Planta
Planta (1988)174:551 560
9 Springer-Verlag 1988
Hormonal regulation of gene expression in barley aleurone layers Induction and suppression of specific genes Randall C. Nolan* and Tuan-Hua David Ho ** Department of Biology and Plant Biology Program, Division of Biology and Biomedical Sciences, Washington University, St. Louis, MO 63130, USA
Abstract. As part of our investigation of the mode of action of plant hormones in barley (Hordeum vulgate L.) aleurone layers, we have studied the expression of five identified and three unidentified m R N A species in the presence of exogenous gibberellic acid (GA3) and abscisic acid. Three of the mRNAs are GA3-inducible, three are suppressed by GA3, and two are constitutive. The extent of the GA3 effect differs considerably for both inducible and suppressible mRNAs. For example, a tenfold higher concentration of GA3 (10 -8 M) is required for full induction of the high-pI group eamylase m R N A than is required for the low-pI e-amylase m R N A (10-9 M). Temporal regulation of m R N A abundance also varies between the two e-amylase isoenzyme groups. The three GA3-suppressible m R N A species studied, alcohol dehydrogenase (ADH1), a probable amylase and protease inhibitor, and an unidentified barley m R N A species also varied in response to GA3. The ADH1 m R N A decreased drastically within 8 h of GA3 treatment, whereas the other two began to decrease in abundance only after 12-16 h of GA3 treatment. Abscisic-acid treatment counteracted the GA3 effects for both the inducible and suppressible m R N A species. Comparison of e-amylase-mRNA levels and e-amylase-synthesis rates showed a strong correlation between the two parameters, the only exception being a lack of e-amylase synthesis in the presence of e-amylase m R N A at low GA3 concentrations. Therefore, the expression of eamylase seems to be regulated primarily by its m R N A levels. * Present address: Department of Biology, Indiana University, Bloomington, IN 47401, USA ** To whom correspondence should be addressed A b b r e v i a t i o n s : A B A = abscisic acid; ADH1 = alcohol dehydrogenase 1; eDNA=copy DNA; GA3=gibberellic acid; PAPI = probable amylase/protease inhibitor
Key words: Abscisic acid and gene expression Aleurone - e-Amylase - Gene expression (barley aleurone) - Gibberellin and gene expression - Hordeum (gene expression).
Introduction
Barley aleurone layers have provided a model system for the study of the regulation of gene expression by plant hormones. Gibberellic acid (GA3), synthesized in the germinating barley embryo, diffuses to the aleurone layer cells, where it triggers the synthesis and secretion of several hydrolytic enzymes. These hydrolases then break down storage materials in the endosperm into smaller components, which are used by the seedling to maintain growth until photosynthesis commences. In isolated aleurone layers, GA3 has been shown to induce synthesis of e-amylase isoenzymes (EC 3.2.1.1), proteases, nuclease (EC 3.1.30.2), fl1,3; 1,4-glucanases (EC 3.2.1.73), and other hydrolases (Jacobsen and Higgins 1982; Callis and Ho 1983; Hammerton and Ho 1986; Brown and Ho 1986; Stuart et al. 1986). The m R N A for e-amylase, fl-1,3 ; 1,4-glucanase, and a putative thiol protease, aleurain, have also been shown to be induced by GA3, (Rogers 1985; Muthukrishnan et al. 1983; Chandler et al. 1984; Stuart et al. 1986; Rogers et al. 1985). Jacobsen and Beach (1985) have demonstrated a tenfold stimulation of transcription of e-amylase genes by run-on transcription studies using nuclei from GA3-responsive aleurone protoplasts. Abscisic acid (ABA) blocks this stimulation of e-amylase transcription. In fact, ABA has been found to antagonize GA3 action in all of the events investigated thus far in barley aleurone layers (Ho 1983).
552
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
A m a j o r goal o f o u r research is to determine the patterns o f c o n t r o l that GA3 and A B A exert on gene expression in barley aleurone layers. T o this end, the effects o f these h o r m o n e s on the levels o f eight different m R N A s have been determined. N o r t h e r n blot experiments were carried out, in which R N A samples were subjected to electrophoresis and then hybridized with specific c D N A s (copy D N A s ) for the different m R N A species (Maniatis et al. 1982; C h u r c h and Gilbert 1984). T h r e e o f these m R N A species, encoding alcohol d e h y d r o g e n a s e ( A D H 1 , E C I . I . I . I ) , a p r o b a b l e eamylase/protease inhibitor (PAPI) and an unidentified m R N A have been f o u n d to be GA3-suppressible, whereas three m R N A s (high-pI e-amylase, low-pI e-amylase, and aleurain) are GA3-inducible, as has previously been r e p o r t e d (Rogers 1985; Rogers et al. 1985). T w o other transcripts were f o u n d to be essentially constitutive in expression. A c o m p l e x p a t t e r n o f regulation is evident for b o t h the suppressible and inducible m R N A s , as s h o w n by time course, GA3 dosage, and GA3 versus A B A experiments. C o m p a r i s o n o f e-amylase-isoenzyme synthesis and the m R N A levels for the two groups o f isoenzymes shows, with one possible exception, a strong correlation between the two parameters. This is f u r t h e r indication that e-amylase expression is regulated primarily at the m R N A level.
Material and methods Tissue preparation. Aleurone layers were prepared from barley (Hordeum vulgare L. cv. Himalaya) caryopses as described by Nolan et al. (1987). Seeds were from the 1981 harvest at Washington State University, Pullman, USA. Ten aleurone layers were incubated in 2 ml of buffer (20 mM Na-succinate pH 5.0, 20 mM CaCI2, 10 ~tg/mlchloramphenicol) for protein-synthesis studies and 75 aleurone layers were incubated in 15 ml of buffer for RNA extraction. Protein analysis. For analysis of protein synthesis, aleurone layers were labelled with 9.4 M Bq/ml of a mixture of [35S]methionine and [35S]cysteine(Trans 35S-label from ICN, Irvine, Cal., USA) for the last hour of incubation. Prior to labelling they were rinsed twice with incubation buffer. The layers were then ground with acid-washed sea sand and 100 lal of 10 IxM leupeptin in a chilled mortar. Non-denaturing gel sample buffer (30% glycerol, 10mM 2-amino-2-(hydroxymethyl)-l,3-propanediol (Tris)-HC1, pH6.8, 0.01% Bromophenol blue) was added to the extracts and trichloroacetic-acid (TCA)-precipitable counts were determined as described by Mans and Novelli (1960). Synthesis of a-amylase isoenzymes was analyzed by non-denaturing polyacrylamide gel electrophoresis, followed by fluorography, as described in Nolan et al. (1987). Aliquots containing equal TCA-precipitable counts were added to each gel lane. Intensity of a-amylase bands on fluorograms therefore indicates their percentage of total protein-synthesis activity (differences between lanes do not necessarily reflect differences in the absolute rates of 0c-amylasesynthesis).
Analysis of RNA. Total RNA was isolated as described by Belanger et al. (1986). Northern analysis was performed as described in Nolan et al. (1987). Briefly, 10 ~tg of total RNA was subjected to electrophoresis in formaldehyde-agarose gels as described in Maniatis etal. (1982) and blotted onto "Gene Screen" membranes (New England Biolabs, Boston, Mass., USA), using 10x SSC (3 M NaC1, 0.3 M Na-citrate). The membranes were baked at 80~ C for 2 h and hybridized to nicktranslated cDNAs as described by Church and Gilbert (1984). Relative levels of specific mRNAs were then shown by autoradiography of the filter hybridizations as described by Nolan et al. (1987). Barley cDNAs for high-p! co-amylase(pM/C), lowpI a-amylase (clone E), a putative thiol protease termed aleurain (G7i), and a probable amylase/protease inhibitor (Alf) were obtained from Dr. J. Rogers, Washington University Medical School (Rogers 1985; Rogers and Milliman 1984; Rogers et al. 1985; Mundy and Rogers 1986). Adh-1 cDNA from maize (pZmL793) was obtained from Dr. M. Sachs, Department of Biology, Washington University (Dennis et al. 1984), A ribosomal-RNA cDNA probe from soybean (pGmr-1) was obtained from Dr. L. Zimmer, Departnaent of Biochemistry, Louisiana State University, Baton Rouge, USA. Three other cDNAs were obtained from a pUC 18cDNA library from 4-h ABA-treated barley aleurone layers and had previously been found to be unaffected by ABA treatment (B. Hong and S. Uknes, Washington University, personal communication). They have been termed pHVU-I, pHVU-2, and pHVU-3.
Results Time course o f GA3-mediated gene expression. Gibberellic acid has varying effects on the expression o f different genes in barley aleurone layers as shown in Fig. 1. Barley e-amylase consists o f two groups o f isoenzymes, with almost identical molecular weight (44 kilodaltons, k D a ) but different isoelectric points (Jacobsen and Higgins 1982; Callis and H o 1983). Messenger R N A for a high-pI eamylase g r o u p ( p M / C p r o b e ) was virtually undetectable in the absence o f G A 3 , b e c a m e detectable by 4 h o f GA3 treatment, and reached a m a x i m u m level by 12-16 h o f GA3 treatment. As shown previously ( N o l a n e t a l . 1987), high-pI e-amylase m R N A started to decline after a r o u n d 24 h o f GA3 t r e a t m e n t and was greatly reduced by 40 h o f GA3 t r e a t m e n t (Fig. 1). L o w - p I e-amylase has been shown to be constitutively expressed at a low level (Rogers 1985). W e f o u n d that low-pI e-amylase m R N A (clone E p r o b e ) was evident in the 1-h time-point samples b o t h in the presence and absence o f GA3 (Fig. 1). Levels o f low-pI e-amylase m R N A decreased with time in the absence o f added G A 3 , b u t increased greatly in a b u n d a n c e by 8 h o f GA3 treatment. After 24 h o f GA3 t r e a t m e n t there was n o appreciable decline in the levels o f low-pI e-amylase transcripts with time as there was for high-pI e-amylase m R N A . High-pI and low-pI e-amylase m R N A therefore differed b o t h in degree o f induction by GA3 and in their expression over
R.C. Nolan and T.-H.D. Ho : Hormone-regulated gene expression in barley aleurone layers
553
Fig. 1. Northern analysis, showing the effects of incubation of barley aleurone layers in the absence (lanes 1~) or presence (lanes 6-15) of 10 - 6 M GA3 on the levels of specific R N A species. The number of above each lane gives the hours of incubation of the aleurone layers with or without GA3. Parallel blots of RNA, separated by electrophoresis, were hybridized with nick-translated c D N A plasmids, each panel showing the hybridization pattern for a different cDNA. Ten micrograms of total R N A was added to each lane. The X-ray films were exposed to give approximately equivalent intensities between panels. A 1-d exposure was normally sufficient for c~-amylase hybridization, whereas 4-7 d exposure was required for ADH1 hybridizations
extended periods of time. Our findings are quite different from those of Deikman and Jones (1986). They found that levels of low-pI e-amylase mRNA, as evidenced by hybridization to clone E, increased with incubation in the absence of GA3. We observed the opposite pattern for low-pI eamylase m R N A in aleurone layers incubated in the absence of GA3. Treatment of aleurone layers with GA3 produced similar patterns for synthesis of the two
groups of e-amylase isoenzymes as it did for eamylase-mRNA levels (Fig. 2), as has been shown previously (Nolan et al. 1987). Neither group of isoenzymes was synthesized when the aleurone layers were incubated for extended periods of time in the absence of GA3 (lane 1). Synthesis of both isoenzyme groups was detectable by 8 h of GA3 treatment. Synthesis of high-pI e-amylase decreased after around 24 h of GA3 treatment, whereas synthesis of low-pI e-amylase continued
554
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
Fig. 2. Time course of the synthesis of ~amylase isoenzymes in barley aleurone layers in the presence of 10 .6 M GA3. Non-denaturing polyacrylamide gel electrophoresis of extracts of [35S]methionine-labelled aleurone layers was performed, followed by fluorography. Equal TCA-precipitable counts were added to each lane. The top a r r o w shows the location on the gel of the high-pI a-amylase isoenzymes and the l o w e r t w o a r r o w s show low-pI ~amylase isoenzymes. Duplicate samples were run for each of the GAa time points, with the hours of incubation in GA3 listed above the brackets
at a high rate even at 40 h of GA3 treatment. These data indicate that the rates of synthesis of c~-amylase isoenzymes are determined by the abundance of their respective mRNAs under these ~:onditions. Protease activity has been shown in several studies to be induced by GA3 in aleurone layers (Jacobsen and Varner 1967; Hammerton and Ho 1986). A GAa-inducible aleurone mRNA encoding a putative protease, termed aleurain, has been cloned by Rogers et al. (1985) and shown to have sequence homology to the mammalian thiol protease cathepsin H (Rogers et al. 1985; Mundy and Rogers 1986). Northern analysis for aleurain (G7i probe) produced a pattern of expression similar to that of low-pI c~-amylase (Fig. 1), but with a lower degree of induction of aleurain by GA3. Substantial amounts of aleurain mRNA were present in aleurone layers at the time when they were peeled away from the starchy endosperm (data not shown). We found that the abundance of this mRNA greatly decreased with time in the absence o f GA3, whereas there was a modest increase in aleurain mRNA after 12 h of incubation in the presence of GA3. One interpretation of these data is that a gibberellin or some other inductive factor is initially present upon isolation of aleurone layers
and declines with time in the absence of added GA3. Alcohol dehydrogenase (ADH1) has previously been shown to be constitutively expressed in barley aleurone layers in the absence of GA3 (Hanson and Jacobsen 1984). We also found continued expression of ADH1 in the absence of GA3. However, after approx. 8 h of GA3 treatment, ADH1 mRNA decreased drastically. At 24 h of incubation there were measurable levels of ADH1 mRNA in the absence of GA3, but very low levels in the presence of GA3. The PAPI-mRNA levels have previously been shown not to be affected by hormonal treatment in barley aleurone layers (Mundy and Rogers 1986). However, we observed that the abundance of PAPI mRNA decreased after around 16 h of GA3 treatment, and was greatly reduced after 24 h of GA3 incubation. Another transcript (pHVU-I) was also found to decrease in abundance after around 8 h of GA3 treatment. Two aleurone transcripts were found (pHVU-2 and pHVU-3) to be neither induced or suppressed by GA3. The levels of ribosomal RNA were also unaffected by incubation or GA3 treatment (Fig. 1). Baulcombe and Buffard (1983) showed that six mRNAs species (one of which was
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
555
Fig. 3. The effect of GA3 concentration on R N A levels in barley aleurone layers. Northern analysis was performed as described in the legend of Fig. 1. For lanes I 6, the aleurone layers were incubated for 8 h in the different hormone concentrations and for lanes 711 the aleurone layers were incubated for 16 h. For lane 1, 2-10 -5 M ABA was added to the incubation buffer and for lanes 2 and 7 neither GA3 or A B A was added. The different molarities of GA3 added to the other incubations are listed above each lane
a-amylase) were GAa-inducible and one unidentified m R N A was suppressed by GA3 in wheat. Our finding is, to the best of our knowledge, the first demonstration of suppression of an identified m R N A species by GA3 in barley aleurone layers.
Effects of GAs concentration of gene expression. Different sensitivities to GA3 concentration in the incubation medium were also observed for the different genes studied. A concentration of 1 0 - 9 M GA3 was adequate for induction of low-pI a-amylase mRNA, while l 0 - s M GA3 was required for
expression of high-pI 0c-amylase m R N A (Fig. 3). Muthukrishnan et al. (1983) found the same GA3concentration requirement, 10-8 M GA3, for the expression of high-pI 0~-amylase, but did not analyze levels of low-pI 0~-amylase mRNA. Aleurain m R N A showed basically the same hybridization pattern as low-pI e-amylase m R N A : it was induced by all four GA3 concentrations tested. A higher GA3 concentration was also required for synthesis of the high-pI a-amylase isoenzymes than was necessary for the low-pI ~-amylase-isoenzyme group (Fig. 4). This is in agreement with the
556
R.C. Nolan and T.-H.D. Ho : Hormone-regulatedgene expressionin barley aleurone layers
Fig. 4. The effectof GA3 concentration on the synthesisof a-amylase isoenzymes. Non-denaturing polyacrylamidegel electrophoresis was performed as described under Fig. 2. Lanes I 6 are for 8-h incubations and lanes 7-11 are for 16-h incubations. For lane 1, 2.10 -5 M ABA was added to the incubation buffer and for lanes 2 and 7 neither GA3 or ABA was added. The differentmolarities of GA3 added to the other incubations are listed above each Iane. ~-Amylaseisoenzymetocafionsare indicated by arrows on the right results of Jacobsen and Higgins (1982), who analyzed the accumulation of e-amylase-isoenzyme activities. Additionally, 10-9 M GA3 induced low-pI e-amylase m R N A , but low-pI e-amylase was not consistently synthesized at appreciable rates at this GA3 concentration. A higher GA3 concentration thus seems to be required for translation or posttranslational processing of low-pI e-amylase than is required for the accumulation of low-pI e-amylase m R N A for these time points. This pattern also seems to be true for the high-pI e-amylase isoenzymes. These data thus contrast with the previous two figures, which showed a strong correlation between the patterns of synthesis of e-amylase isoenzymes and the respective levels of e-amylase mRNA. Ribosomal R N A levels were again unaffected by the addition of ABA or GA3 to aleurone layers (Fig. 3). The ADH1 m R N A was present in the ABA-treated layers and the layers incubated 16 h in the absence of added hormones, but was not
found in any of the 8- and 16-h GA3-treated layers. Therefore, 10 -9 M GA3 was adequate to cause a decrease in ADH1 mRNA. The levels of pHVU-1 transcripts were not affected by 8 h of GA3 treatment but decreased in all of the 16-h GA3-treated layers (data not shown). The PAPI m R N A was also not affected by GA3 of ABA over 8 h of treatment, but was slightly less abundant in the 16-h GA3-treated layers. These data therefore corroborate the GAa time-course data; GA3 both positively and negatively regulates m R N A abundances, with some m R N A species being more highly regulated than others. Abscisic acid c o u n t e r a c t s the effects o f GA3 treatm e n t . Previous investigations have shown that
when ABA and GA3 are added simultaneously to aleurone layers, appearance of translatable e-amylase m R N A is greatly delayed (Mozer 1980; Higgins et al. 1982). The synthesis of e-amylase was even more negatively influenced by ABA addition.
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
557
Fig. 5. Addition of ABA counteracts the effects of GA3 on m R N A abundance in barley aleurone layers. N o r t h e r n analysis was performed as described in Fig. 1. For lanes 1-7, the aleurone layers were incubated for 8 h in the different hormone concentrations and for lanes 814 the aleurone layers were incubated for 16 h. For lanes I and 8, no hormone was added; 1 ~tM GA3 was added in all of the other incubations. The A B A molarities are listed above the appropriate lanes
It was suggested from these data that ABA might in some cases be exerting translational control on e-amylase expression. This possibility was reinvestigated for both groups of e-amylase isoenzymes by incubating aleurone layers in 1 0 - 6 M GA3 and varying ABA concentrations and then analyzing e-amylase-mRNA levels (Fig. 5) and e-amylase synthesis (Fig. 6). If extensive translational control were occurring, one would see a more drastic effect of ABA on protein synthesis than on m R N A levels for the e-amylase isoenzymes.
As shown in Figs. 5 and 6, high ABA concentrations prevented the expression of both e-amylase-isoenzyme groups at both the m R N A - and protein-synthesis levels. High ABA concentrations (10 -4 M and 10 -2 M ABA) counteracted the effects of 10 - 6 M GA3 (lanes 2, 3, 9, 10), whereas lower ABA concentrations were less effective. Since e-amylase synthesis and m R N A levels were affected to the same extent by ABA, no evidence for translational control by ABA was apparent. Addition of ABA affected the other m R N A s
558
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
Fig. 6. Abscisic acid inhibits the induction of ~-amylase-isoenzyme synthesis in barley aleurone layers by GA3. The incubation times were 8 h for lanes 1-7 and 16 h for lanes 8-14; 1 I.tM GA3 was added for lanes 2-7 and ~ 1 4 . The molarities of ABA for lanes 2-6 and ~ 1 3 are listed above these lanes. Nondenaturing polyacrylamide gel electrophoresis was performed as described in the legend for Fig. 2. The locations of c~-amylase isoenzymes are indicated by arrows on the right
tested in a manner consistent with its role as a GA3 antagonist. Expression of A D H 1 was suppressed by GA3 and high A B A concentrations overcame the effect of GA3. The levels of PAPI m R N A also decreased with 16 h of GA3 incubation, and high A B A concentrations blocked this effect of GA3. At 8 h of incubation, induction of aleurain m R N A was reduced by high A B A concentrations. However, with 16 h of incubation none of the A B A concentrations decreased aleurain expression. Discussion
Our results show that there is a great deal of variation in the responses of different genes to GA3 in barley aleurone layers. At least two classes of GA3-inducible genes are apparent, varying in response to GA3 concentration and to duration of GA3 treatment. Two GA3-suppressible genes were identified. They also exhibit quite different responses to GA3. The mechanisms by which these differences in
hormonal response occur remain to be ascertained. Stimulation of run-on transcription of a-amylase genes has been demonstrated by Jacobsen and Beach (1985), using nuclei isolated from GA3-responsive barley aleurone protoplasts. Similar results have been obtained in wild-oat (Avenafatua) protoplasts (Zwar and Hooley 1986). The large degree of induction of 0~-amylase and other genes by GA3 therefore seems to be at least partly the result of transcriptional activation. It is not however evident whether GA3 also increases the m R N A half-life of certain transcripts in barley aleurone layers. Estrogen has been shown to increase both the transcription rate and m R N A halflife of vitellogenin m R N A in Xenopus liver (Brock and Shapiro 1983), so hormonal regulation of m R N A stability is not without precedent. A hormonally-induced increase in m R N A half-life has not yet been demonstrated in a plant system. However, Flores and Tobin (1986) have suggested that benzyladenine exhibits a positive post-transcriptional regulation of chlorophyll a/b-binding protein m R N A in Lemna gibba. Differences in GA3
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
response may result from regulation at both the transcriptional and post-transcriptional levels. The decrease in the levels of certain m R N A s after GA3 treatments is the major new finding of this research. Investigation of the mode of action of GA3 in aleurone layers may eventually need to include gene suppression as a part of the integrated effects of this hormone. The mechanism of decrease in m R N A abundance of specific genes is completely unknown. It is not known, for instance, whether A D H I and PAPI genes are being transcribed in isolated aleurone layers. Both ADH1 and PAPI m R N A s are present in the aleutone layers of immature barley seeds (Hanson et al. 1984; Mundy and Rogers 1986), and their presence in mature aleurone layers could be the result of previous m R N A synthesis. Messenger-RNA degradation is probably involved in the decreases in m R N A levels, and GA3 treatment may cause either specific or general decreases in m R N A stability. The effects of GA3 and ABA on m R N A stability are currently being studied in more detail. It is apparent that GA3 and ABA regulate gene expression primarily by regulating m R N A abundance. This does not mean, however, that regulation of translation does not occur. A requirement of calcium for translation of high-pI a-amylase m R N A in barley aleurone layers has, in fact, been shown by Deikman and Jones (1986). After 24 h of GA3 treatment, total a-amylase synthesis accounts for 40-60% of [35S]methionine incorporation into protein, a-Amylase mRNA, however, only represents around 20% of total translatable m R N A under these conditions (Ho 1980). This provides some indication that c~-amylase is preferably translated in aleurone layers. Gibberellic-acidinduced cellular changes, e.g. increase in endoplasmic reticulum (Jones 1969a, b; Belanger etal. 1986), may be required for efficient translation of a-amylase mRNA. A higher GA3 level may be required for these processes than for synthesis of ~amylase mRNA. The control of gene expression by GA3 varies greatly among genes. Future work investigating GA3 receptors must take into account the complexities of the GA3 response. Abscisic acid seems to counteract GA3 at early regulatory steps, since it is able to block GA3 action for all of the genes studied. A crucial finding for understanding hormonal action in barley aleurone layers will be to ascertain the steps, if any, at which both hormones act. We wish to thank Dr. John Rogers for the use of cDNA clones isolated in his laboratory. This work was supported by a National Science Foundation grant DCB-8702299.
559
References Baulcombe, D.C., Buffard, D. (1983) Gibberellic-acid-regulated expression of e-amylase and six other genes in wheat aleurone layers. Planta 157, 493 501 Belanger, F.C., Brodl, M.R., Ho, T.-H.D. (1986) Heat shock causes destabilization of specific mRNAs and destruction of endoplasmic reticulum in barley aleurone cells. Proc. Natl. Acad. Sci. USA 83, 1354-1358 Brock, M.L., Shapiro, D.J. (1983) Estrogen stabilizes vitellogenin mRNA against cytoplasmic degradation. Cell 17, 319-325 Brown, P.H., Ho, T.-H.D. (1986) Barley aleurone layers secrete a nuclease in response to gibberellic acid. Plant Physiol. 82, 801-806 Callis, J_, Ho, T.-H.D. (1983) Multiple molecular forms of the gibberellin-induced e-amylase from the aleurone layer of barley seeds. Arch. Bioehem. Biophys. 224, 224-234 Chandler, P.M., Zwar, J.A., Jacobsen, J.V., Higgins, TJ.V., Inglis, A.S. (1984) The effects of gibberellic acid and abscisic acid on e-amylase mRNA levels in barley aleurone layer studies using an e-amylase cDNA clone. Plant Mol. Biol. 3, 407-418 Chrispeels, M.J., Varner, J.E. (1967) Hormonal control of enzyme synthesis: On the mode of action of gibberellic acid and abscisin in aleurone layers of barley. Plant Physiol. 42, 1008-1016 Church, G.M., Gilbert, W. (1984) Genomic sequencing. Proc. Natl. Acad. Sci. USA 81, 1991-1995 Deikman, J., Jones, R.L. (1986) Regulation of the accumulation of mRNA for e-amylase isoenzymes in barley aleurone. Plant Physiol. 80, 67~675 Dennis, E.S., Gerlach, W.L., Pryor, A.J., Bennetzen, J.L., Inglis, A., Llewellyn, D.L., Sachs, M.M., Ferl, R.J., Peacock, W.J. (1984) Molecular analysis of the Alcohol dehydrogenase (Adhl) gene of maize. Nucl. Acid. Res. 9, 3983-4000 Flores, S., Tobin, E.M. (1986) Benzyladenine modulation of the expression of two genes for nuclear-encoded chloroplast in Lemna gibba: Apparent post-transcriptional regulation. Planta 168, 340-349 Hammerton, R.W., Ho, T-H.D. (1986) Hormonal regulation of the development of protease and carboxypeptidase activities in barley aleurone layers. Plant Physiol. 80, 692-697 Hanson, A.D., Jacobsen, J.V., Zwar, J.A. (1984) Regulated expression of three Alcohol dehydrogenase genes in barley aleurone layers. Plant Physiol. 75, 573-581 Higgins, T.J.V., Jacobsen, J.V., Zwar, J.A. (1982) Gibberellic acid and abscisic acid modulate protein synthesis and mRNA levels in barley aleurone layers. Plant Mol. Biol. 1,191-215 Ho, T-H.D. (1980) Hormonal and genetic regulation of e-amylase synthesis in barley aleurone cells. In: Genome organization and expression in plants, pp. 147-158, Leaver, C.J., ed. Plenum, New York London Ho, T-H.D. (1983) Biochemical mode of action of abscisic acid. In: Abscisic acid, pp. 142170, Addicott, F.A., ed. Prager, New York Jacobsen, J.V., Beach, L.R. (1985) Control of transcription of e-amylase and rRNA genes in barley aleurone layer protoplasts by gibberellic acid and abscisic acid. Nature 316, 275-277 Jacobsen, J.V., Higgins, T.J.V. (1982) Characterization of the c~-amylase synthesized by aleurone layers of Himalaya barley in response to GA3. Plant Physiol. 70, 1647-1653 Jacobsen, J.V., Varner, J.E. (1967) Gibberellic acid induced synthesis of protease by isolated aleurone layers of barley. Plant Physiol. 42, 1596-1600
560
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
Jones, R.L. (1969a) Gibberellie acid and the fine structure of barley aleurone layers. 1. Changes during the lag phase of e-amylase synthesis. Planta 87, 119-133 Jones, R.L. (1969b) Gibberellic acid and the fine structure of barley aleurone layers. 2. Changes during the synthesis and secretion of e-amylase. Planta 88, 73-86 Maniatis, T., Fritsch, E.F., Sambrook, J. (1982) Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Habor, N.Y., USA Mans, R.J., Novelli, G.D. (1960) A convenient, rapid, and sensitive method for measuring the incorporation of amino acids into protein. Biochem. Biophys. Res. Commun. 3, 540-548 Mozer, T.J. (1980) Control of protein synthesis in barley aleurone layers by the plant hormones gibberellic acid and abscisic acid. Cell 20, 479-485 Mundy, J., Rogers, J.C. (1986) Selective expression of a probable amylase/protease inhibitor in barley aleurone cells: Comparison to the barley amylase/subtilisin inhibitor. Planta 169, 51-63 Muthukrishnan, S., Chandra, G.R., Albaugh, G.P. (1983) Modulation by abscisic acid and S-2-aminoethyl-L-cysteine of e-amylase mRNA in barley aleurone cells. Plant Mol. Biol. 2, 249-258
Nolan, R.C., Lin, L-S., Ho, T-H.D. (1987) The effect of abscisic acid on the differential expression of e-amylase isoenzymes in barley aleurone layers. Plant Mol. Biol. 8, 13-22 Rogers, J.C. (1985) Two barley e-amylase gene families are regulated differently in aleurone cells. J. Biol. Chem. 260, 3731-3738 Rogers, J.C., Dean, D., Heck, G.R. (1985) Aleurain: A barley thiol protease closely related to mammalian cathepsin H. Proc. Natl. Acad. Sci. USA 82, 6512-6516 Rogers, J.C., Milliman, C. (1984) Coordinate increase in major transcripts from the high-pI e-amylase multigene family in barley aleurone cells stimulated with gibberellic acid. J. Biol. Chem. 259, 1223-1240 Stuart, I.M., Loi, L., Fincher, G.B. (1986) Development of (1-3, 1-4)-fl-D-glucan endohydrolase in isolated scutella and aleurone layers of barley (Hordeum vulgate). Plant Physiol. 80, 310-314 Zwar, J.Z., Hooley, R. (1986) Hormonal regulation of e-amylase gene transcription in wild oat (Arena fatua L.) protoplasts. Plant Physiol. 80, 459-463
Received 9 November 1987; accepted 18 January 1988
Planta (1988)174:551 560
9 Springer-Verlag 1988
Hormonal regulation of gene expression in barley aleurone layers Induction and suppression of specific genes Randall C. Nolan* and Tuan-Hua David Ho ** Department of Biology and Plant Biology Program, Division of Biology and Biomedical Sciences, Washington University, St. Louis, MO 63130, USA
Abstract. As part of our investigation of the mode of action of plant hormones in barley (Hordeum vulgate L.) aleurone layers, we have studied the expression of five identified and three unidentified m R N A species in the presence of exogenous gibberellic acid (GA3) and abscisic acid. Three of the mRNAs are GA3-inducible, three are suppressed by GA3, and two are constitutive. The extent of the GA3 effect differs considerably for both inducible and suppressible mRNAs. For example, a tenfold higher concentration of GA3 (10 -8 M) is required for full induction of the high-pI group eamylase m R N A than is required for the low-pI e-amylase m R N A (10-9 M). Temporal regulation of m R N A abundance also varies between the two e-amylase isoenzyme groups. The three GA3-suppressible m R N A species studied, alcohol dehydrogenase (ADH1), a probable amylase and protease inhibitor, and an unidentified barley m R N A species also varied in response to GA3. The ADH1 m R N A decreased drastically within 8 h of GA3 treatment, whereas the other two began to decrease in abundance only after 12-16 h of GA3 treatment. Abscisic-acid treatment counteracted the GA3 effects for both the inducible and suppressible m R N A species. Comparison of e-amylase-mRNA levels and e-amylase-synthesis rates showed a strong correlation between the two parameters, the only exception being a lack of e-amylase synthesis in the presence of e-amylase m R N A at low GA3 concentrations. Therefore, the expression of eamylase seems to be regulated primarily by its m R N A levels. * Present address: Department of Biology, Indiana University, Bloomington, IN 47401, USA ** To whom correspondence should be addressed A b b r e v i a t i o n s : A B A = abscisic acid; ADH1 = alcohol dehydrogenase 1; eDNA=copy DNA; GA3=gibberellic acid; PAPI = probable amylase/protease inhibitor
Key words: Abscisic acid and gene expression Aleurone - e-Amylase - Gene expression (barley aleurone) - Gibberellin and gene expression - Hordeum (gene expression).
Introduction
Barley aleurone layers have provided a model system for the study of the regulation of gene expression by plant hormones. Gibberellic acid (GA3), synthesized in the germinating barley embryo, diffuses to the aleurone layer cells, where it triggers the synthesis and secretion of several hydrolytic enzymes. These hydrolases then break down storage materials in the endosperm into smaller components, which are used by the seedling to maintain growth until photosynthesis commences. In isolated aleurone layers, GA3 has been shown to induce synthesis of e-amylase isoenzymes (EC 3.2.1.1), proteases, nuclease (EC 3.1.30.2), fl1,3; 1,4-glucanases (EC 3.2.1.73), and other hydrolases (Jacobsen and Higgins 1982; Callis and Ho 1983; Hammerton and Ho 1986; Brown and Ho 1986; Stuart et al. 1986). The m R N A for e-amylase, fl-1,3 ; 1,4-glucanase, and a putative thiol protease, aleurain, have also been shown to be induced by GA3, (Rogers 1985; Muthukrishnan et al. 1983; Chandler et al. 1984; Stuart et al. 1986; Rogers et al. 1985). Jacobsen and Beach (1985) have demonstrated a tenfold stimulation of transcription of e-amylase genes by run-on transcription studies using nuclei from GA3-responsive aleurone protoplasts. Abscisic acid (ABA) blocks this stimulation of e-amylase transcription. In fact, ABA has been found to antagonize GA3 action in all of the events investigated thus far in barley aleurone layers (Ho 1983).
552
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
A m a j o r goal o f o u r research is to determine the patterns o f c o n t r o l that GA3 and A B A exert on gene expression in barley aleurone layers. T o this end, the effects o f these h o r m o n e s on the levels o f eight different m R N A s have been determined. N o r t h e r n blot experiments were carried out, in which R N A samples were subjected to electrophoresis and then hybridized with specific c D N A s (copy D N A s ) for the different m R N A species (Maniatis et al. 1982; C h u r c h and Gilbert 1984). T h r e e o f these m R N A species, encoding alcohol d e h y d r o g e n a s e ( A D H 1 , E C I . I . I . I ) , a p r o b a b l e eamylase/protease inhibitor (PAPI) and an unidentified m R N A have been f o u n d to be GA3-suppressible, whereas three m R N A s (high-pI e-amylase, low-pI e-amylase, and aleurain) are GA3-inducible, as has previously been r e p o r t e d (Rogers 1985; Rogers et al. 1985). T w o other transcripts were f o u n d to be essentially constitutive in expression. A c o m p l e x p a t t e r n o f regulation is evident for b o t h the suppressible and inducible m R N A s , as s h o w n by time course, GA3 dosage, and GA3 versus A B A experiments. C o m p a r i s o n o f e-amylase-isoenzyme synthesis and the m R N A levels for the two groups o f isoenzymes shows, with one possible exception, a strong correlation between the two parameters. This is f u r t h e r indication that e-amylase expression is regulated primarily at the m R N A level.
Material and methods Tissue preparation. Aleurone layers were prepared from barley (Hordeum vulgare L. cv. Himalaya) caryopses as described by Nolan et al. (1987). Seeds were from the 1981 harvest at Washington State University, Pullman, USA. Ten aleurone layers were incubated in 2 ml of buffer (20 mM Na-succinate pH 5.0, 20 mM CaCI2, 10 ~tg/mlchloramphenicol) for protein-synthesis studies and 75 aleurone layers were incubated in 15 ml of buffer for RNA extraction. Protein analysis. For analysis of protein synthesis, aleurone layers were labelled with 9.4 M Bq/ml of a mixture of [35S]methionine and [35S]cysteine(Trans 35S-label from ICN, Irvine, Cal., USA) for the last hour of incubation. Prior to labelling they were rinsed twice with incubation buffer. The layers were then ground with acid-washed sea sand and 100 lal of 10 IxM leupeptin in a chilled mortar. Non-denaturing gel sample buffer (30% glycerol, 10mM 2-amino-2-(hydroxymethyl)-l,3-propanediol (Tris)-HC1, pH6.8, 0.01% Bromophenol blue) was added to the extracts and trichloroacetic-acid (TCA)-precipitable counts were determined as described by Mans and Novelli (1960). Synthesis of a-amylase isoenzymes was analyzed by non-denaturing polyacrylamide gel electrophoresis, followed by fluorography, as described in Nolan et al. (1987). Aliquots containing equal TCA-precipitable counts were added to each gel lane. Intensity of a-amylase bands on fluorograms therefore indicates their percentage of total protein-synthesis activity (differences between lanes do not necessarily reflect differences in the absolute rates of 0c-amylasesynthesis).
Analysis of RNA. Total RNA was isolated as described by Belanger et al. (1986). Northern analysis was performed as described in Nolan et al. (1987). Briefly, 10 ~tg of total RNA was subjected to electrophoresis in formaldehyde-agarose gels as described in Maniatis etal. (1982) and blotted onto "Gene Screen" membranes (New England Biolabs, Boston, Mass., USA), using 10x SSC (3 M NaC1, 0.3 M Na-citrate). The membranes were baked at 80~ C for 2 h and hybridized to nicktranslated cDNAs as described by Church and Gilbert (1984). Relative levels of specific mRNAs were then shown by autoradiography of the filter hybridizations as described by Nolan et al. (1987). Barley cDNAs for high-p! co-amylase(pM/C), lowpI a-amylase (clone E), a putative thiol protease termed aleurain (G7i), and a probable amylase/protease inhibitor (Alf) were obtained from Dr. J. Rogers, Washington University Medical School (Rogers 1985; Rogers and Milliman 1984; Rogers et al. 1985; Mundy and Rogers 1986). Adh-1 cDNA from maize (pZmL793) was obtained from Dr. M. Sachs, Department of Biology, Washington University (Dennis et al. 1984), A ribosomal-RNA cDNA probe from soybean (pGmr-1) was obtained from Dr. L. Zimmer, Departnaent of Biochemistry, Louisiana State University, Baton Rouge, USA. Three other cDNAs were obtained from a pUC 18cDNA library from 4-h ABA-treated barley aleurone layers and had previously been found to be unaffected by ABA treatment (B. Hong and S. Uknes, Washington University, personal communication). They have been termed pHVU-I, pHVU-2, and pHVU-3.
Results Time course o f GA3-mediated gene expression. Gibberellic acid has varying effects on the expression o f different genes in barley aleurone layers as shown in Fig. 1. Barley e-amylase consists o f two groups o f isoenzymes, with almost identical molecular weight (44 kilodaltons, k D a ) but different isoelectric points (Jacobsen and Higgins 1982; Callis and H o 1983). Messenger R N A for a high-pI eamylase g r o u p ( p M / C p r o b e ) was virtually undetectable in the absence o f G A 3 , b e c a m e detectable by 4 h o f GA3 treatment, and reached a m a x i m u m level by 12-16 h o f GA3 treatment. As shown previously ( N o l a n e t a l . 1987), high-pI e-amylase m R N A started to decline after a r o u n d 24 h o f GA3 t r e a t m e n t and was greatly reduced by 40 h o f GA3 t r e a t m e n t (Fig. 1). L o w - p I e-amylase has been shown to be constitutively expressed at a low level (Rogers 1985). W e f o u n d that low-pI e-amylase m R N A (clone E p r o b e ) was evident in the 1-h time-point samples b o t h in the presence and absence o f GA3 (Fig. 1). Levels o f low-pI e-amylase m R N A decreased with time in the absence o f added G A 3 , b u t increased greatly in a b u n d a n c e by 8 h o f GA3 treatment. After 24 h o f GA3 t r e a t m e n t there was n o appreciable decline in the levels o f low-pI e-amylase transcripts with time as there was for high-pI e-amylase m R N A . High-pI and low-pI e-amylase m R N A therefore differed b o t h in degree o f induction by GA3 and in their expression over
R.C. Nolan and T.-H.D. Ho : Hormone-regulated gene expression in barley aleurone layers
553
Fig. 1. Northern analysis, showing the effects of incubation of barley aleurone layers in the absence (lanes 1~) or presence (lanes 6-15) of 10 - 6 M GA3 on the levels of specific R N A species. The number of above each lane gives the hours of incubation of the aleurone layers with or without GA3. Parallel blots of RNA, separated by electrophoresis, were hybridized with nick-translated c D N A plasmids, each panel showing the hybridization pattern for a different cDNA. Ten micrograms of total R N A was added to each lane. The X-ray films were exposed to give approximately equivalent intensities between panels. A 1-d exposure was normally sufficient for c~-amylase hybridization, whereas 4-7 d exposure was required for ADH1 hybridizations
extended periods of time. Our findings are quite different from those of Deikman and Jones (1986). They found that levels of low-pI e-amylase mRNA, as evidenced by hybridization to clone E, increased with incubation in the absence of GA3. We observed the opposite pattern for low-pI eamylase m R N A in aleurone layers incubated in the absence of GA3. Treatment of aleurone layers with GA3 produced similar patterns for synthesis of the two
groups of e-amylase isoenzymes as it did for eamylase-mRNA levels (Fig. 2), as has been shown previously (Nolan et al. 1987). Neither group of isoenzymes was synthesized when the aleurone layers were incubated for extended periods of time in the absence of GA3 (lane 1). Synthesis of both isoenzyme groups was detectable by 8 h of GA3 treatment. Synthesis of high-pI e-amylase decreased after around 24 h of GA3 treatment, whereas synthesis of low-pI e-amylase continued
554
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
Fig. 2. Time course of the synthesis of ~amylase isoenzymes in barley aleurone layers in the presence of 10 .6 M GA3. Non-denaturing polyacrylamide gel electrophoresis of extracts of [35S]methionine-labelled aleurone layers was performed, followed by fluorography. Equal TCA-precipitable counts were added to each lane. The top a r r o w shows the location on the gel of the high-pI a-amylase isoenzymes and the l o w e r t w o a r r o w s show low-pI ~amylase isoenzymes. Duplicate samples were run for each of the GAa time points, with the hours of incubation in GA3 listed above the brackets
at a high rate even at 40 h of GA3 treatment. These data indicate that the rates of synthesis of c~-amylase isoenzymes are determined by the abundance of their respective mRNAs under these ~:onditions. Protease activity has been shown in several studies to be induced by GA3 in aleurone layers (Jacobsen and Varner 1967; Hammerton and Ho 1986). A GAa-inducible aleurone mRNA encoding a putative protease, termed aleurain, has been cloned by Rogers et al. (1985) and shown to have sequence homology to the mammalian thiol protease cathepsin H (Rogers et al. 1985; Mundy and Rogers 1986). Northern analysis for aleurain (G7i probe) produced a pattern of expression similar to that of low-pI c~-amylase (Fig. 1), but with a lower degree of induction of aleurain by GA3. Substantial amounts of aleurain mRNA were present in aleurone layers at the time when they were peeled away from the starchy endosperm (data not shown). We found that the abundance of this mRNA greatly decreased with time in the absence o f GA3, whereas there was a modest increase in aleurain mRNA after 12 h of incubation in the presence of GA3. One interpretation of these data is that a gibberellin or some other inductive factor is initially present upon isolation of aleurone layers
and declines with time in the absence of added GA3. Alcohol dehydrogenase (ADH1) has previously been shown to be constitutively expressed in barley aleurone layers in the absence of GA3 (Hanson and Jacobsen 1984). We also found continued expression of ADH1 in the absence of GA3. However, after approx. 8 h of GA3 treatment, ADH1 mRNA decreased drastically. At 24 h of incubation there were measurable levels of ADH1 mRNA in the absence of GA3, but very low levels in the presence of GA3. The PAPI-mRNA levels have previously been shown not to be affected by hormonal treatment in barley aleurone layers (Mundy and Rogers 1986). However, we observed that the abundance of PAPI mRNA decreased after around 16 h of GA3 treatment, and was greatly reduced after 24 h of GA3 incubation. Another transcript (pHVU-I) was also found to decrease in abundance after around 8 h of GA3 treatment. Two aleurone transcripts were found (pHVU-2 and pHVU-3) to be neither induced or suppressed by GA3. The levels of ribosomal RNA were also unaffected by incubation or GA3 treatment (Fig. 1). Baulcombe and Buffard (1983) showed that six mRNAs species (one of which was
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
555
Fig. 3. The effect of GA3 concentration on R N A levels in barley aleurone layers. Northern analysis was performed as described in the legend of Fig. 1. For lanes I 6, the aleurone layers were incubated for 8 h in the different hormone concentrations and for lanes 711 the aleurone layers were incubated for 16 h. For lane 1, 2-10 -5 M ABA was added to the incubation buffer and for lanes 2 and 7 neither GA3 or A B A was added. The different molarities of GA3 added to the other incubations are listed above each lane
a-amylase) were GAa-inducible and one unidentified m R N A was suppressed by GA3 in wheat. Our finding is, to the best of our knowledge, the first demonstration of suppression of an identified m R N A species by GA3 in barley aleurone layers.
Effects of GAs concentration of gene expression. Different sensitivities to GA3 concentration in the incubation medium were also observed for the different genes studied. A concentration of 1 0 - 9 M GA3 was adequate for induction of low-pI a-amylase mRNA, while l 0 - s M GA3 was required for
expression of high-pI 0c-amylase m R N A (Fig. 3). Muthukrishnan et al. (1983) found the same GA3concentration requirement, 10-8 M GA3, for the expression of high-pI 0~-amylase, but did not analyze levels of low-pI 0~-amylase mRNA. Aleurain m R N A showed basically the same hybridization pattern as low-pI e-amylase m R N A : it was induced by all four GA3 concentrations tested. A higher GA3 concentration was also required for synthesis of the high-pI a-amylase isoenzymes than was necessary for the low-pI ~-amylase-isoenzyme group (Fig. 4). This is in agreement with the
556
R.C. Nolan and T.-H.D. Ho : Hormone-regulatedgene expressionin barley aleurone layers
Fig. 4. The effectof GA3 concentration on the synthesisof a-amylase isoenzymes. Non-denaturing polyacrylamidegel electrophoresis was performed as described under Fig. 2. Lanes I 6 are for 8-h incubations and lanes 7-11 are for 16-h incubations. For lane 1, 2.10 -5 M ABA was added to the incubation buffer and for lanes 2 and 7 neither GA3 or ABA was added. The differentmolarities of GA3 added to the other incubations are listed above each Iane. ~-Amylaseisoenzymetocafionsare indicated by arrows on the right results of Jacobsen and Higgins (1982), who analyzed the accumulation of e-amylase-isoenzyme activities. Additionally, 10-9 M GA3 induced low-pI e-amylase m R N A , but low-pI e-amylase was not consistently synthesized at appreciable rates at this GA3 concentration. A higher GA3 concentration thus seems to be required for translation or posttranslational processing of low-pI e-amylase than is required for the accumulation of low-pI e-amylase m R N A for these time points. This pattern also seems to be true for the high-pI e-amylase isoenzymes. These data thus contrast with the previous two figures, which showed a strong correlation between the patterns of synthesis of e-amylase isoenzymes and the respective levels of e-amylase mRNA. Ribosomal R N A levels were again unaffected by the addition of ABA or GA3 to aleurone layers (Fig. 3). The ADH1 m R N A was present in the ABA-treated layers and the layers incubated 16 h in the absence of added hormones, but was not
found in any of the 8- and 16-h GA3-treated layers. Therefore, 10 -9 M GA3 was adequate to cause a decrease in ADH1 mRNA. The levels of pHVU-1 transcripts were not affected by 8 h of GA3 treatment but decreased in all of the 16-h GA3-treated layers (data not shown). The PAPI m R N A was also not affected by GA3 of ABA over 8 h of treatment, but was slightly less abundant in the 16-h GA3-treated layers. These data therefore corroborate the GAa time-course data; GA3 both positively and negatively regulates m R N A abundances, with some m R N A species being more highly regulated than others. Abscisic acid c o u n t e r a c t s the effects o f GA3 treatm e n t . Previous investigations have shown that
when ABA and GA3 are added simultaneously to aleurone layers, appearance of translatable e-amylase m R N A is greatly delayed (Mozer 1980; Higgins et al. 1982). The synthesis of e-amylase was even more negatively influenced by ABA addition.
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
557
Fig. 5. Addition of ABA counteracts the effects of GA3 on m R N A abundance in barley aleurone layers. N o r t h e r n analysis was performed as described in Fig. 1. For lanes 1-7, the aleurone layers were incubated for 8 h in the different hormone concentrations and for lanes 814 the aleurone layers were incubated for 16 h. For lanes I and 8, no hormone was added; 1 ~tM GA3 was added in all of the other incubations. The A B A molarities are listed above the appropriate lanes
It was suggested from these data that ABA might in some cases be exerting translational control on e-amylase expression. This possibility was reinvestigated for both groups of e-amylase isoenzymes by incubating aleurone layers in 1 0 - 6 M GA3 and varying ABA concentrations and then analyzing e-amylase-mRNA levels (Fig. 5) and e-amylase synthesis (Fig. 6). If extensive translational control were occurring, one would see a more drastic effect of ABA on protein synthesis than on m R N A levels for the e-amylase isoenzymes.
As shown in Figs. 5 and 6, high ABA concentrations prevented the expression of both e-amylase-isoenzyme groups at both the m R N A - and protein-synthesis levels. High ABA concentrations (10 -4 M and 10 -2 M ABA) counteracted the effects of 10 - 6 M GA3 (lanes 2, 3, 9, 10), whereas lower ABA concentrations were less effective. Since e-amylase synthesis and m R N A levels were affected to the same extent by ABA, no evidence for translational control by ABA was apparent. Addition of ABA affected the other m R N A s
558
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
Fig. 6. Abscisic acid inhibits the induction of ~-amylase-isoenzyme synthesis in barley aleurone layers by GA3. The incubation times were 8 h for lanes 1-7 and 16 h for lanes 8-14; 1 I.tM GA3 was added for lanes 2-7 and ~ 1 4 . The molarities of ABA for lanes 2-6 and ~ 1 3 are listed above these lanes. Nondenaturing polyacrylamide gel electrophoresis was performed as described in the legend for Fig. 2. The locations of c~-amylase isoenzymes are indicated by arrows on the right
tested in a manner consistent with its role as a GA3 antagonist. Expression of A D H 1 was suppressed by GA3 and high A B A concentrations overcame the effect of GA3. The levels of PAPI m R N A also decreased with 16 h of GA3 incubation, and high A B A concentrations blocked this effect of GA3. At 8 h of incubation, induction of aleurain m R N A was reduced by high A B A concentrations. However, with 16 h of incubation none of the A B A concentrations decreased aleurain expression. Discussion
Our results show that there is a great deal of variation in the responses of different genes to GA3 in barley aleurone layers. At least two classes of GA3-inducible genes are apparent, varying in response to GA3 concentration and to duration of GA3 treatment. Two GA3-suppressible genes were identified. They also exhibit quite different responses to GA3. The mechanisms by which these differences in
hormonal response occur remain to be ascertained. Stimulation of run-on transcription of a-amylase genes has been demonstrated by Jacobsen and Beach (1985), using nuclei isolated from GA3-responsive barley aleurone protoplasts. Similar results have been obtained in wild-oat (Avenafatua) protoplasts (Zwar and Hooley 1986). The large degree of induction of 0~-amylase and other genes by GA3 therefore seems to be at least partly the result of transcriptional activation. It is not however evident whether GA3 also increases the m R N A half-life of certain transcripts in barley aleurone layers. Estrogen has been shown to increase both the transcription rate and m R N A halflife of vitellogenin m R N A in Xenopus liver (Brock and Shapiro 1983), so hormonal regulation of m R N A stability is not without precedent. A hormonally-induced increase in m R N A half-life has not yet been demonstrated in a plant system. However, Flores and Tobin (1986) have suggested that benzyladenine exhibits a positive post-transcriptional regulation of chlorophyll a/b-binding protein m R N A in Lemna gibba. Differences in GA3
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
response may result from regulation at both the transcriptional and post-transcriptional levels. The decrease in the levels of certain m R N A s after GA3 treatments is the major new finding of this research. Investigation of the mode of action of GA3 in aleurone layers may eventually need to include gene suppression as a part of the integrated effects of this hormone. The mechanism of decrease in m R N A abundance of specific genes is completely unknown. It is not known, for instance, whether A D H I and PAPI genes are being transcribed in isolated aleurone layers. Both ADH1 and PAPI m R N A s are present in the aleutone layers of immature barley seeds (Hanson et al. 1984; Mundy and Rogers 1986), and their presence in mature aleurone layers could be the result of previous m R N A synthesis. Messenger-RNA degradation is probably involved in the decreases in m R N A levels, and GA3 treatment may cause either specific or general decreases in m R N A stability. The effects of GA3 and ABA on m R N A stability are currently being studied in more detail. It is apparent that GA3 and ABA regulate gene expression primarily by regulating m R N A abundance. This does not mean, however, that regulation of translation does not occur. A requirement of calcium for translation of high-pI a-amylase m R N A in barley aleurone layers has, in fact, been shown by Deikman and Jones (1986). After 24 h of GA3 treatment, total a-amylase synthesis accounts for 40-60% of [35S]methionine incorporation into protein, a-Amylase mRNA, however, only represents around 20% of total translatable m R N A under these conditions (Ho 1980). This provides some indication that c~-amylase is preferably translated in aleurone layers. Gibberellic-acidinduced cellular changes, e.g. increase in endoplasmic reticulum (Jones 1969a, b; Belanger etal. 1986), may be required for efficient translation of a-amylase mRNA. A higher GA3 level may be required for these processes than for synthesis of ~amylase mRNA. The control of gene expression by GA3 varies greatly among genes. Future work investigating GA3 receptors must take into account the complexities of the GA3 response. Abscisic acid seems to counteract GA3 at early regulatory steps, since it is able to block GA3 action for all of the genes studied. A crucial finding for understanding hormonal action in barley aleurone layers will be to ascertain the steps, if any, at which both hormones act. We wish to thank Dr. John Rogers for the use of cDNA clones isolated in his laboratory. This work was supported by a National Science Foundation grant DCB-8702299.
559
References Baulcombe, D.C., Buffard, D. (1983) Gibberellic-acid-regulated expression of e-amylase and six other genes in wheat aleurone layers. Planta 157, 493 501 Belanger, F.C., Brodl, M.R., Ho, T.-H.D. (1986) Heat shock causes destabilization of specific mRNAs and destruction of endoplasmic reticulum in barley aleurone cells. Proc. Natl. Acad. Sci. USA 83, 1354-1358 Brock, M.L., Shapiro, D.J. (1983) Estrogen stabilizes vitellogenin mRNA against cytoplasmic degradation. Cell 17, 319-325 Brown, P.H., Ho, T.-H.D. (1986) Barley aleurone layers secrete a nuclease in response to gibberellic acid. Plant Physiol. 82, 801-806 Callis, J_, Ho, T.-H.D. (1983) Multiple molecular forms of the gibberellin-induced e-amylase from the aleurone layer of barley seeds. Arch. Bioehem. Biophys. 224, 224-234 Chandler, P.M., Zwar, J.A., Jacobsen, J.V., Higgins, TJ.V., Inglis, A.S. (1984) The effects of gibberellic acid and abscisic acid on e-amylase mRNA levels in barley aleurone layer studies using an e-amylase cDNA clone. Plant Mol. Biol. 3, 407-418 Chrispeels, M.J., Varner, J.E. (1967) Hormonal control of enzyme synthesis: On the mode of action of gibberellic acid and abscisin in aleurone layers of barley. Plant Physiol. 42, 1008-1016 Church, G.M., Gilbert, W. (1984) Genomic sequencing. Proc. Natl. Acad. Sci. USA 81, 1991-1995 Deikman, J., Jones, R.L. (1986) Regulation of the accumulation of mRNA for e-amylase isoenzymes in barley aleurone. Plant Physiol. 80, 67~675 Dennis, E.S., Gerlach, W.L., Pryor, A.J., Bennetzen, J.L., Inglis, A., Llewellyn, D.L., Sachs, M.M., Ferl, R.J., Peacock, W.J. (1984) Molecular analysis of the Alcohol dehydrogenase (Adhl) gene of maize. Nucl. Acid. Res. 9, 3983-4000 Flores, S., Tobin, E.M. (1986) Benzyladenine modulation of the expression of two genes for nuclear-encoded chloroplast in Lemna gibba: Apparent post-transcriptional regulation. Planta 168, 340-349 Hammerton, R.W., Ho, T-H.D. (1986) Hormonal regulation of the development of protease and carboxypeptidase activities in barley aleurone layers. Plant Physiol. 80, 692-697 Hanson, A.D., Jacobsen, J.V., Zwar, J.A. (1984) Regulated expression of three Alcohol dehydrogenase genes in barley aleurone layers. Plant Physiol. 75, 573-581 Higgins, T.J.V., Jacobsen, J.V., Zwar, J.A. (1982) Gibberellic acid and abscisic acid modulate protein synthesis and mRNA levels in barley aleurone layers. Plant Mol. Biol. 1,191-215 Ho, T-H.D. (1980) Hormonal and genetic regulation of e-amylase synthesis in barley aleurone cells. In: Genome organization and expression in plants, pp. 147-158, Leaver, C.J., ed. Plenum, New York London Ho, T-H.D. (1983) Biochemical mode of action of abscisic acid. In: Abscisic acid, pp. 142170, Addicott, F.A., ed. Prager, New York Jacobsen, J.V., Beach, L.R. (1985) Control of transcription of e-amylase and rRNA genes in barley aleurone layer protoplasts by gibberellic acid and abscisic acid. Nature 316, 275-277 Jacobsen, J.V., Higgins, T.J.V. (1982) Characterization of the c~-amylase synthesized by aleurone layers of Himalaya barley in response to GA3. Plant Physiol. 70, 1647-1653 Jacobsen, J.V., Varner, J.E. (1967) Gibberellic acid induced synthesis of protease by isolated aleurone layers of barley. Plant Physiol. 42, 1596-1600
560
R.C. Nolan and T.-H.D. Ho: Hormone-regulated gene expression in barley aleurone layers
Jones, R.L. (1969a) Gibberellie acid and the fine structure of barley aleurone layers. 1. Changes during the lag phase of e-amylase synthesis. Planta 87, 119-133 Jones, R.L. (1969b) Gibberellic acid and the fine structure of barley aleurone layers. 2. Changes during the synthesis and secretion of e-amylase. Planta 88, 73-86 Maniatis, T., Fritsch, E.F., Sambrook, J. (1982) Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Habor, N.Y., USA Mans, R.J., Novelli, G.D. (1960) A convenient, rapid, and sensitive method for measuring the incorporation of amino acids into protein. Biochem. Biophys. Res. Commun. 3, 540-548 Mozer, T.J. (1980) Control of protein synthesis in barley aleurone layers by the plant hormones gibberellic acid and abscisic acid. Cell 20, 479-485 Mundy, J., Rogers, J.C. (1986) Selective expression of a probable amylase/protease inhibitor in barley aleurone cells: Comparison to the barley amylase/subtilisin inhibitor. Planta 169, 51-63 Muthukrishnan, S., Chandra, G.R., Albaugh, G.P. (1983) Modulation by abscisic acid and S-2-aminoethyl-L-cysteine of e-amylase mRNA in barley aleurone cells. Plant Mol. Biol. 2, 249-258
Nolan, R.C., Lin, L-S., Ho, T-H.D. (1987) The effect of abscisic acid on the differential expression of e-amylase isoenzymes in barley aleurone layers. Plant Mol. Biol. 8, 13-22 Rogers, J.C. (1985) Two barley e-amylase gene families are regulated differently in aleurone cells. J. Biol. Chem. 260, 3731-3738 Rogers, J.C., Dean, D., Heck, G.R. (1985) Aleurain: A barley thiol protease closely related to mammalian cathepsin H. Proc. Natl. Acad. Sci. USA 82, 6512-6516 Rogers, J.C., Milliman, C. (1984) Coordinate increase in major transcripts from the high-pI e-amylase multigene family in barley aleurone cells stimulated with gibberellic acid. J. Biol. Chem. 259, 1223-1240 Stuart, I.M., Loi, L., Fincher, G.B. (1986) Development of (1-3, 1-4)-fl-D-glucan endohydrolase in isolated scutella and aleurone layers of barley (Hordeum vulgate). Plant Physiol. 80, 310-314 Zwar, J.Z., Hooley, R. (1986) Hormonal regulation of e-amylase gene transcription in wild oat (Arena fatua L.) protoplasts. Plant Physiol. 80, 459-463
Received 9 November 1987; accepted 18 January 1988
