Plant Molecular Biology 2:269 278 (1983). 9 Martinus Ni/hoff/Dr W. Junk Publishers, The Hague, Printed in the Netherlands.
Identification o f a multigene family for small heat shock proteins in soybean and physical characterization of one individual gene coding region Friedrich SchOffij & Joe L. Key 2
l Universitiit Bielefeld, Fakultiitfiir Biologie VI(Genetik), Postfach 8640, D-4800 Bielefeld 1, F.R.G. 2 University of Georgia, Botany Department, Athens, GA 30602, U.S.A.
Keywords: heat shock genes, hybrid selection, in vitro translation, multigene family, R-loop, soybean
Abstract When soybean seedlings are tranferred from 28 to 40 ~ C, a heat shock (hs) response is elicited. This is characterized by the synthesis of a new set of proteins (hs-proteins) and by cessation of normal protein synthesis (8). At the level of poly(A)mRNA, a new class of highly abundant RNAs appears which encodes a group of hs-proteins in the low molecular weight range of 15-18 kD (11). The classification of these proteins/genes into several sub-classes is based on a complex sequence relationship for class I protein/genes. This was confirmed by both the complexity and the similarity of southern blot hybridization patterns of genomic D N A digests with class I cDNA-probes. Genomic D N A clones (obtained from )~-libraries by screening with cDNA-probes) for the class 1 gene 1968 showed cross hybridization with all other class I cDNA-probes. Higher specificity of gene/protein correlation was obtained by variation of hybridization criteria. The specificity of c D N A clone 1968 for the genomic D N A clone ~hs68-7 was demonstrated by thermal stability of hybridization at 55 ~ C and 65 ~ C in 50% formamide compared to other cross-reacting probes. The correlation of clone 1968 with a specific hs-protein was obtained by temperature dependent release of hybrid selected hs-mRNAs at 50, 60, 70 and 85 * C followed by in vitro translation and two-dimensional gel analysis. The coding regions of hs-genes on genomic D N A clones were mapped by R-loop formation. The position of R-loops was mapped relative to certain restriction sites on subclones of ~hs68-7 DNA. The polarity of hs-genes was determined by attaching ~3X 174RF-DNA'labels' to the 3' poly(A)-tails of the m R N A s of R-loops.
Introduction Changes of protein synthesis in eucaryotes in response to temperature have great potential in the study of environmentally-regulated gene expression. This phenomenon of protein synthesis at elevated temperature (heat shock or hs) well above optimum growth conditions has been much studied in Drosophila (for review see (1)), but within recent years its study has also spread to many other biological systems (2). Little is known about the physiological role of hs-responses, although it has been shown to provide protection from thermal killing in Drosophila (3), Dictostelium (5), yeast (4), soybean (6) and fibroblasts (7).
In higher plants, the hs-phenomenon was first discovered at the level of protein synthesis in soybean (8, 9). It has been shown that elevation of temperature from 28 to 40 ~ induced de novo synthesis of several major groups of hs-proteins (hsp) whose molecular weights resemble those found for Drosophila. However, there is a marked difference in the complexity of the low molecular weight (lmw) group of hsp's between these two organisms. Drosophila synthesizes four hsp's of 22 000, 23 000, 26 000 and 27 000 daltons; soybean produces more than 20 hsp's in the molecular weight range of 15 000 18 000 daltons. The genes for these four Drosophila hsp's comprise a small hs-gene family with similar sequences which are
270 also related to that of a-crystallin (10), implying that certain structural domains (possibly for functional aggregation) are shared by these proteins. The lmw-hsp genes in soybean are the most actively expressed and coordinately regulated genes under hs conditions (11). Their hsp's are commonly associated with purified nuclei at high temperature, however, and disaggregate at low temperature (12). This indicates a c o m m o n function for these proteins in hs-response which is possibly related to c o m m o n structural features in proteins and genes. The lmw-hsp genes are subdivided into several classes defined by sequence homologies among poly(A)mRNA's (l 1, 12). Two of the eight classes are of particular interest for molecular studies of gene expression, because they represent the extreme exponents of the lmw-hsp genes. These are designated classes I and II; I consists of 13 closely related hsp's/genes, II comprises only one hsp which has no known sequence homology to other hs-genes. We have initiated the analysis of genomic D N A and cloning of hs-genes to study gene structure and expression in more detail. Here we describe the identification of a small multi-gene family for class I hs-genes and the characterization of one individual member by R-loop mapping and determination of the relative polarity of the transcript. We present experimental data which allow the assignment of specific gene/protein correlations for closely related members of a multigene family. Materials and m e t h o d s
Bacterial strains and plasmids Escherichia coli K12 derivatives used in this
study are: SK1590, (13) host strain for hs-cDNA clones; 294 (pro, r , m +, su +) host strain for genomic DNA subclones LE392, (14) host strain for propagation of phage k charon 4A derivatives (Table l).
DNA and R N A isolation from soybean High molecular weight soybean DNA was isolated from purified nuclei essentially as described by Nagao et al. (16). Poly(A)RNA purification from total soybean hypocotyl R N A after hs (40 ~ C) or untreated (28 ~ was carried out according to SchOffi & Key (11).
Plasmid and X-DNA isolation Plasmid D N A was isolated from E. coli cells according to the sarcosyl method (17) as described by Schi3ffl & Ptihler (18). Growth of k Charon 4A derivatives and X-DNA purification was performed as described by Williams & Blattner (19).
Restriction endonuclease digestion and subcloning o f DNA fragments in pBR322 Assay conditions for DNA digestions with the restriction endonucleases EcoRI, HindIII and PstI were according to Maniatis et al. (14), and standard electrophoresis of D N A fragments on 1% agarose gels was carried out according to Schi3ffl & Ptihler (18). Ten #g/lane was applied for soybean chromosomal D N A digests and about 0.5 #g/lane for plasmid or ,k-DNA digests. Completion of digestion was tested for soybean chromosomal D N A by southern blot hybridization with soybean rNDAprobes, kindly provided by Dr. R. Nagao, Universi-
Table 1. Plasmid pBR322 derivatives carrying soybean hs-gene sequences.
Plasmid
Class
Cloning site
Cloned insert size (kb)
Reference
cDNA clones pFS1968 pFS1920 pFS2005 pFS2059 pFS2019 pFS2026 pEC53
I 1 I I JI :I1 VII
PstI Pstl PstI PstI Pstl Pstl PstI
0.43 0.37 0.40 0.50 0.40 0.40 0.45
(11) (11) (11) (11) (1 I) (11) (12, 15)
Genomic clones phs6810 phs6871
I 1
EcoRI EcoRI/Hindlll
10.5 7.1
This paper This paper
271 ty of Georgia (data not shown). Fragment sizes were generally determined by comparison with XD N A digests (EcoRI, Hindlll, EcoR1/HindlII) run on the same gel. Subcloning ofEcoRI/HindIII fragments of genomic soybean DNA of Xhs68-7 into the respective sites of pBR322 was carried out as described by Maniatis et al. (14). Potential combinant reclones were screened by sizing the cloned D N A fragments on agarose gels using Xhs68-7 restriction fragments as a reference. Specific clones were identified by southern blot hybridization using cDNA-probes of clone 1968 (11).
Screening of a soybean genomic DNA-library Screening of a soybean genomic DNA-library, cloned into the EcoRI site o f a X Charon 4A vector (kindly provided by Dr R. Nagao, University of Georgia), was carried out as described by Nagao et al. (16), using a radioactively labelled insert probe of c D N A clone 1968 as described by Sch6ffl & Key (11).
Hybridization probes and blot hybridization methods c D N A inserts, isolated electrophoretically (20) and plasmids were labelled to high specific activity with [c~-32P]-dCTP (11) by nick translation (21). Unincorporated nucleotide was separated from labelled D N A on Sephadex G50. Southern blot hybridization (22) and autoradiography were carried out under the conditions described by Sch6ffi & Key (11). For D N A : D N A dot blot hybridizations, heat denatured Xhs68-7 D N A (0.1-0.0125/~g/5 #1) was spotted onto dry nitrocellulose filters (presoaked in 20 X SSC), baked for 2 h at 80 ~ and then hybridized as for southern blot hybridizations (11), except at different temperatures as indicated in the text.
Hybrid selection of mRNAs, in vitro translation and 2D-gel analysis of proteins Hybrid selection of hs poly(A)mRNA from total cellular poly(A)RNA was carried out as described by Sch6ffl & Key (11). The release of bound R N A from hybrids was modified by using an elution buffer consisting of 2 mM E D T A pH 7.6 and 10 #g/ ml wheat germ tRNA. RNA was eluted stepwise at
increasing temperatures (for 2 • 5 min each) as indicated in the text. RNA, present in the different fractions, was concentrated (11), translated in vitro in wheat germ $30 preparations (23) in the presence of [3H]-leucine (8) and proteins were analysed by 2D-O'Farrell gel electrophoresis (24).
Electron microscopy of R-loops, determination of polarity R-loops were formed by a procedure based on that of T h o m a s et al. (25) as modified by Kabak et al. (26). Linearized or fragmented plasmid D N A was crosslinked at a concentration of 50 # g / m l with UV-light in the presence of 1 # g / m l trioxalen. This D N A was then incubated for 16 h at 55 ~ at a concentration of 4-8 Fzg/ml with total cellular soybean hs-poly(A)RNA (2 4 # g / m l ) in 0.4 M NaC1, 0.01 M PIPES pH 7.8, 0.01 M EDTA, 50 100 # g / m l carrier RNA, 70% formamide. R-loops were stabilized by glyoxal treatment according to K a b a k et aL (26). The sample was diluted 10-fold in a solution of 50% formamide, 0.1 M Tris pH 8.5, 0.01 M EDTA, 50 # g / m l cytochrome C and spread on a hypophase of 15% formamide, 0.01 M Tris pH 8.5, 0.001 M E D T A (26). Samples of the film produced by spreading were picked up on parlodion-coated grids, stained with uranyl-acetate in 90% ethanol and rotary shadowed with platinum-palladium. Molecules were visualized and photographed with a Hitachi HS9 electron microscope, molecular lengths were measured and converted to kb by comparison with a Q X 1 7 4 R F I I - D N A standard. Gene polarity was determined by annealing dTtailed (27) O X 1 7 4 R F I I - D N A (28) to the poly(A)tails on R-loops (29) by incubation for 16 h at 4 o C. For these experiments, R-loops were purified in advance by passing the mixture through Sepharose 4B (26). R-loop fractions were recovered by visualization in the electron microscope. Results
Restriction enzyme analysis of soybean chromosomal DNA using clones hs-cDNA's C h r o m o s o m a l D N A isolated from purified nuclei was subjected to restriction enzyme analysis and southern blot hybridizations using cloned hscDNA's as hybridization probes. Figure 1 shows
272
Fig. l. Hybridization of soybean genomic DNA fragments with soybean hs cDNA clones. DNA was digested with EcoRl (E) and HindllI (H), electrophoresed on l% agarose, blotted to nitrocellulose, hybridized with radioactively labelled cDNA insert probes as indicated and autoradiographed.
the patterns obtained for EcoRI and HindIII digests of D N A when probed with different class I and class II cDNA's. All the class I probes (2005, 2059, 1920, 1968) hybridze very similarly but the patterns are complex and distinct from the simple hybridization patterns obtained with the class II probes (2019, 2026). When comparing the EcoRI generated patterns of hybridization bands obtained for class I clones, a total of about 6-7 bands adds up to roughly 25 kb, ind!cating the existence of a mul-. tigene family. A clonal specificity for class I probes is shown by marked intensity differences of certain EcoRI bands within class I patterns. In contrast to this, class II genes seem to be located only on bn~,, (possibly a doublet band) EcoRI fragment of about 2 kb on the soybean genome. Gene copy number reconstitution experiments indicate a unique D N A or at least a low copy number reiteration for individual class I hs-genes (data not shown).
Identification o f a genomic D N A clone containing sequences homologous to class I hs-cDNA's A soybean genomic D N A library, cloned into the EeoRI site of a ?~ Charon 4A vector (16), was screened for hs-specific D N A sequences using a radioactively labelled cDNA-insert of class I cDNA clone pFS1968. Two out of several ,k-clones identified initially by high signori'intensity in plaque hybridization were purified and further analysed. They~turned out" to be identical in their restriction fragment patterns and in hybridi'zation patterns obtained for EcoRI, HindIII and E c o R I / H i n d I I I digests with probe 1968. One clone (Jkhs68-7)is shown in Fig. 2A: The restriction map of this clone for the respective enzymes and the maps of two subclones of the soybean~genomic D N A in pBR322 are shown i n Fig. 2B. EU]'ther southern blot analysis of)~hs687'DNA using otherclass 1 and class II c D N A probes
273
Fig. 2. Characterization of genomic D N A clone Ahs68-7, carrying hs-specific sequences. A: Fragment and hybridization patterns for digests of Xhs68-7 with EcoRl (E), Hindlll (H) and EcoRI/Hindlll (EH). The three lanes on the left show the ethidium bromide staining of fragments, the three lanes on the right show the autoradiograph of a blot hybridized with probe 1968. B: Maps of genomic clone Ahs68-7 and two plasmid subclones in pBR322. Boxed portions of the maps represent soybean genomic DNA, solid line the A-vector, dashed line the pBR322 vector DNA. C: Specificity of Ahs68-7 D N A for different hs c D N A clones by dot blot hybridization. Identical concentration series of immobilized DNA (a: 0.1 ~g; b: 0.05 #g; c: 0.025 #g; d: 0.0125 ~g) were hybridized at different temperatures (as indicated) with different c D N A probes ((1): 1968; (2): 1920; (3): 2059; (4): 53) and auloradiographed.
(see Table 1), revealed sequence homologies of both EcoRI/HindIII fragments of the genomic D N A with all class I probes (identical patterns to that shown in Fig. 2A), but not with class II probes (data not shown). It is important to note that neither of the c D N A clones used as probes in this study contains a HindIII site (unpublished observation). Hence, there are at least two different locations of class I homologous DNA-sequences on khs68-7. The specificity of this genomic D N A clone for c D N A clone 1968 was compared with other crosshybridizing c D N A clones by dot blot hybridization. Three different temperatures (45, 55 and 65 ~ were used in this experiment to vary the
stringencies of hybridizations between ,khs68-7 (identical dilution series spotted onto nitrocellulose filters) and four different cDNA-probes (1968, 1920, 2059, 53). The results shown in Fig~ 2C demonstrates the higher specificity of )~hs68-7 for c D N A 1968 by hybridization up to 65 o C (for concentration a), compared to the hybridizations with the other cDNAs which are less temperature stable. However, their cross-hybridization under normal stringency (45 ~ C) with ,khs68-7 D N A indicates the close relationship of lmw-hsp genes a n d / o r multiple locations of hs-specific D N A sequences at this locus.
274
Assignment o f a specific hs-protein f o r class I c D N A clone pFS1968 Standard hybrid selection/translation protocols were insufficient to d e m o n s t r a t e a specific gene/ p r o t e i n c o r r e l a t i o n for class I hsp's (11). This c o m p l i c a t i o n in m o l e c u l a r studies, caused by the sequence h o m o l o g y o f class I genes, is resolved for D N A : D N A h y b r i d i z a t i o n s as described above. A m o r e specific selection of h s - m R N A s for hybrid release t r a n s l a t i o n s was o b t a i n e d by v a r y i n g the h y b r i d release t e m p e r a t u r e . C o n d i t i o n s for h y b r i d selection using c D N A clone 1968 were identical to those d e s c r i b e d by Sch6ffl & Key (1 I) for the same clone. H o w e v e r , the h y b r i d i z e d p o l y ( A ) m R N A was g r a d u a l l y released by increasing t e m p e r a t u r e s to 50, 60, 70 and 85 ~ C. The elution p a t t e r n of trans-
latable R N A s was d e t e r m i n e d by assaying for 3Hleucine i n c o r p o r a t i o n into proteins synthesized in vitro in the wheat g e r m system using the m R N A s of the i n d i v i d u a l fractions as templates. In this experiment, a b o u t 40% of the total i n c o r p o r a t i o n resulted f r o m R N A eluted at 50 o C, 30% from 60 ~ R N A , 20% f r o m 7 0 ~ and 10% f r o m 8 5 ~ The p r o t e i n s synthesized f r o m the 4 different R N A fractions were s u b s e q u e n t l y a n a l y s e d by 2D-gel a n a l y sis. F i g u r e 3 shows the c a t a l o g u e of f l u o r o g r a m s of s e p a r a t e d p r o t e i n s for the different temperatures. The m a j o r spots in all f o u r f l u o r o g r a m s correlate with l m w - h s p ' s of the s o y b e a n (11). At 50 ~ a n u m b e r of a d d i t i o n a l m i n o r spots a p p e a r which do not belong to the s t a n d a r d p a t t e r n of class I hsp's, as synthesized f r o m 60 * C R N A . A t 70 ~ C, R N A for m a i n l y three hsp's out of the s t a n d a r d p a t t e r n is
Fig. 3. Fluorograms of in vitro synthesized proteins separated on 2D-gels. PolY(A)mRNAs, selected by hybridization to cDNA clone 1968, were eluted stepwise from hybrids at 50, 60, 70 and 85 ~ and subsequently~ranslated in a wheat germ system in the presence of [3H]-leucine. The molecular weight standards used in the S DS-d[rnensiOn of gel electrophoresis are shown on the left side of each gel. Their molecular weights are fro m top to bottom: 92 500 (degraded); 69 000~.46 000; 30 000; 12 300 dalton, q~hearrowhead marks hsp68.
275 released; 85 ~ RNA reduces this pattern even more to only one major hsp, defined as hsp68. This experiment shows that by the criterion of thermal stability of m R N A : c D N A hybridization, a specific gene/protein correlation can be obtained. The data also further allow subclassification of class I genes/ proteins by identifying two hsp's which are closely related to hsp68 (the two additional proteins in the 70 ~ pattern). Consequently, the 10 remaining members of class I have to be considered as more distantly related to hsp68.
hybridize to duplex D N A by displacing the identical strand and forming R-loops (25, 26). By measuring and comparing the position of R-loops on these two plasmids, it was possible to locate them unambiguously with respect to the restriction map of the plasmids as shown in Fig. 4A and B. The loop distributions on both plasmids determine only one single location, starting about 1 kb away from the H i n d I I I site, with an average loop size of 0.45 kb. This is just about the size needed for a coding capacity of a lmw-hsp, but only half the size of the m R N A s found for these proteins (11). This apparent discrepancy may result either from an overestimate of the RNA-size due to the D N A standards used for its determination ( l 1) or from an unavoidable underestimate of loop sizes due to the partial sequence homology of class I mRNAs. The participation of heterologous RNAs in R-loop formation at this locus is indicated by the large variations in size of R-loops (0.5 +0.13 kb and 0.4 +0.17 kb) contributing to the histograms in Fig. 4. The polarity of the transcripts at this gene locus was deter-
R-loop m a p p i n g o f a class I hs-gene on subclones o f Xhs68-7
A hs-specific DNA region on genomic clone hhs68-7 has been precisely mapped by electron microscopy of R-loops using the plasmid subclones phs6810 and phs6871 (see Fig. 2B). E c o R I digested phs6810 and H i n d I I I digested phs6871 were separately incubated with total poly(A)RNA from hshypocotyls under conditions, which allow R N A to
E
H i
E!
0 loop size distances from E
.10
.
.
.
.
w
A
0.5+ 0.13kb 3.95+_0.27kb 5.13+0.40kb
l-
9
_e 0
E
loop s i z e distance from H
-10
0
0.41_0.17
B
kb
1. O0 __.0 . 4 0 kb
c"
map !
i H
9
,
,
.
|
9
9
9
9
w
.
.
.
.
!
.
.
.
.
units 1
9
-~-
0
E
H
Fig. 4. Histogramsof R-loopslocalized on subclones of soybeangenomicDNA carryinghs-genes. R-loops,formedby hybridizationof hs-poly(A)RNA on the 10.5 kb EcoRI fragment of phs6810 (A) and on HindIII linearized phs6871 (B) were visualized by electron microscopy, measured and aligned to each other with respect to their commonrestriction sites E for EcoRI and H for HindII1.
276
Fig. 5. Electron micrograph of R-loops identifyingthe coding region and polarity of a soybean hs-gene. Soybeanhs-poly(A)RNAwas hybridized to EcoRI digested phs-6810(A) and HindII1 digested phs6871 (B). The 3' ends of RNAsin R-loopswerelabelled by attaching circular dT-tailed @X174RFII-DNAcirclesto the poly(A)-tails.The line drawings showthe interpretation of R-loopstructures for both molecules.
mined by labelling the 3' poly ends of RNA in R-loops by dT-tailed @X174RF|I-DNA (27 29). The result for the two different types of R-loops is shown in Fig. 5. In both cases, @X174-markers are attached to the same ends relative to the maps in Fig. 4. This result defines the 5' terminus at the proximal end and the 3' terminus at the distal end of R-loops with respect to the H i n d l I I site. The gene does not contain introns detectable by this method. R-loop formation Was not observed in experiments when the hs-poly(A)RNA fraction was replaced by total control poly(A)RNA, indicating the hs-specific activation of this gene.
Discussion
The data presented in this paper confirm the
predictions made from earlier observations (11) on the existence of a multigene family for part of the lmw-group of soybean hsp's. This was shown by southern blot hybridization, where about 6 7 different E c o R I fragments could be identified as potential bearers of class I hs-genes for a total of 13 different proteins. The finding that fewer fragments hybridize than there are genes available suggests a certain clustering of hs-genes. One indication for such an organization of related genes on genomic DNA is provided by the hybridization analysis of clone ~.hs68-7. This clone carries a hs-gene, identified by R-loop mapping, in the middle of a 10.5 kb E c o R I fragment and another portion of hs-gene sequences is located on this fragment at the end proximal to the HindIII-site, possibly cut by E c o R I (see Fig. 2B, unpublished observation). It is not yet known to what degree clusters of class I genes are
277 interspersed with other hs-genes. Similar organization of hs-genes has been described for Drosophila at different chromosomal locations, such as, the multiple gene copies for hsp70 at 87 A/87 C (30 34) or the four different genes hsp22, hsp23, hsp26 and hsp27 organized at the 67B locus (27, 35, 36). The hs-genes at the 67B locus are interspersed with three additional genes of unknown function; differential expression of genes at this cluster is under developmental control (38, 39). Drosophila hs-genes appear to have no intervening sequences, with the exception of the hsp83 gene (35, 40). This general feature holds true also for the only other hs-gene of higher eucaryotic origin characterized so far (this paper), but its seems to apply for other soybean hs-genes of the lmw-hsp group as well (unpublished data). The relatively broad variation in R-loop sizes described in this paper is interpreted as the result of partial sequence homology of this gene with other class I hsmRNAs, which compete in hybridization at the same gene locus. The asymmetrical distribution of the R-loops at this site suggest a preference for the 5' half of c o m m o n sequences on class I genes. The methods applied in the studies specifying gene/protein correlation of hsp68 can be generally used for further subclassification of the class 1 multigene family based on sequence relationships. Final proof of these predictions has to come from DNA sequence analysis to show the structural and functional significance of'hs-domains' and to provide the basis for investigating the regulation of hs-gene expression.
Acknowledgements We thank Dr P Nagao for providing a soybean genomic library and his assistance in screening. We also wish to acknowledge the excellent technical assistance of Sieglinde Angermtiller, G0tz Baumann and Cheryl Mothershed. We are grateful to Drs V. Krishnapillai and M. O'Connell for critical reading and Susanna Malmivaara for typing of the manuscript. This work was supported by the Deutsche Forschungsgemeinschaft (grant Scho 242-4).
References l. Ashburner M, Bonnet J J: The induction of gene activity in Drosophila by heat shock. Cell 17:241-254, 1979. 2. Schlesinger M J, Ashburner M, Tissibres A: Heat shock, from bacteria to man. Cold Spring Harbor Laboratory, 1982. 3. Mitchell HK, Moiler G, Peterson NS, Lipps-Sarimento L: Specific protection from phenocopy induction by heat shock. Dev Genet 1:181-192, 1979. 4. McAllister L, Finkelstein DB: Heat shock proteins and thermal resistance in yeast. Biochem Biophys Res C o m m u n 93:819-824, 1980. 5. Loomis WT, Wheeler S: Heat shock response of Dictostelium. Dev Biol 79:399 408, 1980. 6. Key JL, Lin CY, Ceglarz E, Sch6ffl F: The heat shock response in soybean seedlings. In: Ciferri O, Dure LS (eds) The Structure and Function of Plant Genomes, in press. 7. Li GC, Werb Z: Correlation between synthesis of heat shock proteins and development of thermotolerance in Chinese hamster fibroblasts. Proc Natl Acad Sci USA 79:3218-3222, 1982. 8. Key JL, Lin CY, Chen YM: Heat shock proteins of higher plants. Proc Natl Acad Sci USA 78:3526-3530, 1981. 9. Barnett T, Altschuler M, McDaniel CN, Mascarenhas JP: Heat shock induced proteins in plant cells. Dev Gen 1:331-340, 1980. 10. lngolia TD, Craig EA: Four small Drosophila heat shock proteins are related to each other and to mammalian c~-cryst a l l m Proc Natl Acad Sci USA 79:2360 2364, 1982. 1 l. Sch~ffl F, Key J L: An analysis of m R N A s for a group of heat shock proteins of soybean using cloned cDNAs. J Mol Appl Genet 1:301-314, 1982. 12. Key JL, Lon CY, Ceglarz E, SchOffl F: The heat shock response in plants: Physiological considerations. In: Schlesinger M J, A s h b u r n e r M, Tissibres A (eds) Heat shock, From Bacteria to Man. Cold Spring Harbor la'boratory, 1982, pp 329-336. 13. Kushner SR: An improved method for transformation in Escherichia coli with ColE1 derived plasmids. In: Boyer HW, Nicosia S (eds) Genetic Engineering. Elsevier/North Holland Biomedical Press, Amsterdam. 1978, pp 17-23. 14. Maniatis T, Fritsch EF, Sambrook J: Molecular Cloning, a Laboratory Manual, Cold Spring Harbor Laboratory, 1982. 15. Czarnecka E, Edelman L, Sch/3ffl F, Key Jk: Analysis of physical stress response in soybean using cloned heat shock cDNA's. 1983 (submitted for publication). 16. Nagao RT, Shah DM, Eckenrode VK, Meagher RB: Multigene family of actin-related sequences isolated from a soybean genomic library. D N A 2:1 9, 1981. 17. Bazaral M, Helinski DR: Circular forms of colicinogenic factors El, E2 and E3 from Escherichia coll. J Mol Biol 36:185-194, 1968. 18. Scb6ffl F, Ptihler A: The plasmid system of Escherichia coli strain U R 12644: identification and molecular characteristics of transpnsons involved in the generation of endogeneous R-plasmids. Genet Res Camb 34:287-301, 1979.
278 19. Williams BW, Blattner FR: Construction and characterization of the hybrid bacteriphage lambda Charon vectors for D N A cloning. J Virol 29:555-575, 1979. 20. McDonell MW, Simon MN, Studier FW: Analysis of restriction fragments of T7 D N A and determination of molecular weights by electrophoresis in natural and alkaline gels. J MolBio1110:119 146, 1977. 211Rigby PWJ, Dieckmann M, Rhodes C, Berg P: Labelling deoxyribonucleic acid to high specific activity in vitro by nick translation with DN A polymerase 1. J Mol Biol 113:337-351, 1977. "22. Southern EM: Detection~f specific sequences a m o n g DNA fragments. J MoI Biol 98:'503-517, 1975. 23. Roberts BE, Paterson BM: Efficient translation of tobacco mosaic virus R N A and rabbit globin RNA in a cell free system from commercial wheat germ. Proc Natl Acad Sci USA 70:2330 2334, 1973. 24. O'Farrell PH: High resolution two-dimensional electrophoresis of proteins. J Biol Chem 250:4007-4021, 1975. 25. T h o m a s M, White RL, Davis I~W: Hybridization of R N A to double-stranded DNA: formation of R-loops. Proc Natl Acad Sci USA 73:2294 2298, 1976. 26. Kaback DB, Angerer LM, Davidson N: Improved methods for the formation and stabilization of R-loops. Nucleic Acids Res 6:2499 2517, 1979. 27. Bender W, Davidson N: Mapping of poly(A) sequences in electron microscope reveals unusual structure of type C oncornavirus R N A molecules. Cell 7:595 607, 1976. 28. Fisher RL, Goldberg RB: Structure and flanking regions of soybean seed protein genes. Cell 29:651 660, 1982. 29. Bender W, Davidson N, Kindle KL, Taylor WC, Silverman M, Firtel RA: The strticture of M6, a recombinant plasmid containing Dictostelium D N A homologous to actin messenger RNA. Cell 15:779 788, 1978. 30. Livak K J, Freund R, Schweber M, Wensink PC, Meselson M: Sequence organization and transcription at two heat shock loci in Drosophila. Proc Natl Acad Sci USA 75: 5613-5617, 1978. 31. Lis JT, Prestidge L, Hogness DS: A novel arrangement of
tandemly repeated genes at a major heat shock site in D. melanogaster. Cell 14:901-919, 1978. 32. Schedl P, Artavania-Tsakonas S, Steward R, Gehring W, Mirault M-E, Goldschmidt-Clermont M, Moran L, Tissi~res A: Two hybrid plasmids with D. melanogaster D N A sequences complementary to m R N A coding for the major heat shock protein. Cell 14:921 929, 1978. 33. Artavanis-Tsakonas S, Schedl P, Mirault M-E, Moran L: Genes for the 70 000 dalton heat shock protein in two cloned D. melanogaster DNA segments. Cell 17:9-18, 1979. 34. Craig EA, McCarthy BJ, Wadsworth SC: Sequence organization of two recombinant plasmids containinggenes for the major heat shock protein of D. melanogaster. Cell 16: 575 588, 1979. 35. Corces V, Holmgren R, Freund R, Morimoto R, Meselson M: Four heat shock" proteins of Drosophila melanogaster coded within a 12 kilobase region in chromosome subdivision 67B. Proc Natl Acad Sci USA 77:5390 5393, 1980. 36. Wadsworth SC, Craig EA, McCarthy BJ: Genes for three Drosophila heat schock induced proteins at a single locus. Proc Natl Acad Sci USA 77:2134 2137, 1980. 37. Voellmy R, Goldschmidt-Clermont M, Southgate R, Tissi~res A, Levis R, Gehring W: A D N A segment isolated from chromosomal site 67B in D. melanogaster contains four closely lined heat shock genes. Cell 23:261 270, 1981. 38. Sirotkin K, Davidson N: Developmentally regulated transcription from Drosophila melanogaster chromosomal site 67B. Dev Biol 89:196 210, 1982. 39. Z i m m e r m a n n JL, Petri W, Meselson M: Accumulation of a specific subset of D. melanogaster heat shock m R N A s in normal development without heat shock. Cell 32:1161-1170, 1983. 40. Holmgren R, Corces V, Morimoto R, Blackman R, Meselson M: Sequence homologies in the 5' regions of four Drosophila heat shock genes. Proc Natl Acad Sci USA 78: 3775 3778, 1981. Received 24 May 1983; in revised from 22 August 1983; accepted 30 August 1983.
Identification o f a multigene family for small heat shock proteins in soybean and physical characterization of one individual gene coding region Friedrich SchOffij & Joe L. Key 2
l Universitiit Bielefeld, Fakultiitfiir Biologie VI(Genetik), Postfach 8640, D-4800 Bielefeld 1, F.R.G. 2 University of Georgia, Botany Department, Athens, GA 30602, U.S.A.
Keywords: heat shock genes, hybrid selection, in vitro translation, multigene family, R-loop, soybean
Abstract When soybean seedlings are tranferred from 28 to 40 ~ C, a heat shock (hs) response is elicited. This is characterized by the synthesis of a new set of proteins (hs-proteins) and by cessation of normal protein synthesis (8). At the level of poly(A)mRNA, a new class of highly abundant RNAs appears which encodes a group of hs-proteins in the low molecular weight range of 15-18 kD (11). The classification of these proteins/genes into several sub-classes is based on a complex sequence relationship for class I protein/genes. This was confirmed by both the complexity and the similarity of southern blot hybridization patterns of genomic D N A digests with class I cDNA-probes. Genomic D N A clones (obtained from )~-libraries by screening with cDNA-probes) for the class 1 gene 1968 showed cross hybridization with all other class I cDNA-probes. Higher specificity of gene/protein correlation was obtained by variation of hybridization criteria. The specificity of c D N A clone 1968 for the genomic D N A clone ~hs68-7 was demonstrated by thermal stability of hybridization at 55 ~ C and 65 ~ C in 50% formamide compared to other cross-reacting probes. The correlation of clone 1968 with a specific hs-protein was obtained by temperature dependent release of hybrid selected hs-mRNAs at 50, 60, 70 and 85 * C followed by in vitro translation and two-dimensional gel analysis. The coding regions of hs-genes on genomic D N A clones were mapped by R-loop formation. The position of R-loops was mapped relative to certain restriction sites on subclones of ~hs68-7 DNA. The polarity of hs-genes was determined by attaching ~3X 174RF-DNA'labels' to the 3' poly(A)-tails of the m R N A s of R-loops.
Introduction Changes of protein synthesis in eucaryotes in response to temperature have great potential in the study of environmentally-regulated gene expression. This phenomenon of protein synthesis at elevated temperature (heat shock or hs) well above optimum growth conditions has been much studied in Drosophila (for review see (1)), but within recent years its study has also spread to many other biological systems (2). Little is known about the physiological role of hs-responses, although it has been shown to provide protection from thermal killing in Drosophila (3), Dictostelium (5), yeast (4), soybean (6) and fibroblasts (7).
In higher plants, the hs-phenomenon was first discovered at the level of protein synthesis in soybean (8, 9). It has been shown that elevation of temperature from 28 to 40 ~ induced de novo synthesis of several major groups of hs-proteins (hsp) whose molecular weights resemble those found for Drosophila. However, there is a marked difference in the complexity of the low molecular weight (lmw) group of hsp's between these two organisms. Drosophila synthesizes four hsp's of 22 000, 23 000, 26 000 and 27 000 daltons; soybean produces more than 20 hsp's in the molecular weight range of 15 000 18 000 daltons. The genes for these four Drosophila hsp's comprise a small hs-gene family with similar sequences which are
270 also related to that of a-crystallin (10), implying that certain structural domains (possibly for functional aggregation) are shared by these proteins. The lmw-hsp genes in soybean are the most actively expressed and coordinately regulated genes under hs conditions (11). Their hsp's are commonly associated with purified nuclei at high temperature, however, and disaggregate at low temperature (12). This indicates a c o m m o n function for these proteins in hs-response which is possibly related to c o m m o n structural features in proteins and genes. The lmw-hsp genes are subdivided into several classes defined by sequence homologies among poly(A)mRNA's (l 1, 12). Two of the eight classes are of particular interest for molecular studies of gene expression, because they represent the extreme exponents of the lmw-hsp genes. These are designated classes I and II; I consists of 13 closely related hsp's/genes, II comprises only one hsp which has no known sequence homology to other hs-genes. We have initiated the analysis of genomic D N A and cloning of hs-genes to study gene structure and expression in more detail. Here we describe the identification of a small multi-gene family for class I hs-genes and the characterization of one individual member by R-loop mapping and determination of the relative polarity of the transcript. We present experimental data which allow the assignment of specific gene/protein correlations for closely related members of a multigene family. Materials and m e t h o d s
Bacterial strains and plasmids Escherichia coli K12 derivatives used in this
study are: SK1590, (13) host strain for hs-cDNA clones; 294 (pro, r , m +, su +) host strain for genomic DNA subclones LE392, (14) host strain for propagation of phage k charon 4A derivatives (Table l).
DNA and R N A isolation from soybean High molecular weight soybean DNA was isolated from purified nuclei essentially as described by Nagao et al. (16). Poly(A)RNA purification from total soybean hypocotyl R N A after hs (40 ~ C) or untreated (28 ~ was carried out according to SchOffi & Key (11).
Plasmid and X-DNA isolation Plasmid D N A was isolated from E. coli cells according to the sarcosyl method (17) as described by Schi3ffl & Ptihler (18). Growth of k Charon 4A derivatives and X-DNA purification was performed as described by Williams & Blattner (19).
Restriction endonuclease digestion and subcloning o f DNA fragments in pBR322 Assay conditions for DNA digestions with the restriction endonucleases EcoRI, HindIII and PstI were according to Maniatis et al. (14), and standard electrophoresis of D N A fragments on 1% agarose gels was carried out according to Schi3ffl & Ptihler (18). Ten #g/lane was applied for soybean chromosomal D N A digests and about 0.5 #g/lane for plasmid or ,k-DNA digests. Completion of digestion was tested for soybean chromosomal D N A by southern blot hybridization with soybean rNDAprobes, kindly provided by Dr. R. Nagao, Universi-
Table 1. Plasmid pBR322 derivatives carrying soybean hs-gene sequences.
Plasmid
Class
Cloning site
Cloned insert size (kb)
Reference
cDNA clones pFS1968 pFS1920 pFS2005 pFS2059 pFS2019 pFS2026 pEC53
I 1 I I JI :I1 VII
PstI Pstl PstI PstI Pstl Pstl PstI
0.43 0.37 0.40 0.50 0.40 0.40 0.45
(11) (11) (11) (11) (1 I) (11) (12, 15)
Genomic clones phs6810 phs6871
I 1
EcoRI EcoRI/Hindlll
10.5 7.1
This paper This paper
271 ty of Georgia (data not shown). Fragment sizes were generally determined by comparison with XD N A digests (EcoRI, Hindlll, EcoR1/HindlII) run on the same gel. Subcloning ofEcoRI/HindIII fragments of genomic soybean DNA of Xhs68-7 into the respective sites of pBR322 was carried out as described by Maniatis et al. (14). Potential combinant reclones were screened by sizing the cloned D N A fragments on agarose gels using Xhs68-7 restriction fragments as a reference. Specific clones were identified by southern blot hybridization using cDNA-probes of clone 1968 (11).
Screening of a soybean genomic DNA-library Screening of a soybean genomic DNA-library, cloned into the EcoRI site o f a X Charon 4A vector (kindly provided by Dr R. Nagao, University of Georgia), was carried out as described by Nagao et al. (16), using a radioactively labelled insert probe of c D N A clone 1968 as described by Sch6ffl & Key (11).
Hybridization probes and blot hybridization methods c D N A inserts, isolated electrophoretically (20) and plasmids were labelled to high specific activity with [c~-32P]-dCTP (11) by nick translation (21). Unincorporated nucleotide was separated from labelled D N A on Sephadex G50. Southern blot hybridization (22) and autoradiography were carried out under the conditions described by Sch6ffi & Key (11). For D N A : D N A dot blot hybridizations, heat denatured Xhs68-7 D N A (0.1-0.0125/~g/5 #1) was spotted onto dry nitrocellulose filters (presoaked in 20 X SSC), baked for 2 h at 80 ~ and then hybridized as for southern blot hybridizations (11), except at different temperatures as indicated in the text.
Hybrid selection of mRNAs, in vitro translation and 2D-gel analysis of proteins Hybrid selection of hs poly(A)mRNA from total cellular poly(A)RNA was carried out as described by Sch6ffl & Key (11). The release of bound R N A from hybrids was modified by using an elution buffer consisting of 2 mM E D T A pH 7.6 and 10 #g/ ml wheat germ tRNA. RNA was eluted stepwise at
increasing temperatures (for 2 • 5 min each) as indicated in the text. RNA, present in the different fractions, was concentrated (11), translated in vitro in wheat germ $30 preparations (23) in the presence of [3H]-leucine (8) and proteins were analysed by 2D-O'Farrell gel electrophoresis (24).
Electron microscopy of R-loops, determination of polarity R-loops were formed by a procedure based on that of T h o m a s et al. (25) as modified by Kabak et al. (26). Linearized or fragmented plasmid D N A was crosslinked at a concentration of 50 # g / m l with UV-light in the presence of 1 # g / m l trioxalen. This D N A was then incubated for 16 h at 55 ~ at a concentration of 4-8 Fzg/ml with total cellular soybean hs-poly(A)RNA (2 4 # g / m l ) in 0.4 M NaC1, 0.01 M PIPES pH 7.8, 0.01 M EDTA, 50 100 # g / m l carrier RNA, 70% formamide. R-loops were stabilized by glyoxal treatment according to K a b a k et aL (26). The sample was diluted 10-fold in a solution of 50% formamide, 0.1 M Tris pH 8.5, 0.01 M EDTA, 50 # g / m l cytochrome C and spread on a hypophase of 15% formamide, 0.01 M Tris pH 8.5, 0.001 M E D T A (26). Samples of the film produced by spreading were picked up on parlodion-coated grids, stained with uranyl-acetate in 90% ethanol and rotary shadowed with platinum-palladium. Molecules were visualized and photographed with a Hitachi HS9 electron microscope, molecular lengths were measured and converted to kb by comparison with a Q X 1 7 4 R F I I - D N A standard. Gene polarity was determined by annealing dTtailed (27) O X 1 7 4 R F I I - D N A (28) to the poly(A)tails on R-loops (29) by incubation for 16 h at 4 o C. For these experiments, R-loops were purified in advance by passing the mixture through Sepharose 4B (26). R-loop fractions were recovered by visualization in the electron microscope. Results
Restriction enzyme analysis of soybean chromosomal DNA using clones hs-cDNA's C h r o m o s o m a l D N A isolated from purified nuclei was subjected to restriction enzyme analysis and southern blot hybridizations using cloned hscDNA's as hybridization probes. Figure 1 shows
272
Fig. l. Hybridization of soybean genomic DNA fragments with soybean hs cDNA clones. DNA was digested with EcoRl (E) and HindllI (H), electrophoresed on l% agarose, blotted to nitrocellulose, hybridized with radioactively labelled cDNA insert probes as indicated and autoradiographed.
the patterns obtained for EcoRI and HindIII digests of D N A when probed with different class I and class II cDNA's. All the class I probes (2005, 2059, 1920, 1968) hybridze very similarly but the patterns are complex and distinct from the simple hybridization patterns obtained with the class II probes (2019, 2026). When comparing the EcoRI generated patterns of hybridization bands obtained for class I clones, a total of about 6-7 bands adds up to roughly 25 kb, ind!cating the existence of a mul-. tigene family. A clonal specificity for class I probes is shown by marked intensity differences of certain EcoRI bands within class I patterns. In contrast to this, class II genes seem to be located only on bn~,, (possibly a doublet band) EcoRI fragment of about 2 kb on the soybean genome. Gene copy number reconstitution experiments indicate a unique D N A or at least a low copy number reiteration for individual class I hs-genes (data not shown).
Identification o f a genomic D N A clone containing sequences homologous to class I hs-cDNA's A soybean genomic D N A library, cloned into the EeoRI site of a ?~ Charon 4A vector (16), was screened for hs-specific D N A sequences using a radioactively labelled cDNA-insert of class I cDNA clone pFS1968. Two out of several ,k-clones identified initially by high signori'intensity in plaque hybridization were purified and further analysed. They~turned out" to be identical in their restriction fragment patterns and in hybridi'zation patterns obtained for EcoRI, HindIII and E c o R I / H i n d I I I digests with probe 1968. One clone (Jkhs68-7)is shown in Fig. 2A: The restriction map of this clone for the respective enzymes and the maps of two subclones of the soybean~genomic D N A in pBR322 are shown i n Fig. 2B. EU]'ther southern blot analysis of)~hs687'DNA using otherclass 1 and class II c D N A probes
273
Fig. 2. Characterization of genomic D N A clone Ahs68-7, carrying hs-specific sequences. A: Fragment and hybridization patterns for digests of Xhs68-7 with EcoRl (E), Hindlll (H) and EcoRI/Hindlll (EH). The three lanes on the left show the ethidium bromide staining of fragments, the three lanes on the right show the autoradiograph of a blot hybridized with probe 1968. B: Maps of genomic clone Ahs68-7 and two plasmid subclones in pBR322. Boxed portions of the maps represent soybean genomic DNA, solid line the A-vector, dashed line the pBR322 vector DNA. C: Specificity of Ahs68-7 D N A for different hs c D N A clones by dot blot hybridization. Identical concentration series of immobilized DNA (a: 0.1 ~g; b: 0.05 #g; c: 0.025 #g; d: 0.0125 ~g) were hybridized at different temperatures (as indicated) with different c D N A probes ((1): 1968; (2): 1920; (3): 2059; (4): 53) and auloradiographed.
(see Table 1), revealed sequence homologies of both EcoRI/HindIII fragments of the genomic D N A with all class I probes (identical patterns to that shown in Fig. 2A), but not with class II probes (data not shown). It is important to note that neither of the c D N A clones used as probes in this study contains a HindIII site (unpublished observation). Hence, there are at least two different locations of class I homologous DNA-sequences on khs68-7. The specificity of this genomic D N A clone for c D N A clone 1968 was compared with other crosshybridizing c D N A clones by dot blot hybridization. Three different temperatures (45, 55 and 65 ~ were used in this experiment to vary the
stringencies of hybridizations between ,khs68-7 (identical dilution series spotted onto nitrocellulose filters) and four different cDNA-probes (1968, 1920, 2059, 53). The results shown in Fig~ 2C demonstrates the higher specificity of )~hs68-7 for c D N A 1968 by hybridization up to 65 o C (for concentration a), compared to the hybridizations with the other cDNAs which are less temperature stable. However, their cross-hybridization under normal stringency (45 ~ C) with ,khs68-7 D N A indicates the close relationship of lmw-hsp genes a n d / o r multiple locations of hs-specific D N A sequences at this locus.
274
Assignment o f a specific hs-protein f o r class I c D N A clone pFS1968 Standard hybrid selection/translation protocols were insufficient to d e m o n s t r a t e a specific gene/ p r o t e i n c o r r e l a t i o n for class I hsp's (11). This c o m p l i c a t i o n in m o l e c u l a r studies, caused by the sequence h o m o l o g y o f class I genes, is resolved for D N A : D N A h y b r i d i z a t i o n s as described above. A m o r e specific selection of h s - m R N A s for hybrid release t r a n s l a t i o n s was o b t a i n e d by v a r y i n g the h y b r i d release t e m p e r a t u r e . C o n d i t i o n s for h y b r i d selection using c D N A clone 1968 were identical to those d e s c r i b e d by Sch6ffl & Key (1 I) for the same clone. H o w e v e r , the h y b r i d i z e d p o l y ( A ) m R N A was g r a d u a l l y released by increasing t e m p e r a t u r e s to 50, 60, 70 and 85 ~ C. The elution p a t t e r n of trans-
latable R N A s was d e t e r m i n e d by assaying for 3Hleucine i n c o r p o r a t i o n into proteins synthesized in vitro in the wheat g e r m system using the m R N A s of the i n d i v i d u a l fractions as templates. In this experiment, a b o u t 40% of the total i n c o r p o r a t i o n resulted f r o m R N A eluted at 50 o C, 30% from 60 ~ R N A , 20% f r o m 7 0 ~ and 10% f r o m 8 5 ~ The p r o t e i n s synthesized f r o m the 4 different R N A fractions were s u b s e q u e n t l y a n a l y s e d by 2D-gel a n a l y sis. F i g u r e 3 shows the c a t a l o g u e of f l u o r o g r a m s of s e p a r a t e d p r o t e i n s for the different temperatures. The m a j o r spots in all f o u r f l u o r o g r a m s correlate with l m w - h s p ' s of the s o y b e a n (11). At 50 ~ a n u m b e r of a d d i t i o n a l m i n o r spots a p p e a r which do not belong to the s t a n d a r d p a t t e r n of class I hsp's, as synthesized f r o m 60 * C R N A . A t 70 ~ C, R N A for m a i n l y three hsp's out of the s t a n d a r d p a t t e r n is
Fig. 3. Fluorograms of in vitro synthesized proteins separated on 2D-gels. PolY(A)mRNAs, selected by hybridization to cDNA clone 1968, were eluted stepwise from hybrids at 50, 60, 70 and 85 ~ and subsequently~ranslated in a wheat germ system in the presence of [3H]-leucine. The molecular weight standards used in the S DS-d[rnensiOn of gel electrophoresis are shown on the left side of each gel. Their molecular weights are fro m top to bottom: 92 500 (degraded); 69 000~.46 000; 30 000; 12 300 dalton, q~hearrowhead marks hsp68.
275 released; 85 ~ RNA reduces this pattern even more to only one major hsp, defined as hsp68. This experiment shows that by the criterion of thermal stability of m R N A : c D N A hybridization, a specific gene/protein correlation can be obtained. The data also further allow subclassification of class I genes/ proteins by identifying two hsp's which are closely related to hsp68 (the two additional proteins in the 70 ~ pattern). Consequently, the 10 remaining members of class I have to be considered as more distantly related to hsp68.
hybridize to duplex D N A by displacing the identical strand and forming R-loops (25, 26). By measuring and comparing the position of R-loops on these two plasmids, it was possible to locate them unambiguously with respect to the restriction map of the plasmids as shown in Fig. 4A and B. The loop distributions on both plasmids determine only one single location, starting about 1 kb away from the H i n d I I I site, with an average loop size of 0.45 kb. This is just about the size needed for a coding capacity of a lmw-hsp, but only half the size of the m R N A s found for these proteins (11). This apparent discrepancy may result either from an overestimate of the RNA-size due to the D N A standards used for its determination ( l 1) or from an unavoidable underestimate of loop sizes due to the partial sequence homology of class I mRNAs. The participation of heterologous RNAs in R-loop formation at this locus is indicated by the large variations in size of R-loops (0.5 +0.13 kb and 0.4 +0.17 kb) contributing to the histograms in Fig. 4. The polarity of the transcripts at this gene locus was deter-
R-loop m a p p i n g o f a class I hs-gene on subclones o f Xhs68-7
A hs-specific DNA region on genomic clone hhs68-7 has been precisely mapped by electron microscopy of R-loops using the plasmid subclones phs6810 and phs6871 (see Fig. 2B). E c o R I digested phs6810 and H i n d I I I digested phs6871 were separately incubated with total poly(A)RNA from hshypocotyls under conditions, which allow R N A to
E
H i
E!
0 loop size distances from E
.10
.
.
.
.
w
A
0.5+ 0.13kb 3.95+_0.27kb 5.13+0.40kb
l-
9
_e 0
E
loop s i z e distance from H
-10
0
0.41_0.17
B
kb
1. O0 __.0 . 4 0 kb
c"
map !
i H
9
,
,
.
|
9
9
9
9
w
.
.
.
.
!
.
.
.
.
units 1
9
-~-
0
E
H
Fig. 4. Histogramsof R-loopslocalized on subclones of soybeangenomicDNA carryinghs-genes. R-loops,formedby hybridizationof hs-poly(A)RNA on the 10.5 kb EcoRI fragment of phs6810 (A) and on HindIII linearized phs6871 (B) were visualized by electron microscopy, measured and aligned to each other with respect to their commonrestriction sites E for EcoRI and H for HindII1.
276
Fig. 5. Electron micrograph of R-loops identifyingthe coding region and polarity of a soybean hs-gene. Soybeanhs-poly(A)RNAwas hybridized to EcoRI digested phs-6810(A) and HindII1 digested phs6871 (B). The 3' ends of RNAsin R-loopswerelabelled by attaching circular dT-tailed @X174RFII-DNAcirclesto the poly(A)-tails.The line drawings showthe interpretation of R-loopstructures for both molecules.
mined by labelling the 3' poly ends of RNA in R-loops by dT-tailed @X174RF|I-DNA (27 29). The result for the two different types of R-loops is shown in Fig. 5. In both cases, @X174-markers are attached to the same ends relative to the maps in Fig. 4. This result defines the 5' terminus at the proximal end and the 3' terminus at the distal end of R-loops with respect to the H i n d l I I site. The gene does not contain introns detectable by this method. R-loop formation Was not observed in experiments when the hs-poly(A)RNA fraction was replaced by total control poly(A)RNA, indicating the hs-specific activation of this gene.
Discussion
The data presented in this paper confirm the
predictions made from earlier observations (11) on the existence of a multigene family for part of the lmw-group of soybean hsp's. This was shown by southern blot hybridization, where about 6 7 different E c o R I fragments could be identified as potential bearers of class I hs-genes for a total of 13 different proteins. The finding that fewer fragments hybridize than there are genes available suggests a certain clustering of hs-genes. One indication for such an organization of related genes on genomic DNA is provided by the hybridization analysis of clone ~.hs68-7. This clone carries a hs-gene, identified by R-loop mapping, in the middle of a 10.5 kb E c o R I fragment and another portion of hs-gene sequences is located on this fragment at the end proximal to the HindIII-site, possibly cut by E c o R I (see Fig. 2B, unpublished observation). It is not yet known to what degree clusters of class I genes are
277 interspersed with other hs-genes. Similar organization of hs-genes has been described for Drosophila at different chromosomal locations, such as, the multiple gene copies for hsp70 at 87 A/87 C (30 34) or the four different genes hsp22, hsp23, hsp26 and hsp27 organized at the 67B locus (27, 35, 36). The hs-genes at the 67B locus are interspersed with three additional genes of unknown function; differential expression of genes at this cluster is under developmental control (38, 39). Drosophila hs-genes appear to have no intervening sequences, with the exception of the hsp83 gene (35, 40). This general feature holds true also for the only other hs-gene of higher eucaryotic origin characterized so far (this paper), but its seems to apply for other soybean hs-genes of the lmw-hsp group as well (unpublished data). The relatively broad variation in R-loop sizes described in this paper is interpreted as the result of partial sequence homology of this gene with other class I hsmRNAs, which compete in hybridization at the same gene locus. The asymmetrical distribution of the R-loops at this site suggest a preference for the 5' half of c o m m o n sequences on class I genes. The methods applied in the studies specifying gene/protein correlation of hsp68 can be generally used for further subclassification of the class 1 multigene family based on sequence relationships. Final proof of these predictions has to come from DNA sequence analysis to show the structural and functional significance of'hs-domains' and to provide the basis for investigating the regulation of hs-gene expression.
Acknowledgements We thank Dr P Nagao for providing a soybean genomic library and his assistance in screening. We also wish to acknowledge the excellent technical assistance of Sieglinde Angermtiller, G0tz Baumann and Cheryl Mothershed. We are grateful to Drs V. Krishnapillai and M. O'Connell for critical reading and Susanna Malmivaara for typing of the manuscript. This work was supported by the Deutsche Forschungsgemeinschaft (grant Scho 242-4).
References l. Ashburner M, Bonnet J J: The induction of gene activity in Drosophila by heat shock. Cell 17:241-254, 1979. 2. Schlesinger M J, Ashburner M, Tissibres A: Heat shock, from bacteria to man. Cold Spring Harbor Laboratory, 1982. 3. Mitchell HK, Moiler G, Peterson NS, Lipps-Sarimento L: Specific protection from phenocopy induction by heat shock. Dev Genet 1:181-192, 1979. 4. McAllister L, Finkelstein DB: Heat shock proteins and thermal resistance in yeast. Biochem Biophys Res C o m m u n 93:819-824, 1980. 5. Loomis WT, Wheeler S: Heat shock response of Dictostelium. Dev Biol 79:399 408, 1980. 6. Key JL, Lin CY, Ceglarz E, Sch6ffl F: The heat shock response in soybean seedlings. In: Ciferri O, Dure LS (eds) The Structure and Function of Plant Genomes, in press. 7. Li GC, Werb Z: Correlation between synthesis of heat shock proteins and development of thermotolerance in Chinese hamster fibroblasts. Proc Natl Acad Sci USA 79:3218-3222, 1982. 8. Key JL, Lin CY, Chen YM: Heat shock proteins of higher plants. Proc Natl Acad Sci USA 78:3526-3530, 1981. 9. Barnett T, Altschuler M, McDaniel CN, Mascarenhas JP: Heat shock induced proteins in plant cells. Dev Gen 1:331-340, 1980. 10. lngolia TD, Craig EA: Four small Drosophila heat shock proteins are related to each other and to mammalian c~-cryst a l l m Proc Natl Acad Sci USA 79:2360 2364, 1982. 1 l. Sch~ffl F, Key J L: An analysis of m R N A s for a group of heat shock proteins of soybean using cloned cDNAs. J Mol Appl Genet 1:301-314, 1982. 12. Key JL, Lon CY, Ceglarz E, SchOffl F: The heat shock response in plants: Physiological considerations. In: Schlesinger M J, A s h b u r n e r M, Tissibres A (eds) Heat shock, From Bacteria to Man. Cold Spring Harbor la'boratory, 1982, pp 329-336. 13. Kushner SR: An improved method for transformation in Escherichia coli with ColE1 derived plasmids. In: Boyer HW, Nicosia S (eds) Genetic Engineering. Elsevier/North Holland Biomedical Press, Amsterdam. 1978, pp 17-23. 14. Maniatis T, Fritsch EF, Sambrook J: Molecular Cloning, a Laboratory Manual, Cold Spring Harbor Laboratory, 1982. 15. Czarnecka E, Edelman L, Sch/3ffl F, Key Jk: Analysis of physical stress response in soybean using cloned heat shock cDNA's. 1983 (submitted for publication). 16. Nagao RT, Shah DM, Eckenrode VK, Meagher RB: Multigene family of actin-related sequences isolated from a soybean genomic library. D N A 2:1 9, 1981. 17. Bazaral M, Helinski DR: Circular forms of colicinogenic factors El, E2 and E3 from Escherichia coll. J Mol Biol 36:185-194, 1968. 18. Scb6ffl F, Ptihler A: The plasmid system of Escherichia coli strain U R 12644: identification and molecular characteristics of transpnsons involved in the generation of endogeneous R-plasmids. Genet Res Camb 34:287-301, 1979.
278 19. Williams BW, Blattner FR: Construction and characterization of the hybrid bacteriphage lambda Charon vectors for D N A cloning. J Virol 29:555-575, 1979. 20. McDonell MW, Simon MN, Studier FW: Analysis of restriction fragments of T7 D N A and determination of molecular weights by electrophoresis in natural and alkaline gels. J MolBio1110:119 146, 1977. 211Rigby PWJ, Dieckmann M, Rhodes C, Berg P: Labelling deoxyribonucleic acid to high specific activity in vitro by nick translation with DN A polymerase 1. J Mol Biol 113:337-351, 1977. "22. Southern EM: Detection~f specific sequences a m o n g DNA fragments. J MoI Biol 98:'503-517, 1975. 23. Roberts BE, Paterson BM: Efficient translation of tobacco mosaic virus R N A and rabbit globin RNA in a cell free system from commercial wheat germ. Proc Natl Acad Sci USA 70:2330 2334, 1973. 24. O'Farrell PH: High resolution two-dimensional electrophoresis of proteins. J Biol Chem 250:4007-4021, 1975. 25. T h o m a s M, White RL, Davis I~W: Hybridization of R N A to double-stranded DNA: formation of R-loops. Proc Natl Acad Sci USA 73:2294 2298, 1976. 26. Kaback DB, Angerer LM, Davidson N: Improved methods for the formation and stabilization of R-loops. Nucleic Acids Res 6:2499 2517, 1979. 27. Bender W, Davidson N: Mapping of poly(A) sequences in electron microscope reveals unusual structure of type C oncornavirus R N A molecules. Cell 7:595 607, 1976. 28. Fisher RL, Goldberg RB: Structure and flanking regions of soybean seed protein genes. Cell 29:651 660, 1982. 29. Bender W, Davidson N, Kindle KL, Taylor WC, Silverman M, Firtel RA: The strticture of M6, a recombinant plasmid containing Dictostelium D N A homologous to actin messenger RNA. Cell 15:779 788, 1978. 30. Livak K J, Freund R, Schweber M, Wensink PC, Meselson M: Sequence organization and transcription at two heat shock loci in Drosophila. Proc Natl Acad Sci USA 75: 5613-5617, 1978. 31. Lis JT, Prestidge L, Hogness DS: A novel arrangement of
tandemly repeated genes at a major heat shock site in D. melanogaster. Cell 14:901-919, 1978. 32. Schedl P, Artavania-Tsakonas S, Steward R, Gehring W, Mirault M-E, Goldschmidt-Clermont M, Moran L, Tissi~res A: Two hybrid plasmids with D. melanogaster D N A sequences complementary to m R N A coding for the major heat shock protein. Cell 14:921 929, 1978. 33. Artavanis-Tsakonas S, Schedl P, Mirault M-E, Moran L: Genes for the 70 000 dalton heat shock protein in two cloned D. melanogaster DNA segments. Cell 17:9-18, 1979. 34. Craig EA, McCarthy BJ, Wadsworth SC: Sequence organization of two recombinant plasmids containinggenes for the major heat shock protein of D. melanogaster. Cell 16: 575 588, 1979. 35. Corces V, Holmgren R, Freund R, Morimoto R, Meselson M: Four heat shock" proteins of Drosophila melanogaster coded within a 12 kilobase region in chromosome subdivision 67B. Proc Natl Acad Sci USA 77:5390 5393, 1980. 36. Wadsworth SC, Craig EA, McCarthy BJ: Genes for three Drosophila heat schock induced proteins at a single locus. Proc Natl Acad Sci USA 77:2134 2137, 1980. 37. Voellmy R, Goldschmidt-Clermont M, Southgate R, Tissi~res A, Levis R, Gehring W: A D N A segment isolated from chromosomal site 67B in D. melanogaster contains four closely lined heat shock genes. Cell 23:261 270, 1981. 38. Sirotkin K, Davidson N: Developmentally regulated transcription from Drosophila melanogaster chromosomal site 67B. Dev Biol 89:196 210, 1982. 39. Z i m m e r m a n n JL, Petri W, Meselson M: Accumulation of a specific subset of D. melanogaster heat shock m R N A s in normal development without heat shock. Cell 32:1161-1170, 1983. 40. Holmgren R, Corces V, Morimoto R, Blackman R, Meselson M: Sequence homologies in the 5' regions of four Drosophila heat shock genes. Proc Natl Acad Sci USA 78: 3775 3778, 1981. Received 24 May 1983; in revised from 22 August 1983; accepted 30 August 1983.
