Oxygen Evolving Complex in C4 Species of Flaveria - NCBI

4 downloads 0 Views 2MB Size Report
activity and associated polypeptidesin C4-like and C4 Flaveria species were approximately one-half those observed in meso- phyll cells but equivalent to those ...
Received for publication September 13, 1991 Accepted November 8, 1991

Plant Physiol. (1992) 98, 1154-1162 0032-0889/92/98/11 54/09/$01 .00/0

Characterization of the Expression of the Photosystem 11Oxygen Evolving Complex in C4 Species of Flaveria' Susan L. Ketchner2 and Richard T. Sayre* Departments of Biochemistry (S.L.K., R. T.S.) and Plant Biology (R. T.S.), The Ohio State University, Columbus, Ohio 43210

generation of elevated CO2 concentrations in the BSC, leading to a reduction in photorespiratory activities (6). The tissue-specific expression of C3 and C4 cycle enzymes is determined by developmentally regulated processes as well as by environmental (light) signals and may be regulated at the transcriptional and/or posttranscriptional level (4, 10, 12, 15, 20, 23, 25, 28). In addition, the expression of some enzymes, e.g. Rubisco, requires the coordinate synthesis and assembly of subunits encoded by two separate genomes (chloroplast and nuclear) (4, 15, 25, 26, 28). To date, the molecular characterization of the expression of multi-subunit enzyme complexes in C4 plants has largely focused on this two subunit enzyme complex, Rubisco (4, 25, 27). Another multi-subunit enzyme complex that is expressed in a tissue-specific manner in NADP-ME type C4 plants is the PSII complex (5-7, 11, 26, 28). The PSII complex is composed of more than 10 polypeptides which are both nuclear and chloroplast encoded and is expressed predominantly in the MC of NADP-ME enzyme type C4 plants (17, 23, 26). Studies on the localization and expression of the PSII activity and polypeptides in C4 plants have, however, been limited to monocots such as maize and sorghum. In this study we have characterized the pattern of expression of PSII activity and polypeptides in C3, "C4-like," and C4 species of Flaveria (dicot).

ABSTRACT We have determined the levels of photosystem 11 activity and polypeptide abundance in whole leaves and isolated bundle sheath and mesophyll cells of C4, "C4-like," and C3 species of the genus Flaverla (Asteraceae). On a chlorophyll basis, the whole leaf levels of the DI, D2, and 34-kilodalton photosystem 11 polypeptides were similar for each Flaveria species. Photosystem II activity varied twofold, but was not correlated with photosynthetic type (C3 or C4). The bundle sheath cell levels of photosystem 11 activity and associated polypeptides in C4-like and C4 Flaveria species were approximately one-half those observed in mesophyll cells but equivalent to those in bundle sheath cells of the C3 species, Flaveria cronquistll. Analyses of the steady-state levels of transcripts encoding photosystem 11 polypeptides indicated that there were no differences in transcript abundance between mesophyll and bundle sheath cells of the C4 Flaveria species. This pattem was in contrast to the three- to tenfold higher levels of transcripts encoding photosystem 11 polypeptides in mesophyll versus bundle sheath cells of maize. It is apparent that the higher mesophyll cell to bundle sheath ratio of photosystem 11 polypeptides in C4- and C4-like species of Flaveria is the result of higher levels of photosystem 11 expression in mesophyll cells rather than lower levels of expression in bundle sheath cells.

MATERIALS AND METHODS

A characteristic feature of C4 plants is the differentiation of the photosynthetic tissues of leaves into two distinct cell types, MC3 and BSC. These cells contain different complements of enzymes active in photosynthetic CO2 fixation (5). For example, PEPCase, the C4 cycle enzyme that catalyzes primary CO2 fixation to form C4 acids, is localized in the MC, whereas the C3 cycle enzyme, Rubisco, which fixes CO2 derived from the decarboxylation of C4 acids, is localized in the BSC. The spatial separation of these and other enzymes facilitates the

Plant Material

Cuttings of Flaveria brownii and Flaveria palmeri were obtained from Dr. Lauren Mets, University of Chicago. Flaveria cronquistii cuttings and Flaveria trinervia seeds were obtained from Dr. B. de Moore, Washington State University. Maize seeds were obtained from Liberty Seed Company, New Philadelphia, Ohio, and spinach was obtained from a local market. Flaveria species were maintained by clonal propagation of cuttings to ensure genetic homogeneity. Plants were grown in growth chambers on a 16-h photoperiod (250 uE/ m2/s, 300-700 nm) at 35°C and 8-h dark period at 25°C.

' Supported in part by the National Institutes of Health Grant GM 40703 and the Colleges of Biological Sciences and Agriculture. 2Present address: Service de Photosynthese-IBPC, Centre National de la Recherche Scientifique, 13 rue Pierre et Marie Curie, 75005 Paris, France. 3Abbreviations: MC, mesophyll cell; BSC, bundle sheath cell; PEPCase, phosphoenolpyruvate carboxylase; NADP-ME, NADPmalic enzyme; OEC, oxygen evolving complex; PMSF, phenylmethylsulfonylfluoride; TPCK, N-tosyl-L-phenylalanine chloromethyl

Antibodies Polyclonal antibodies against the 16- and 34-kD OEC proteins were prepared against spinach proteins (26). Antibodies against Dl and D2 PSII reaction center proteins were made against synthetic peptides corresponding to specific regions of the Dl (amino acids 333-342) and D2 (amino acids 19-28,

ketone; DMBQ, dimethylbenzoquinone; DPIP, dichlorophenolindophenol; TE, 10 mm Tris-HCl, pH 8.0, 1.0 mM EDTA. 1154

LOCALIZATION Of PSII IN C4 PLANTS

58-67, 98-107, 303-312) PSII reaction center proteins of spinach (22). Dr. David Oliver, University of Idaho, provided polyclonal antibodies against the 98-kD subunit (P-protein) of glycine decarboxylase from pea leaves and Dr. Hernando Ramirez, Center for International Tropical Agriculture, Cali, Colombia, provided the polyclonal antibodies against PEPCase of cassava (Manihot esculentum). Isolation of Whole Leaf Thylakoids

Thylakoids were isolated from 5 g of fully expanded leaves by grinding in 50 mL of 0.3 M sorbitol, 0.05 M Tricine, pH 7.4, 3.0 mM MgCl2, 1.0 mm sodium ascorbate, 0.5% BSA, and 0.1 mM PMSF in a mortar and pestle with sand. The homogenate was filtered through two layers of Miracloth. The Miracloth was rinsed with 25 mL of grind buffer, and the pooled preparation was centrifuged at 300g for 1 min to pellet debris. The supernatant was centrifuged at 5,000g for 10 min to pellet thylakoids. The thylakoids were then gently homogenized in 15 mL of buffer without PMSF and BSA and repelleted as above. The final pellet was resuspended in a small amount of buffer to give a final Chl concentration of 1.5 mg Chl/mL as determined by the method of Arnon (1). Thylakoids were used immediately or stored frozen at -80°C following addition of glycerol to give a final concentration of 20% (w/v). Isolation of BSC and MC

BSC and MC protoplasts of Flaveria species were prepared by procedures similar to those of Moore (18). Four to five grams of leaf material were cut into 1 x 0.5 mm sections and digested in 30 mL of 2% (w/v) "Onozuka" cellulase RS, 0.3% (w/v) Macerozyme R-10, 0.5 M sorbitol, 50 mM Mes, pH 5.5, 10 mm sodium ascorbate, 1 mM CaCl2, and 0.2% (w/v) BSA for 2.5 to 3.0 h at 25°C. The protoplasts were filtered through 120 ,m and 72 Mm nylon nets and collected by pelleting at 120g for 1 min. After resuspension in 16% dextran (w/v) in 0.5 M sucrose, 5 mM Hepes, pH 7.0, 1 mM CaCl2, and 0.2% (w/v) BSA (buffered osmoticum), 3 mL of protoplasts were overlaid with 2 mL each of 12%, 8%, and 1% (w/v) dextran in buffered osmoticum followed by a final layer containing 0.125 M KCl, 0.25 M sorbitol, 5 mM Hepes, pH 7.0, 1 mM CaCl2, and 0.2% (w/v) BSA (KCl-sorbitol) and centrifuged at 5OOg for 10 min. The BSC protoplasts were isolated from the 12% to 8% (w/v) dextran interface and the MC protoplasts from the 1% (w/v) dextran-KCl-sorbitol interface. MC and BSC protoplasts from the C3 species, F. cronquistii, were isolated by a modification of this procedure as published by Moore et al. (19). MC and BSC protoplasts of maize were prepared according to the procedure of Sheen and Bogorad (25). Intact chloroplasts were isolated from protoplasts which were ruptured by several passages through a 26G syringe followed by pelleting at 300g for 3 min. The supernatant, corresponding to cytoplasmic components, was used for determination of MC and BSC purity (see below). The chloroplast pellet was then washed in 0.35 M sorbitol, 5.0 mM Hepes, pH 7.5, 1.0 mM MgCl2, 0.1 mm EDTA, 0.1% (w/v) PVP and 0.2% (w/v) BSA, repelleted, and frozen in liquid N2 for storage

1155

at -80°C until use. Chl concentrations were determined as above (1) and protein concentrations were determined by the Bradford method (2). SDS-PAGE

Proteins were separated by SDS-PAGE (22). Samples were loaded on the basis of equal amounts of Chl (5-25 ,g) or on the basis of protein (50-100 Mg) for cytoplasmic fractions (see figure legends). The separated proteins were electrophoretically transferred to nitrocellulose or Immobilon-P at 250 mA for 4.5 h by the method of Towbin et al. (29). Antigens were labeled with specific antibodies and detected with protein A-'25I according to previously described procedures (16). Densitometric scanning of autoradiographs was done using an LKB gel scanner. Electron Transport Studies

Thylakoids from whole leaves were prepared as described in the preceding section. MC and BSC thylakoids were prepared from isolated protoplasts which were suspended in 0.4 M sucrose, 50 mM Tricine, pH 7.5, 5 mM MgCl2, 1.0 mM EDTA, 10 mM sodium ascorbate, 0.1 mM TPCK, and 0.1 mM PMSF and ruptured by grinding with a small pestle. The thylakoids were then pelleted at 5,000g for 3 min and washed once in the above buffer minus PMSF and resuspended at a Chl concentration 20.5 mg Chl/mL. MC and BSC preparations having cross-contamination (based on marker enzyme levels) greater than 10% were discarded. PSII and PSI activities were determined polarographically in assay buffer containing 0.4 M sucrose, 50 mM Hepes, pH 7.0, and 5 mm MgCl2. For the measurement of PSII activity the assay buffer also contained 30 mm methylamine, 1.0 mM ferricyanide, 200 Mm DMBQ and thylakoids at 1.5 Mg Chl/mL. PSI activity was determined by the rate of 02 uptake in assay buffer containing 0.3 Mig Chl/mL, 30 mM methylamine, 10 AM DCMU, 50 Mm methyl viologen, 5 mm DPIP, 5 mM Na-ascorbate, and 0.4 mM KCN. RNA Isolation

Total RNA was isolated from Flaveria species by the method of Sheen and Bogorad (28) using 12 g of fully expanded leaf tissue. Isolated MC and BSC were ruptured in RNA lysis buffer (5.0 M guanidinium thiocyanate, 50 mM Tris, pH 9.0,25 mm EDTA, 25 mM EGTA, 2% [w/v] Sarkosyl and 0.1 M f,-mercapthoethanol), frozen in liquid N2, and thawed in the presence of 3 volumes of TE-saturated phenol. To this solution an equal volume of CHCl3 was added, and the mixture was vortexed vigorously for 5 min. After centrifugation for 10 min in a microfuge, the upper layer was removed to a new tube, an equal volume of TE-saturated phenol and CHC13 was added and the mixture vortexed as before. The upper layer was extracted with 2 volumes of CHC13, transferred to a new tube and 2.5 volumes of ethanol were added. The RNA was precipitated overnight at -20°C and pelleted in a microfuge for 30 min. The RNA was then washed with 70% ethanol and dried under a vacuum. The purity of each MC and BSC fraction used for RNA isolations

1156

KETCHNER AND SAYRE

was confirmed by western blot analyses. RNA was isolated from maize MC and BSC by the method of Sheen and Bogorad (25). RNA was separated under denaturing conditions on a formaldehyde agarose gel using standard protocols and transferred to nitrocellulose (14). DNA fragments encoding Rubisco large subunit (rbcL), actin (mac), D2 (psbD), the 34-kD subunit of the OEC (oeel) and the 16-kD subunit of the OEC (oee3) were cloned from maize (13, 24, 26) and DNA encoding the Dl protein (psbA) was cloned from Amaranthus hybridus (9). rbcL, psbA, psbD, and mac DNA fragments were radioactively labeled by the random hexamer method and oeel and oee3 were labeled by in vitro generation of radioactively labeled RNA with a Stratagene RNA synthesis kit. Northern blots were prehybridized in a solution containing 5 x SSPE (20 x SSPE = 174 g/L NaCl, 27.6 g/L NaH2PO4.H20, 7.4 g/L EDTA; pH 7.4), 50% formamide, 1% SDS, and 100 ,ug/mL sheared calf thymus DNA for 2 h at the specified temperatures. Hybridizations with mac and psbD probes were carried out at 37°C, hybridizations with psbA and rbcL were at 42°C, and hybridizations with oeel and oee3 were carried out at 65°C. All blots were incubated overnight. The blots were washed with 2 x SSPE at room temperature for 5 min and then washed twice at temperatures specified below in 2 x SSPE and 1% SDS for 15-30 min. For blots probed with psbD and mac, the wash temperature was 55°C; for all other probes the blots were washed at 65°C. The blots were then rinsed with 2 x SSPE at room temperature for 15 min prior to exposure to Kodak XAR-5 x-ray film. The length of exposure of the film was limited for purposes of obtaining linear response curves for quantification by densitometry using an LKB Ultrascan XL model 2222-022 scanner. Each blot was probed no more than five times to prevent anomalous results due to loss of bound mRNA.

RESULTS AND DISCUSSION In previous studies on NADP-ME type C4 monocots, it was found that PSII activity was approximately 10-fold higher in MC than BSC on a Chl basis (5, 6, 8, 11, 26, 28). It was not known, however, whether this pattern of expression of PSII activity occurs in NADP-ME type C4 dicots. Previous analyses of the level of thylakoid stacking in MC and BSC chloroplasts of the C4 dicot Flaveria trinervia had indicated that the distribution of PSII complexes between the two cell types may be equivalent (L. Mets, personal communication). To determine the overall levels and localization of PSII in C4 and C4like Flaveria species, we have measured PSII activity, polypeptides, and associated transcript abundance in whole leaves and isolated MC and BSC. It was expected that if PSII activity (on a Chl basis) is expressed predominantly in only one cell type (MC), as in maize, then the whole leaf levels of PSII activity in C4 Flaveria species would be less than in the C3 species. As shown in Table I, a twofold variation in whole leaf PSII activity was observed among the Flaveria species. The levels of PSII activity were not, however, correlated with the photosynthetic type. F. trinervia, a C4 species, had whole leaf PSII rates (209 Amol 02/mg Chl/h) that were 23% higher than the C3 species, F. cronquistii (170 ,umol 02/mg Chl/h), whereas F. brownii,

Plant Physiol. Vol. 98, 1992

Table 1. PSII and PSI Electron Transport Activities in Flaveria Thylakoids from Whole Leaves Data are means ± SE of three replications. PSII PSI Species H2O-+DMBQ Ascorbate/DPIP-Methyl viologen F. brownii 122 ± 27 517 ± 10

(C4-like) F. cronquistii (C3) F. palmeri

170 ± 28 137 ± 30

490 ± 55 488 ± 72

209 ± 12

649 ± 51

(C4-like) F. trinervia (C4)

which has been described as a C4-like species, had the lowest level of PSII activity (122 ,mol 02/mg chl/h). Similarly, rates of PSI activity (649 ,umol 02/mg Chl/h) were 32% higher for F. trinervia (C4) than for the C3 species, F. cronquistii (490 Amol 02/mg Chl/h). The C4-like species, F. brownii and F. palmeri, had PSI activities that were essentially identical to that of F. cronquistii. Thus, there was no correlation between photosynthetic type (C3, C4-like, or C4) and the whole leaf level of PSII or PSI activity. We have quantified the levels of specific PSII polypeptides by western blot analyses in an attempt to determine the biochemical basis for the species-specific differences in whole leaf levels of PSII activity. We chose to focus on the levels of the Dl, D2, 34-kD, and 16/17-kD PSII polypeptides for several reasons: (a) the Dl and D2 polypeptides are thought to coordinate or bind the reaction center Chl, pheophytins, and quinones of the PSII complex (17), (b) the 34-kD protein is part of the extrinsic OEC and is required for stabilization of the manganese complex (16, 17), (c) the 16/17-kD protein is probably the most peripheral of the three extrinsic OEC polypeptides and may be the last polypeptide to assemble with the PSII complex resulting in functionality (16, 17), and (d) the four polypeptides are encoded by separate genomes. The Dl and D2 polypeptides are chloroplast encoded whereas the 16/17- and 34-kD proteins are nuclear encoded (22, 26). As shown in Figure 1, the Dl antibody, prepared against a synthetic peptide corresponding to amino acids 333-342 of the spinach Dl sequence, cross-reacted with a single polypeptide of 36 kD in all Flaveria species, maize, and spinach (22). Because the region of the Dl protein used for generation of the antibody is conserved between different species (3), we would not expect species-specific differences in cross reactivity with this antibody. Similarly, the D2 antibody was prepared against a highly conserved peptide corresponding to amino acids 98-107 ofthe spinach D2 sequence. As shown in Figure 1, the D2 antibody recognized a 34-kD band in Flaveria, which was similar in size to that detected in spinach and maize. Unlike the Dl antibody, however, a second band having a molecular weight of 28-kD band was also detected in each Flaveria species. Other D2 antibodies, generated against synthetic peptides corresponding to different regions of the D2 protein, also recognized the same 28-kD band with similar affinity, indicating that this protein was in fact D2 (data not shown). The addition of protease inhibitors (TPCK and PMSF) had no effect on the abundance of the 28-kD

LOCALIZATION Of PSII IN C4 PLANTS

D2

S 1 2 3 4 5

Dl

S

1

1157

2 3 4 5 '-36 kD

,34 kD

I-

26 kD

o

:--,,-

ow*"*Awwft'-

OWN6.44oft - omok

34kD

S

34 kD

N

16kD

1 2 3 4 5

2

S

3

49W o-w "

4

*---

&-A#

K

band, indicating that the 28-kD band is probably a different conformer of the 34-kD D2 band and not a proteolytic fragment. Finally, as shown in Figure 1, polyclonal antibodies against both the 16- and 34-kD OEC proteins of spinach recognized proteins of similar size in Flaveria species and maize. It is not known, however, whether the antibodies have similar affinities for polypeptides from each Flaveria species (see discussion of transcript homologies below). The relative amounts of each protein in leaves of the different Flaveria species are summarized in Figure 2. There were no significant differences between C3, C4-like, or C4 Flaveria species in the levels of the Dl or D2 proteins. Similar results were also observed for the 34-kD protein, except for the C4 species, F. trinervia. The reduced levels of the 34-kD protein in F. trinervia are probably due to proteolytic degradation. This conclusion is based on the observation that the addition of the protease inhibitor TPCK (0.1 mM) to the thylakoid isolation buffer increased the intensity of the 34kD band and decreased the intensity of the lower mol wt,

-2

D2 Protein

- 3

5 17 kD

16 kD

Figure 1. Identification of PSII thylakoid polypeptides isolated from whole leaves by western blot analysis. Thylakoids proteins were loaded on the basis of equal amounts of Chl (25 Asg) and proteins were separated by SDS-PAGE. Proteins were transferred to Immobilon-P, probed with antibodies, and detected with protein A-1251 by autoradiography. Lane S, spinach; lane 1, F. brownie (C4-like); lane 2, F. palmeri (C4-like); lane 3, F. trinervia (C4); lane 4, F. cronquistfi (C3); lane 5, maize. Different lanes may not be from the same blot. For quantifications refer to Figure 2.

cross-reacting polypeptides. Even in the presence of TPCK, however, lower mol wt polypeptides were detected in F. trinervia preparations. In contrast to the Dl, D2, and 34-kD polypeptides the levels of the 16/17- kD protein ranged from to nearly 4 times that observed for the C3 species, F. cronquistii (it is noted that the 17-kD protein migrates at 16 kD in Flaveria species, Fig. 1). Three of the Flaveria species had identical levels of the 16-kD protein, however, the C4-like species, F. brownie, had levels that were nearly fourfold higher than the other species. One possible explanation for the variability in the abundance of the 16-kD protein is that the protein is not tightly bound to the PSII complex and is washed off to different extents during thylakoid isolation. In support of this interpretation we note that there was substantial variability in the level of the 16-kD protein from F. brownie (see Fig. 2). Based on these results, it was apparent that there was no correlation between PSII activity and the abundance of any particular polypeptide screened in whole leaf preparations

Dl Protein

3

s0

Je C.) 2 as

.0

1

1

2-34

= 2

2

3

4

Figure 2. Quantification of PSII polypeptide abundance detected by western blot analysis of thylakoids polypeptides isolated from whole leaves. Amounts are normalized to F. cronquistii (C3) on the basis of equal amounts of Chi. Bar 1, F. brown (C4-like); bar 2, F. palmer (C4-like); bar 3, F. trinervia (C4); bar 4, F. cronquistli (C3). Data are means + SE (bars).

5

kD Protein co

0

Z

1-O

o

o -i a-

IL: .....

Im

1

2

3

4

5

1

2

3

4

5

KETCHNER AND SAYRE

1158

A.

M

B

m

t

3 4f

u,!

S

/

kV

M

-w

_

.V

Plant Physiol. Vol. 98, 1992

M

_

Figure 3. Identification of PEPCase and glycine decarboxylase in isolated MC and BSC cytoplasmic polypeptide fractions by western blot analysis. Cell extracts were loaded on the basis of equal amounts of protein (50 ,g protein) and proteins were separated by SDS-PAGE. Proteins were transferred to Immobilon-P, probed with antibodies, and detected with protein A-1251 by autoradiography. The figures for different species are not all from the same blot but were chosen from separate blots that were overexposed to demonstrate tissue-specific differences. A, Western blot probed with PEPCase antibodies (100 and 90 kD) and Rubisco antibodies (55 kD); B, western blot probed with glycine decarboxylase Psubunit antibodies. M denotes MC extracts; B denotes BSC extracts. Lane 1, F. brownie (C4-like); lane 2, F. palmer (C4-like); lane 3, F. trinervia (C4); lane 4, F. cronquistii (C3); lane 5, maize.

(Table I and Fig. 2). We observed that species that had substantially different rates of PSII activity had similar levels of the Dl, D2, and 34-kD polypeptides as determined by western blots. In addition, species specific differences in the levels of the 16-kD polypeptide were not correlated with differences in PSII activity in whole leaves. We suspect that the discrepancy between PSII activity and polypeptide abundance is due to differences in stability or numbers of active PSII complexes rather than to differences in total numbers of PSII complexes. Because the levels of all PSII polypeptides were not determined, however, it is possible that the abundance of functional PSII complexes may not be accurately reflected by the levels of the D1, D2, 34-kD, or 16-kD polypeptides. To determine whether the lack of detectable differences in PSII polypeptide levels in whole leaves masked differences in PSII polypeptide levels in the separate cell types, it was necessary to determine the level of PSII activity and polypeptides in isolated MC and BSC. The purity of our MC and BSC preparations was determined on the basis of the distribution of glycine decarboxylase and PEPCase in each cell type preparation. Assuming that the level of the cell specific markers, PEPCase (MC) and glycine decarboxylase (BSC), in

the BSC and MC, respectively, was no more than 5% of the level in MC and BSC, respectively, it was determined that level of cross-contamination of our MC and BSC preparations was no more than 10% (Fig. 3). A comparison of the PSII activity in isolated cell types (Table II) and whole leaf levels (Table I) indicated that there was some loss of PSII activity that occurred during cell separation for each species, with the possible exception of F. palmer. In contrast to PSII activity, the levels of PSI activity were unaffected by the cell separation procedure. The differential loss of PSII and PSI activity is probably due to the greater susceptibility of PSII to damage than PSI. For comparative purposes, however, we assumed that such damage was similar for each cell type from a given species. Based on MC/BSC ratios for PSII activity (Table II), only two Flaveria species, F. trinervia and F. palmer, exhibited any difference in PSII activity between cell types (MC/BSC PSII activity ratios >1.5). Significantly, C4 and C4-like Flaveria species having higher MC/BSC PSII activity ratios did not have reduced levels of PSII activity in the BSC (relative to the C3 species, F. cronquistii) as occurs in maize, but had higher levels of MC PSII activity. In contrast to maize, there were

Table II. Electron Transport Activities in Isolated MC and BSC Data are means ± SE of two replications except where noted. F.

cronquistli

MC

H2O-+DMBQ PS11

F. brownil

(C3)

Reaction

98 ± 18

(C4-like) BSC

97

MC

16

89 ± 8

F. palmer (C4-like)

MC JmoI 02/mgChl/h 101 ± 27 169 ± 40 BSC

BSC

F. trinervia (C4)

MC

94 ± 34 147 ± 17

BSC

Z. mays

(C4) MC

86 ± 30 174 ± 17

BSC

16 ± 8

Ascorbate/DPIP-+MV 790 ± 42 767 ± 65 557 ± 70 564 ± 69 655 ± 92 683 ± 11 524 ± 31 500 ± 70 408 ± 63 702 ± 27 PSI 65 65 H2CO-MVa 128 128 148 94 163 148 NDb ND Ratio PSII MC/BSC 1.0 0.9 1.8 1.7 10.9 Ratio PSI MC/BSC 1.0 1.0 1.0 1.0 0.6 b one a Only replication. ND, not determined.

LOCALIZATION Of PSII IN C4 PLANTS

D2

1

s

2

4

3

Dl

5

36 kD

n s 30D MiB M

34 kD

B

M

B

M

B

B

B

2 k4Bm5 M~~~~~~~~~ B

M

M

M

-"

16kD

s

M

1

B

M

2

B

M

3

B

M

4

B

M

59

_-

s

B

M

2

3

4

B

M

B

..l.l.E

V

17 kD__ 16 kD -_ I-,~J

no differences in PSI activity between MC and BSC for any species of Flaveria. To determine whether the tissue specific differences in PSII electron transport activity were associated with differences in the abundance of PSII polypeptides, we quantified the levels of several PSII polypeptides in isolated MC and BSC by western blot analysis (Fig. 4). The results from densitometry scans of multiple western blots are presented in Table III. Consistent with the results obtained from electron transport assays the levels of PSII polypeptides were 1.2- to 2.5-fold higher in MC than BSC for F. trinervia and F. palmeri but equivalent in each cell type for F. brownie and F. cronquistii. The apparent correlation between PSII activity and PSII polypeptide abundance in isolated cell types was in contrast to measurements made on whole leaves. The MC/BSC PSII polypeptide ratios for the Flaveria species were, however, substantially less than those obtained for maize (this study and ref. 8). The levels of PSII polypeptides in maize MC were 4- to 11-fold higher than in BSC, results that were consistent with the level of PSII activity in maize MC and BSC, i.e., PSII activity (16 gmol 02/mg Chl/h) in maize BSC was approximately 10% of that found in MC (170 ,gmol 02/mg Chl/h) (8, 26). In contrast to the cell specific pattern of expression of PSII activity in some Flaveria species, there was no difference in PSI activity between MC and BSC for any

%

5B

1159

Figure 4. Identification of PSII polypeptides from isolated BSC and MC thylakoids. Thylakoid proteins were loaded on the basis of equal amounts of Chl (10 Asg) and proteins were separated by SDS-PAGE. Proteins were transferred to Immobilon-P, probed with antibodies, and detected with protein A-1251 by autoradiography. M denotes MC extracts; B denotes BSC extracts. Lane S, spinach. Lane 1, F. brownie (C4-like); lane 2, F. palmer (C4-like); lane 3, F. trinervia (C4); lane 4, F. cronquistii (C3); lane 5, maize. Not all lanes were from identical blots. See Table IV for quantification of polypeptide levels.

Flaveria species examined. In maize, however, PSI activity was 2-fold higher in BSC than in MC (8). Overall, these results suggest that expression of PSII activity in Flaveria is quite different than that in maize and is not coupled to differences in the tissue specific expression of photorespiratory and C4 pathway enzymes. To characterize the regulatory mechanisms that account for the patterns of PSII expression in Flaveria, we determined the steady-state transcript levels for several messages encoding PSII polypeptides. Figure 5 shows results obtained from northern blots of total RNA isolated from whole leaves. Overall, there was no difference in the sizes of transcripts surveyed between the Flaveria species and maize. In only one case was a polymorphism in transcript size observed and that was for the psbD (D2 protein) transcript. For each Flaveria species, as well as maize, two transcripts of 4.2 and 3.5 kb were detected (7). As shown in Figure 6, the levels of transcripts encoding PSII polypeptides were similar in all the Flaveria species examined. Although the relative abundance of the psbA (Dl protein) and psbD transcripts in Flaveria species and maize was similar, the apparent abundance of the oee 1 (34 kD) and oee3 (16 kD) transcripts in maize was much higher (3.0-6.0 times) than in Flaveria. The apparent difference in oeel and oee3 transcript abundance between Flaveria species and

Table Ill. PS/I Polypeptide Levels in MC and BSC Amounts standardized to levels of F. cronquistii mesophyll levels. Data are means ± SE of three replications. Protein F. brownie (C4-like) F. cronquistii (C3) F. pa/meri (C4-like) F. trinervia (C4) D1 MC 1.0 0.5±0.2 0.6±0.1 0.6±0.3 BSC 0.6 ± 0.1 0.9 ± 0.2 0.4 ± 0.3 0.5 ± 0.3 1.7 MC/BSC 0.6 1.2 1.5 D2 MC 1.0 1.1 ±0.1 1.3±0.3 1.2±0.5 BSC 1.7 ± 0.3 1.2 ± 0.2 0.6 ± 0.4 0.8 ± 0.3 MC/BSC 0.6 0.9 2.2 1.5 34 kD MC 1.0 0.5 ± 0.2 1.0 ± 0.6 0.5 ± 0.1 BSC 0.8 ± 0.1 0.4 ± 0.1 0.5 ± 0.1 0.3 ± 0.02 1.3 MC/BSC 1.0 1.7 2.5 16 kD MC 1.0 2.1 ± 0.3 0.5 ± 0.1 1.9 ± 0.3 BSC 0.9 ± 0.1 1.4 ± 0.1 0.6 ± 0.2 0.8 ± 0.2 1.1 MC/BSC 0.8 1.5 2.4

Z. mays (C4)

1.3±0.1 0.3 ± 0.02 4.3

0.9±0.1 0.1 ± 0.04 9.0 7.1 ± 0.3 0.1 ± 0.02 7.1 1.1 ± 0.01 0.1 ± 0.01 11

KETCHNER AND SAYRE

1160

psbD ''' '4

QsbUA

oeel 1I

oee3

2 *0

34 1.

_

6

maize may be due to differences in sequence identity between the Flaveria transcripts and the maize gene probes. Northern blots washed at various stringencies indicated that the sequence identity between the maize gene probes oee 1 and oee3 and Flaveria transcripts was low but did not differ between the four Flaveria species (data not shown). For all transcripts of PSII polypeptides studied, there again was no correlation between steady-state transcript abundance and photosynthetic type for whole leaves. In contrast to the results obtained for transcripts encoding PSII polypeptides, the whole leaf levels of rbcL transcripts in C4 Flaveria species were lower than in the C3 species, F. cronquistii (data not shown). The species specific differences in whole leaf levels of rbcL transcripts and the transcripts encoding PSII polypeptides may stem from the fact that rbcL is specifically expressed only in BSC of C4 Flaveria species, whereas transcripts encoding the PSII polypeptides lack this distinct tissue-specific expression (see below). Additionally, the greater abundance of MC than BSC in whole leaves may account for the apparent lower steady-state levels of rbcL detected in whole leaves of C4 Flaveria species. To determine whether cell specific differences in the abundance of PSII polypeptides were correlated with differences in transcript abundance, we measured select transcript pool sizes in isolated MC and BSC of Flaveria species. The species used for this analysis were F. palmeri, F. trinervia, F. cronquistii, and the C4 monocot maize (Table IV). Due to the small amounts of total RNA that could be isolated from each protoplast preparation it was necessary to load the entire quantity of RNA on the gel for northern blot analysis. As a

4 5

1

Q* _4

rbcL

Figure 5. Identification of transcripts encoding the chloroplast encoded psbA and psbD and nuclear encoded oeel and oee3 genes from Northern blots of whole leaf RNA extracts. Total RNA equivalent to 10 Ag was loaded in each lane. Lane 1, F. brown/i (C4-like); lane 2, F. palmer (C4-like); lane 3, F. trinervia (C4); lane 4, F. cronquistii (C3); lane 5, maize. Not all lanes were from identical blots. See Figure 6 for quantification purposes.

=

-01 3 I* or 0 C > 0 0 -j 0 -

P.9.

psbD

3

0

C co 0 > IN 0

>

21

-

.. I.

--Hilflll

1

1 =

0

.3

all.~~~~~Hl

....

X.....

m

Co

psbA

U:

0 0 -

Figure 6. Relative transcript abundance for the chloroplast encoded rbcL, psbA and psbD and nuclear encoded oeel and oee3 transcripts in total RNA extracts from whole leaves of Flaveria species and maize. Levels were standardized on the basis of the level of mac transcript and compared to the levels in F. cronquistil. Lane 1, F. cronquistii (C3); lane 2, F. brownie (C4-like); lane 3, F. palmer (C4-like); lane 4, F. trinervia (C4); and lane 5, maize.

Plant Physiol. Vol. 98, 1992

3

2

4

5

1

6- oeel

"

8

co

6-

2

3

4

5

2

3

4

5

oee3

CT

0r-

>0 0%. -i

U4-

-ft

.s 00

Or_ to I. %-

IV

2

-

|u

2

L

IIM 0

cc

-9.

1

2

3

4

1

5