Proc. Natl. Acad. Sci. USA Vol. 86, pp. 7919-7923, October 1989 Biology Transcriptional regulation and DNA methylation of nuclear for photosynthesis in nongreen cells* (Acer pseudoplatanus/suspension-cultured celis/nudear expression/HeLa cell lysate/in vitro ) JARUNYA NGERNPRASIRTSIRIt, HIRoKAzu KOBAYASHIt, AND TAKASHI AKAZAWAt§ tResearch Institute for Biochemical Regulation, School of Agriculture and *Radioisotope Research Center, Nagoya University, Chikusa, Nagoya 464-01, Japan Communicated by Jack R. Harlan, July 14, 1989

ABSTRACT The transcripts of nuclear genes for the small (rbcL) encoding the large subunit of RbuP2Case is little subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase expressed in the amyloplasts (8). However, the amyloplasts (rbcS), chlorophyll a/b-binding protein (cab), and extrinsic were competent to import and proteolytically process the 33-kDa protein involved in photosystem II water oxidation precursors ofpolypeptides synthesized in the cytoplasm such (woxA) were not detectable in the white wild cultured cells of as the small subunit of RbuP2Case encoded by the nuclear sycamore (Acerpseudoplatanus), in contrast to their high levels gene rbcS (9). Thus, it is imperative for us to examine in the sibling green mutant cells and the constitutive expression whether these genes are transcribed in the white wild cells; of genes (act) in both cell types. We have examined the if not, a pertinent question to be answered is what regulatory template activities of nuclear DNAs using the HeLa cell in vitro mechanisms may govern the overall processes. transcription system. All ofthe three photosynthesis genes from In the present investigation, we have attempted to study the green cell line and act from both cell types were well the rate-limiting step involved and have examined the tran- transcribed in vitro, but these photosynthesis genes from the scriptional template activities of nuclear genes from both the white cell line were not, indicating that the transcriptional white wild and green mutant cell lines ofsycamore. We report regulation is ascribable to DNA templates. Digestion of nuclear here experimental results supporting the view that the selec- DNA with methyl-sensitive and -insensitive isoschizomeriz en- tive methylation of nuclear DNA plays a causative role in the donucleases and the subsequent Southern hybridization transcriptional suppression of nuclear genes for photosyn- showed that each gene has the identical recognition sites of thesis in the nonphotosynthetic plant cells. restriction enzymes in the green and white cell lines, but some of the sites were methylated only in the photosynthesis genes in the white cells. There was observed a clear inverse relationship MATERIALS AND METHODS between the level of expressed transcripts and the extent of Cell Cultures. The white wild cell line of sycamore was DNA methylation. Thus, it is inferred that the selective meth- grown in a liquid medium by supplementation with sucrose ylation of DNA is a likely mechanism for suppressing tran- and 2,4-dichlorophenoxyacetic acid (2,4-D) as described by scription of nuclear genes for photosynthesis in the nonphoto- Bligny (10). The green mutant cell line of sycamore originally synthetic plant cells. isolated by Lescure (6) was grown without 2,4-D under continuous illumination with fluorescent light (2,000 lux) at Differentiation of various types of such as the 250C. At the exponential stage of growth, the cells were developmental formation of in greening seed- harvested and processed for the isolation ofnuclear DNA and lings and the transitional formation of from total cellular RNA. chloroplasts during tomato fruit ripening contains compli- Nuclear DNA and Total Cellular RNA Preparation. To cated processes. There exists interaction of nuclear and purify nuclear DNA, nuclei were isolated from protoplasts of genes, and one or more regulatory mechanisms op- the two cell lines, and the DNA was purified by CsCl density erating therein are not thoroughly understood, although it is gradient centrifugation as reported (11). Each batch of puri- well recognized that the coordinated expression ofthese two fied nuclear DNA was ascertained to be free from contami- types is required in the overall process. nation of plastid DNA by Southern blotting with a labeled The transcriptional, posttranscriptional (1, 2), and trans- rbcL gene probe. Total cellular RNA from both cell lines was lational (3, 4) controls as well as posttranslational regulation isolated by the reported method (7) in the presence of an (5) have been postulated as mechanisms governing the RNase inhibitorfrom placenta (Takara Shuzo, Kyoto, expression of photosynthesis genes during plastid differen- Japan). tiation, and along this line of research we have selected Characterization of RNA and DNA. Purified cellular RNA liquid-cultured cells of sycamore, plane maple (Acerpseudo- was denatured by glyoxal and electrophoresed in an agarose platanus) as an experimental system. The white wild cell line gel. The digestion of DNA with several restriction endonu- of sycamore grows heterotrophically, whereas the sibling cleases and subsequent agarose gel electrophoresis were green mutant cell line originally isolated from the former by performed by the conventional techniques as described (8). the mutagen treatment (6) can be cultured under complete Transfer of DNA and RNA to Zeta-Probe (Bio-Rad) mem- autotrophic condition and absolutely requires light illumina- branes and subsequent hybridization were as described in tion (7). We have previously reported that some polypeptides Bio-Rad's instruction manual(s). The radioactive bands on such as ribulose-1,5-bisphosphate carboxylase/oxygenase the membranes were detected by Fuji x-ray film RXO-G with [RbuP2Case; 3-phospho-D-glycerate carboxy-lyase (dimeriz- screen GRENEX and exposure at -80°C. ing), EC 4.1.1.39] cannot be detected in the white wild cells Fuji intensifying (7), and it was subsequently found that the chloroplastic gene Abbreviation: RbuP2Case, ribulose-1,5-bisphosphate carboxylase/ oxygenase. The publication costs of this article were defrayed in part by page charge *This is paper no. 82 in the series "Structure and Function of payment. This article must therefore be hereby marked "advertisement" Proteins." in accordance with 18 U.S.C. §1734 solely to indicate this fact. §To whom reprint requests should be addressed.

7919 Downloaded by guest on September 26, 2021 7920 : Ngernprasirtsiri et al. Proc. Natl. Acad. Sci. USA 86 (1989)

Plasmids bearing pea leaf cDNA clones for rbcS (pSS15) a b c d and cab, which encodes the chlorophyll a/b-binding protein * r ^ f--n ---) 1 2, r3 - 4 5 6 7 8 9 10 11 (pAB96) (12, 13), and for woxA (see ref. 14 for gene desig- S MW MW MW MW MW nation), which encodes the extrinsic 33-kDa protein involved in the photosystem II water oxidation (pWOX338) (15), were used throughout this investigation. The containing 18S rRNA-encoding DNA from the Alternaria alter- nata (pABM4; unpublished data), specifically hybridizable with higher plant 18S ribosomal DNAs, was also used. DNA probes were labeled with [a-32P]dCTP by using the Klenow fragment of Escherichia coli DNA polymerase I and primers 28S 23S in the oligolabeling kit of Pharmacia. Oligonucleotides com- plementary to conserved sequences ofdicotyledon actin gene 18S (act) corresponding to near N-terminal regions (16-18), 5'- 16S ACCATTCC(AT)GTTCCATTGTC(AG)CAAACAAG-3' in which parentheses show mixtures of two bases in these positions, was synthesized by a DNA synthesizer (Applied rbcS cab woxA 18S Biosystems model 381A); the oligonucleotide was purified by rDNA OPC Cartridge (Applied Biosystems) and was labeled at the FIG. 1. RNA-DNA hybridization analysis of total cellular RNAs 3' end by T4 polynucleotide kinase with [y-32P]ATP (222 isolated from green mutant (lanes M) and white wild (lanes W) cell TBq/mmol). lines of sycamore with cDNA probes. The agarose gel (1.2%) In Vitro Transcription by HeLa Cell Lysate. The HeLa cell electrophoretic patterns of purified total cellular RNA (20 jg each) lysate was supplied by BRL (19). The in vitro transcription from two cell lines are represented in lanes 2 and 3. (a-d) Autora- assay of nuclear DNA from both cell lines was carried out diograms of the RNA-DNA hybridization, with nuclear cDNA with the HeLa cell lysate at 30°C for 1 hr following the probes specified at the bottom of the lanes. The sizes of RNA size published procedure (20) except that RNase inhibitor from markers from HeLa cells (28S, 18S rRNA) and E. coli (23S, 16S human placenta was added. The resultant radioactive tran- rRNA) are indicated at the left (lane 1). scripts purified as described (8) were assayed to determine the incorporation of [a-32P]UTP into RNA. Ten microliters of hybridization with the membrane-blotted contain- the extract after phenol/chloroform treatment was spotted on ing DNA probes. a Whatman GF/C glass-fiber disc, which was washed five The addition of a-amanitin (0.5 pug/ml), an inhibitor of times with cold 5% (wt/vol) trichloroacetic acid, twice with RNA polymerase II (19), drastically inhibited the 32p incor- 95% ethanol, and finally dried. poration (99%), showing that RNA polymerase II entails the Size Analysis ofin Vitro Transcription Products. Nonlabeled transcriptional activity (Table 1). The degree ofincorporation transcripts were produced by the HeLa in vitro transcription was indistinguishable between nuclear DNAs derived from system as described above, except for the use of nonradio- the two sycamore cell lines. The cell line-specific factor(s) active UTP. To compare the size of transcripts, both in vitro inhibiting transcription appeared to be absent in DNA prep- transcripts and total cellular RNA were subjected to dena- arations, since the total incorporation of 32p was barely turation with glyoxal and subsequent electrophoresis in a altered when the mixture ofthe two preparations was used as 1.2% agarose gel, followed by RNADNA hybridization with template in the in vitro transcription system. labeled DNA probes as described (7). In the case of hybrid- The labeled transcripts for rbcS, cab, and woxA were ization with oligonucleotide-end-labeled probes, the blots obtained from the in vitro transcription system when the were hybridized as described in Bio-Rad's instruction manual nuclear DNA from green mutant was used as the template. In for 24 hr at 56°C (21). contrast, none of these products were detectable with the nuclear DNA of white wild cells (data not shown). It should be noted that the 18S rDNA was shown not to serve as a RESULTS template in the in vitro transcription system, consistent with Absence of Transcripts from Nuclear Genes for Photosyn- the previous report showing that rDNA can be transcribable thesis in White Wild Cell Line. We first examined the expres- only by RNA polymerase I (22). sion of nuclear genes for photosynthesis in both white wild Size Analysis of Transcripts Synthesized in Vitro. The sizes and green mutant cells of sycamore. The levels of transcripts of the transcripts produced in the in vitro transcription for rbcS, cab, woxA, and 18S rRNA genes in total cellular RNA preparations were determined by RNADNA hybrid- Table 1. Template activity of nuclear DNAs obtained from green ization analysis. The green mutant (M lanes 4, 6, 8, and 10) mutant and white wild cultured cells of sycamore in the HeLa in displayed clear positive signals with the DNA probes tested vitro transcription system (Fig. 1). In contrast, transcripts for rbcS, cab, and woxA were 32p incorporationt cpm not detectable in the white wild cells (W lanes 5, 7, and 9). The With hybridization with 18S rRNA gene probe showed that 18S Without a-amanitin rRNA was present in both the green mutant (lane 10) and white wild (lane 11) cells. DNA templates* a-amanitin (0.5 ug/ml) Template Activity of Nuclear DNA by HeLa in Vitro Tran- No DNA 4.71 x 10 4.84 x 10 scription System. To examine mechanisms of transcriptional SV40 DNAt 8.28 x 106 1.24 x i01 regulation, we determined the template activities of nuclear Green mutant nuclear DNA 2.67 x 106 2.40 x 104 DNAs from both cell lines using a heterologous in vitro White wild nuclear DNA 2.59 x 106 2.85 x 104 transcription system by HeLa cell lysate. In the assay, Pst Green mutant DNA I-cleaved simian 40 was used as a positive control as + white wild nuclear DNA 2.69 x 106 2.62 x 104 described by Handa et al. (20). The reaction was carried out *PurifiedDNA (1 Mg) was used in each assay. with [a-32P]UTP as substrate, and the resultant radioactive tTotal incorporation of 32p into trichloroacetic acid-insoluble frac- products were subjected to analysis for 32p incorporation into tions following the in vitro transcription assay was as described. the trichloroacetic acid-insoluble fraction and subsequent tPst I-cleaved SV40 DNA was used as a quality control. Downloaded by guest on September 26, 2021 Cell Biology: Ngernprasirtsiri et al. Proc. Natl. Acad. Sci. USA 86 (1989) 7921 scripts of the white wild cells (lanes 2 and 4). However, the a rbcS 1 2"3 4 presence oftranscripts ofact indicated that the housekeeping 18S - 16S - _ genes in white wild nuclear DNA are actively transcribed - 1.0 kb (Fig. 2d, lanes 2 and 4). Identical Restriction Sites in Nuclear DNAs from Two Cell b cab Lines and Methylated Sequences in Lowly Expressed Genes. 18S - To examine the identity of nucleotide sequences between the 16S- _ 1.2 kb nuclear DNAs of both cell lines and the distribution of methylated bases, we performed the Southern blot analysis of c woxA DNA fragments produced by three pairs of methyl-sensitive 18S _ and -insensitive isoschizomeric restriction enzymes-i.e., 16S- m 1.2 kb BstNI/EcoRII, Msp I/Hpa II, and Sau3AI/Mbo I. It is known that BstNI can cleave the 5'-CCA/TGG-3' sequences d act when the internal cytosine residue is 5-methylcytosine, while 18S V-- - 1.7 kb EcoRII cannot. Msp I and Hpa II can cleave the 5'-CCGG-3' 16S sequences containing a methylated cytosine residue at dif- ferent positions. In another pair of isoschizomers, Sau3AI EM W M W. can cleave the 5'-GATC-3' sequences containing N6- methyladenine, but Mbo I cannot (25). The hybridization data presented in Fig. 3 clearly show multiple signals with respect to each individual cDNA probe FIG. 2. Size analysis of in vitro transcripts and total cellular tested, indicating the presence of multigene families as re- RNAs for rbcS (a), cab (b), woxA (c), and act (d). RNA-DNA ported (26). The digestion of DNAs with the methyl- hybridization of cellular RNAs (0.5 pug) from green mutant (M lanes insensitive enzymes exhibited the completely identical pro- 1) and white wild (W lanes 2) cells and nonradioactive in vitro files between the two cell lines. Sizes of DNA fragments transcripts generated by using 4 Aug each ofnuclear DNAs from green mutant (M lanes 3) and white wild (W lanes 4) cells as templates in produced by the pairs of methyl-sensitive and -insensitive the HeLa in vitro transcription system. Autoradiograms of hybrid- isoschizomers hybridizable with each specific gene probe ization with labeled DNA probes are shown. Migrating positions of were identical in the green mutant, showing the absence of RNA size markers, 18S rRNA from HeLa cells and 16S rRNA from methylated bases in the recognition sequence. In contrast, E. coli, are shown at the left. The approximate sizes oftranscripts are the profiles were different from that of the nuclear DNA of indicated at the right. white wild cell line, showing the presence of 5-methylcy- tosine or N6-methyladenine in the recognition sites in rbcS, system were compared with those in the total cellular RNA cab, and woxA regions (Fig. 3) but not in the act region (data as shown in Fig. 2. The approximate sizes of transcripts for not shown). It must be emphasized that the unique fragments rbcS (Fig. 2a), cab (Fig. 2b), woxA (Fig. 2c), and act (Fig. 2d) produced by the methyl-sensitive enzymes persisted regard- were 1.0, 1.2, 1.2, and 1.7 kilobases (kb), respectively. The less of the extended time of digestion and the amount of sizes of transcripts for rbcS, cab, and act were nearly enzymes; absence of any inhibitors for the restriction en- identical to those reported in several other plant species (23, zymes was ensured by the codigestion of plasmid pBR322 24). Furthermore, they are identical between the total cellular and the subsequent hybridization experiments (data not RNA (lanes 1 and 2) and the in vitro transcription products shown). using the nuclear DNA derived from both cell lines (lanes 3 Methylated Base Compositions of Nuclear DNA. The puri- and 4). The transcripts of rbcS, cab, and woxA were not fied nuclear DNAs from both cell lines were completely detectable either in the cellular RNA or the in vitro tran- hydrolyzed and subjected to base composition analysis by

a rbcS b cab c woxA

1 2 3 4 5 6 78 931112 1 2 3 4 5 6 78 931112 1 2 34 5 6 7 8910ti12 kbp kbp kbp

23.13 - - 23.13_- 23.13_- 9.42 - 9.42 _- 9.42,- -, --m -so 6.56-- a-. mOm_ 3533?C:-- W 6.56 - mm omgm A 6.56- NOan _n MMA - - m obbb mm_ -~ mw~mm m~ 4.37.- _ mm emm 4.37.- 4.37. 'U -U..o min --. - m C = 2.32- mftqm --Mw n 2.32.- m_xm 2.32.- __,_ - S;; 2.03 - 2.03-- 2.03'.- awg m OMI _ xam _0 _M4

0.56 - 0.56.- 0.56.- 0.13- 0.13- 0.13-

B E 8 E MsH NiMSHLS R B E B EWH MsHS WSIL B E B E MsH MsH S MS Mb M W M W MW M W M W M W M W M W M W

FIG. 3. Methylation of nuclear DNA regions containing photosynthesis genes. Nuclear DNAs (1 ,g) from green mutant (lanes M) and white wild (lanes W) cell lines were digested with three pairs of isoschizomeric endonucleases: BstNI (B lanes 1 and 3)/EcoRlI (E lanes 2 and 4); Msp I (Ms lanes 5 and 7)/Hpa II (H lanes 6 and 8); and Sau3AI (S lanes 9 and 11)/Mbo I (Mb lanes 10 and 12). DNA fragments electrophoresed in agarose gel (0.7%) were transferred to Zeta-Probe membranes and subjected to the Southern hybridization with the DNA probes as shown. Different specific cleavage sites observed only in the white wild nuclear DNA are marked by white circles. The sizes of the DNA fragments are shown at the left in kilobase pairs. Downloaded by guest on September 26, 2021 7922 Cell Biology: Ngernprasirtsiri et al. Proc. Natl. Acad. Sci. USA 86 (1989) reverse-phase HPLC (8). The mol % ofbases were calculated DNAs (27). However, so far there is little information avail- by standardization of each peak area with respect to the able concerning possible functions of this modified base. standard bases (Table 2). A variety of methylated bases were Recently, it has been reported that the methylation of genes detected, and the mol % of methylated bases were indistin- for tissue-specific storage proteins in maize correlates with guishable between both cell lines. The base compositions of their reduced transcriptional activities (39). Studies on the salmon sperm DNA in the analysis always agree with the T-DNA of crown gall tumors (40) and the ribosomal genes published data, and the G+C content of sycamore nuclear (41) also showed that there is an inverse relationship between DNA was found to be in a range of ordinary values in higher the magnitude ofmethylation and the transcriptional activity. (27). Among several methylated bases, 5-methylcy- We have reported that DNA methylation serves as a mech- tosine was shown to be most abundant in the DNAs from both anism for suppressing the transcription of photosynthetic cell lines, accounting for 25% of the total cytosine residues. genes in nongreen plastids-i.e., amyloplasts in white wild cultured cells of sycamore (7, 8) and chromoplasts in ripened red tomato fruits (42). Furthermore, a role of DNA methyl- DISCUSSION ation has been found in the differential expression of C4 It is absolutely necessary to use the nuclear in vitro tran- photosynthesis genes in mesophyll and bundle sheath cells of scription system to examine whether or not a regulatory greening maize leaves (43). factor is associated with the DNA template. A reliable in vitro Bearing in mind the limitations of the heterologous in vitro transcription system of plant origin is not readily available, transcription assay system used in the present investigation, although wheat germ is reported to be a useful material (28, we hypothesize from our experimental results that the direct 29). In the present study, we have used a heterologous in vitro interaction between the region ofnuclear DNA containing the transcription system containing HeLa cell lysate. Transcrip- selectively methylated bases and either RNA polymerase or tion of the maize zein gene by the HeLa cell lysate has been common transcription-regulating factor(s) is involved in the reported (30). As presented in Fig. 2, the lysate was able to specific transcriptional suppression of photosynthesis genes transcribe effectively four kinds ofplant genes. Interestingly, in the white wild sycamore cells. To our knowledge, there has termination, polyadenylylation, and correct splicing ap- not been reported any evidence showing the direct interac- peared to proceed as ascertained from the determined tran- tion between methylated plant DNA and RNA polymerase II, script sizes. Although it has been reported that the HeLa cell whereas there are some contradictory observations in lysate entails the activities of termination (31), polyadenylyl- cells. It has been reported that methylated simian virus 40 and ation (32), and splicing (33) of animal or plant genes, these adenovirus 2 DNA can be normally transcribed by HeLa cell reactions had not been examined in a single reaction system. lysate (44, 45), whereas the methylated DNA microinjected Thus, this report shows the most satisfactory use ofthe HeLa into the frog oocytes was poorly transcribed (44, 46). The cell lysate in plant gene transcription and subsequent RNA interaction between DNA and nuclear proteins but not RNA processing. polymerase II is postulated to be affected by DNA methyl- The postreplicative methylation of DNA has been postu- ation in vertebrates (47). Attempts to determine the specific lated to entail many important biological functions including sites of methylated sequences around the promoter region of its regulatory role in the gene expression in both prokaryotic nuclear DNA interacting with RNA polymerase II or other and eukaryotic cells (34). Although it has been proposed that factors in the white wild cells of sycamore stand a good the transcriptional activities of certain vertebrate genes are chance of clarifying the complicated regulatory process of closely correlated with the reduced levels of methylation gene expression in plant cells in general at the molecular (34-36), detailed molecular mechanisms have not yet been level. fully understood (37). Various patterns of methylation of nuclear DNA have been reported in plant cells (38), and it has We are much indebted to A. Watanabe, N.-H. Chua, and T. Tsuge indeed been known for a long time that plant DNAs contain for providing us cDNA and gene clones. This study was supported in part by Grants-in-Aid from the Ministry ofEducation, Science and abundant 5-methylcytosine compared to the mammalian Culture (Monbusho) of Japan and by a grant of the Nissan Science Foundation (Tokyo). J.N. is a recipient of the predoctoral student Table 2. Base compositions of nuclear DNAs obtained from fellowship provided by the Hitachi Scholarship Foundation (Tokyo). green mutant and white wild cultured cells of sycamore Base composition, mol % 1. Mullet, J. E. & Klein, R. R. (1987) EMBO J. 6, 1571-1579. 2. Gruissem, W., Barkan, A., Deng, X. W. & Stern, D. (1988) Salmon Green mutant White wild Trends Genet. 4, 258-263. Base sperm cells cells 3. Berry, J. O., Nikolau, B. J., Carr, J. P. & Klessig, D. F. (1985) Cytosine 18.01 13.29 13.16 Mol. Cell Biol. 5, 2238-2246. 3-Methylcytosine 0.56 1.42 1.44 4. Inamine, G., Nash, B., Weissbach, H. & Brot, N. (1985) Proc. 5-Methylcytosine 1.68 5.65 5.79 Natl. Acad. Sci. USA 82, 5690-5694. 5. Kobayashi, H., Bogorad, L. & Miles, C. D. (1987) Plant Guanine 21.59 18.26 18.17 Physiol. 85, 757-767. 1-Methylguanine ND ND ND 6. Lescure, A.-M. (1969) Physiol. Veg. 7, 237-250. 3-Methylguanine ND ND ND 7. Ngernprasirtsiri, J., Macherel, D., Kobayashi, H. & Akazawa, 7-Methylguanine 0.11 0.24 0.27 T. (1988) Plant Physiol. 86, 137-142. N2-Methylguanine ND 1.01 1.05 8. Ngernprasirtsiri, J., Kobayashi, H. & Akazawa, T. (1988) Proc. Adenine 28.74 29.68 29.11 Natl. Acad. Sci. USA 85, 4750-4754. 1-Methyladenine ND ND ND 9. Strzalka, K., Ngernprasirtsiri, J., Watanabe, A. & Akazawa, T. N6-Methyladenine ND ND 0.49 (1987) Biochem. Biophys. Res. Commun. 149, 799-806. 29.31 30.45 30.52 10. Bligny, R. (1977) Plant Physiol. 59, 502-505. Thymine 11. Ohyama, K., Pelcher, L. E. & Horn, D. (1977) Plant Physiol. Total 100.00 100.00 100.00 60, 179-181. The mol % values were calculated by standardization of each 12. Broglie, R., Bellemare, G., Bartlett, S. G., Chua, N.-H. & elution peak area (A254) relative to that of a standard mixture offour Cashmore, A. (1981) Proc. Natl. Acad. Sci. USA 78, 7304- major bases and eight methylated bases (see figure 3 in ref. 8). Acid 7308. hydrolysates of DNA (5 ug) were subjected to analysis, and each 13. Coruzzi, G., Broglie, R., Cashmore, A. & Chau, N.-H. (1983) value represents an average of three separate determinations. ND, J. Biol. Chem. 258, 1399-1402. not detectable (<0.02%). 14. Kuwabara, T., Reddy, K. J. & Sherman, L. A. (1987) Proc. Downloaded by guest on September 26, 2021 Cell Biology: Ngernprasirtsiri et al. Proc. Natl. Acad. Sci. USA 86 (1989) 7923 NatI. Acad. Sci. USA 84, 8230-8234. 30. Boston, R. S. & Larkins, B. A. (1986) Plant Mol. Biol. 7, 15. Watanabe, A., Minami, E., Murase, M., Shinohara, K., Ku- 71-79. wabara, T. & Murata, N. (1987) in Progress in Photosynthetic 31. Resnekov, 0. & Aloni, Y. (1989) Proc. Nail. Acad. Sci. USA Research, ed. Biggins, J. (Nijhoff, Dordrecht, The Nether- 86, 12-16. lands), Vol. 4, pp. 629-636. 32. Manley, J. J. (1984) in Transcription and : A Prac- 16. Narin, C. J., Winesett, L. & Ferl, R. J. (1988) Gene 65, tical Approach, eds. Hames, B. D. & Higgins, S. J. (IRL, 247-257. Oxford, U.K./Washington, DC), pp. 71-88. 17. Shah, D. M., Hightower, R. C. & Meagher, R. B. (1983) J. 33. Brown, J. W. S., Feix, G. & Frendewey, D. (1986) EMBO J. 5, Mol. Appl. Genet. 2, 111-126. 2749-2758. 18. Drouin, G. & Dover, G. A. (1987) Nature (London) 328, 34. Doerfler, W. (1983) Annu. Rev. Biochem. 52, 93-124. 557-558. 35. Razin, A. & Cedar, H. (1984) Int. Rev. Cytol. 92, 159-185. 19. Manley, J. L., Fire, A., Cano, A., Sharp, P. A. & Gefter, 36. Cedar, H. (1988) Cell 53, 3-4. M. L. (1980) Proc. Natl. Acad. Sci. USA 77, 3855-3859. 37. Bird, A. P. (1984) Nature (London) 307, 503-504. 20. Handa, H., Kaufman, R. J., Manley, J., Gefter, M. & Sharp, 38. Gruenbaum, Y., Many, T. N., Cedar, H. & Razin, A. (1981) P. A. (1981) J. Biol. Chem. 256, 478-482. Nature (London) 292, 860-862. 21. Davis, L. G., Dibner, M. D. & Battey, J. F. (1986) in Basic 11, Methods in Molecular Biology (Elsevier, London), pp. 75-78. 39. Bianchi, M. W. & Viotti, A. (1988) Plant Mol. Biol. 22. Wollgiehn, R. (1982) in Nucleic Acids and Proteins in Plant, 203-214. eds. Parthier, B. & Boulter, D. (Springer, Berlin/Heidelberg/ 40. Hepburn, A. G., Clarke, L. E., Pearson, L. & White, J. (1983) New York), Vol. 2, pp. 125-170. J. Mol. Biol. 2, 315-329. 23. Piechulla, B., Pichersky, E., Cashmore, A. R. & Gruissem, W. 41. Watson, J. C., Kaufman, L. S. & Thompson, W. F. (1987) J. (1986) Plant Mol. Biol. 7, 367-376. Mol. Biol. 193, 15-26. 24. Hightower, R. C. & Meagher, R. B. (1985) EMBO J. 4, 1-8. 42. Ngernprasirtsiri, J., Kobayashi, H. & Akazawa, T. (1988) Plant 25. McCelland, M. & Nelson, M. (1985) Nucleic Acids Res., Suppl. Physiol. 88, 16-20. 13, r201-r207. 43. Ngernprasirtsiri, J., Chollet, R., Kobayashi, H., Sugiyama, T. 26. Pichersky, E., Bernatzky, R., Tanksley, S. D. & Cashmore, & Akazawa, T. (1989) J. Biol. Chem. 264, 8241-8248. A. R. (1986) Proc. Natl. Acad. Sci. USA 83, 3880-3884. 44. Fradin, A., Manley, J. L. & Prives, C. L. (1982) Proc. Natl. 27. Shapiro, H. S. (1976) in Handbook ofBiochemistry and Mo- Acad. Sci. USA 79, 5142-5146. lecular Biology: Nucleic Acids, ed. Fasman, G. D. (CRC, 45. Jove, R., Sperber, D. E. & Manley, J. M. (1984) Nucleic Acids Cleveland), 3rd Ed., Vol. 2, p. 259. Res. 12, 4715-4730. 28. Yamazaki, K. & Imamoto, F. (1987) Mol. Gen. Genet. 209, 46. Vardimon, L., Kressmann, A., Cedar, H., Maechler, M. & 445-452. Doerfler, W. (1982) Proc. Natl. Acad. Sci. USA 79, 1073-1077. 29. Job, D., Marmillot, P., Job, C. & Jovin, T. M. (1988) Biochem- 47. Keshet, I., Lieman-Hurwitz, J. & Cedar, H. (1986) Cell 44, istry 27, 6371-6378. 535-543. Downloaded by guest on September 26, 2021