The Plant Cell, Vol. 2, 857-866, September 1990 O 1990 American Society of Plant Physiologists

A Metal-Dependent DNA-Binding Protein lnteracts with a Constitutive Element of a Light-Responsive Bromoter

Eric Lam,’ Yur iko Kano-Murakami, Phi l ip Gilmartin, Bettina Niner, and Nam-Hai Chua2

Laboratory of Plant Molecular Biology, The Rockefeller University, 1230 York Avenue, New York, New York 10021 -6399

We have used DNase I footprinting to characterize nuclear factors that bind to the light-responsive promoter of pea rbcS-3A, one member of the gene family encoding the small subunit of ribulose-l,5-bisphosphate carboxylase. A sequence-specific binding activity, designated 3AF1, binds to an AT-rich sequence present at the -45 region of the rbcS-3A promoter. A tetramer of the 3AF1 binding site, designated as Box VI, can form multiple complexes with tobacco leaf and root nuclear extracts. Mutations of 3 base pairs in Box VI severely reduce DNA-protein complex formation in vitro. The wild-type Box VI tetramer, but not the mutant tetramer, is active in transgenic tobacco plants when placed upstream of the cauliflower mosaic virus 35s promoter truncated at -90. These results correlate binding of 3AF1 to the in vivo function of Box VI. The Box VI tetramer/35S chimeric construct confers expression in diverse cell types and organs and its activity is not dependent on light. By using the Box VI tetramer as a probe to screen a cDNA expression library, we have obtained a putative cDNA clone for the 3AF1 DNA-binding activity. Lysogen extracts of Escherichia col i expressing the cDNA clone give sequence-specific complexes with Box VI. The deduced amino acid sequence of the protein encoded by the cDNA contains two stretches of about 100 residues that are 80% homologous. Moreover, in each of the two repeats, there is an arrangement of histidines and cysteines, which may be related to the two known types of zinc-finger motifs found in many DNA-binding proteins. Consistent with the expectation that metal coordination plays an important role in DNA binding by this protein, we found that 1,lO-phenanthroline can abolish the formation of DNA-protein complexes. Interestingly, we found that the same treatment did not abolish the DNA binding activity of 3AF1 in crude nuclear extracts of tobacco. These data indicate that the nuclear 3AF1 activity is likely due to multiple DNA-binding proteins all interacting with Box VI in vitro. RNA gel blot analysis shows that multiple transcripts homologous to this cDNA clone are expressed in different tobacco organs.

INTRODUCTION

The small subunit of ribulose-l,5-bisphosphate carboxyl- ase, the key enzyme for carbon fixation in higher plants, is encoded by a multigene family (rbcs), and its expression is regulated by light (Dean et al., 1989). The promoters for several rbcS from different plant species have been shown to contain sufficient information for light responsiveness and organ specificity (reviewed in Dean et al., 1989). Severa1 different plant nuclear factors have been reported to bind to some of the conserved sites found in rbcS promoters. One of these, GT-1, binds to multiple sites in the upstream sequences of several pea rbcS (Green et al., 1987, 1988; Kuhlemeier et al., 1988b). The other factor, GBF, binds to a “G-box” upstream of several tomato rbcS (Giuliano et al., 1988).

Among the pea rbcS, the rbcS-3A promoter has been studied most extensively. This promoter contains se-

’ Current address: AgBiotech Center, Waksman lnstitute of Mo- lecular Biology, P.O. Box 759, Rutgers State University, Piscata- way, NJ 08855. To whom correspondence should be addressed.

quence information that confers light responsiveness as well as tissue specificity (Fluhr et al., 1986a, 1986b; Aoyagi et al., 1988). Through deletion analyses and the use of chimeric promoter constructs, we have identified at least three distinct regions that are involved in the light-respon- sive expression of rbcS-3A. Two 5’ regions (-41 O to -1 70 and -170 to -50) can confer light responsiveness and tissue specificity when placed upstream of the homologous and a heterologous promoter (Kuhlemeier et al., 1988a, 1989). The third region, located between -50 and +15, has been shown to confer light-responsive but not organ- specific expression when potentiated by a heat-inducible element (Kuhlemeier et al., 1989). Sequence comparison of several pea rbcS promoters has identified five conserved sequences, designated as Boxes I to V, that are located between -170 and +1 (Nagy et al., 1986). GT-1 binds to two of these conserved sequences (Box II and Box 111) located within the -1 50 to -1 O0 region of several rbcS promoters (Green et al., 1987, 1988; also see Figure 1 B). The same factor also binds to sequences homologous to Boxes II and 111 in the rbcS-3A upstream region (-410 to

858 The Plant Cell

-170). Site-specific mutagenesis of the GT-1 binding sites,Boxes II and III, showed a clear correlation between GT-1binding in vitro and activity of the rbcS-3A promoter intransgenic plants (Kuhlemeier et al., 1988a). Recently, wehave demonstrated that a synthetic tetramer of Box IIalone is able to confer light-responsive and tissue-specificexpression when fused to a truncated 35S promoter (-90to +8) from the cauliflower mosaic virus (CaMV) (Lam andChua, 1990). Together, these results support the notionthat GT-1 is involved in the light responsiveness of therbcS-3A upstream elements.

The expression conferred by the -50 to +15 region ofthe rbcS-3A promoter is likely to be mediated by factorsdifferent from those of the two upstream regions becauseit does not contain any GT-1 binding sites. Moreover, whenpotentiated by a heat-shock element, this promoter regionis active in root as well as in leaf (Kuhlemeier et al., 1989).In the present work, we have characterized a DMA-bindingactivity, 3AF1, that interacts with sequences between —51and -31, designated as Box VI, of the rbcS-3A promoter.Experiments with transgenic plants show that Box VIfunctions as a constitutive element. We have isolated froma tobacco leaf expression library a putative cDNA clonefor 3AF1. Our data suggest that the protein encoded bythis cDNA clone requires metal for DMA binding and thatdistinct transcripts homologous to this cDNA clone arepresent in different tobacco organs.

RESULTS

Characterization of a Nuclear Factor Binding to the -45Region of the rbcS-3A Promoter

The -50 to +15 region of the rbcS-3A promoter has beenfound to contain sufficient information for light responsive-ness (Kuhlemeier et al., 1989). We are interested in definingnuclear factors that interact specifically with sequenceswithin this region. Figure 1A shows the protection ofsequences from -51 to -31 of the rbcS-3A promoter fromDNase I cleavage after incubation with nuclear extractsfrom tobacco leaves. The protected site is directly up-stream of the putative TATA-box, and we named this siteBox VI. Figure 1 shows the location of Box VI in relationto Box II and Box III. We also designated the bindingactivity that recognizes Box VI as 3AF1.

We synthesized a tetramer of the Box VI sequence foruse as a binding site probe. As a control for sequencespecificity, we also synthesized a mutant tetramer with 3altered base pairs. Figure 2 (bottom panel) shows thenucleotide sequence of the wild-type and the mutant bind-ing site. When incubated with either tobacco leaf or rootnuclear extracts, the wild-type tetramer forms multiplecomplexes. The mobilities of some of these complexes are

clearly different between leaf and root nuclear extractsand the positions of the complexes are found to be variablein different extract preparations (data not shown). At pres-ent, the relationship between the different complexes isnot clear. They may be the result of different gene productsthat can bind to Box VI or they may be different forms ofthe same protein resulting from either differential mRNAprocessing, post-translational modification, or proteolysisduring extract preparation. These possibilities are not mu-tually exclusive and are likely to contribute to the complex-ity and variability of the gel shift patterns that we haveobserved. We will address these possibilities in the future.

C E G C E G

-47

-371-

B

Upper

-151 Box II _

Lower

138 Bo*170 .—————————i r______iAaCAAAATTTCAAATCTfoTGTGGTTAATATGJCTGCAAACTTIftTCATTTTCAtJT

ATCTAACAAMTTGGTACTAGGCAGTAGCTAAGTACCACAATATTAAGACCATAATAT-51 B°"VI -31 *1TSa^AATASATAAAIAAAAACATI^TATATAGCAAGTTTTAGCAGAAGCTTTGCAATT

CATACAGAAGTSASAAA+20

Figure 1. Detection of 3AF1 by DNase I Footprinting.

(A) An rbcS-3A promoter fragment (from -170 to +8) was labeledat either the upper or lower strand by Klenow fill-in with all four32P-nucleotide triphosphates. The end-labeled probe was incu-bated with either tobacco nuclear extract (E) or control buffer (C)before DNase I treatment. Positions of guanines as determinedby the method of Maxam and Gilbert (1980) are indicated (G). Thebracketed areas indicate the sequences protected from DNase Idigestion by incubation with tobacco nuclear extracts.(B) The sequence between -170 and +20 of the rbcS-3A pro-moter is shown. Nucleotides were numbered according to Fluhret al. (1986b). Boxes II and III are GT-1 binding sites (Green et al.,1987, 1988).

Characterization of an AT-Rich Binding Factor 859

Leaf Root are not as sensitive as LB1 and LB2 to the mutations thatwe have introduced.

—— LB2

•RB1

probe: W M W M

WILD-TYPE

AAATAGATAAATAAAAACATTTTTATCTATTTATTTTTGTAA

-51 -31AAATAGA£AI£TAAAAACATTTTTATCTCTAGATTTTTGTAA

MUTANTFigure 2. Detection of 3AF1 in Tobacco Nuclear Extracts withSynthetic Box VI Tetramers.

A tetramer of the 21 -bp sequence of Box VI (5'-AAATAGATAAA-TAAAAACATT-3') was synthesized with Hindlll and Xhol restric-tion sites at the 5' and 3' ends, respectively. A Box VI mutanttetramer (5'-AAATAGAGATCTAAAAACATT-3'), with 3 bp of the21 bp altered, was also synthesized. These tetramers were usedto detect sequence-specific DNA-binding protein(s) in tobacco leafand root nuclear extracts by gel retardation assays. W, wild-typeBox VI tetramer; M, mutant Box VI tetramer; fp, free probe. LB1and LB2 indicate DNA-protein complexes observed with tobaccoleaf nuclear extracts. RB1 and RB2 indicate DNA-protein com-plexes observed with tobacco root nuclear extracts. The se-quences of wild-type and mutant Box VI are shown; the mutatedbases are underlined.

In any case, we have designated the dominant complexesin the leaf sample as LB1 and LB2 and those in the rootsample as RB1 and RB2. Figure 2 shows that formationof LB1, LB2, RB1, and RB2 is sensitive to the 3-bpmutation that we introduced into Box VI. These resultsshow that factors in leaf and root nuclei can bind to BoxVI in a sequence-specific manner. Figure 2 shows thatthere are faster migrating complexes than LB1 with to-bacco leaf nuclear extracts. However, these complexes

Functional Characterization of Box VI, the Binding Sitefor 3AF1

To test for the function of Box VI, we fused the wild-type(4F1) and mutant (4F1m) tetramers of this element to thepromoter vectors X-GUS-46 and X-GUS-90 (Benfey andChua, 1989; Lam and Chua, 1989). In these constructs,the 84-bp tetramers were fused upstream of the -46 and-90 derivatives of the CaMV 35S promoter (Benfey andChua, 1989; Lam and Chua, 1989). The coding sequenceof the bacterial enzyme /3-glucuronidase (GUS) was usedas the reporter gene (Jefferson et al., 1987). Agmbacteriumtumefaciens harboring these constructs were used totransform tobacco, and five to 10 independent transgenicplants for each construct were assayed for GUS activity.Table 1 shows the data obtained with leaf and root extractsfrom a representative plant of each construct. With the X-GUS-46 vector, there is little GUS activity above back-ground with either the wild-type or the mutant Box VI

Table 1. Activity of Tetramers of Box VI in Transgenic Tobacco

Construct

X-GUS-464F1-464F1m-46X-GUS-904F1-904F1m-90

Expression

Root

360286320

170070701210

Leaf

63428176

3100620

The X-GUS-46 and X-GUS-90 vectors have been discussed inMethods. Tetramers of the wild-type and mutant Box VI se-quences (shown in Figure 2) were fused upstream of the -46 and-90 truncated derivatives of the CaMV 35S promoter. For thewild-type Box VI tetramer, the resultant constructs are designated4F1 -46 and 4F1 -90, respectively. For the mutant Box VI tetramer,the constructs are designated 4F1m-46 and 4F1m-90, respec-tively. Five to 10 independent transgenic plants of each constructwere used to determine GUS expression levels in leaf and rootextracts by the fluorometric assay (Jefferson et al., 1987). Themean values of the five to 10 plants examined were determined.For each construct, the results from a representative plant thatshowed activities closest to the mean values are shown in thetable. The deviation among individual transgenic plants containingthe same construct varied in the range of threefold to fourfoldfrom the values shown. The values obtained for the X-GUS-46vector containing plants are similar to those obtained from non-transformed tobacco plants and, thus, represent background GUSactivities in tobacco detected by our assay. All values are givenin picomoles of 4-methylumbelliferone per minute per milligram ofprotein.

860 The Plant Cell

Figure 3. Histochemical Localization of GUS Expression Conferred by the 4F1 -90 Synthetic Promoter.Leaf, stem, and root sections from transgenic plants with the 4F1-90 construct were prepared as described previously (Benfey and Chua,1989; Benfey etal., 1989).(A) The midvein region of the leaf.(B) Stem.(C) The lamina portion of the leaf.(D) and (E) Root.GUS activity was localized by the histochemical substrate X-Gluc (Research Organics). EP, epidermis; CH, chlorenchyma; VA, vascularelements; CO, collenchyma; T, trichome; X, xylem; P, phloem; PM, palisade mesophyll; SM, spongy mesophyll; RT, root tip; RH, roothair; C, cortex.

Characterization of an AT-Rich Binding Factor 861

tetramers. The -90 derivative of the 35S promoter showsclear expression above background in the root, as hasbeen reported previously (Benfey et al., 1989). Placementof the Box VI tetramer at -90 enhances the activity of thistruncated 35S promoter in leaf and root. In both organs,the mutant Box VI tetramer (the 4F1 m-90 construct) showsmuch lower activation of the —90 promoter.

Histochemical localization of the GUS gene productshows that the 4F1-90 construct is expressed in most celltypes of leaf, stem, and root of tobacco, as illustrated inFigure 3. Because Box VI is located in a region of therbcS-3A promoter known to mediate light responsiveness,we tested whether the expression of this construct isdependent on light. Comparison of the GUS transcriptlevels by 3' S1 nuclease protection assays showed nodramatic difference between expression of the 4F1-90construct in light-grown or dark-adapted plants, as shownin Figure 4. Thus, in contrast to the Box II tetramer (Lamand Chua, 1990), the Box VI tetramer does not conferlight-responsive expression.

Cloning of Putative 3AF1: A Metal-Dependent DNA-Binding Protein

By using the wild-type Box VI tetramer as a probe, weobtained a putative cDNA clone of 3AF1 by screeningapproximately 200,000 clones from an amplified tobaccoleaf expression library containing approximately 60,000independent clones. Lysogen extracts from Escherichiacoli expressing this cDNA clone were prepared and usedto assay for sequence-specific DNA-binding activity. Thetetramer binding site (as-1) of a previously characterized

LysogenExtract TGAla 3AF1

m —

Probe: A W M A W MFigure 5. Binding Site Specificity of a Putative cDNA Clone for3AF1.

Lysogen extracts from B. co/i expressing a putative cDNA cloneof 3AF1 were used for gel retardation assays with different bindingsite probes. A, a tetramer of as-7 (5'-CTGACGTAAGGGAT-GACGCAC-3'), the 21-bp binding site for ASF-1 (Katagiri et al.,1989; Lam et al., 1989); W, wild-type Box VI tetramer; M, mutantBox VI tetramer (see Figure 2). As a control, we compared proteinbinding to these probes with lysogen extracts from E. coli express-ing a putative cDNA clone for ASF-1 (TGA1 a, left panel). Arrowsindicate the complexes obtained with the different lysogen ex-tracts that are dependent on the cDNA insert. ED, endogenousDNA-binding activity in lysogen extracts.

L D L D L D

Transgenic PlantFigure 4. Expression of the 4F1 -90 Promoter Element Is Insen-sitive to Light.

Total leaf RNA from light grown (L) or 2-days dark-adapted (D)transgenic tobacco plants containing the 4F1-90 construct wasanalyzed by 3' S1 nuclease protection assays using an rbcS-3Cprobe (Fluhr et al., 1986b). Three independent transgenic plantswere analyzed. Ten micrograms of RNA was used for each lane.Only the protected band (230 bases) is shown.

plant factor activation sequence factor (ASF)-1 (Lam et al.,1989) was used as a control. Figure 5 (right panel) showsthat the wild-type tetramer of Box VI gives three com-plexes when incubated with the lysogen extracts. Only thetwo slower complexes are sensitive to the 3-bp change inthe Box VI mutant. In contrast, the fastest migrating DNA-protein complex is apparently not sensitive to the muta-tions. We have designated this complex as "ED" for en-dogenous DNA-binding activity. No significant complex isdetected with the as-7 tetramer using the same lysogenextracts. As a comparison, we also examined binding ofthese probes with lysogen extracts from E. coli expressinga cDNA clone of another plant factor, TGA1 a (Katagiri etal., 1989). We have shown previously that TGAla bindsspecifically to as-1 and is most likely equivalent to thenuclear factor ASF-1 (Katagiri et al., 1989). With an as-1tetramer as probe, as expected, we observed good bindingactivity with extracts from E. coli expressing the TGAlacDNA clone. With the wild-type or mutant Box VI tetramerprobes, only the ED complex was evident with this lysogenextract. Therefore, the complex ED may be attributed to

862 The Plant Cell

B 9 Yp&i&m&gEp

108

Figure 6. Primary Structure of the Putative cDNA Clone for 3AF1.

(A) The nucleotide sequence of the 3AF1 cDNA clone is shown with the predicted amino acid sequence represented in the stand- ard one-letter code above the DNA sequence. The first nucleotide immediately after the EcoRl site in frame with the lacZ coding sequence is designated as nucleotide 1. (B) The predicted amino acid sequence from residue 9 to 107 is compared with that from residue 108 to 206. ldentical amino acid residues between the two sequences are boxed. Functionally conserved amino acid substitutions are underlined, and the posi- tions of the conserved histidines and cysteines are indicated by asterisks r).

an endogenous bacterial protein with affinity for AT-rich sequences. Because the putative 3AF1 protein was syn- thesized as a lacZ fusion protein, it was not possible for us to compare directly the mobility of the DNA-protein complexes obtained with the lysogen extracts and those observed with nuclear extracts. Nevertheless, our data indicate that we have isolated a putative 3AF1 cDNA clone because it encodes a protein with the same sequence binding specificity as the nuclear activity 3AF1.

Sequence analysis showed that the putative cDNA clone for 3AF1 contains an insert of 683 bp with an open reading frame of 227 amino acids in frame with the lacZ coding sequence of the Agt l l vector (Figure 6A). The encoded protein has two interesting features (Figure 6B). First, it contains two repeat sequences with about 80% identity. The homologous sequences are approximately 94 amino acids long and the conservation is greatest at the center of these repeats. Second, in each of the repeats, the

conserved histidines and cysteines are arranged as His- X1z-Cys-Xz-Cys-X14-His-X,z-Cys. This arrangement is rem- iniscent of the two types of known zinc-finger motifs found in transcription factors such as TFlllA and the glucocorti- coid receptors that have the sequence of C ~ S - X ~ - ~ - C ~ S - X12-14-Hi~-X2-4-Hi~ and Cys-Xz~4-Cys-X1z~,4-Cys-X2-4-Cys, respectively (Struhl, 1989). However, the conserved repeat found in the putative 3AF1 protein lacks a histidine pair downstream from the cysteine pair at positions 38 and 41. It should be noted that the second repeat, between amino acids 108 and 206, contains a histidine pair at positions 151 and 155. The presence of this histidine pair, in con- junction with the cysteine pair at positions 137 and 140, raises the possibility that this region may form a conven- tional zinc finger. Thus, if this sequence does conform to the two known types of zinc-finger structures, there will only be one zinc finger in this coding sequence and the N- terminal repeat may not be involved in DNA binding. Based on the similarity of the two repeats in the deduced protein sequence of our cDNA clone and the fact that most known zinc-finger proteins contain multiple fingers (Struhl, 1989), we suggest that both units in the putative 3AF1 protein may be functional in DNA-protein interactions. Future site- specific mutagenesis of this factor may help to resolve these questions.

Because the predicted protein sequence of the cDNA insert suggests that metal coordination may be involved in DNA binding of this protein, we tested the sensitivity of complex formation to the metal chelator 1,l O-phenanthro- lhe. Figure 7 shows that the addition of 1 mM of the chelator completely abolishes binding of the putative 3AF1 to the Box VI probe. As a control, the binding of TGAla to the as-7 site (Katagiri et al., 1989) is insensitive to the chelator .

To examine whether the 3AF1 activity detected in crude nuclear extracts corresponds to the DNA-binding protein encoded by our cDNA clone, we have tested whether 1 , l O-phenanthroline would also inhibit Box VI-binding ac- tivities in these extracts. We found that 1 ,lO-phenanthro- line at a concentration of 4 mM failed to abolish 3AF1 activity in crude nuclear extracts of tobacco leaf (data not shown). This result thus suggests that there are different proteins that can interact with Box VI in vivo. The protein encoded by our cDNA clone corresponds to only one of these DNA-binding activities and is not likely to represent a major portion of the 3AF1 activities in the crude nuclear extracts.

RNA Gel Blot Analysis of Transcripts Homologous to the Putative 3AF1 cDNA Clone

We have also examined the expression of 3AF1 -related gene(s) by RNA gel blot analysis. Figure 8 shows that multiple transcripts ranging in size from about 1.5 kb to 4.5 kb are present in RNA from both young and mature tobacco leaves. At least three transcripts ranging in size

Characterization of an AT-Rich Binding Factor 863

1,10 Phe —— +Extract — + +

—— + DISCUSSION

Characterization of Box VI: A Factor Binding Site inthe rbcS-3A Promoter

The promoter region (-50 to +15) of rbcS-3A contains

—ED

R S YL ML

TGAIct 3AF1Figure 7. 1,10-Phenanthroline Specifically Inhibits Complex For-mation between Box VI Tetramer and the Putative 3AF1 Factor.

Lysogen extracts from E. coli expressing either the cDNA cloneof ASF-1 (TGA1a) or the putative cDNA clone of 3AF1 were usedfor gel retardation assays. For the TGAla-containing extracts, aprobe with a single copy of the as-1 binding site inserted intoposition -55 of the rbcS-3A promoter was used (Katagiri et al.,1989). For the 3AF1-containing extracts, the Box VI tetramer wasused as probe. When indicated (+), 1 mM 1,10-phenanthrolinewas added during the incubation of the extracts with the probe.Arrows indicate the positions of the complexes that are dependenton the cDNA insert. Note that in this particular assay with extractsfrom E. coli expressing the putative 3AF1 clone, the slowestmigrating complex, indicated by the top arrow, is much lower inintensity than the faster migrating complex. This is in contrast tothat shown in Figure 5, where the slower complex is the moreintense species. Upon repeated assays, we found that the relativeintensities of the two complexes are variable. The reason for thisvariability is unknown at present. ED, endogenous DMA-bindingactivity in the lysogen extracts; 1,10 Phe, 1,10-phenanthroline.

from 2 kb to 4 kb are present in RNA from the stem. A 4-kb transcript, similar in size to that observed in stem, isclearly the major species in the root. However, this tran-script is apparently not detectable in leaf RNA under ourconditions (indicated by solid triangles in Figure 8A). As acontrol of the RNA quality, we have also examined thedevelopmental regulation of the expression of two othertobacco DNA-binding proteins, TGA1a and TGA1b (Kata-giri et al., 1989). As reported previously using poly-A+

RNA, we also observe that these genes are preferentiallyexpressed in roots as compared with leaves. Interestingly,we found that transcripts homologous to TGA1 a are morehighly expressed in mature leaves when compared withyoung leaves, whereas the reverse is true for transcriptsencoding TGA1b.

kb

+ 0.2

B+ I.8

+ 2.9

Figure 8. RNA Gel Blot Analysis of Transcripts Homologous tothe Putative 3AF1 cDNA Clone in Different Organs of Tobacco.

(A) Total RNA (40 ^g per lane) from root (R), stem (S), young leaf(YL), and mature leaf (ML) of Nicotians tabacum (cv SR1) wereanalyzed for transcripts homologous to the putative 3AF1 cDNAclone. Solid triangles (A) indicate transcripts that appear to bedifferent in the different organs.(B) The same blot as in (A) was used to assay for the expressionof TGA1a. The arrow indicates the approximate size of thetranscript.(C) The same blot as in (A) was used to assay for the expressionof TGA1b. The arrow indicates the approximate size of thetranscript.

864 The Plant Cell

elements that can confer light-responsive expression but not organ specificity. This region alone, however, does not activate transcription because site-specific mutation or deletion of sequences upstream from -50 completely de- stroys activity (Kuhlemeier et al., 1988, 1989). To identify factors involved in the activity conferred by this region, we have characterized a binding site, Box VI, that is located between -51 and -31. Although rbcS-3A is expressed in chloroplast-containing tissues, nuclear factors from both leaf and root of tobacco bind to Box VI in a sequence- specific manner. A tetramer of Box VI can activate tran- scription of a truncated CaMV 35s promoter, whereas a tetramer of the mutated element cannot. This activity is dependent on sequence elements within the -90 to -46 region of the 35s promoter (cf. Table 1). Previously, we have characterized another tobacco nuclear factor, ASF-1, that binds to the -75 region of the 35s promoter (Lam et al., 1989). Thus, the expression of the Box VI tetramer, when fused to the -90 derivative of the 35s promoter, may be the result of synergistic interactions between 3AF1 and ASF-1. This result suggests that Box VI may be an element that is functional only in the presence of another cis-regulatory element, and is consistent with the previous observation that the -50 derivative of rbcS-3A has no detectable activity (Kuhlemeier et al., 1989). Thus, Box VI is likely to be a class B element as defined by Fromental et al. (1 988). This class of cis-regulatory elements is func- tional only when juxtaposed with certain other elements.

In contrast to the Box II element (Lam and Chua, 1990), the activity conferred by the Box VI element (-51 to -31 of rbcS-3A) is apparently constitutive because it is neither light responsive nor tissue specific, at least in conjunction with the truncated 35s promoter (Figures 3 and 4). This result, together with the previous observation that the -50 to +15 region of the rbcS-3A promoter can confer light responsiveness (Kuhlemeier et al., 1989), suggests that factors that can mediate light-responsive expression are likely to be located elsewhere, probably between -31 and +15 of the rbcS-3A promoter. 3AF1 may function, there- fore, as an accessory factor that interacts with other cell- specific and light-responsive elements, such as Box II (Lam and Chua, 1990), of the rbcS-3A promoter. Because the activity of Box VI depends on other cis-regulatory ele- ments, it may not be the major determinant for the regu- lation of the rbcS-3A promoter.

Our results show that the rbcS-3A promoter, which is light responsive and is preferentially expressed in chloro- phyllous cells (Fluhr et al., 1986a, 1986b; Aoyagi et al., 1988), comprises light-responsive elements, like Box II, as well as constitutive elements, such as Box VI. In this context, it is important to emphasize that Box VI behaves as a class B element, whose activity is dependent on other factors binding elsewhere in the same promoter. This property may enable 3AF1 to participate in the expression of promoters with diverse patterns of regulation because its ability to confer transcriptional activation will depend on

the synergistic interaction of a number of different factors. Recently, we have demonstrated that another plant factor binding site, activation sequence (as)-2, from the -1 O0 region of the CaMV 35s promoter, also behaves as a class B element in vivo (Lam and Chua, 1989). Therefore, this class of cis-acting elements may be rather common and may provide the basis for the generation of tissue specific- ity by combinatorial utilization of different trans-acting factors.

During the preparation of this manuscript, Datta and Cashmore (1 989) have reported the characterization of a pea nuclear factor AT-1. This pea factor binds to an AT- rich motif present in the upstream regions of several light- responsive promoters, including the -294 to -266 region of pea rbcS-3A. However, binding of AT-1 to the pea rbcS- 3A promoter has not been examined, and it is not known whether AT-1 will bind to Box VI or not. In future studies, we will examine the relationships between 3AF1 and other AT-rich sequence-binding activities, including A i -1 , that have been reported for several plant promoters (Jofuku et al., 1987; Jensen et al., 1988; Bustos et al., 1989).

Characterization of a Putative Clone of 3AF1

By screening a tobacco leaf cDNA expression library with the Box VI tetramer as a probe, we have isolated a partial cDNA clone that encodes a sequence-specific DNA-bind- ing activity. Lysogen extracts from E. coli expressing this clone displayed binding activity with the Box VI probe in gel retardation assays. The binding activity was reduced drastically by the mutation of 3 bp in Box VI. No binding activity was detected in these lysogen extracts with as-1, the binding site for ASF-1. Taken together, these results indicate that the partial cDNA apparently encodes a DNA binding domain with sequence specificity similar to that of the nuclear factor 3AF1. Because our partial cDNA clone is likely to be expressed in E. coli as a fusion protein with the LacZ gene product, we cannot determine to which binding activity in the leaf nuclear extracts (Figure 2) the encoded protein corresponds. Antibodies to the cloned gene product will be needed to address this question definitively in the future.

Sequence analysis of the cDNA showed that the en- coded putative 3AF1 factor contains a tandem repeat of about 94 amino acid residues sharing 80% identity. Within these repeats, an interesting array of histidines and cys- teines is conserved. These conserved histidines and cys- teines do not conform to the two known types of zinc- finger motifs found in many transcription factors from animal cells and yeast (Struhl, 1989). The spacing of 12 and 14 amino acids found between the conserved histi- dines and cysteines in the putative 3AF1 protein is very similar to that found between the histidine and cysteine pairs in factors such as TFlllA (Miller et al., 1985) and may, in fact, form the large loop of the zinc-finger motif. In

Characterization of an AT-Rich Binding Factor 865

addition, conserved phenylalanine and leucine residues found in the loop region of most zinc-finger motifs (Struhl, 1989) are also present in the region downstream from the cysteine pairs of the putative 3AF1 protein, although at a slightly different spacing with respect to the 2 cysteines. For these reasons, we favor the model that both homolo- gous repeats in the putative 3AF1 protein are involved in sequence-specific DNA binding, and that the conserved histidines and cysteines form a nove1 type of structure for DNA-protein interactions. Consistent with the proposal that the putative 3AF1 protein may be a zinc-finger protein, we demonstrated that the metal chelator i ,i O-phenanthro- line inhibits complex formation between the putative factor and Box VI.

By RNA gel blot analysis, we found multiple transcripts that are homologous to our cDNA clone. At least one transcript present in leaf RNA is clearly different from the ones found in root and stem. This difference in transcript size is also reflected by the difference in DNA-protein complex mobility observed with nuclear extracts from leaf and root (Figure 2). Thus, distinct forms of 3AF1 may be expressed in these organs. These results raise the possi- bility that the activity of the Box VI tetramer in different organs of tobacco might result from distinct proteins that collectively make up the 3AF1 activity. These different proteins may bind to Box VI with different affinity as well as specificity, and they may be regulated differently during development.

METHODS

Preparation of Extracts and Assays for DNA-Protein lnteraction

Tobacco nuclear extracts were prepared essentially as described by Green et al. (1987), except the Percoll gradient step was omitted. DNase I footprinting was carried out with an end-labeled probe under conditions similar to those described in Lam et al. (1 989). The positions of guanines were determined by the method of Maxam and Gilbert (1 980). Gel retardation assays were carried out in 0.7% agarose/3% polyacrylamide gels under conditions similar to those described by Mikami et al. (1987). 1,lO-Phenan- throline was purchased from Sigma.

Vectors, Clones, and Transgenic Plant Analysis

The X-GUS-46 and X-GUS-90 vectors have been described pre- viously (Benfey and Chua, 1989). Essentially, these are pMON505 derivatives with the GUS coding sequence fused to the -46 and -90 derivatives of the 35s promoter. The 3’ polyadenylation sequence of pea rbcS-3C was fused downstream from the GUS coding sequence to provide for poly-A addition signals. Unique restriction sites for Hindlll and Xhol are present directly upstream of the -46 and the -90 derivatives of the 35s promoter. Synthetic oligonucleotide pairs were used to construct tetramers of Box VI

and the 3-bp mutant. Hindlll and Xhol sites are present at the 5’ and 3’ ends of the tetramers, respectively. These tetramers were first cloned into a pEMBL12 derivative with an Xhol site. After sequence verification of the tetramers, the binding sites were then cloned into X-GUS-46 and X-GUS-90. These vectors were intro- duced into Agrobacterium tumefaciens by triparental mating, and the exconjugants were used to transform Nicotiana fabacum (cv SR1) by the leaf disc technique (Fraley et al., 1985). Primary independent transgenic plants for each construct were used to measure GUS activity in root and leaf extracts. The fluorometric and histochemical assays for the GUS gene product have been described previously (Jefferson et al., 1987; Benfey et al., 1989). RNA from light-grown and dark-adapted tobacco plants was isolated and assayed for GUS transcript levels by 3’ S l nuclease protection as described previously (Fluhr et al., 1986b).

lsolation and Characterization of a Putative cDNA Clone for 3AF1 by Expression Library Screening

The method of Singh et al. (1988) was used to screen for se- quence-specific DNA-binding protein with specificity similar to that of 3AF1. An amplified tobacco leaf library of about 60,000 individ- ual, random-primed cDNA clones in Xgtll was screened with a nick-translated Box VI tetramer. About 200,000 clones were screened, and one clone was identified as a putative clone for 3AF1. Lysogen extracts from this clone were prepared according to Singh et al. (1988) and used in gel retardation assays to confirm the sequence specificity of the putative DNA-binding protein. The cDNA clone was found to have only one EcoRl site; therefore, we subcloned the approximately 3-kb Kpnl/Sstl fragment, containing the cDNA insert, into the Kpnl/Sacl site of M13mp18 and M13mpl9 vectors (New England BioLabs). Single-stranded DNA from these clones was then used to sequence both strands of the cDNA insert. RNA gel blot analysis was carried out with 40 pg of total RNA from different organs of tobacco. Total RNA was denatured in 50% formamide and 6% formaldehyde, and tran- scripts were separated on a 1 % agarose gel with 6% formalde- hyde. After transfer onto nitrocellulose, an EcoRI/EcoRV fragment (298 bp) was nick-translated with 3’P-CTP and the labeled probe was used to detect transcripts.

ACKNOWLEDGMENTS

We thank Fumiaki Katagiri for helpful comments on the manuscript and the tobacco leaf cDNA library and Hugh Williams for photog- raphy. We also thank Philip Benfey for the X-GUS-46 and X-GUS- 90 vectors, as well as many valuable discussions. E.L. and P.M.G. were supported by postdoctoral fellowships from the National lnstitutes of Health and the Science and Engineering Council/ North Atlantic Treaty Organization, respectively. This work was supported by the Monsanto Company and by Grant BRSG 507 RR07065 from the National lnstitutes of Health.

Received June 28, 1990.

866 The Plant Cell

REFERENCES Jensen, E.O., Marcker, K.A., Schell, J., and deBruijn, F.J. (1 988). lnteraction of a nodule specific, trans-acting factor with distinct DNA elements in the soybean leghaemoglobin Ibc, 5’ upstream region. EMBO J. 7, 1265-1271.

Jofuku, K.D., Okamuro, J.K., and Goldberg, R.B. (1987). Inter- action of an embryo DNA binding protein with a soybean lectin gene upstream region. Nature 328, 734-737.

Katagiri, F., Lam, E., and Chua, N.-H. (1989). Two tobacco DNA-binding proteins have homology to CREB. Nature 340,

Kuhlemeier, C., Cuozzo, M., Green, P.J., Goyvaerts, E., Ward, K., and Chua, N.-H. (1988a). Localization and conditional re- dundancy of regulatory elements in rbcS-3A, a pea gene en- coding the small subunit of ribulose-bisphosphate carboxylase. Proc. Natl. Acad. Sci. USA 85, 4662-4666.

Kuhlemeier, C., Fluhr, R., and Chua, N.-H. (1988b). Upstream sequences determine the difference in transcript abundance of pea rbcS genes. MOI. Gen. Genet. 212,405-41 1.

Kuhlemeier, C., Strittmatter, G., Ward, K., and Chua, N.-H. (1 989). The pea rbcS-3A promoter mediates light responsive- ness but not organ specificity. Plant Cell 1, 471 -478.

Lam, E., and Chua, N.-H. (1989). ASF-2: A factor that binds to the cauliflower mosaic virus 35s promoter and a conserved GATA motif in Cab promoters. Plant Cell 1, 11 47-1 156.

Lam, E., and Chua, N.-H. (1990). GT-1 binding site confers light responsive expression in transgenic tobacco. Science 248,

Lam, E., Benfey, P.N., Gilmattin, P.M., Fang, R.X., and Chua, N.-H. (1 989). Site-specific mutations alter in vitro factor binding and change promoter expression pattern in transgenic plants. Proc. Natl. Acad. Sci. USA 86, 7890-7894.

Maxam, A.M., and Gilbert, W. (1 980). Sequencing end-labeled DNA with base-specific chemical cleavages. Methods Enzymol.

Mikami, K., Tabata, T., Takefumi, K., Nakayama, T., and Iwa- buchi, M. (1987). Nuclear protein(s) binding to the conserved DNA hexameric sequence postulated to regulate transcription of wheat histone gene. FEBS Lett. 223, 273-278.

Miller, J., McLachlan, A.D., and Klug, A. (1985). Repetitive zinc- binding domains in the protein transcription factor MA from Xenopus oocytes. EMBO J. 4, 1609-1614.

Nagy, F., Fluhr, R., Kuhlemeier, C., Kay, S.A., Boutry, M., Green, P.J., Poulsen, C., and Chua, N.-H. (1986). Cis-acting elements for selective expression of two photosynthetic genes in trans- genic plants. Philos. Trans. R. SOC. Lond. B 314, 493-500.

Singh, H., LeBowitz, J.H., Baldwin, A.S., and Sharp, P.A. (1988). Molecular cloning of an enhancer binding protein: lsolation by screening of an expression library with a recognition site DNA. Cell 52, 41 5-423.

Struhl, K. (1 989). Helix-turn-helix, zinc-finger, and leucine-zipper motifs for eukaryotic transcriptional regulatory proteins. Trends Biochem. Sci. 14, 137-140.

727-730.

471-474.

65,499-560.

Aoyagi, K., Kuhlemeier, C., and Chua, N.-H. (1988). The pea rbcS-3A enhancer-like element directs cell-specific expression in transgenic tobacco. MOI. Gen. Genet. 213, 179-185.

Benfey, P.N., and Chua, N.-H. (1989). Regulated genes in trans- genic plants. Science 244, 174-1 81.

Benfey, P.N., Ren, L., and Chua, N.-H. (1989). The CaMV 35s enhancer contains at least two domains which can confer different developmental and tissue-specific expression patterns.

Bustos, M.M., Guiltinan, M.J., Jordano, J., Begum, D., Kalkan, F.A., and Hall, T.C. (1 989). Regulation of Psglucuronidase expression in transgenic tobacco plants by an A/T-rich, cis- acting sequence found upstream of a French bean P-phaseolin gene. Plant Cell 1, 839-853.

Datta, N., and Cashmore, A.R. (1 989). Binding of a pea nuclear protein to promoters of certain photoregulated genes is modu- lated by phosphorylation. Plant Cell 1, 1069-1077.

Dean, C., Pichersky, E., and Dunsmuir, P. (1989). Structure, evolution, and regulation of RbcS genes in higher plants. Annu. Rev. Plant Physiol. Plant MOI. Biol. 40, 41 5-439.

Fluhr,, R., Kuhlemeier, C., Nagy, F., and Chua, N.-H. (1986a). Organ-specific and light-induced expression of plant genes. Science 232, 11 06-1 1 12.

Fluhr, R., Moses, P., Morelli, G., Coruzzi, G., and Chua, N.-H. (1 986b). Expression dynamics of the pea rbcs family and organ distribution of the transcripts. EMBO J. 5, 2063-2071.

Fraley, R.T., Rooers, S.G., Horsch, R.B., Eichholtz, J.S., Flick, C.L., Finch, W.L., Hoffmann, N.L., and Sanders, P.R. (1985). The SEV system: A new disarmed Ti plasmid vector system for plant transformation. Bio/Technology 3, 629-635.

Fromental, C., Kanno, M., Nomiyama, H., and Chambon, P. (1 988). Cooperativity and hierarchical levels of functional orga- nization in the SV40 enhancer. Cell 54, 943-953.

Giuliano, G., Pichersky, E., Malik, V.S., Timko, M.P., Scolnik, P.A., and Cashmore, A.R. (1 988). An evolutionary conserved protein binding sequence upstream of a plant light-regulated gene. Proc. Natl. Acad. Sci. USA 85, 7089-7093.

Green, P.J., Kay, S.A., and Chua, N.-H. (1987). Sequence- specific interactions of a pea nuclear factor with light-respon- sive elements upstream of the rbcS-3A gene. EMBO J. 6,

Green, P.J., Yong, M.H., Cuozzo, M., Kano-Murakami, Y., Sil- verstein, P., and Chua, N.-H. (1988). Binding site requirements for pea nuclear protein factor GT-1 correlate with sequences required for light-dependent transcriptional activation of the rbcS-3A gene. EMBO J. 7,4035-4044.

Jefferson, R.A., Kavanagh, T.A., and Bevan, M.W. (1987). GUS fusions: 0-Glucuronidase as a sensitive and versatile fusion marker in higher plants. EMBO J. 6, 3901-3907.

EMBO J. 8,2195-2202.

2543-2549.

DOI 10.1105/tpc.2.9.857 1990;2;857-866Plant Cell

E Lam, Y Kano-Murakami, P Gilmartin, B Niner and N H Chualight-responsive promoter.

A metal-dependent DNA-binding protein interacts with a constitutive element of a

 This information is current as of October 23, 2020

 

Permissions X

https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists