Apr 23, 1984 - ical Research Support Group No. RR07071. This is Paper 9332 of the. Journal Series of the North Carolina Agricultural Research Service,.
Vol. 259, No. 19, Issue of October 10,pp. 12007-12013,1984 Printed in U.S.A.
THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 19% by The American Society of Biological Chemists, Inc.
High-mobility Group Chromosomal Proteins of Wheat* (Received for publication, April23, 1984)
Steven Spiker From the Genetics Department, North CarolinaState Uniuersity, Raleigh, North Carolina27695-7614
can selectively retain rnononucleosomes which contain DNA Fourproteinshavebeen extracted frompurified chromatin of wheat embryos with 0.35 M NaCl. These complementary to messenger RNA. The manner in which proteins are soluble in 2% (w/v) trichloroacetic acid HMG proteins bindto nucleosomes has been studied inmany and thus meet the original operational requirements laboratories to (see Weisbrod, 1982a and Igo-Kemenes et al., be classified as ”high-mobility group”(HMG) chromo- 1982 for review). Because all nucleosomes apparently have somal proteins. The proteins have been characterized two high-affinity HMG binding sites (Mardian et al., 1980; by one- and two-dimensional electrophoresis, amino Sandeen et al., 1980), it is uncertain how the postulated acid analysis, and peptide mapping. Threeof the pro- specificity in HMG-nucleosome binding might occur. Length teins (HMGb, c, and d) share the mammalian HMG of DNA in mononucleosomes isolated in vitromay beinvolved characteristic of being rich in both acidic and basic (Swerdlowand Varshavsky, 1983; Stein andTownsend, 1983) amino acid residues. Unlike their putativemammalian but this still leaves open the question ofhow postulated counterparts, these plant HMG proteins contain less specific binding may occur in uiuo. than 7 mol % proline. Thefourthwheatprotein (HMGa) is rich in both proline and in basic amino acidThe fact that we still do not know thetrue biological residues. This wheat protein, however, contains only functions of HMG proteins has been highlighted by recent about half the proportion of acidic residues found in reports which question their long-assumed roles as structural characteristic also proteins of actively transcribed nucleosomes (Seale et al., mammalian HMG proteins-a found in the trout testis HMG protein, H6. Comparative 1983; Reeves and Chang, 1983). HMG proteins are, in fact, peptide maps show that none of the wheat HMG pro- defined on an operational basis, not on a functional basis teins are degradation products of other HMG proteins (Johns, 1982). The proteins we now call HMG proteins were first noticed as impurities in histone H1 preparations from or the H1 histones. The peptide maps have not, however, been useful in establishing homologies mamwith calf thymus (Johns, 1964). According to the original operamalian HMG proteins. Wheat HMG proteins are re- tional definition presented by Goodwin et al. (1973), HMG leased from DNase I-treated nuclei and co-isolatewith proteins are 1) chromatin proteins, 2) extracted from chromicrococcal nuclease-sensitive chromatinfractions. matin by 0.35 M NaCl, and 3) soluble in 2% (w/v) trichloroHMG proteins of acetic acid. These proteins migrate rapidly in polyacrylamide Similar observations concerning the vertebrateanimalshavebeenconsideredconsistent gel electrophoresis systems; hence, their designation as “highas structural components mobility group” proteins. Further characterization of these with a role for these proteins of actively transcribed chromatin. proteins revealed certain unusual characteristics which Johns (1982) has added to theoperational definition of calf thymus HMG proteins.Theseare 4) high in basic amino acidsapproximately 25mol %, 5) high in acidic amino acidsThe HMG’ chromosomal proteins have attracted consid- approximately 25 mol %, and 6) relatively high in proline-7 erable attention inrecentyears because, in part, of their mol % or more. The operational definitionsof HMG proteins postulated roles as structuralproteins of nucleosomes of ac- are expedient but, as has been stressed by Johns (1982) and tively transcribed genes (reviewedby Weisbrod, 1982a). Chro- Mayes (1982), care shouldbe taken when such definitions are matin containing DNA sequences complementary to messen- applied to organisms and tissues other thancalf thymus. The gerRNA loses its selective sensitivity to DNaseI when operationaldefinitions should aid, nothinder establishing extracted with 0.35 M NaC1-a concentration which removes functional definitions for the HMG proteins. An illustration HMG proteins. Moreover, the selective sensitivitycan be of why the operational definitions cannot be strictly applied restored by reconstituting salt-stripped chromatin with the is provided by the trout testis protein H6. This protein shows 0.35 M NaC1-extracted proteins or with purified HMG14 or striking sequence homology to calf thymus HMG14 and 17 17 (Weisbrod and Weintraub, 1979; Weisbrod et al., 1980; and is almost certainly the functional equivalent of these Gazit et al., 1980). Weisbrod (1982b) and Weisbrod and Wein- mammalian HMG proteins (Dixon, 1982). However, the trout traub (1981) have demonstrated that anHMG affinity column protein does not meet the fifth requirement to be an HMG * This work was supported in part by National Science Foundation protein. It contains only about 13 mol % glutamic plus asparGrant PCM81-05135 and National Institute of Health Grant Biomed- tic acid residues. Proteins thathave properties similar to mammalian HMG ical Research Support Group No. RR07071. This is Paper 9332 of the Journal Series of the North Carolina Agricultural Research Service, proteins have been found in many organisms (Mayes, 1982; Raleigh, NC27695. The costs of publication of this article were Bassuk and Mayfield, 1982; Hamana and Iwai, 1979; Katula, defrayed in part by the payment of page charges. This article must 1983; Marquez et al., 1982; Spiker et al., 1978; Weber and therefore be hereby marked ‘‘aduertkernent” in accordance with 18 Isenberg, 1980). However, the properties of these proteins U.S.C. Section 1734 solely to indicate this fact. ‘The abbreviations used are: HMG, high-mobility group; Ph- correspond in varying degrees to those of calf thymus HMG MeS02F, phenylmethanesulfonyl fluoride; SDS, sodium dodecyl sul- proteins and it is open to question whether they should be fate; HPLC, high-performance liquid chromatography. considered HMG proteins. We have previously reported the
12007
Wheat HMG Proteins
12008
existence of four chromosomal proteins from wheat embryos which meet the salt extractability and acid solubility requirements to be termed “HMG” proteins (Spiker et al., 1978). One of these (wheat HMGb) was additionally shown to be HMG-like initscontent of basic and acidic amino acid residues. We have now further characterized the wheat HMG proteins by amino acid composition, molecular weight, peptide maps, and association with putatively transcriptionallyactive regions of the genome. Comparative amino acid compositions and peptide maps indicate that each of the four plant HMG proteins are genuine proteins and not degradation artifacts of other HMG proteins or H1histones. Amino acid compositions and peptide maps were not conclusive indicators of which of the plantHMG proteins might correspond to thehigh-molecular-weight mammalian HMGl and 2 or the low-molecularweight mammalian HMG 14 and 17. Wheat HMG proteins b, c, and d meet the requirements for being termed HMG proteins in that they are high in both acidic and basic amino acid residues. However, the proline contents of these are all less than 7 mol %. Wheat HMGa meets the requirement of being relatively high in proline (12 mol %) but, like trout H6, it is relatively low in acidic amino acid residues-about 12 mol %. The plant HMG proteins share the characteristics of vertebrate HMG proteins of being released from nuclei treated with DNase I and co-isolating with the susceptible fraction of chromatin obtained by micrococcal nuclease treatment. EXPERIMENTALPROCEDURES
Source of Tissue-Wheat embryos were obtained from General Mills (Minneapolis, MN) or were mechanically isolated from dry, mature grains (Triplett, 1979). The embryos from both sources yielded chromatin with indistinguishable properties. Thymus glands were obtained from 5-week-old pigs and placed on dry ice within 5 min after slaughter. Isolation of Chromatin-Chromatin was isolated from wheat embryos by the method of Simon and Becker (1976) with the following modifications. All solutions contained 0.1 mM PhMeS02F and12 mM NaHS03. A Polytron homogenizer (Brinkmann) was used to disrupt the tissue. The final step inthe purification was centrifugation of the chromatin through 1.7 M sucrose, 15 mM 2-mercaptoethanol, 10 mM Tris-HC1 (pH8), 0.5% Triton X-100 (Rohm andHaas), 12 mM NaHS03,O.l mM PhMeS02Ffor 30 min at 16,000 X g. All procedures were carried out a t 0-2 “C. Isolation of HMG Proteins-HMG proteins were isolated from wheat embryos essentially by the methods used by Goodwin et al. (1973) to isolate calf thymus HMG proteins. These methods are the basis of the operational definition of HMG proteins, i.e. nonhistone proteins extracted from purified chromatin with 0.35 M NaCl and soluble in 2% trichloroacetic cid. Purified wheat embryo chromatin was washed twice in 75 mM NaCl, 25 mM Na-EDTA (pH 7.5), 12 mM NaHS03, 0.1 mM PhMeSOzF. The chromatin pellet was then homogenized in 0.35 M NaC1, 1 mM Tris-HC1 (pH 8). 0.1 mM PhMeS02F using a glass-Teflon homogenizer. Pellet and supernatant fractions were obtained by centrifugation at 12,000 X g for 10 min. The extraction procedure was repeated and the supernatants combined. The combined supernatants were made 2% (w/v) trichloroacetic acid by slow addition of 100% (w/v) trichloroacetic acid. After standing on ice for 30 min, the solution was centrifuged (12,000 X g, 10 min) to remove insoluble material. The supernatant fraction was made 10 mM 2-mercaptoethanol and 0.015 volume of 30% NH,OH was added. All operations to this point were carried out at 0-2 “C. After adding 3 volumes of acetone to thesolution and allowing it to stand overnight a t 5 ‘C, the precipitated HMG proteins were removed by centrifugation, washed once in acetone, 0.1 N HCl (6:1), twice in acetone, and dried in ~ ( I C U Oa t room temperature. Typical yields were 10 mgof HMG proteins/300 g of dry embryos (2-5% of histone yield). HMG proteins were extracted from frozen pig thymus essentially by the method of Sanders (1977). Tissue washomogenized in 10 volumes of0.75 N HCIO,,0.1 mM PhMeS02F using a Polytron homogenizer. The homogenate was centrifuged (12,000 X g, 10 min) and thepellet fraction re-extracted with an additional 10 volumes of 0.75 N HClO,. The combined supernatants were filtered, 3.5 volumes
of acetone was added, and the solution was made 0.07 N HCl by adding concentrated HC1. The resulting H1 histone precipitate was removed by centrifugation and an additional 2.5 volumes of acetone (based on supernatant volumebefore first acetone addition) was added to the resulting supernatant. The HMG precipitate (contaminated by some H1 histone) was allowed to form overnight at 5 “C, collected by centrifugation, then washed and dried in the same manner as were the wheat HMG proteins. Polyacrylamide Gel Electrophmesis-SDS-l8% polyacrylamide gels were run as previously described (Spiker, 1982). Two-dimensional electrophoresis of wheat HMG proteins was carried out essentially according to O’Farrell et al. (1977) using nonequilibrium pH gradient electrophoresis in the first dimension and SDS gels (as above) in the second dimension. Isolation of HMG Fractions-Wheat HMG fractions a, b, c, and d were isolated by preparative electrophoresis (Spiker and Isenberg, 1983) or by HPLC. HPLC was carried out using a combination of reverse-phase (Waters Microbondapak C-18) and cation exchange (Brownlee Labs CX-300). Details will be presented elsewhere. Amino Acid Analysis-Amino acid analysis of purified HMG fractions was carried out using the continuous gradient elution method of Klapper (1982). Peptide Mapping-Peptide maps were obtained by hydrolysis of proteins in gel slices by the method of Cleveland et al. (1977) as modified by Luna et al. (1979). A further modification of this system (Spiker, 1980) allowed 3-mm-thick gel slices to be placed in 3-mm wells of a stacking gel along with the proteolytic enzymes and the resulting peptides to be separated on a 1.5-mm-thick running gel. In the work presented in this paper, a similar modification was used except that the 3-mm-thick stacking gel was poured over a 0.5-mm running gel and a “mini gel” apparatus as described by Matsudaira and Burgess (1978) was used. Salt Extraction of HMG Proteins-The amount of wheat HMG proteins extracted by various concentrations of NaCl wasdetermined as follows. Wheat embryo chromatin purified by centrifugation through 1.7 M sucrose as described above was washed with 75 mM NaCI, 1mM Tris-HC1 (pH 7), 0.1 mM PhMeS02F.An even suspension of chromatin in this buffer was divided into measured aliquots and adjusted to the desired NaCl concentration. Each suspension was then layered over a 15%sucrose solution containing the desired NaCl concentration and centrifuged a t 47,000 rpm in a Ti 60 rotor (Beckman) for 22 h. The pellets were suspended in water, extracted with 0.4 N H~SOI and the acid-soluble proteins precipitated by dialysis against ethanol. The supernatants were exhaustively dialyzed against water containing 0.1 mM PhMeS02F,made 0.4 N &SO4, and further dialyzed against ethanol to precipitate the proteins. PolyacrylamideSDS gels were then run, stained with Coomassie Blue and scanned to quantitate DNA-bound protein from the pellets and free proteins from the supernatants. RESULTSANDDISCUSSION
Electrophoretic Mobility and MolecularWeight Estimution-The electrophoretic mobilities of the wheat HMG proteins arecompared to those of mammalian (pig thymus) HMG proteins in the SDS gel in Fig. 1. In general, the electrophoretic mobilities are similar but no wheat HMG matches any pig HMG in mobility. In Fig. 2 the mobilities of wheat HMG proteins on SDS gels are compared to those of ovalbumin, carbonic anhydrase, myoglobin, cytochrome c, and ubiquitin. Assuming molecular weights of 45,000, 29,000, 17,200, 11,700, and 8,500, respectively, for these standards,the molecular weights of the wheat HMG proteins are calculated to be a = 24,300, b = 20,400, c = 17,400, and d = 14,500. Histones, which have a high proportion of basic residues, migrate anomalously on SDS gels (Panyim and Chalkley, 1971). This anomalous migration leads to an overestimate of their molecular weights when typical molecular weight standards are used for comparison. Like the histones, HMG proteins have a high proportion of basic amino acid residues; but, because they are also high in acidic residues, it is uncertain what standards are appropriate for molecular weight estimation.When pig thymus HMG proteins are used for standards, the molecular weights of the
Wheat HMG Proteins
.-m
c
12009
b
a
U
a l
“ r ” H1 1 2’
“a
14 \
“b
17-
“c
d
-
C
“d
FIG.1. Sodium dodecyl sulfate-polyacrylamide gels of wheat embryo and pig thymus HMG proteins. Wheat embryo and pig thymus HMG proteinswere isolated and separated by SDSpolyacrylamide gel electrophoresis as described in the text and stained with Coomassie Blue. Pig HMG proteins 1, 2, 14, and 17 are labeled and co-migrate with their counterparts from calf thymus (Goodwin FIG.3. Wheat HMG fractions purified by high-performance et al., 1978). Contaminating pig thymusH1 histones are also labeled. liquid chromatography. Wheat HMG proteins a, b, c, and d were Wheat HMG proteins a, b, c, and d are labeled. purified by HPLC asdescribed in the text. Inpan& a-d, respectively, the purified proteins were separated by SDS electrophoresis along with an unfractionated HMG preparation. Scans of the gels showed all preparations to be more than95% pure. TABLE I
- 1
Amino acid content of HMG proteins fromcalf thymus and wheat embryos
-2
Residue
c”
d’
“ d
3 .4 ”5
FIG.2. Molecular-weight estimation ofwheat HMG protein. Sodium dodecyl sulfate-polyacrylamide gels were used as described in the text to electrophoretically separate wheat HMG proof known molecular teins (lefttwo lanes) andstandardproteins weights (right two lanes). 1 = ovalbumin, 45,000; 2 = carbonic anhydrase, 29,000; 3 = myoglobin, 17,200; 4 = cytochrome c, 11,700; 5 = ubiquitin, 8,500.
Calf Wheat thymus”
9.3 10.7 8.1 12.0 Asx 5.2 2.7 3.22.51.4 3.7 1.2 4.2 Thr 5.0 7.4 7.8 2.3 Ser 12.0 15.3 12.7 5.3 10.5 17.1 Glx 18.1 17.5 7.05.83.6 8.9 4.7 12.6 12.9 8.5 Pro 8.6 11.2 6.5 5.3 6.513.7 GIY 11.3 15.9 20.3 18.4 14.5 9.0 8.1 14.8 Ala 5.42.3 4.71.93.6 3.2 2.0 4.2 Val Trace Trace 0.7 CYS 2.3 1.5 0.4 Met 0.5 Ile 1.8 1.3 2.0 2.22.2 4.5 1.0 2.0 Leu TYr 2.0 2.9 0.4 3.1 0.6 3.8 2.7 Phe 3.03.6 15.9 15.9 17.0 13.4 24.3 19.0 21.3 19.4 LYS 0.2 1.3 1.7 0.3 2.0 0.3 His 0.9 2.74.7 4.23.92.5 7.1 4.1 5.6 ‘4% 24.1 19.5 20.5 28.4 24.6 Lys + Arg 25.2 22.5 Asx + Glx 28.8 26.8 25.2 “Goodwin et al. (1978).
wheat HMG proteins are calculated to be a = 23,100, b = 14,500, c = 10,000, and d = 6,300. Purification of Fractions-Wheat HMG fractions were purified by preparative electrophoresis (Spiker and Isenberg, 1983) or by HPLC. In the SDS gels in Fig. 3, the purity of the fractions obtainedby HPLC is demonstrated.A two-step procedure is used. In the first step, a cation-exchange column at pH 6 and a NaCl gradient separates HMGa and b from HMGc andd. In thesecond step, a reverse-phase C-18 column separates HMGa from HMGb and HMGc from HMGd. In most cases, no impurities are detected even on overloaded gels. Some contaminationof HMGb with HMGc can be noted in panel b of Fig. 3. Scans of Coomassie Blue-stained gels show that HMGb is more than 95% pure. Different preparations of mammalian HMG proteins are contaminated with H 1 histones to various degrees (Nicolas and Goodwin, 1982). Such is also the case with wheat HMG preparations which
embryo
HMGl HMG2 HMG14 HMG17 HMGa HMGb HMGc HMGd mol % mol %
7.2
13.9 11.7
7.0
8.3
1.7 2.7 1.9 0.7
10.7 8.2
7.6
9.2
11.8
1.8
0.4 1.5 0.8
0.2 1.1 2.3
1.4 1.2
20.1
18.6
12.5 22.7 29.2 24.4
can be seen as thehigh-molecular-weight band in the unfractionated HMG standards inFig. 3. Amino Acid Compositions and Comparison with Calf HMG Proteins-The amino acid compositions of the four wheat HMG proteins and the four calf thymus HMG proteins are shown in TableI. The mole per centof lysine plus arginine is about 20% for each of the wheat HMGproteins. These values are certainlyhigh but not ashigh as thosefor the calf thymus HMG proteins. The mole percentages for the acidic residues
Wheat HMG Proteins
12010
obviously related for wheat HMGb,c, and d are well within the rangesof those even the “weak” test for relatedness, they are of the calf thymus HMG proteins. Wheat HMGa, however, by sequence. The sequences of these two proteins can be eachproteinresultingin a44% has only 12.5% acidic residues. This low percentage of acidic aligned with twogapsin residues does not conform to the definitions of HMG proteins sequence homology. This homology includesa stretch in proposed by Johns (1982) for calf thymus. However, in light which no gaps are necessary to generate 12 consecutive and of a similar low percentage of acidic residues in trout testis 19 of 22 sequence identities. Conversely, calf thymus HMG2 H6, which is almost certainly an HMG (Dixon, 1982), the and HMG14 meet the“weak” test for relatedness even though possibility that this wheat protein should be considered an there is little sequence homology between the two proteins. HMG protein cannot be ruled out. In fact, if the last charac- Calf HMG2 (Walker, 1982) and HMG14 contain only two teristic of calf thymus HMG proteins, proline content, is an common sequences with as many as four consecutive amino important diagnostic, HMGa is the only wheat HMG which acid residues. meets the criterion of a t least 7 mol % set by Johns (1982). As can be seen in Fig. 4, wheat HMGb and HMGd meet Wheat HMGa has 12.6 mol % proline. The others have less the “strong” test criterion for being related. Wheat HMGc than 7%. meets the “weak” test criterion for being related to wheat In order to makea comparison of amino acid compositions HMGbandHMGd.WheatHMGcandHMGd meet the of HMG proteins which would take all the amino acids into “weak” test for beingrelated tocalf thymus HMG14. No other account, we have used the indexof Marchalonis and Weltman relationships are evident using the criteria of Cornish-Bowden (1971) and the critical values of Cornish-Bowden (1980) for (1980). testing the significance of amino acid composition indexes. Wheat HMGa has some characteristics reminiscent of H 1 The index, SAQ is definedas: SAQ = lo4 ( X A - X,#, where histones. It hashigh proline and alanine content.However, it XiAis the mole fraction of amino acid residues of the ith type is much lower in lysine and higher in arginine than wheat H1 in protein A and X,- is the mole fraction of residues of the histones (Spiker et al., 1976). No sequences are available for ith type in protein B. Cornish-Bowden (1980) has developed wheat H 1 histones. However, we have used the partial setables of “critical values” for predicting “relatedness” of pro- quence information and amino acid compositions of aminoteins based on amino acid composition comparison by SAQ. and carboxy-terminal regions of maize H1 histone (Hurley The values are dependent upon the total number of amino and Stout, 1980) to determine if the amino acid composition acid residues in the larger of the two proteins (N). For each of HMGa could be consistent with that of a simple degradaN , critical values for a “strong” test and for a “weakn test tion product of H1 histone. Using this approach, we could have been calculated. According to Cornish-Bowden, the cal- find noevidence that HMGa isderived from H1. Comparative culated value of SAQ nearly always will be greater than the peptide maps of wheat HMGa and total H1 histones (data critical value forthe “strong” test if the proteins are unrelated.not shown) also provide no evidence that HMGa is derived Calculated values of SAQ will almost always be smaller than from H1. When the amino acid compositions of HMGa and the criticalvalue forthe “weak” test if the proteins are related. total H1 histone are compared by the method of CornishWe have calculated SAQ for all pairs of wheat and calf Bowden (1980), the two proteins do not meet the “weak” test HMG proteins and compared the values to the criticalvalues for being related. By the sameprocedures, none of the wheat of Cornish-Bowden (1980) in an attempt to determine if any HMG proteins have aminoacid compositions which are “reof the wheat HMG proteins are more closely related to the lated” to anyof the wheat H2 histones. larger calf HMG proteins (1 and 2) or to the smaller calf Two-dimemiowl Electrophoresis-Commercially available HMG proteins (14 and17). The results of these calculations ampholines have not proved usefulin characterizing the wheat are summarized in Fig. 4. In this figure if two proteins meet HMG proteins by isoelectric focusing. Thus, we have used the “weak” test requirement for being related, they areconnonequilibrium pH gradient electrophoresis(O’Farrell et al., nected by a dotted line. If they meet the “strong” test for being 1977). In Fig. 5, a two-dimensional gel is shown in which a related,theyareconnected bya solid line. It shouldbe emphasized that we make no claims that proteins arenecessarily related by sequence if they meet the“weak” or “strong” test for relatedness. Nor do we claim that proteins which do not meet the “weak” test are necessarily unrelated. For example, calf HMG14 and HMG17 have been entirely sequenced (Walker etal., 1977,1979), and even though they do not meet HMG a __,,(.._... /’’_ ...’
I
HMG &____._....._........ HMG d
HMG 14
HMG I
::
HMG 17
-HMG 2
] ]
H1 WHEAT
CALF
FIG. 4. Similarities ofamino acid composition of wheat and calf thymus HMG proteins. The index of Marchalonis and Weltman (1971) was calculated for all pairs of calf and wheat HMG proteins. If a pair meets the“weak” test requirementfor being related (Cornish-Bowden, 1980), the proteins are connectedby a dotted line. If a pair also meets the “strong” test requirement for being related, the proteins are connected witha solid line.
a
FIG.5. Two-dimensional electrophoresis of wheat HMG proteins. A preparation of wheat HMG proteins was separated by two-dimensional electrophoresis essentially according to O’Farrell et al. (1977) using nonequilibrium pH gradient electrophoresis in the first dimension and NaDodSO, electrophoresis in thesecond dimension. An aliquot of the same HMG preparation which had not been separated in the first dimensionwas separated in the second dimension for comparison. Wheat HMGa (labeled a) co-migrates with the wheat H1 contaminant in the firstdimension. Wheat HMG proteins b, c, and d (unlabeled in figure) are each resolved into two species in the first dimension.
Wheat HMG Proteins
a
b
b d
a c cd b d
12011
c d
1
2
-.
.IC*-.) I__._
FIG. 6. Comparative one-dimensional peptide maps of wheat and pig thymus HMG proteins. Peptide maps of wheat and pig thymus HMG proteins were made using Staphylococcal V8 protease by the method of Cleveland et al. (1977) modified as described in the text. The l a n e s in the five panels in this figure are labeled with the proteins mapped ( a = HMGa; b = HMGb; c = HMGc; d = HMGd; I = HMG1; 2 = HMGZ). The peptides are stained with Coomassie Blue. The two, common, high-molecular-weight bands in each lane are due to theproteolytic enzyme.
Salt Extraction of Wheat HMG Proteins-We have used extraction of purified wheat chromatin with 0.35 M NaCl in order to establish that the proteinswe have studied meet the original operational criteria for HMG proteins. However, we notice that when such procedures are used, some of the HMG proteinsarenotextracted from chromatin. In Fig.7, the percentages of HMG proteins extractedby various concentrations of NaCl are shown. Well over half of the total HMG proteins are extracted by 0.35 M NaCl. However, about 10% of the HMG proteins remained bound to chromatin at 0.4 M .2 .3 4 .5 NaCl and all detectableHMGproteins were not removed until 0.45 M NaCl. NaCl (MI Association with Putatiuely Actiue Chromatin-Some of the FIG. 7. Salt extraction of wheat HMG protein. The percent- first, but certainly not conclusive, indications that vertebrate ageof total chromatin-bound HMG proteins extracted from chromatin by various concentrations of NaCl was determined as described HMG proteins are structural proteins of active chromatin were the observations that these proteins are released from in the text. DNase I-treated chromatin (Vidali et al., 1977; Levy-Wilson et al., 1977). We have previously shown (Spiker et al. 1983) wheat HMG preparation is separated by nonequilibrium pH gradient electrophoresisin the firstdimension and SDSelec- that DNase I treatment of nuclei releases the wheat proteins trophoresis in the second dimension. HMGa and contaminat- which we have called HMG proteins. Micrococcal nuclease ing histone H1 are labeled in the figure. Calf thymus HMG also more readily attacks chromatin containing transcribed proteins have secondary modifications and thus are separatedsequences than it attacks bulk DNA (Bloom and Anderson, into multiple forms by isoelectric focusing (Nicolas and Good- 1978). Jackson etal. (1979) have developed a procedure which uses micrococcal nuclease to obtainachromatinfraction win, 1982). Similarly, wheat HMGbhas two forms-one which migrates slowly in the first dimension and one which which contains HMG proteins inamounts stoichiometric with migrates nearly as rapidly as HMGa. HMGc and d also have the histone fractions. When the procedure of Jackson et al. (1979) is used on wheat chromatin, fractions similarly highly two forms each when separated by this procedure. Peptide Mapping-We have used the procedures of Cleve- enriched in the proteinswe call HMG proteins are obtained. land et al. (1977) to compare the peptide maps of each of the Fig. 8 shows SDS gels of proteins of the wheat chromatin wheat HMG proteins with the others, with each of the pig fractions prepared by the procedures of Jackson etal. (1979). thymus HMG proteins, and with wheat H1 histones. Some of In lane 1 of this figure are the proteins of the micrococcal the maps are shown in Fig. 6. When the maps of the wheat nuclease-resistant fraction. The only prominent proteins in HMG proteins are compared, a few bands of common mobility this fraction are the histones. In lane 2 of this figure are the can be demonstrated. However, there is no indication that proteins of the micrococcal nuclease-susceptible fraction. This any of the lower-molecular-weight HMG proteins arederived fraction contains histones but italso contains very prominent from any of the higher-molecular-weight ones or that exten- proteins which co-migrate with the wheat proteins we have sive sequence homology exists in the wheat HMG proteins. called HMG proteins. This fraction also contains a number Similarly, in comparison of wheat HMG maps with those of of higher-molecular-weight proteins which may be analogous the pig thymus HMG proteins, no evidence of extensive to the proteins studied by Prior et al. (1983) or by Emerson and Felsenfeld (1984). homology can be found.
Wheat HMG Proteins
12012
1
2
3
+a
4-b 4 - C
3+
4-d
FIG. 8. Wheat HMG proteins co-isolate with micrococcal nuclease-sensitive chromatin.Wheat embryo chromatin was separated into micrococcal nuclease-sensitive andmicrococcal nucleaseresistant fractions by the method of Jackson et al. (1979). The proteins of each fraction werethen separated by SDS-polyacrylamide gel electrophoresis. Lane 1 is the proteins from the micrococcal nuclease-resistant fraction.The only prominent proteinsin this fraction are the histones. Histones H3 and H4 (labeled 3 and 4 ) appear as single bands. Histones HI (labeled I ) and H2A and H2B (labeled 2) appear as multiple bands. Lane 2 is the proteins from the micrococcal nuclease-sensitive fraction. The histones are also present in this fraction but they are accompaniedby prominent proteinswhich co-migrate with the HMG proteins in lane 3. Bands corresponding in mobility to HMGa, b, and d can be seen in this gel. Two-dimensional gels (data not shown) indicate that HMGc is also in the nucleasesensitive fraction. In addition to the histones and HMG proteins, two prominent groups of high-molecular-weight proteins appear (arrowed) with the micrococcal nuclease-sensitive chromatin. Summary-Four proteins are extracted from purifiedwheat embryo chromatin with 0.35 M NaCl and are soluble in 2% (w/v) trichloroacetic acid. Thus,theseproteinsmeetthe original operational criteria for being “HMG” proteins. Because the true biological roles of mammalian HMG proteins have not been established, we cannot as yet determineif the plant HMG proteins are the functional equivalentsof mammalian HMG proteins. The plant HMG proteins do have several characteristics, however, which are consistent with the idea that they do have the samerole in chromatin structure as mammalian HMG proteins. They are rich in basic amino acid residues-approximately 20 mol %. Three arerich in acidic amino acid residues-approximately 23-29 mol %. The fourth wheat HMG has only about 12 mol % asparate asparagine and glutamate glutamine residues. Thus, this protein is uncharacteristicall; low in acidic residues.. The ”
+
+
weight of this observation must beconsidered in light of similar acidic amino acid residue content of trout testis protein H6which, on the basisof amino acid sequence, is almost certainly an HMG protein (Dixon, 1982). The wheat HMG which is uncharacteristicallylow in acidic amino acidresidues, HMGa, is the only wheat HMG which meets the prolinecontent criterion suggested by Johns (1982) for calf thymus HMGproteins.WheatHMGacontainsabout13 mol % proline residues. The other three wheat HMG proteins contain 4-6 mol %. We were unable to assign any of the plant HMG proteins to the larger HMG class (corresponding to calf thymus HMGl and 2) or to the smaller HMG class (corresponding to calf thymus HMG14 and 17) either by simple inspection of the amino acid contents or by using the procedures of Cornish-Bowden (1980). Wheat HMGc and HMGd do, however, meet the“weak” requirement of Cornish-Bowden forbeing “related” to calf thymus HMG14. Comparative peptide mapping also failed to establish extensive homology between any of the plant and vertebrate HMG proteins. The results of the mapping studies did, however, indicate that the wheat HMG proteins were not degradation products of highermolecular-weight HMG proteins orof the H1histones. The same HMG proteins which are extracted from wheat embryo purified chromatincan be extracted from various differentiated tissues of wheat with 0.5 N HClO, and these proteins are associated with mononucleosomes separated on sucrosegradients(datanotshown).Theobservationthat wheat HMG proteins are associated with DNase I- and micrococcal nuclease-sensitive chromatin is consistent with the idea that they are the functional equivalents of mammalian HMG proteins. Complete sequence data on the plant HMG proteins would be helpful in establishing such a correspondence. However, it is obvious that correspondence in function cannot be reliably established until the truein vivo functions of mammalian HMG proteins and plant HMG proteins are established. Acknowledgments-Thanks are due to Marcia Bates and Laura Arwood for technical assistance. REFERENCES Bassuk, J. A., and Mayfield, J. E. (1982) Biochemistry 21,1024-1027 Bloom, K. S., and Anderson, J. N. (1978) Cell 15,141-150 Cleveland, D.W., Fischer, S. G., Kirschner, M. W., and Laemmli, U. K. (1977) J. Bwl. C h m . 252,1102-1106 Cornish-Bowden, A. (1980) Anal. Biochem. 105,233-238 Dixon, G. H. (1982) in The HMG Chromosomal Proteins (Johns, E. W., ed) pp. 149-192, Academic Press, New York Emerson, B. M., and Felsenfeld, G. (1984) Proc. Natl. Acad. Sci. U. S. A. 81,95-99
Gazit, B., Panet, A., and Cedar, H. (1980) Proc. Natl. Acad. Sci. U. S. A. 77,1787-1790 Goodwin,G. H., Sanders, C., and Johns, E. W.(1973) Eur. J. Biochem. 38,14-19
Goodwin, G. H., Walker, J. M., and Johns, E. W. (1978) in The Cell Nucleus (Busch, H., ed) Vol. VI, pp. 182-219, Academic Press, New York Hamana,.K., and Iwai, K. (1979) J. Biochem. (Tokyo) 86,789-794 Hurley, C. K., and Stout, J. T. (1980) Biochemistry 19,410-416 Igo-Kemenes,T.,Horz,W., and Zachau,H. G . (1982) Annu. Rev. Biochem. 51,89-121
Jackson, J. B., Pollock, J. M., and Rill,R. L. (1979) Biochemistry 18, 3739-3748
Johns, E. W. (1964) Bwchem. J. 92.55-59 Johns, E. W. (1982) in The HMG Chromosomal Proteins (Johns, E. W., ed) pp. 1-7, Academic Press, New York Katula, K. S. (1983) Dev. Bwl. 98, 15-27 Klapper,D. G. (1982) in Methods inProteinSequenceAnalysis (Elzinga, M., ed) pp. 509-515, Humana Press, Clifton, NJ Levy-Wilson,B., Wong, N. C. W., and Dixon,G. H. (1977) Proc. Natl.Acad.Sci. U. S-A. 74, 2810-2814
Wheat HMG Proteins Luna, E. J., Kidd, G . H., and Branton, D. (1979) J . Biol. Chem. 264, 2526-2532 Marchalonis, J. J., and Weltman, J. K. (1971) Comp. Biochem. Physiol. 3 8 B , 609-625 Mardian, J. K. W., Paton, A. E., Bunick, G . J., and Olins, D. E. (1980) Science (Wash. D.C.) 2 0 9 , 1534-1536 Marquez, G., Moran, F., Franco, L., and Montero, F. (1982) Eur. J. Biochem. 123,165-170 Matsudaira, P. T., and Burgess, D. R. (1978) Anal. Biochem. 87,386396 Mayes, E. L. V. (1982) in The HMG Chromosomal Proteins (Johns, E.W., ed) pp. 9-40, Academic Press, New York Nicolas, R. H., and Goodwin, G. H. (1982) in The HMG Chromosomal Proteins (Johns, E. W., ed) pp. 41-68, Academic Press, New York O’Farrell, P. Z., Goodman, H. M., and O’Farrell, P. H. (1977) Cell 12,1133-1142 Panyim, S., and Chalkley, R. (1971) J. Biol. Chem. 246, 7557-7560 Prior, C. P., Cantor, C. R., Johnson, E. M., Littau, V. C., and Allfrey, V. G . (1983) Cell 3 4 , 1033-1042 Reeves, R., and Chang, D. (1983) J. Biol. Chem. 268,679-687 Sandeen, G., Wood, W. I., and Felsenfeld, G . (1980) Nucleic Acids Res. 8,3757-3778 Sanders, C. (1977) Biochem. Biophys. Res. Commun. 7 8 , 1034-1042 Seale, R. L., Annunziato, A. T., and Smith,R. D. (1983) Biochemistry 22,5008-5015 Simon, J., and Becker, W. M. (1976) Biochim. Biophys. Acta 4 5 4 , 154-171 Spiker, S. (1980) J. Chromatogr. 1 9 8 , 169-171
12013
Spiker, S. (1982) J. Biol. Chem. 267, 14250-14255 Spiker, S., and Isenberg, I. (1983) Methods Enzymol. 9 1 , 214-226 Spiker, S., Key, J. L., and Wakim, B. (1976) Arch. Biochem. Biophys. 176,510-518 Spiker, S., Mardian, J. K.W., and Isenberg, I. (1978) Biochem. Biophys. Res. Commun. 8 2 , 129-135 Spiker, S., Murray, M. G., and Thompson, W. F. (1983) Proc. Natl. Acad. Sci. U.S. A. 80,815-819 Stein, A., and Townsend, T. (1983) Nucleic Acids Res. 11,6803-6819 Swerdlow, P. S., and Varshavsky, A. (1983) Nucleic Acids Res. 1 1 , 387-401 Triplett, B. A. (1979) Dissertation, Oregon State University, Corvallis, OR Vidali, G., Boffa, L. C., and Allfrey, V. G . (1977) Cell 12, 409-415 Walker, J. M. (1982) in The HMG Chromosomal Proteins (Johns, E. W., ed) pp. 69-87, Academic Press, New York Walker, J. M., Hastings, J. R. B., and Johns, E. W. (1977) Eur. J. Biochem. 7 6 , 461-468 Walker, J. M., Goodwin, G . H., and Johns, E. W. (1979) FEBS Lett. 100,394-398 Weber, S., and Isenberg, I. (1980) Biochemistry 19,2236-2240 Weisbrod, S. (1982a) Nature (Lord)2 9 7 , 289-295 Weisbrod, S. (1982b) Nucleic Acids Res. 10, 2017-2042 Weisbrod, S., and Weintraub, H. (1979) Proc. Natl. Acad. Sci. U. S. A. 7 6 , 630-634 Weisbrod, S., and Weintraub, H. (1981) Cell 2 3 , 391-400 Weisbrod, S., Groudine, M., and Weintraub, H. (1980) Cell 1 9 , 289301