Mar 18, 1994 - the following sequence: a --). a2 --> CY~P -+ a&P' (for review, see ... ries Inc., and nonfat powdered milk was from Carnation Company.
Val. 269, No.38, Issue of September 23, pp. 23655-23660, 1994 Printed in U.S.A.
JOURNAL OF BIOLOGICAL CHEMISTBY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc. THE
Epitope Mapping and Functional Characterizationof Monoclonal Antibodies Specific for the a! Subunit of Escherichia coli RNA Polymerase* (Received for publication, March 18, 1994, and in revised form, June 27,1994)
Karim A. Sharie, Noboyuki FujitaJ,Ruzhong Jins, Kazuhiko IgarashiO, Akira Ishihamag,and Joseph S. Krakow$ll From the $Department of Biological Sciences,Hunter College of the City University of New York, New York, New York 10021 and the $Department of Molecular Genetics, National Institute of Genetics, Mishimu, Shizuoka 411, Japan
The epitopes have been localized for a set of mono- 1991; Igarashi etal., 1991). A number of a mutants with point clonalantibodiesspecificforthe a subunit of the mutations have been isolated that are defective in positive Escherichia coli RNA polymerase. The antibodies are control by transcriptionalactivators(SunshineandSauer, classified into three groups based on their epitopic as-1975; Dale et al., 1986; Garrett andSilhavy, 1987; Matsuyama signments. Group1, mAb 123C2, maps in theN terminus and Mizushima, 1987; Giffard and Booth, 1988; Slauch et al., of a between amino acids1 and 23; Group 2 antibodies 1991; Lombardoet al., 1991; Zou et al., 1992). Most of the rpoA (mAb 129C4, mAb 124D1 and mAb 121C5) map in the mutations affecting response to transcription activators (class I 190 and 210; Group factors) cluster in the C-terminal portion of a (contact site I) central region between amino acids 3 antibodies (mAb 130B1 and mAb 126C6) map in the C (Slauch et al., 1991; Lombardo et al., 1991; Zou et al., 1992). terminus between amino acids310 and 320. mAb 130C2 These studies also indicated that the protein-protein contact is anomalous since it maps to the N terminus between composed of sites on the CY subunit for transcription factors are amino acids 1 and 23 as well as to the C terminus beonly about 10 amino acid residues and, thus, are as short as tween amino acids 320 and 329. The antibodies were used to investigate the roleof a in transcription activa- antigenic epitopes recognized by antibodies (for review, see tion withCAMP receptor protein-dependent promoters. Ishihama, 1993). This work defines thelocation of epitopes for a set of monoThree antibodies (130C2,121C6, and 125C6) inhibited clonal antibodies directed against the a subunit of the E. coEi C A M P receptor protein-dependent initiation with lac P+ but not withlac UV5 or gal I?+. Inhibition was observed RNA polymerase. The epitopes were mapped in three different with f'ree RNApolymerase and the closed promoter com-regions of the CY subunit. Three of the antibodies stronglyint P+ plex; the preformed open promoter complex was insen-hibited transcription directedby c ~ P / C R P ' - d e p e n d e n Zac sitive. Only Z a c P' was sensitive to these anti-cu antibod- but did not affect the lac W5 or thegal P' directed reaction. ies supporting the concept that the mode of interaction ~~ERIMEN PROCEDURES T~ of RNA polymerase withCAMPreceptor protein differs between lac P' and gal P', MateriuZs-Reagents were obtained as follows. CsCland USBioclean were from U. S.Biochemical Corp.; ethidium bromide, aprotinin, and protein A-Sepharose were from Sigma. Restriction endonucleases and in vitro rabbit reticulocyte lysate translation kit (Type11) was from The DNA-dependent RNA polymerase of Escherichia coli is a Boehringer Mannheim. DMEM and fetal bovine serum were from Life Technologies, Inc., and the in vitro transcription kit was from Stratmultisubunit enzyme composed of a catalytically competent core enzyme (a,PP') and one of several molecular species of the agene. Tran3'S-label and [3H]leucine were from ICN Biochemicals, and goat anti-mouseIgG was from Sigma and Cappel. Acrylamide was from CT subunit. The enzyme is engaged in a complex sequence of Serva Fine Chemicals Inc. 5-30 Lysate System and RNasin were from steps for RNA synthesis. The role of each of the RNA polymer- Promega; DNA amplification kit was from Perkin Elmer Corp. Goat ase subunits has not beencompletely characterized. anti-mouse IgG-phosphatase was from Kirkegaard& Perry LaboratoThe CY subunit of 329 amino acid residues is encoded by the ries Inc., and nonfat powdered milk was from Carnation Company. rpoA gene. The assemblyof core RNApolymerase takesplace in the Preparation of Proteins-RNApolymerasewaspreparedby the following sequence: a a2--> C Y ~-+ P a&P' (for review, see method indicated in Fujita et al. (1987). Anti-u monoclonal antibodies Ishihama et al. (1987)). Analysis o f mutations in the rpoA gene were prepared according to Rockwellet al. (1985). Polyclonal anti-u antibody was prepared in rabbits as indicated in Iwakuraet al. (1974). affecting RNA polymerase assembly (Ishihama et al., 1980; DNA Preparations-PlasmidspTAX185,pTAD235,pTAD176,and Kawakami and Ishihama,1980; Igarashi etal., 1990; Hayward pTAD150 (Igarashi et al., 1990) and pGEW185, pGEMAD296, and et aZ. 1991)indicated that the N-terminal region of CY plays an pGEMAD256 (Igarashi and Ishihama,1991) were purifiedby CsCl denimportant role in core enzyme assembly. Zn uitro studies of sity gradient centrifugation as described by Daviset al. (1986). TheZac deletion mutants indicated thatthe C-terminal domainof CY is P', lac W5, and gal F"used forthe transcription assays were prepared et al. (1991). involved in transcription activation where activators bind up- as described by Igarashi and Ishihama (1991) and Igarashi For PCR, pTAX185 was digested with XbaI to excise the rpoA gene and stream of the promoter -35 region (Igarashi and Ishihama, resolvedbyelectrophoresison a 0.8%agarosegel.The1.2-kilobase fragment bearingthe rpoA gene wascut outand purified usingUSBioextraction and ethanol precipita* This work was supported in part by a Research Grant GM22619 clean. Following pheno~chlorofo~ from the National Institutes of Health (toJ. S. K.) and grants-in-aid tion, the DNA fragment was dissolved inTE buffer (5 mM Tris-KC1 (pH from the Ministry of Education, Science,and Culture of Japan (toA. I.) 8.01, 0.5 mM EDTA). The costs of publication of this article were defrayed in part by the payment of page charges. Thisarticle must therefore be hereby marked The abbreviations used are:CRP, CAMP receptor protein; DMEM, ~ a ~ u e F t ~ e in ~ eaccordance n~" with 18 U.S.C. Section 1734 solely to Dulbecco's Modified Eagle Medium; mAb, monoclonal antibody; RP, indicate this fact. closed promoter complex; RP,, open promoter complex;WT, wild type: 1R ' J whom correspondence should be addressed. PCR, polymerase chain reaction. --).
23655
A b s as Probes of the RNA Polymerase cy Subunit
23656 -
Primer name
CY-T~RBS 100-107 a-185-190R a-205-210R a-T7RBS 210-217 a-305-310R a-315-320R a-324-329R a
TABLEI Primers used forPCR reactions Sequence" GAA TTC TAATACGAC TCA GGA GATATACATATG CTT TTA TAACGCTGC TTC AAC TTATTATGT GCCCGTGGT GAA TTC TAA TAC GAC TCA GGA GATATACATATGACAATCGAT TTATTA ACG GGA AGC CAG TTATTAGTT TTC CAGGCG TTA TTA CTC GTC A G C ~GAT
CTA ACC ATT T T CTA
TAGGGTTAA CTT TAA TTG AAT AAA TCT GGC GTA C m TAG GGT TAA CTT TAA CCT GAA GAG GCG CAC GTC CAT GCC GCT TGC
GAA A m GAA
ATT
Underlining indicates rpoA sequence.
In Vitro Synthesis of Duncated a Fragments-In vitro transcription 37 "C. Cells from 0.5 ml of culture medium werespun down in a microand capping was carried out according to the procedure given in the centrifuge, washed with 50 mM Tris-HCl (pH 8.01, and suspended in Stratagene instruction manual (Cat no. 200350, 1939) using T7 RNA Laemmli buffer. After boiling for 5 min, the supernatant was loaded polymerase and 2 pg of circular plasmid DNA(pGEMAX185,pGEonto a SDS, 10% polyacrylamide gel. Following electrophoresis, the "3296, p G E ~ 2 5 6 ,or NheI-linearized pTAD235, pTAR176, proteins were transferred onto a nitroce~~ulose membrane (0.2 p d using pTAD150) in the presence of 7m-GpppG (Nielson and Shapiro, 1936). a ~ n i B l o (Millipore). ~ S ~ ~ The nitrocellulose membrane was treated This produced mRNAs coding fora-WT, a-296,0-256, a-235, a-176 and with 5 mg/ml heparin and then blocked in 5%nonfat dry milk in Trisa-150, respectively. Digestion of pTAx185 with StyI, NciI, or Cfrl3I buffered saline (Luo and Krakow, 1992). The membrane was incubated generated templates for the preparation of mRNAs coding for a-23, with one of the anti-a monoclonal antibodies overnight at 4 "6. After a-51, and a-107, respectively. mRNA corresponding to a210-310 was washing three times with washing buffer (0.5% nonfat dry milk in produced by transcribing the PCR-generated DNA template with T7 Tris-butTered saline), the nitrocellulosestrips were incubated with goat RNA polymerase. The mRNAs were translated using a rabbit reticulo- anti-mouse I&-phosphatase for 2 h at room temperature. The strips cyte lysate system (250 pl of reaction mixture) supplemen~dwith 15pl were washed three times with washing buffer and developed in nitro of radioactive amino acid mixture (0.2 mCi of Tran3*S?abelc o n t a i ~ n ~blue tetra~o~ium, 5-bromo-4-chloro-3-indolylphosphate solution to vis[~Slmethionine+ [35Slcysteine)under the conditions specified in the ualize the hands. protocol for the reticulocyte translation kit. Since the amino acid mixIn Vitro ~ a n s c r ~ ~ ~ i o n - Treaction he mix (final volume 50 111) conture contains cysteine, the protein products are primarily labeled with tained 50 mM Tris-HC1 (pH 7.81, 3.0 mM magnesium acetate, 0.1 mM [35Slmethionine. After a 120-min incubation a t 30 "C, aliquots were EDTA, 0.1 mM dithiothreitol, 50 mM NaCl, 12.5 pg bovine serum albudirectly subjected to SDS-polyacrylamide gelelectrophoresis (Laemmli, min, 0.1 pmol each of lac P,lac UV5, and gal P', 3 pmolofRNA 1970) and autoradio~aphyto verify synthesis of the respective a polymerase, 2.5 pmolof CRP, and 100 pv CAMP.ARer incubating for 30 min at 37 "C, 15 pl of a mix c o n t a i ~ n g0.5 mM each of ATP, GTP, GTP, polypeptides. PCR ~e~hods-primersfor PCR were synthesized using an Applied and 0.16 mM UTP + 2 pCi EaS"PIUTPand 10 pg of heparin was added, Biosystems Model 380BDNAS~thesizer(Table I).The upstream prim- Where indicated, 12 pmol of one of the monoclonal antibodies was n carried out for 5 min at 37 "C ers contain a T7promoter and a ribosomal binding site attachedas a 5' added. Single-round ~ a n s c ~ p t i owas extension. The presence of these sequences are designated as T7RBS". and stopped hy the addition of 60 pl of a mix containing 40 m EDTA, The numbers represent the amino acid residues in the a protein. The 13 pg of E. coli tRNA and 0.5 M sodium acetate (pH 5.2). Following primers designated "R" represent reverse complements of 3'-terminal ethanol precipitation, the transcripts were resolved using 8 M ureafragments of the a gene supplemented with tandem stop codons. PCR polyacrylamide geleiectrophoresis followed by autoradiography. was carried out essentially using the protocol of Lesley et ul. (1991) using the XbuI f r a ~ e n containing t the rpoA gene as template. RESULTS 5-30 ~anscription-7f.anslattottReactions-The DNA templates gen~ p ~ t Q ~ pu~ p by~ ImmunoprecipitatioR-Each ~ ~ g of the erated by PCR weretranslated in anE. coli 5-30 fraction (Fromega)by a modification of the procedure of Lesley et ai. (1991) in the presence seven indicated anti-a monoclonal antibodies were tested for of 5 pl (0.07 mCi) of Tran3'S-labe1 in a final volume of 50 pl. Products their ability to bind to the 35S-labeledC-terminal truncated CY were resolved hy SDS-polyacrylamide gelelectrophoresis. Autoradiog- fragments. AI1of the a fragments were precipitated by the raphy was performed to veri@ the synthesis of the desired truncated a anti-cy polyclonal antibody (Fig. 1).Tvvo of the monoclonal anfragments. tibodies, mAb 130Bl and mAb 125C6, bound onlyto intact a. Immunoprecipitution--S5S-or 3H-labeled a p o l ~ e p t ~ din e s443 111of Three antibodies, mAb 121C5, mAb 124D1, and mAb 129C4, the reticulocyte lysate or 5-30 mix were added to 0.5 mlofDMEM containing 5% fetal bovine semm and 0.02% SDS and incubated over- cross-reacted with CY-235and longer fragments. OnlymAb night at 5 "C with 2 pg of one of the monoclonal antibodies. After 123C2 (data not shown) and mAb 130C2 bound to all of the addition of 5 pg of goat anti-mouse IgG, the mixture was incubated for fragments including a-150, m e results show that theepitopes 2 h at 5 "C. Immune compfexes formed were recovered after the addi- reside in threedistinct regions of the a subunit: the epitopes for tion of 50 $ of 1 mg/ml protein A-Sepharose in phosphate-buffered mAb 125C6 and mAb 130Bl in the extreme C terminus besaline (10 mM potassium phosphate (pH 7.2) and 150 mu NaCI) followed tween amino acids 296-329 (Group 3 antibodies);those for mAb by incubation for 10 min a t 5 "C and centrifugation for 2 min in a microcent~fuge.The pellet was washed five times with RIPA b-er (20 121C5, mAb 124D1, and mAb 129C4 in thecentral region bew Tris-HCI (pH 8.0), 0.15 hi NaCl, 0.5% sodium deoxy~olate,1% tween amino acids 176-235 (Group 2 antibodies); and those for Triton X-100,5 mM MgCl,) and then resuspended in Laemmli buffer (0.2 mAb 130C2 and mAb 123C2 in theN-terminal region between M Tris-HCl, pH8.00,1%SDS, 1%2-mercaptoethanol, 10% glycerol, 0.1% amino acids 1-150 (Group 1 antibodies). bromphenol blue). ARer SDS-polyacrylamide gel electrophoresis To further localize the epitopes for the Group 1 antibodies (Laemmli, 1970),the gel wasstained with Coomassie Bluein 10% acetic (mAb 123C2 and mAb 130C2), shorter N-terminal a fragments acid, 50% methanol. After soaking the gel in 1M sodium salicylate for 60 were used for immunoprecipitation. As expected mAb 129C4 min (Din et al., 19901, the gel was dried for a u t o r a d i o ~ p h y . For analysis of a-51 and a-23, the amounts of labeled peptide and did not bind to anyofthe N-terminal a fragments used (Fig. 2) monoclonal antibody were increased 3-fold, 15 pg of aprotinin and 0.1 since its epitope is not located between amino acids 1 and 150. mM phenylmethylsulfonylfluoride were added, and the time of incuba- Both mAb 130C2 and mAb 123C2 wereable to cross-react with tion for formation of the immune complex was increased to 36 h at 5 "C. an a fragment as short as23 amino acid residues. The results After electrophoresis, the gel was microwaved for 40 s in Coomassie indicate that theepitopes for these two antibodies are located stain to enhance the fixation of these small fragments. Western Blotting of C-lkrmiml lEuncated a-E. coli transformed in the extreme N-terminal region between amino acids 1-23. The results using truncated a indicate that theepitopes for with p G E ~ 2 9 was 6 grown to an absorbance of 0.3 a t 600 nm and induced with 0.4 mM isopropyl-l-thio-~-D-galactop~anoside for 2 h a t the Group 2 antibodies, mAb 124D1, mAb 129C4, and mAb
mAbs as Probes of the RNA Polymerase
CY
Subunit
mAb 124D1 mAb
n-
wt
Q-296 Q-256-
23657 12964
mAb 121
C5
-
a- 235 a - 1 7 6a - 150 -
Flo. 1. Immunoprecipitation of the C-terminal truncated a fragments with anti-cy monoclonal antibodies. 4 pl each of .‘‘Sslabeled a-150, a-176, a-235, and a-256; 8 p1 of a-296; and 6 pl of a-WT were mixed in 0.5 ml of DMEM containing 5% fetal bovine serum and 0.02% SDS and then immunoprecipitated with the indicated monoclonal antibodies as described under “Experimental Procedures.”
B - a-wt
FIG.3. Immunoprecipitation of the internally truncated cy fragments with anti-a monoclonal antibodies. The a fragment spanning amino acid 100-329, was produced by coupled transcriptiontranslation of PCR-generated templates. Equal amounts of trichloroacetic acid-precipitable radioactivity of a100-190,a100-210, and a210329 in 0.3 ml of DMEM containing 5% fetal bovine serum were treated with the indicated monoclonal antibodies as described under “Experimental Procedures.”
no Ab . . is,
-a-150 -a
- IO7
mAb 13081 0
0‘)
2
2
0
0
7
N U
2
mAb 125C6
-+
z3 z
+
N U
0
2
0
6Ui
Y
U
FIG.2. Immunoprecipitation of the C-terminal truncated cy fragments with the indicated anti-cy monoclonal antibodies. Immunoprecipitation was carried out as described under “Experimental Procedures.”
121C5, are located between amino acids 176-235, and thosefor the Group 3 antibodies, mAb 130B1, and mAb 125C6, are present within the C-terminal region between amino acids 296329. To ensure that the removal of these segments wasdirectly responsible for the loss of immunoreactivity and to further define the positions of the epitopes, a set of internal a fragments was examined.As shown in Fig. 3, Group 2 antibodies, mAb 124D1, mAb 129C4, and mAb 121C5, bound to a100-210; none of these antibodiescross-reacted with a100-190 or a210329. On the other hand, as shown in Fig. 4, Group 3 antibodies, mAb 130B1 and mAb 12326, bound to both a210-329 and a210-320, while neither antibody bound to a210-310. The results indicate that the epitopes for Group 2 antibodies are located between amino acids190 and 210. The epitopes forboth of the Group 3 antibodies reside within the 10-amino acid Cterminal segment spanning amino acids310-320. Epitope Mapping by Western Blotting-To confirm the epitope map derived by immunoprecipitation, we carried out Western blot analysis. Fig. 5 shows the results usinga-WT and a-296. As expected, the Group1 and 2 antibodies bound to both full-length and truncated a-296, while Group 3 bound only to full-length a. The more rapidly migratingprotein is probably a
FIG.4. Immunoprecipitationof the internally truncated a fragments with anti-a monoclonal antibodies. The Q fragment spanning amino acid 210-329, was produced by coupled transcriptiontranslation of PCR-generated templates. Equal amounts of trichloroacetic acid-precipitable radioactivity of a210-310, ~ 2 1 0 - 3 2 0 ,and a210329 in 0.3 ml of DMEM containing 5% fetal bovine serum were treated with the indicated monoclonal antibodies as described under “Experimental Procedures.”
degraded form of the a-296. The results presented in Fig. 2 derived from immunoprecipitation indicated that mAb 130C2 could bind to N-terminal fragmentsas short as 23 amino acids. It was surprisingt o find that thisantibody was unable tobind to a-296 (Fig. 51, although this truncated protein retains the using N-terminal sequence of a. Western blotting is carried out SDS-denatured proteins and, thus, the native conformation of the N-terminal epitopic domain essential for binding ofmAb 130C2 may be disrupted. It is known that some monoclonal antibodiesdiscriminate between thenativeanddenatured forms of an epitope. Apparently mAb 130C2 cannot bind to the N-terminal epitope on a-296 after denaturation but is still able to bindto the full-sized a. The results indicated that mAb 130C2 recognizes another epitope located in the C-terminal
23658
mAbs as Probes of the RNA Polymerase a Subunit
-a-wt , - a-wt 1-
U-296
I
- U-210-329
- a-210-320 021 0-320 a210-329 FIG.5. Western blotting analysis of intact a and a-296.Western blotting was done a s described under “Experimental Procedures.”
FIG.6. Immunoprecipitation of the internally truncated a fragments with anti-a monoclonal antibody.The a fragment spanning amino acids 210-329 was produced by coupled transcription-translation of PCR-generated templates. Equal amounts of trichloroacetic acid-precipitable radioactivity of a210-320 and a210-329 in 0.3 ml of DMEM + 5% fetal bovine serum were treated with the indicated monoclonal antibody a s described under “Experimental Procedures.”
region of a. To confirm this prediction, immunoprecipitation with mAb 130C2 was performed using the C-terminal a fragments, a210-320 and a210-329. mAb 130C2 boundonly to mutations within this contact site I region of the C-terminal a210-329 (Fig. 6). We concluded that mAb 130C2 recognizes a domain of a render mutant RNA polymerases insensitive to sequence between amino acids 320 and 329 as well a s a con- transcription activation by class I activators (Lombard0 et al., formational epitope involving amino acids 1-23 (Fig. 7). 1991; Slauch et al., 1991; Thomas and Glass, 1991; Zou et al., Functional Mapping-Single-round in vitro mixed transcrip- 1992). However, some mutationshave alsobeenidentified tion assays were carried out in a reaction mixture containing within the N-terminalregion. Results obtained using a library lac P+,lac UV5, and gal P’ DNA templates asdescribed under of C-terminal and N-terminal random mutations in a showed “Experimental Procedures.” Theantibodies were incubated that most of the Lac- mutants carried a mutation within a with RNA polymerase, RP, or RP, complexes prior to the addi- narrow region of CY between amino acids 265-270, but about tion of the substrates for RNA synthesis. Of the antibodies 10% of the Lac- mutants were found t o carry mutations in the examined, mAb 130C2, mAb 121C5, mAb 125C6, and mAb N-terminal region.” Most of the a mutations that made cells 130B1 had previously been shownt o strongly inhibit initiation defective in ompR regulation are located in the C-terminal from lac P’ and partially withlac U V 5 when RNA polymerase domain; however, two of the mutations are near Nthe terminus was preincubated with these antibodies (Dalla Venezia and (Slauch et al., 1991). Oneof the mutants,rpoA150, is altereda t Krakow, 1990). The results presented inFig. 8 show that none Pro-240 and Leu-28 (it is thought that significant the mutation of the five monoclonal antibodies tested inhibited transcription is Pro-240 and not Leu-28), while the other mutant,rpoA53, is initiation from the lacUV5 promoter. This suggested that these mutated in the extreme N terminus a t Gly-3. The apparent monoclonal antibodies do not interfere with the catalyticfunc- ompR defect was considered to be a consequence of the overextion of RNA polymerase. In contrast, mAb 130C2, mAb 121C5, pression of the mutant a subunit. However, our epitope mapand mAb 125C6 strongly and mAb 130B1 partially inhibited ping studies suggest that both the C-terminal and the N-tercAMP/CRP-dependent initiation from lac P1 when these anti- minal regions may be involved in molecular communication bodies were preincubated with eitherholoenzyme or preformed with regulatory proteins. RP,. After the open complexes were formed, these monoclonal The epitopes for two Group 3 monoclonal antibodies (mAb antibodies were without apparent effect on transcription ini- 125C6 and mAb 130B1), which inhibit CRP-dependent lac P1 tiation from lac P1. Transcription from the cAMP/CRP-depend- transcription, are located near the C terminus of a between ent gal P1 promoter was insensitive to all the monoclonal an- amino acids 310 and 320 (Fig. 7). The epitope for another intibodies tested, supporting the concept that CRP makes contacthibitory monoclonal antibody (Group 1mAb 130C2) maps near with site I on a for lac transcription (Igarashi and Ishihama, both the N-terminal region between amino acids 1 and 23 and 1991) but with siteI1 on u for gal transcription (Kumar et al., the C-terminal region between amino acids 320 and 329. The 1994). One of the Group 1 antibodies, mAb 123C2, which rec- abilityto inhibit CRP-dependent lac transcriptionis also ognizes the N-terminal epitope, did not inhibit anyof the pro- shown by the Group 2 mAb 124D1 (data not shown) and mAb moter-dependent reactions.’ 121C5, whose epitopes are mapped in the central region of a between amino acids 190 and 210. This is a region of a for which mutants defective in CRP-mediated regulation have not The involvement of the a subunit in transcription regulation as yet been demonstrated. Anti-a antibodies, which do not inis a topic of considerable interest.C-terminaltruncations, hibit transcription with lac P+,have epitopes that mapclose to which still allow for assembly of the altereda into RNA polym- the regions where theepitopes for inhibitory antibodies reside. erase, are no longer able to support CRP-dependent lac tran- Located in close proximity are the epitopes for the inhibitory scription (Igarashi and Ishihama, 1991). Likewise, most tran- mAb 124D1 and mAb 121C5 as well a s mAb 129C4, which was scription factors thatbind upstream of the promoter are unable shown to block reassembly of RNA polymerase (Riftina et al., to activate mutant RNA polymerases containing a C-terminal 1989).This suggests that the inhibitory effect of the monoclonal truncated a (Igarashi et al., 1991). Amino acid substitution antibodies may be specific rather than merely a steric effect of the monoclonal antibodies binding to a in RNA polymerase. DISCUSSION
The slowestmoving band seen in Fig. 8 is only present when the lac W 5 template is used and probably is the end to end transcript produced from this template.
C. Zou and A. Ishihama, unpublished results.
mAbs as
Probes of the RNA Polymerase cx Subunit
CRP OmpR Frr OmpR 240 265 270 290 311 322
OmpR 3 28
FIG.7. Epitope map for the anti-a monoclonal antibodies. Thenumhers correspond to the amino acid residues in the n subunit; *: indicates inhibitory antibodies; + indicates mAb that inhibits reconstitution. The upper bur shows a map of some mutant n proteins defective for different activator proteins.
23659
I
a 1
a
OmpU
2
1
3 I
190
210
I
1
1
I
I
121 C5' 124D1' 129C4+
130C2' 12362
Y
I
I
I
310 320 329 I
13081' 125C6'
I
13OC2'
mAb:RP, mAb:RP, RNP:mAb
Y?
lac P 2 -
1 -lac P I
L . -lac
uv5
FIG.8. Effects of the anti-a monoclonal antibodies on transcription from the CRP-dependentlac and gal promoters. Transcriptionreaction was carried out as indicated under "ExperimentalProcedures."Each of the monoclonal antibodies was incubated with RNA polymerase or RP, overnight a t 5 "C or with RP, for 30 min at 37 "C to form antigen-antibody complexes.
gal P 2*
-galPI
mAb 130C2, mAb 121C5, and mAb 125C6 strongly inhibit iniFive epitopes (for proteins other than a ) have been charactiation when incubated with free RNA polymerase or RNA po- terized by determining the crystal structure of an antigenlymerase in the closed complex. However, they fail to inhibit monoclonal antibody Fab complex. All of these epitopes occupy initiation once the open complex is formed. The resistance of large areascomposed of 15-22 amino acid residues and several the RP, complex could be a consequenceof the maskingof these surface loops, indicating that the antigenicity of the native on its conformation (for review, see epitopes by CRP. It is also possible that the antibody can no protein is strictly dependent Laver et al. (1990)). Epitopes of 4-7 aminoacids in length longer block an essentialconformational change inRNA polymmapped by using synthetic peptides may represent only parts erase once the RP, complex has formed. The a subunit plays an important role in the interaction of CRP with theholoenzyme of the epitopes exposed on denatured forms. In the present for promoters having a CRP binding site upstream of the -35 study, all of the truncated a proteins have been immunoprecipitated a t least by some monoclonal antibodies, indicating region. that some epitopes can assume the native conformation on Mutations affecting the response to CRP activation have these truncatedderivatives. Where a particular antibody failed been primarily mapped within a narrow region of a between to precipitatea truncated fragment dueto the presumed loss of amino acids 265 and 270 (Zou et dl., 1992). Cross-linking be- the epitope sequence, a PCR-amplified DNA fragment bearing tween CRP and the C-terminal region of a has been demonthe target sequence coded for a peptide fragment, which was strated by Yan et al. (1994). The inhibitory antibodies bind to bound by the monoclonal antibody. However, the possibility epitopes present in regions of a distal from the proposed CRP that theregions mapped represent only a major portionbut not contact region. Antibody binding may indirectly result in inhi- all of the epitope sequence cannot be ruled out. One of the bition by blocking an essentialconformational change ina con- monoclonal antibodies, mAb 130C2, binds to epitopes that are sequent to its interaction with CRP. It is also possible that located in both the N and C termini of the N subunit. Since antibody binding could alter the conformation of a such that there isno apparent sequence homology for these two regions of the proposed contact site (Zou et al., 1992) is inaccessible to a, it is possible that protein folding may form afunctional CRP. Benjamin et al. (1992) haveproposed that an antigen may domain. undergo long range conformational changesdistanttothe Acknowledgment-We thank Dr. Jianying Luo for helpful discusepitopic site upon binding t o the antibody. In that case, the sions. different monoclonal antibodies upon binding to different a domains may still be inhibitory, and, hence, the actual siteon REFERENCES the a subunit that interacts withCRP on the lac P1 type proBenjamin, D.C., Williams, D. C.. Jr., Smith-Gill. S. J.. and Rule, G. S. (1992) moter cannot be defined by the epitopes. None of the anti-a Riochernisfry 31, 9539-9545 monoclonal antibodies inhibit CRP-dependentgalP' transcrip- Dale, E. C., Christie, G. E., and Calendar, R. (1986) J. Mol. B i d . 192, 793-803 Dalla Venezia, N., and Krakow, J. S. (1990) J . Riol. Chem. 265, 8122-8126 tion; similarly no a mutants have been presently shown to Davis, L. G., Dibner, M. D., and Rattey. J. F. (1986) Basic "chods i n Molecular Biology, Elsevier Science Publishing Co., Inc., N e w York affect this reaction supporting thenotion that the target site of S., Brill, S. J., Fairman. M. P., and Stillman, R. (1990) Genes & Deu. 4, CRP for gal activation is located on the d o subunit (Kumaret Din,968-957 al., 1994). Fujita, N., Nomura, T., and Ishihama, A. (1987)J . B i d . C h e n ~262, . 18365-1859
23660
mAbs as Probes ofthe RNA Polymerase
Garrett, S., and Silhavy, T. M. (1987)J. Bacteriol. 169,1379-1385 Giffard, P. M., and Booth, I. R. (1988)Mol. & Gen. Genet. 214, 148-152 Hayward, R. S., Igarashi, K, and Ishihama, A. (1991)J. Mol. Biol. 221,23-29 fgarashi, K., and Ishihama,A. (1991)Cell 65, 1015-1022 Igarashi, K., Fujita, N., and Ishihama,A. (1990)Nucleic Acids Res. IS,5945-5948 Igarashi, K., Hanamura, A., Makino, K., Aiha, H., Mizuno, T., Nakata, A,, and Ishihama, A. (1991)Proc. Nutl. Acad. Sei. U. S.A. 88,8958-8962 Ishihama, A. (1993)J. Bacteriol. 176,2483-2489 Ishihama, A,, Shimamoto, N., Aiba, H., Kawakami, X-, Nashimoto, H., Tsugawa, A,, and Uchida, H. (1980)J. Mol. Bid. 137,137-150 Ishihama, A., Fujita, N., and Glass, R. E. (1987)Proteins Struct. Funct. Genet. 2, 42-53 Iwakura, Y., Ito, K., and Ishihama, A. (1974MoZ. & Gen. Genet. 153, 1-23 Kawakami, K., and Ishihama, A. (1980)Nucleic Acids Res. 19,3491-3495 Kumar, A., Grimes, B., Fujita, N., Makino, K., Malloch, R. A., Hayward, R. S., and Ishihama, A. (19941J. Mol. Biol. 235,405-413 Laemmli, U. K. (1970)Nature 227,680485
(Y
Subunit
Laver, W. G., Air, G. M., Webster, R.G., and Smith-Gill, S. J. 119901 Cell 61, 553-556 Lesley, S. A., Brow,M. A. D., and Burgess, R. R. (19911J. Biol. Chem. 266, 2632-2638 Luo, J., and Krakow, J. S . (1992)J. Biol. Chern. 287, 18175-18181 Lombardo, M.J., Bagga, D., and Miller, C . G. (1991)J. Bacteriol. 173, 7511-7518 Matsuyama, S., and Mizushima, S. (1987)J. Mol. Biol. 195,847-853 Nielson, D. A,, and Shapiro, D. J. 11986)Nucleic Acids Res. 14,5936 Riftina, F., DeFalco, E., and Krakow, 3. S. (1989)B i ~ 28,32993305 ~ ~ ~ Rackwell, P.,Beasley, E., and Krakow, J. S. (1985jB i ~ h e 24, ~ ~ 3240-3245 s ~ ~ Slauch, J. M., Russo, F. D., and Silhavy, T. J. (1991)J. Bacteriol. 173, 7501-7510 Sunshine, M. G., and Sauer, B. 11975)Proc. Natl. Acad. Sci. U. S. A. 72,2770-2774 Thomas, M. S., and Glass, R. E. (1991)Mol. ~icrobiol.5, 2719-2726 Yan, C., Ehright, Y. W., and Ebright, R. H. (1994)Science 265,90-92 Zou, C., Igarashi, K., Fujita, N., and Ishihama,A. (1992)Mol. ~ j c r o b ~6,2599o~. 2605
t
~
