Effect of Acetylation of Histone H4 on Communication in Chromatin

Long-distance interaction plays an important role in the regulation of eukaryotic genes. Chromatin structure is involved in the process, but the role of histone modifications has not been studied. In the present work, the role of acetylation Н4K16 (Н4K16-Ac) to enhancer-promoter communication (EPC) was analyzed. This modification is associated with euchromatin and is involved in the decompaction of chromatin fibers. We have shown that the effect of H4K16-Ac on EPC in vitro depends on the level of chromatin assembly. EPC in chromatin, which lacks nucleosomes at random positions on DNA, is inhibited in the presence of H4K16-Aс. At the same time, EPC in chromatin, in which nucleosomes occupy all available positions on DNA, is somewhat stimulated in the presence of H4K16-Aс.

Transcription of the majority of eukaryotic genes is regulated by enhancers, which are the DNA sequences that bind the transcription factors and upregulate transcription from remote promoters [1,2]. To be activated, genes should be located within the region of a so-called "open chromatin" or euchromatin [3], which is the DNA-histone complex characterized by relaxed structure and certain histone modifications (for example, Н4K16-Ac and H3K27-Ac) [4]. It is also known that most enhancers provide in cis activation of their target promoters, and the DNA fragment dividing them forms a loop [4]. These and other studies showed the importance of chromatin structure to support interactions between the enhancer and promoter [2,5]. Moreover, it was shown that DNA condensation during the chromatin modeling and, thus, the decrease in the average distance between the enhancer and chromatin are not the only reasons why chromatin stimulates interactions between the genomic regions [6].
To understand the exact mechanisms of the influence of individual chromatin components on its plasticity and dynamics, as well as the possibility to sustain effective enhancer-promoter interaction (EPI), we have developed an in vitro transcription system that allowed quantitative analysis of the EPI rate. We obtained a template that contained a promoter and a remote enhancer. A long fragment of DNA between them included 13 "601" synthetic sequences going one after another, characterized by high affinity to his-tones and providing accurate assembly of nucleosomes and formation of the chromatin fiber [7,8]. This experimental system allowed us to demonstrate the important role of the terminal regions ("tails") of the core histones and free DNA regions [9] in sustaining of effective EPI [10].
The present study was aimed to determine the effect of covalent modification of one of the histones (Н4K16-Ac) on the EPI in chromatin. It was previously shown that this modification is associated with euchromatin [11] and affects its structure via decondensation of the chromatin fiber [12].

MATERIALS AND METHODS
Proteins. All proteins of the transcription complex were purified as described in [13].
Chromatin template assembly. Unmodified histone octamers, as well as the octamers with histone H4 acetylated by Lys-16, were obtained as described in [14]. Chromatin with histone terminal domains removed was obtained by treatment of chromatin isolated from chicken erythrocytes with trypsin [15].
Nucleosomes were assembled on the linearized pYP05 plasmid at different octamer and template ratios (0.9 : 1 and 1 : 1) using the gradient dialysis against the buffer that contained from 2 M to 10 mM NaCl [16].
Assessment of the quality of chromatin template assembly. Verification of the expected positions of

RESEARCH ARTICLE
nucleosomes on the used templates was performed by the analysis of the enzyme cleavage sensitivity. The assembled chromatin was incubated with the AluI and ScaI restriction endonucleases (New England Biolabs, United States). DNA was purified and used as the template for polymerase chain reaction (PCR) with a single radioactively labeled primer (method of primer elongation) and DNA Taq-polymerase (New England Biolabs, United States). The primer was annealed right prior the promoter [7].
Transcription. Conditions for the in vitro transcription were optimized to provide maximal use of the chromatin template. Transcription was performed as described in [13].
A closed initiation complex (RPc) was assembled on the template. One transcription round was carried out in 50 μL reaction mixture, which contained Trisacetate buffer (50 mM, pH 8.0), 100 mM potassium acetate, 8 mM magnesium acetate, 27 mM ammonium acetate, 0.7% PEG-8000, 0.2 mM ditiothreitol, 1 nM chromatin or DNA, 10 nM RNA polymerase, 300 nM sigma-factor 54 and the NtrC and NtrB protein regulators of nitrogen metabolism in concentrations of 120 and 400 nM, respectively. All components were first mixed together in the transcription buffer (total volume 40 μL) and incubated at 37°С for 15 min in order to form the closed complex. Next, 5 μL of 40 mM ATP dissolved in the transcription buffer were added to the reaction mixture to a final concentration of 4 mM. The reaction was carried out for 2 min at 37°C in order to form the open initiation complex (RPo). Then, a mixture of four ribonucleosidetriphosphates (NTP) (4 mM each) in the transcription buffer supplemented with 2.5 μCi [α-32 P]-GTP (3000 Ci/mM) and 2 mg/mL heparin was added to the reaction in order to initiate elongation and limit the number of transcription cycles to one round. The reaction was carried out at 37°C for 15 min and was stopped by the phenol : chloroform mixture (1 : 1). Radioactively labeled RNA was purified and analyzed in denaturating polyacrylamide gel. The gel was dried and transferred onto the radiosensitive screen, which was then scanned. The data obtained were analyzed with the OptiQuant (PerkinElmer, USA) program. All experiments were repeated thrice. The mean value of the transcription efficiency and the standard deviation were calculated.

RESULTS AND DISCUSSION
In the present study, we used the experimental in vitro system, which allowed us to perform quantitative analysis of the EPI efficiency on the polynucleosome templates with precisely positioned nucleosomes, which spontaneously formed chromatin fibers under physiological conditions [7,8]. To perform the experiments, we used either the templates totally covered with nucleosomes (Figs. 1a, 1e, template 13) or the templates that lacked one of the nucleosomes at a random position (Fig. 1e, template 13-1). These tem-plates contained either intact histones or histone H4 acetylated at Lys-16. The level of nucleosome assembly was assessed by the analysis of sensitivity of free DNA fragments and fragments bound to proteins to the AluI restriction endonuclease (Figs. 1b, 1c). The presence and location of DNA breaks were estimated by the method of primer elongation.
Studies carried out with the previously developed in vitro transcription system, which allowed us to assess quantitatively the EPI rate on the chromatin template (Fig. 1d), showed that incomplete saturation of DNA with nucleosomes (Fig. 1e, template 13-1one of thirteen histone octamers is absent) leads to an inhibitory effect of Н4K16-Ac on EPI (Figs. 1e, 1f, template 13-1). This results in the distortion of the structure of chromatin fiber, because the Н4K16-Ac distorts the contacts between the positively charged Nterminal domain of the H4 and negatively charged Н2А/H2В part of the neighbor nucleosome. The activation is less expressed on the templates with a higher level of nucleosome assembly (Figs. 1e, 1f, template 13). The data obtained are consistent with the data on the role of histone terminal domains in maintaining EPI [10].
The nucleosome free regions of DNA work as "joints" and enable the compact chromatin fibril to make a turn or loop in order to maintain the effective EPI [9]. Therefore, this structure is characterized by optimal dynamics and remains compact. Acetylation of the Н4K16 leads to chromatin decondensation and decreases the rate of EPI (Fig. 1f, template 13-1).
A higher level of chromatin assembly (Fig. 1, template 13) results in the formation of more rigid chromatin fiber, which is characterized by a nonoptimal dynamics and a higher level of compaction. Acetylation of the Н4K16 distorts the level compaction but, most probably, increases flexibility of the fibril because of opening of free DNA regions. All these events lead to a small upregulation effect of the Н4K16-Ac on the EPI at the saturated level of chromatin assembly (Figs. 1e, 1f, template 13). It is noteworthy that, previously, a decrease in the EPI efficiency was observed if the N-terminal domains of histones were removed on template 13 (Figs. 1a, 1e) [2]. This resulted in loss of not only compactness but also several electrostatic interactions inside the chromatin fiber, which, apparently, affect the sustainment of effective EPI. Hence, the Н4K16-Ac on template 13 induces an effect opposite to that observed after the removal of the N-terminal domains of histones.
The data obtained support the hypothesis that the Н4K16-Ac provides controversial effects on the EPI in chromatin depending on the level of its assembly. This is consistent with the data obtained in vivo [17], which showed that the Н4K16-Ac is located inside relatively small regions of active chromatin and, apparently, opens access to the transcription factor binding sites and provides the chromatin fiber with flexibility   M M similar to that of a "joint" in order to provide effective EPI. Therefore, the rate of EPI in chromatin may increase in two ways: removal of individual nucleosomes [9] or via the acetylation at the Н4K16-Ac as shown in the present study.
FUNDING This work was supported by the Russian Scientific Foundation, project no. 19-74-30003.

COMPLIANCE WITH ETHICAL STANDARDS
This study used neither animals nor people as subjects.