Cis and Trans Internucleosomal Interactions of H3 and H4 Tails in Tetranucleosomes Nathan P. Nurse, Chongli Yuan School of Chemical Engineering, Purdue University, West Lafayette, IN 47906 Received 13 July 2014; accepted 22 August 2014 Published online 5 September 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/bip.22560
ABSTRACT: Chromatin structure and the transcriptional state of a gene can be modulated by modifications made on H3 and H4 tails of histones. Elucidating the internucleosomal interactions of these tails is vital to understanding
detailed molecular interactions mediated by epigenetic C 2014 Wiley modification of flexible histone tails. V
Periodicals, Inc. Biopolymers 103: 33–40, 2015. Keywords: histone tail; nucleosome array; FCS; PCH; FFS
epigenetic regulation. Differentiation between cis (intranucleosomal) and trans (inter-nucleosomal) interactions is often difficult with conventional techniques since H3 and H4 tails are flexible. The distinction, however, is important because these interactions model short- and
This article was originally published online as an accepted preprint. The “Published Online” date corresponds to the preprint version. You can request a copy of any preprints from the past two calendar years by emailing the Biopolymers editorial office at [email protected].
long-range chromatin interactions respectively and have different bearings in biological processes. Combining FCS
INTRODUCTION
and PCH analysis, we can decouple the contribution of
n eukaryotes, histone epigenetic marks play a significant role in modulating chromatin structure.1 The compaction of chromatin acts as a steric hindrance that can occlude cellular machinery such as transcription factors, thus modulating gene expression.2 Understanding the epigenetic control of chromatin is paramount in understanding diseases, such as cancers and Alzheimer’s disease,3,4 in which this level of control is often abnormal. Histone tail modifications, especially modifications that occur on charged residues (lysine and arginine), can modulate chromatin structure by affecting nucleosomal interactions. Internucleosomal interactions are commonly categorized as either cis or trans interactions, with cis referring to internucleosomal interactions within the same nucleosome array and trans referring to internucleosomal interactions between different arrays. Specifically, cis interactions contribute to the compaction of arrays while trans interactions contribute to array oligomerization. While both interactions can be affected by the presence of histone tail modifications, changes in these two types of interactions can have different implications on chromatin structure and consequently lead to different gene transcription states. The addition of divalent cations (e.g., Mg21)
histone tails to cis and trans effects. A Mg21 gradient was employed to facilitate compaction and oligomerization of tetranucleosomes with H3 and/or H4 tail truncations. H3 tails were found to play a multifunctional role and exhibit the ability to partake in both attractive cis and trans interactions simultaneously. H4 tails partake in attractive cis and repulsive trans interactions at low [Mg21]. These interactions are diminished at higher [Mg21]. Simultaneous H3 and H4 tail truncation inhibited array oligomerization but maintained local array compaction at relatively high [Mg21]. The established experimental approach can be extended to study the Additional Supporting Information may be found in the online version of this article. Correspondence to: Chongli Yuan; e-mail:[email protected] Contract grant sponsor: Purdue Engineering School faculty start-up package and the Purdue Showalter’s Trust Contract grant number: 11098479 C 2014 Wiley Periodicals, Inc. V
Biopolymers Volume 103 / Number 1
I
33
34
Nurse and Yuan
is a common technique used to study histone tail interactions by modulating both cis and trans interactions involved in condensing and compacting chromatin.5–13 The role of histone tail interactions on chromatin structure has been studied using a variety of techniques including differential centrifugation,5,6 ultracentrifugation,6–9 chemical crossorster Resonance Energy Transfer (FRET),14,15 linking,10–12 F€ FRET-fluorescence correlation spectroscopy (FRET-FCS),14,15 small angle X-ray scattering,16 and atomic force microscopy (AFM).13 These studies have revealed that tails from all four core histone proteins, namely H2A, H2B, H3, and H4, contribute to array condensation in an independent and additive manner and that tailless nucleosomes do not form oligomers or aggregates.5,8,13 These studies have also shown that modifications on H3 and H4 tails contribute more significantly to internucleosomal interaction than H2A and H2B tails.6 Biologically, H3 and H4 tails are also of particular interest due to the large number of positively charged residues they contain and the variety of enzymes that are known to modify them.17 One of the primary challenges in deciphering the role of histone tails on nucleosomal interactions is to distinguish between cis and trans internucleosomal interactions. Distinguishing between these two types of interactions is important because cis interactions account for short range interactions between nucleosomes while trans interactions are thought to imitate longer range fiber–fiber interactions.18 Due to the flexible nature of histone tails, the biophysical and biochemical characterizations of histone tail mediated interaction typically cannot easily distinguish between these two types of interactions with a few exceptions. For example, cis and trans interactions have been distinguished previously using cross-linking experiments and found that both H3 and H4 tails are capable of participating in trans internucleosomal interactions under Mg21 induced oligomerization.11,12 The cross-linking experiments, however, typically requires the introduction of chemical modification to specific histone locations, which can also potentially affect the experimental outcome. In addition, crosslinking experiments do not reveal structural changes as a result of the interactions and thus can not reveal the strength of the interaction and may overlook repulsive interactions. To unambiguously determine the contributions of histone tails to cis and trans internucleosomal interactions, we demonstrated the use of fluorescence fluctuation spectroscopy (FFS) to reveal the cis and trans interactions simultaneously in this study. FFS records the fluctuations of fluorescence signals within a small focal volume.19 The fluctuation data can be analyzed by two complementary techniques, namely fluorescence correlation spectroscopy (FCS) and photon counting histogram (PCH)20 The FCS analysis reveals information about the translational diffusivity of the labeled molecules which can be
related to molecular size.21 The revealed hydrodynamic size of the molecule is dependent on both compaction (resulting from cis internucleosomal interactions) and oligomerization (resulting from trans internucleosomal interactions). The same fluorescence fluctuation data can be analyzed using PCH. PCH analysis reveals the molecular brightness of the labeled molecules which is proportional to the oligomerization only. PCH analysis has been used to analyze the protein aggregation and oligomerization process in the literature.22 Combined with FCS, the PCH analysis can be used to reveal the effects of cis and trans internucleosomal interactions respectively. We used a model tetranucleosome array in our study to evaluate the tail effect of histone H3 and H4. Tetranucleosomes were considered as the repeat unit of the dodecanucleosomal 30-nm fiber.23 FFS experiments were performed on nucleosome arrays containing octamers with H3 and/or H4 tail truncations. FFS data were analyzed using both FCS and PCH. Our results suggest that H3 tails have a multifunctional role and can simultaneously partake in both attractive cis and attractive trans interactions. The first 10 residues of H4 tails partake in attractive cis interactions and repulsive trans interactions at low [Mg21] (0–0.32 mM). Increasing [Mg21] diminishes the strength of these H4 tail interactions. Tetranucleosomes with simultaneous truncation of H3 and H4 tails do not oligomerize yet maintain intra array compaction at level similar to wild type arrays.
RESULTS FFS Results of Tetranucleosome Arrays With Different Tail Modifications Among four types of core histone proteins, H3 and H4 tails are known to be directly involved in modulating chromatin structure.5,8–12,24 Most of these studies, however, do not distinguish between cis and trans interactions. To reveal the roles of the flexible region of the H3 and H4 tails on nucleosome array compaction (modulated by cis interactions) and oligomerization (modulated by trans interactions), FFS measurements were performed on nucleosome arrays reconstituted with wild type, DH3, DH4, and DH3DH4 octamers. Array compaction and oligomerization are facilitated by the addition of MgCl2. Mg21 cations are commonly used in chromatin studies to bridge the interactions between neighboring nucleosomes both in short and long ranges.5–7,25 The collected FFS data were analyzed using both FCS and PCH approaches. The diffusivity revealed using FCS accounts for the hydrodynamic radius of arrays and is affected by both array compaction and oligomerization. Increases in attractive cis interactions (or decreases in repulsive cis interactions) can lead to the compaction of Biopolymers
H3 and H4 effects on internucleosomal interactions
FIGURE 1 (a) Diffusivity and (b) Oligomerization number of wild type (squares), truncated H3 tail (circles), truncated H4 tail (triangles), and truncated H3 and H4 tail (diamonds) tetranucleosomes with increasing [Mg21]. Statistical significance was determined by a Student’s t-test with p 5 0.05. Statistical significance between tail truncation statuses is depicted on the chart with symbols indicating the corresponding truncation statuses that are significantly different under identical [Mg21]. The symbols indicating statistical difference from a particular tail truncation status are (*) for wild type, (1) for DH3, (#) for DH4, and () for DH3DH4. Error bars represent standard error. Oligomerization number for DH3DH4 arrays can be found in Supporting Information Figure S2.
nucleosome arrays and an increase in diffusivity. Increases in attractive trans interactions (or decreases in repulsive trans interactions), on the other hand, can result in a decrease in translational diffusivity. It is difficult to decouple the contributions of these interactions without the molecular level details of the formed complex. Molecular brightness, as revealed by Biopolymers
35
PCH, however, is dependent only on oligomerization. Increases in array compactness do not change the molecular brightness. The trends from PCH can, thus, be used to unambiguously determine the trans-interactions. Combining FCS and PCH analysis, we can decouple the effects of cis and trans interactions in determining the diffusivity of array complexes. The results of FCS and PCH analysis are shown in Figures 1a and 1b, respectively. In the absence of Mg21, nucleosome arrays typically exist in a monomeric form.11 As a precaution, the samples were also centrifuged before performing FFS measurements to remove any aggregation that may be in the sample. With no Mg21, unfolded arrays would assume an extended structure and exhibit similar diffusivities. Nevertheless, as shown in Figure 1a, wild type arrays show a 21%, 17%, and 13% higher diffusivity (p < 0.02) than DH3, DH4, and DH3DH4 arrays respectively. This implies that wild type arrays adopt a more compact folding than DH3, DH4, and DH3DH4 arrays. With increasing Mg21, nucleosome arrays, with the exception of DH3DH4, demonstrate a similar diffusivity trend with a decrease in diffusivity at high [Mg21]. This decrease in diffusivity is due to increased oligomerization of the arrays. Significantly, DH3DH4 arrays adopt a conformation that has a 19% higher diffusivity at 0.77 mM Mg21 than at 0 mM Mg21 (p 0.05). (Supporting Information Figure S2) This suggests that the double tail truncation significantly reduces the ability of the tetranucleosomes to oligomerize. This result is consistent with previous studies suggesting that nucleosome arrays lacking H3 and H4 tails do not oligomerize.5,6,9 Due to the lack of oligomerization, the diffusivity of DH3DH4 arrays, as shown in Figure 1a, does not show a dramatic decrease in diffusivity with increasing [Mg21], different from the other three constructs. In order to compare DH3DH4 array diffusivity to wild type array diffusivity, oligomerization number must be accounted for. The diffusivity of proteins can be considered as inversely proportional to the cube root of molecule weight.19,33 Although tetranucleosomes can partake in a range of conformations, the folded structure of the tetranucleosomes has been found to be a truncated two start helix34 which is comparable with a globular shape, so this relationship should be a good approximation here. The expected diffusivity of tetranucleosomes in a monomeric form (DM) can be calculated by correcting the measured diffusivity (Di) by oligomerization number (Ni) according to Eq. (2): 1
DM 5Di Ni3
(2)
The comparison of the diffusivity of DH3DH4 arrays and wild type in monomeric form is shown in Figure 2. At low Mg21 concentrations (
ABSTRACT: Chromatin structure and the transcriptional state of a gene can be modulated by modifications made on H3 and H4 tails of histones. Elucidating the internucleosomal interactions of these tails is vital to understanding
detailed molecular interactions mediated by epigenetic C 2014 Wiley modification of flexible histone tails. V
Periodicals, Inc. Biopolymers 103: 33–40, 2015. Keywords: histone tail; nucleosome array; FCS; PCH; FFS
epigenetic regulation. Differentiation between cis (intranucleosomal) and trans (inter-nucleosomal) interactions is often difficult with conventional techniques since H3 and H4 tails are flexible. The distinction, however, is important because these interactions model short- and
This article was originally published online as an accepted preprint. The “Published Online” date corresponds to the preprint version. You can request a copy of any preprints from the past two calendar years by emailing the Biopolymers editorial office at [email protected].
long-range chromatin interactions respectively and have different bearings in biological processes. Combining FCS
INTRODUCTION
and PCH analysis, we can decouple the contribution of
n eukaryotes, histone epigenetic marks play a significant role in modulating chromatin structure.1 The compaction of chromatin acts as a steric hindrance that can occlude cellular machinery such as transcription factors, thus modulating gene expression.2 Understanding the epigenetic control of chromatin is paramount in understanding diseases, such as cancers and Alzheimer’s disease,3,4 in which this level of control is often abnormal. Histone tail modifications, especially modifications that occur on charged residues (lysine and arginine), can modulate chromatin structure by affecting nucleosomal interactions. Internucleosomal interactions are commonly categorized as either cis or trans interactions, with cis referring to internucleosomal interactions within the same nucleosome array and trans referring to internucleosomal interactions between different arrays. Specifically, cis interactions contribute to the compaction of arrays while trans interactions contribute to array oligomerization. While both interactions can be affected by the presence of histone tail modifications, changes in these two types of interactions can have different implications on chromatin structure and consequently lead to different gene transcription states. The addition of divalent cations (e.g., Mg21)
histone tails to cis and trans effects. A Mg21 gradient was employed to facilitate compaction and oligomerization of tetranucleosomes with H3 and/or H4 tail truncations. H3 tails were found to play a multifunctional role and exhibit the ability to partake in both attractive cis and trans interactions simultaneously. H4 tails partake in attractive cis and repulsive trans interactions at low [Mg21]. These interactions are diminished at higher [Mg21]. Simultaneous H3 and H4 tail truncation inhibited array oligomerization but maintained local array compaction at relatively high [Mg21]. The established experimental approach can be extended to study the Additional Supporting Information may be found in the online version of this article. Correspondence to: Chongli Yuan; e-mail:[email protected] Contract grant sponsor: Purdue Engineering School faculty start-up package and the Purdue Showalter’s Trust Contract grant number: 11098479 C 2014 Wiley Periodicals, Inc. V
Biopolymers Volume 103 / Number 1
I
33
34
Nurse and Yuan
is a common technique used to study histone tail interactions by modulating both cis and trans interactions involved in condensing and compacting chromatin.5–13 The role of histone tail interactions on chromatin structure has been studied using a variety of techniques including differential centrifugation,5,6 ultracentrifugation,6–9 chemical crossorster Resonance Energy Transfer (FRET),14,15 linking,10–12 F€ FRET-fluorescence correlation spectroscopy (FRET-FCS),14,15 small angle X-ray scattering,16 and atomic force microscopy (AFM).13 These studies have revealed that tails from all four core histone proteins, namely H2A, H2B, H3, and H4, contribute to array condensation in an independent and additive manner and that tailless nucleosomes do not form oligomers or aggregates.5,8,13 These studies have also shown that modifications on H3 and H4 tails contribute more significantly to internucleosomal interaction than H2A and H2B tails.6 Biologically, H3 and H4 tails are also of particular interest due to the large number of positively charged residues they contain and the variety of enzymes that are known to modify them.17 One of the primary challenges in deciphering the role of histone tails on nucleosomal interactions is to distinguish between cis and trans internucleosomal interactions. Distinguishing between these two types of interactions is important because cis interactions account for short range interactions between nucleosomes while trans interactions are thought to imitate longer range fiber–fiber interactions.18 Due to the flexible nature of histone tails, the biophysical and biochemical characterizations of histone tail mediated interaction typically cannot easily distinguish between these two types of interactions with a few exceptions. For example, cis and trans interactions have been distinguished previously using cross-linking experiments and found that both H3 and H4 tails are capable of participating in trans internucleosomal interactions under Mg21 induced oligomerization.11,12 The cross-linking experiments, however, typically requires the introduction of chemical modification to specific histone locations, which can also potentially affect the experimental outcome. In addition, crosslinking experiments do not reveal structural changes as a result of the interactions and thus can not reveal the strength of the interaction and may overlook repulsive interactions. To unambiguously determine the contributions of histone tails to cis and trans internucleosomal interactions, we demonstrated the use of fluorescence fluctuation spectroscopy (FFS) to reveal the cis and trans interactions simultaneously in this study. FFS records the fluctuations of fluorescence signals within a small focal volume.19 The fluctuation data can be analyzed by two complementary techniques, namely fluorescence correlation spectroscopy (FCS) and photon counting histogram (PCH)20 The FCS analysis reveals information about the translational diffusivity of the labeled molecules which can be
related to molecular size.21 The revealed hydrodynamic size of the molecule is dependent on both compaction (resulting from cis internucleosomal interactions) and oligomerization (resulting from trans internucleosomal interactions). The same fluorescence fluctuation data can be analyzed using PCH. PCH analysis reveals the molecular brightness of the labeled molecules which is proportional to the oligomerization only. PCH analysis has been used to analyze the protein aggregation and oligomerization process in the literature.22 Combined with FCS, the PCH analysis can be used to reveal the effects of cis and trans internucleosomal interactions respectively. We used a model tetranucleosome array in our study to evaluate the tail effect of histone H3 and H4. Tetranucleosomes were considered as the repeat unit of the dodecanucleosomal 30-nm fiber.23 FFS experiments were performed on nucleosome arrays containing octamers with H3 and/or H4 tail truncations. FFS data were analyzed using both FCS and PCH. Our results suggest that H3 tails have a multifunctional role and can simultaneously partake in both attractive cis and attractive trans interactions. The first 10 residues of H4 tails partake in attractive cis interactions and repulsive trans interactions at low [Mg21] (0–0.32 mM). Increasing [Mg21] diminishes the strength of these H4 tail interactions. Tetranucleosomes with simultaneous truncation of H3 and H4 tails do not oligomerize yet maintain intra array compaction at level similar to wild type arrays.
RESULTS FFS Results of Tetranucleosome Arrays With Different Tail Modifications Among four types of core histone proteins, H3 and H4 tails are known to be directly involved in modulating chromatin structure.5,8–12,24 Most of these studies, however, do not distinguish between cis and trans interactions. To reveal the roles of the flexible region of the H3 and H4 tails on nucleosome array compaction (modulated by cis interactions) and oligomerization (modulated by trans interactions), FFS measurements were performed on nucleosome arrays reconstituted with wild type, DH3, DH4, and DH3DH4 octamers. Array compaction and oligomerization are facilitated by the addition of MgCl2. Mg21 cations are commonly used in chromatin studies to bridge the interactions between neighboring nucleosomes both in short and long ranges.5–7,25 The collected FFS data were analyzed using both FCS and PCH approaches. The diffusivity revealed using FCS accounts for the hydrodynamic radius of arrays and is affected by both array compaction and oligomerization. Increases in attractive cis interactions (or decreases in repulsive cis interactions) can lead to the compaction of Biopolymers
H3 and H4 effects on internucleosomal interactions
FIGURE 1 (a) Diffusivity and (b) Oligomerization number of wild type (squares), truncated H3 tail (circles), truncated H4 tail (triangles), and truncated H3 and H4 tail (diamonds) tetranucleosomes with increasing [Mg21]. Statistical significance was determined by a Student’s t-test with p 5 0.05. Statistical significance between tail truncation statuses is depicted on the chart with symbols indicating the corresponding truncation statuses that are significantly different under identical [Mg21]. The symbols indicating statistical difference from a particular tail truncation status are (*) for wild type, (1) for DH3, (#) for DH4, and () for DH3DH4. Error bars represent standard error. Oligomerization number for DH3DH4 arrays can be found in Supporting Information Figure S2.
nucleosome arrays and an increase in diffusivity. Increases in attractive trans interactions (or decreases in repulsive trans interactions), on the other hand, can result in a decrease in translational diffusivity. It is difficult to decouple the contributions of these interactions without the molecular level details of the formed complex. Molecular brightness, as revealed by Biopolymers
35
PCH, however, is dependent only on oligomerization. Increases in array compactness do not change the molecular brightness. The trends from PCH can, thus, be used to unambiguously determine the trans-interactions. Combining FCS and PCH analysis, we can decouple the effects of cis and trans interactions in determining the diffusivity of array complexes. The results of FCS and PCH analysis are shown in Figures 1a and 1b, respectively. In the absence of Mg21, nucleosome arrays typically exist in a monomeric form.11 As a precaution, the samples were also centrifuged before performing FFS measurements to remove any aggregation that may be in the sample. With no Mg21, unfolded arrays would assume an extended structure and exhibit similar diffusivities. Nevertheless, as shown in Figure 1a, wild type arrays show a 21%, 17%, and 13% higher diffusivity (p < 0.02) than DH3, DH4, and DH3DH4 arrays respectively. This implies that wild type arrays adopt a more compact folding than DH3, DH4, and DH3DH4 arrays. With increasing Mg21, nucleosome arrays, with the exception of DH3DH4, demonstrate a similar diffusivity trend with a decrease in diffusivity at high [Mg21]. This decrease in diffusivity is due to increased oligomerization of the arrays. Significantly, DH3DH4 arrays adopt a conformation that has a 19% higher diffusivity at 0.77 mM Mg21 than at 0 mM Mg21 (p 0.05). (Supporting Information Figure S2) This suggests that the double tail truncation significantly reduces the ability of the tetranucleosomes to oligomerize. This result is consistent with previous studies suggesting that nucleosome arrays lacking H3 and H4 tails do not oligomerize.5,6,9 Due to the lack of oligomerization, the diffusivity of DH3DH4 arrays, as shown in Figure 1a, does not show a dramatic decrease in diffusivity with increasing [Mg21], different from the other three constructs. In order to compare DH3DH4 array diffusivity to wild type array diffusivity, oligomerization number must be accounted for. The diffusivity of proteins can be considered as inversely proportional to the cube root of molecule weight.19,33 Although tetranucleosomes can partake in a range of conformations, the folded structure of the tetranucleosomes has been found to be a truncated two start helix34 which is comparable with a globular shape, so this relationship should be a good approximation here. The expected diffusivity of tetranucleosomes in a monomeric form (DM) can be calculated by correcting the measured diffusivity (Di) by oligomerization number (Ni) according to Eq. (2): 1
DM 5Di Ni3
(2)
The comparison of the diffusivity of DH3DH4 arrays and wild type in monomeric form is shown in Figure 2. At low Mg21 concentrations (
