Portal de Periódicos da CAPES

Conformational Changes of DNA Induced by Binding ofChironomus High Mobility Group Protein 1a (cHMG1a)

1997; Elsevier BV; Volume: 272; Issue: 32 Linguagem: Inglês

10.1074/jbc.272.32.19763

1083-351X

Ewa Heyduk, Tomasz Heyduk, Peter Claus, Jacek R. Wiśniewski,

Animal Genetics and Reproduction

High mobility group (HMG) proteins are thought to facilitate assembly of higher order chromatin structure through modulation of DNA conformation. In this work we investigate the bending of a 30-base pair DNA fragment induced by Chironomus HMG1 (cHMG1a), and HMGI (cHMGI) proteins. The DNA bending was measured in solution by monitoring the end-to-end distance between fluorescence probes attached to opposite ends of the DNA fragment. The distance was measured by fluorescence energy transfer using a novel europium chelate as a fluorescence donor. These measurements revealed that the end-to-end distance in the 30-base pair DNA was decreased from ∼100 Å in free DNA to ∼50.5 Å in cHMG1a·DNA complex. The most probable DNA bending angle consistent with these distance measurements is about 150°. The deletion of the charged regulatory domains located close to the C terminus of the HMG1 box domain of cHMG1a protein had no effect on the induced bend angle. The ability to induce a large DNA bend distinguishes the cHMG1 from the cHMGI protein. Only small perturbation of the DNA conformation was observed upon binding of the cHMGI protein. A strong DNA bending activity of cHMG1a and its relative abundance in the cell suggests that this protein plays a very important role in modulation of chromatin structure. High mobility group (HMG) proteins are thought to facilitate assembly of higher order chromatin structure through modulation of DNA conformation. In this work we investigate the bending of a 30-base pair DNA fragment induced by Chironomus HMG1 (cHMG1a), and HMGI (cHMGI) proteins. The DNA bending was measured in solution by monitoring the end-to-end distance between fluorescence probes attached to opposite ends of the DNA fragment. The distance was measured by fluorescence energy transfer using a novel europium chelate as a fluorescence donor. These measurements revealed that the end-to-end distance in the 30-base pair DNA was decreased from ∼100 Å in free DNA to ∼50.5 Å in cHMG1a·DNA complex. The most probable DNA bending angle consistent with these distance measurements is about 150°. The deletion of the charged regulatory domains located close to the C terminus of the HMG1 box domain of cHMG1a protein had no effect on the induced bend angle. The ability to induce a large DNA bend distinguishes the cHMG1 from the cHMGI protein. Only small perturbation of the DNA conformation was observed upon binding of the cHMGI protein. A strong DNA bending activity of cHMG1a and its relative abundance in the cell suggests that this protein plays a very important role in modulation of chromatin structure. Local modulation of DNA structure is an important event in the organization of chromatin. The TATA box-binding regulatory protein induces large conformational changes in DNA, in particular helix unwinding, widening of the minor groove and DNA bending (1Juo Z.S. Chiu T.K. Leiberman P.M. Baikalov I. Berk A.J. Dickerson R.E. J. Mol. Biol. 1996; 261: 239-254Crossref PubMed Scopus (287) Google Scholar). Proteins carrying conserved HMG1 box domain (HMG1-BD) 1The abbreviations used are: HMG1-BD, HMG1 box domain; HMG, high mobility group protein; AMCA-NHS, 7-amino-4-methylcoumarin-3-acetic acid, succinimidyl ester; AMCA(13)-S-S-Py,N-[6–7-amino-4-methylcoumarin-3-acetamido)hexyl]-3′-(2′-pyridyldithio)propionamide; DTPA, diethylenetriaminepentaacetic acid; FRET, fluorescence resonance energy transfer; cHMG1, Chironomus HMG protein 1; cHMGI,Chironomus HMG protein I; bp, base pair(s); SRY, sex-determining region Y factor; LEF, lymphoid enhancer-binding factor. and the (K/R)XRGRP (AT-hook) motif have been found to modulate the DNA conformation by interacting primarily with the minor groove of DNA (for a review, see Ref. 2Bustin M. Reeves R. Prog. Nucleic Acid Res. Mol. Biol. 1996; 54: 35-100Crossref PubMed Google Scholar). The ∼80-amino acid HMG1-BD DNA-binding element is found in eukaryotic proteins involved in diverse physiological processes, including regulation of transcription and development (3Grosschedl R. Giese K. Pagel J. Trends Genet. 1992; 10: 94-100Abstract Full Text PDF Scopus (736) Google Scholar). Each of the HMG1-BDs falls in one of the two classes: those that do or do not recognize specific DNA motifs. The HMG1-BDs of sex-determining region Y factor (SRY) and lymphoid enhancer-binding factor (LEF-1) are the best understood members of the first class. The highly abundant HMG1/2-type proteins and the upstream binding factor are examples of the second class. A common feature of all HMG1-BDs is their ability to distort the conformation of B-DNA. The induced DNA bend by HMG1-BD-carrying proteins is thought to facilitate organization of higher order chromosomal structure (for reviews, see Refs. 2Bustin M. Reeves R. Prog. Nucleic Acid Res. Mol. Biol. 1996; 54: 35-100Crossref PubMed Google Scholar and 3Grosschedl R. Giese K. Pagel J. Trends Genet. 1992; 10: 94-100Abstract Full Text PDF Scopus (736) Google Scholar). The angles of the DNA bend induced by binding of HMG1-BDs of SRY and LEF-1 proteins have been determined by the circular permutation assay (4Giese K. Cox J. Grosschedl R. Cell. 1992; 69: 185-195Abstract Full Text PDF PubMed Scopus (558) Google Scholar, 5Giese K. Pagel J. Grosschedl R. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 3368-3372Crossref PubMed Scopus (108) Google Scholar, 6Lnenicek-Allen M. Read C.M. Crane-Robinson C. Nucleic Acids Res. 1996; 24: 1047-1051Crossref PubMed Scopus (53) Google Scholar) and NMR spectroscopy (7Love J.J. Li X. Case D.A. Giese K. Grosschedl R. Wright P.E. Nature. 1995; 376: 791-795Crossref PubMed Scopus (521) Google Scholar, 8Werner M.H. Huth J.R. Gronenborn A.M. Clore G.M. Cell. 1995; 81: 705-714Abstract Full Text PDF PubMed Scopus (433) Google Scholar). The NMR studies revealed that the HMG1-BDs of human SRY and mouse LEF-1 bend their target DNAs by 70–80° and 107–127°, respectively. The bending of DNA induced by binding of sequence nonspecific HMG1-BD has been only indirectly demonstrated, using the ligase-mediated circularization assay (9Pil P.M. Chow C.S. Lippard S.J. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 9465-9469Crossref PubMed Scopus (196) Google Scholar, 10Paull T.T. Haykinson M.J. Johnson R.C. Genes Dev. 1993; 7: 1521-1534Crossref PubMed Scopus (311) Google Scholar, 11Churchill M.E.A. Jones D.N.M. Glaser T. Hefner H. Searles M.A. Travers A.A. EMBO J. 1995; 14: 1264-1275Crossref PubMed Scopus (94) Google Scholar, 12Teo S.-H. Grasser K.D. Thomas J.O. Eur. J. Biochem. 1995; 230: 943-950Crossref PubMed Scopus (98) Google Scholar, 13Wiśniewski J.R. Heßler K. Claus P. Zechel K. Eur. J. Biochem. 1997; 243: 151-157Crossref PubMed Scopus (18) Google Scholar). The sequences flanking the HMG1-BD at its C terminus are able to modulate its affinity for DNA. Changes in the number of positive and negative charges within the C-terminal portion of the insect HMG1 protein (14Wiśniewski J.R. Schulze E. J. Biol. Chem. 1992; 267: 17170-17177Abstract Full Text PDF PubMed Google Scholar) increase and decrease the DNA-binding affinity of the protein, respectively (15Wiśniewski J.R. Schulze E. J. Biol. Chem. 1994; 269: 10713-10719Abstract Full Text PDF PubMed Google Scholar). Similarly, a C-terminal addition of mainly positively charged residues to the HMG1-BD (box B) of mammalian HMG1 (12Teo S.-H. Grasser K.D. Thomas J.O. Eur. J. Biochem. 1995; 230: 943-950Crossref PubMed Scopus (98) Google Scholar) and of human LEF-1 (6Lnenicek-Allen M. Read C.M. Crane-Robinson C. Nucleic Acids Res. 1996; 24: 1047-1051Crossref PubMed Scopus (53) Google Scholar, 16Giese K. Amsterdam A. Grosschedl R. Genes Dev. 1991; 5: 2567-2578Crossref PubMed Scopus (215) Google Scholar) resulted in an increase of their DNA-binding affinity. The presence of this extension has also been suggested to be essential for DNA bending by HMG1-BD of LEF-1 (6Lnenicek-Allen M. Read C.M. Crane-Robinson C. Nucleic Acids Res. 1996; 24: 1047-1051Crossref PubMed Scopus (53) Google Scholar). An AT-hook DNA-binding motif is the characteristic feature of the proteins of the HMGI/Y family, which comprises four structurally related proteins: the mammalian proteins HMGI and HMGY (17Johnson K.R. Lehn D.A. Elton T.S. Barr P.J. Reeves R. J. Biol. Chem. 1988; 263: 18338-18342Abstract Full Text PDF PubMed Google Scholar, 18Johnson K.R. Lehn D.A. Reeves R. Mol. Cell. Biol. 1989; 9: 2114-2123Crossref PubMed Scopus (224) Google Scholar), HMGI-C (19Manfioletti G. Giancotti V. Bandiera A. Buratti E. Sautiere P. Cary P. Crane-Robinson C. Coles B. Goodwin G.H. Nucleic Acids Res. 1991; 19: 6793-6797Crossref PubMed Scopus (167) Google Scholar), and an insect cHMGI (20Claus P. Schulze E. Wiśniewski J.R. J. Biol. Chem. 1994; 269: 33042-33048Abstract Full Text PDF PubMed Google Scholar). Proteins containing several AT-hook motifs have been found in plants (for a review, see Ref. 21Grasser K.D. Plant J. 1995; 7: 185-192Crossref PubMed Scopus (74) Google Scholar) andDrosophila (D1 protein; Ref. 22Ashley T.C. Pendleton C.G. Jennings W.W Saxena A. Glover C.V.C. J. Biol. Chem. 1989; 264: 8394-8401Abstract Full Text PDF PubMed Google Scholar). Binding of HMGI to a DNA fragment of the positive regulatory domain II of the β-interferon gene resulted in a reduction of a intrinsic curvature of a fragment by 13° (23Falvo J.V. Thanos D. Maniatis T. Cell. 1995; 83: 1101-1111Abstract Full Text PDF PubMed Scopus (277) Google Scholar). In this report we analyzed the conformational changes of a 30-bp DNA upon binding of the Chironomus HMG1a and HMGI proteins. Employing fluorescence resonance energy transfer (FRET) distance measurements, we found that the cHMG1a protein bent the 30-bp DNA fragment by about 150°. We also show that the changes in the DNA-binding affinity of cHMG1a produced by deletion of the charged sequences flanking the HMG-BD have no apparent effect on the angle of DNA bending. The binding of the cHMGI protein induced only a small perturbation of DNA conformation. AMCA-NHS (7-amino-4-methylcoumarin-3-acetic acid, succinimidyl ester) was purchased from Boehringer Mannheim, AMCA(13)-S-S-Py (N-[6–7-amino-4-methylcoumarin-3-acetamido)hexyl]-3′-(2′-pyridyldithio)propionamide) was from Pierce, and CY5 (monofunctionalN-hydroxysuccinimide ester) was from Amersham Corp. Diethylenetriaminepentaacetic acid anhydride (DTPA anhydride) and EuCl3 were from Aldrich. All other chemicals were of highest purity commercially available. The thiol-reactive europium chelate, DTPA-AMCA(13)-S-S-Py, was prepared as described previously (24Heyduk E. Heyduk T. Anal. Biochem. 1997; 248: 216-227Crossref PubMed Scopus (78) Google Scholar). A polymerase chain reaction was used to introduce NdeI restriction site to the N-terminal end of the coding region of the cHMGI cDNA clone (pCW126, Ref. 20Claus P. Schulze E. Wiśniewski J.R. J. Biol. Chem. 1994; 269: 33042-33048Abstract Full Text PDF PubMed Google Scholar). The polymerase chain reaction product was cloned into the vector pGEM-T (Promega), cut withNdeI/EcoRI and ligated with theNdeI-EcoRI fragment of pET3a to give the expression constructs. The cloned DNA inserts were sequenced to verify that only the intended changes had occurred. For the induction of expression, the constructs were transferred to Escherichia coli BL21(DE3). Bacteria carrying constructs for the expression of the cHMG1a protein and its deletion mutant cHMG1a/102 were prepared as described previously (15Wiśniewski J.R. Schulze E. J. Biol. Chem. 1994; 269: 10713-10719Abstract Full Text PDF PubMed Google Scholar). Transformed E. coli BL21(DE3) cells were grown at 37 °C in 5 ml of LB medium in the presence of 300 μg of ampicillin on a rotary shaker with 240 revolutions/min. A 12-h-old culture was used to inoculate 200 ml of fresh LB medium containing 20 mg of ampicillin. The culture was grown to an optical density of 0.4 measured at 600 nm. The expression of the HMG proteins was then induced by adding isopropyl-1-thio-β-d-galactopyranoside to a concentration of 1 mm. The induced culture was shaken for an additional 2 h under the same conditions. The cells were harvested in a centrifuge at 4 °C with 5000 × g. Cell pellets were frozen at −20 °C. Proteins were extracted from bacteria with 5% HClO4 by three thawing-freezing cycles. The supernatants were acidified with HCl to 0.35 m, precipitated with 6 volumes of acetone, and dried under vacuum. The extracted proteins were chromatographed on a MONO S column (Pharmacia, Uppsala, Sweden) using a NaCl gradient in 25 mm boric acid/NaOH, pH 9.4. The fractions containing recombinant proteins eluting as a major peak at about 0.4 mNaCl were pooled, desalted on a PD-10 column (Pharmacia), and lyophilized. Finally, the HMG proteins were re-chromatographed on a Zorbax 300 SB C-18 reverse-phase column using an acetonitrile gradient in 0.1% trifluoroacetic acid/water. The HMG1-BD of cHMG1a (cHMG1a/84) was prepared by a limited digestion of the cHMG1a with trypsin as described previously (15Wiśniewski J.R. Schulze E. J. Biol. Chem. 1994; 269: 10713-10719Abstract Full Text PDF PubMed Google Scholar). All oligonucleotides were obtained from Midland Certified Reagent Co. (Midland, TX). The donor-labeled upper strand was prepared by using two approaches resulting in two different donor-labeled DNA molecules, differing in a size of the linker between DNA and the fluorochrome. In a first approach a reactive thiol group was introduced to a 5′ end of the oligonucleotide by a post-synthetic modification with cystamine of the 5′-phosphorylated oligonucleotide (25Heyduk T. Lee J.C. Biochemistry. 1992; 31: 3682-3688Crossref PubMed Scopus (39) Google Scholar) and subsequent modification with thiol-reactive europium chelate (DTPA-AMCA(13)-S-S-Py; structure and properties described in Ref. 24Heyduk E. Heyduk T. Anal. Biochem. 1997; 248: 216-227Crossref PubMed Scopus (78) Google Scholar). The product was purified by two successive Sephadex G-25 spin columns as described previously (24Heyduk E. Heyduk T. Anal. Biochem. 1997; 248: 216-227Crossref PubMed Scopus (78) Google Scholar). This modification reaction produced a donor-labeled oligonucleotide with a 17 atom linker between the 5′ phosphorus atom and the coumarin ring of the europium chelate. To prepare a donor-labeled oligonucleotide with a shorter linker, a novel two-step procedure was used. The procedure described below allows europium chelate modification of an aliphatic amino group-containing oligonucleotides. In the first step, the 5′ aliphatic amino group was introduced by a post-synthetic modification with ethylenediamine of a 5′-phosphorylated oligonucleotide using procedures analogous to those used for cystamine modification (25Heyduk T. Lee J.C. Biochemistry. 1992; 31: 3682-3688Crossref PubMed Scopus (39) Google Scholar). The 5′-amino oligonucleotide was then modified with AMCA-NHS by incubation of 10 μm oligonucleotide in 0.1m sodium bicarbonate buffer (pH 8.3) for 4 h at room temperature with 2 mm AMCA-NHS. The excess of unreacted AMCA-NHS was removed on a Sephadex G-25 spin column, and the labeled oligonucleotide was further purified by high performance liquid chromatography using a 4.1 mm × 150-mm PRP-1 reverse phase column (Hamilton, Reno, NV). The column was run at 1 ml/min, and a linear gradient of 50 ml of 10–60% “B” was used. Buffer “A” was 25 mm triethylammonium acetate (pH 7.0), and buffer “B” was 25 mm triethylammonium acetate (pH 7.0) containing 50% acetonitrile. Fractions containing AMCA-labeled oligonucleotide were pooled and were dried in Speed-Vac (Savant Instruments Inc., Holbrook, NY). In the second step, 7-amino group of DNA-bound AMCA was modified with DTPA anhydride. Dry AMCA-labeled oligonucleotide was dissolved in 0.2 m HEPES buffer (pH 7.7) and was reacted with DTPA anhydride (20 mg/ml) for 1 h at room temperature. The excess of unreacted DTPA was removed by running the sample twice through a Sephadex G-25 spin column. This two-step procedure results in donor-labeled DNA with a 6-atom linker between 5′ phosphorus atom and the coumarin ring of europium chelate. The structure of the europium chelate itself (Fig. 1 B) is identical to the thiol-reactive europium chelates described previously (24Heyduk E. Heyduk T. Anal. Biochem. 1997; 248: 216-227Crossref PubMed Scopus (78) Google Scholar).Figure 1A, modified 30-bp DNA used in our energy transfer experiments. B, the structure of a donor europium chelate fluorochrome. Europium is coordinated by carboxylic acid groups of the DTPA component. C, binding of the cHMG1a protein (circles) and cHMGI protein (squares) to fluorochrome-modified 30-bp DNA. The binding was monitored by measuring changes in fluorescence polarization (40Heyduk T. Lee J.C. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 1744-1748Crossref PubMed Scopus (210) Google Scholar) of the acceptor (CY5). The excitation was at 647 nm, and the emission was observed at 670 nm.View Large Image Figure ViewerDownload Hi-res image Download (PPT) CY5-labeled complementary oligonucleotide was prepared and purified using methods described previously (24Heyduk E. Heyduk T. Anal. Biochem. 1997; 248: 216-227Crossref PubMed Scopus (78) Google Scholar). To prepare fluorochrome-labeled DNA duplexes, equimolar amounts of europium chelate-labeled 30-nucleotide oligonucleotide were mixed with unlabeled complementary strand (donor-only 30-bp DNA) or with CY5-labeled complementary 30-nucleotide oligonucleotide (donor-acceptor 30-bp DNA) in 50 mm Tris buffer (pH 7.8), 100 mm NaCl containing 10 μm EDTA. The mixtures were heated to 95 °C for 30 s and were allowed to cool slowly to room temperature. Europium chloride (EuCl3) was added to hybridized DNA samples to a concentration of 5 μm, and after 30 min of incubation at room temperature, the excess of europium was removed on a Sephadex G-25 column. The labeled DNA was stored at −20 °C until needed. Steady-state fluorescence and polarization measurements were performed on a SLM 500C spectrofluorometer (SLM Instruments, Inc., Urbana, IL) equipped with an L-format polarization accessory. Luminescence lifetime measurements were performed on a laboratory-built two-channel spectrofluorometer with a pulsed nitrogen laser (LN300, Laser Photonics, Orlando, FL) as an excitation source (24Heyduk E. Heyduk T. Anal. Biochem. 1997; 248: 216-227Crossref PubMed Scopus (78) Google Scholar). This instrument allows simultaneous acquisition of donor decay (at 620 nm) and sensitized acceptor decay curves (at 670 nm). The decay curves were analyzed according to Equation 1.I=∑αiexp(−t/τi)Equation 1 I is the measured intensity at time tafter the excitation pulse, and α i and τ i are the amplitude and the lifetime of the ith decay component, respectively. The decay curves were fitted to Equation 1 by a non-linear regression using SCIENTIST (Micromath Scientific Software, Salt Lake City, UT). The energy transfer (E) between europium chelate and CY5 was calculated from measurements of luminescence lifetime of a donor in the absence (τd) and in the presence of acceptor (τda).E=1−τd/τdaEquation 2 The distances between donor and acceptor were calculated according to Förster theory (26Förster T. Ann. Phys. 1948; 2: 55-75Crossref Scopus (4983) Google Scholar).R6=Ro6(1−E)/EEquation 3 R is a distance between a donor and an acceptor, andR o is a distance at which the energy transfer is 0.5. The R o was calculated using standard procedures (28Stryer L. Annu. Rev. Biochem. 1978; 47: 819-846Crossref PubMed Scopus (1978) Google Scholar, 29Clegg R.M. Methods Enzymol. 1992; 211: 353-388Crossref PubMed Scopus (658) Google Scholar, 30Selvin P.R. Methods Enzymol. 1995; 246: 300-334Crossref PubMed Scopus (511) Google Scholar) and assumptions described by Selvin et al. (27Selvin P.R. Hearst J.E. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10024-10028Crossref PubMed Scopus (275) Google Scholar,31Selvin P.R. Rana T.M. Hearst J.E. J. Am. Chem. Soc. 1994; 116: 6029-6030Crossref Scopus (190) Google Scholar), according to Equation 4.Ro=9.7×103(Jκ2n−4qd) 1/6(Å)Equation 4 The orientation factor (κ2) was assumed to be 2/3 (24Heyduk E. Heyduk T. Anal. Biochem. 1997; 248: 216-227Crossref PubMed Scopus (78) Google Scholar, 27Selvin P.R. Hearst J.E. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10024-10028Crossref PubMed Scopus (275) Google Scholar, 31Selvin P.R. Rana T.M. Hearst J.E. J. Am. Chem. Soc. 1994; 116: 6029-6030Crossref Scopus (190) Google Scholar). The quantum yield of europium in europium chelates was assumed to be 1 in 100% D2O (27Selvin P.R. Hearst J.E. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10024-10028Crossref PubMed Scopus (275) Google Scholar). The quantum yields of europium chelates in water and in water/D2O mixtures were calculated from Equation 5.qd=τd/τd(100%D2O)Equation 5 τd is a lifetime in water or water/D2O mixture, and τd(100% D2O) is a lifetime in 100% D2O. All fluorescence experiments were performed in 10 mm Tris buffer (pH 7.8), 80 mm NaCl, 5% glycerol, 10 μm EDTA. In some experiments 1 mm MgCl2 was included, but the results were essentially identical to those in buffer with no magnesium. The overall design of the experiments was to use the changes in the end-to-end distance of a DNA fragment induced by binding of the cHMG proteins to determine the extent of DNA deformation (bending). The end-to-end distance was measured in solution using FRET between fluorescence probes attached to the opposite ends of a DNA fragment. Fig.1 (A and B) shows a design of the fluorochrome-labeled 30-bp DNA molecules. The term donor-only DNA is used in this work to refer to a 30-bp DNA molecule with an europium chelate attached to the 5′ end of the upper strand. The donor-acceptor DNA refers to a 30-bp DNA with an europium chelate attached to the 5′ end of the upper strand and CY5 fluorochrome attached to the 5′ end of the lower strand. We used europium chelates as donors in FRET measurements because these luminescence probes offer several important advantages compared with organic dye fluorochromes traditionally used for FRET experiments (24Heyduk E. Heyduk T. Anal. Biochem. 1997; 248: 216-227Crossref PubMed Scopus (78) Google Scholar, 27Selvin P.R. Hearst J.E. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10024-10028Crossref PubMed Scopus (275) Google Scholar, 30Selvin P.R. Methods Enzymol. 1995; 246: 300-334Crossref PubMed Scopus (511) Google Scholar). TheR 0 value of europium chelate-CY5 donor-acceptor pair calculated from absorbance and fluorescence emission spectra of the 30-bp fluorochrome-labeled DNA samples (data not shown) was 55 Å, a value identical to the one reported previously (24Heyduk E. Heyduk T. Anal. Biochem. 1997; 248: 216-227Crossref PubMed Scopus (78) Google Scholar). Both the cHMG1a and cHMGI proteins were able to form a complex with fluorochrome-labeled 30-bp DNA, as illustrated by DNA binding experiments in solution using polarization of fluorochrome-labeled DNA fluorescence as an indicator of protein-DNA complex formation (Fig.1 C). Fluorescence anisotropy values of DNA complexes with cHMG1a and cHMGI were different, although the molecular mass values of these proteins are similar (12.9 kDa (14Wiśniewski J.R. Schulze E. J. Biol. Chem. 1992; 267: 17170-17177Abstract Full Text PDF PubMed Google Scholar) and 10.4 kDa (20Claus P. Schulze E. Wiśniewski J.R. J. Biol. Chem. 1994; 269: 33042-33048Abstract Full Text PDF PubMed Google Scholar), respectively), suggesting that the conformation of protein-DNA complex in both cases was different. The stoichiometry of cHMG1a·DNA complexes is not well established. In the presence of cHMG1a, there are probably two major species in solution (free DNA and a 1:1 protein-DNA complex) because: (i) protein concentrations used (0–1100 nm) for FRET measurements were not saturating (see Fig.1 C), (ii) the binding of cHMG1a to a short DNA fragment was shown to be non-cooperative (15Wiśniewski J.R. Schulze E. J. Biol. Chem. 1994; 269: 10713-10719Abstract Full Text PDF PubMed Google Scholar), and (iii) cHMG1a was shown to preferentially bind to AT-rich sequences (14Wiśniewski J.R. Schulze E. J. Biol. Chem. 1992; 267: 17170-17177Abstract Full Text PDF PubMed Google Scholar). Our 30-bp DNA fragment contains a 11-bp stretch of A and T nucleotides at its center, which should promote preferential binding of a single cHMG1a molecule to the center of the DNA fragment (binding site size for cHMG1a was shown to be 11–13 bp; Ref. 15Wiśniewski J.R. Schulze E. J. Biol. Chem. 1994; 269: 10713-10719Abstract Full Text PDF PubMed Google Scholar). Fig.2 shows the decay curves for donor-only 30-bp DNA in the absence and in the presence of cHMG1a. The decays were rigorously single-exponential in the absence and in the presence of cHMG1a, and the lifetimes of donor decay were not significantly changed in the presence of the protein. These data suggest that the donor molecules were not significantly perturbed by the binding of the protein to DNA. Such insensitivity of europium chelate decay to its microenvironment when attached to a macromolecule was observed previously (24Heyduk E. Heyduk T. Anal. Biochem. 1997; 248: 216-227Crossref PubMed Scopus (78) Google Scholar). Single-exponential decays of donor-only samples allow straightforward analysis of the effects observed in donor-acceptor DNA. Fig. 3 shows the decay curves of europium luminescence in donor-acceptor labeled 30-bp DNA in the absence of cHMG1a, with 250 nm cHMG1a, and with 1100 nmcHMG1a (Fig. 3, A–C, respectively). Without the protein, the donor decay was rigorously single-exponential and the determined lifetime was slightly smaller in comparison to donor-only sample (TableI). With cHMG1a, the decay could no longer be fitted to a single-exponential function and a satisfactory fit could only be obtained using double-exponential function. The longer lifetime component decayed with a lifetime essentially identical to the one observed in the absence of the protein. The amplitude of this component decreased with an increase in protein concentration (Fig. 3, B and C). Therefore, we concluded that this component corresponded to a population of DNA molecules that were not bound to the protein. As already discussed, the concentrations of cHMG1a used in luminescence decay experiments were not saturating; thus, the presence of free DNA was expected. The shorter lifetime component (∼200 μs) appears only in the presence of cHMG1a with donor-acceptor DNA, and its amplitude increases with an increase of cHMG1a concentration. Thus, we concluded that this component corresponded to donor-acceptor 30-bp DNA complexed with cHMG1a. The smaller value of the lifetime of this component compared with a lifetime of donor-only DNA molecule suggests that the end-to-end distance of 30-bp DNA in complex with cHMG1a was drastically decreased compared with free DNA.Table ISummary of energy transfer distance measurements in 30-bp DNA fragment in the absence and in the presence of cHMG1a protein and its deletion mutantsSample1-aThe results obtained with donor attached to DNA using a shorter linker (6-atom) are shown in this table.Lifetime1-bIn the case of multiexponential decays (donor-acceptor + protein samples), the lifetime shown is a component assigned to a protein-DNA complex.Energy transferDistanceμsEÅDonor602.9 ± 2Donor + cHMG1a616.7 ± 7Donor-acceptor586.4 ± 30.027 ± 0.01100.0 ± 10Donor-acceptor + cHMG1a226.0 ± 220.63 ± 0.0450.3 ± 1.5Donor + cHMG1a/102614.0 ± 1Donor-acceptor + cHMG1a/102233.0 ± 230.62 ± 0.0450.7 ± 1.5Donor + cHMG1a/84606.0 ± 4Donor-acceptor + cHMG1a/84243.9 ± 470.60 ± 0.0851.4 ± 3.01-a The results obtained with donor attached to DNA using a shorter linker (6-atom) are shown in this table.1-b In the case of multiexponential decays (donor-acceptor + protein samples), the lifetime shown is a component assigned to a protein-DNA complex. Open table in a new tab The assignment of the ∼200-μs component to a DNA in complex with cHMG1a was corroborated by two additional experiments. First, the decay of a sensitized acceptor emission in donor-acceptor DNA was measured in the presence and absence of cHMG1a (Fig.4). In the case of the europium chelate-CY5 pair used in our studies, a lifetime of a donor is in the microsecond range and the lifetime of an acceptor is in the nanosecond range. Thus, any emission of the acceptor recorded in microsecond time scale would be originating from acceptors that were excited by FRET (24Heyduk E. Heyduk T. Anal. Biochem. 1997; 248: 216-227Crossref PubMed Scopus (78) Google Scholar, 27Selvin P.R. Hearst J.E. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10024-10028Crossref PubMed Scopus (275) Google Scholar, 31Selvin P.R. Rana T.M. Hearst J.E. J. Am. Chem. Soc. 1994; 116: 6029-6030Crossref Scopus (190) Google Scholar). Consequently, the lifetime of sensitized acceptor fluorescence should reflect the lifetime of the donor engaged in FRET. Without the protein (Fig. 3 A), only a small background fluorescence is observed (the energy transfer in free DNA is very small, about 2% see Table I). Addition of the protein produced a large increase in the sensitized emission of the acceptor (Fig. 4,B and C) confirming a drastic increase of FRET in donor-acceptor 30-bp DNA with cHMG1a. The major component of sensitized acceptor emission decayed with a lifetime of ∼200 μs, essentially identical to the lifetime of a short component of donor decay. These data confirmed the assignment of a ∼200-μs component to DNA in complex with cHMG1a. A very short-lived component (∼20 μs) could be also observed in sensitized acceptor decays with the protein. Its nature is unclear, since this short-lived component could be observed only in sensitized emission of acceptor but not in donor decay measured directly. Therefore, the ∼2