The Plant Cell, Vol. 4, 971-981, August 1992 O 1992 American Society of Plant Physiologists

A Rice cab Gene Promoter Contains Separate cis-Acting Elements That Regulate Expression in Dicot and Monocot Plants

Sheng Luan’ and Lawrence Bogorad2 Department of Cellular and Developmental Biology, Harvard University, 16 Divinity Avenue, Cambridge, Massachusetts 02138

The major light-harvesting chlorophyll alb binding proteins of the photosynthetic apparatus are encoded by families of nuclear cab genes. The expression of most cab genes is tissue specific and photoregulated in angiosperms. In trans- genic tobacco plants, expression of the reporter gene p-glucuronidase (GUS) is photoregulated and tissue specific from 5’ upstream sequences of the rice cablR gene; deletion of sequences upstream from position -170 with respect to the transcription start site eliminates the enhanced and photoregulated expression in the transgenic plants. Using an in situ transient expression assay, we have determined that the sequence OCT-R, an octamer repeat that lies within the -269 to -170 region of cablR, is essential for photoregulated expression of the chimeric GUS gene in leaf cells of maize and rice but is not required for expression in illuminated tobacco leaves. Conversely, box lll*- and G-box-like sequences found near OCT-R in cablR are necessary for high-leve1 transient expression of the reporter gene in tobacco leaf tissue but are not required for transient expression in maize or rice leaves.

INTRODUCTION

A number of plant nuclear genes are activated by light. The best studied examples are genes of the families encoding chlo- rophyll a/b binding proteins (cab) and the small subunit of ribulose-1 ,Bbisphosphate carboxylase (rbcS) (for reviews, see Tobin and Silverthorne, 1985; Dean et al., 1989). Because light regulation has been determined to be mainly at the transcrip- tional leve1 (Gallagher and Ellis, 1982; Silverthorne and Tobin, 1984), studies in the past few years have been directed at iden- tifying cis- and trans-acting elements that mediate light responsiveness to delineate the pathway of light signal trans- duction in plants.

In addition to ‘TATA” and “CAAT’ boxes, a few other conserved sequences have been found in the 5‘ upstream regions of light- responsive genes (Kuhlemeier et al., 1987; Gilmartin et al., 1990). The GT-1 binding sites, which are present in most rbcS genes, are among the best characterized (Green et al., 1987; Lam and Chua, 1990). These sequence elements include box 11, box 111, box 11: and box 111: They bind to the same protein factor and are functionally redundant in the rbcS9A promoter. The tetramer of box II can confer light responsiveness and tis- sue specificity on the expression of a truncated cauliflower mosaic virus (CaMV) 35s promoter (-90 version) in transgenic

‘ Current address: Department of Chsmistry, Harvard University, 12 Oxford Street, Cambridge, MA 02138.

To whom correspon.dence should be addressed.

tobacco (Lam and Chua, 1990). The G-box is another con- served sequence in rbcS genes that is required for high activity of the Arabidopsis rbcS-1A promoter in transgenic tobacco (Giuliano et al., 1988; Donald and Cashmore, 1990). GATA mo- tifs conserved in both rbcS and cab genes (Castresana et al., 1987; Gidoni et al., 1989) are also involved in high-leve1 ex- pression by illuminated plants (Donald and Cashmore, 1990). Other sequence elements, such as AT-1 binding sites (Datta and Cashmore, 1989) and 3AF1 binding sites (Lam et al., 1990), may also play important roles in light-regulated expression of some genes.

The functional studies of dicot genes described above were conducted in transgenic dicot plants. Severa1 light-responsive genes have been isolated from monocot plants (Lamppa et al., 1985a; Luan and Bogorad, 1989; Matsuoka and Sanada, 1991). Although some of these genes can be regulated properly in transgenic tobacco (Lamppa et al., 1985b; Matsuoka and Sanada, 1991), others cannot (Keith and Chua, 1986). Some short sequences in the promoter regions of monocot genes have limited homology to light-responsive elements of dicot genes. However, none of the sequence elements has been characterized functionally in monocot plants because of the difficulty of producing transgenic monocots.

We have analyzed the roles of sequences in the 5‘ portion of the photoregulated rice cablR gene (Luan and Bogorad, 1989) by studying the expression of a reporter gene in trans- genic tobacco and in in situ transient expression assays. In

972 The Plant Cell

A

- 7 5 0 ____ Bsl I1 AGATCTAGACATCACmTGATTGGGATTAAGGTAATGAGCCCTATCTGATGTCAG

TGGGGATTGmACAGTACCGCAGCACACTGACGTATGGGTCTGGACCCATATG

. -584 T T A G C C A C C G C T A C T G C A T C A G C A G T A T T G C A G A G A A m G C T G C

TCCTCCTCTTAACTATAACTTATATTCAATTTATGTCTCTCGAAAATAGATATGAA

CATACTTTTTTAAAAAATAATACTACATATTGTGAATTTGTGATCCTTACCTTTAC

__ Cla I ATTGAGTTATGACGAACAAClTATCGATTATATAAAAGAMGGATGACTTCTTAT

. -346 -CCAAACAAATCCTATAGTAATGTCTTTTTAACTTTCAGTGACTAACATATACC

. -269 ATCAAACGAGTCCATATTAGGATAATACTACGAAGAATTGTCATCCCACm

GACGGTGTAT * G-box(?) JTACACTGCCACATATCAGTTTrAAAATGAAAACCfiGCTCACCCCAAGCTCACCAAG BoxIII*(?) OCT-R

. -170 AATCTTCGAG-CTCCGCCGAAAAATCTCGGACAAACCCGCGGCTCA

3AF1-binding site (7) ~ Bal I

CACGCCTCCTCGCACCACCACCTAGAATATCCTCT%TTGGCCACGCGCGCA

. -80 :40 CATCAGCTCCCAATCTCCCGCCCCAGGAGGCAATCCCCCCTCGClTCCCGCGCTA

. +I mAAACTCCCGCGCCATCTCCAACTCCC~CTCACACTCGCTCGCTCATCGCCATC

TCTCTCAGCTCTCACAGCTCACTGCATCAATGGCCGCG~CACCATG~GCTCTCCT .+I14

~ Htnc II ’

CCCCGGTWTGGCCCGCGCGGCGCCGTCG/\C

B

-75 +I14

Cabl R-GUS CablR I GUS InosT I

35s-GUS 35s GUS nosT I 35S-Rlc 35s GUS nosT I

binary vector

-750 +114 Figure 1. The Sequence of the cablR Gene Promoter and Diagrams of Constructs Used in the Experiments.

(A) The sequence from -750 to +114 (derived from Bglll/Hincll diges- tion) in relation to the transcription start site of cablR is shown. The 5’ends of the deletion constructs at -750, -584, -346, -269, -170, -80, and -40 are indicated. The fragment between the Clal and Ball

the course of this work, we have identified a 100-bp segment of the rice cablR gene that is required for light-regulated ex- pression in both transgenic tobacco and bombarded leaves of tobacco, maize, and rice. An octamer repeat within this 100- bp region has been characterized and shown to be required for photoresponsiveness of the cablR promoter in maize and rice leaves. However, elimination of the octamer repeat had no effect on expression in tobacco leaves. Other sequences, including previously recognized dicot control regions, affect expression in leaves of this dicot plant.

RESULTS

Deletion Analysis of the cablR Promoter in Transgenic Tobacco

To investigate the functional elements of the cablR promoter sequence, chimeric gene constructs containing different 5’to 3’deletions of the promoter region fused to the P-glucuronidase (GUS) reporter, asshown in Figure 1, were introduced into the genome of tobacco SR1 plants by leaf disc transformation via Agrobacterium tumefaciens (Horsch et al., 1985). As a posi- tive control, the CaMV 35s-GUS construct pB1121 (Jefferson et al., 1987) was introduced into SR1 plants.

As shown in Figure 2A, the average GUS activity in leaves of parenta1 transgenic plants with the 35s-GUS construct is about three times greater than in the plants with the cablR ”fll promoter” (-750/+114-GUS), designated cablR-GUS, con- struct shown in Figure 1B. A deletion mutant containing only 269 bp of 5‘ upstream sequence also exhibited near maximal GUS activity, but no GUS activity was observed from the -170 construct. Comparable results were obtained when GUS ac- tivities of F1 seedlings containing the various constructs were tested histochemically (Figure 28). F1 seeds from transgenic plants with different constructs were surface sterilized and grown in the light for 10 days on MS (Murashige and Skoog, 1962) solidified medium. The tiny seedlings were dropped into a solution of the X-gluc substrate for GUS and kept in the dark for 16 hr. As shown in Figure 28, the cotyledons of the light- grown seedlings with cablR-, -584-, -346-, and -269-GUS as well as 35s-GUS turned blue showing GUS activity, but the -170-GUS-containing transformant did not.

To determine whether GUS activity in transgenic tobacco conferred by cablR is light regulated and, if so, which part (or parts) of the promoter is required for the light responsive- ness, 10-day-old etiolated seedlings with cablR-GUS and

sites was used to make 295A/295B (Figure 6A). The sequence ele- ments of interest are underlined except for the G-box-like sequence. The complementary strand of this sequence is compared with the G-box consensus in Figure 6A, section i. (6) The procedures used to make the constructs cablR-GUS, 35s- GUS, 35S-Rlc, and the binary vector are described in Methods.

Rice cai>1 R Promoter 973

A50

40 -

> 30 -

20H

10-

I •V V

« i • .3 5 S - 7 5 0 - 5 8 4 - 3 4 6 - 2 6 9 - 1 7 0

Constructs

Figure 2. Expression of the GUS Gene in Tobacco Plants Transformedwith 35S-GUS and catoIR Deletion Constructs.(A) GUS activity of leaves from parental transgenic plants was mea-sured by a fluorometric assay (Jefferson, 1987). GUS activity wasassayed in leaves of eight to 12 plants (data from each is shown asa dot) with each construct. GUS activity is expressed as production

35S-GUS constructs were assayed for GUS activity histochem-ically, as described for light-grown seedlings. Plants withcai>1R-GUS did not show GUS activity (Figure 2C), whereasthose with 35S-GUS turned completely blue. GUS activity wasdetectable within 24 hr after etiolated seedlings with ca£>1R-,-584-, -346-, and -269-GUS had been transferred to the light(only a -750-GUS-containing seedling is shown in Figure 2C).Seedlings with the -170-GUS construct did not express GUSactivity even in the light. Also, GUS activity was detected onlyin the green tissues of transgenic plants with the cab1R-GUSconstructs (Figures 2B and 2C). In contrast, GUS activity inplants with 35S-GUS was detectable in stem vascular tissuesand roots as well as in green tissues of both illuminated andunilluminated seedlings. The pattern of cab1R-GUS expres-sion is consistent with the tissue specificity of cab transcriptaccumulation in rice plants (Luan, 1991).

Deletion Analyses of the cablR Promoter inBombarded Leaves of Tobacco, Maize, and Rice

Although transgenic rice and maize plants have been produced(Shimamoto et al., 1989; Gordon-Kamm et al., 1990), the pro-cess is far from routine and is very slow. Therefore, we haveturned to an in situ transient expression assay to analyze thecablR promoter in these plants. To determine whether ahistochemical spot-counting assay of transient expression ofGUS activity in tobacco leaves would yield the same resultsas analyses of transgenic plants, we delivered DNA of thecao1R-GUS deletion series into tobacco leaves by bombard-ment with DNA-coated tungsten microprojectiles using the DuPont PDS 1000 biolistic apparatus. To quantitate GUS activityin the bombarded leaves, a histochemical spot-counting as-say was compared with the results of trangenic tobaccoexperiments. For the histochemical analyses (see Methods),following 48 hr of incubation after bombardment, leaf segmentsfrom each shot were sliced into smaller pieces, put into amicrocentrifuge tube containing 0.8 mL of GUS substrate so-lution (Jefferson, 1987), and incubated in the dark at 37°C for16 hr before counting the number of blue spots under the mi-croscope, as shown in Figure 3.

As shown in Figure 4A, the number of blue spots per shotwas sharply reduced when all of the sequences upstream of-170 were deleted. This is consistent with the expression pat-tern in transgenic tobacco in which promoter activity wasabolished by deleting down to -170 (Figure 2A). Because

of nanomoles of methylumbelliferone per miligram of protein per minute.(B) Histochemical assay of 10-day-old, light-grown F, seedlings shownafter incubation for 24 hr with GUS substrate. From left to right areseedlings with -750, -584, -346, -269, -170, and 35S-GUSconstructs.(C) From left to right are a 10-day-old, dark-grown 35S-GUS seedlingand two cab1R-GUS seedlings, one illuminated for 24 hr and the nextmaintained in darkness before incubation with GUS substrate.

974 The Plant Cell

Figure 3. The Appearance of a Tobacco Leaf Segment after BeingBombarded with catrtR-GUS Construct and Incubated with Histochem-ical GUS Substrate.

The spots were counted and served as a criterion for quantifying GUSactivity.

etiolated tobacco plants have very tiny cotyledons, light/darktransient expression experiments were not performed.

The parallels between the results of GUS expression in trans-genic tobacco leaves and the spot-counting assay opened theway for using the latter to study in situ transient expressionfrom cablR gene promoter constructs in maize and rice leaves.As can be seen from the data in Figure 4A, GUS gene expres-sion from various cab promoter constructs in tobacco leavescan be quantified using the spot-counting assay by perform-ing a number of replicates, but we tested to see whether aninternal control could further aid quantitation. A convenient,unique, visible marker established for maize tissues (Ludwiget al., 1990) proved to be an excellent internal control. The Rlcgene encodes a protein factor that induces cell anthocyaninautonomous pigmentation for most maize tissues (Ludwig etal., 1989). After delivering DNAs of cab1R-GUS and 35S-Rlc(Figure 1B) into living maize leaf tissue, the expression levelof the Rlc protein can be assessed by counting the numberof pigmented cells (Ludwig et al., 1990), and this result canbe compared with the number of blue spots that develop afterthe leaf segments are incubated with X-gluc.

Although we were interested in ultimately testing cablR pro-moter constructs in rice leaves, we chose maize leaves as themajor target material for other reasons as well: (1) maize seed-lings grow well in darkness providing large amounts of leafmaterial as target for bombardment in general and for study-ing the effects of illumination on expression in particular; and(2) maize is closely related to rice but is a C4 plant that con-tains two types of photosynthetic leaf cells, mesophyll andbundle sheath, thus providing the possibility of studying thebehavior of a C3 plant gene (cablR) in cells of a C4 plant.

To determine whether the cablR promoter is light respon-sive in the transient assay in maize leaf tissues, we introducedcab1R-GUS or 35S-GUS into etiolated maize leaves togetherwith 35S-Rlc by bombardment (see Methods). In Figure 4B,the relative GUS activity is given as the number of blue spotsper 100 red pigmented cells. The cab1R-GUS gene was ex-pressed about fourfold more actively in illuminated than inunilluminated leaves of dark-grown maize seedlings, whereasGUS activity was expressed from 35S-GUS equally well in ei-ther condition.

Because a flash is generated by the gun powder in the up-per chamber of the PDS1000 but some light can leak into thesample chamber through a small hole, a mock shot (nomicroprojectiles and no DNA) was made to monitor the effectof the flash on the expression of endogenous cab genes inetiolated maize leaves. The cab mRNA level was not detect-ably different in the bombarded and unbombarded tissues after6 hr of incub Vion in the dark (data not shown). The single flashwas apparently not sufficient to trigger a detectable increasein expression of cab genes in the leaf tissues. Accordingly,far-red light illumination (4 u.mol rn~2 sec~1 for 5 min) did notsignificantly change the level of GUS activity in the samplesincubated in darkness after bombardment (data not shown).

To delineate the region of the promoter that contains ele-ments involved in the photoregulation of cablR, we deliveredGUS constructs with different 5' deletions of the cablR pro-moter into etiolated maize and rice leaves. The deletionconstruct with only 269 bp upstream of the transcription startsite of the cablR gene can still express the GUS reporter genein a light-responsive manner (Figure 4C). The -170 constructexpressed GUS activity at about the same low level as thelonger 5' sequences did in unilluminated tissues but, unlikethe longer promoter sequences, illumination did not result inany significant increase in expression. GUS was expressedfrom constructs with CAAT and TATA sequences (-80) orwith only the TATA box (-40) but at a much lower level thanfrom the -170 construct in both maize (Figure 4C) and rice(Figure 4D).

Dissection of the c/s-Acting Elements in the PromoterRegion (-269 to -170)

The deletion analyses described above showed that sequencesrequired for light-responsive expression of cablR in maize, rice,and tobacco leaves are located between positions -170 and-269. To assess the ability of DNA, including this sequence,to confer photoresponsive expression on a heterologous pro-moter, the cablR sequence from positions -94 to -389 wasligated in both orientations to a GUS reporter gene (Jeffersonet al., 1987) under the control of a truncated CaMV 35S RNApromoter carrying a 5' deletion terminating 89 bp upstreamfrom the transcription start site. This promoter, designated-90/35S, retains TATA and CAAT boxes but not the upstreamenhancer sequences (Fang et al., 1989). The -90 version ofthe 35S promoter has been used frequently as a heterologous

Rice cablR Promoter 975

"1 T T T

-750 -584 -346 -269 -170 -80

Constructs

0 Light

dark

300

200

100

O

35SGUS Cabl R-GUS

C

4w

0 light

-750 -584 3 4 6 -269 -170 -80 -40

Constructs

D

150 T

L u) 100 8 W - n & 50 n

z

.A-

E,

O -750

Figure 4. Transient Expression of cablR-GUS in Maize, Rice, and Tobacco Leaves.

(A) Deletion analysis of the cablR promoter in bombarded tobacco leaves. (6) Light-regulated expression of cablR-GUS in maize leaves. (C) Light responsiveness of the deletion constructs in maize leaves. (D) Light-regulated expression of the deletion constructs in rice leaves. The bombardment procedure is described in Methods and in the text.

,584

T light

346 -269 -170 -80 -40

Constructs

basic promoter for analyzing enhancer activity in plants (Chen et al., 1988; Thomas and Flavell, 1990). When delivered into etiolated maize leaf tissue by the particle gun, both gene fu- sions, designated 295A and 2958, as shown in Figure 5A, expressed GUS activity in a light-dependent manner, as shown in Figure 58. The expression of the GUS gene was threefold higher in illuminated than in unilluminated leaves of dark-grown maize seedlings. On the other hand, the control construct, -90/35S-GUS, expressed GUS activity at a low basal leve1 that was about the same in light and darkness.

In comparing the cablR promoter region with sequences of the light-responsive elements identified earlier (Gilmartin et al., 1990), we found sequences, as shown in Figure 6A, that

have significant homology to the box IW, G-box, and 3AF1.- binding site, sequences previously discovered and analyzed in dicot genes (Green et al., 1987; Giuliano et al., 1988; Lam et al., 1990). In addition, we noticed a sequence containing an octamer repeat (OCT-R) (Figures 1A and 6A). Using poly- merase chain reaction (see Methods), we generated specific mutations in each of these sequences (Figure 6A). The wild- type (295A) and mutant constructs were tested for their capacity to confer photoresponsive expression on the -90/35S-GUS gene in etiolated maize leaves.

As shown in Figure 68, mutations in the sequences resem- bling box 111: G-box, and 3AF1 binding sites (Figure 6A) did not affect the activity or light responsiveness of the GUS

976 The Plant Cell

CablR 35s GUS

A

-90-GUS

295A

2958

B

: B -0

O O r c

P - J

d 5

o - 2

-90

lzT7i-l

I T

T

lo0: O

i 0 Light I

I T

295A 2950 -90-GUS

Figure 5. Analyses of the Light-Responsive Enhancer Element in Maize Leaves.

(A) Diagrams of constructs -90/35S-GUS and 295A/2958. (8) The expression of these constructs in illuminated or unilluminated leaf tissue of dark-grown maize seedlings.

constructs in maize leaves, but deletion of the 15 bp contain- ing most of the repeated octamer sequences (Figure 6A, D34) eliminated the capacity of the gene to exhibit light-dependent GUS expression. Similar results were obtained in rice leaves (data not shown).

To determine whether both repeats are required for the en- hancer activity, two mutant plasmids were constructed, 0“-1 and OCTm-2 (Figure 6A). In each of the mutant constructs, one of the octamer repeats was deleted from 295A. As shown in Figure 66, deletion of either repeat reduced GUS expression to the dark leve1 and eliminated the response to light. Dele- tion of one or both of the octamer repeats changed the spacing of sequences in the -170 to -269 region. The distance be- tween two interacting cis-acting elements has been shown to be important for regulation of some genes (Gilmartin and Chua, 1990). To examine the possibility that the changes in spacing rather than elimination of the octamer sequences in the cablR

gene affects photoregulation, base replacements were gener- ated in the octamer repeats. Two base changes (CC to GG) were made in each octamer sequence (Figure 6A, CG-1). These alterations had about the Same effect as deleting the repeated sequences, as shown in Figure 66 (CG-1).

To study the function of the octamer repeat in the context of the original cablR promoter instead of with the -90/35S sequence, we deleted the 15 bp from the “full” promoter of cablR (-750kablR). A Bglll-Hincll fragment containing 750

A I.

Box-lll’ G-box 3AF1-blndlng slte

Consensus CATITACACT CACGTGGCA AAATAGATAAA C a b l R TmrACACT TATGTGGCA’ AAACTTATAAA Mutants TTGTMBCT TATGTaCA AAACTTAGAE

* The O-boX sequence i6 derived from lhe mmplmrnw strend 01 lhe sequence indicated in Figure i A .

ii.

(Blll-m) (GB-m) (3AF1-m)

WT (OCT-R) AGCTCACCCCAAGCTCACCAAGAAT CAAGAAT D 3 4 AGC

OCTm-1 AGCTCACCCCAAG AGAAT AGCTCACCAAGAAT OCTm-2 AGC

CG-1 AGCTCAQQCC AAGCTC AGGAAGAAT

iii.

Pea rbcSA CACA CA TTTACACT CTT CACAT I I I I 1 1 1 1 I l l l I I I I I

CablA CACA TT TlTACACT GC CACATA

300

200

100

o i F 1 - m OCT.1 OCT-2 CG-1 - 9 0

Constructs

Figure 6. Mutagenesis of Severa1 cisActing Elements.

(A) Various sequences and mutations discussed in the text. Section i shows a comparison of cablR sequences with consensus sequences (Gilmartin et al., 1990). Mutations were made to the underlined bases. Section ii shows the sequence containing the octamer repeat and var- ious mutants of it, D34, OCTm-1, OCTm-2, and CG-1. Deletions are represented as spaces; base changes are underlined. Section iii shows alignment of part of the pearbcS-3A promoter region (-261 to -240; Gilmartin et al., 1990) with a segment of cablR promoter in the -170 to -269 region (-254 to -234 in Figure 1). (8) The effect of mutations on GUS expression in maize leaves.

Rice cablR Promoter 977

Cabl R-GUS I

GUS nosT Cabl R

J

8 400-

e

u) c

U

O

300- - O P

e 3 200- o

TET

[7 Light

Dark

-90/35S-GUS

I I I c

cabl R CD34 -80

Figure 7. Effect of the OCT-R Deletion on the Expression of cablR- GUS in Maize Leaves.

(A) Diagram of the construct with the 15-bp deletion from the -750 to +114 segment of cablR. (B) Elimination of photoregulated expression of cablR-GUS by OCT-R deletion.

bp upstream of the transcription start site and part of the cod- ing region (-750 to +114; Figure 1A) of cablR wascloned into BamHVSmal sites of pBluescript SK+. After deleting the same 15-bp sequence as in D34 (Figure 6A), a GUS reporter gene from pBI101 (Jefferson et al., 1987) was excised by PstllEcoRl digestion and fused in frame with either the wild-type or mu- tated cablR sequence (-750 to +114) in pBluescript SK+ to form cablR-GUS and CD34, as shown in Figure 7A. 60th wild- type cablR-GUS and CD34 DNAs were delivered into etiolated maize leaves. The light-dependent expression of cablR-GUS was completely eliminated by the 15-bp deletion (Figure 78).

The Repeated Sequence Is Not Sufficient to Confer Light Responsiveness

To determine whether OCT-R might be sufficient for the cablR promoter to respond to light, we tested the capacity of a tetramer of the octamer repeat to confer photoresponsiveness on a het- erologous basal promoter. In Figure 8A, a tetramer of the 22-bp

element containing both octamer sequences was made and fused to -90/35S-GUS to generate the construct TET-90/35S- GUS. Using -90/35S-GUS as a negative control and 295A as a positive control, we tested the ability of TET-90/35S-GUS to drive photoresponsive expression in maize leaves. Whereas the sequence that included the wild-type cablR enhancer (-389 to -94) consistently increased GUS expression at least threefold in a photoregulated manner, TET-90/35S-GUS was not capable of responding to light and expressed GUS at a low leve1 like -90/35S-GUS (Figure 8B). This suggests that other sequence elements within the -389 to -94 region are also required for light responsiveness.

The Octamer Repeat Does Not, but Other Sequences Do, Affect Expression of cablR in Tobacco Leaf Tissue

The 5'deletion experiments showed that sequences between -170 and -269 are required for photoregulated expression of cablR-GUS in both transgenic tobacco and bombarded leaves of rice and maize. OCT-R within this region is required

A

AGCTCACCCCAAGCTCACCAAG

B

1 dark

295A -90/35S TET-90135S

Figure 8. Functional Analyses of OCT-R Sequence.

(A) TET-90/35S-GUS construct. (E) The tetramer of the OCT-R sequence is not enough to confer light- regulated expression on a minirnum promoter in maize leaves.

978 The Plant Cell

2 0 5 A 0 3 4 GB-m B111'-m 3 A F l - m - 9 0

Constructs

Figure 9. Expression of Different Constructs in Tobacco Leaf Tissues.

Expanding tobacco leaves were taken from 6- to Week-old plants grow- ing in sterile culture and cut into quarters for bombardment. After bombarded leaves had been incubated at 25OC under fluorescent lamps (250 pmol m-* se&) for 36 hr, leaf tissues were sliced and incubated in the GUS substrate solution for 16 hr. Blue spots were then counted. The mean number of spots and the SE of eight replicates from two separate experiments are shown.

for light-regulated expression of the reporter gene in maize leaves as shown above. In Figure 9, however, deletion of the OCT-R sequences (Figure 6A) had no effect on the capacity of the -170 to -269 region to drive the expression of the GUS gene in tobacco leaves. Conversely, mutations in G-box-like and box 111'-like sequences (Figure 6A) each reduced expres- sion in tobacco by about 50% (Figure 9) but had no effect on expression of the GUS reporter in maize leaves (Figure 66). Because etiolated tobacco cotyledons are so small, we did not study the effects of illumination on expression from mu- tated promoter regions.

DlSCUSSlON

In searching for sequences required for photoresponsiveness of the cablR promoter, we found a sequence element contain- ing an octamer repeat in the -269 to -170 region. Deletion of either or both of the repeated sequences or replacement of bases (CACC to CAGG) in both octamers eliminated the light- dependent expression of the GUS reporter in maize or rice leaves. This suggests that the sequence containing the oc- tamer repeats, AGCTCACCCCAAGCTCACCAAG, is required for the cablR promoter to respond to light. The effect of muta- tion of CC to GG shows that these bases are important for presumptive trans-acting factor(s) to bind and function. How- ever, the data presented in this report do not reveal exactly which of the other bases in this 22-bp sequence are also criti- cal for its function. In addition, there are other nearly identical

symmetrical sequences in the sequence, such as CCAGCT- CACCCCAAGCTCACC; additional mutations will be required to fully define this sequence element. Because deletion of ei- ther half of the element abolished light regulation, there could be an interaction between the putative trans-acting factors that may bind to the two octamers or the two octamers could act as a single protein binding element. A similar repeated se- quence is present in the promoter region of another rice cab gene (cab2R) that is not expressed as strongly as cablR (Luan and Bogorad, 1989; Luan, 1991). However, in cab2R the repeats are separated by a sequence of 40 bp in length. It would be interesting to determine whether this repeat is important in light- regulated expression from the cab2R promoter.

A number of cases have been reported in which a promoter- containing DNA segment of a monocot gene fused to a reporter sequence can be expressed and regulated properly in a widely heterologous system, such as a transgenic dicot (for exam- ple, Lamppa et al., 1985b). We have found this to be true of the cablR promoter segment: cablR-GUS is expressed in a light-responsive and organ-specific manner in transgenic tobacco. Because the sequence between -269 and -170 is required for expression in tobacco as well as in maize or rice, it seemed possible that common cis-acting sequences might exist for light regulation in these plants. However, more detailed analyses have shown that the cabl R promoter contains sepa- rate sequences for expression of cablR-GUS in tobacco.and photoregulated expression in the monocot plants maize and rice. Although deletion of a sequence containing an octamer repeat completely eliminated light-dependent expression in maize or rice leaves, it had no effect on expression in illumi- nated tobacco leaves.

Box 111* and G-box sequences have been implicated in light- regulated expression in dicot plants (reviewed in Gilmartin et al., 1990). In comparing the cablR promoter segment (-269 to -170) with the promoter sequence of pea rbcS3A, we found that box lH*-like and G-box-like sequences are next to each other in both promoters (Figure 1A and Figure 6A, section iii). Replacement of bases in either box Ill*-like or G-box-like se- quences reduced expression of the GUS gene in tobacco leaves by about 50%, but such changes did not alter expres- sion in maize or rice leaves. Box Ill*-like and G-box-like sequences may be required for proper regulation of the cablR promoter in transgenic tobacco, whereas regulation in a more closely related plant, maize, as well as in rice, requires an- other set of sequence elements including OCT-R. It is possible that the box lH*- and G-box-like sequences function in some maize or rice tissues under some circumstances. The fact that the rice cablR gene has regulatory sequences for expression in diverse species implies both genetic conservation (retain- ing sequences from dicot plant genes) and genetic diversity (new elements for specific control) during evolution.

These experiments have also shown that a histochemical in situ spot-counting transient assay in which plant tissues are bombarded with DNA-coated microprojectiles can be used to study the functions of a promoter of a monocot gene in homol- ogous or heterologous plants in cases where the production

Rice cablR Promoter 979

of transgenic plants is difficult or very lengthy. However, we cannot exclude the possibility that some differences may ex- ist between the in situ transient assay and transgenic plant systems. In our case, the -170 construct gave hardly detect- able GUS activity in transgenic tobacco but was expressed at a higher leve1 in bombarded tissues. This might be attrib- uted to differences in the fluorometric and histochemical assays for GUS activity (Luan, 1991).

In conclusion, we have found that there are separate cis- acting elements in the rice cablR gene 5'upstream region that are required for expression in tobacco versus in maize and rice. Curiously, this gene also contains separate sequences, resembling some previously described elements in dicot genes, that regulate expression in tobacco. More detailed analysis of the monocot-specific sequence elements and identification of protein factors binding to them should yield further insights into the mechanisms for light-regulated gene expression in diverse species of plants. These would be important steps toward delineating the differences and similarities in light sig- na1 transduction pathways among these monocot and dicot plant species.

METHODS

Plant Materials

Maize (W23 r-g, AbC, PI) kernels (kindly provided by Dr. E. Coe) and rice (Labelle) seeds (kindly provided by Dr. Charles M. Bollich) were surface sterilized and imbibed in water for 24 hr at room temperature. 60th maize and rice seedlings were grown in the dark at about 28OC for 9 days before their leaves were used for bombardment. Expanded tobacco leaves from plants grown sterilely on MS medium (Murashige and Skoog, 1962) were taken for bombardment.

Constructs and Deletions

35S-GUS was made by cloning the 3.0-kb Hindlll-EcoRI fragment con- taining the CaMV 35s promoter-controlled GUS gene from pB1121 (Jefferson et al., 1987) into pUCl9. To generate cablR-GUS, a Bglll- Hincll fragment from cablR extending from positions -750 to +114 (in relation to the transcription start site; see sequence in Figure 1) was cloned into BamHI/Smal sites of pEluescript SK+ and then fused in frame with a GUS cassette from pBllOl (Jefferson et al., 1987). To construct 35S-Rlc, the GUS cassette was removed from 35s-GUS by BamHl and Sacl, and an Xbal fragment from an Rlc cDNA clone (Ludwig et al., 1989) was inserted in place of GUS. The 5' end dele- tions of the cablR-GUS were generated by Exollllmung bean nucleases from Stratagene following the manufacturer's protocol. For Ti plas- mid-mediated transformation, a cassette containing promder deletions and the GUS gene was cloned into a binary vector pBIN19 (Eevan, 1984) (see Figure 16).

leaves of 6- to 8-week-old explants of tobacco SR1. After incubation for 36 hr with Agrobacterium tumefaciens harboring both the disarmed Ti plasmid (pGV2260) (Deblaere et al., 1985) and the binary vector pEIN19 (Bevan, 1984) containing the transgene constructs, leaf discs were thoroughly washed in liquid B5 medium and placed on regener- ating medium (0.1 pglmL indoleacetic acid, 0.15 pglmL kinetin, and 250 pglmL cefotaxime in 85 medium with 3% sucrose and 0.8% agar) for 4 days before transfer to selection medium (50 pglmL kanamycin in regenerating medium). Plantlets were transferred to rooting medium (selection medium without hormones) until they grew into plants with four to five leaves. Leaves at the same position on 10 to 15 transgenic plants generated from each construct were analyzed. The leve1 of GUS activity varied up to 10-fold among different plants with the same con- struct. DNA gel blot analyses indicated that most plants contained only one copy of the transgene. GUS activity was not related to the appar- ent number of copies of the integrated genes (data not shown). The plants were then transferred into pds and raised to flowering in a growth chamber (14 hr of IightllO hr of dark, 26OC). Some of the flowers were self-pollinated to produce F, seeds.

Fluorometric GUS Assay

A 1-cm-wide segment from leaves at the same position of each of the transgenic tobacco plants was pulverized in a microcentrifuge tube in the presence of liquid nitrogen and lysed in GUS assay buffer (Jeffer- son, 1987). After centrifugation at 12,000 rpm in a microcentrifuge (Fisher) for 10 min, aliquots of the supernatant were taken for enzyme assay, as described by Jefferson (1987).

Mlcroprojectile Bombardment and Histochsmical GUS Assay

A biolistic system (PDS 1000; Du Pont) was used for all the microprojec- tile bombardments. Three-centimeter-long middle sections from the second leaves of etiolated 9-day-old maize seedlings were placed on MS agar (0.6%) plates (three pieces per 4.5-cm-diameter plate). Each plate was bombarded with 2 pL of a 10-pL mixture containing 2 pg of DNA precipitated onto 1.5 mg of tungsten particles with an average diameter of 1.2 fim (for details, see Klein et al., 1987, 1989). For bom- barding maize leaves, 35s-Rlc DNA was mixed with GUS constructs in a molar ratio of 1:2. In lightldark experiments, tissues were bom- barded under a green safelight in adarkroom. After incubation at 25OC for 48 hr under white fluorescent lamps (250 pmol m-* sec-') or in darkness, red dots of anthocyanin generated by Rlcinduced pigmen- tation were visible in maize leai tissue. The red dots were counted under the microscope. The tissues were then cut into about 4-mm-wide seg- ments and put into a microcentrifuge tube containing 0.8 mL of GUS substrate solution (Jefferson, 1987) and incubated at 37% for 24 hr. The number of blue spots was then counted under the microscope, and the ratio of blue spots to red dots was calculated. This was the basis for the quantitative GUS assay. For bombardment of rice leaves, 3-cm-long leaf segments were arranged on an agar plate and bom- barded as described above. Tobacco leaves (about 4 cm wide and 8 cm long) from sterile culture were split into halves along the midvein and cut into quarters; two quarters served as the target for one shot.

Transgenic Tobacco Production and Analyses

The leaf disc infection procedure (Horsch et al., 1985) WaS employed to produce transgenic tobacco plants. Briefly, discs were made from

Site-Directed Mutagenesls

A region of the cablR promder (-389 to -94)~was obtained by re- striction digestion of -750lcablR DNA using Clal and Ball (see

980 The Plant Cell

Figure 1). After filling in the ends by T4 polymerase, this DNA frag- ment was ligated in two orientations to -90135s-GUS in the pBluescript vector to form the constructs 295A and 2958 (Figure 5A). The muta- genesis procedure was basically as described by Sarkar and Sommer (1990), except for some minor changes. Briefly, the DNA fragment be- ing mutagenized was inserted into the pBluescript SK+ vector. In the mutagenesis reactions, three primers were used in the mo-round poly- merase chain reactions (PCRs). One primer contained the desired mutations (mutant primer), and the other two were generated from vector sequences (vector primers)-in this case, the M13 -20 primer se- quence and the reverse primer sequence. The first PCR round was performed using the mutant primer and one of the two vector primers under the reaction conditions of Sarkar and Sommer (1990), except that the temperature and time for the annealing process were changed depending upon the mutations. To generate the 15-bp deletion, for in- stance, we annealed at 32OC for 2 min per cycle. The product of the first PCR was then used as a primer together with the other vector primer in the second PCR round to produce the whole insert sequence containing the desired mutations. The mutations were verified by se- quencing, and the mutated insert sequence was cloned back into the constructs containing the -90/35S-GUS cassette. The 295-bp fragment was mutagenized at various places using essentially the pro- cedure described above while cloned in pBluescript SK+. The mutated -389 to -94 sequence was then excised and cloned back upstream of -90135s-GUS to generate the various mutants: D34, OCTm-1, OCTm-2, Bll ltm, GB-m, and 3AF1-m. The cablR promoter in pBluescript SK+ was used as a template for producing the CD34 mutant.

ACKNOWLEDGMENTS

We thank Dr. Susan Wessler for the Ic cDNA clone. We are grateful to Dr. James DeCamp for critical reading of the manuscript. This work was supported in part by a research grant from the National lnstitute of General Medical Sciences. Early portions of this work were sup- ported in part by the Rockefeller Foundation.

Received March 10, 1992; accepted June 19, 1992.

REFERENCES

Bevan, M.W. (1984). Binary Agrobacterium vectors for plant transfor- mation. Nucl. Acids Res. 12, 8711-8721.

Castresana, C., Garcia-Luque, I., Alonso, E., Malik, V.S., and Cashmore, A.R. (1987). Both positive and negative regulatory ele- ments mediate expression of a photoregulated CAB gene from Nicotiana plumbaginifolia. EMBO J. 7, 1929-1936.

Chen, L L . , Pan, N A , and Beachy, R.N. (1988). A DNA sequence element that confers seed-specific enhancement to a constitutive promoter. EMBO J. 7, 297-302.

Datta, N., and Cashmore, A.R. (1989). Binding of a pea nuclear pro- tein to promoters of certain photoregulated genes is modulated by phosphorylation. Plant Cell 1, 1069-1077.

Dean, C, Pichersky, E., and Dunsmuir, P. (1989). Structure, evolution

and regulation ofrbcS genes in higher plants. Annu. Rev. Plant Phys- iol. 40, 415-439.

Deblaere, R., Bytebier, E., De Greve, H., Deboeck, F., Schell, J., Van Montagu, M., and Leemans, J. (1985). Efficient octopine Ti plasmid-derived vectors for Agrobacterium-mediated gene transfer to plants. NucI. Acids Res. 13, 4777-4788.

Donald, R.G.K., and Cashmore, A.R. (1990). Mutation of either G box or I box sequences profoundly affects expression from the Arabidopsis rbcS-1A promoter. EMBO J. 9, 1717-1726.

Fang, R.-X., Nagy, F., Sivasubramanlam, S., and Chua, NrH. (1989). Multiple cis regulatory elements for maximal expression of the cauliflower mosaic virus 35s promoter in transgenic plants. Plant Cell 1, 141-150.

Gallagher, T.F., and Ellis, R.J. (1982). Light-stimulated transcription of genes for two chloroplast polypeptides in isolated pea leaf nuclei.

Gidoni, D., Brosio, P., Bond-Nutter, D., Bedbrook, J., and Dunsmulr, P. (1989). Nove1 cis-acting elements in petunia cab gene promoters. MOI. Gen. Genet. 215, 337-344.

Gilmartin, P.M., and Chua, N.-H. (1990). Spacing between W-l binding sites within a light-responsive element is critical for transcriptional activity. Plant Cell 2, 447-455.

Gilmartin, P.M., Sarokin, L., Memelink, J., and Chua, N.-H. (1990). Molecular light switches for plant genes. Plant Cell 2, 369-378.

Giullano, O., Pichersky, E., Malik, V.S., Timko, M.P., Scolnik, P.A., and Cashmore, A.R. (1988). A evolutionarily consewed protein bind- ing sequence upstream of plant light-regulated genes. Proc. Natl. Acad. Sci. USA 85, 7089-7093.

Gordon-Kamm, W.J., Spencer, TA., Mangano, M.L., Adams, T.R., Daines, R.J., Start, W.G., OBrlen, J.V., Chambers, S.A., Adams, W.R., Jr., Willetts, N.G., Rlce, T.B., Mackey, C.J., Krueger, R.W., Kausch, A.P., and Lemaux, P.G. (1990). Transformation of maize cells and regeneration of fertile transgenic plants. Plant Cell 2,

Green, P.J., Kay, S.A., and Chua, N.-H. (1987). Sequence-specific interactions of a pea nuclear factor with light-responsive elements upstream of the rbcS-3A gene. EMBO J. 6, 2543-2549.

Horsch, R.B., Fry, J.E., Hoffmann, N.L., Eichholtz, D., Rogers, S.G., and Fraley, R.T. (1985). A simple and general method for transfer- ring genes into plants. Science 227, 1229-1231.

Jefferson, R.A. (1987). The GUSgenefusion system. Plant MOI. Biol. Rep. 5, 387-405.

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

Keith, E., and Chua, N.-H. (1986). Monocot and dicot pre-mRNAs are processed with different efficiencies in transgenic tobacco. EMBO J. 5, 2419-2425.

Kleln, T. M., Wolf, E.D., Wu, R., and Sanford, J.C. (1987). High ve- locity microprojectiles for delivering nucleic acid into living cells. Nature 327, 70-73.

Klein, T.M., Roth, B.A., and Fromm, M.E. (1989). Regulation of an- thothyanin biosynthetic genes introduced into intact maize tissues by microprojectiles. Proc. Natl. Acad. Sci. USA 86, 6681-6685.

Kuhlemeier, C., Green, P.J., and Chua, N.-H. (1987). Regulation of gene expression in higher plants. Annu. Rev. Plant Physiol. 38,

EMBO J. 1, 1493-1498.

603-618.

221-257.

Rice cablR Promoter 981

Lam, E., and Chua, N.-H. (1990). GT-1 binding site confers light- responsive expression in transgenic tobacco. Science 248,471-473.

Lam, E., Kano-Murakami, Y., Gllmartin, P., Niner, E., and Chua, N.-H. (1990). A metal-dependent DNA-binding Protein interacts with a constitutive element of a light-responsive promoter. Plant Cell2,

Ludwig, S.R., Bowen, E., Beach, L., and Wessler, S.R. (1990). A regulatory gene as a nove1 visible marker for maize transformation. Science 247, 449-450.

Matsuoka, M., and Sanada, Y. (1991). Expression of photosynthetic genes from C4 plant, maize, in tobacco. MOI. Gen. Genet. 225,

857-866. 41 1-41 9. Lamppa, G., Morelli, G., and Chua, N.-H. (1985a). Structure and de-

velopmental regulation of a wheat gene encoding the major chlorophyll alb binding polypeptide. MOI. Cell. Biol. 5, 1370-1376.

Lamppa, O., Nagy, F., and Chua, N.-H. (1985b). Light-regulated and tissue-specific expression of a wheat cab gene in transgenic tobacco. Nature 315, 750-753.

Luan, S. (1991). Rice Genes Encoding Light-Harvesting Chlorophyll alb-Binding Proteins: Molecular Characterization of Photoregulated Expression. Ph.D. Dissertation (Cambridge, MA: Harvard University).

Luan, S., and Bogorad, L. (1989). Nucleotide sequences of two genes encoding light-harvesting chlorophyll alb binding proteins in rice.

Murashige, T., and Skoog, F. (1962). A revised medium for rapid growth and bioassays with tobacco cultures. Physiol. Plant. 15, 473-497.

Sarkar, G., and Sommer, S.S. (1990). The “megaprimer” method of site-directed mutagenesis. BioTechniques 8, 404-407.

Shimamoto, K., Terada, R., Izawa, T., and Fujimoto, H. (1989). Fer- tile transgenic rice plants regenerated from transformed protoplasts. Nature 338, 274-276.

Silverthorne, J., and Tobin, E.M. (1984). Demonstration of transcrip- tional regulation of specific genes by phytochrome action. Proc. Natl. Acad. Sci. USA 81, 1112-1116.

Nucl. A i d s Res. 17, 2357-2358. Ludwlg, S.R., Habera, L.F., Dellaporta, S.L., and Wessler, S.R.

(1989). Lc, a member of the maize R gene family responsible for tissue-specific anthocyanin production, encodes a protein similar to transcriptional activators and contains a myc-homology region. Proc. Natl. Acad. Sci. USA 86, 7092-7097.

Thomas, M.S., and Flavell, R.B. (1990). ldentification of an enhancer element for the endosperm-specific expression of high molecular weight glutenin, Plant Cell 2, 1171-1180,

Tobln, E.M., and Sllverthorne, J. (1985). Light regulation of gene ex- pression in higher plants. Annu. Rev. Plant Physiol. 36, 569-593.

DOI 10.1105/tpc.4.8.971 1992;4;971-981Plant Cell

S Luan and L Bogoradand monocot plants.

A rice cab gene promoter contains separate cis-acting elements that regulate expression in dicot

 This information is current as of March 2, 2021

 

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

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