- Email: [email protected]
Promoter activity of the 5′ flanking region of a tobacco glycolate oxidase gene1 in transgenic tobacco plants
Biochimica et Biophysica Acta 1399 (1998) 105^110
Promoter paper
Promoter activity of the 5P £anking region of a tobacco glycolate oxidase gene1 in transgenic tobacco plants Simon Barak, Ali Nejidat, Micha Volokita * The Jacob Blaustein Institute for Desert Research, Ben-Gurion University of the Negev, Sede Boqer Campus 84990, Israel Received 27 April 1998; accepted 11 May 1998
Abstract A clone harboring the 5P flanking region of a tobacco glycolate oxidase (GLO) gene was isolated from a VEMBL3-tobacco genomic DNA libarary. Primer extension analysis indicated two major transcripts with 76 and 81 bp 5P UTRs. An RT-PCR assay mapped the major mRNA transcription initiation site to thymine at position 81 upstream of the translation initiation codon. A putative TATA box spanning positions 356 to 350 upstream of the transcription initiation site was found. Promoter activity of the 5P flanking region (33.0 kb to +82 bp) was demonstrated in tobacco plants transformed with a GLO-L-glucuronidase (GUS) chimeric gene. Furthermore, in these transgenic plants, GUS expression patterns mimicked the expression patterns of the endogenous GLO. ß 1998 Elsevier Science B.V. All rights reserved. Keywords: Glycolate oxidase; Promoter; Transcription initiation site; Light-dependent expression ; Transgenic plant
The photorespiratory glycolate pathway in C3 plants is linked to the photosynthetic carbon-reduction pathway by the bifunctional chloroplast enzyme Rubisco [1] and results in the loss of a large fraction of ¢xed carbon in the form of CO2 [2]. Glycolate oxidase (GLO) is a peroxisomal enzyme that catalyzes the oxidation of photosynthetic glycolate to yield glyoxylate and H2 O2 in green leaves [3]. GLO is expressed in a tissue-speci¢c manner in cells whose plastids have developed into functional chloroplasts [4]. During the light-induced greening of etiolated seedlings and leaves of various C3 plants, glycolate
* Corresponding author. Fax: +972 (7) 659-6704; E-mail: [email protected] 1 The sequence data reported in this paper have been submitted to the GeneBank Data Libraries under the accession number U62485.
oxidase mRNA accumulates [5,6] and its activity increases [7]. These data indicate that GLO gene expression is principally regulated at the transcriptional level, as has been found for other plant genes involved in photorespiration and photosynthesis [8,9]. As an initial step in elucidating the GLO gene mode of expression, we report the cloning of the 5P £anking region of a tobacco glycolate oxidase gene and its functional promoter activity in stably transformed lines of tobacco. Approximately 1U106 phage plaques in a tobacco cv. Samson NN genomic DNA library (kindly provided by Dr. R. Fluhr, The Weizmann Institute of Science, Rehovot, Israel) were screened with spinach 32 P-labeled cDNA encoding glycolate oxidase [10]. Two positive clones (VGLO76 and VGLO91) that had an identical 17 kb DNA insert were isolated. This genomic DNA insert contains a partial glycolate oxidase gene consisting of the 5P £anking region and
0167-4781 / 98 / $19.00 ß 1998 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 7 8 1 ( 9 8 ) 0 0 0 8 3 - 9
BBAEXP 91147 24-7-98
106
S. Barak et al. / Biochimica et Biophysica Acta 1399 (1998) 105^110
about two-thirds of the coding region as determined by DNA sequencing (see footnote 1 for sequence data). Fig. 1A schematically shows the approximate 6 kbp BamHI/SalI fragment that contains the partial coding region and about 3 kbp of the 5P £anking region upstream of the ¢rst translation initiation codon. The partial coding region encompasses the 3P end of the original 17 kb DNA insert and consists of seven exons and introns, the last intron of relatively large size. The putative intron splice junctions (data not shown) conform to the AT^GT rule [11]. We speculate that we failed to isolate the rest of the gene because the exons that encode approximately the last 153 amino acid residues of the tobacco glycolate oxidase are small and scattered among relatively long introns. The coding region yielded a predicted polypeptide of 217 amino acids displaying over 90% identity when compared with the ¢rst 216 amino acid residues deduced from the primary glycolate oxidase sequences of spinach cDNA [10] and lentil cDNA [5]. Fig. 1B depicts a comparison of the amino acid sequence deduced from the ¢rst exon of the tobacco gene and the N-terminus amino acid sequences of the spinach and the lentil cDNAs. The tobacco GLO gene contains an additional glutamine residue at position 3 downstream of the N-terminus (Fig. 1B). A search for known regulatory elements in approximately 800 bp of the 5P £anking region upstream of the major GLO transcription initiation site (see below) yielded only a putative I-box [12] (Fig. 1C). Primer extension analysis was performed to determine the transcription initiation site of the mature GLO mRNA. A 32 P-end-labeled oligonucleotide complementary to nucleotides +76 to +102 that encompass the ¢rst translated ATG codon was annealed with the total RNA isolated from green leaves of 2-month-old tobacco plants. The primer was extended by MMLV reverse transcriptase and the cDNA products were analyzed by electrophoresis on a denaturing polyacrylamide gel. As shown in Fig. 2, two major putative initiation sites were detected. Using the same primer for sequencing the 3.5 kb HindIII/SalI fragment in VGLO76 maps thymine and cytosine to positions 81 and 76 upstream of the ¢rst translated ATG, as the 5P ends of the 5P UTR. Besides these two major 5P ends of the transcripts, some other minor halts of the polymerase
were observed on both upstream and downstream sequences. A putative TATA box was found [13], spanning positions 356 to 350 relative to the thymine at position 81 upstream of the translation initiation codon (Fig. 1C). In order to verify the transcription start site as determined by the primer extension assay and to eliminate the possibility that an intron(s) may interrupt the 5P UTR as was shown for several other plant genes [14,15] a PCR assay was designed. PCR products were generated by the same primer pairs, using either genomic DNA or single-stranded cDNA synthesized from total leaf RNA (RT-PCR) as templates. When the primer pair A and B (see Fig. 3A) were used with genomic DNA, the expected 440 bp size PCR product was yielded (Fig. 3B, lane 3). This PCR product consists of 260 bp of exons I and II, 97 bp of the ¢rst deduced GLO intron and 81 bp of the 5P £anking region upstream of the ¢rst translation codon (see Fig. 3A). The same primer pair produced an approximately 340 bp size RT-PCR product (Fig. 3B, lane 4) as could be estimated by comparison of its size to the genomic PCR product of 357 bp produced by the primer pair A/D (Fig. 3B, lane 2). The estimated di¡erence of 100 bp between the genomic PCR and the RT-PCR products generated by primer pair A/B, as based on the agarose gel electrophoresis results, is in agreement with the deduced 97 bp size of the ¢rst GLO intron. Based on these ¢gures, the size of the 5P UTR of the GLO transcript is about 80 bp. This number is in agreement with the result of the primer extension assay. Thus, these results ruled out the possibility of the existence of a sizable intron between the translation initiation codon and the transcription start site as mapped by the primer extension assay. Moreover, the primer pair A/C failed to yield an RT-PCR product (Fig. 3B, lane 6) while the same primer pair produced the expected 460 bp genomic PCR product (Fig. 3B, lane 5). The most plausible interpretation of these results is that the genomic sequence complementary to 5P-TAACTCATCCAGCATCTTAATGTG-3P (oligonucleotide B) is transcribed and comprises the 5P end of the mature GLO mRNA. It should be noted that in the present study we were interested in the particular cloned GLO gene and therefore the annealing temperature for the PCR was set to the highest value to allow annealing of
BBAEXP 91147 24-7-98
S. Barak et al. / Biochimica et Biophysica Acta 1399 (1998) 105^110
107
Fig. 1. Structure of the tobacco glycolate oxidase gene (partial). (A) Restriction enzyme mapping of the coding region and its adjacent 5P £anking region. The blank boxes denote exons. Restriction enzyme sites are: B, BamHI; E, EcoRI; H, HindIII; S, SalI; and Sp, SphI. (B) Comparison of the deduced amino acid sequence of the tobacco GLO exon I with the deduced N-terminus sequences of the spinach [10] and lentil [5] glycolate oxidase, respectively. (C) Nucleotide sequence of 1045 bp of the SphI/EcoRI tobacco GLO fragment that contains the ¢rst exon and its 5P £anking region. For numbering, the major transcription start point (T), denoted by an asterisk, was set as +1. The putative TATA and I boxes are underlined and boxed, respectively, and the ¢rst ATG codon is in bold face.
the primers to sequences that are near 100% homology. Thus the possibility of positive reaction with other putative glycolate oxidase gene family members was quite low. Based on the RT-PCR and the primer extension experiments we assigned the transcription
initiation start site to the thymine at nucleotide +1 as shown in Fig. 1C. The RT-PCR experiment clearly demonstrated the presence of transcripts related to the partially cloned GLO gene in tobacco leaves. The next step was to
BBAEXP 91147 24-7-98
108
S. Barak et al. / Biochimica et Biophysica Acta 1399 (1998) 105^110
Fig. 2. Primer extension analysis of the transcription start site of GLO. A synthetic oligonucleotide, 5P-GACATTTGTTACTTCCTCCATCTTTAG-3P which is complementary to nucleotides +76 to +103 (see Fig. 1C) was radiolabeled at its 5Pterminus. Total leaf RNA from 10-day-old tobacco seedlings was used as the template for primer extension. The size of the extension product was analyzed on a sequencing gel (lane 1). The sequencing ladder was generated by using the same primer in dideoxy sequencing reactions with the 3.5 kb HindIII/SalI fragment of the GLO genomic clone.
determine whether the 5P £anking region (33.0 kb to +82 bp) of this gene is functioning as a promoter. For this purpose, the 5P £anking region of the tobacco GLO gene was fused to the bacterial uidA gene encoding the L-glucuronidase (GUS) enzyme. The GLO-GUS construct was subcloned into the binary vector pBI101 by replacing the original GUS gene in this vector [16]. Genetically stable transformed tobacco (Nicotiana tabacum, var. Samsun) plants were obtained by essentially using the Agrobacterium tumefaciens-mediated leaf disc transformation method as described by Kavanagh et al. [17]. Four independent transgenic tobacco lines containing the chimeric gene GLO-GUS (as proved by PCR; data not shown) were obtained and found to have GUS activity, thus demonstrating the promoter activity of the cloned 3 kb fragment. In these transgenic lines, GUS-speci¢c activity in the cotyledons varied within the range of two orders of magnitude. In order to con¢rm that the 3 kb promoter confers similar expression patterns on both the reporter and the endogenous GLO genes, GUS and GLO activities were compared in the transgenic plants. As with
Fig. 3. PCR mapping of the 5P end of the GLO transcript (A) Schematic diagram of the GLO gene region in the vicinity of the ¢rst translational codon. Primer A, 5P-TTCAGGATGTGCCATTTTCTGCAT-3P; primer B, 5P-TAACTCATCCAGCATCTTAATGTG-3P; primer C, 5P-TGAGCTTAGATAATTATAAGACAA-3P; primer D, 5P-ATGGAGGAAGTAACAAATGTCATG-3P. (B) PCR products ampli¢ed by di¡erent primer pairs using genomic or single-stranded cDNA as templates. RT-PCR analysis was performed on RNA isolated from leaves of 10-week-old tobacco plants. The RNA was treated with RNase-free DNase prior to the reverse transcriptase reaction. The RNA was converted to single-stranded cDNA by reverse transcription with random hexamer primers. Primers speci¢c for GLO were used in the polymerase chain reaction with the cDNA. For genomic DNA analysis, PCR was performed on DNA extracted from leaves of 10-week-old tobacco plants. The 360 bp size product of the PCR, performed with primers A and D on genomic DNA, was used for size calibration of the RT-PCR product.
BBAEXP 91147 24-7-98
S. Barak et al. / Biochimica et Biophysica Acta 1399 (1998) 105^110
Fig. 4. Light-dependent induction of GUS and endogenous glycolate oxidase activities in transgenic tobacco seedlings. GLO and GUS activities [16,18] were determined in crude homogenates prepared from seedlings grown for 10 days after sowing in the light (50 WE cm32 s31 ) or in the dark. L, light; D, dark; D - L, 9-day-old etiolated seedlings transferred to the light regime for 24 h. For ease of comparison, data for the two transgenic lines were normalized; the highest speci¢c activity determined for each transgenic line was set as 100% (28.9 þ 8.9 and 0.42 þ 0.03 nmol min31 mg31 protein for GLO and GUS activities, respectively, in GLO-GUS line; and 34.8 þ 2.9 and 71.3 þ 4.5 nmol min31 mg31 protein for GLO and GUS activities, respectively, in the CaMV35S-GUS transgenic line). The data are the mean of three replicates. Each replicate consists of ca. 30 pooled seedlings. The bars represent the standard errors of the mean. (A) GLO-GUS transgenic plants. (B) CaMV35SGUS transgenic plants.
the endogenous glycolate oxidase enzyme, GUS was expressed in the cotyledons and leaves, but not in the roots. Furthermore, all transgenic lines displayed light-induced GUS activity (data not shown). Four
109
independent control transgenic tobacco lines expressing the CaMV35S-GUS construct displayed a high light-independent GUS activity in roots and leaves (data not shown) as expected from a constitutive promoter [16]. A comparison between the apparent light-dependent expression patterns of GUS and of the endogenous glycolate oxidase was carried out in one each of the GLO-GUS and CaMV35S-GUS expressing lines (Fig. 4). Glycolate oxidase and GUS activities, in cotyledons of GLO-GUS seedlings germinated and grown in complete darkness, were about 15^20% of the activities measured in light-grown seedlings. Within 24 h of illumination, cotyledons of dark-grown seedlings responded with a 2-fold increase in the activities of both enzymes (Fig. 4A). In contrast, the pattern of GUS activity in the CaMV35S-GUS transgenic plants was not correlated with that of endogenous glycolate oxidase and was not induced by light (Fig. 4B). Taken together, the results obtained from the transgenic plants suggest that the 3 kb 5P £anking region is functioning as a GLO promoter conferring similar expression patterns on the GUS gene as on the endogenous glycolate oxidase gene. It is suggested that this region contains the regulatory sequences of the GLO promoter. We would like to thank Gila Granot for excellent technical assistance. A.N. was supported by a Deichmann fellowship. This study was supported in part by the Israel Science Foundation founded by the Israel Academy of Sciences and Humanities.
References [1] G.H. Lorimer, Ann. Rev. Plant Physiol. 32 (1981) 349^ 383. [2] C.W. Husic, D.H. Husic, N.E. Tolbert, Crit. Rev. Plant Sci. 5 (1987) 45^100. [3] N.N. Artus, S.C. Somerville, C.R. Somerville, Crit. Rev. Plant Sci. 4 (1986) 121^147. [4] R. Boldt, S. Koshuchowa, W. Gross, T. Bo«rner, C. Schnarrenberger, Plant Sci. 124 (1997) 33^40. [5] C. Ludt, H. Kindl, Plant Physiol. 94 (1990) 1193^1198. [6] M. Esaka, N. Yamada, M. Kitabayashi, Y. Setoguchi, R. Tsugeki, M. Kondo, M. Nishimura, Plant Mol. Biol. 33 (1997) 141^155. [7] J. Feierabend, H. Beevers, Plant Physiol. 49 (1972) 28^32. [8] J.M. Greenler, W.M. Becker, Plant Physiol. 94 (1990) 1484^ 1487.
BBAEXP 91147 24-7-98
110
S. Barak et al. / Biochimica et Biophysica Acta 1399 (1998) 105^110
[9] C. Kuhlemeier, P. Green, N.-H. Chua, Ann. Rev. Plant Physiol. 38 (1987) 221^257. [10] M. Volokita, C.R. Somerville, J. Biol. Chem. 262 (1987) 15825^15828. [11] S.M. Mount, Nucleic Acids Res. 10 (1981) 168^181. [12] P.M. Gilmartin, L. Sarokin, J. Memelink, N.-H. Chua, Plant Cell 2 (1990) 369^378. [13] C.P. Joshi, Nucleic Acids Res. 15 (1987) 6643^6653. [14] R. Nakano, T. Matsumura, H. Sakakibara, T. Sugiyama, T. Hase, Plant Cell Physiol. 38 (1997) 1167^1170.
[15] S. Huang, Y.-Q. An, J.M. McDowell, E.C. McKinney, R.B. Meagher, Plant Mol. Biol. 33 (1997) 125^139. [16] R.A. Je¡erson, T.A. Kavanagh, M.W. Beavan, EMBO J. 6 (1987) 3901^3907. [17] T.A. Kavanagh, R.A. Je¡erson, M.W. Beavan, Mol. Gen. Genet. 215 (1988) 38^45. [18] A.L. Baker, N.E. Tolbert, Methods Enzymol. 9 (1966) 338^ 342.
BBAEXP 91147 24-7-98
Promoter paper
Promoter activity of the 5P £anking region of a tobacco glycolate oxidase gene1 in transgenic tobacco plants Simon Barak, Ali Nejidat, Micha Volokita * The Jacob Blaustein Institute for Desert Research, Ben-Gurion University of the Negev, Sede Boqer Campus 84990, Israel Received 27 April 1998; accepted 11 May 1998
Abstract A clone harboring the 5P flanking region of a tobacco glycolate oxidase (GLO) gene was isolated from a VEMBL3-tobacco genomic DNA libarary. Primer extension analysis indicated two major transcripts with 76 and 81 bp 5P UTRs. An RT-PCR assay mapped the major mRNA transcription initiation site to thymine at position 81 upstream of the translation initiation codon. A putative TATA box spanning positions 356 to 350 upstream of the transcription initiation site was found. Promoter activity of the 5P flanking region (33.0 kb to +82 bp) was demonstrated in tobacco plants transformed with a GLO-L-glucuronidase (GUS) chimeric gene. Furthermore, in these transgenic plants, GUS expression patterns mimicked the expression patterns of the endogenous GLO. ß 1998 Elsevier Science B.V. All rights reserved. Keywords: Glycolate oxidase; Promoter; Transcription initiation site; Light-dependent expression ; Transgenic plant
The photorespiratory glycolate pathway in C3 plants is linked to the photosynthetic carbon-reduction pathway by the bifunctional chloroplast enzyme Rubisco [1] and results in the loss of a large fraction of ¢xed carbon in the form of CO2 [2]. Glycolate oxidase (GLO) is a peroxisomal enzyme that catalyzes the oxidation of photosynthetic glycolate to yield glyoxylate and H2 O2 in green leaves [3]. GLO is expressed in a tissue-speci¢c manner in cells whose plastids have developed into functional chloroplasts [4]. During the light-induced greening of etiolated seedlings and leaves of various C3 plants, glycolate
* Corresponding author. Fax: +972 (7) 659-6704; E-mail: [email protected] 1 The sequence data reported in this paper have been submitted to the GeneBank Data Libraries under the accession number U62485.
oxidase mRNA accumulates [5,6] and its activity increases [7]. These data indicate that GLO gene expression is principally regulated at the transcriptional level, as has been found for other plant genes involved in photorespiration and photosynthesis [8,9]. As an initial step in elucidating the GLO gene mode of expression, we report the cloning of the 5P £anking region of a tobacco glycolate oxidase gene and its functional promoter activity in stably transformed lines of tobacco. Approximately 1U106 phage plaques in a tobacco cv. Samson NN genomic DNA library (kindly provided by Dr. R. Fluhr, The Weizmann Institute of Science, Rehovot, Israel) were screened with spinach 32 P-labeled cDNA encoding glycolate oxidase [10]. Two positive clones (VGLO76 and VGLO91) that had an identical 17 kb DNA insert were isolated. This genomic DNA insert contains a partial glycolate oxidase gene consisting of the 5P £anking region and
0167-4781 / 98 / $19.00 ß 1998 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 7 8 1 ( 9 8 ) 0 0 0 8 3 - 9
BBAEXP 91147 24-7-98
106
S. Barak et al. / Biochimica et Biophysica Acta 1399 (1998) 105^110
about two-thirds of the coding region as determined by DNA sequencing (see footnote 1 for sequence data). Fig. 1A schematically shows the approximate 6 kbp BamHI/SalI fragment that contains the partial coding region and about 3 kbp of the 5P £anking region upstream of the ¢rst translation initiation codon. The partial coding region encompasses the 3P end of the original 17 kb DNA insert and consists of seven exons and introns, the last intron of relatively large size. The putative intron splice junctions (data not shown) conform to the AT^GT rule [11]. We speculate that we failed to isolate the rest of the gene because the exons that encode approximately the last 153 amino acid residues of the tobacco glycolate oxidase are small and scattered among relatively long introns. The coding region yielded a predicted polypeptide of 217 amino acids displaying over 90% identity when compared with the ¢rst 216 amino acid residues deduced from the primary glycolate oxidase sequences of spinach cDNA [10] and lentil cDNA [5]. Fig. 1B depicts a comparison of the amino acid sequence deduced from the ¢rst exon of the tobacco gene and the N-terminus amino acid sequences of the spinach and the lentil cDNAs. The tobacco GLO gene contains an additional glutamine residue at position 3 downstream of the N-terminus (Fig. 1B). A search for known regulatory elements in approximately 800 bp of the 5P £anking region upstream of the major GLO transcription initiation site (see below) yielded only a putative I-box [12] (Fig. 1C). Primer extension analysis was performed to determine the transcription initiation site of the mature GLO mRNA. A 32 P-end-labeled oligonucleotide complementary to nucleotides +76 to +102 that encompass the ¢rst translated ATG codon was annealed with the total RNA isolated from green leaves of 2-month-old tobacco plants. The primer was extended by MMLV reverse transcriptase and the cDNA products were analyzed by electrophoresis on a denaturing polyacrylamide gel. As shown in Fig. 2, two major putative initiation sites were detected. Using the same primer for sequencing the 3.5 kb HindIII/SalI fragment in VGLO76 maps thymine and cytosine to positions 81 and 76 upstream of the ¢rst translated ATG, as the 5P ends of the 5P UTR. Besides these two major 5P ends of the transcripts, some other minor halts of the polymerase
were observed on both upstream and downstream sequences. A putative TATA box was found [13], spanning positions 356 to 350 relative to the thymine at position 81 upstream of the translation initiation codon (Fig. 1C). In order to verify the transcription start site as determined by the primer extension assay and to eliminate the possibility that an intron(s) may interrupt the 5P UTR as was shown for several other plant genes [14,15] a PCR assay was designed. PCR products were generated by the same primer pairs, using either genomic DNA or single-stranded cDNA synthesized from total leaf RNA (RT-PCR) as templates. When the primer pair A and B (see Fig. 3A) were used with genomic DNA, the expected 440 bp size PCR product was yielded (Fig. 3B, lane 3). This PCR product consists of 260 bp of exons I and II, 97 bp of the ¢rst deduced GLO intron and 81 bp of the 5P £anking region upstream of the ¢rst translation codon (see Fig. 3A). The same primer pair produced an approximately 340 bp size RT-PCR product (Fig. 3B, lane 4) as could be estimated by comparison of its size to the genomic PCR product of 357 bp produced by the primer pair A/D (Fig. 3B, lane 2). The estimated di¡erence of 100 bp between the genomic PCR and the RT-PCR products generated by primer pair A/B, as based on the agarose gel electrophoresis results, is in agreement with the deduced 97 bp size of the ¢rst GLO intron. Based on these ¢gures, the size of the 5P UTR of the GLO transcript is about 80 bp. This number is in agreement with the result of the primer extension assay. Thus, these results ruled out the possibility of the existence of a sizable intron between the translation initiation codon and the transcription start site as mapped by the primer extension assay. Moreover, the primer pair A/C failed to yield an RT-PCR product (Fig. 3B, lane 6) while the same primer pair produced the expected 460 bp genomic PCR product (Fig. 3B, lane 5). The most plausible interpretation of these results is that the genomic sequence complementary to 5P-TAACTCATCCAGCATCTTAATGTG-3P (oligonucleotide B) is transcribed and comprises the 5P end of the mature GLO mRNA. It should be noted that in the present study we were interested in the particular cloned GLO gene and therefore the annealing temperature for the PCR was set to the highest value to allow annealing of
BBAEXP 91147 24-7-98
S. Barak et al. / Biochimica et Biophysica Acta 1399 (1998) 105^110
107
Fig. 1. Structure of the tobacco glycolate oxidase gene (partial). (A) Restriction enzyme mapping of the coding region and its adjacent 5P £anking region. The blank boxes denote exons. Restriction enzyme sites are: B, BamHI; E, EcoRI; H, HindIII; S, SalI; and Sp, SphI. (B) Comparison of the deduced amino acid sequence of the tobacco GLO exon I with the deduced N-terminus sequences of the spinach [10] and lentil [5] glycolate oxidase, respectively. (C) Nucleotide sequence of 1045 bp of the SphI/EcoRI tobacco GLO fragment that contains the ¢rst exon and its 5P £anking region. For numbering, the major transcription start point (T), denoted by an asterisk, was set as +1. The putative TATA and I boxes are underlined and boxed, respectively, and the ¢rst ATG codon is in bold face.
the primers to sequences that are near 100% homology. Thus the possibility of positive reaction with other putative glycolate oxidase gene family members was quite low. Based on the RT-PCR and the primer extension experiments we assigned the transcription
initiation start site to the thymine at nucleotide +1 as shown in Fig. 1C. The RT-PCR experiment clearly demonstrated the presence of transcripts related to the partially cloned GLO gene in tobacco leaves. The next step was to
BBAEXP 91147 24-7-98
108
S. Barak et al. / Biochimica et Biophysica Acta 1399 (1998) 105^110
Fig. 2. Primer extension analysis of the transcription start site of GLO. A synthetic oligonucleotide, 5P-GACATTTGTTACTTCCTCCATCTTTAG-3P which is complementary to nucleotides +76 to +103 (see Fig. 1C) was radiolabeled at its 5Pterminus. Total leaf RNA from 10-day-old tobacco seedlings was used as the template for primer extension. The size of the extension product was analyzed on a sequencing gel (lane 1). The sequencing ladder was generated by using the same primer in dideoxy sequencing reactions with the 3.5 kb HindIII/SalI fragment of the GLO genomic clone.
determine whether the 5P £anking region (33.0 kb to +82 bp) of this gene is functioning as a promoter. For this purpose, the 5P £anking region of the tobacco GLO gene was fused to the bacterial uidA gene encoding the L-glucuronidase (GUS) enzyme. The GLO-GUS construct was subcloned into the binary vector pBI101 by replacing the original GUS gene in this vector [16]. Genetically stable transformed tobacco (Nicotiana tabacum, var. Samsun) plants were obtained by essentially using the Agrobacterium tumefaciens-mediated leaf disc transformation method as described by Kavanagh et al. [17]. Four independent transgenic tobacco lines containing the chimeric gene GLO-GUS (as proved by PCR; data not shown) were obtained and found to have GUS activity, thus demonstrating the promoter activity of the cloned 3 kb fragment. In these transgenic lines, GUS-speci¢c activity in the cotyledons varied within the range of two orders of magnitude. In order to con¢rm that the 3 kb promoter confers similar expression patterns on both the reporter and the endogenous GLO genes, GUS and GLO activities were compared in the transgenic plants. As with
Fig. 3. PCR mapping of the 5P end of the GLO transcript (A) Schematic diagram of the GLO gene region in the vicinity of the ¢rst translational codon. Primer A, 5P-TTCAGGATGTGCCATTTTCTGCAT-3P; primer B, 5P-TAACTCATCCAGCATCTTAATGTG-3P; primer C, 5P-TGAGCTTAGATAATTATAAGACAA-3P; primer D, 5P-ATGGAGGAAGTAACAAATGTCATG-3P. (B) PCR products ampli¢ed by di¡erent primer pairs using genomic or single-stranded cDNA as templates. RT-PCR analysis was performed on RNA isolated from leaves of 10-week-old tobacco plants. The RNA was treated with RNase-free DNase prior to the reverse transcriptase reaction. The RNA was converted to single-stranded cDNA by reverse transcription with random hexamer primers. Primers speci¢c for GLO were used in the polymerase chain reaction with the cDNA. For genomic DNA analysis, PCR was performed on DNA extracted from leaves of 10-week-old tobacco plants. The 360 bp size product of the PCR, performed with primers A and D on genomic DNA, was used for size calibration of the RT-PCR product.
BBAEXP 91147 24-7-98
S. Barak et al. / Biochimica et Biophysica Acta 1399 (1998) 105^110
Fig. 4. Light-dependent induction of GUS and endogenous glycolate oxidase activities in transgenic tobacco seedlings. GLO and GUS activities [16,18] were determined in crude homogenates prepared from seedlings grown for 10 days after sowing in the light (50 WE cm32 s31 ) or in the dark. L, light; D, dark; D - L, 9-day-old etiolated seedlings transferred to the light regime for 24 h. For ease of comparison, data for the two transgenic lines were normalized; the highest speci¢c activity determined for each transgenic line was set as 100% (28.9 þ 8.9 and 0.42 þ 0.03 nmol min31 mg31 protein for GLO and GUS activities, respectively, in GLO-GUS line; and 34.8 þ 2.9 and 71.3 þ 4.5 nmol min31 mg31 protein for GLO and GUS activities, respectively, in the CaMV35S-GUS transgenic line). The data are the mean of three replicates. Each replicate consists of ca. 30 pooled seedlings. The bars represent the standard errors of the mean. (A) GLO-GUS transgenic plants. (B) CaMV35SGUS transgenic plants.
the endogenous glycolate oxidase enzyme, GUS was expressed in the cotyledons and leaves, but not in the roots. Furthermore, all transgenic lines displayed light-induced GUS activity (data not shown). Four
109
independent control transgenic tobacco lines expressing the CaMV35S-GUS construct displayed a high light-independent GUS activity in roots and leaves (data not shown) as expected from a constitutive promoter [16]. A comparison between the apparent light-dependent expression patterns of GUS and of the endogenous glycolate oxidase was carried out in one each of the GLO-GUS and CaMV35S-GUS expressing lines (Fig. 4). Glycolate oxidase and GUS activities, in cotyledons of GLO-GUS seedlings germinated and grown in complete darkness, were about 15^20% of the activities measured in light-grown seedlings. Within 24 h of illumination, cotyledons of dark-grown seedlings responded with a 2-fold increase in the activities of both enzymes (Fig. 4A). In contrast, the pattern of GUS activity in the CaMV35S-GUS transgenic plants was not correlated with that of endogenous glycolate oxidase and was not induced by light (Fig. 4B). Taken together, the results obtained from the transgenic plants suggest that the 3 kb 5P £anking region is functioning as a GLO promoter conferring similar expression patterns on the GUS gene as on the endogenous glycolate oxidase gene. It is suggested that this region contains the regulatory sequences of the GLO promoter. We would like to thank Gila Granot for excellent technical assistance. A.N. was supported by a Deichmann fellowship. This study was supported in part by the Israel Science Foundation founded by the Israel Academy of Sciences and Humanities.
References [1] G.H. Lorimer, Ann. Rev. Plant Physiol. 32 (1981) 349^ 383. [2] C.W. Husic, D.H. Husic, N.E. Tolbert, Crit. Rev. Plant Sci. 5 (1987) 45^100. [3] N.N. Artus, S.C. Somerville, C.R. Somerville, Crit. Rev. Plant Sci. 4 (1986) 121^147. [4] R. Boldt, S. Koshuchowa, W. Gross, T. Bo«rner, C. Schnarrenberger, Plant Sci. 124 (1997) 33^40. [5] C. Ludt, H. Kindl, Plant Physiol. 94 (1990) 1193^1198. [6] M. Esaka, N. Yamada, M. Kitabayashi, Y. Setoguchi, R. Tsugeki, M. Kondo, M. Nishimura, Plant Mol. Biol. 33 (1997) 141^155. [7] J. Feierabend, H. Beevers, Plant Physiol. 49 (1972) 28^32. [8] J.M. Greenler, W.M. Becker, Plant Physiol. 94 (1990) 1484^ 1487.
BBAEXP 91147 24-7-98
110
S. Barak et al. / Biochimica et Biophysica Acta 1399 (1998) 105^110
[9] C. Kuhlemeier, P. Green, N.-H. Chua, Ann. Rev. Plant Physiol. 38 (1987) 221^257. [10] M. Volokita, C.R. Somerville, J. Biol. Chem. 262 (1987) 15825^15828. [11] S.M. Mount, Nucleic Acids Res. 10 (1981) 168^181. [12] P.M. Gilmartin, L. Sarokin, J. Memelink, N.-H. Chua, Plant Cell 2 (1990) 369^378. [13] C.P. Joshi, Nucleic Acids Res. 15 (1987) 6643^6653. [14] R. Nakano, T. Matsumura, H. Sakakibara, T. Sugiyama, T. Hase, Plant Cell Physiol. 38 (1997) 1167^1170.
[15] S. Huang, Y.-Q. An, J.M. McDowell, E.C. McKinney, R.B. Meagher, Plant Mol. Biol. 33 (1997) 125^139. [16] R.A. Je¡erson, T.A. Kavanagh, M.W. Beavan, EMBO J. 6 (1987) 3901^3907. [17] T.A. Kavanagh, R.A. Je¡erson, M.W. Beavan, Mol. Gen. Genet. 215 (1988) 38^45. [18] A.L. Baker, N.E. Tolbert, Methods Enzymol. 9 (1966) 338^ 342.
BBAEXP 91147 24-7-98