Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (277K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (17)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Kuznetsova, S.
Right arrow Articles by Shabarova, Z.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kuznetsova, S.
Right arrow Articles by Shabarova, Z.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 1995 Oxford University Press 4783-4790

Footnote

Crosslinking of double-stranded oligonucleotides containing O -methyl-substituted pyrophosphate groups to the HNF1 transcription factor in nuclear cell extract

Crosslinking of double-stranded oligonucleotides containing O -methyl-substituted pyrophosphate groups to the HNF1 transcription factor in nuclear cell extract Svetlana A. Kuznetsova 1,* , Catherine Clusel , Edgardo Ugarte , Isabelle Elias , Marc Vasseur , Marta Blumenfeld and Zoe A. Shabarova1

GENSET, 1 rue R. et S.Delaunay, 75011 Paris , France and 1 Joint Laboratory GENSET-Laboratory of Nucleic Acid Chemistry, Moscow State University, Moscow 119899, Russia

Received July 30, 1996 ; Revised and Accepted October 17, 1996

ABSTRACT

Probing of the HNF1 (hepatocyte nuclear factor I) DNA-binding region using a set of DNA duplexes containing pyrophosphate or O -methyl-substituted pyrophosphate internucleotide groups at different positions of the HNF1 recognition sequence was performed. The histidine-tagged HNF1/1-281 DNA binding domain and nuclear extract from rat liver were used. We showed that HNF1 from these species specifically binds to modified DNA duplexes. A correlation in binding affinity of both types of duplexes was detected. Crosslinking of the HNF1 DNA-binding domain and HNF1 in nuclear liver extract to DNA duplexes carrying O -methyl-substituted pyrophosphate groups was observed. The crosslinking efficiency of HNF1 in liver extract to substituted pyrophosphate-modified DNA duplex, containing a reactive internucleotide group between nucleotides G and T of the GT dinucleotide immediately 5 ' to the TAAT recognition sequence, amounts to 40% of the efficiency of non- covalent association. Nonspecific crosslinking of the reactive DNA duplexes to other components of nuclear extract was not observed. These results indicate that DNA duplexes carrying substituted pyrophosphate internucleotide groups can specifically bind and crosslink with DNA-binding proteins, especially transcription factors in crude preparations and could constitute a potential tool to control the expression of disease-causing genes.

INTRODUCTION

The development and differentiation of eukaryotic organisms are complex phenomena that require specific regulation of the expression of particular genes. This control operates predominantly at the transcriptional level and to a great extent depends on cellular sequence-specific transcription factors which interact with DNA sequences usually located in the 5'-flanking region of the gene ( 1 ). Transcription factors are attractive targets for regulating gene expression ( 2 , 3 ). Their inhibition by short double-stranded oligodeoxynucleotides containing recognition sequences can interfere with the gene expression.

We have previously demonstrated that double-stranded oligonucleotides carrying recognition sites for DNA transcription factors can specifically inhibit gene expression at nM concentrations ( 4 ). To improve the efficiency of oligonucleotide competitors, we suggest using a crosslinking procedure. Specific crosslinking of reactive double-stranded oligonucleotides to targeted factors results in their irreversible inhibition and may be considered a potential therapeutic tool to regulate the expression of harmful genes involved in viral diseases or cancer.

Crosslinking procedures are extensively used to study nucleic acid-protein interactions. Photochemical crosslinking of proteins to nucleic acids containing 5-bromouridine ( 5 , 6 ), 4-thiouridine ( 7 - 9 ), 6-thiodeoxyguanosine ( 9 ) or azidonucleotides ( 10 - 12 ) have been utilized. This procedure has been successfully applied to study transcription factors. Photochemical crosslinking has been performed to identify four subunits of Saccharomyces cerevisae transcription factor IIIC and to explore the proximity of each of these subunits to different parts of a yeast tRNA ( 11 ). This procedure involved the use of DNA probes containing arylazide derivatives of deoxyuridine. Photochemical crosslinking has been developed to identify contacts between 5S DNA and Xenopus TF IIIA. 5-Azido-2'-deoxyuridine substituted DNA has been used in this investigation ( 13 ). However, photochemical crosslinking requires additional activators such as light or heat. Additionally, aside from the propensity of arylazides to yield reactive nitrenes upon exposure to activators, the reaction pathways utilized by the generated nitrene vary according to the composition and environment of the aryl group ( 10 ).

Lately, more efficient crosslinking procedures using platinum complexes as crosslinking reagents have been introduced ( 14 ). These reagents function within the physiological pH range and require no additional activators ( 15 ). Crosslinking of transcription factors CREB and JUN to their recognition sequences with PtII complexes has been performed ( 14 ). However, to obtain high crosslinking efficiency it is necessary to use phosphorothioate-containing oligonucleotides because crosslinking is much less efficient in the absence of phosphorothioate groups ( 14 ).

Recently, we have suggested a novel method for inserting active groups in double-stranded DNAs to permit crosslinking with proteins ( 16 , 17 ). Our studies revealed that DNAs carrying substituted pyrophosphate internucleotide groups react in near-physiological conditions with different nucleophiles, including nucleophilic amino acids, according to the following scheme ( 16 , 18 ):


Scheme 1. NuH indicates nucleophile.

Modified DNA duplexes containing these reactive groups in the recognition site of DNA-binding protein interact with amino acids which are close to the DNA interface. As a result a crosslink between protein and modified duplex is formed. This approach has been successfully employed for the affinity modification of some enzymes participating in DNA recognition, such as Eco RI and Rsr I restriction and modification enzymes ( 19 ), Eco RII and Sso II restriction endonucleases ( 20 , 21 ) and for affinity modification of the NF-kB p50 subunit ( 22 ).

Potential advantages of this approach are: (i) well-developed and simple synthesis of DNA duplexes containing substituted pyrophosphate groups; (ii) reactive groups can be introduced at any desired position in nucleic acid; (iii) modified DNA duplexes are sufficiently stable and can be kept without loss of activity for long periods; (iv) crosslinking is possible only to amino acids which are placed at `zero' distance from reactive groups thus allowing identification of amino acid residues that have been implicated in direct interaction with phosphate groups of DNA; (v) crosslinking occurs in near-physiological conditions within a pH range of 6.0-8.0 without any additional activators. Thus this approach can be an attractive tool for both research and therapeutical applications.

In the present work we have developed an experimental approach for crosslinking sequence-specific DNA binding transcription factors to double-stranded oligonucleotides carrying substituted pyrophosphate groups. As a model, we have used the rat liver transcription factor HNF1. HNF1 is a well characterized liver- enriched transcription factor required for the expression of many hepatic specific genes ( 23 , 24 ). It binds as a dimer to a 15 bp pseudopalindromic site g/aGTTAATNATTAACc/a present in the promoters or enhancers of these genes ( 23 , 25 ). In order to design a crosslinking reagent, i.e. to determine the position of reactive groups capable of crosslinking to HNF1, probing of the HNF1 DNA-binding region has been carried out using a set of modified DNA duplexes containing reactive substituted pyrophosphate or stable pyrophosphate groups at the different positions of the recognition site. Here, we show that HNF1 is able to specifically bind to the modified duplexes and crosslink to the reactive ones. We demonstrate that this approach is generally applicable to cell extracts. This is of principal importance because cell extracts contain many components that could interfere with the crosslinking procedure.

MATERIALS AND METHODS

Oligonucleotides

Oligonucleotides and DNA duplexes used in this study are depicted in Figure 1 . Oligonucleotides [1-32] forming DNA duplexes I-XVI and I'-XVI' and oligonucleotides [33-38] forming DNA duplexes double-stranded (ds) PE56, ds HF30 and mut ds HF30 were synthesized by GENSET (Paris, France) using an automated DNA synthesizer (Applied Biosystems 394/8). 5'-Phosphate oligonucleotides (5'-p) were synthesized using 5' Phosphate-On cyanoethyl phosphoramidite (Clontech) as the phosphorylating reagent. Oligonucleotides carrying a 3'-terminal phosphate group (3'-p) were obtained by the phosphoramidite method, using the 5' phosphorylation reagent as previously described ( 19 ). 5'-End labeling of oligonucleotides was carried out with T4 polynucleotide kinase and [[gamma]- 32 P]ATP, following standard procedures ( 26 ).


Figure 1 . ( A ) Structures of oligonucleotides and DNA duplexes used in this study. The 15 bp pseudopalindromic binding site of HNF1 is underlined. Arrows indicate the positions of modified internucleotide groups. Roman figures indicate the numbers of corresponding modified duplexes. DNA duplexes I-XVI contain substituted pyrophosphate groups at the positions which are indicated by adjacent arrows. DNA duplexes I'-XVI' contain pyrophosphate internucleotide groups at the same positions. Each modified duplex contains only one modified internucleotide group. Arabic figures indicate the numbers of corresponding oligonucleotides. ( B ) Structures of O -methyl-substituted pyrophosphate and pyrophosphate groups.

PE56 ds oligonucleotide (ds PE56) contains the rat albumin promoter sequence -63/-41 which includes the HNF1 binding site ( 23 ). Oligonucleotides [33] and [34] correspond to the upper and lower strands of ds PE56, respectively. HF30 ds oligonucleotide (ds HF30) corresponds to the rat albumin promoter sequence -68/-39 containing a more extended HNF1 binding site. Oligonucleotides [35] and [36] correspond to the upper and lower strands of ds HF30, respectively. Mut HF30 ds oligonucleotide (mut ds HF30) contains an HNF mutated site that abolishes HNF1 binding ( 23 ). Oligonucleotides [37] and [38] correspond to the upper and lower strands of mut HF30, respectively.

Synthesis of oligonucleotides carrying an O -methyl- substituted 3 ' -phosphate group (3 ' -pOCH 3 )

Synthesis of 3'-pOCH 3 oligonucleotides [1-7] was performed as described in ( 27 ). Synthesis of more extended 3'-pOCH 3 oligonucleotides [8-16] was performed as follows: 0.01-0.1 [mu]mol of 3'-p oligonucleotides [8-16] were incubated for 12 h at 4oC in 110 [mu]l of reaction mixture containing 0.25 M MES pH 4.5, 0.5 M MgCl 2 (Buffer A), 35% CH 3 OH and 11 mg (57.4 [mu]mol) N- (3-dimethylaminopropyl)- N '-ethylcarbodiimide (EDC). After incubation, oligonucleotides were recovered by precipitation with 10 volumes of 2% LiClO 4 in acetone, and were further reprecipitated three times by resuspension in 2 M LiClO 4 and addition of 10 volumes of acetone. 3'-pOCH 3 oligonucleotides were isolated by reverse-phase HPLC on a Delta Pak 300 Å column (3.9 * 150 mm, 7 [mu] particle size), using a linear gradient of acetonitrile in 0.1 M triethylammonium acetate buffer pH 7.0. At the next step 3'-pOCH 3 oligonucleotides were desalted by multiple evaporations to dryness from 50% ethanol at 50oC. Methylation efficiency was 75-95% and the final yield after HPLC purification was 60-75%.

Synthesis of oligonucleotides carrying pyrophosphate groups

Oligonucleotides containing internucleotide pyrophosphate groups (p-p) at the positions indicated in Figure 1 were synthesized by template-directed chemical ligation ( 28 ) of the corresponding 3'-phosphate upstream oligonucleotide and 5'-[ 32 P]-phosphate downstream oligonucleotide. Briefly, equimolar amounts (total nucleotide concentration = 10 -3 M) of the 3'-phosphate and the radioactive 5'-phosphate oligonucleotides to be ligated, and the complementary 30mer template (oligonucleotides [36], HF 30 lower strand or [35], HF 30 upper strand) in 0.05 M MES pH 6.0, 0.02 M MgCl 2 (Buffer B) were treated with 0.2 M EDC for 16 h at 4oC in the dark. Ligation products corresponding to the HF30 upper or lower strands carrying p-( 32 P)p pyrophosphate groups at the ligation sites (see Fig. 1 ), were isolated by electrophoresis on a 20% denaturing polyacrylamide gel, followed by elution with 2 M LiClO 4 and precipitation with 5 volumes of acetone.

Synthesis of oligonucleotides carrying O -methyl-substituted pyrophosphate groups

Synthesis of oligonucleotides containing internucleotide O -methyl-substituted pyrophosphate groups (pOCH 3 -p) at the positions indicated in Figure 1 was performed as described for the pyrophosphate-modified oligonucleotides (see above), except that the 3'-phosphate upstream oligonucleotide was replaced by the corresponding O -methyl-substituted 3'-phosphate oligonucleotide (3'-pOCH 3 ).

Plasmid construction

The plasmid pET-HNF1/1-281 contains the cDNA sequence coding for a complete DNA binding domain of HNF1 ( 29 - 31 ), with no additional C-terminal amino acids. This plasmid was constructed as follows. The DNA fragment encoding the region between Met 1 and Leu 281 of rat HNF1 was generated by PCR on pRSV-HNF1 template ( 29 ) using a 5' primer containing an Nde I site overlapping the Met 1 codon, and a 3' primer in which the Leu 281 codon was followed by a stop codon and a Bam HI restriction site. The PCR fragment cut with Nde I and Bam HI was inserted between the corresponding sites in pET-14b (Novagen, Inc.) and sequenced by the dideoxy method.

Nuclear extracts and proteins

Nuclear extracts from rat liver were prepared as described previously ( 23 ). Histidine-tagged HNF1 DNA binding domain was expressed in E.coli strain BL21(DE3)pLysS ( 31 ) transformed with pET-HNF1/1-281, and purified as follows. Cells were grown to mid-log phase and HNF1/1-281 expression was induced by adding 0.4 mM IPTG during 3 h at 37oC. After induction, cells were harvested and lysed by sonication in a buffer containing 5 mM imidazole, 20 mM Tris-HCl pH 7.9, 500 mM NaCl, 1 mM PMSF (Buffer C). Insoluble material was removed by centrifugation, and the supernatant was filtered through a 0.45 [mu]m membrane. The soluble protein extract was then applied to a Ni 2+ affinity column (His-Bind Resin, Novagen Inc.). After extensive washing with buffer C containing 60 mM imidazole, His-tagged HNF1/1-281 was eluted with buffer C containing 1 M imidazole. Protein was >95% pure at this stage, as determined by SDS-PAGE. The purified protein was supplemented with 20% glycerol and was stored at -80oC.

Binding

Binding of the DNA duplexes I-XVI carrying O -methyl-substituted pyrophosphate groups (pOCH 3 -[ 32 P]p) was carried out as previously described ( 4 ). Three fmoles of 32 P-labeled duplex was added to 1.4 ng of His-tagged HNF1 DNA binding domain or 0.5-2 [mu]g nuclear extract in 14 [mu]l buffer containing 10 mM Hepes pH 7.9, 50 mM KCl, 0.1 mM EDTA, 0.5 mM DTT, 10% (vol/vol) glycerol, 0.25 mM PMSF, 2.5 [mu]g/ml aprotinin, 2.5 [mu]g/ml leupeptin, 6 mM MgCl 2 , 6 mM spermidine, 1.5 [mu]g poly(dI-dC) . poly(dI-dC) and 250 ng sonicated salmon sperm DNA. After incubation at 4oC for 10 min, binding was detected by gel shift assay on 6% non-denaturing PAGE gels containing 0.25 * TBE.

Binding of the non-modified duplexes dsPE56, dsHF30, mut ds HF30 and modified DNA duplexes I'-XVI' containing pyrophosphate groups was carried out in the same way. Both 5'-ends of dsPE56, dsHF30, mut ds HF30, or on pyrophosphate group (p-[ 32 P]p) of duplexes I'-XVI' were 32 P-labelled.

Crosslinking experiments

DNA duplexes I-XVI were incubated with purified HNF1 or liver nuclear extract as described above. After incubation at 4oC for 10 min, at 4oC for 12 h or at 20oC for 10 min or 12 h the mixtures were directly loaded on 0.1% SDS-10% PAGE and the crosslinked product separated from non-crosslinked duplex as described earlier ( 23 ). The contribution of a 30mer DNA duplex in the protein mobility is equivalent to 16 kDa.

RESULTS AND DISCUSSION

Synthesis of a set of modified DNA duplexes containing substituted pyrophosphate or pyrophosphate inter- nucleotide groups at different positions of the HNF1 binding site

Crosslinking of ds oligonucleotides carrying substituted pyrophosphate internucleotide groups to DNA-binding proteins can be achieved if nucleophilic amino acid residues in protein are implicated in direct interactions with reactive internal groups. Therefore to design crosslinking reagents on the base of substituted pyrophosphate-modified ds oligonucleotides it is necessary to determine phosphate groups that are in close proximity with such amino acids.


Figure 2 . Electrophoretic analysis of the reaction mixtures after synthesis of upper strands of DNA duplexes I-XIII containing O -methyl-substituted pyrophosphate groups. 1-13, reaction mixtures after synthesis of upper strands of DNA duplexes I-XIII, respectively. For structures see Figure 1; for conditions see Materials and Methods. The 32 P-label was introduced into the 5'-ends of starting downstream oligonucleotides [29-17] forming DNA duplexes I-XIII.

By comparing the X-ray structure of an atypical homeodomain present in the DNA-binding region of HNF1 with the known structures of classical homeodomains some phosphate backbone contacts have been proposed ( 33 ). As an example HNF1-Lys36, HNF1-Arg39 and HNF1-Arg82 are believed to make contacts to the phosphate backbone. In order to determine phosphates from the HNF1 recognition site in contact with nucleophilic amino acid residues, probing of the HNF1 DNA-binding region was performed using substituted pyrophosphate-modified DNA duplexes I-XVI with reactive groups at the positions depicted in Figure 1 . Modified groups were introduced in both upper and lower strands of HNF1 recognition site.

We used a methoxy group as a nonnucleotide substitute. We have previously demonstrated that cleavage of a substituted pyrophosphate group is faster when the non-nucleotide substitute is an alcoxy group ( 17 ). It follows that crosslinking must also be more efficient in this case. The methoxy group was selected also as it had the lowest distorting effect on DNA structure compared with other alcoxy groups.

Synthesis of oligonucleotides carrying O -methyl-substituted pyrophosphate groups was performed by template-induced chemical ligation of an oligonucleotide carrying an OCH 3 substituted 3'-p to another oligomer bearing a 5'-p as described earlier ( 16 , 17 ). Reactions were performed according the following scheme:


Scheme 2. X, Y and Z indicate starting oligonucleotides.

Synthesis of oligonucleotides carrying an O -methyl-substituted 3'-phosphate group (3'-pOCH 3 ) was performed in an aqueous buffer solution with excess of methanol under the action of EDC. The method of terminal phosphate group alkylation in oligonucleotides was developed by Ivanovskaya and co-authors ( 27 ). The reaction routinely proceeds with a quantitative yield within 4-5 h. This is the case for up to 14mer oligonucleotides [1-7]. For more extended oligonucleotides [8-16] we changed slightly the methylation procedure to avoid precipitation of oligonucleotide material during methylation. By decreasing methanol concentration from 50 to 35% and increasing the reaction time to 12 h, we managed to achieve a high methylation efficiency of 3'-phosphorylated oligonucleotides [8-16].

Chemical ligation and isolation of products from reaction mixtures were performed as described in Materials and Methods. As shown in Figure 2 , ligation products were formed in all cases. Ligation efficiency, giving rise to the HF30 upper or lower strand carrying pOCH 3 -[ 32 P]p pyrophosphate groups at the ligation sites (Fig. 1 ), was 10-15%.


Figure 3 . Electrophoretic analysis of the reaction mixtures after synthesis of upper strands of DNA duplexes I'-XIII' containing pyrophosphate groups. 1-13, reaction mixtures after synthesis of upper strands of DNA duplexes I'-XIII', respectively. For structures see Figure 1; for conditions see Materials and Methods. The 32 P-label was introduced into the 5'-ends of starting downstream oligonucleotides [29-17] forming DNA duplexes I'-XIII'.

DNA duplexes I-XVI were obtained after annealing of corresponding pOCH 3 -[ 32 P]p-containing oligonucleotides with an equimolar amount of the complementary strand. The reactivity of these modified duplexes was confirmed as described earlier ( 16 , 18 ). All duplexes I-XVI were selectively cleaved under the action of aqueous solutions of N- methylimidazole or ethylendiamine at pH 8.0 according to the mechanism of nucleophilic substitution at the phosphorus atom as we determined previously ( 16 ). As a result starting oligonucleotides has been observed (data not shown).

In order to determine whether HNF1 would bind the pyrophosphate-modified oligonucleotides and to determine the influence of methyl substitute on the binding efficiency, we obtained a set of DNA duplexes I'-XVI' that were analogous to DNA duplexes I-XVI but contained a stable pyrophosphate groups. Their structures are depicted in Figure 1 .


Figure 4 . Binding of DNA duplexes I-XIII and I'-XIII' to the HNF1 DNA binding domain. ( A ) Autoradiogram of a gel retardation assay using 3 fmol of 5'-end-labeled dsHF30 or body-labeled DNA duplexes I-XIII containing O -methyl-substituted pyrophosphate groups (designated as met1-met13) in the absence (-) or presence (+) of a 1.4 ng of HNF1 DNA binding domain, as indicated. ( B ) Autoradiogram of a gel retardation assay using 3 fmol of 5'-end-labeled dsHF30 or body-labeled DNA duplexes I'-XIII' containing pyrophosphate groups (designated as pp1-pp13) in the absence (-) or presence (+) of a 1.4 ng of HNF1 DNA binding domain. Structures of DNA duplexes HF30, I-XIII, I'-XIII' are depicted in Figure 1; for conditions see Materials and Methods. The arrows indicate the positions of the HNF1 specific complex and the free probes, respectively. ( C ): Schematic representation of the results of binding. The arrows indicate the positions of pyrophosphate (p-p) or O -methyl- substituted pyrophosphate (pOCH 3 -p) groups. Figures indicate the numbers of corresponding duplexes. The binding site of HNF1 is underlined. Strong (+), weak (") binding or the absence of binding (-) are indicated.

Synthesis of oligonucleotides carrying pyrophosphate groups at the positions indicated in Figure 1 was performed also by chemical ligation, except that the O -methyl-substituted 3'-phosphate upstream oligonucleotide was replaced by the corresponding 3'-phosphate. As shown in Figure 3 , ligation efficiency (45-85%) depended on the place of modification.

Specific binding of the DNA duplexes containing pyro- phosphate or O -methyl-substituted pyrophosphate groups to the histidine-tagged HNF1 DNA binding domain

It has been previously demonstrated that ds oligonucleotides containing the HNF1 binding site can specifically interact with HNF1 ( 4 , 23 ). In order to determine if HNF1 is able to specifically bind DNA duplexes carrying pyrophosphate or substituted pyrophosphate groups instead of natural phosphate ones both types of modified DNA duplexes I-XVI and I'-XVI' were tested for binding to the His-tagged DNA binding domain which contains all of the determinants necessary for DNA recognition ( 29 ). Binding was detected by gel retardation shift assay.


Figure 5 . Binding of DNA duplexes I-XII and I'-XIII' to the HNF1 in rat liver nuclear extract. ( A ) Autoradiogram of a gel retardation assay using 3 fmol of 5'-end-labeled dsPE56 or body-labeled DNA duplexes I-XII (designated as 1-12) and 1 [mu]g of rat liver nuclear extract. ( B ) Autoradiogram of a gel retardation assay using 3 fmol of 5'-end-labeled dsPE56 or body-labeled DNA duplexes I'-XIII' (designated as 1-13) and 1 [mu]g of rat liver nuclear extract. For structures of duplexes PE56, I-XII and I'-XIII' see Figure 1; for conditions see Materials and Methods. The arrows indicate the positions of the HNF1 specific complex and the free probes, respectively. Asterisks indicate the position of nonspecific complexes. ODNpp and ODNmet designate duplexes containing pyrophosphate or OCH 3 -substituted pyrophosphate groups.

Figure 4 shows an autoradiogram of 6% non-denaturing PAGE gel after binding of DNA duplexes I-XIII and I'-XIII' to the DNA binding domain of HNF1. As shown in Figure 4 , when we incubated radioactive duplexes with protein, DNA-protein complexes corresponding to the HNF1 DNA binding domain retarded band were observed in most cases. For DNA duplexes I, II, VIII, X, XI and I', II', VIII', XI' the binding efficiency of HNF1 to modified duplexes was similar to the binding efficiency of ds HF30, which does not contain any modifications. For duplexes III, V, XII, XIII and III', V', X', XII', XIII' the binding efficiency was lower and for duplexes IV, VI, VII, IX, XIV-XVI and IV', VI', VII', IX', XIV'-XVI' binding was abolished (for XIV-XVI, XIV'-XVI' data not shown). These data indicate that HNF1 can recognize and effectively bind DNA duplexes containing pyrophosphate or O -methyl-substituted pyrophosphate groups within the recognition site. A correlation between binding affinity of both types of modified duplexes was observed. This means that the presence of methyl substituent at the pyrophosphate group did not influence binding. The results of binding are summarized in Figure 4 C. As shown in Figure 4 C, specific binding efficiency depended on the place of modification and was higher when the ends of the HNF1 binding site were modified. On the contrary, when modified groups were placed near the center of dyad symmetry of the pseudopalindromic binding site, binding was much less effective or was abolished. This can be explained by lengthening of the internucleotide bond in the place of modification and consequently by local change of double helix geometry near the central nucleotide pair of palindromic sequence. This can interfere with the conformational change (presumably bending) of DNA, which requires binding of HNF1 to a B-DNA palindromic sequence ( 32 ). We suggest that this conformational change involves a part of palindromic sequence adjacent to the center of dyad symmetry.


Figure 6 . Competition gel retardation assay for NHF1 binding. The binding was performed using 1 [mu]g of liver nuclear extract and 3 fmol of 5' end-labeled ds PE56 (lanes 1-7) or (pOCH 3 -[ 32 P]p)-labeled duplex X (lanes 8-15), in the presence of indicated amounts of unlabeled ds PE56 (lanes 2-4 and 10-12) or duplex X (lanes 5-7 and 13-15). Lanes designed 0 contained no competing oligonucleotides. The arrows indicate the positions of the HNF1 specific complex and the free probes, respectively.The abolition of HNF1 binding to DNA duplexes containing modified phosphate groups in the center of the HNF1 binding site could limit the usefulness of the method. However, it should be noted that the replacement of phosphate group on pyrophosphate or substituted pyrophosphate ones within the recognition site leading to the abolition of binding has not been previously observed. Effective crosslinking was detected for all investigated proteins ( 18 - 20 , 22 ).

Additionally, it should be noted that after binding of some pyrophosphate-modified DNA duplexes (II', IV'-VIII' and X') to the HNF1 binding domain additional bands were observed in the gel (Fig. 4 B). These bands were formed when the binding procedure is carried out in the absence of HNF1. After addition of the protein in some cases these nonspecific bands disappeared (see, for example, duplex II'). This can be explained by nonspecific binding of these modified duplexes to the components present in the buffer for binding procedure such as aprotinin or leupeptine. The following addition of HNF1 could compete for the binding to their target sequences. However, binding of O -methyl-substituted pyrophosphate DNA duplexes was highly specific.

Specific binding of the DNA duplexes containing pyro- phosphate or O -methyl-substituted pyrophosphate groups to the HNF1 in rat liver nuclear extract

In order to demonstrate the ability of substituted pyrophosphate-modified DNA duplexes to bind specific transcription factors in crude preparations we studied the binding of duplexes I-XVI and I'-XVI' to the HNF1 in rat liver nuclear extract as described above. It was particularly important because cell extracts contain many components that could interfere with the binding procedure. As shown in Figure 5 , after incubation of labeled DNA duplexes with the liver nucleic protein, DNA-protein complexes corresponding to the HNF1 retarded band were detected in all cases mentioned above. However, in general the efficiency of binding was lower. It should be noted that the data obtained for liver nuclear extract correlated with those obtained for the HNF1 DNA binding domain. The binding efficiency decreased from the ends of the HNF1 binding site to their center. Near the center of dyad symmetry binding was abolished.


Figure 7 . SDS-PAGE analysis of crosslinking reactions. ( A) Autoradiogram of 10% SDS gel showing crosslinking of DNA duplexes I-XIII (designated as met1-met13) to the HNF1 DNA binding domain. (-) or (+) indicate the absence or the presence of protein; the arrows indicate the position of the free probes. ( B ) Autoradiogram of 10% SDS gel showing crosslinking of DNA duplex X (designated as met10) to the HNF1 in rat liver nuclear extract. Reactions were performed as described in Materials and Methods at 4oC during 10 min, at 4oC for 12 h (ON) or at 20oC (RT) for 10 min, at 20oC during ON. Arrows indicate the positions of the free probes. Asterisk indicates the position of specific crosslinking complex.

After incubation of DNA-duplexes containing O -methyl- substituted pyrophosphate groups with nuclear cell extract we did not detect any additional bands corresponding to nonspecific complexes of these DNA duplexes with other nuclear proteins. On the contrary, after incubation of pyrophosphate-modified DNA duplexes I'-XVI' with liver extract intensive additional bands were observed almost in all cases (Fig. 5 B). Thus, introduction of methyl substitute into the pyrophosphate internucleotide group not only leads to the formation of reactive compounds that are able to crosslink to DNA-binding proteins, but provides the high specificity of binding.

Sequence specificity of binding of HNF1 to modified DNA duplexes containing pyrophosphate or substituted pyrophosphate groups within the recognition site

The interaction between protein and modified DNA duplexes in the cases when it was observed was shown to be sequence specific by two independent criteria. First, an excess of unlabeled ds PE56 or duplex X acted as a competitors in the binding experiments (Fig. 6 ). A 50-fold excess of competitors almost completely blocked the attachment of labeled modified DNA duplex. At the same time the presence of mut ds HF30 containing an HNF1 mutated site that totally abolishes HNF1 binding ( 23 ) under the same conditions did not compete with labeled modified duplexes for HNF1 binding (data not shown).

Crosslinking of DNA duplexes containing O -methyl- substituted pyrophosphate groups to the HNF1 DNA binding domain or to HNF in rat liver nuclear extract

We have previously demonstrated that ds oligonucleotides carrying substituted pyrophosphate internal groups have been successfully applied to the crosslinking to some DNA-binding proteins ( 18 - 21 ). In order to choose the effective crosslinking reagent for transcription factor HNF1 we used a similar crosslinking procedure.

Histidine-tagged HNF1 DNA binding domain or nuclear extract from rat liver were incubated with body-labeled DNA duplexes I, II, VIII, X, XI, XII and XIII, which are shown to bind specifically to the HNF1 under standard conditions described above. Crosslinking products were separated from non-crosslinked duplexes in SDS-polyacrilamide gels. As shown in Figure 7 , the incubation of O -methyl-substituted DNA duplex X with liver extract or with HNF1 DNA binding domain resulted in crosslinking. In the other cases the formation of the crosslinking products was not observed. Because the covalent attachment of short oligonucleotides has a minor measurable effect on the mobility of this protein in SDS-polyacrylamide gel electrophoresis (see Materials and Methods), these experiments indicate that: (i) in the case of liver extract the crosslinking complex gave rise to a ~100 kDa band corresponding to the 87-93 kDa protein linked to the 16 kDa oligonucleotide, which is in agreement with the data observed earlier ( 23 , 29 ); (ii) the crosslinking complex with the HNF1 DNA binding domain gave rise to a ~60 kDa band corresponding to the ~45 kDa His-tagged HNF1/1-281 ( 29 ) to the 16 kDa DNA duplex.

Crosslinking should to be specific because binding of HNF1 with O -methyl-substituted pyrophosphate-modified DNA duplexes resulted in only one specific DNA-HNF1 complex (see above).

Crosslinking efficiency depended on the reaction time and temperature (Fig. 7 ) and was higher when the reaction mixture was incubated overnight at room temperature. The crosslinking effficiency was 40% of the efficiency of non-covalent association. In order to improve the crosslinking yield we added to the reaction mixture 1 [mu]l of 0.4 M N- methylimidazole (N-MeIm) pH 8.0. It has been established previously that the addition of N-MeIm can lead to the increase of crosslinking efficiency ( 18 , 19 ). However it did not influence the extent of covalent attachment (data not shown).

Thus, DNA duplex X specifically crosslinks with the HNF1 transcription factor in cell extract. As shown in Figure 7 , additional bands corresponding to nonspecific crosslinking to the other nuclear proteins could not be detected. Consequently, our crosslinking procedure is generally applicable to the nuclear cell extract and could constitute a potential therapeutic tool for inhibition of the expression of harmful genes.

Additionally, our results indicate that DNA duplex X contained a substituted pyrophosphate group between nucleosides G and T adjacent to the TAAT recognition sequence. Consequently, nucleophilic amino acid (or amino acids) from the DNA binding region of HNF1 have a specific contact with the phosphate group located between G and T bases in the binding site. This could be lysine or arginine since; as we have described earlier, DNA duplexes containing substituted pyrophosphate groups cleaved efficiently under the action of such amino acids ( 18 ). Hystidine seems to be less probable because in this case phosphoroimidazolide of oligonucleotide should be formed as a result of crosslinking. However, such compounds are unstable in aqueous solution and cleave immediately giving starting materials ( 34 ).

In the natural cognate sequences the GT dinucleotide 5'-adjacent to the TAAT recognition sequence is strictly conserved ( 25 ). In agreement with this it has been suggested that amino acid residues from DNA binding region have specific contacts with GT bases in DNA ( 33 ). Our experimental data confirm that the phosphate group between G and T has a specific contact with residue(s) from DNA binding region of rat liver HNF1.

REFERENCES

1 Maniatis, T., Goodbourn, S. and Fischer, J.A. (1987) Science, 236, 1237-1244. MEDLINE Abstract

2 Wu, H., Holcenberg, J.S., Tomich, J., Chen, J., Jones, P.A., Huang, S.-H. and Calame, K.L. (1990) Gene, 89, 203-209. MEDLINE Abstract

3 Berkowitz, L.A., Riabowd, K.T. and Giulman, M.Z. (1989) Mol. Cell. Biol., 9, 4272-4281.

4 Clusel, C., Ugarte, E., Enjolras, N., Vasseur, M. and Blumenfeld, M. (1993) Nucleic Acids Res., 21, 3405-3411. MEDLINE Abstract

5 Willis, M.C., LeCuyer, K.A., Meisenheimer, K.M., Uhlenbeck, O.C. and Koch, T.Y. (1994) Nucleic Acids Res., 22, 4947-4952. MEDLINE Abstract

6 Lio, J., Sodeoka, M., Lane, W.S. and Verdine, G.L. (1994) Proc. Natl. Acad. Sci. USA, 91, 908-912.

7 Dubreuil, Y.L., Expert - Bezancon, A. and Favre, A. (1991) Nucleic Acids Res., 19, 3653-3660. MEDLINE Abstract

8 Graifer, D.M., Juzumiene, D.I., Wollenzien, P. and Karpova, G.G. (1994) Biochemistry, 3878-3884.

9 Nikiforov, T.T. and Connolly, B.A. (1992) Nucleic Acids Res., 20, 1209-1214. MEDLINE Abstract

10 Sylvers, L.A. and Wower, J. (1993) Bioconjugate Chemistry, 4, 411-418. MEDLINE Abstract

11 Bartholomew, B., Kassavetis, G.A., Braun, B.R. and Geiduschek, E.P. (1990) EMBO J., 9, 2197-2205. MEDLINE Abstract

12 Mayer, A.N. and Barany, F. (1995) Gene, 153, 1-8. MEDLINE Abstract

13 Lee, D.K., Evans, R.K., Blanco, J., Gottesfeld, J. and Johnson, J.D. (1991) J.Biol.Chem., 266, 16478-16784.

14 Chu, B.C.F. and Orgel, L.E. (1992) Nucleic Acids Res., 20, 2497-2502.

15 Gruff, E.S. and Orgel, L.E. (1991) Nucleic Acids Res., 19, 6849-6854. MEDLINE Abstract

16 Kuznetsova, S.A., Ivanovskaya, M.G. and Shabarova, Z.A. (1990) Bioorganicheskaya Khimia, 16, 219-225.

17 Kuznetsova, S.A., Kanevskii, I.E., Florent'ev, V.L., Mirzabekov, A.D. and Shabarova, Z.A. (1994) Mol. Biol., 28, 190-195.

18 Shabarova, Z.A., Sheflyan, G.Y., Kuznetsova, S.A., Kubareva, E.A., Sysoev, O.N., Ivanovskaya, M.G. and Gromova, E.S. (1994) Bioorganicheskaya Khimia, 20, 413-419.

19 Purmal, A.A., Shabarova, Z.A. and Gumport, R.I. (1992) Nucleic Acids Res., 20, 3713-3719. MEDLINE Abstract

20 Sheflyan, G.Y., Kubareva, E.A., Volkov, E.M., Oretskaya, T.S., Gromova, E.S. and Shabarova, Z.A. (1995) Gene, 157, 187-190. MEDLINE Abstract

21 Sheflyan, G.Y., Kubareva, S.A., Kuznetsova, S.A., Karyagina, A.S., Nikolskaya, I.I., Gromova, E.S. and Shabarova, Z.A. (1996) FEBS Lett., 390, 307-310. MEDLINE Abstract

22 Kubareva, E.A., Fedorova, O.A., Gottikh, M.B., Tanaka, H., Malvy, C. and Shabarova, Z.A. (1996) FEBS Lett., 381, 35-38. MEDLINE Abstract

23 Cereghini, S., Blumenfeld, M. and Yaniv, M. (1988) Genes Dev., 2, 957-974. MEDLINE Abstract

24 Blumenfeld, M., Maury, M., Chouard, T., Yaniv, M. and Condamine, H. (1991) Development, 113, 589-599. MEDLINE Abstract

25 Frain, M., Swart, G., Monaci, P., Nicosia, A., Stampfli, S., Frank, R. and Cortese, R. (1989) Cell, 59, 145-157. MEDLINE Abstract

26 Ausubel, F.M., Brent, R., Kungston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A. and Struhl, K. (1989) Current Protocols in Molecular Biology. Greene Publishing Associates and Wiley-Interscience, New York.

27 Ivanovskaya, M.G., Gottikh, M.B. and Shabarova, Z.A. (1987) Nucleosides and Nucleotides, 6, 913-934.

28 Shabarova, Z.A. Biochimie (1988), 70, 1323-1334.

29 Chouard, T., Blumenfeld, M., Bach, I., Vandekerckhove, J., Cereghini, S. and Yaniv, M. (1990) Nucleic Acids Res., 18, 5853-5863. MEDLINE Abstract

30 Nicosia, A., Monaci, P., Tomei, L., De Francesco, R., Nuzzo, M., Stunnenberg, H. and Cortese, R. (1990) Cell, 61, 1225-1236.

31 Tomei, L., Cortese, R. and De Francesco, R. (1992) EMBO J., 11, 4119-4129. MEDLINE Abstract

32 Studier, F.W., Rosenberg, A.H., Dunn, J.J. and Dubendorff, J.W. (1990) Methods Enzymol., 185, 60-89. MEDLINE Abstract

33 Ceska, T.A., Lamers, M., Monaci, P., Nicosia, A., Cortese, R. and Suck, D. (1993) EMBO J., 12, 1805-1810. MEDLINE Abstract

34 Isagulyants, M.G., Ivanovskaya, M.G., Potapov, V.K. and Shabarova, Z.A. (1985) Bioorganicheskaya Khimia, 11, 239-247.

Return

*To whom correspondence should be addressed. Tel: +7 095 939 31 53; Fax: +7 095 939 3181; Email: kuznetsova@biorg.chem.msu.su
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
M. Rogacheva, A. Ishchenko, M. Saparbaev, S. Kuznetsova, and V. Ogryzko
High Resolution Characterization of Formamidopyrimidine-DNA Glycosylase Interaction with Its Substrate by Chemical Cross-linking and Mass Spectrometry Using Substrate Analogs
J. Biol. Chem., October 27, 2006; 281(43): 32353 - 32365.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
V. G. Metelev, O. A. Borisova, E. M. Volkov, T. S. Oretskaya, and N. G. Dolinnaya
New chemically reactive dsDNAs containing single internucleotide monophosphoryldithio links: reactivity of 5'-mercapto-oligodeoxyribonucleotides
Nucleic Acids Res., October 1, 2001; 29(19): 4062 - 4069.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (277K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (17)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Kuznetsova, S.
Right arrow Articles by Shabarova, Z.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kuznetsova, S.
Right arrow Articles by Shabarova, Z.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?