This article was downloaded by: [New York University] On: 15 May 2015, At: 04:57 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Biomolecular Structure and Dynamics Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tbsd20

Labelling of Histone H5 and Its Interaction with DNA. 2. Cooperative Binding of Histone H5 to DNA as Probed By Steady-State Fluorescence and Diffusion-Enhanced Energy Transfer a

a

M. Lerho , M. Favazza & C. Houssier

a

a

Laboratoire de Chimie Macromoléculaire et Chimie Physique , Université de Liège , Sart-Tilman (B6), B-4000 , Liège , Belgium Published online: 21 May 2012.

To cite this article: M. Lerho , M. Favazza & C. Houssier (1990) Labelling of Histone H5 and Its Interaction with DNA. 2. Cooperative Binding of Histone H5 to DNA as Probed By Steady-State Fluorescence and Diffusion-Enhanced Energy Transfer, Journal of Biomolecular Structure and Dynamics, 7:6, 1301-1319, DOI: 10.1080/07391102.1990.10508567 To link to this article: http://dx.doi.org/10.1080/07391102.1990.10508567

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with

primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.

Downloaded by [New York University] at 04:57 15 May 2015

This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Journal ofBiomolecular Structure & Dynamics, ISSN 0739-1102 Volume 7, Issue Number 6 (1990), eAdenine Press (1990).

Labelling of Histone HS and Its Interaction with DNA. 2. Cooperative Binding of Histone HS to DNA as Probed By Steady-State Fluorescence and Diffusion-Enhanced Energy Transfer Downloaded by [New York University] at 04:57 15 May 2015

M. Lerho, M. Favazza and C. Roussier* Laboratoire de Chimie Macromoteculaire et Chimie Physique Universite de Liege, Sart-Tilman (B6) B-4000 Liege, Belgium Abstract The interaction ofhistone HSlabelled with fluorescein isothiocyanate (FITC) with DNA has been studied by fluorescence titration, and diffusion-enhanced fluorescence energy transfer (DEFET) measurements with Th(III) lanthanide chelates as donors. Analysis ofthe binding data by the model of Schwarz and Watanabe(.!. Mol. Bioi. 163,467-484 (1983)) yielded a mean stoichiometry of 60 nucleotides per HS molecule, independently of ionic strength, in the range of 3 to 300 mM NaCl, at very low DNA concentration (6 J.IM in mononucleotide). It ensues an approximate electroneutrality of the saturated complexes. Histone HS molecules appeared to be clustered along the DNA lattice in clusters containing on average 3 to 4 HS molecules separated by about 79 base pairs, at mid-saturation of the binding sites. The interaction process was found highly cooperative but the cooperativity parameter was also insensitive to ionic strength in the above range. DEFET experiments indicated an important decrease of accessibility of the FITC label to the ThHED3A' and ThEDTA- chelates with ionic strength in the 0 to 100 mM NaCI range. In the presence of DNA, HS appears already folded at low ionic strength so that the FITC probe is also not accessible to the donor chelate. The present study constitutes an indispensable preliminary step to further studies on the localization of histone HS in condensed chromatin structures.

Introduction The globular domains of his tones H 1 and HS were shown to interact primarily with the chromatosome assembly at the entry and exit points of the two superhelical DNA turns (1,2). On the other hand, the highly basic flanking C-terminal region interacts strongly with the linker DNA which joins the individual nucleosomes and is thus responsible for chromatin condensation into the 30 nm fibre (3,4). Even

*Auth,or to whom correspondence should be addressed

1301

1302

Lerho etal.

Downloaded by [New York University] at 04:57 15 May 2015

though the histone Hl/H5-DNA complexes have been thoroughly characterized for a long time (5-8), the binding mechanism is not fully understood. Actually, in most reaction conditions described in the literature, mixtures ofHl with DNA precipitated before complete charge neutralization of DNA was achieved. Electron micrographs of these aggregated complexes revealed highly variable morphological features, including double fibres networks, cables, stem-like structures or doughnuts (6-9). On the contrary, soluble complexes were obtained with short DNA fragments (10,11) and very dilute solution of high molecular-weight DNA(> 1X 106 Daltons) (12). In the latter case, extensive charge neutralization was attained (Hl!DNA input ratio ~ 1.4). Moreover, well-defined doughnut-shaped or toroidal particles were observed, characteristic of DNA collapse produced by non-specific polycation condensation (12-14). In this work, we have determined the binding characteristics of the interaction of histone H5 with high molecular-weight DNA H5 was chosen as model histone of the H 1-class, as it is dominant and effective stabilizer of condensed chromatin in birds erythrocytes nuclei. The binding has been studied by following the fluorescence quenching of covalently labelled H5. In the accompanying paper, we discussed the labelling conditions of H5 with fluorescein isothiocyanate (FITC) and verified the structural state of the labelled histone as compared to native one. Since the binding ofH5 to DNA was unambiguously demonstrated to be cooperative at any ionic strength (11 ), we used the formalism developed by Schwarz and Watanabe (15) for the non-specific interaction oflarge ligands with a DNA lattice. We have applied the method of diffusion-enhanced fluorescence energy transfer (DEFET) as a new tool for investigating the localization oflinker histone in extended and condensed chromatin chains. The efficiency of energy transfer is provided by the long lifetime of the excited state ofthe donor, a Tb(III) lanthanide chelate which is free to diffuse in the bulk solution and thus encounters many acceptor molecules (16,17). The FITC chromophore is very suitable as histone-bound acceptor, because its excitation spectrum offers excellent overlap with the 490 nm emission band of terbium. This method has proved to be very useful in the determination of the accessibility of chromophores imbedded in membranes or in macromolecules (17-22). For example, Wensel et al. (21) compared the rates of energy transfer for a number of DNA ligands, namely ethidium, acridine orange and Co(III) bleomycin. A measure of chromophore accessibility is directly provided by the energy transfer rate constant, using a neutral terbium chelate TbHED3A, while positively charged chelates allow to estimate the electrostatic potential in the vicinity of each DNA-bound acceptor. Also ofinterest for our concern is the possibility to infer that an acceptor chromophoric binding site is located inside a protein, or fully exposed at its surface (18,19).1t could thus be expected that the FITC labels, attached to the e-NH 2 lysyl groups of H5, would be strategically located at possible binding sites of the DNA phosphodiester backbone. Therefore, this powerful technique has been aimed at investigating labelled H5, in free form or interacting with DNA, as an indispensable preliminary step to further studies on the localization of Hl/H5 mainly in condensed chromatin structures.

Histone H5 Labeling

1303

Materials and Methods A. Preparation of Fluorescein-Labelled H5 (H5-FITC)

The preparation ofHS-FITC is described in the accompanying paper (23). Labelling ratios from 0.5 to 3.4 were used.

Downloaded by [New York University] at 04:57 15 May 2015

B. Interaction of H5-FITC with DNA and Fluorescence titrations

Titration of DNA by HS-FITC was performed by microvolumes addition of a concentrated HS-FITC solution to calf-thymus DNA in 1 mM sodium cacodylate, at pH 6.5. The highly polymerized DNA sample used (calf-thymus DNA, type I from Sigma), was purified by the detergent method until a residual protein content ofless than 1% was attained. Separate experiments at DNA concentrations of2.6 to 50 flM in mononucleotide were conducted. Mter each HS-FITC addition, the solution was mixed by gentle shaking. Fluorescence intensity measurements were performed by fixed time integration on a Edinburgh modell99 photon-counting fluorescence instrument, using a nitrogen-filled flash lamp as light source, with respective excitation and emission wavelengths of340 and 516 nm. A thorough discussion of the applicability and limitations of various procedures for the analysis of cooperative binding of proteins to nucleic acid lattices has been recently given by Kowalczykowski et a/. (24). C. Terbium Chelate and Accessibility Experiments

The media used throughout this study were 1 mM Na-cacodylate pH 6.5, without (buffer A) or with 0.5 mM Na-EDTA (buffer B). Protein-DNA complexes for accessibility measurements were prepared by direct addition of DNA to a 1-2 flM HS-FITC solution, followed by rapid stirring. The final DNA concentration ranged in this case from 4X 10-4 to 2X 10- 3 M. The complexes were then left at room temperature for about one hour prior to the spectroscopic measurements. The lanthanide complex used for·diffusion-enhanced fluorescence energy transfer measurements, terbium N-(2-hydroxyethyl)-ethylenediaminetriacetate (TbHED3A), has been prepared by Dr. E. Merciny (Laboratoire de Chimie Analytique, Universite de Liege) by direct attack of terbium oxide Tb40 7 . This chelate was added at a fixed final concentration of 1 mM, to the protein-DNA complex just prior to the measurement. The accessibility of the acceptor chromophore (FITC) relative to the neutral donor chelate molecule (TbHED3A) was determined from the changes in terbium fluorescence lifetime measured with the Edinburgh photon-counting instrument equipped with a microsecond Xenon flashlight. Excitation and emission wavelengths were set at 370 and 546 nm, respectively. Relative accessibilities were deduced from the decay rate using the expression (18,19,22):

Lerho eta/.

1304 a

b

I

I I

I

I I

";

.!

E

.,;;:;c

I I

c;

I

?:-

"! :t

.s

I

I

I

I I

I I

I

I

I

I

I

I I

I

..

I

~I

,...

I g Downloaded by [New York University] at 04:57 15 May 2015

I

I I

I

I

I

cJI I

~

I

~

.

: 9•

9 •

i! 0 .i!

l'e

..

..••

1.·

,rJ.

~t 0

ID 107• (Ll, IMI

1.0

Figure 1: Fluorescence titration of a fixed amount of calf-thymus DNA (CP = 2.6 ~by increasing amounts ofH5-FITC a tO mM NaCI (a) and 100 mM NaCl (b), in 1 mM Na-cacodylate pH 6.5. The lattice saturation fraction 9 is given by d/ and w were obtained from the X- and Y-axis intercepts of the straight lines, respectively.

1

1 + k [A] 2

- ="t

"to

[1]

where 1:0 and 1: are the terbium chelate lifetimes in the absence and presence of acceptor, respectively, [A] the acceptor molar concentration and k2 (M- 1s- 1) the second-order energy transfer rate constant. Terbium lifetime 1:0 was found invariant on NaCl and DNA addition, within experimental accuracy(< 1%). The product k2 [A], which represents the energy transfer rate constant kT, is related to the overlap integral between the donor emission and the acceptor absorption spectra. This spectral overlap integral is defined as (16,19):

f

F(/..) e(/..)1._4 d/.. J = ---,;o---fF(/..) d/..

[2]

where F(/..) is the relative fluorescence intensity of the donor at the wavelength A, and e(/..) the molar extinction coefficient of the acceptor at the same wavelength.

Results Interaction of Labelled H5 with DNA

In order to determine the binding parameters for H5-DNA interaction, we titrated a

1305

Downloaded by [New York University] at 04:57 15 May 2015

Histone H5 Labeling

-4

0

2

3

I08x[L] (M)

'

2

3

'

Figure 1 continued

fixed amount ofDNA with H5-FITC. Figure 1 shows typical titration curves without NaCl added (a) and at 100 mM NaCl (b). Labelled histone H5 (L) was added to a solution of DNA 2.6 tJM in mononucleotide (CP) at variable input ratios Cv'[L] 0 with [L] 0 the total H5-FITC concentration. The titration curve consists of two parts: the non-linear left-hand part is characteristic of complex formation while the linear right-hand part, with a steeper slope is parallel to the line obtained for the blank H5FITC added to the buffer, in the absence of DNA Such a profile should display a "break" at the intersection of the two lines, reflecting the fact that the titration has reached the stoichiometric conditions. The titration data were analyzed according to the model of McGhee and von Hippel (25) for non-specific cooperative binding, in the simplified form given by Schwarz and Watanabe (15,26). The site size n is given by the molar ratio of nucleotide residues to ligand saturation concentration, n = CP/[L]s, where the latter is given by the intercept of the saturation asymptote with the [LJo-axis. A mean stoichiometry n of about 60 nucleotides per H5 molecule was found. This value was independent of NaCl concentration, within experimental accuracy, in the range ofO to 0.3 M NaCl. The fluorescence data were analyzed in the following way: the lattice saturation fraction 9 was obtained from the titration curve (9 = d/d 00 , see Figure 1a) and the free protein concentration [L] was given by the mass conservation law: [L) = [L] 0 - 9(C/n). Assuming a highly cooperative interaction (i.e., (w/n)lh. ~ 1), the intrinsic association constant K and the cooperativity parameter w could be readily evaluated from the equation (15,27): (2 9 - 1) [9 (1 - 9)rlh.

=

(w/n)lh. (K w [L] - 1)

[3)

Plots of the left-hand side of this equation against free protein concentration [L] are

Lerho eta/.

1306

Downloaded by [New York University] at 04:57 15 May 2015

Table I Influence of Labelling Ratio and DNA Concentration CP on the Binding Parameters for the Interaction ofH5-FITC with DNA Cr(IJM)

Label. ratio

2.6 2.6 5.0 5.0 25.5 25.5 25.5 50.0

1.37 1.37 3.40 3.40 0.49 0.97 1.38 3.40

NaCI (M) 0.1 0.1

0.1

n (nucl.) 64 ± 12 51± 12 52± 8 65 ± 13 67 ± 5 66 ± 9 58± 12 87 ± 10

Kw (107M- 1)

(J)

390 ± 840 ± 560 ± 380 ± 190 ± 560 ± 340 ± 540 ±

200 220 330 100 70 160 160 100

5.2 5.1 21.5 7.0 2.0 1.2 1.2 0.5

± ± ± ± ± ± ± ±

1.5 0.4 6.3 1.2 0.3 0.2 0.2 0.1

K (104M-I)

nb of measur.

16.2 ± 10 6.1 ± 1.8 47.0 ± 14 18.1 ± 5 9.9 ± 4 4.5 ±I 4.2 ± 1.7 0.9 ± 0.1

2 I 2 2 I 1 2 1

expected to give straight lines with abscissa and ordinate intercepts respectively equal to 1/Kc.o and -(w/n)'~'. In practice, this equation was used for win> 3 and e lying between 0.2 and 0.8 (Figure lc,d). Note that in the papers referenced (15, 27), "K" represents the apparent binding constant, i.e., Kw; this could give rise to some confusion in further comparisons with other works. Table I shows the influence of DNA concentration and labelling ratio on the binding parameters, in the range of2.6 to 50~ in mononucleotide and 0.49 to 3.4 FITC per HS molecule. Considering the experimental uncertainty or reproducibility of these measurements, no significant correlation between these two factors and n, w can be detected. However the trend towards lower values for Kc.o when DNA concentration was raised may be attributed to DNA-DNA interactions ("intermolecular aggregation") at the highest concentrations (26). This is evidenced by the binding isotherm, e versus [L] in Figure 2a. These isotherms are reasonably superimposed for CP