The Plant Cell, Vol. 1, 105-1 14, January 1989, @ 1989 American Society of Plant Physiologists

The opaque-2 Mutation of Maize Differentially Reduces Zein Gene Transcription Robert Kodrzycki," Rebecca S. Boston,b and Brian A. Larkins".'

a Department of Botany and Plant Pathology, Purdue University, West Lafayette, Indiana 47907 Department of Botany, North Carolina State University, Raleigh, North Carolina 27695 Department of Plant Sciences, University of Arizona, Tucson, Arizona 85721

Zeins, the storage proteins of maize seed, are encoded by a large multigene family that is regulated developmentally and expressed in a tissue-specific manner during endosperm development. The synthesis of these proteins is affected by mutations, such as opaque-2, that cause a reduction in the accumulation of zein proteins and mRNAs. We used nuclear run-on transcription assays to analyze the expression of zein genes in developing normal and opaque-2 endosperms and to map the origin of these transcripts with respect to the coding and noncoding regions of the genes. These analyses demonstrate that zein gene expression is regulated transcriptionally and posttran- scriptionally in developing endosperm. Transcription of genes encoding a-zeins is inhibited significantly in opaque- 2 mutants, with expression of those encoding the M, 22,000 proteins being almost totally blocked. Other gene subfamilies were affected but to a lesser extent.

INTRODUCTION

The storage proteins of maize seed are a group of alcohol- soluble polypeptides that are synthesized in developing maize endosperm between 10 and 45 d after pollination (DAP). These proteins, called zeins, can be divided into four structurally distinct types. a-Zeins, which are generally the most abundant, are proteins of M, 19,000 and 22,000 that are encoded by a large multigene family (Hagen and Rubenstein, 1981). The P-zein, M, 14,000 (Pedersen et al., 1986), y-zeins, M, 27,000 and 16,000 (Prat et al., 1987), and 6-zein, M, 10,000 (Kirihara et al., 1988), are sulfur-rich proteins that are encoded by genes present in only one or two copies. The expression of zein genes is coordinately regulated and zein mRNAs accumulate to high concentra- tions during early stages of endosperm development (Marks et al., 1985a; Boston et al., 1986). Typically, zeins account for approximately 50% of the total endosperm protein at maturity.

Severa1 mutations have been identified that cause a reduction in zein synthesis. One of these, opaque-2, causes about a 50% reduction in zein proteins compared with the normal genotype (Tsai et al., 1978). This mutation affects primarily the synthesis of the a-zeins; there is a significant reduction of M, 19,000 polypeptides, and the M, 22,000 polypeptides are synthesized in only trace amounts (Jones et al., 1977). Another mutation, opaque- 7, also causes a reduction in a-zein synthesis, but the reduction is greater in the M, 19,000 proteins (DiFonzo et al., 1980). The floury-2 mutation causes a reduction of

' To whom correspondence should be addressed.

both the M, 19,000 and 22,000 a-zeins in proportional amounts (Jones, 1978).

The mechanisms by which these mutations affect zein synthesis are unknown. Previous studies have shown that the reduction of zein proteins is correlated with reduced amounts of the corresponding mRNAs (Pedersen et al., 1980; Soave et al., 1981; Burr and Burr, 1982; Marks et al., 1985a; Boston et al., 1986), suggesting that these mutations alter transcription. Furthermore, since the a- zein genes are located on different chromosomes (Soave and Salamini, 1984), it is likely that trans-acting regulatory factors are affected by these mutations.

To characterize better the regulation of zein gene expression, we used nuclear run-on assays to investigate transcription during endosperm development. These stud- ies show that genes encoding zein proteins are regulated transcriptionally. In general, there is good agreement be- tween the rate of transcription of various zein genes and the leve1 of the corresponding polysomal mRNAs; how- ever, some genes appear to be transcribed more actively than others.

RESULTS

Accumulation of Zein Proteins in Normal and opaque-2 Endosperms

Zeins are synthesized in developing normal maize endo- sperm between 1 O and 45 DAP. At 12 DAP, the M, 19,000

106 The Plant Cell

and 22,000 a-zeins and M, 27,000 7-zein were detectedin SDS-PAGE separations of total zein proteins as shownin Figure 1A. The other zeins were barely detectable atthis stage but were apparent by 14 DAP. a-Zeins werepresent in the highest concentration, with the M, 19,000polypeptides the most abundant. The staining intensity ofthe M, 27,000 protein was similar to that of the M, 22,000zeins. The M, 14,000 (3-zein and the MT16,000 7-zein weresynthesized coordinately, but the amount of Mr 14,000protein was greater. The M, 10,000 6-zein was present inthe lowest amount.

As illustrated in Figure 1B, synthesis of zein proteins inthe opaque-2 mutant was delayed and reduced relative tonormal genotype. The M, 27,000 7-zein was detected by12 DAP and represented the most abundant speciesthroughout development. The a-zeins, which predominatein the normal genotype, were reduced significantly. The M,19,000 proteins were not detected until 14 DAP and theM, 22,000 zeins were found in only trace amounts. Onlysmall amounts of /3- and 6-zeins were synthesized; theamount of the M, 16,000 7-zein, relative to the M, 14,000(3-zein, was higher in the mutant compared with the normalgenotype.

Analysis of Zein Gene Transcription

Previous studies showed that zein mRNAs are detectedin maize endosperm tissue (Kriz et al., 1987). To investi-gate whether or not zein gene expression is regulatedtranscriptionally, we compared nuclear run-on transcriptionassays from endosperm, leaf, root, and shoot tissue. Whenadded to an in vitro run-on transcription reaction, nucleifrom these tissues incorporated radioactively labeled nu-cleotides for 15 to 20 min with kinetics similar to thosedetermined in a previous study (Beach et al., 1985). Asillustrated in Tables 1 and 2, transcripts corresponding toa- and 7-zeins were synthesized in endosperm nuclei at14 and 20 DAP but were not detected in leaf nuclei ornuclei from shoot or root tissue (data not shown). Thepresence of transcripts corresponding to chlorophyll a/bbinding protein (Cab) and the small subunit of ribulosebisphosphate carboxylase/oxygenase (Rbc S) in leaf butnot endosperm tissue demonstrated that the inability todetect zein transcripts from leaf nuclei was not due toRNA degradation. Transcription of both zein- and leaf-specific genes was sensitive to the addition of 8 to 10 ̂ g/ml a-amanitin, indicating that these genes were transcribedby RNA polymerase II. Transcripts corresponding to rRNAwere detected in all four tissues and their synthesis wasinsensitive to this concentration of a-amanitin.

Zein Gene Expression Is Regulated Transcriptionally inDeveloping Maize Endosperm

We investigated the transcription of zein RNAs in endo-sperm tissue by comparing the synthesis of transcripts at

10 12"DAP

14 16 18 20

27.000

12 14 16 18 20 28

10,OOO

27,000

22.00019.000

16.OOO14,000

10,000

Figure 1. Developmental Regulation and Tissue Specificity of ZeinGene Expression.

SDS-PAGE of zeins extracted with 70% EtOH, 2% 2-mercapto-ethanol from (A) normal and (B) opaque-2 developing endo-sperms. Each developmental stage is listed across the top forboth genotypes, and the apparent molecular weights are givenalong the side. Each sample represents the amount of zeinextracted from 1 ̂ g of normal or 2 ̂ g of opaque-2 kernels.

different stages of development. Nuclei were isolated fromdeveloping normal seeds between 10 and 20 DAP. Yieldsof nuclei at later stages of development were insufficientfor in vitro transcription. The incorporation of radioactivityinto RNAs corresponding to a-, /3-, 7-, and 5-zeins, sucrosesynthase (SS), UDP-glucosyl transferase (Wx), Cab, andRbc S is shown in Table 3, and these data are expressedas hybridized counts per minute per microgram of DNA inFigure 2. Because of variation in incorporation of radionu-cleotides between experiments, the data shown were nor-malized to total counts per minute per hybridization andamount of DNA per reaction.

The analysis presented in Figure 2 shows that zein genetranscription is regulated developmentally. Transcriptswere detected by 10 DAP and their amounts increaseduntil 16 DAP. Subsequently, there was an apparent de-crease in transcription, but we were unable to examinestages older than 20 DAP. The levels of «-zein transcripts(19ab, 19c, 19d, and 22) generally were higher than thosefor the genes encoding /?-, 7-, and 5-zeins, which is con-sistent with the larger numbers of genes encoding theseproteins. The onset of SS transcription was similar to thatof the zein genes and its expression slightly exceeded that

Gene Transcription in Maize Endosperm 107

Table 1. Gene Sequences Used as Probes

Coding Sequences DNA lnsert Size Source

Zein Probes 10 kD 600 Kirihara et al., 1988 14 kD 91 5 Pedersen et al., 1986 19 kD, aba 1010 Kriz et al., 1987 19 kD, c 880 Marks et al., 1985b 19 kD, d 850 Marks et al., 1985b 22 kD, b 890 Marks and Larkins 1982 27 kD 920 Wang and Essen, 1986

Sucrose Synthase (SS) 1300 Werr et al., 1985 UDP-Glucosyltransferase (WX)~ 21 O0 Wessler and Marguerite, 1985 Ribosomal DNA (rDNA) 9700 Messing et al., 1984 Chlorophyl a/b Binding Protein (Cab) 1200 Martineau and Taylor, 1985 Small Subunit of Rubisco (Rbc Sp

From waxy locus of maize. Pstl subclone of the 3’ terminus of cDNA clone. Rubisco = ribulose-l,5-bisphosphate carboxylase/oxygenase.

Non-Zein Probes

600 Nelson et al., 1984

a SnaBI/Xbal subclone of coding region, see Figure 6.

of Wx. The leve1 of transcription for both of these genes, which are associated with starch synthesis was similar to that of 8-zein.

Transcription of cY-Zein Genes 1s Delayed and Reduced by the opaque-2 Mutation

The opaque-2 mutation causes about a 50% reduction in the amount of a-zeins (Tsai et al., 1978); the M, 19,000 proteins are reduced greatly and the M, 22,000 proteins are barely detectable (Figure 16). This reduction in protein synthesis is associated with a decrease in the levels of the corresponding mRNAs (Marks et al., 1985a; Boston et al., 1986). To investigate the extent to which the reduction of a-zein mRNAs is regulated transcriptionally, we performed nuclear run-on transcription assays with nuclei from de- veloping opaque-2 endosperms.

The incorporation of radioactivity into RNAs correspond- ing to the various zein genes, SS, and Wx is presented in Table 4 and shown graphically in Figure 2. The synthesis of zein transcripts in opaque-2 nuclei showed a develop- mental pattern similar to normal; however, transcription was delayed by severa1 days. The rate of transcription of a- and y-zein genes increased substantially between 12 and 18 DAP, and transcription rates of the p- and 6-zeins increased slightly during this time. The most notable dif- ference between opaque-2 and normal nuclei was the reduced rate of transcription of a-zeins. The levels of transcripts from the y- and 6-zeins were similar to that of the normal genotype, but there was a decrease in the transcription of the p-zein.

A comparison of the rates of zein gene transcription normalized to that of the SS gene in normal and opaque-

2 nuclei is illustrated in Figure 3. In the normal genotype, this normalization shifted in the pattern of expression such that the peak of transcriptional activity was at 18 DAP rather than 16 DAP for all gene subfamilies except 19ab, where the peak was at 12 DAP. Normalizing zein gene transcription to that of SS in opaque-2 resulted in a similar developmental pattern of expression. This analysis dem- onstrates clearly that the rate of transcription of genes encoding M, 22,000 a-zeins is reduced by more than 90% in opaque-2 nuclei. Transcription of the M, 19,000 a-zeins is reduced by 40 and 75%, depending on the gene subfam- ily. The rate of transcription of p- and 6-zeins is not affected greatly and there is little change in the rate of y-zein transcription.

Table 2. Tissue-Specific Transcription of Maize Zein Genes

Leaf (days after germination)

Endosperm (days after pollination)

Probe 14 20 18 18” 14 14”

cpm in coding strand - cpm in noncoding strandb

19 kD,ab 247 53 346 25 3 6 27 kD 155 54 210 6 3 3 Cab 8 4 - - 154 5 Rbc S 4 3 - - 17 2 rDNA 5017 2323 2517 2877 1660 1626

a Transcription includes 8 pg/ml a-amanitin.

strand ranaed between 1 O and 40 cDm. Averaged for three hybridizations. Hybridization to noncoding

108 The Plant Cell

Table 3. lncorporation from Run-On Transcription of Normal W64A Endosperm Nuclei

Days After Pollination

Probe 10 12 14 16 18 20

cpm in coding strand - cpm in noncoding strand" 10 kD 37 173 77 144 75 7 14 kD 66 146 87 175 112 51 19 kD,ab 1 o1 1258 51 3 896 346 111 19 kD,c 79 1820 621 1284 853 128 19 kD,d 15 506 165 421 320 52 22 kD 62 1990 936 2401 1324 161 27 kD 159 360 155 435 21 o 54 ss 24 153 72 181 55 25 wx 14 133 16 52 31 26

4 Cab 3 Rbc S

cpm/Hyb" 1.2 x 106 5.2 X 106 1.9 x 106 1.8 X 106 1.7 x 10' 8.1 x 105 DNA, figb 75 160 176 131 139 123

a Averaged for three hybridizations. Hybridization to noncoding strand ranged between 1 O and 40 cpm.

- - - - 8 4 - - - -

Total microqrams of DNA Der run-on transcription.

Transcriptional Activity of Genes Varies between Zein Classes

a-Zeins are encoded by a large multigene family. To assess the transcriptional activity for each gene, we compared the transcription rate per microgram of DNA (Figure 2) with the predicted gene copy number for each zein class. This analysis, shown in Table 5, revealed that the transcriptional activity per gene was highest consistently for that encoding the M, 27,000 y-zein protein. The high copy number a- zein genes (19ab, 19c, 22) showed the lowest transcrip- tional activity per gene. The 19d group of a-zeins, which are encoded by only two genes, was second highest in transcriptional activity. The genes encoding the M, 10,000 and 14,000 zeins had transcriptional activities that were between the 19d subclass and the high copy number a- zein genes.

Comparison of Nuclear and Polysomal Levels of Zein Transcripts lndicates Posttranscriptional Regulation

The relative rates of transcription determined by nuclear run-on assays with normal nuclei did not correspond to the intensity of protein staining observed by SDS-PAGE. The relative amount of M, 19,000 a-zein proteins was much greater than that of M, 22,000 a-zeins (Figure 1A). However, nuclear run-on transcripts corresponding to the M, 22,000 zeins were equal roughly to the sum of those from the M, 19,000 zein genes. This implies that posttran- scriptional mechanisms may affect the levels of these proteins.

0 14 DAP

c

.- c ~ ta 18 DAP

V 10 14 19ab 19c 19d 22 27 SS W x

10 14 19ab 19c 19d 22 27 SS W x

Figure 2. Leve1 of Zein Transcripts in lsolated Nuclei from Devel- oping Maize Endosperm.

Nuclei were isolated from developing normal and opaque-2 maize kernels between 1 O and 20 DAP and added to a run-on transcrip- tion reaction as described in "Methods." The amount of transcripts for zein and other endosperm-specific genes is expressed as counts per minute hybridizing to the coding strand minus counts per minute hybridizing to the noncoding strand/total counts per minute in the hybridization reaction. This value was normalized to the amount of DNA in the assay. 10, 14,19ab, 19c, 19d, 22, and 27 represent the corresponding zein genes; SS, sucrose syn- thase; Wx, UDP-glucosyltransferase.

Gene Transcription in Maize Endosperm 1 O9

~

Table 4. lncorporation from Run-On Transcription of opaque-2 W64A Endosperm Nuclei

Days ARer Pollination

Probe 12 14 16 18

10 kD 14 kD 19 kD,ab 19 kD,c 19 kD,d 22 kD 27 kD ss wx Cab Rbc S cpm/Hyb" DNA. uab

cpm in coding strand - cpm in noncoding strand" 11 139 49 76 34 27 53 75 59 75 299 480 98 114 339 562 29 34 51 77 12 10 19 56

286 125 179 301 162 125 179 301

- - - 3 2 - - -

3.0 X 106 3.1 X 106 2.4 X 106 1.5 X 106 170 177 125 120

a Averaged for three hybridizations. Hybridization to noncoding strand ranged between 1 O and 40 cpm. Total micrograms of DNA per run-on transcription.

To characterize further the relationship between rates of transcription and steady-state levels of mRNAs, we determined the concentrations of polysomal mRNAs cor- responding to various zein clones. Poly(A)+ mRNAs from total polysomes isolated at 12 and 18 DAP were used to synthesize radioactive cDNAs, and these were hybridized subsequently to the zein clones. As illustrated in Figure 4, with the exception of genes encoding the M, 22,000 a-, p-, and y-zeins, the fractional contribution of steady-state mRNAs and nuclear run-on transcripts was similar. The fractional contribution of nuclear run-on transcripts of the M, 22,000 a-zeins was more than double that of the steady-state polysomal mRNAs. For the genes encoding the p- and y-zein, the proportion of steady-state mRNAs was double that of nuclear run-on transcripts.

Transcripts for an o-Zein Subfamily Do not Map to Far Upstream Noncoding Regions

Previous studies of gene transcription suggested that zein RNAs can originate from 5'-flanking regions that are nearly 1 kb upstream of the coding sequence (Langridge et al., 1982a, 1982b; Langridge and Feix, 1983; Brown et al., 1986). To determine the prevalence of these large RNAs and map their origin relative to the coding region, we prepared subclones of the 5'-flanking and coding regions of a gene encoding an M, 19,000 a-zein. As shown in Figure 5, these subclones encompass the following re- gions: (A) 2160 base pairs (bp) of the 5' noncoding region from -2388 to -228, relative to the initiation codon; (6)

1 O 1 2 nucleotides extending from -228 through the coding region with 81 nucleotides with the 3' noncoding se- quence: (C) 429 nucleotides extending from -97 through the first half of the coding region; and (D) 452 nucleotides extending from the middle of the coding region through 81 nucleotides of the 3' noncoding sequence.

The data in Table 6 show that each of these subclones hybridizes to nuclear run-on transcripts isolated from 14 a 20 DAP nuclei. At both stages hybridization to the region including the coding sequence and the first 228 nucleotides of 5' noncoding sequence was 20-fold greater than to the 21 60 nucleotides preceding this region. The amount of radioactivity hybridizing to the 3' end of B (probe D) was approximately twice that hybridizing to the 5' end of B (probe C). At both developmental stages the sum of the counts hybridizing to probes C and D was nearly equal to the counts per minute hybridizing to probe 6. Similar results were obtained regardless of whether or not the RNase treatment was included following the hybridization reaction .

DlSCUSSlON

Previous studies of the legume embryos (Beach et al., 1985; Walling et al., 1986) have shown that genes encod- ing storage proteins are regulated developmentally and

Wild type

li

C O 10 14 19ab 19c 19d 22 27 SS Wx

A . t '

2 O

10 14 19ob 19c 19d 22 27 SS Wx

Figure 3. Comparison of Zein Nuclear Transcripts in the Normal and opaque-2 Mutant with Sucrose Synthase Nuclear Transcripts in Each Genotype.

The data presented are the same as those in Figure 2, but are normalized to the leve1 of sucrose synthase transcripts at each developmental stage.

11 O The Plant Cell

Table 5. TranscriDtional Activitv Per Gene"

Gene copy Days After Pollination Zein Clone Number 10 12 14 16 18 20

lOkD l b 42.0 21.0 23.0 62.0 33.0 7.0 14 kD 2" 37.0 9.0 13.0 38.0 24.5 26.0 19 kD,ab 20" 4.6 6.3 7.8 15.9 6.3 4.6 19 kD,c 20" 2.9 6.5 6.3 18.3 12.4 4.3 19 kD,d 2" 8.5 30.5 25.0 89.5 70.0 26.5 22 kD 24" 1.1 3.8 4.5 16.1 9.0 2.6 27 kD l d 178.0 43.0 47.0 186.0 92.0 54.0

~

a Data are expressed as normal genotype transcription level (from Figure 2) divided by the estimated gene copy number. bKirihara et al., 1988.

Marks et al., 1984. Gallardo et al.. 1988.

transcriptionally. Our results demonstrated this is also true of zein gene expression during maize endosperm devel- opment. We found no evidence for transcription of any zein gene subfamily in leaf, root, or shoot tissue, nor did we find significant expression of genes involved in photo- synthesis in endosperm tissue. The rates of zein gene transcription increased in endosperm nuclei between 1 2 and 18 DAP. Transcription may decline at later stages of development, but we were unable to measure this because yields of nuclei from developmental stages later than 20 DAP were poof and had little activity in run-on transcription assays. This problem appeared to result primarily from the increase in starch content; however, nuclear instability associated with endoreduplication and enlargement that occurs during later stages of endosperm development (Knowles and Phillips, 1985) may be responsible. We were unable to purify nuclei efficiently by Percoll gradient cen- trifugation, which may have been a result of variation in nuclear size. Yields of nuclei from early stages of devel- opment were low, but this could be compensated for by increasing the amount of tissue.

Although the expression of various zein gene subfamilies was regulated coordinately, pronounced differences in their apparent rates of transcription were observed. Rela- tive to the genes encoding the M, 19,000 a-zeins, those encoding the M, 22,000 a-zeins were more actively tran- scribed (Figure 2). DNA sequence analyses suggest there may be a number of pseudogenes among the a-zeins (Spena et al., 1983; Kridl et al., 1984), which complicates the evaluation of this data. However, assuming most de- tected sequences are active genes, significant differences in the rates of transcription among zein subfamilies are evident. Comparing rates of transcription with gene copy numbers (Table 5) may bias the higher rates of transcrip- tion toward genes present in low copy numbers, but the same trends would be true even if only half of the predicted number of a-zein genes were active transcriptionally.

The variation in rates of transcription compared with the amoants of steady-state mRNAs (Figure 4) and proteins (Figure 1) indicates that zein gene expression is also regulated at the posttranscriptional level. Such differences in the prevalence of specific nuclear run-on transcripts on steady-state levels of mRNAs have been cited as evidence for posttranscriptional regulatory events (Beach et al., 1985; Walling et al., 1986). In the case of storage globulin synthesis in developing legume cotyledons, sulfur metab- olism appears to play a major role in controlling mRNA stability (Chandler et al., 1983; Beach et al., 1985; Holo- wach et al., 1986). Although the mechanisms involved in the posttranscriptional regulation of zeins are unknown, variation of mRNA stability, structure, or transport may be important.

The difference in transcription rate of genes encoding the M, 22,000 and 19,000 zeins is surprising since they share the same putative transcriptional regulatory ele- ments (Thompson and Larkins, 1989). There is no obvious explanation for the smaller amount of polysomal mRNAs

0.5 I

Nuclear

Polysomol 0.4 1 12 DAP

0.3

0.2

E 0.1

Z n

Q Z

c .- , - Y - 10 14 19ob 19c 19d 22 27 0 0.5 C 18 DAP O .- Z0.4

0.3

0.2

o. 1

O

U

L L L

10 14 19ob 19c 19d 22 27

Figure 4. Comparison of Zein Nuclear Transcripts with Amounts of Polysomal mRNAs.

Nuclear run-on transcripts were determined as described in Figure 2; the amount of transcripts for each zein sequence is expressed as a percent of the total. The amount of zein mRNA in total polysomes was determined as described previously (Marks et al., 1985a). Poly(A)+ mRNAs were used to synthesize radioactive cDNAs and the amount of each mRNA was determined by hybrid- ization in DNA excess with cloned zein sequences. The radioac- tivity corresponding to each type of mRNA is expressed as a percent of the total.

Gene Transcription in Maize Endosperm 11 1

I) N I _ N m - , I &

N

n “ f

+ + E? N

o

A I I

Figure 5. Restriction Enzyme Map of the Genomic Clone gzl9abll (Kriz et al., 1987).

Shaded area represents the coding region of the gene. Region containing the 5’ noncoding region with the P1 promoter (A), the coding region plus the P2 promoter (B), the 5‘ end of the coding sequence (C), and the 3‘ end of the coding sequence (D) were subcloned into M13mpl8/19 to obtain both coding and noncoding strands. These DNA segments were used as probes in a nuclear run-on transcription assay as described in “Methods.”

of the M, 22,000 relative to the M, 19,000 a-zeins in the normal genotype. Structural differences in mRNAs would not account for this variation, since sequences encoding both types of a-zein genes are related and thought to have arisen from a common ancestral gene (Wilson and Larkins, 1984; Marks et al., 1985b). The reduction in transcription of genes encoding the M, 22,000 relative to the M, 19,000 a-zeins in opaque-2 mutants may be related to the mech- anism responsible for this difference.

Although previous studies noted the reduction of a-zein mRNAs in opaque-2 mutants (Pedersen et al., 1980; Burr and Burr, 1982), it was not clear thet this mutation acted at the transcriptional level. However, results of the nuclear run-on transcription analysis demonstrate this clearly. Be- cause direct comparisons of transcriptional data are sub- ject to experimental variation, we normalized the expres- sion of zein genes to that of sucrose synthase (Figure 3). This comparison shows a substantial reduction in a-zein transcription in opaque-2 relative to the normal genotype. We believe this is a valid comparison, since there is no evidence to suggest that opaque-2 affects expression of the sucrose synthase gene.

As suggested by the differences in protein and mRNA levels, opaque-2 affects differentially the transcription of genes encoding M, 22,000 and 19,000 a-zein proteins. This indicates that, although these genes share common sequence elements that are responsible for their coordi- nate expression, there must also be specific regulatory factors that control their transcription. The complexity of transcriptional regulation for the zein gene family is shown further by the fact that the opaque-2 mutation appears to have little effect on transcription of other gene subfamilies. This is especially evident for the gene encoding the M,

27,000 zein, which shares many of the same conserved sequences in the 5’ noncoding region as the genes encod- ing the a-zeins (Boronat et al., 1986). The recent cloning of opaque-2 locus (Schmidt et al., 1987) should lead to an eventual understanding of how the protein encoded by this locus affects transcription of a-zein genes.

Severa1 previous studies concluded that high concentra- tions of zein precursor RNAs are transcribed in maize endosperm (Langridge et al., 1982a, 1982b). At least some of these large RNAs were found to map to two putative promoters (P1 and P2) that are separated by approxi- mately 900 bp (Langridge et al., 1982b; Langridge and Feix, 1983). S1 nuclease mapping of mRNA 5‘ ends and transcription in heterologous systems have given ambig- uous results regarding the utilization of these promoters (Langridge et al., 1982a, 1982b; Langridge et al., 1984; Brown et al., 1986), as well as the concentration of these RNAs in maize endosperm (Kriz et al., 1987). Based on our mapping of nuclear run-on transcripts, a-zein RNAs arising from the far upstream promoter (Pl) constitute, at most, a relatively minor percentage of the gene transcripts. There was a 20-fold difference between the extent of hybridization with the 21 60-bp probe that preceded the coding region by 228 nucleotides and the hybridization with the probe that contained the coding region and 228 nucleotides of 5’ noncoding sequence. Although there was a polar effect with regard to the labeling of run-on tran- scripts, it amounted to no more than a twofold difference between 5‘ and 3‘ ends of the coding region. Furthermore, the extent of hybridization to the 2-kb probe that preceded the P2 promoter was near the background level.

The hybridization results with the 19ab gene used for this experiment are representative of genes encoding a- zeins. This genomic clone corresponds to one of the largest gene subfamilies, with approximately 20 members, and previous studies have shown these genes have more than 90% sequence identity in their coding and 5‘ noncod- ing sequences (Kriz et al., 1987). It is unlikely that specific- ity of hybrid formation affected these results significantly, because similar values were obtained regardless of whether or not the filters were treated with ribonuclease following the hybridization reaction.

Table 6. Detection of Transcripts Corresponding to 19ab Zein Gene Subclones.

Probe” 14 DAPb O/O Probe B 20 DAPb % Probe B

A 11 5 5 5 B 204 1 O0 110 1 O0 C 76 37 32 29 D 181 89 62 56

a Described in Figure 5.

strand, average of three hybridizations. Data expressed as cpm in coding strand - cpm in noncoding

1 12 The Plant Cell

It remains to be established whether or not other mu- tations, such as opaque-7 and floury-2, also alter zein gene transcription. However, it is clear that transcriptional as well as posttranscriptional mechanisms play important roles in regulating the synthesis of maize storage proteins.

METHODS

Materials

DNase I and Proteinase K were purchased from Boehringer Mannheim (Indianapolis, IN). a3'P-GTP (2000 Ci/mmol) was ob- tained from ICN (Irvine, CA) and P~'P-ATP (3000 Ci/mmol) was purchased from Amersham (Arlington Heights, IL). RNase A and a-amanitin were obtained from Sigma (St. Louis, MO). Reverse transcriptase was purchased from Life Science Inc. (St. Peters- burg, FL). Redistilled phenol and restriction endonucleases were obtained from Bethesda Research Laboratories (Gaithersburg, MD). All other chemicals were reagent grade and were purchased from either Fisher (Springfield, NJ), S/P-Mallinckrodt (McGaw Park, IL), or Sigma.

W64A normal and opaque-2 corn (Zea mays L.) were grown during the summer of 1987. Ears of different developmental stages were harvested and frozen in liquid nitrogen immediately.

Extraction of Zeins and SDS-PAGE

Zeins were extracted from developing or mature seed by grinding endosperms in 70% ethanol plus 2% 2-mercaptoethanol at 65°C. The homogenate was centrifuged 5 min at lOOOg in a Sorvall SS- 34 rotor to pellet insoluble material, and the supernatant was diluted with sample buffer, boiled for 5 min, loaded onto a discon- tinuous SDS-polyactylamide gel, and separated as described by Laemmli (1 970).

Preparation of DNA Probes

Coding regions of clones described in Table 1 (except rDNA) were excised by restriction endonuclease digestion and religated into the multiple cloning region of M13mpl8/19 by standard methods (Yanisch-Perron et al., 1985). The rDNA clone was a double- stranded plasmid in the vector pUC8. The m13 clones were constructed such that single-stranded clones of both coding and noncoding strands were obtained.

DNA was bound to nitrocellulose (Schleicher & Schuell, Keene, NH) with a dot blot manifold (Bethesda Research Laboratories) at a concentration of 1.5 pg of DNA per spot. For hybridization to nuclear run-on transcripts, filter discs with bound DNA were removed from nitrocellulose sheets with a paper punch. For quantification of mRNAs, discs were not removed from nitrocel- lulose prior to hybridization with cDNAs.

lsolation of Nuclei

Procedures for nuclei isolation were carried out at 4OC. Thirty to 50 frozen kernels, 4 to 8 g each, depending on developmental

stage, were chopped with a razor blade in nuclei isolation buffer (NIB) (0.25 M sucrose, 25 mM Tris [pH 7.8 at 25"C], 10 mM MgCI2, 2.5% [w/v] Ficoll, 5.0% [w/v] dextran, and 10 mM P-mercaptoeth- anol) (Beach et al., 1985) and then ground very gently in a mortar and pestle. The homogenate was filtered through 100-pm Nitex mesh (Tetko Inc., Elmsford, NY) into 15-ml tubes, and nuclei'were pelleted by centrifugation for 5 min at lOOOg in a Sorvall HB-4 rotor. The pellet of starch and nuclei was resuspended in NIB, centrifuged as above, resuspended in NIB plus 20% glycerol, and centrifuged again. The supernatant was decanted and the tubes were placed upright for 5 to 10 min to allow the residual buffer to settle for resuspension of the pellet. The crude nuclear preparation contained a large amount of starch, but this did not interfere with the nuclear run-on transcription reaction. The starch content, which was considered excluded volume, was determined by centrifuging an aliquot of nuclei and determining the supernatant volume. Crude nuclear preparations were typically 20% to 60% starch. A second aliquot was taken from each preparation for measurement of the DNA content. Nuclei were stained with 4,6- diamidino-2-phenylindole and examined by fluorescence micros- copy to determine integrity and quantity.

Seedling leaf, root, and shoot nuclei were isolated by grinding tissue gently in liquid nitrogen. The nitrogen was allowed to evaporate and the powder homogenized in a buffer similar to that used to isolate endosperm nuclei, except the sucrose concentra- tion was 0.44 mM (leaf NIB) and Triton X-100 was included at 0.1 O/O (v/v). The homogenate was filtered through two layers of cheesecloth and then filtered twice through Miracloth (Behring Diagnostics, La Jolla, CA) into 15-ml tubes and centrifuged as above. The pellet, consisting of nuclei, plastids, and starch, was washed with leaf-NIB without Triton, followed by leaf-NIB plus 20% glycerol as described for endosperm nuclei.

Nuclear Run-On Transcription and Hybridization of RNA Tran- scripts

Synthesis of nuclear run-on transcripts was similar to the proce- dure of Beach et al. (1 985). The volume of the reaction was based on the volume of nuclei (less the starch content) obtained from each developmental stage. The reaction mixture, which ranged from 0.5 to 1 .O ml, contained 320 mM (NH4),S04, 0.35 mrv each ATP, CTP, and UTP, 4.3 p~ GTP, and 500 pCi of u-~'P-GTP. Freshly prepared nuclei in NIB (or leaf-NIB) plus 20% glycerol were diluted ninefold in this mixture. Typically, the DNA content was 75 to 180 pg per reaction. For some experiments, a-amanitin was included at 8 to 1 O pg/ml.

Transcription was carried out for 20 min at 25°C and followed by treatment with 80 pg/ml DNase I for 10 min. The mixture was adjusted to 5 mM Tris (pH 7.6 at 25"C), 0.5 mM EDTA, 0.5% SDS, and digested with 8 pg/ml Proteinase K 1 hr at 42OC. The RNA transcripts were extracted with phenol/chloroform/isoamyl alcohol (25:24:1, v/v) twice and tRNA was added to a final concentration of 0.1 pg/ml as carrier. RNA was precipitated in 0.3 M NaOAc (pH 5.2 at 25OC) with 2.5 volumes of etharlol. The resulting pellet was resuspended in 1 O mM sodium pyrophosphate and precipitated with 10% trichloroacetic acid. The RNA was washed with 95% ethanol and resuspended in 10 mM Tris (pH 7.6 at 25T) containing 1 mM EDTA. After an additional cycle of ethanol precipitation, the pellet was washed with 95% ethanol and resuspended in water.

Gene Transcription in Maize Endosperm 1 13

Prior to hybridization, filter discs with single-stranded DNA corresponding to both coding and noncoding strands of cloned sequences were incubated in triplicate in screw-capped cryotubes (Nunc, Neptune, NJ) in a buffer containing 10 mM Hepes (pH 7.8 at 25OC), 1 mM EDTA, 30% deionized formamide, 0.3 M NaCI, 0.2% (w/v) SDS, 0.2 mg/ml wheat germ tRNA, 0.1 mg/ml polyri- boadenylic acid, 1 mM sodium pyrophosphate, 0.02% (w/v) poly- vinylpyrrolidone. and 0.02% (w/v) Ficoll for 3 to 6 hr at 42OC. This solution was removed and 200 pl of fresh buffer containing nuclear run-on transcripts was added. Mineral oil was overlaid to prevent evaporation. The hybridization was carried out at 42OC for 36 to 42 hr. Following hybridization, the nitrocellulose discs were washed three time for 15 min at 65°C in 2 x SSC (1 x SSC =

0.15 M sodium chloride, 0.015 M sodium citrate [pH 7.0 at 25'C]), 0.5% SDS. The filters were then treated with 2 pg/ml RNase for 30 min at 37OC in 2 x SSC, followed by washing twice for 30 min in 2 x SSC. The filters were dried by blotting on Whatman 3MM paper, transferred individually to scintillation vials, and placed under heat lamps for 1 O to 12 min to dry. Scintillation mixture (Omnifluor, Du Pont-New England Nuclear, Boston, MA) was added and radioactivity determined for each filter disc by liquid scintillation spectroscopy.

DNA Measurement

The DNA content of nuclear preparations was determined by a colorimetric diphenylamine assay (Levya and Kelley, 1974) with sheared calf thymus DNA as a standard. Aliquots of nuclei were washed twice with 95% ethanol to remove sucrose, which could interfere with the measurement.

Analysis of Polysomal RNA

Total polysomes were isolated (Larkins and Hurkman, 1978). and poly(A)' RNA was purified by reversible binding to mAP Hybond affinity paper as described by the manufacturer (Amersham). The complexity of zein mRNAs was analyzed as described previously (Marks et al., 1985a). Briefly, cDNA was synthesized from the poly(A)+ RNA with reverse transcriptase by priming with oligo(dT). The cDNA was radiolabeled with (Y-~'P-~ATP and used as a probe in DNA excess blot hybridizations. After the hybridization reaction, each DNA dot was removed and the radioactivity was determined by liquid scintillation spectroscopy.

ACKNOWLEDGMENTS

We thank Drs. S. Wessler, W. Taylor, C. Hanna, J. Messing, A. Esen, and 1. Rubenstein for making their cDNA and genomic clones available to us. This research was supported by a grant from the National lnstitutes of Health (GM 36790) (to B.A.L.) R.S.B. was the recipient of a National lnstitutes of Health post- doctoral fellowship. R.K. was the recipient of a United States Department of Agriculture National Needs fellowship. This is journal paper number 11802 from the Purdue Agricultura1 Exper- iment Station.

REFERENCES

Beach, L.R., Spencer, D., Randall, P.J., and Higgins, T.J.V. (1 985). Transcriptional and post-transcriptional regulation of storage protein gene expression in sulfur-deficient pea seeds. Nucleic Acids Fies. 13, 999-1 O1 3.

Boronat, A., Martinez, M.C., Reina, M., Puigdomenech, P. and Palau, J. (1986). lsolation and sequencing of a 28-kD glutelin- 2 gene from maize. Common elements in the 5' flanking region among zein and glutelin genes. Plant Sci 47, 95-102.

Boston, R.S., Kodrzycki, R., and Larkins, B.A. (1986) Transcrip- tional regulation of maize zein genes. In Molecular Biology of Seed Storage Proteins and Lectins. Ninth Symposium in Plant Physiology, L. M. Shannon and M. J. Chrispeels, eds (Rockville, MD: American Society of Plant Physiologists), pp. 117-126.

Brown, J.W.S., Wandelt, C., and Feix, G. (1986). The upstream regions of zein genes: Sequence analysis and expression in the unicellular green alga acetabularia. Eur. J. Cell. Biol. 42, 161- 170.

Burr, F.A., and Burr, B. (1982). Three mutations in Zea mays affecting zein accumulation: A comparison of zein polypeptides, in vitro synthesis and processing, mRNA levels, and genomic organization. J. Cell Biol. 94, 201-206.

Chandler, P.M., Higgins, T.J.V., Randall, P.J., and Spencer, D. (1 983). Regulation of legumin levels in developing pea seeds under conditions of sulfur deficiency. Plant Physiol. 71, 47-54.

Difonro, N., Gentinetta, E., Salamini, F., and Soave, C. (1980). Action of the opaque-7 mutation on the accumulation of storage products in maize endosperm. Plant. Sci. Lett. 14, 345-354.

Gallardo, D., Reina, M., Rigau, J., Boronat, A., and Palau, J. (1988). Genomic organization of the 28 kDa glutelin-2 gene from maize. Plant. Sci. 54, 21 1-218.

Hagen, C., and Rubenstein, 1. (1981). Compex organization of zein genes in maize. Gene (Amst.) 13, 239-249.

Holowach, L.P., Madison, J.T., and Thompson, J.F. (1986). Studies on the mechanism of regulation of the mRNA leve1 for a soybean storage protein subunit by exogenous L-methionine. Plant Physiol. 80, 561-567.

Jones, R.A., Larkins, B.A., and Tsai, C.Y. (1977). Storage pro- teins synthesis in maize II. Reduced synthesis of a major zein component by the opaque-2 mutant of maize. Plant Physiol.

Jones, R.A. (1 978). Effects of floury-2 locus on zein accumulation and RNA metabolism during endosperm development. Biochem. Genet. 16, 27-38.

Kirihara, J.A., Hunsperger, J.P., Mahony, W.C., and Messing, J.W. (1988). Differential expression of a gene for a methionine- rich storage protein in maize. MOI. Gen. Genet. 211, 477-484.

Knowles, R.V., and Phillips, R.L. (1985). DNA amplification pat- terns in maize endosperm nuclei during kernel development. Proc. Natl. Acad. Sci. USA 82,7010-7014.

Kridl, J.C., Vieira, J.,. Rubenstein, I., and Messing, J. (1984). Nucleotide sequence of a zein genomic clone with a short open reading frame. Gene (Amst.) 28, 1 13-1 18.

Krir, A.L., Boston, R.S., and Larkins, B.A. (1 987). Structural and transcriptional analysis of DNA sequences flanking genes that encode 19 kilodalton zeins. MOI. Gen. Genet. 207, 90-98.

Laemmli, U.K. (1 970). Cleavage of structural protein during the

59,525-529.

Received October 26, 1988.

1 14 The Plant Cell

assembly of the head of bacteriophage T4. Nature 227, 680- 685.

Langridge, P., and Feix, G. (1983). A zein gene of maize is transcribed from two widely separated promoter regions. Cell.

Langridge, P., Pintor-Toro, J.A., and Feix, G. (1 982a). Transcrip- tion effects of the opaque-2 mutation of Zea mays L. Planta (Berl.) 156, 166-1 70.

Langridge, P., Pintor-Toro, J.A., and Feix, G. (1 982b). Zein precursor mRNAs from maize endosperm. MOI. Gen. Genet.

Langridge, P., Eibel, H., Brown, J.W.S., and Feix, G. (1984). Transcription from maize storage protein gene promoters in yeast. EMBO J. 3,2467-2471.

Larkins, B.A., and Hurkman, W.J. (1978). Synthesis and depo- sition of zein in protein bodies of maize endosperm. Plant Physiol. 62, 256-263.

Leyva, A., Jr., and Kelley, W.N. (1974). Measurement of DNA in cultured human cells. Anal. Biochem. 62, 173-179.

Marks, M.D., and Larkins, B.A. (1982). Analysis of sequence microheterogeneity among zein messenger RNAs. J. Biol. Chem. 257,9976-9983.

Marks, M.D., Pedersen, K., Wilson, D.R., DiFonzo, N., and Larkins, B.A. (1984). Molecular biology of the maize seed storage proteins. Curr. Top. Plant Biochem. Physiol. 3, 9-18.

Marks, M.D., Lindell, J.S., and Larkins, B.A. (1985a). Quantita- tive analysis of the accumulation of zein mRNA during maize endosperm development. J. Biol. Chem. 260, 16445-1 6450.

Marks, M.D., Lindell, J.S., and Larkins, B.A. (1985b). Nucleotide sequence analysis of zein mRNAs from maize endosperm. J. Biol. Chem. 260,16451 -1 6459.

Martineau, B., and Taylor, W.C. (1 985). Photosynthetic gene expression and cellular differentiation in developing maize leaves. Plant Physiol. 78, 399-404.

Messing, J., Carlson, J., Hagen, G., Rubenstein, I., Oleson, A. (1984). Cloning and sequencing of the ribosomal RNA genes in maize: The 17s region. DNA (NY) 3, 31-40.

Nelson, T., Harpster, M.H., Mayfield, P., and Taylor, W.C. (1 984). Light-regulated gene expression during maize leaf de- velopment. J. Cell Biol. 98, 558-564.

Pedersen, K., Bloom, K.S., Anderson, J.N., Glover, D.V., and Larkins, B.A. (1980). Analysis of the complexity and frequency of zein genes in the maize genome. Biochemistry 19, 1644- 1650.

34,1015-1022.

187,432-438.

Pedersen, K., Argos, P., Naravana, S.L.V., and Larkins, B.A. (1 986). Sequence analysis and characterization of a maize gene encoding a high-sulfur zein protein of M, 15.000. J. Biol. Chem.

Prat, S., Perez-Grau, L., and Puigdomenech, P. (1987). Multiple variability in the sequence of a family of maize endosperm proteins. Gene (Amst.) 52, 41-49.

Schmidt, R.J., Burr, F.A., and Burr, 6. (1987). Transposon tag- ging and molecular analysis of the maize regulatory locus opaque-2. Science 238,960-963.

Soave, C., and Salamini, F. (1984). Organization and regulation of zein genes in maize endosperm. Phil. Trans. R. SOC. Lond.

Soave, C., Tardani, L., DiFonzo, N., and Salamini, F. (1981). Zein leve1 in maize endosperm depends on a protein under control of the opaque-2 and opaque-6 loci. Cell 27, 403-41 O.

Spena, A., Viotti, A., and Pirotta, V. (1983). Two adjacent gen- omic zein sequences: Structure, organization, and tissue-spe- cific restriction pattern. J. MOI. Biol. 169, 799-81 1.

Thompson, G.A., and Larkins, B.A. (1 989). Structural elements regulating zein gene expression. Bioessays, in press.

Tsai, C.Y., Larkins, B.A., and Glover, D.V. (1978). lnteraction of the opaque-2 gene with starch-forming mutant genes on the synthesis of zein in maize endosperm. Biochem. Genet. 16,

Yanisch-Perron, C., Vieira, J., and Messing, J. (1 985). lmproved M13 phage cloning vectors and host strains: Nucleotide se- quences of the M13mp18 and pUC19 vectors. Gene (Amst.)

Wang, S.-Z., and Esen, A. (1 986). Primary structure of a proline- rich zein and its cDNA. Plant Physiol. 81, 70-74.

Werr, W., Frommer, W.-B., Mass, C., and Starlinger, P. (1985). Structure of the sucrose synthase gene on chromosome 9 of Zea mays L. EMBO J. 4, 1373-1380.

Walling, L., Drews, G.N., and Goldberg, R.B. (1 986). Transcrip- tional and post-transcriptional regulation of soybean seed pro- tein mRNA levels. Proc. Natl. Acad. Sci. USA 83, 2123-2127.

Wessler, S.R., and Marguerite, V.J. (1985). Molecular basis of mutations at the waxy locus of maize: Correlation with the fine structure genetic map. Proc. Natl. Acad. Sci. USA 82, 4177- 41 81.

Wilson, D.R., and Larkins, B.A. (1984). Zein gene organization

261,6279-6284.

(B) 304,341-347.

883-896.

33,103-109.

in maize and related grasses. J. MOI. Evol. 20, 330-340.

DOI 10.1105/tpc.1.1.105 1989;1;105-114Plant Cell

R Kodrzycki, R S Boston and B A LarkinsThe opaque-2 mutation of maize differentially reduces zein gene transcription.

 This information is current as of July 11, 2020

 

Permissions X

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

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

Sign up for eTOCs at:

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

Sign up for CiteTrack Alerts at:

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

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

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