cell , 1989, vol. 1, no. 3, p. 361-371
DOI : 10.1105/tpc.1.3.361 PMID : 2535507
Available at:
http://archive-ouverte.unige.ch/unige:148885
Disclaimer: layout of this document may differ from the published version.
1 / 1
Molecular and Biophysical Analysis of Herbicide- Resista n t M u t a n t s of Chlam ydomonas reínhardtií :
Structure-Function Relationship of the Photosystem II D 1 Poly peptide
Jeanne M. Erickson,ai b,' Klaus Pfister," Michele Rahire,b Robert K. Togasaki," Laurens Mets,' and Jean-David Rochaixb
a Department of Biology, University of California,
Los
Angeles, California 90024Departments of Molecular Biology and Plant Biology, University of Geneva, CH1211, Geneva 4, Switzerland Ciba-Geigy AG, Agricultura1 Division, CH4002, Basel, Switzerland
Department of Biology, Indiana University, Bloomington, Indiana 47405
e Department of Molecular Genetics and Cell Biology, University of Chicago, Chicago, lllinois 60637
Plants and green algae can develop resistance to herbicides that block photosynthesis by competing with quinones in binding to the chloroplast photosystem II (PSII)
D1
polypeptide. Because numerous herbicide-resistant mutants of Chlamydomonas reinhardtii with different patterns of resistance to such herbicides are readily isolated, this system provides a powerful tool for examining the interactions of herbicides and endogenous quinones with the photosynthetic membrane, and for studying the structure-function relationship of the D1 protein with respect to PSll electron transfer. Here we report the results of DNA sequence analysis of theD1
gene from four mutants not previously characterized at the molecular level, the correlation of changes in specific amino acid residues of theD1
protein with levels of resistance to the herbicides atrizine, diuron, and bromacil, and the kinetics of fluorescence decay for each mutant, which show that changes at two different amino acid residues dramatically slow PSll electron transfer. Our analyses, which identify a region of 57 amino acids of the
D1
polypeptide involved in herbicide binding and which define aD1
binding niche for the second quinone acceptor, Qe of PSII, provide a strong basis of support for structural and functional models of the PSll reaction center.INTRODUCTION
Photosynthesis, the chemical process by which light en- ergy is converted into chemical emergy, takes place in the chloroplast of plants and algae. Photosystem II (PSII) is the thylakoid membrane protein-pigment complex that me- diates the initial light-driven reactions in which electrons are removed from water, yielding oxygen and hydrogen ions (for general review, see Clayton, 1980; Dismukes, 1988). The electrons are transferred through a series of intermediate acceptors to the first and second stable qui- none acceptors of PSII, Qn and Qe, respectively (Velthuys, 1981). Much recent work has focused on describing the structural composition of the PSll complex, and, in partic- ular, of the PSll reaction center. Our interest lies in eluci- dating how the structure of the reaction center relates to its function in mediating the transfer of electrons from water to Qe.
A number of chemically different classes of herbicides block the photosynthetic electron transpor? chain on the
'
To whom correspondence should be addressed.reducing side of PSII. These include the s-triazines, sub- stituted ureas, and phenolic derivatives, all of which can be used as probes to study the function of the thylakoid membrane (for reviews, see Trebst, 1980; Arntzen et al., 1982). The different herbicides compete with each other and with quinone for binding to thylakoid membranes (Tischer and Strotmann, 1977), and they appear to disrupt electron transfer by displacing the bound plastoquinine, Q, (Vermaas et al., 1983). Photoaffinity labeling of thylakoid membranes with ''C-azidoatrazine indicates that the her- bicide analog binds, in herbicide-sensitive plants and algae, and with greatly reduced affinity in herbicide-resistant strains, to a PSll polypeptide of 32,000 apparent molecular weight (Pfister et al., 1981; Steinback et al., 1981; Tellen- bach et al., 1983; Erickson et al., 1984a). This polypeptide, called D1 in Chlamydomonas reinhardtii (Chua and Gillham, 1977), is encoded by the chloroplast psbA gene (Zurawski et al., 1982), is highly conserved in all species examined to date (see Erickson et al., 1985b, for review), and is an integral part of the active PSll reaction center core (Barber
et al., 1987; Danielius et al., 1987; Marder et al., 1987;
Nanba and Satoh, 1987).
The role of the D1 polypeptide in determining resistance to triazine herbicides was demonstrated by the isolation and characterization of psbA from plants and algae resist- ant to high levels of the herbicide atrazine. DNA sequence analysis of these mutant genes showed that a single base pair change in psb A, compared with the sequence of psbA from herbicide-sensitive strains, resulted in a single amino acid substitution at residue 264 which changed the wild- type serine residue to alanine in Chlamydomonas (Erickson et al., 1984a) and to glycine in Amaranthus hybridus and Solanum nigrum (Hirschberg and Mclntosh, 1983; Hirsch- berg et al., 1984; Goloubinoff et al., 1984). Transformation of cyanobacteria with a mutant cyanobacterial psbA gene coding for alanine at position 264 produced herbicide- resistant cyanobacteria (Golden and Haselkorn, 1 985) and provided definitive confirmation that such mutations result in the herbicide-resistant phenotype. These molecular ge- netic results were consistent with classical genetic evi- dente that the herbicide-resistance trait is maternally in- herited (Souza-Machado et al., 1978,; Galloway and Mets, 1982, 1984; Erickson et al., 1984a) and, coupled with the physical characterization of the herbicide-resistant and sensitive strains, strongly suggested that the psbA gene product was the “herbicide-binding” protein and served as the Q,-binding protein of PSII.
Our previous work in characterization of herbicide-re- sistant mutants of C. reinhardtii revealed additional muta- tion sites in psbA that alter D1 amino acid residues 219 and 255 and result in unique patterns of resistance to the herbicides atrazine and diuron (Erickson et al., 1985a). In contrast to the changes described at residue 264 which dramatically slow electron transfer in both plants and algae (Bowes et al., 1980; Erickson et al., 1984a), the amino acid substitutions at residues 219 and 255 do not appear to affect the rate of Qa to Qe electron transfer (Galloway and Mets, 1984). Hence, herbicide-resistant mutants that have single amino acid substitutions in the D1 protein are important tools for studying PSll electron transfer and, indirectly, quinone binding. Unlike the photosynthetic bac- teria1 reaction centers, which have been successfully crys- tallized and analyzed by x-ray diffraction (Deisenhofer et al., 1985; Allen et al., 1987), the chloroplast PSll reaction center has resisted a similar type of analysis, and PSll crystals that can be effectively analyzed by x-ray diffraction have not been obtained. Thus, the combined biochemical, molecular, and genetic approach provides a powerful means of dissecting structure-function relationships in the PSll reaction center.
To further explore the functional regions of the D1 protein and to delineate the extent of the Qs-binding site more precisely, we have characterized four additional uni- parenta1 mutants of Chlamydomonas. Here we report the results of sequence analysis of the D1 gene, psbA, from
four mutants, Ar204, Br24, Br202, and Drl8, and the identification of two new mutation sites which result in amino acid substitutions at residues 256 and 275 of the D1 protein. In addition, we have studied the rates of electron transfer between QA and Qe, in these four mutants and in the three we previously characterized, by observing fluorescence induction kinetics and fluorescence decay kinetics. (In the strict sense we mean decay in the fluores- cence yield, but we use the term fluorescence decay for simplicity.) We have also compared the relative resistance of all seven mutants to the herbicides atrazine, bromacil, and diuron. These data allow us to correlate the presence of specific amino acid substitutions in D1 with herbicide resistance patterns and with the rate of QA to QB electron transfer.
RESULTS
lsolation and Genetic Characterization of Herbicide- Resistant Mutants
Dr18 was obtained by mutagenizing C. reinhardtii wild- type (wt) strain 137c(+) with N-methyl-N ‘-nitro-N-nitroso- guanidine (Gillham, 1965) and selecting on diuron (DCMU) gradient plates of solid minimal medium as previously described by McBride et al. (1977). Br24 was obtained following a similar procedure, except that bromacil gradient plates were substituted for the DCMU gradient plates. In both cases, no attempt was made to reduce the copy number of the chloroplast genome by treatment with 5- fluoro-2’-deoxyuridine before mutagenesis. Analysis of 1 7 tetrads from matings of Dr18 with wt cells and eight tetrads from matings of Br24 with wt cells showed a uniparental mode of inheritance for both herbicide-resistant pheno- types. In addition, both mutants showed resistance to herbicide in vitro, monitored by Hill activity of chloroplast membrane preparations (Lien et ai., 1977). The isolation and phenotypic characterization of uniparental herbicide- resistant mutants of C. reinhardtii selected for their resist- ance to atrazine (mutants Ar204, Ar207, Galloway and Mets, 1984), diuron (mutant DCMU4, Erickson et al., 1984a; mutant Dr2, Galloway and Mets, 1984), and bro- macil (mutant Br202, Galloway and Mets, 1984) have been previously reported as noted.
Molecular Characterization of psbA from Four Herbicide-Resistant Mutants
psbA, the chloroplast gene coding for D1, contains five exons spanning approximately 7 kb and is present in two copies per circular chloroplast molecule of C. reinhardtii (Erickson et ai., 1984b). Total DNA from the mutants was
isolated and digested with BamHl and Bglll, generating BamHI-Bglll restriction fragments of approximately 1 O kb and 12 kb which each contain one copy of psbA. Digested DNA from Ar204, Br202, Br24, and Dr18 was cloned in Escherichia coli using plasmid vectors as described in
“Methods,” and recombinant clones containing psbA were identified by colony hybridization. Because all three of the herbicide-resistant mutation sites previously characterized (Erickson et al., 1984a, 1985a) were localized in exons 4 and 5 of psbA, only these two exons were sequenced for each of the four new mutants listed above, using the Maxam-Gilbert sequence strategy diagrammed in Figure 1 and described in “Methods.” We find that an adenine to cytosine transversion changes the wild-type leucine codon 275 to phenylalanine in mutant Br202. Mutants Ar204 and Br24 both have an identical mutation in which a guanine to adenine transition changes the wild-type glycine codon 256 to aspartic acid. These latter two mutants were iso- lated in different laboratories and selected for resistance to different herbicides (atrazine in the case of Ar204, bromacil in the case of Br24). Sequence analysis of psbA from mutant Drl8 reveals a guanine to adenine transition which changes valine codon 219 to isoleucine in the mu- tant. This same change was reported previously in the independently isolated diuron-resistant mutant Dr2 (Erick- son et al., 1985a). Table 1 shows the results of psbA sequence analysis of the four new mutants, as well as the three previously characterized, and summarizes the change deduced in the D1 protein for each mutant. Except for the single mutation observed in each mutant as indi- cated in Table 1, no other changes were found in exons 4 and 5 in any of these mutants. In contrast to the extensive molecular analysis of our previous work, no attempt was
Hf Hf .
.-._
K303 h 175
Figure 1. Restriction Map of psbA from C. reinhardtii.
The map shows the five exons (I to V, black boxes) and the Maxam-Gilbert DNA sequencing strategy for exons 4 and 5 contained in the 1300-bp Xbal-EcoRI fragment and the 1100-bp Hindlll-Kpnl fragment, respectively. The sizes of the four small DNA fragments sequenced are indicated in base pairs. Arrows show the approximate extent of sequencing from each labeled end. Each fragment was labeled separately at both the 5‘ and 3’
end, so that both strands corresponding to the entire exon 4 and 5 regions were sequenced for each mutant. Restriction sites are as follows: D, Ddel; H, Hindlll; Hf, Hinfl; K, Kpnl; R, EcoRI; X, Xbal.
Table 1. psbA Mutations in Herbicide-Resistant Strains of C.
reinhardtii
Strain psbA Sequence Codon Change Position WT Dr2, DR18
WT Ar207
WT Ar204, Br24
WT DCMW
WT Br202
tta GTT act Val + Ile 21 9 tta ATT act
tac TTT ggt Phe -+ Tyr 255 tac TAT ggt
ttt GGT cgt Gly -* Asp 256 ttt GAT cgt
gct TCT ttc Ser + Ala 264 gct GCT ttc
ttc TTA gct Leu -+ Phe 275 ttc TTC gct
made to sequence both copies of psbA from these four new mutants. When a mutation eliminates or creates a restriction enzyme site, the presence of that mutation can be verified for a large population of cells by performing the diagnostic restriction digestion followed by a genomic DNA gel blot, as was done for DCMW (Erickson et al., 1985a) and for Br202, Br24, Dr2, and Dr18 (J.M. Erickson, unpub- lished). In all cases, overexposure of the blot showed that all detectable copies of psbA contained the mutation.
Genetic evidence in cyanobacteria strongly suggests that herbicide resistance is a recessive trait, i.e. that a mixture of wild-type and mutant gene products results in a herbi- cide-sensitive phenotype (Pecker et al., 1987). Hence, we find, as expected, only mutant copies of psbA in our strains that were selected for herbicide resistance.
Relative Resistance of Mutants to Herbicides
All seven herbicide-resistant mutants were tested in vivo for their relative levels of resistance to atrazine, bromacil, and diuron by plating comparable numbers of cells on minimal agar plates containing varying amounts of a given herbicide, and comparing their growth to that of wild-type 737c(+) cells. Bromacil and atrazine were added to a series of plates at concentrations of 1 p ~ , 2 p ~ , 5 p ~ , 1 O
pM, 20 wM, 50 p M , 100 gM, and 500 pM. Diuron was added to plates at final concentrations of 0.1 p ~ , 0.2 p ~ , 0.5 p ~ , 1 pM, 2 pM,
5
pM, 10 pM, 20 p M , 50 pM, 100 pM, and 250PM. Growth of wild-type cells was inhibited by 1 p~ diuron, 2 p~ bromacil, and 5 p~ atrazine. The resistance levels of all mutants are summarized in Table 2.
Table 2. Summary of Properties of Herbicide-Resistant Mutants of C. reinhardtii
Codons Resistance Fluorescence
No. WT MT Strains A B D lnduction Decay
(L)
(Fsec)21 9 Val Ile Dr2 2x 2x 20x Normal 404 k 53 (5)
Val Ile Drl8 2x 2x 20x Normal 394 f 10 (3)
255 Phe Tyr Ar207 15X 1 x 0.5X Normal 373 +. 31 (5)
256 GlY ASP Ar204 15X 1 ox 3 x Rapid >1 O00 (5)
GlY ASP Br24 15X 1 ox 3x Rapid >1 O00 (3)
275 Leu Phe Br202 1 x 5x 5x Normal 374 f 53 (5)
264 Ser Ala DCMW 1
oox
250X 20x Rapid >1 O00 (6)Wild-type l x 1 x 1 x Normal 294 c 50 (7)
Resistance of C. reinhardtii mutant strains to atrazine (A), bromacil (B), and diuron (D) is indicated relative to the wild-type strain as described in “Results.” Codon numbers are for the D1 polypeptide. Amino acid residues in wild-type (WT) and mutant (MT) strains are indicated. Fluorescence induction and decay measurements are as described in “Methods” and in the legends to Figures 2 to 4. Tqf2 is the half-time of fluorescence decay (see Figure 3 and text). The number of independent measurements of decay half-time is indicated in parentheses for each mutant.
Electron Transfer from QA to Qe in Herbicide-Resistant Mutants
fluorescence lnduction Kinetics
Electron transfer on the reducing side of photosystem II was monitored in the seven herbicide-resistant mutants whose properties are described in Table 2 by analyzing the kinetics of fluorescence induction as described in
“Methods.” Figure 2 depicts the fluorescence induction kinetics for 81-24 and wild-type cells. The inset in Figure 2 shows the initial fluorescence rise on an expanded time scale, which clearly depicts the difference between the two cell lines. Our measurements for mutants Ar204 and DCMW show a more rapid rise in the fluorescence induc- tion curve when compared with wild-type cells (data not shown), which is consistent with previous results (Erickson et al., 1984a; Galloway and Mets, 1984). Fluorescent rise curves of this type have also been found for the atrazine- resistant biotypes of Amarantbus and have been inter- preted as a slowing of electron transfer between Q A and QB (Bowes et al., 1980; Arntzen et al., 1982). In contrast, our measurement of fluorescence induction transients for the herbicide-resistant mutants Ar207, Br202, Dr2, and Dr18 were not significantly different from those of wild- type cells (data not shown).
Fluorescence Decay Curves
To determine the rate of electron transfer frOm QA to Qe more precisely, the kinetics of reoxidation of QA by Qe was measured with a pulse-modulated fluorimeter (Schreiber, 1986). Analysis of single, isolated colonies on agar plates or liquid cells in suspension gave similar results. Fluores- cence decay curves following an actinic saturating flash
were determined using a pulsed low-intensity measuring beam (see “Methods”). Plotting the decay of fluorescence on a logarithmic scale allows one to determine, by linear extrapolation of the initial part of the curve In ala-x versus time, the decay half-time, where a represents the maximum fluorescence and a-x the fluorescence as a function of time. Such a plot is shown in the bottom half of Figure 3. Only the initial part of the curve can be linearized in a first-order plot. This value is taken as a specific indicator of the reoxidation of Q A by Qs. The latter part of the curve
(~300
psec), however, is also QA:QB dependent. This was demonstrated by the complete inhi- bition of the fluorescence decay in the presence of a PSll inhibitor (data not shown). Figures 3 and 4 show the fluorescence decay curves obtained for wild-type cells and the seven mutants Dr18, Dr2, 61-202, Ar207, DCMW, Br24, and Ar204. The corresponding values for the decay half-times are indicated in Table 2. The decay curves and T,,* values obtained for Dr18 and Dr2 are very similar (Figure 3), as expected, since the two mutants carry the same chloroplast mutation (Table 1). Mutants Ar204 and Br24, which were isolated independently and which con- tain identical mutations, both show an identical and signif- icant decrease in the rate of Q A reoxidation relative to wild- type cells (Figure 4, Table 2), indicating a slowing of electron transfer between Q A and QB. The same is true for the DCMW mutant (Figure 4). The fluorescence decay curves obtained with all mutants were highly reproducible.The estimated TIj2 values for Ar207,Br202, Dr2, and Drl8 are slightly but consistently higher than that seen for wild- type cells (Figures 3 and 4, Table 2).
DlSCUSSlON
Analysis of the psbA genes from this characterization of herbicide-resistant mutants and from our previous studies
500 msec
Figure 2. Fluorescence lnduction Kinetics of Dark-Adapted Cells of Chlamydomonas.
Wild-type strain 737c (solid line); herbicide-resistant mutant Br24 (dashed line). Relative fluorescence was measured as indicated in “Methods.” The inset shows the initial fluorescence rise on an expanded time scale. Bars indicate the different time scales.
(Erickson et al., 1984a, Erickson et al., 1985a) have to- gether identified five different mutation sites in psbA that result in amino acid substitutions at five distinct amino acid residues of the D1 reaction center polypeptide of PSII.
Each of these mutations can be correlated with a specific pattern of resistance to the PSll inhibitors atrazine, bro- macil, and diuron, and with specific rates of electron trans- fer from QA to QB (Table 2) in the absence of herbicide. An additional psbA mutation affecting alanine codon 251 has recently been identified in a metribuzin-resistant mutant of Chlamydomonas (Johanningmeier et al., 1987), bringing the total number of mutant sites found in the Chlamydo- monas D1 polypeptide to six. These mutations serve as probes for assessing the role of specific amino acid resi- dues of D1 in quinone-binding and electron transfer on the reducing side of PSII.
Figure 5 shows the locations of these six sites in the D1 polypeptide of C. reinhardtii. This particular membrane- spanning model, with five transmembrane domains (after Trebst, 1987), is based on information obtained from a variety of molecular, biochemical, and physical studies of chloroplast and bacterial reaction centers. Active PSll core complexes capable of efficient primary charge separation have recently been isolated from thylakoid membranes of spinach an‘d pea (Barber et al., 1987; Danielius et ai., 1987;
Marder et ai., 1987; Nanba and Satoh, 1987). These PSll cores contain the D1 and D2 polypeptides, cytochrome b559, chlorophyll a, pheophytin, and p-carotene. The chlo- roplast PSll reaction center is related both structurally and functionally to the reaction centers of the purple photosyn- thetic bacteria (for review, see Barber, 1987; Rochaix and Erickson, 1988). In each case, pheophytin acts as an
intermediate electron acceptor and a very similar quinone- iron complex exists that contains the primary and second- ary stable quinone electron acceptors Q, and QB. In the bacterial reaction centers, QA and QB are bound to the M and
L
reaction center polypeptides, respectively. TheL
subunit has also been shown to bind PSll herbicides (Brown et al., 1984; DeVitry and Diner, 1984; Okamura, 1984). The molecular structures of the reaction center of Rhodopseudomonas viridis (Deisenhofer et ai., 1985) and Rhodobacter sphaeroides (Allen et al., 1987) have recently been determined by x-ray crystallographic analysis. In both of these organisms, the
L
and M subunits form a pair,h I
I ’
- a, L
Y
a, o C a, o IA a, L O 3
-
I - I
I
V
5 0 0 u s e c
-
Figure 3. Fluorescence Decay Measurements.
Upper: Fluorescence decay curves for dark-adapted cells of the C. reinhardtii wild-type strain 737c (solid line), herbicide-resistant mutants Dr2 (dashed line), and Dr18 (dotted line), following an actinic saturating flash as described in “Methods.” Curves are normalized to the same maximum fluorescence to eliminate influ- entes of sample thickness and chlorophyll concentration.
Lower: Semi-logarithmic plot of fluorescence decay. a = maximal fluorescence; a-x = fluorescence leve1 as a function of time. The broken line extrapolates the initial part of the curve and is taken as a measure of the kinetics of QA to Qe electron transfer as discussed in the text. Bar represents time scale.
I
A
-
ov L
o U C o
U H o L O 3 L
-
500 usec
I---
h I
,. -
oL v
o U C o 0 H
o L O 3 G
-
w - 500 usec
Figure 4. Fluorescence Decay Curves for Dark-Adapted Cells of C. reinhardtii.
Wild-type strain 137c compared with herbicide-resistant mutants Ar207 and Br202 (upper panel) and DCMW, Br24, and Ar204 (lower panel), following an actinic flash as described in “Methods.”
Curves are normalized to the same maximum fluorescence. Bar represents time scale.
each folded through the membrane with five transmem- brane domains that put the two quinone binding sites in close proximity. DNA sequence analysis of the genes coding for the chloroplast D1 and D2 polypeptides of Chlamydomonas showed that these two proteins are of similar size (352 amino acid residues each), have several highly homologous regions, and have nearly identical hy- dropathic profiles (Rochaix et al., 1984). Moreover, a local sequence homology is apparent between the D1, D2, and the
L
and M subunits of the bacterial reaction center (Rochaix et al., 1984; Youvan et al., 1984; Erickson et al., 1985b; Williams et al., 1986). Based on the above infor- mation, as well as the structural and functional homology between the two bacterial and PSll subunits, it has been proposed that D2 and D1 are the apoproteins that bind Q A and Qe, respectively, and that these two proteins can be folded with five transmembrane domains each to form a PSll core with chlorophyll, pheophytin, and quinone that resembles the bacterial reaction center (Trebst, 1987).lmmunologic probing of thylakoid membranes with an- tibodies specific for proposed stromal and lumenal resi- dues of the D1 polypeptide (Sayre et al., 1986) gives results that strongly support the general model for D1 shown in Figure 5. This model, however, is in conflict with previous studies that suggest that the COOH terminus of D1 is in the stroma and not the lumen (Marder et al., 1984), and with our previously published model (Erickson et al., 1985a), based on Rao et al. (1983), which predicted seven transmembrane helices for D1. Although more work on the structure of D1 is required, the current model is most consistent with available data. Given this model, we see that four of the D1 residues that are changed in the herbicide-resistant mutants (Ala-251, Phe-255, Gly-256, and Ser-264) are localized on the stromal side of the membrane in a 14-amino acid region of D1 between the fourth and fifth transmembrane domains. Although the remaining two residues, Val-219 and Leu-275 are further away in the primary amino acid sequence of D1, the membrane-folding model places them in close proximity to the other altered residues on the stromal side of the thylakoid membrane. Recent analysis of herbicide-resist- ant mutants in cyanobacteria have identified phenylalanine codon 211 as another site at which a change confers resistance to atrazine and diuron (Gingrich et al., 1988).
Thus, all these mutation sites fall in one localized region of the protein and define the Qe-binding pocket. X-ray diffrac- tion analysis of the bacterial reaction center crystals sug- gests that amino acid residues His-217, Trp-250, and Ala- 258 of the M subunit are contacts for binding QA (Deisen- hofer et al., 1985; Trebst, 1987). Based on similarities of M with D2, and D2 with D1, the corresponding residues of the D1 protein are His-215, Phe-255, and Ser-264.
These three residues are close to or coincide exactly with residues that are altered in our herbicide-resistant mutants.
Although several herbicide-resistant mutants have been examined at the leve1 of psbA sequence in higher plants, algae, and cyanobacteria as previously noted, as well as in Euglena (Johanningmeier and Hallick, 1987), no changes in other parts of D1 or in other PSll core polypeptides have been reported in conjunction with herbicide resistance.
This raises the possibility that the conformation of the herbicide and/or the Qe-binding site is not significantly influenced by other regions of the D1 polypeptide or other peptides. PSll inhibitors, however, can have effects on the donor side of PSll which appear to be mediated through the transmembrane helices (Renger, 1973; Carpentier et al., 1985; Trebst, 1987). It has also been reported that the metribuzin-resistant mutant of C. reinhardtii, which has alanine codon 251 changed to valine, is affected on the donor side of PSll (Johanningmeier et al., 1987).
In this study we have used measurements of chlorophyll fluorescence decay curves to explore the rate of QA reox- idation in seven herbicide-resistant mutants. In some of these mutants, the rate of electron transfer from QA to QB is strongly reduced, whereas in others it is barely affected.
Figure 5. Model of the C. reinhardtii D1 Protein with Five Transmembrane Helices (I to V, Boxed Regions) and a Helix Parallel to the Membrane which Contains Residues 251 to 261 (Boxed), after Trebst (1987).
The amino-terminal end (NH2) is in the stroma, whereas the carboxyl terminus (COOH) is in the lumen. Circles represent amino acid residues, as indicated by the single letter code. Numbers correspond to the amino acid residue. The six residues that are altered in herbicide-resistant mutants as described in the text are marked with dark circles, and are all localized toward the stromal side of the thylakoid membrane. Histidine residues 215 and 272 (circles with dark edges) are involved in iron binding (Fez+). The azido-atrazine binding area between residues 214 and 224 (Wolber et al., 1986) is indicated by a black curving line.
Mutants of the latter group include Dr2, Drl8, Ar207, and Br202 (see Table 2). In these mutants, the fluorescence decay half-time is consistently longer than in wild-type cells, indicating a small decrease in the rate of electron transfer. The consistent difference between the fluores- cence decay curves of these mutants and wild-type cells observed at later times, e.g. 1 to 2 msec (Figures 3 and 4) is difficult to interpret. This may be because the measure- ments in this time frame are affected by properties or components of PSII, such as structural and functional heterogeneities, other than the QA:QB interaction. Our re- sults on Dr2 are at variance with those of Haworth and Steinback (1 987), who concluded from fluorescence induc- tion transients that the rate of electron transfer from QA to Qe is strongly reduced in this mutant. We do not have an explanation for this discrepancy, but note that chlorophyll fluorescence decay provides a more direct measurement for rates of electron transfer between the two quinones.
Although Dr2 is moderately resistant to diuron and other PSll inhibitors except for s-triazines, and shows an ele- vated 15,, value for diuron inhibitors of the in vitro Hill reaction (Galloway and Mets, 1984), surprisingly, no differ- ence in binding of radioactive diuron to the thylakoids of Dr2 and wild-type cells could be found (Haworth and Steinback, 1987).
Fluorescence decay measurements in the mutants DCMW, Ar204, and Br24 indicate that the rate of Qn to
QB electron transfer is considerably slower in these mu- tants than in wild-type cells. It has also been shown previously that the decay of fluorescence in chloroplasts from resistant biotypes of Amaranthus is more than 10- fold slower than in sensitive strains (Bowes et al., 1980).
The mutation in DCMU4 changes serine 264 of the D1 protein to alanine, and the mutant strain is resistant to high levels of atrazine and low levels of diuron. In both Ar204 and Br24, glycine residue 256 is changed to aspartic
acid as the result of a point mutation in psbA. These two mutants are resistant to moderate levels of atrazine and bromacil (Table 2). The metribuzin-resistant mutant MZ2 has a change of alanine to valine at codon 251, is also resistant to atrazine and diuron, and has a slowed Q, to
QB electron transfer (Johanningmeier et al., 1987). Hence, in C. reinhardfii, amino acid substitutions have been found at D1 residues 251, 256, and 264 that significantly reduce the rate of electron transfer on the reducing side of PSII.
The drastic effect of the glycine 256 change on electron transfer is understandable since the mutation introduces a negative charge and thereby seriously disrupts the ionic environment of this region. In the model shown in Figure 5, alanine 251, phenylalanine 255, and glycine 256 are part of the surface parallel helix formed between the fourth and fifth transmembrane helices.
Although herbicides that compete with quinone for bind- ing to the thylakoid membrane have been useful probes for exploring structure-function relationships on the reduc- ing side of PSII, studies with herbicide-resistant mutants have obviously been limited to the specific regions of D1 involved in herbicide binding. Because compounds that inhibit electron transfer on the oxidizing side of PSll are often highly toxic to cells (Carpentier et al., 1985), a similar extension of this approach to studying structure-function relationships in other regions of the D1 protein has not been possible. However, a recent report has demonstrated that the chloroplast genome of C. reinhardfii can be trans- formed with exogenous DNA (Boynton et al., 1988). Using this technology, we have transformed a mutant strain of C. reinhardfii, which lacks the chloroplast gene coding for the D1 polypeptide, with the cloned D1 gene from our Br202 herbicide-resistant mutant and obtained transform- ants that have the herbicide-resistant phenotype of the original Br202 mutant (J. M. Erickson and M. C. Yee, unpublished results). Boynton et al. (1988) also report the recovery of herbicide-resistant transformants after trans- formation of the same strain with the cloned psbA gene from our DCMU4 mutant. These results not only confirm that the chloroplast herbicide resistance trait is a direct result of a single amino acid substitution in the D1 poly- peptide, but also show that psbA can be successfully transformed into the chloroplast genome of a psbA dele- tion mutant. Thus it is now possible, using site-directed mutagenesis and subsequent transformation, to extend structure-function analyses of the D1 polypeptide to any region of interest in this important PSll reaction center component.
METHODS
Alga1 and Bacterial Strains
Chlamydomonas reinhardtii wild-type strain 737c(+) and the her- bicide-resistant mutant strains were maintained in dim light on
Tris-acetate-phosphate (TAP) agar plates (Gorman and Levine.
1965). Br24 and Dr18 were isolated as bromacil- and diuron- resistant mutants, respectively, as described in “Results.” The isolation of mutants DCMW, Ar204, Ar207, Br202, and Dr2 has been previously described (Erickson et al., 1984a; Galloway and Mets, 1984). Escherichia coli strains JM83 or JM1 O1 were used for the propagation of plasmid vectors and recombinant plasmids.
lsolation of DNA and Cloning of psbA
Total DNA was prepared from 40-ml TAP cultures of the different C. reinhardfii mutants using a modification of the procedure of Weeks et al. (1986) as previously described (Erickson et al., 1986).
The DNAs were double-digested with BamHl and Bglll and ligated with DNA from the plasmid pUC19 vector or the Bluescript KS+
vector (Stratagene), after linearization of the vector DNA with BamHI, treatment with calf intestine alkaline phosphatase, and phenol extraction. Calcium-shocked E. coli cells were transformed with the ligated DNA and plated on L agar plates containing 100 pg/ml of ampicillin. Colonies appearing after incubation at 37°C were transferred to nitrocellulose filters (Grunstein and Hogness, 1975) and hybridized with a 32P nick-translated DNA fragment from the wild-type psbA gene (gel-isolated 1.3-kb Xbal/EcoRI fragment, Erickson et al., 1984b). Hybridizing colonies were iso- lated and rapid preparations of plasmid DNA from 1.5-ml cultures were performed by a modification of the Holmes and Quigley (1981) procedure as previously described (Erickson et al., 1984a).
DNA Sequence Analysis
The 1.3-kb Xbal-EcoRI and the 1.1-kb Hindlll-Kpnl fragments of psbA (Erickson et al., 1984b) were isolated by digesting the cloned psbA gene of each mutant with these restriction enzymes and electroeluting the fragments from preparative polyacrylamide gels.
After digestion of these fragments with Ddel or Hinfl, as shown in Figure 1, the digested fragments were labeled at either the 3’ or 5’ ends as previously described (Erickson et al., 1984b), strand- separated by electrophoresis on a polyacrylamide gel, and se- quenced by the chemical cleavage method of Maxam and Gilbert (1 980). Reaction products were fractionated by electrophoresis on 6% and 20% polyacrylamide gels containing 6 M urea and visualized by autoradiography with Kodak XR5 x-ray film.
Fluorescence Measurements
Kinetic studies of chlorophyll fluorescence decay were performed with a pulse-modulated fluorimeter (Walz Co., Effeltrich, Federal Republic of Germany), which has been described in detail else- where (Schreiber, 1986). During the measuring period of 2 msec, the weak noninductive monitoring beam (A = 660 nm) was mod- ulated at a frequency of 100 kHz. The common end of the 4- branched fiber optic system was positioned about 1 mm above the surface of agar plates containing the alga1 strains. The total area analyzed on the plates was 1 13 mm’.
Chlorophyll fluorescence decay curves representing the QA reoxidation were recorded following an actinic, saturating xenon flash of 8 psec width (T1,2) obtained from an FX-800 flash tube (EGcG, Salem, MA) operated at an energy of 2 joules. The actinic
light passed through a Calflex Cyanll (Balzers, Liechtenstein) and a BG 28, Schott (Duryea, PA) glass filter before it was coupled into the light guide system. Due to interna1 gating of the pulse- modulated fluorimeter detector, the first signal point was recorded at 110 psec after triggering of the actinic flash. The fluorescence signal was stored in a digital oscilloscope (Explorer 111, Nicolet, Madison, Wl) using 4096 samples at 12-bit resolution and then subsequently transferred on-line to a Hewlett Packard 9920 com- puter for further analysis. For each decay curve, 10 to 20 tran- sients were averaged by flashing the sample repetitively every 8 sec. After replotting the data in a first-order analysis, a linear regression line was fitted to the first 150 Fsec of the data for calculation of the decay half-time.
Fluorescence induction transients were measured in continu- ous actinic blue light (excitation filter: Corning 4-96) derived from a Schott KL 500 fiberoptic light source. The actinic light passed through a mechanical shutter (opening time, 1 msec) and was directed by fiber optics to the surface of the agar plate. Fluores- cence was isolated from scattered actinic light by a Schott RG 665 red glass filter and detected by a UDT 500 photodiode/
amplifier combination (United Detector Technology, Hawthorne, CA ). Storage and processing of the transients were as described above.
ACKNOWLEDGMENTS
We thank O. Jenni for drawings and photographs, J.D. McBride for the isolation of Br24 and genetic analysis of Br24 and Dr18, and J. Van Dillewijn and L. Jung for technical assistance. This work was supported by the Swiss National Science Foundation grant 3.328.086 (to J.D.R.) and U.S. Department of Agriculture grant AGR 5901-410-8-186 (to R.K.T.).
Received January 6, 1989.
REFERENCES
Allen, J.P., Feher, G., Yeates, T.O., Komiya, H., and Rees, D.C.
(1987). Structure of the reaction center from Rhodobacter sphaeroides R-26: The cofactors. Proc. Natl. Acad. Sci. USA Arntren, C.J., Pfister, K., and Steinback, K.E. (1982). The mech- anism of chloroplast triazine resistance: Alterations in the site of herbicide action. In Herbicide Resistance in Plants, H. Le Baron and J. Gressel, eds (New York: John Wiley & Sons), pp.
Barber, J. (1 987). Photosynthetic reaction centres: A common link. Trends Biochem. Sci. 12, 321-326.
Barber, J., Chapman, D.J., and Telfer, A. (1987). Characterisa- tion of a PS II reaction centre isolated from the chloroplasts of Pisum sativum. FEBS Lett. 220, 67-73.
Bowes, J., Crofts, AR., and Arntzen, C.J. (1980). Redox reac- tions on the reducing side of photosystem II in chloroplasts with altered herbicide binding properties. Arch. Biochem. Biophys.
84,5730-5734.
185-21 4.
200,303-308.
Boynton, J.E., Gillham, N.W., Harris, E.H., Hosler, J., Johnson, A.M., Jones, A.R., Randolph-Anderson, B., Robertson, D.,
Klein, T., Shark, K., and Sanford, J. (1988). Chloroplast trans- formation in Chlamydomonas with high velocity microprojec- tiles. Science 240, 1534-1 538.
Brown, A.R., Gilbert, C.W., Guy, R., and Arntren, C.J. (1984).
Triazine herbicide resistance in the photosynthetic bacterium Rhcdopseudomonas sphaeroides. Proc. Natl. Acad. Sci. USA Carpentier, R., Fuerst, P., Nakatani, H.Y., and Arntzen, C.J.
(1985). A second site for herbicide action in photosystem II.
Biochim. Biophys. Acta 808, 293-299.
Chua, N.-H., and Gillham, N.W. (1977). The sites of synthesis of the principal thylakoid membrane polypeptides in Chlamydo- monas reinhardtii. J. Cell Biol. 74, 441 -447.
Clayton, R.K. (1980). Photosynthesis: Physical Mechanisms and Chemical Patterns. (IUPAB Biophysics Series; Cambridge:
Cambridge University Press).
Danielius, R.V., Kimiyuki, S., van Kan, P.J.M., Plijter, J.J., Nuijs, A.M., and van Gorkom, H.J. (1987). The primary reaction of photosystem II in the D1 -D2-cytochrome b-559 complex. FEBS Lett. 213, 241-244.
Deisenhofer, J., Epp, O., Miki, K., Huber, R., and Michel, H.
(1985). X-ray structure analysis at 3A resolution of a membrane protein complex: Folding of the protein subunits in the photo- synthetic reaction center from Rsp. viridis. Nature (Lond.) 318, De Vitry, C., and Diner, B. (1984). Photoaffinity labelling of the
azidoatrazine receptor site in reaction centers of Rhodopseu- domonas sphaeroides. FEBS Lett. 167,327-331.
Dismukes, G.C. (1 988). The spectroscopically derived structure of the manganese site for photosynthetic water oxidation and a proposal for the protein-binding sites for calcium and man- ganese. chem. Scr. 28A, 99-104.
Erickson, J.M., Rahire, M., Bennoun, P., Delepelaire, P., Diner,
B., and Rochaix, J.-D. (1 984a). Herbicide resistance in Chla- mydomonas reinhardtii results from a mutation in the chloro- plast gene for the 32 kilodalton protein of photosystem II. Proc.
Natl. Acad. Sci. USA 81,3617-3621.
Erickson, J.M., Rahire, M., and Rochaix, J.-D. (1 984b). Chla- mydomonas reinhardtii gene for the Mr 32000 protein of pho- tosystem II contains four large introns and is located entirely within the chloroplast inverted repeat. EMBO J. 3, 2753-2762.
Erickson, J.M., Rahire, M., Rochaix, J.-D., and Mets, L. (1985a).
Herbicide resistance and cross-resistance: Changes at three distinct sites in the herbicide-binding protein. Science 228,204- 207.
Erickson, J.M., Delpelaire, P., and Rochaix, J.-D. (1 985b). Analy- sis of genes encoding two photosystem II proteins of the 30- 34 kD size class. In Molecular Biology of the Photosynthetic Apparatus, K. Steinback, S. Bonitz, C.J. Arntzen, and L. Bo- garad, eds (Cold Spring Harbor, NY: Cold Spring Harbor Lab- oratory), pp. 53-65.
Erickson, J.M., Rahire, M., Malnoe, P., Girard-Bascou, J., Pierre, Y., Bennoun, P., and Rochaix, J.-D. (1986). Lack of the D2 protein in a Chlamydomonas reinhardtii psbD mutant affects photosystem II stability and D1 expression. EMBO J. 5, Galloway, R., and Mets, L. (1982). Nonmendelian inheritance of
81,631 0-631 4.
618-624.
1745-1754.
3-(3,4-dichlorophenyI)l,l -dimethylurea resistant thylakoid membrane properties in Chlamydomonas. Plant Physiol. 70, Galloway, R., and Mets, L. (1 984). Atrazine, bromacil and diuron resistance in Chlamydomonas: A single non-mendelian genetic locus controls the structure of the thylakoid binding site. Plant Physiol. 74, 469-474.
Gillham, N.W. (1 965). lnduction of chromosomal and nonchro- mosomal mutations in Chlamydomonas reinhardtii with N- methyl-A!’-nitro-N-nitrosoguanidine. Genetics 52, 529-537.
Gingrich, J.C., Buzby, J.S., Stirewalt, V.L., and Bryant, D.A.
(1988). Genetic analysis of two new mutations resulting in herbicide resistance in the cyanobacterium Synechococcus sp.
PCC 7002. Photosynthesis Res. 16, 83-99.
Golden, S., and Haselkorn, R. (1985). Mutation to herbicide resistance maps within the psbA gene of Anacysfis nidulans R2. Science 229,1104-1 107.
Goloubinoff, P., Edelman, M., and Hallick, R. (1984). Chloro- plast-coded atrazine resistance in Solanum nigrum: psbA loci from susceptible and resistant biotypes are isogenic except for a single codon change. Nucleic Acids Res. 12, 9489-9496.
Gorman, D.S., and Levine, R.P. (1965). Cytochrome F and Plas- tocyanin: Their sequence in the photosynthetic electron trans- port chain of Chlamydomonas reinhardfii. Proc. Natl. Acad. Sci.
Grunstein, M., and Hogness, D. (1975). Colony hybridization: A method for the isolation of cloned DNAs that contain a specific gene. Proc. Natl. Acad. Sci. USA 72, 3961-3965.
Haworth, P., and Steinback, K. (1987). lnteraction of herbicides and quinone with the Qb-protein of the diuron-resistant Chla- mydomonas reinhardtii mutant Dr2. Plant Physiol. 83, 1027- 1031.
Hirschberg, J., and Mclntosh, L. (1983). Molecular basis of herbicide resistance in Aramanfhus hybridus. Science 222, Hirschberg, J., Bleecker, A., Kyle, D.J., Mclntosh, L., and Arntzen, C.J. (1 984). The molecular basis of triazine-herbicide resistance in higher plant chloroplasts. Z. Naturforsch. 39c, Holmes, D.S., and Quigley, M. (1981). A rapid boiling method for the preparation of bacterial plasmids. Anal. Biochem. 114,193- 197.
Johanningmeier, U., and Hallick, R.B. (1987). The psbA gene of DCMU resistant Euglena gracilis has an amino acid substitution at serine codon 265. Curr. Genet. 12,465-470.
Johanningmeier, U., Bodner, U., and Wildner, G.F. (1987). A new mutation in the gene coding for the herbicide-binding protein in Chlamydomonas. FEBS Lett. 211, 221-224.
Lien, S., McBride, J.C., McBride, A.C., Togasaki, R.K., and San Pietro, A. (1977). A comparative study of photosystem II spe- cific inhibitors: The differential action on a DCMU resistant mutant strain of C. reinhardfii. Plant Cell Physiol. (Special Issue:
Photosynthetic Organelles), 243-256.
Marder, J.B., Goloubinoff, P., and Edelman, M. (1 984). Molec- ular architecture of the rapidly metabolized 32-kilodalton protein of photosystem II. J. Biol. Chem. 259, 3900-3908.
Marder, J.B., Chapman, D.J., Telfer, A., Nixon, P.J., and Barber, J. (1987). ldentification of psbA and psbD gene products, D1 1673-1 677.
USA 54,1665-1 669.
1346-1349.
41 2-420.
and D2, as reaction centre proteins of photosystem 2. Plant MOI Biol 9, 325-333.
Maxam, A., and Gilbert, W. (1980). Sequencing end-labelled DNA with base-specific chemical cleavages. Methods Enzymol. 65, McBride, J.C., McBride, A.C., and Togasaki, R.K. (1977). Isola- tion of Chlamydomonas reinhardtii mutants resistant to the herbicide DCMU. Plant Cell Physiol. (Special Issue: Photosyn- thetic Organelles), 239-241 .
Nanba, O., and Satoh, K. (1987). lsolation of a photosystem II reaction center consisting of D-1 and D-2 polypeptides and cytochrome b-559. Proc. Natl. Acad. Sci. USA 84, 109-1 12.
Okamura, M.Y. (1984). On the herbicide binding site in bacterial reaction centers. In Biosynthesis of the Photosynthetic Appa- ratus: Molecular Biology, Development and Regulation, J.P.
Thornber, L.A. Staehelin, and R.B. Hallick, eds (New York: Alan R. Liss Inc.), pp. 381-390.
Pecker, I., Ohad, N., and Hirschberg, J. (1987). The chloroplast- encoded type of herbicide resistance is a recessive trait in cyanobacteria. In Progress in Photosynthetic Research, J. Big- gins, ed (Dordrecht, The Netherlands: Martinus Nijhoff) Vol. 111, Pfister, K., Steinback, K.E., Gardner, G., and Arntzen, C.J.
(1981). Photoaffinity labelling of a herbicide receptor protein in chloroplast membranes. Proc. Natl. Acad. Sci. USA 78, 981- 985.
Rao, J.K.M., Hargrove, P.A., and Argos, P. (1983). Will the seven helix bundle be a common structure for integral membrane proteins? FEBS Lett. 156, 165-169.
Renger, G. (1 973). The action of 3-(3,4-dichIorophenyI)-l,l -di- methylurea on the water-splitting enzyme system Y of photo- synthesis. Biochim. Biophys. Acta 314, 113-1 16.
Rochaix, J.-D., and Erickson, J. (1 988). Function and assembly of photosystem II: Genetic and molecular analysis. Trends Biochem. Sci. 13, 56-59.
Rochaix, J.-D., Dron, M., Rahire, M., and Malnoe, P.M. (1984).
Sequence homology between the 32kDa and D2 chloroplast membrane polypeptides of Chlamydomonas reinhardtii. Plant MOI. Biol. 3, 363-370.
Sayre, R.T., Andersson, B., and Bogorad, L. (1986). The topol- ogy of a membrane protein: The orientation of the 32 kd Qb- binding chloroplast thylakoid membrane protein. Cell 47, 601 - 608.
Schreiber, U. (1986). Detection of rapid induction kinetics with a new type of high frequency modulated chlorophyll fluorometer.
Photosynthesis Res. 9, 261-272.
Souza-Machado, V., Bandeen, J.-D., Stephenson, G.R., and Lavigne, P. (1 978). Uniparental inheritance of chloroplast atra- zine tolerance in Brassica campestris. Can. J. Plant Sci. 58, Steinback, K.E., Mclntosh, L., Bogorad, L., and Arntzen, C.J.
(1981). ldentification of the triazine receptor protein as a chlo- roplast gene product. Proc. Natl. Acad. Sci. USA 78, 7463- 7467.
Tellenbach, M., Gerber, A., and Boschetti, A. (1983). Herbicide binding to thylakoid membranes of a DCMU-resistant mutant of Chlamydomonas reinhardfii. FEBS Lett. 158, 147-1 50.
Tischer, W., and Strotmann, H. (1 977). Relationship between 499-560.
pp. 81 1-81 4.
977-981.
inhibitor binding of chloroplasts and inhibition of photosynthetic electron transport. Biochim. Biophys. Acta 460, 113-125.
Trebst, A. (1980). lnhibitors in electron flow: Tools for the func- tional and structural localization of carriers and energy conser- vation sites. Methods Enzymol. 69,675-715.
Trebst, A. (1 987). The three-dimensional structure of the herbicide binding niche on the reaction center polypeptides of photosys- tem 11. Z. Naturforsch. 42c, 742-750.
Velthuys, B.R. (1 981). Electron-dependent competition between plastoquinone and inhibitors for binding to photosystem li.
FEBS Lett. 126, 277-281.
Vermaas, W.F.J., Renger, G., and Arntren, C.J. (1983). Herbi- cide(quinone binding interactions in photosystem II. Z. Natur-
Weeks, D., Beerman, N., and Griffith, O.M. (1986). A small-scale forsch. 3 9 ~ , 368-373.
five-hour procedure for isolating multiple samples of CsCI- purified DNA. Anal. Biochem. 152, 376-385.
Williams, J.C., Steiner, L., and Feher, G. (1986). Primary struc- ture of the reaction center from Rhodopseudomonas sphae- roides. Proteins 1, 312-325.
Wolber, P.K., Eilmann, M., and Steinback, K.E. (1986). Mapping of the trizaine binding site to a highly conserved region of the Qb-Protein. Arch. Biochem. Biophys. 248,224-233.
Youvan, DE., Bylina, E.J., Alberti, M., Begusch, H., and Hearst, J.E. (1 984). Nucleotide and deduced polypeptide sequences of the photosynthetic reaction center 8870 antennae and flanking polypeptides from Rsp. capsulata. Cell 37, 949-957.
Zurawski, G., Bohnert, H., Whitfeld, P., and Bottomley, W.
(1982). Nucleotide sequence of the gene for the Mr 32,000 thylakoid membrane protein from Spinacia oleracea and Nico- tiana debneyi predicts a totally conserved primary translation protein of Mr 38,950. Proc. Natl. Acad. Sci. USA 79, 7699- 7703.
