Plant Molecular Biology 2:189 198 (1983). 9 Martinus NijhofJ/Dr W. Junk Publishers, The Hague. Printed in the Netherlands.
D e v e l o p m e n t a l biochemistry of cottonseed embryogenesis and germination x v I . Analysis of the principal cotton storage protein gene family with cloned cDNA probes Glenn A. Galau l, Caryl A. Chlan & Leon Dure III*
Department of Biochemistry, University of Georgia, Athens, GA 30602, U.S.A. t present address: Department of Botany, University of Georgia, U.S.A.
Keywords: cotton storage proteins, closed cDNA, in vitro translation, colony hybridization, hybrid selection, hybrid-arrested translation
Abstract D N A s complementary to the m R N A s coding for the major cotton 48 kD and 52 kD storage proteins have been cloned and used to characterize the principal cotton storage protein gene family. The principal storage proteins are found to emanate from three subsets of genes that share some homology, as shown by c o m m o n antigenic determinants shared by the proteins themselves, but that are distinguishable by nucleic acid hybridization. A single sequence subfamily of 2.26 kb m R N A s codes for the 69 kD preproteins (precursors to the mature 48 kD proteins) and two sequence subfamilies of 1.96 kb m R N A s each code for 60 kD preproproteins (precursors to the mature 52 kD proteins). Hybrid arrested translation shows that cloned members of these three subfamilies hybridize only with the m R N A s of a single subfamily at moderate criterion. These three subfamilies comprise all of the principal storage protein m R N A s detectable by in vitro translation. With hybridization at low criterion, some homology has been detected between the two 1.96 kb m R N A families, although no homology has yet been detected between the 2.26 kb m R N A family and either of the two 1.96 kb m R N A families.
Abbreviations Poly (A) +, polyadenylated S SC, standard saline citrate, 0.15 M NaC1, 0.015 M trisodium citrate, sodium dodecyl sulfate, SDS, ethylene-diamine tetraacetic acid, EDTA, kb, kilobases, kilo Daltons kD,
Introduction The principal nutritional storage proteins that accumulate in the cotyledons of the cotton seed during embryogenesis are composed of two sets of isoelectric variants whose apparent molecular *To w h o m reprint requests should be addressed.
weights when mature are 48 and 52 kD (9). These proteins are initially synthesized as precursor sets of about 69 and 60 kD, and undergo considerable processing before they are deposited in their final form in the protein bodies of the tissue. The proteins of each molecular weight set have been considered members of a multigene family. The immunologic cross reactivity and the similarity in amino acid composition between the two molecular weight sets (9, 11) suggest that the two families are but subsets of a larger, more distantly related multigene family. T o explore the relationship among members of this putative family, the possible reiteration and clustering of their genes, and to follow their expression during cotyledon development, we have made use of cloned c D N A sequences constructed from storage protein mRNAs. We describe here the construction and identification of the c D N A probes, and present data indicating that the storage pro-
190 teins emanate from three related but distinct gene subsets. In a companion paper, the use of the cDNA probes in mapping the developmental time span and extent of expression of the storage protein genes is presented (13).
Methods and materials
100 m M NaCI, 0,1 m M E D T A (pH 7.6). Equi-molar quantities of excluded c D N A (>300 nucleotides in length) and Pst I-cleaved plasmid were mixed, heated3 m i n a t 6 5 ~ 1 h a t 5 5 ~ and 1 h a t 4 5 ~ and ethanol precipitated. The D N A was then used to transform E, coli SK 1590 as described (19). Yields were about 2 • l 0 4 transformants (>95% recombinant) per #g recombinant plasmid.
General methods
DNA restriction and labeling in vitro
Gossypium hirsutum, variety Coker 201, was greenhouse grown. The preparation of total and poly (A)+RNA (15), the in vivo labeling of cotton proteins (11), the in vitro translation of RNA in wheat germ extracts, the electrophoresis of protein in one and two dimensional Laemmli gels (9) and the immunological techniques used (9) have been described.
Transformed bacteria were grown to stationary phase in a nutrient medium with 20 ~g ml -j of selective antibiotic and crude plasmid D N A was prepared by a modified alkaline SDS lysis method (4). Further purification, when employed, was by isopycnic centrifugation in CsCl-ethidium bromide. The treatment of recombinant plasmid DNAs with restriction enzymes followed their supplier's recommendations. Nick translation of DNAs with [32P]dCTP was essentially as described (27), but with the inclusion of 100 #g ml 1 RNAse A (Worthington) when plasmid DNAs had not been purified by centrifugation in CsC1. Labeled inserts were prepared by nick translation of electrophoretically separated inserts or by nick translation of the entire plasmid with subsequent removal of non-insert radioactive D N A by preparative hybridization with plasmid D N A (Galau, in preparation). 32p-labeled c D N A used for filter hybridization was synthesized as described above except that the labeled deoxyribonucleotide triphosphate was reduced to 15 #M, and the R N A was pretreated with 5 m M methylmercuric hydroxide as described (26). Labeled DNAs were routinely sheared by sonication to 0.5 kb average length.
Cloning of cDNAs Double-stranded c D N A was prepared from young cotton embryo total poly (A) + R N A essentially as described (6, 32) with the inclusion of placental ribonuclease inhibitor (Biotech, Madison WI) at 103 units ml I. Reverse transcriptase (Life Sciences, Inc), repurified by CM-Sephadex chromatography (24), was used at ten units per ~g RNA. After second strand synthesis, reactions were deproteinized with phenol, desalted on Sephadex G100 in 5 m M a m m o n i u m acetate and evaporated to dryness. Hairpin structures were cleaved with 700 units ml I S-1 nuclease (P. L. Biochemicals) in 0.3 M NaC1, 0.03 M sodium acetate, 0.003 M zinc acetate (pH 4.5) for 1 h at 37 ~ C. The reactions were deproteinated, desalted on Sepharose CL-2B and concentrated by evaporation. The tailing of cDNA inserts with dC and the tailing of Pst I-cleaved pBR325 (5) with dG was performed in 140 m M cacodylic acid, 30 m M T r i s base(adjusted to pH 6.9 with KOH), 1 m M dithiothreitol, 100/ag ml -l bovine serum albumin, 1 m M COC12, 15 m M deoxynucleotide triphosphate, 5 n M 3' termini and 350 units ml 1terminal deoxynucleotide transferase (P. L, Biochemicals). Between 14 and 16 residues o f d C were added in 1 min at 17 ~ C and about the same number o f d G in 2 min at 37 ~ C. Reaction mixtures were deproteinated with phenol and chromatographed on Sepharose CL-6B in 20 m M Tris-HC1,
Electrophoresis of nucleic acids' and northern blotting RNA and DNA were electrophoresed in denaturing conditions in I 0 mMmethylmercuric hydroxideagarose gels (2) in modified E buffer (7) or in 3% or 6% formaldehyde-agarose gels (21). Acridine orange at 0.5 #g ml I was occasionally included in formaldehyde gels from which RNA was to be transferred. Gels were destained in 3% formaldehyde-phosphate buffer and photographed prior to transfer. Transfer of R N A from formaldehyde gels was directly onto nitrocellulose (Schleicher and
191 Schue11, B-83) in 10X SSC by the method of Southern (28). After drying the filters were baked in vacuo at 80 ~ C for 90 min (30). Electrophoresis of D N A in non-denaturing conditions was in agarose or polyacrylamide in 90 m M Tris-borate, 3 m M EDTA, pH 8.3 (1).
for l0 min each in SSC, 0.1% sodium pyrophosphate, 0.1% SDS, at the SSC concentration and temperature of hybridization. Dried filters were exposed to fogged (20) K o d a k X - A R film with Dupont Cronex Lightening Plus Screens. Hybrid-selected and hybrid-arrested translation
Colony replicas
For screening large numbers of colonies, transformants (95% of which proved to be recombinant) were transferred with a toothpick to a chilled 23 • 23 cm petri plate containing nutrient broth and chloramphenicol (20 #g m1-1) in 2% agar, and grown 16 h at 37 ~ C. Colonies were replica-plated via a velvet cloth onto nitrocellulose on new plates and grown 11 h at 37 ~ C. Plasmid D N A was amplified (16, 18) for 16 h at 37 ~ after transfer of the filter to new plates containing 40 #g ml I spectinomycin. Colonies on the replica filters were lysed and prepared for hybridization by slight modifications of several methods (16, 29). Filter hybridization
The hybridization of all nitrocellulose filters was performed essentially as described by Maniatis et al. (22). Filters were wetted in 4X SSC, 0.1% SDS for 15 min at 68 ~ and incubated at 68 ~ for several hours in a solution containing 4X SSC, 10 m M sodium phosphate buffer, 10 m M E D T A , 0.1% SDS and 5X Denhardt's solution (pH 7.5). Denhardt's solution is 0.2% bovine serum albumin, 0.2% Ficoll (4 • 105 daltons) and 0.2% polyvinylpyrrolidone (3.7 • 105 daltons) (8). Filters were next incubated in hybridization buffer several hours and then in fresh hybridization buffer containing the radioactive D N A probe. Hybridization buffer contained 10 m M sodium phosphate buffer, 10 m M EDTA, 0.1% sodium pyrophosphate, 0.1% SDS, 100 #g ml I denatured, sonicated herring sperm D N A (0.5 kb), 10% dextran sulfate (5 X 105 daltons) (31), and from 1 to 6X SSC (pH 7.5). Purified formamide at 50% (v:v) was also present in some instances. The concentration of SSC and the hybridization temperature are detailed in the Figure Legends. After incubation with 32P-labeled D N A probe for 12 48 h, the filters were washed at the hybridization temperature 3 times for 10 min each with hybridization buffer and finally 4 times
To equate c D N A clones with specific mRNAs, recombinant plasmids were digested with Pst I which liberated the cloned c D N A from the plasmid vehicle. The D N A was then alkali denatured, neutralized, and bound to nitrocellulose discs as described by McGrogan et al. (23). The hybridization of young embryo poly(A) + R N A with the D N A filters, and subsequent thermal elution of the complexed R N A were also as described (23). The eluates from the first two washes at each temperature were pooled and ethanol precipitated two times. These released R N A s were resuspended in water and translated in the wheat germ in vitro translation system as previously described (11). Hybrid-arrested translation experiments were essentially as described by Paterson et al. (25) with the following modifications. Ten #g of Pst I-digested recombinant plasmid D N A and 40 #g of total R N A from young embryo cotyledons were mixed in 5 ~1 of water, heat denatured and quick chilled. Mixtures were b r o u g h t up to a final volume of 100 ~1 of 80% formamide (deionized and 2 times recrystalized), 10 m M Pipes - K O H , 0.4 M NaC1 (pH 7.5), and incubated at 40 ~ C for 1 h. The reactions were stopped by the addition of 0.8 ml of ice cold water and each reaction was divided into two fractions. One fraction was immediately ethanol precipitated, and the other was heat denatured by boiling for 30 s, quick chilled and ethanol precipitated. The fractions were ethanol precipitated several times prior to the translation of an aliquot in the wheat germ translation system (11).
Results
Immunochemical discrimination between storage protein subfamilies
The one dimensional electrophoretic display of the abundant proteins of embryo cotyledons at a developmental stage of m a x i m u m storage protein
192
Fig. 1. One dimensional gel electrophoresis of cotton cotyledon proteins. Lane 1: stained proteins of immature embryocotyledons. Lane 2: proteins synthesized in vitro from RNA extracted from immature embryocotyledons.Lane3: proteins synthesized in vivo by immature embryocotyledons.Lane4: in vitro synthesized proteins immunoprecipitated in a 5-min incubation by antibodies prepared against the 52 kD storage proteins. Lane 5: in vitro synthesized proteins immunoprecipitated in a 5-min incubation by antibodies prepared against the 48 kD storage proteins. Size estimates are in kD.
synthesis is presented in Fig. 1, lane 1. In lane 3 are displayed proteins labeled in v i v o when identical embryos are excised from the developing cotton boll and incubated in radioactive amino acids for a long period (24 h). It is clear that a long exposure of the tissue to the labeled amino acids yields a synthesis profile that mimics the concentration of the extant proteins of the tissue. This is not unexpected since these abundant proteins are chiefly the seed storage proteins and lectins that presumably do not
turn over during embryogenesis. When total or poly (A) + RNA extracted from the cotyledons of equivalent age embryos is translated in the wheat germ system, the profile of in v i t r o synthesis obtained is shown in lane 2. The protein species marked on this figure introduce the several intermediates thought to be involved in the formation of the mature storage proteins. Actually all of the intermediate and mature forms of these proteins are comprised of sets of isoelectric variants (9, 11); however, since no other abundant proteins with these apparent molecular weights occur in the cotyledon tissue, one dimensional gel electrophoresis is adequate up to a point for describing storage protein synthesis in these embryos. F r o m the data previously published (10, 11) we have postulated that the 69 kD initial translation products (lane 2) give rise to long-lived intermediates of 67 kD through the loss of a signal peptide during the movement of the initial translates to their deposition in the protein body. The 67 kD intermediates are observable on stained gels (lane 1) and on gels showing in v i v o synthesis (lane 3). We have postulated that these intermediates are slowly cleaved to the 48 kD mature proteins and a smaller protein set (not identified). It was also postulated analogously that the 60 kD initial translation products seen in lane 2 lose a signal peptide. These intermediates are not seen in stained gels or gels of in v i v o synthesis products because they are rapidly processed in v i v o to a higher apparent molecular weight set of intermediates of 70 kD by glycosylation. A very slow cleavage of this set was postulated to yield the glycosylated 52 kD protein set and a smaller protein set (not identified). Since the initial translation products lose a signal peptide and subsequently are cleaved to low molecular weight forms, they can be considered preproproteins. All the designated proteins in Fig. 1 (both in v i v o and in v i t r o labeled species and the stainable species) are precipitable by antibodies prepared against either the mature 52 kD species or the mature 48 kD species (11). Furthermore, the amino acid compositions of these two molecular weight sets are strikingly similar (9). For these reasons all the isoelectric variants of the 69 and 60 kD initial translation products are considered members of a multigene family indistinguishable by routine immunochemical procedures. However, since there are differences in apparent molecular weight and in
193 subsequent glycosylation between these two sets, the extent of homology/non-homology between the two sets of genes and their proteins was pursued. The first indication of major sequence differences between the 69 and 60 kD proteins was the fact that antibody to the 48 kD mature proteins precipitates ahnost exclusively the in vitro synthesized 69 kD preproproteins if the antigen/antibody incubation is limited to 5 min (Fig. 1, lane 5). Antibody to the 52 kD mature proteins does not show this short term selectively (lane 4), nor does either antibody when the incubation with the translation reaction mixture is carried out overnight. This more rapid reaction of the 48 kD antibody with the 69 kD preproproteins reinforces the proposition that the 48 kD mature proteins are derived from the 69 kD set. Identification o f storage protein messenger R N A s Both in vitro translation (12) and c D N A - m R N A hybridization kinetics (14) suggest that storage protein m R N A s together comprise a large fraction (>30%) of the young embryo mRNA. Thus these mRNAs should be resolved as obvious bands upon electrophoresis of poly(A) + RNA. Two m R N A bands are indeed visible in poly(A) + R N A upon electrophoresis in formaldehyde-agarose or methylmercury-agarose, each comprising about 15% of the total non-rRNA ethidium bromide-binding material (Fig. 2, lane 2). Their sizes are 1.96 _+0.02 kb and 2.26 + 0.02 kb (N = 6) as measured in either gel system in a wide variety of agarose concentrations using cotton rRNA (Fig. 2, lane 1) or cotton r R N A and E. coli r R N A as size markers (Fig. 2, lane 3). Besides their abundance in young embryo mRNA, several other considerations suggest these m R N A s are the storage protein mRNAs. The abundance of these two mRNAs in other developmental stage RNAs, as seen on stained gels of total poly (A) + R N A (data not shown) is c6rrelated with the storage protein m R N A content deduced from in vitro translation (12) and from c D N A - m R N A hybridization kinetics (14). Since the initial translation precursors are of two molecular weight classes and are structurally different though related as judged by immunological criteria (Fig. 1), one might expect two m R N A size species, one each for the 60 kD and 69 kD protein size classes. The 1.96 kb and 2.26 kb m R N A s seen in Fig, 2 exceed
Fig. 2. Identification of two abundant mRNAs in young embryo poly(A)+ RNA. RNAswereelectrophoresed in 3% formaldehyde-2-2% agarose gels and stained with ethidium bromide. Lane h 0.8 #g young embryo total RNA. Lane 2:2.5 #g young embryototal poly(A)+ RNA. Lane3: sameas lane2, but with0.6 ,ug E. coli rRNA included. Size estimates are in kb.
by about 0.4 kb the size required to code for these proteins. Construction and identification o f cloned storage protein c D N A s Our approach to obtaining eDNA clones for these m R N A S was to clone double-stranded cDNA made from young embryo m R N A , the stage at which they are at maximum concentration (12, 14), and to screen the recombinant colonies with eDNA from this developmental stage and a stage at which these m R N A s are greatly reduced. Colonies giving strong radioactive signals with the homologous c D N A probe but undetectable signals with the heterologous c D N A probe were used for the isolation of putative storage protein cDNAs. In the synthesis of double-stranded cDNA from young embryo cDNA, those double-stranded molecules of a discrete size and abundancy (shown by electrophoresis) sufficient to carry storage protein sequences were found to have internal Eco RI sites
194 b u t no P s t I sites. T o c o n s t r u c t r e c o m b i n a n t c D N A p l a s m i d s c a r r y i n g s t o r a g e p r o t e i n sequences whose inserts c o u l d be liberated intact f r o m the p l a s m i d vehicle, t o t a l y o u n g e m b r y o d o u b l e - s t r a n d e d c D N A was tailed with dC a n d a n n e a l e d to d G tailed Pst I-cleaved pBR325 as described in M e t h ods a n d m a t e r i a l s a n d used to t r a n s f o r m E. coli. N i t r o c e l l u l o s e replicas of a p p r o x i m a t e l y 1 700 rec o m b i n a n t colonies were h y b r i d i z e d with the two r a d i o a c t i v e c D N A p r e p a r a t i o n s m e n t i o n e d above. I n the first case, t o t a l y o u n g e m b r y o c D N A r e a c t e d s t r o n g l y with a b o u t 35% of the colonies, as one w o u l d p r e d i c t if such colonies c o n t a i n e d storage p r o t e i n sequences. T h a t these a b u n d a n t sequences are, in fact, y o u n g e m b r y o - s p e c i f i c was d e m o n s t r a t e d by the failure of dry seed c D N A to react with these colonies u n d e r equivalent h y b r i d i z a t i o n c o n d i t i o n s . S t o r a g e p r o t e i n m R N A sequences are r e d u c e d several h u n d r e d - f o l d at the dry seed stage (14).
T h e p l a s m i d inserts of a b o u t 270 of the y o u n g embryo-specific, strongly h y b r i d i z i n g colonies were sized after Pst I digestion. As predicted, less t h a n 2% of these plasmids a p p e a r e d to contain a Pst I site in the insert. I n 65 of the p l a s m i d s the inserts were l o n g e r t h a n 1.2 kb. T o detect the n u m b e r of separate sequences present in this p o o l of y o u n g embryo-specific, long i n s e r t - c o n t a i n i n g r e c o m b i n a n t s , these colonies were t r a n s f e r r e d to new plates and screened with r a n d o m l y selected m e m b e r s of the pool. F i g u r e 3 shows the r e a c t i o n of these colonies with three p l a s m i d inserts. Three groups of h y b r i d i z ing colonies were detected which t o g e t h e r c o m p r i s e m o r e t h a n 95% of the pre-selected colonies. In fairly extensive tests, not s h o w n here, no p l a s m i d insert was f o u n d to hybridize with m o r e t h a n one g r o u p of colonies at a m o d e r a t e criterion. In the clone identification tests described below, several r e p r e s e n t a tives f r o m each g r o u p were h y b r i d i z e d to y o u n g e m b r y o R N A with similar results. F o r the sake of
Fig. 3. Autoradiograph of colony hybridization of selected recombinant colonies with three cDNA plasmid inserts at two different criteria. Inserts were separated from the plasmid DNA before hybridization. The specific radioactivity of the insert DNA and exposure times were roughly the same in alle cases. Hybridization and filter washing conditions were l x SSC at 68 o C for (A C), and 6X SSC at 55 ~ for (D-F). Plasmid inserts used are identified on the left.
195 simplicity we present results only for one plasmid from each plasmid group in all of these tests. These are indicated in Fig. 3 and are C-72, C-94 and C-134.* Their insert lengths are about 2.3 kb, 1.9 kb, and 2.1 kb, respectively. A summation of the number of recombinant colonies carrying the young embryo c D N A library that hybridized with each of these plasmid groups suggests that sequences homologous to C-72, C-94 and C-131 comprise about 16, 12 and 6% of young embryo poly(A) + mRNA, respectively. The concentration of these sequences in total cotyledon m R N A during embryogenesis is dealt with in the companion paper (13). In Fig. 4 the three representative plasmids were hybridized to northern blots of electrophoretically separated young embryo total poly(A) + RNA. C-72 hybridized with the abundant 2.26 kb m R N A and both C-94 and C-134 hybridized with the abundant 1.96 kb mRNA. The proteins for which these m R N A s code were determined by the hybrid-select technique which involved translation in vitro of RNAs which hybridized with immobilized plasmid carrying one of the three inserts after recovery by thermal elution. In Fig. 5, mRNAs that hybridized with C-72 proved to code exclusively for the 69 kD proteins and both C-94 and C-134 hybridized with mRNAs that code exclusively for 60 kD storage proteins, as would be predicted by the northern blot results of Fig. 4. These products synthesized with hybrid-selected m R N A s have not been immuno-precipitated with anti-storage protein antibodies. However, their identity has been established by their behavior in one and two dimensional electrophoresis (data not shown). Curiously, the apparent melting temperatures of the three m R N A - D N A hybrids are not the same, the 2.26 kb m R N A - D N A hybrids melting about 10 ~ lower than the 1.96 kb m R N A - D N A hybrids. As a final demonstration that the three plasmid inserts do, in fact, represent three distinct subfamilies of storage protein genes, hybrid-arrested translation (25) was carried out with each of the three plasmids. In Fig. 6, C-72 is shown to arrest completely the translation of all the mRNAs coding for the 69 kD storage proteins, whereas neither C-94 nor C-134 alone completely arrest translation of *The precise notation of these plasmids is pGhSP C-72, pGhSP C-94, and pGhSP C-134.
Fig. 4. Autoradiographof hybridization of RNA with the three
plasmids used in Fig. 3. Young embryo poly(A)+ RNA was electrophoresed in a 6% formaldehyde-l.8% agarose gel, the RNAs blotted onto nitrocellulose paper whichwas then cut into strips corresponding to each electrophoresis lane. Strips were hybridized with one of the radioactive plasmids as indicated at the bottom of the figure. Plasmid specific radioactivities and autoradiograph exposure times were roughly equivalent. Hybridization and post-hybridization washing conditions were 4X SSC at 68 ~C. Sizes in kb are of cotton rRNAs and mRNAsthat were visualized in an adjacent lane by ethidium bromide staining.
m R N A s coding for the 60 kD storage protein. However, when present together (each at one-half the concentration as when present alone) they completely arrest the translation of these mRNAs. In all experiments the translation of the mRNAs arrested by the plasmid is restored upon denaturation of the m R N A - D N A hybrids prior to translation in vitro. These experiments have been performed at several R N A to D N A ratios with identical results (data not shown); therefore, the complete arrest of m R N A s coding for the 60 kD protein only with both C-94 and C-134 together is not the result of incomplete saturation of complementary mRNAs with either plasmid alone. Thus, at the hybridization criterion used here and in the colony hybridization experi-
196
Fig. 5. Autoradiograph of electrophoretically separated proteins synthesizedin vitro from rnRNA selected by hybridization with each of the three plasmids used in Fig. 3. Lane h proteins synthesized in vitro from young embryo cotyledon total RNA (control). Lanes 2-4: proteins synthesizedfrom RNA hybridized with plasmid C-72 and thermally released by incubation at 55 ~C (lane 2), 65 ~C (lane 3) and 75 ~C (lane 4). Lanes 5-7: as in lanes 2-4 using plasmid C-94. Lanes 8 10: as in lanes 2-4 using plasmid C-134. Apparent sizes in kD of principal storage protein preproproteins are indicated on left.
m e n t s in Fig. 3B, C, the C-94 a n d C-134 groups of plasmid inserts represent two distinct sets of sequences, b o t h of which code for subfamilies of 60 k D storage proteins. There are no other m R N A subfamilies for these proteins since, at the criterion used, these three groups of cloned c D N A s together arrest all m R N A sequences coding for 69 and 60 kD storage proteins. The issue of relatedness of these m R N A sequences has been p u r s u e d in some detail by m e a s u r i n g the h y b r i d i z a t i o n of colonies, the h y b r i d i z a t i o n of R N A on n o r t h e r n blots, and in hybrid-arrest and hybrid-select t r a n s l a t i o n carried out with m a n y cloned c D N A s u n d e r varied conditions. At no criterion e m p l o y e d (as low as 6 • SSC, at 55 ~ C, or 50% f o r m a m i d e , 4 X SSC, at 32 ~ did sequences present in 2.26 kb m R N A a n d 1.96 kb m R N A p l a s m i d inserts show a n y cross reaction (e.g. Fig.
Fig. 6. Autoradiograph of electrophoretically separated proteins synthesized in vitro from mRNA after hybridization with
each of the three plasmids used in Fig. 3. Lane 1: as in Fig. 5. Lanes 2, 4, 6: in vitro synthesis after incubation with plasmids, C-72, C-94 and C-134, respectively. Lane 8: after incubation with both C-94 and C-134. Lanes 3, 5, 7, 9 are of in vitro synthesis after mRNAs that were hybridized to the plasmids were thermally released by heating the incubation mixture to 100 ~
3D F), However, h o m o l o g y has been detected between the two 1.96 kb m R N A families via c o l o n y h y b r i d i z a t i o n at a 6 X SSC, at 55 o C, as s h o w n by a close e x a m i n a t i o n of Fig. 4 E - F .
Discussion T h e fact that the c o t t o n storage proteins are encoded in three sets of m R N A s rather than two came as a surprise. O n e a n d two d i m e n s i o n a l gel electrophoresis of the proteins suggest two sets as do the i m m u n o - p r e c i p i t a t i o n data which show only a kinetic d i s t i n c t i o n between the antigenicity of the two
197 molecular weight groups of proteins (Fig. 1). Initial hybrid selection experiments were first performed with plasmids C-72 a n d C-94 a n d were initially interpreted as indicating only two sets. Only when the screening of r e c o m b i n a n t colonies c o n t a i n i n g selected y o u n g e m b r y o c D N A inserts with r a n d o m ly selected m e m b e r s of that p o p u l a t i o n (such as in Fig. 3) was a third set of a b u n d a n t m R N A s represented by C-134) considered to be a possibility. F u r t h e r , hybrid-select experiments using C-134 showed that m R N A s for the 60 k D proteins were also selected by this plasmid (Fig. 5). Finally, the hybrid-arrest experiments revealed that both C-94 a n d C-134 were required to arrest fully the translation of the 60 k D proteins (Fig. 6). A m o n g l e g u m i n o u s species there are several examples of storage proteins that differ in molecular weight, in degree of glycosylation, a n d that show a n u m b e r of isoelectric variants, yet belong to a single multigenic family. All the vicilins of Phaseolus vulgaris are closely related as s h o w n recently by the s e q u e n c i n g of c D N A clones representing each varia n t (17). In Glycine max the different vicilins show evidence for a c o m m o n genetic b a c k g r o u n d by their i m m u n o c h e m i c a l cross reactivity. However, this h o m o l o g y is not manifested in the cross hybridizability of c D N A clones representing the various vicilin groups (3), much the same situation as we find for the cotton proteins. I n c o t t o n the degree of h o m o l o g y between all three gene subfamilies is still unclear. All three sets of proteins share antigenic sites, yet c D N A s for the 2.2 a n d 1.9 kb m R N A s do n o t hybridize a n d the c D N A s for the two 1.9 kb m R N A subfamilies hybridize only at a low criterion. These relationships can only be uncovered by c D N A sequencing as has been d o n e for the Phaseolus a n d Glycine proteins. The reiteration of these sequences, their g e n o m i c o r g a n i z a t i o n a n d analysis of their possible evolut i o n a r y relationships awaits the sequencing of gen o m i c clones.
Acknowledgements T h e technical assistance of J a n a Pyle a n d Steve Morris is gratefully acknowledged. This work was supported in part by grants from the N a t i o n a l Institutes of Health, C o t t o n Inc., and the C h e v r o n Research F o u n d a t i o n .
References 1. Air, AM, Sanger, F, Coulson, AR, 1976. Nucleotide and amino acid sequences of gene G of ~X174. J. Mol. Biol. 108:519 533. 2. Bailey,JM & Davidson, N, 1976. Methylmercuryas a reversible denaturing agent for agarose gel electrophoresis. Anal. Biochem. 70:75-85. 3. Beachy, RN, Doyle, J J, Ladin, BF & Schuler, MA (in press). Structure and expression of genes encoding the soybean 7S seed storage proteins. In: O Ciferri & LS Dure, (eds.) The Structure and Function of Plant Genomes. Plenum Press, New York, N.Y. 4. Birnboim, HC & Doly, J, 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucl. Acid Res. 7:1513-1523. 5. Bolivar, F, 1978. Construction and characterization of new cloning vehicles. III Derivativesof plasmid pBR322 carrying unique Eco RI sites for selection of Eco RI generated recombinant DNA molecules. Gene 4:12l- 136. 6. Buell, GN, Wickens, MP, Payvar, F & Schimke, RT, 1978. Synthesis of full length cDNAs from four partially purified oviduct mRNAs. J. Biol. Chem. 253:2471-2482. 7. Chandler, PM, Rimkus, D & Davidson, N, 1979. Gel electrophoretic fractionation of RNAs by partial denaturation with methylmercurichydroxide. Anal. Biochem.99:200-206. 8. Denhardt, DT, 1966. A membrane-filter technique for the detection of complementary DNA. Biochem. Biophys. Res. Comm. 23:641-646. 9. Dure III, LS & Chlan, CA, 1981. Developmental biochemistry of cottonseed embryogenesisand germination XII Purification and properties of principal storage proteins. Plant Physiol. 68:180-186. 10. Dure IlI, LS, Chlan, CA & Galau, GA (in press). Cottonseed storage proteins as a tool for developmental biology. In: O Ciferri & LS Dure, (eds.). The Structure and Function of Plant Genomes. Plenum Press, New York, N.Y. 11. Dure Ill, LS & Galau, GA, 1981. Developmentalbiochemistry of cottonseed embryogenesisand germination X I11Regulation of biosynthesis of principal storage proteins. Plant Physiol. 68:187-194. 12. Dure Ill, LS, Greenway, SC & Galau, GA, 1981. Developmental biochemistry of cottonseed embryogenesisand germination: changing messengerribonucleic acid populations as shown by in vitro and in vivo protein synthesis. Biochemistry 20:4162-4168. 13. Dure Ill, LS, Pyle, JB, Chlan, CA, Baker, JD & Galau, GA (companion paper) Developmental biochemistry of cottonseed embryogenesis and germination XVII Developmental expression of genes for the principal storage proteins. (Submitted to Plant Mol. Biol.). 14. Galau, GA & Dure III, LS, 1981. Developmental biochemistry of cottonseed embryogenesisand germination: changing messengerribonucleic acid populations as shown by reciprocal heterologous complementary deoxyribonucleic acidmessenger ribonucleic acid hybridization. Biochemistry 20:4169 4178. I 15. Galau, GA, Legocki, AB, Greenway, SC & Dure II1, LS, 1981. Cotton messenger RNA sequences exist in both po-
198 lyadenylated and nonpolyadenylated forms. J. Biol. Chem. 256:2551 2560. 16. Grunstein, M & Hogness, DS, 1975. Colony hybridization: a method for the isolation of cloned DNAs that contain a specific gene. Proc. Natl. Acad. Sci. USA 72:3961 3965. 17. Hall, TC, Slightom, JL, Ersland, DR, Murray, MG, Hoffman, LM, Adang, MJ, Brown, JWS, Ma, Y, Matthews, JA, Cramer, JH, Barker, RF, Sutton, DW, Kemp, JD. Phaseolin: nucleotide sequence explains molecular weight and charge heterogeneity of a small multigene family and also assists vector construction for gene expression in alien tissue. In Cifferri, O & Dure, LS, eds. The structure and function of plant genomes. Plenum Press, New York, N.Y. (in press). 18. Hanahan, D & Meselson, M, 1980. Plasmid screening at high colony density. Gene 10:63-67. 19. Kushner, SR, 1978. An improved method for transformation of Escherichia coli with cnlEl derived plasmids. In Boyer, HW & Nicosia, S, eds. Genetic Engineering. pp. 17 23. Elsevier-North Holland Biomedical Press, Amsterdam. 20. Laskey, RA & Mills, AD, 1975. Quantitative film detection of 3 H and 14C in polyacrylamide gels by fluorography. Eur. J. Biochem. 56:335 341. 21. Lehrach, H, Diamond, D, Wozney, JM & Boedtker, H, 1977. RNA molecular weight determinations by gel electrophoresis under denaturing conditions, a critical reexamination. Biochemistry 16:4743-4751. 22. Maniatis, T, Hardison, RC, Lacy, E, Lauer, J, O'Connell, C, Quon, D, Sin, GK & Efstratiadis, A, 1978. The isolation of structural genes from libraries of eucaryotic DNA. Cell 15:687 701. 23. McGrogan, M, Spector CJ, Halbert P & Raskas H J, 1979. Purification of specific adenovirus 2. RNAs by preparative hybridization and selective thermal elution. Nucleic Acid Res. 6:593 607. 24. Myers, JC, Ramirez, F, Kacian, DL, Flood, M, & Spiegel-
man, S, 1980. A simple purification of avian myeloblastosis virus reverse transcriptase for full-length transcription of 35 S RNA. Anal. Biochem. 101:88 96. 25. Paterson, BM, Roberts, BE & Kuff, EL, 1977. Structural gene identification mapping by DNA. mRNA hybrid-arrested cell-free translation. Proc. Natl. Acad. Sci. USA 74: 4370 4374. 26. Payvar, F & Schimke, RT, 1979. Methylmercury hydroxide enhancement of translation and transcription of ovalbumin and conalbumin mRNAs. J. Biol. Chem. 254:7636-7642. 27. Rigby, PWJ, Dieckmann, M, Rhodes, C & Berg, P, 1977. Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J. Mol. Biol. 113:237-251. 28. Southern, EM, 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503-517. 29. Thayer, RE, I979. An improved method for detecting foreign DNA in plasmids of Escherichia coli. Anal. Biochem. 98:60-63. 30. Thomas, PS, 1980. Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc. Natl. Acad. Sci. USA 77:5201-5205. 31. Wahl, GM, Stern M & Stark, GR, 1979. Efficient transfer of large DNA fragments from agarose gels to diazobenzyloxymethyl-paper and rapid hybridization by using dextran sulfate. Proc. Natl. Acad. Sci. USA 76:3683 3687 32. Wickens, MP, Buell, GN & Schimke, RT, 1978. Synthesis of double-stranded DNA complementary to lysozyme, ovamucoid, and ovalbumin mRNAs. Optimization for full length second strand synthesis by Escherichia coli DNA polymerase I. J. Biol. Chem. 253:2483-2495. Received 24 May 1983; in revised form and accepted 2 August 1983.
D e v e l o p m e n t a l biochemistry of cottonseed embryogenesis and germination x v I . Analysis of the principal cotton storage protein gene family with cloned cDNA probes Glenn A. Galau l, Caryl A. Chlan & Leon Dure III*
Department of Biochemistry, University of Georgia, Athens, GA 30602, U.S.A. t present address: Department of Botany, University of Georgia, U.S.A.
Keywords: cotton storage proteins, closed cDNA, in vitro translation, colony hybridization, hybrid selection, hybrid-arrested translation
Abstract D N A s complementary to the m R N A s coding for the major cotton 48 kD and 52 kD storage proteins have been cloned and used to characterize the principal cotton storage protein gene family. The principal storage proteins are found to emanate from three subsets of genes that share some homology, as shown by c o m m o n antigenic determinants shared by the proteins themselves, but that are distinguishable by nucleic acid hybridization. A single sequence subfamily of 2.26 kb m R N A s codes for the 69 kD preproteins (precursors to the mature 48 kD proteins) and two sequence subfamilies of 1.96 kb m R N A s each code for 60 kD preproproteins (precursors to the mature 52 kD proteins). Hybrid arrested translation shows that cloned members of these three subfamilies hybridize only with the m R N A s of a single subfamily at moderate criterion. These three subfamilies comprise all of the principal storage protein m R N A s detectable by in vitro translation. With hybridization at low criterion, some homology has been detected between the two 1.96 kb m R N A families, although no homology has yet been detected between the 2.26 kb m R N A family and either of the two 1.96 kb m R N A families.
Abbreviations Poly (A) +, polyadenylated S SC, standard saline citrate, 0.15 M NaC1, 0.015 M trisodium citrate, sodium dodecyl sulfate, SDS, ethylene-diamine tetraacetic acid, EDTA, kb, kilobases, kilo Daltons kD,
Introduction The principal nutritional storage proteins that accumulate in the cotyledons of the cotton seed during embryogenesis are composed of two sets of isoelectric variants whose apparent molecular *To w h o m reprint requests should be addressed.
weights when mature are 48 and 52 kD (9). These proteins are initially synthesized as precursor sets of about 69 and 60 kD, and undergo considerable processing before they are deposited in their final form in the protein bodies of the tissue. The proteins of each molecular weight set have been considered members of a multigene family. The immunologic cross reactivity and the similarity in amino acid composition between the two molecular weight sets (9, 11) suggest that the two families are but subsets of a larger, more distantly related multigene family. T o explore the relationship among members of this putative family, the possible reiteration and clustering of their genes, and to follow their expression during cotyledon development, we have made use of cloned c D N A sequences constructed from storage protein mRNAs. We describe here the construction and identification of the c D N A probes, and present data indicating that the storage pro-
190 teins emanate from three related but distinct gene subsets. In a companion paper, the use of the cDNA probes in mapping the developmental time span and extent of expression of the storage protein genes is presented (13).
Methods and materials
100 m M NaCI, 0,1 m M E D T A (pH 7.6). Equi-molar quantities of excluded c D N A (>300 nucleotides in length) and Pst I-cleaved plasmid were mixed, heated3 m i n a t 6 5 ~ 1 h a t 5 5 ~ and 1 h a t 4 5 ~ and ethanol precipitated. The D N A was then used to transform E, coli SK 1590 as described (19). Yields were about 2 • l 0 4 transformants (>95% recombinant) per #g recombinant plasmid.
General methods
DNA restriction and labeling in vitro
Gossypium hirsutum, variety Coker 201, was greenhouse grown. The preparation of total and poly (A)+RNA (15), the in vivo labeling of cotton proteins (11), the in vitro translation of RNA in wheat germ extracts, the electrophoresis of protein in one and two dimensional Laemmli gels (9) and the immunological techniques used (9) have been described.
Transformed bacteria were grown to stationary phase in a nutrient medium with 20 ~g ml -j of selective antibiotic and crude plasmid D N A was prepared by a modified alkaline SDS lysis method (4). Further purification, when employed, was by isopycnic centrifugation in CsCl-ethidium bromide. The treatment of recombinant plasmid DNAs with restriction enzymes followed their supplier's recommendations. Nick translation of DNAs with [32P]dCTP was essentially as described (27), but with the inclusion of 100 #g ml 1 RNAse A (Worthington) when plasmid DNAs had not been purified by centrifugation in CsC1. Labeled inserts were prepared by nick translation of electrophoretically separated inserts or by nick translation of the entire plasmid with subsequent removal of non-insert radioactive D N A by preparative hybridization with plasmid D N A (Galau, in preparation). 32p-labeled c D N A used for filter hybridization was synthesized as described above except that the labeled deoxyribonucleotide triphosphate was reduced to 15 #M, and the R N A was pretreated with 5 m M methylmercuric hydroxide as described (26). Labeled DNAs were routinely sheared by sonication to 0.5 kb average length.
Cloning of cDNAs Double-stranded c D N A was prepared from young cotton embryo total poly (A) + R N A essentially as described (6, 32) with the inclusion of placental ribonuclease inhibitor (Biotech, Madison WI) at 103 units ml I. Reverse transcriptase (Life Sciences, Inc), repurified by CM-Sephadex chromatography (24), was used at ten units per ~g RNA. After second strand synthesis, reactions were deproteinized with phenol, desalted on Sephadex G100 in 5 m M a m m o n i u m acetate and evaporated to dryness. Hairpin structures were cleaved with 700 units ml I S-1 nuclease (P. L. Biochemicals) in 0.3 M NaC1, 0.03 M sodium acetate, 0.003 M zinc acetate (pH 4.5) for 1 h at 37 ~ C. The reactions were deproteinated, desalted on Sepharose CL-2B and concentrated by evaporation. The tailing of cDNA inserts with dC and the tailing of Pst I-cleaved pBR325 (5) with dG was performed in 140 m M cacodylic acid, 30 m M T r i s base(adjusted to pH 6.9 with KOH), 1 m M dithiothreitol, 100/ag ml -l bovine serum albumin, 1 m M COC12, 15 m M deoxynucleotide triphosphate, 5 n M 3' termini and 350 units ml 1terminal deoxynucleotide transferase (P. L, Biochemicals). Between 14 and 16 residues o f d C were added in 1 min at 17 ~ C and about the same number o f d G in 2 min at 37 ~ C. Reaction mixtures were deproteinated with phenol and chromatographed on Sepharose CL-6B in 20 m M Tris-HC1,
Electrophoresis of nucleic acids' and northern blotting RNA and DNA were electrophoresed in denaturing conditions in I 0 mMmethylmercuric hydroxideagarose gels (2) in modified E buffer (7) or in 3% or 6% formaldehyde-agarose gels (21). Acridine orange at 0.5 #g ml I was occasionally included in formaldehyde gels from which RNA was to be transferred. Gels were destained in 3% formaldehyde-phosphate buffer and photographed prior to transfer. Transfer of R N A from formaldehyde gels was directly onto nitrocellulose (Schleicher and
191 Schue11, B-83) in 10X SSC by the method of Southern (28). After drying the filters were baked in vacuo at 80 ~ C for 90 min (30). Electrophoresis of D N A in non-denaturing conditions was in agarose or polyacrylamide in 90 m M Tris-borate, 3 m M EDTA, pH 8.3 (1).
for l0 min each in SSC, 0.1% sodium pyrophosphate, 0.1% SDS, at the SSC concentration and temperature of hybridization. Dried filters were exposed to fogged (20) K o d a k X - A R film with Dupont Cronex Lightening Plus Screens. Hybrid-selected and hybrid-arrested translation
Colony replicas
For screening large numbers of colonies, transformants (95% of which proved to be recombinant) were transferred with a toothpick to a chilled 23 • 23 cm petri plate containing nutrient broth and chloramphenicol (20 #g m1-1) in 2% agar, and grown 16 h at 37 ~ C. Colonies were replica-plated via a velvet cloth onto nitrocellulose on new plates and grown 11 h at 37 ~ C. Plasmid D N A was amplified (16, 18) for 16 h at 37 ~ after transfer of the filter to new plates containing 40 #g ml I spectinomycin. Colonies on the replica filters were lysed and prepared for hybridization by slight modifications of several methods (16, 29). Filter hybridization
The hybridization of all nitrocellulose filters was performed essentially as described by Maniatis et al. (22). Filters were wetted in 4X SSC, 0.1% SDS for 15 min at 68 ~ and incubated at 68 ~ for several hours in a solution containing 4X SSC, 10 m M sodium phosphate buffer, 10 m M E D T A , 0.1% SDS and 5X Denhardt's solution (pH 7.5). Denhardt's solution is 0.2% bovine serum albumin, 0.2% Ficoll (4 • 105 daltons) and 0.2% polyvinylpyrrolidone (3.7 • 105 daltons) (8). Filters were next incubated in hybridization buffer several hours and then in fresh hybridization buffer containing the radioactive D N A probe. Hybridization buffer contained 10 m M sodium phosphate buffer, 10 m M EDTA, 0.1% sodium pyrophosphate, 0.1% SDS, 100 #g ml I denatured, sonicated herring sperm D N A (0.5 kb), 10% dextran sulfate (5 X 105 daltons) (31), and from 1 to 6X SSC (pH 7.5). Purified formamide at 50% (v:v) was also present in some instances. The concentration of SSC and the hybridization temperature are detailed in the Figure Legends. After incubation with 32P-labeled D N A probe for 12 48 h, the filters were washed at the hybridization temperature 3 times for 10 min each with hybridization buffer and finally 4 times
To equate c D N A clones with specific mRNAs, recombinant plasmids were digested with Pst I which liberated the cloned c D N A from the plasmid vehicle. The D N A was then alkali denatured, neutralized, and bound to nitrocellulose discs as described by McGrogan et al. (23). The hybridization of young embryo poly(A) + R N A with the D N A filters, and subsequent thermal elution of the complexed R N A were also as described (23). The eluates from the first two washes at each temperature were pooled and ethanol precipitated two times. These released R N A s were resuspended in water and translated in the wheat germ in vitro translation system as previously described (11). Hybrid-arrested translation experiments were essentially as described by Paterson et al. (25) with the following modifications. Ten #g of Pst I-digested recombinant plasmid D N A and 40 #g of total R N A from young embryo cotyledons were mixed in 5 ~1 of water, heat denatured and quick chilled. Mixtures were b r o u g h t up to a final volume of 100 ~1 of 80% formamide (deionized and 2 times recrystalized), 10 m M Pipes - K O H , 0.4 M NaC1 (pH 7.5), and incubated at 40 ~ C for 1 h. The reactions were stopped by the addition of 0.8 ml of ice cold water and each reaction was divided into two fractions. One fraction was immediately ethanol precipitated, and the other was heat denatured by boiling for 30 s, quick chilled and ethanol precipitated. The fractions were ethanol precipitated several times prior to the translation of an aliquot in the wheat germ translation system (11).
Results
Immunochemical discrimination between storage protein subfamilies
The one dimensional electrophoretic display of the abundant proteins of embryo cotyledons at a developmental stage of m a x i m u m storage protein
192
Fig. 1. One dimensional gel electrophoresis of cotton cotyledon proteins. Lane 1: stained proteins of immature embryocotyledons. Lane 2: proteins synthesized in vitro from RNA extracted from immature embryocotyledons.Lane3: proteins synthesized in vivo by immature embryocotyledons.Lane4: in vitro synthesized proteins immunoprecipitated in a 5-min incubation by antibodies prepared against the 52 kD storage proteins. Lane 5: in vitro synthesized proteins immunoprecipitated in a 5-min incubation by antibodies prepared against the 48 kD storage proteins. Size estimates are in kD.
synthesis is presented in Fig. 1, lane 1. In lane 3 are displayed proteins labeled in v i v o when identical embryos are excised from the developing cotton boll and incubated in radioactive amino acids for a long period (24 h). It is clear that a long exposure of the tissue to the labeled amino acids yields a synthesis profile that mimics the concentration of the extant proteins of the tissue. This is not unexpected since these abundant proteins are chiefly the seed storage proteins and lectins that presumably do not
turn over during embryogenesis. When total or poly (A) + RNA extracted from the cotyledons of equivalent age embryos is translated in the wheat germ system, the profile of in v i t r o synthesis obtained is shown in lane 2. The protein species marked on this figure introduce the several intermediates thought to be involved in the formation of the mature storage proteins. Actually all of the intermediate and mature forms of these proteins are comprised of sets of isoelectric variants (9, 11); however, since no other abundant proteins with these apparent molecular weights occur in the cotyledon tissue, one dimensional gel electrophoresis is adequate up to a point for describing storage protein synthesis in these embryos. F r o m the data previously published (10, 11) we have postulated that the 69 kD initial translation products (lane 2) give rise to long-lived intermediates of 67 kD through the loss of a signal peptide during the movement of the initial translates to their deposition in the protein body. The 67 kD intermediates are observable on stained gels (lane 1) and on gels showing in v i v o synthesis (lane 3). We have postulated that these intermediates are slowly cleaved to the 48 kD mature proteins and a smaller protein set (not identified). It was also postulated analogously that the 60 kD initial translation products seen in lane 2 lose a signal peptide. These intermediates are not seen in stained gels or gels of in v i v o synthesis products because they are rapidly processed in v i v o to a higher apparent molecular weight set of intermediates of 70 kD by glycosylation. A very slow cleavage of this set was postulated to yield the glycosylated 52 kD protein set and a smaller protein set (not identified). Since the initial translation products lose a signal peptide and subsequently are cleaved to low molecular weight forms, they can be considered preproproteins. All the designated proteins in Fig. 1 (both in v i v o and in v i t r o labeled species and the stainable species) are precipitable by antibodies prepared against either the mature 52 kD species or the mature 48 kD species (11). Furthermore, the amino acid compositions of these two molecular weight sets are strikingly similar (9). For these reasons all the isoelectric variants of the 69 and 60 kD initial translation products are considered members of a multigene family indistinguishable by routine immunochemical procedures. However, since there are differences in apparent molecular weight and in
193 subsequent glycosylation between these two sets, the extent of homology/non-homology between the two sets of genes and their proteins was pursued. The first indication of major sequence differences between the 69 and 60 kD proteins was the fact that antibody to the 48 kD mature proteins precipitates ahnost exclusively the in vitro synthesized 69 kD preproproteins if the antigen/antibody incubation is limited to 5 min (Fig. 1, lane 5). Antibody to the 52 kD mature proteins does not show this short term selectively (lane 4), nor does either antibody when the incubation with the translation reaction mixture is carried out overnight. This more rapid reaction of the 48 kD antibody with the 69 kD preproproteins reinforces the proposition that the 48 kD mature proteins are derived from the 69 kD set. Identification o f storage protein messenger R N A s Both in vitro translation (12) and c D N A - m R N A hybridization kinetics (14) suggest that storage protein m R N A s together comprise a large fraction (>30%) of the young embryo mRNA. Thus these mRNAs should be resolved as obvious bands upon electrophoresis of poly(A) + RNA. Two m R N A bands are indeed visible in poly(A) + R N A upon electrophoresis in formaldehyde-agarose or methylmercury-agarose, each comprising about 15% of the total non-rRNA ethidium bromide-binding material (Fig. 2, lane 2). Their sizes are 1.96 _+0.02 kb and 2.26 + 0.02 kb (N = 6) as measured in either gel system in a wide variety of agarose concentrations using cotton rRNA (Fig. 2, lane 1) or cotton r R N A and E. coli r R N A as size markers (Fig. 2, lane 3). Besides their abundance in young embryo mRNA, several other considerations suggest these m R N A s are the storage protein mRNAs. The abundance of these two mRNAs in other developmental stage RNAs, as seen on stained gels of total poly (A) + R N A (data not shown) is c6rrelated with the storage protein m R N A content deduced from in vitro translation (12) and from c D N A - m R N A hybridization kinetics (14). Since the initial translation precursors are of two molecular weight classes and are structurally different though related as judged by immunological criteria (Fig. 1), one might expect two m R N A size species, one each for the 60 kD and 69 kD protein size classes. The 1.96 kb and 2.26 kb m R N A s seen in Fig, 2 exceed
Fig. 2. Identification of two abundant mRNAs in young embryo poly(A)+ RNA. RNAswereelectrophoresed in 3% formaldehyde-2-2% agarose gels and stained with ethidium bromide. Lane h 0.8 #g young embryo total RNA. Lane 2:2.5 #g young embryototal poly(A)+ RNA. Lane3: sameas lane2, but with0.6 ,ug E. coli rRNA included. Size estimates are in kb.
by about 0.4 kb the size required to code for these proteins. Construction and identification o f cloned storage protein c D N A s Our approach to obtaining eDNA clones for these m R N A S was to clone double-stranded cDNA made from young embryo m R N A , the stage at which they are at maximum concentration (12, 14), and to screen the recombinant colonies with eDNA from this developmental stage and a stage at which these m R N A s are greatly reduced. Colonies giving strong radioactive signals with the homologous c D N A probe but undetectable signals with the heterologous c D N A probe were used for the isolation of putative storage protein cDNAs. In the synthesis of double-stranded cDNA from young embryo cDNA, those double-stranded molecules of a discrete size and abundancy (shown by electrophoresis) sufficient to carry storage protein sequences were found to have internal Eco RI sites
194 b u t no P s t I sites. T o c o n s t r u c t r e c o m b i n a n t c D N A p l a s m i d s c a r r y i n g s t o r a g e p r o t e i n sequences whose inserts c o u l d be liberated intact f r o m the p l a s m i d vehicle, t o t a l y o u n g e m b r y o d o u b l e - s t r a n d e d c D N A was tailed with dC a n d a n n e a l e d to d G tailed Pst I-cleaved pBR325 as described in M e t h ods a n d m a t e r i a l s a n d used to t r a n s f o r m E. coli. N i t r o c e l l u l o s e replicas of a p p r o x i m a t e l y 1 700 rec o m b i n a n t colonies were h y b r i d i z e d with the two r a d i o a c t i v e c D N A p r e p a r a t i o n s m e n t i o n e d above. I n the first case, t o t a l y o u n g e m b r y o c D N A r e a c t e d s t r o n g l y with a b o u t 35% of the colonies, as one w o u l d p r e d i c t if such colonies c o n t a i n e d storage p r o t e i n sequences. T h a t these a b u n d a n t sequences are, in fact, y o u n g e m b r y o - s p e c i f i c was d e m o n s t r a t e d by the failure of dry seed c D N A to react with these colonies u n d e r equivalent h y b r i d i z a t i o n c o n d i t i o n s . S t o r a g e p r o t e i n m R N A sequences are r e d u c e d several h u n d r e d - f o l d at the dry seed stage (14).
T h e p l a s m i d inserts of a b o u t 270 of the y o u n g embryo-specific, strongly h y b r i d i z i n g colonies were sized after Pst I digestion. As predicted, less t h a n 2% of these plasmids a p p e a r e d to contain a Pst I site in the insert. I n 65 of the p l a s m i d s the inserts were l o n g e r t h a n 1.2 kb. T o detect the n u m b e r of separate sequences present in this p o o l of y o u n g embryo-specific, long i n s e r t - c o n t a i n i n g r e c o m b i n a n t s , these colonies were t r a n s f e r r e d to new plates and screened with r a n d o m l y selected m e m b e r s of the pool. F i g u r e 3 shows the r e a c t i o n of these colonies with three p l a s m i d inserts. Three groups of h y b r i d i z ing colonies were detected which t o g e t h e r c o m p r i s e m o r e t h a n 95% of the pre-selected colonies. In fairly extensive tests, not s h o w n here, no p l a s m i d insert was f o u n d to hybridize with m o r e t h a n one g r o u p of colonies at a m o d e r a t e criterion. In the clone identification tests described below, several r e p r e s e n t a tives f r o m each g r o u p were h y b r i d i z e d to y o u n g e m b r y o R N A with similar results. F o r the sake of
Fig. 3. Autoradiograph of colony hybridization of selected recombinant colonies with three cDNA plasmid inserts at two different criteria. Inserts were separated from the plasmid DNA before hybridization. The specific radioactivity of the insert DNA and exposure times were roughly the same in alle cases. Hybridization and filter washing conditions were l x SSC at 68 o C for (A C), and 6X SSC at 55 ~ for (D-F). Plasmid inserts used are identified on the left.
195 simplicity we present results only for one plasmid from each plasmid group in all of these tests. These are indicated in Fig. 3 and are C-72, C-94 and C-134.* Their insert lengths are about 2.3 kb, 1.9 kb, and 2.1 kb, respectively. A summation of the number of recombinant colonies carrying the young embryo c D N A library that hybridized with each of these plasmid groups suggests that sequences homologous to C-72, C-94 and C-131 comprise about 16, 12 and 6% of young embryo poly(A) + mRNA, respectively. The concentration of these sequences in total cotyledon m R N A during embryogenesis is dealt with in the companion paper (13). In Fig. 4 the three representative plasmids were hybridized to northern blots of electrophoretically separated young embryo total poly(A) + RNA. C-72 hybridized with the abundant 2.26 kb m R N A and both C-94 and C-134 hybridized with the abundant 1.96 kb mRNA. The proteins for which these m R N A s code were determined by the hybrid-select technique which involved translation in vitro of RNAs which hybridized with immobilized plasmid carrying one of the three inserts after recovery by thermal elution. In Fig. 5, mRNAs that hybridized with C-72 proved to code exclusively for the 69 kD proteins and both C-94 and C-134 hybridized with mRNAs that code exclusively for 60 kD storage proteins, as would be predicted by the northern blot results of Fig. 4. These products synthesized with hybrid-selected m R N A s have not been immuno-precipitated with anti-storage protein antibodies. However, their identity has been established by their behavior in one and two dimensional electrophoresis (data not shown). Curiously, the apparent melting temperatures of the three m R N A - D N A hybrids are not the same, the 2.26 kb m R N A - D N A hybrids melting about 10 ~ lower than the 1.96 kb m R N A - D N A hybrids. As a final demonstration that the three plasmid inserts do, in fact, represent three distinct subfamilies of storage protein genes, hybrid-arrested translation (25) was carried out with each of the three plasmids. In Fig. 6, C-72 is shown to arrest completely the translation of all the mRNAs coding for the 69 kD storage proteins, whereas neither C-94 nor C-134 alone completely arrest translation of *The precise notation of these plasmids is pGhSP C-72, pGhSP C-94, and pGhSP C-134.
Fig. 4. Autoradiographof hybridization of RNA with the three
plasmids used in Fig. 3. Young embryo poly(A)+ RNA was electrophoresed in a 6% formaldehyde-l.8% agarose gel, the RNAs blotted onto nitrocellulose paper whichwas then cut into strips corresponding to each electrophoresis lane. Strips were hybridized with one of the radioactive plasmids as indicated at the bottom of the figure. Plasmid specific radioactivities and autoradiograph exposure times were roughly equivalent. Hybridization and post-hybridization washing conditions were 4X SSC at 68 ~C. Sizes in kb are of cotton rRNAs and mRNAsthat were visualized in an adjacent lane by ethidium bromide staining.
m R N A s coding for the 60 kD storage protein. However, when present together (each at one-half the concentration as when present alone) they completely arrest the translation of these mRNAs. In all experiments the translation of the mRNAs arrested by the plasmid is restored upon denaturation of the m R N A - D N A hybrids prior to translation in vitro. These experiments have been performed at several R N A to D N A ratios with identical results (data not shown); therefore, the complete arrest of m R N A s coding for the 60 kD protein only with both C-94 and C-134 together is not the result of incomplete saturation of complementary mRNAs with either plasmid alone. Thus, at the hybridization criterion used here and in the colony hybridization experi-
196
Fig. 5. Autoradiograph of electrophoretically separated proteins synthesizedin vitro from rnRNA selected by hybridization with each of the three plasmids used in Fig. 3. Lane h proteins synthesized in vitro from young embryo cotyledon total RNA (control). Lanes 2-4: proteins synthesizedfrom RNA hybridized with plasmid C-72 and thermally released by incubation at 55 ~C (lane 2), 65 ~C (lane 3) and 75 ~C (lane 4). Lanes 5-7: as in lanes 2-4 using plasmid C-94. Lanes 8 10: as in lanes 2-4 using plasmid C-134. Apparent sizes in kD of principal storage protein preproproteins are indicated on left.
m e n t s in Fig. 3B, C, the C-94 a n d C-134 groups of plasmid inserts represent two distinct sets of sequences, b o t h of which code for subfamilies of 60 k D storage proteins. There are no other m R N A subfamilies for these proteins since, at the criterion used, these three groups of cloned c D N A s together arrest all m R N A sequences coding for 69 and 60 kD storage proteins. The issue of relatedness of these m R N A sequences has been p u r s u e d in some detail by m e a s u r i n g the h y b r i d i z a t i o n of colonies, the h y b r i d i z a t i o n of R N A on n o r t h e r n blots, and in hybrid-arrest and hybrid-select t r a n s l a t i o n carried out with m a n y cloned c D N A s u n d e r varied conditions. At no criterion e m p l o y e d (as low as 6 • SSC, at 55 ~ C, or 50% f o r m a m i d e , 4 X SSC, at 32 ~ did sequences present in 2.26 kb m R N A a n d 1.96 kb m R N A p l a s m i d inserts show a n y cross reaction (e.g. Fig.
Fig. 6. Autoradiograph of electrophoretically separated proteins synthesized in vitro from mRNA after hybridization with
each of the three plasmids used in Fig. 3. Lane 1: as in Fig. 5. Lanes 2, 4, 6: in vitro synthesis after incubation with plasmids, C-72, C-94 and C-134, respectively. Lane 8: after incubation with both C-94 and C-134. Lanes 3, 5, 7, 9 are of in vitro synthesis after mRNAs that were hybridized to the plasmids were thermally released by heating the incubation mixture to 100 ~
3D F), However, h o m o l o g y has been detected between the two 1.96 kb m R N A families via c o l o n y h y b r i d i z a t i o n at a 6 X SSC, at 55 o C, as s h o w n by a close e x a m i n a t i o n of Fig. 4 E - F .
Discussion T h e fact that the c o t t o n storage proteins are encoded in three sets of m R N A s rather than two came as a surprise. O n e a n d two d i m e n s i o n a l gel electrophoresis of the proteins suggest two sets as do the i m m u n o - p r e c i p i t a t i o n data which show only a kinetic d i s t i n c t i o n between the antigenicity of the two
197 molecular weight groups of proteins (Fig. 1). Initial hybrid selection experiments were first performed with plasmids C-72 a n d C-94 a n d were initially interpreted as indicating only two sets. Only when the screening of r e c o m b i n a n t colonies c o n t a i n i n g selected y o u n g e m b r y o c D N A inserts with r a n d o m ly selected m e m b e r s of that p o p u l a t i o n (such as in Fig. 3) was a third set of a b u n d a n t m R N A s represented by C-134) considered to be a possibility. F u r t h e r , hybrid-select experiments using C-134 showed that m R N A s for the 60 k D proteins were also selected by this plasmid (Fig. 5). Finally, the hybrid-arrest experiments revealed that both C-94 a n d C-134 were required to arrest fully the translation of the 60 k D proteins (Fig. 6). A m o n g l e g u m i n o u s species there are several examples of storage proteins that differ in molecular weight, in degree of glycosylation, a n d that show a n u m b e r of isoelectric variants, yet belong to a single multigenic family. All the vicilins of Phaseolus vulgaris are closely related as s h o w n recently by the s e q u e n c i n g of c D N A clones representing each varia n t (17). In Glycine max the different vicilins show evidence for a c o m m o n genetic b a c k g r o u n d by their i m m u n o c h e m i c a l cross reactivity. However, this h o m o l o g y is not manifested in the cross hybridizability of c D N A clones representing the various vicilin groups (3), much the same situation as we find for the cotton proteins. I n c o t t o n the degree of h o m o l o g y between all three gene subfamilies is still unclear. All three sets of proteins share antigenic sites, yet c D N A s for the 2.2 a n d 1.9 kb m R N A s do n o t hybridize a n d the c D N A s for the two 1.9 kb m R N A subfamilies hybridize only at a low criterion. These relationships can only be uncovered by c D N A sequencing as has been d o n e for the Phaseolus a n d Glycine proteins. The reiteration of these sequences, their g e n o m i c o r g a n i z a t i o n a n d analysis of their possible evolut i o n a r y relationships awaits the sequencing of gen o m i c clones.
Acknowledgements T h e technical assistance of J a n a Pyle a n d Steve Morris is gratefully acknowledged. This work was supported in part by grants from the N a t i o n a l Institutes of Health, C o t t o n Inc., and the C h e v r o n Research F o u n d a t i o n .
References 1. Air, AM, Sanger, F, Coulson, AR, 1976. Nucleotide and amino acid sequences of gene G of ~X174. J. Mol. Biol. 108:519 533. 2. Bailey,JM & Davidson, N, 1976. Methylmercuryas a reversible denaturing agent for agarose gel electrophoresis. Anal. Biochem. 70:75-85. 3. Beachy, RN, Doyle, J J, Ladin, BF & Schuler, MA (in press). Structure and expression of genes encoding the soybean 7S seed storage proteins. In: O Ciferri & LS Dure, (eds.) The Structure and Function of Plant Genomes. Plenum Press, New York, N.Y. 4. Birnboim, HC & Doly, J, 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucl. Acid Res. 7:1513-1523. 5. Bolivar, F, 1978. Construction and characterization of new cloning vehicles. III Derivativesof plasmid pBR322 carrying unique Eco RI sites for selection of Eco RI generated recombinant DNA molecules. Gene 4:12l- 136. 6. Buell, GN, Wickens, MP, Payvar, F & Schimke, RT, 1978. Synthesis of full length cDNAs from four partially purified oviduct mRNAs. J. Biol. Chem. 253:2471-2482. 7. Chandler, PM, Rimkus, D & Davidson, N, 1979. Gel electrophoretic fractionation of RNAs by partial denaturation with methylmercurichydroxide. Anal. Biochem.99:200-206. 8. Denhardt, DT, 1966. A membrane-filter technique for the detection of complementary DNA. Biochem. Biophys. Res. Comm. 23:641-646. 9. Dure III, LS & Chlan, CA, 1981. Developmental biochemistry of cottonseed embryogenesisand germination XII Purification and properties of principal storage proteins. Plant Physiol. 68:180-186. 10. Dure IlI, LS, Chlan, CA & Galau, GA (in press). Cottonseed storage proteins as a tool for developmental biology. In: O Ciferri & LS Dure, (eds.). The Structure and Function of Plant Genomes. Plenum Press, New York, N.Y. 11. Dure Ill, LS & Galau, GA, 1981. Developmentalbiochemistry of cottonseed embryogenesisand germination X I11Regulation of biosynthesis of principal storage proteins. Plant Physiol. 68:187-194. 12. Dure Ill, LS, Greenway, SC & Galau, GA, 1981. Developmental biochemistry of cottonseed embryogenesisand germination: changing messengerribonucleic acid populations as shown by in vitro and in vivo protein synthesis. Biochemistry 20:4162-4168. 13. Dure Ill, LS, Pyle, JB, Chlan, CA, Baker, JD & Galau, GA (companion paper) Developmental biochemistry of cottonseed embryogenesis and germination XVII Developmental expression of genes for the principal storage proteins. (Submitted to Plant Mol. Biol.). 14. Galau, GA & Dure III, LS, 1981. Developmental biochemistry of cottonseed embryogenesisand germination: changing messengerribonucleic acid populations as shown by reciprocal heterologous complementary deoxyribonucleic acidmessenger ribonucleic acid hybridization. Biochemistry 20:4169 4178. I 15. Galau, GA, Legocki, AB, Greenway, SC & Dure II1, LS, 1981. Cotton messenger RNA sequences exist in both po-
198 lyadenylated and nonpolyadenylated forms. J. Biol. Chem. 256:2551 2560. 16. Grunstein, M & Hogness, DS, 1975. Colony hybridization: a method for the isolation of cloned DNAs that contain a specific gene. Proc. Natl. Acad. Sci. USA 72:3961 3965. 17. Hall, TC, Slightom, JL, Ersland, DR, Murray, MG, Hoffman, LM, Adang, MJ, Brown, JWS, Ma, Y, Matthews, JA, Cramer, JH, Barker, RF, Sutton, DW, Kemp, JD. Phaseolin: nucleotide sequence explains molecular weight and charge heterogeneity of a small multigene family and also assists vector construction for gene expression in alien tissue. In Cifferri, O & Dure, LS, eds. The structure and function of plant genomes. Plenum Press, New York, N.Y. (in press). 18. Hanahan, D & Meselson, M, 1980. Plasmid screening at high colony density. Gene 10:63-67. 19. Kushner, SR, 1978. An improved method for transformation of Escherichia coli with cnlEl derived plasmids. In Boyer, HW & Nicosia, S, eds. Genetic Engineering. pp. 17 23. Elsevier-North Holland Biomedical Press, Amsterdam. 20. Laskey, RA & Mills, AD, 1975. Quantitative film detection of 3 H and 14C in polyacrylamide gels by fluorography. Eur. J. Biochem. 56:335 341. 21. Lehrach, H, Diamond, D, Wozney, JM & Boedtker, H, 1977. RNA molecular weight determinations by gel electrophoresis under denaturing conditions, a critical reexamination. Biochemistry 16:4743-4751. 22. Maniatis, T, Hardison, RC, Lacy, E, Lauer, J, O'Connell, C, Quon, D, Sin, GK & Efstratiadis, A, 1978. The isolation of structural genes from libraries of eucaryotic DNA. Cell 15:687 701. 23. McGrogan, M, Spector CJ, Halbert P & Raskas H J, 1979. Purification of specific adenovirus 2. RNAs by preparative hybridization and selective thermal elution. Nucleic Acid Res. 6:593 607. 24. Myers, JC, Ramirez, F, Kacian, DL, Flood, M, & Spiegel-
man, S, 1980. A simple purification of avian myeloblastosis virus reverse transcriptase for full-length transcription of 35 S RNA. Anal. Biochem. 101:88 96. 25. Paterson, BM, Roberts, BE & Kuff, EL, 1977. Structural gene identification mapping by DNA. mRNA hybrid-arrested cell-free translation. Proc. Natl. Acad. Sci. USA 74: 4370 4374. 26. Payvar, F & Schimke, RT, 1979. Methylmercury hydroxide enhancement of translation and transcription of ovalbumin and conalbumin mRNAs. J. Biol. Chem. 254:7636-7642. 27. Rigby, PWJ, Dieckmann, M, Rhodes, C & Berg, P, 1977. Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J. Mol. Biol. 113:237-251. 28. Southern, EM, 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503-517. 29. Thayer, RE, I979. An improved method for detecting foreign DNA in plasmids of Escherichia coli. Anal. Biochem. 98:60-63. 30. Thomas, PS, 1980. Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc. Natl. Acad. Sci. USA 77:5201-5205. 31. Wahl, GM, Stern M & Stark, GR, 1979. Efficient transfer of large DNA fragments from agarose gels to diazobenzyloxymethyl-paper and rapid hybridization by using dextran sulfate. Proc. Natl. Acad. Sci. USA 76:3683 3687 32. Wickens, MP, Buell, GN & Schimke, RT, 1978. Synthesis of double-stranded DNA complementary to lysozyme, ovamucoid, and ovalbumin mRNAs. Optimization for full length second strand synthesis by Escherichia coli DNA polymerase I. J. Biol. Chem. 253:2483-2495. Received 24 May 1983; in revised form and accepted 2 August 1983.
