Sp140 is an IFNg-inducible leukocyte-specific member of the Sp100 family proteins. Along with the PML tumor suppressor, Sp100 family proteins are major components of PML-nuclear bodies (PML-NBs), they interact with DNA and various regulatory factors and may be involved in chromatin-dependent transcriptional activation or repression. Sp140 is expressed in all human mature B cells and plasma cell lines, as well as some T cells, suggesting an important role in cellular functions that are peculiar to these cells. In mammalian cells Sp140 behaves as a transcription co-activator for a reporter gene (Bloch et al, 2000). Despite the involvement in B-cell Chronic Lymphocytic Leukaemia (CLL, OMIM151400) (Di Bernardo et al, 2008) and HIV-1 replication (Guldner et al, 1992), until now Sp140 physiological and pathological role is unknown and unexplored, both at cellular and molecular level. The predicted Sp140 amino acid sequence, 867 residues in length, indicates a modular structure similar to other Sp100 family members (Bloch et al, 1996). The N-terminus HSR domain (amino acids 36-157) is responsible both for PML-NBs targeting and homo/hetero-dimerization. The region between residues 306-404 is strongly negatively charged, while the central portion contains a putative bipartite nuclear localization signal, a SAND domain (residues 588-661), a PHD finger (residues 692-734) and a bromodomain (residues 756-859). In this thesis we investigated Sp140 PHD finger both at functional and structural level. Structural determination was performed in solution by means of NMR spectroscopy. Analysis of 1H-15N HSQC spectra of Sp140 15N PHD finger wild type and 15N PHD finger P45A mutant revealed peptidyl-prolyl cis trans isomerization of the T44-P45 peptide bond and a ratio between cis and trans conformers of 1:2. As the NetPhos 2.0 server predicts phosphorylation of the T44 residue by the p38 mitogen-activated protein kinase (p38 MAPK), we suppose that the phosphorylated T44-P45 bond might be a site of regulation of the domain structure through the activity of PIN-1, an enzyme that efficiently and specifically bind to and isomerizes the phosphorylated S/T-P motifs in proteins (Wulf et al, 2005). Using triple resonance NMR experiments we assigned 95% of backbone resonances, 96% of side chain 1H resonances and 73% of side chain non-1H resonances (side chain resonances of OH, SH, CO, NH and NH2 groups were not assigned) of the Sp140 PHD finger trans conformer. The tautomeric state of the two histidines was determined, analyzing a 1H-15N long range correlation HSQC spectrum. The 3J(HNHa) coupling constants and the corresponding phi dihedral angles were calculated by analysis of the 3D HNHA spectrum and were then compared to those predicted by TALOS+ software. We manually assigned more than 850 NOE cross-peaks of the 2D and 3D NOESY spectra. NOEs restraints and 12 dihedral angles were employed for structure calculation of the Sp140 trans conformer by means of ARIA 2.1.3 software. The best NMR ensemble achieved up to now showed a compact globular fold with two short a-helices (stretches 12-EVCR-15 and 50-FCRM-53) as the sole elements of secondary structure. Analysis of the electrostatic surface potential revealed the strong negative charged character of the Sp140 PHD finger trans conformer, with a positive patch only on one side of the domain. Heteronuclear NOE experiments revealed that loop 2, containing the T44-P45 bond which undergoes cis trans isomerization is flexible from E40 to N47. Also the N-terminal tail of the domain is flexible up to residue L8. Flexibility caused a limited number of available NOE restraints in these two regions, especially in loop 2, and consequently an overall high backbone RMSD of the NMR ensemble. In order to increase the number of NOE restraints to calculate the structure we have recently acquired NOESY spectra at higher magnetic fields (900 MHz 1H frequency) that will be soon analyzed. We next investigated the functional role of Sp140 PHD finger and verified its ability to work as histone binding modules. According to sequence alignments, Sp140 PHD finger belongs to the PHD finger subclass specifically recognizing the H3K4me0 epigenetic mark. Indeed, it has an N-terminal conserved aspartic acid which should be involved in electrostatic interactions with the unmodified histone K4 side-chain, as previously demonstrated by both AIRE PHD1 (Chignola et al, 2009) and BHC80 PHD finger (Lan et al, 2007) structures in complex with an unmodified H3 peptide. Unexpectedly tryptophan fluorescence and NMR titrations of Sp140 PHD finger with a 15mer H3K4me0 peptide showed no binding, indicating that the presence of the N-terminal acidic hallmark is not sufficient to predict binding to the H3K4me0 epigenetic mark. We next explored the possibility that SP140 PHD finger might recognize other epigenetic modifications, we therefore extended our binding assays to other synthetic peptides carrying different types of epigenetic marks (such as methylation or acetylation). We also applied a large screening approach using the MODifiedTM Histone Peptide Arrays, which allows the screening of 384 unique modification combinations on the N-terminal tails of histone H3 (up to residue 45), H4 (up to residue 30), H2A and H2B (up to residue 19). However, until now we identified no specific binding of Sp140 PHD finger to the unmodified histone tails and to any combination of histone epigenetic marks spotted on the array (acetylation, methylation, phosphorylation and citrullination). Taken together these data suggest that Sp140 PHD finger is not a classical epigenetic reader and other functions might be attributed to this domain, such as a SUMO E3 ligase activity towards the adjacent bromodomain. This hypothesis was supported by the presence of a typical KxE SUMOylation site at the N-terminus of the Sp140 bromodomain (761-LKCE-764) and by recent data on KAP1 PHD finger, which functions as SUMO E3 ligase for the adjacent bromodomain (Ivanov et al, 2007; Zeng et al, 2008). In support to this hypothesis NMR titration experiments revealed binding of recombinant SUMO E2 ligase Ubc9 to a small, well defined surface of Sp140 PHD finger. We next expressed and purified in E.coli Sp140 PHD finger – bromodomain (PB) tandem and through in vitro SUMOylation test and mass spectrometry analysis we found that the construct was SUMOylated on lysine K837. This residue constitutes an atypical SUMOylation site, located at the bromodomain BC loop. Importantly, sequence alignment reveals that this position corresponds to one of the four SUMOylation sites indentified in KAP1 bromodomain. The intramolecular SUMO E3 ligase activity of Sp140 PHD finger is also supported by the observation that Sp140 PHD finger and bromodomain interact with each other, as deduced by superposition of the 1H-15N HSQC spectra of Sp140 PHD finger alone and PB tandem. Interaction could occur through the hydrophobic core made by stretch 765-FLLLKV-770 in the bromodomain and V695, F712, F718, F732 residues in the PHD finger. Indeed, this is the same core that mediates KAP1 PHD finger and bromodomain interaction, leading to a structural and functional unit whose integrity is fundamental for the KAP1 PHD finger SUMO E3 liagse activity (Zeng et al, 2008). Through NMR titrations we also demonstrated binding of SUMO-1 to both Sp140 PHD finger alone and PB tandem. The binding surface mapped on the homology model of the PB tandem is in agreement with the covalent addition of one SUMO-1 moiety to both the bromodomain K873 and K762 (in the typical KxE SUMOylation site at the bromodomain N-terminus). We are currently performing mutagenesis experiments to confirm the two Sp140 PB tandem SUMOylation sites by means of in vitro SUMOylation tests and mass spectrometry analysis on the Sp140 PB tandem single mutants K762R and K837R. To further validate the intramolecular SUMO E3 ligase activity of the PHD finger we will perform in vitro SUMOylation tests on a PB tandem mutant, in which the function of the PHD finger is inhibited trough an unfolding mutation. In this thesis we have collected strong evidences that in vitro Sp140 PHD finger has SUMO E3 ligase activity for the adjacent bromodomain. This finding might have an important biological relevance in the context of the full length protein: first of all the SUMO E3 ligase activity correlates well with Sp140 localization in leukocytes PML-NBs. PML-NBs are nuclear sub-compartments in which 48% of their protein components show one or more SUMOylation sites which, once they are SUMOylated, work as PML-NBs recruitment signal for these proteins (Van Damme et al, 2010). A similar mechanism could be hypothesized for Sp140 protein, whereby the PHD finger mediates the SUMOylation of the adjacent bromodomain in order to enable the protein recruitment to PML-NBs. SUMOylation might also have a role in the Sp140 transcription co-activating function, other than localization. It is conceivable that Sp140 SUMOylation might recruit proteins able to interact non-covalently with the conjugated SUMO-1 through their SIMs (SUMO Interaction Motifs). Hereby SUMOylated Sp140 protein might serve as a platform for the assembly of a leukocyte-specific transcription complex, whose target genes are still to be identified. Importantly, a similar mechanism has been observed for the co-repressor KAP1; SUMOylation is required for KAP1-mediated gene silencing, because the SUMOylated bromodomain serves as a scaffold and recruits the SETDB1 histone methyltransferase and the CHD3/Mi2 component of the NuRD complex through recognition of the SUMO moieties by the SIMs of these proteins (Ivanov et al, 2007). Through NMR and ITC titrations we found that Sp140 PB tandem is able to bind to a 15mer H3K9ac synthetic peptide. We excluded any PHD finger involvement in this interaction, because when we titrated the H3K9ac peptide to Sp140 PHD finger alone we did not see any binding. Therefore Sp140 bromodomain seems to retain the characteristic ability of bromodomains to bind to acetylated histone tails, despite it does not show the conserved Tyr, Tyr and Asn residues employed by bromodomains in order to bind to acetylated lysines. If confirmed, these results suggest that Sp140 bromodomain is characterized by a new, peculiar mode of recognition of acetylated histone tails. In conclusion, we are collecting important functional and structural data on two domains (PHD finger and bromodomain) of Sp140, a leukocyte-specific nuclear protein involved in B-cell Chronic Lymphocytic Leukaemia and HIV-1 replication, but unexplored up to now. Our functional data strongly indicate that the two domains interact with each other constituting a structural and functional unit in which the PHD finger mediates SUMOylation of the bromodomain. We are solving the NMR solution structure of the Sp140 PHD finger trans conformer. With the exception of the T44-P45 peptidyl-prolyl cis trans isomerization, we are not observing great differences with the solved structures of PHD fingers recognizing histone H3 epigenetic marks, on the contrary of Sp140 PHD finger. Therefore, our data further support the concept that PHD fingers are versatile domains able to perform different activities according subtle but significant structural differences.

THE PHD FINGER OF SP140: A STRUCTURAL AND FUNCTIONAL STUDY / C. Zucchelli ; tutor : Giovanna Musco ; Marcello Maria Duranti; coordinatore del dottorato: Francesco Bonomi. Universita' degli Studi di Milano, 2010 Dec 09. 23. ciclo, Anno Accademico 2010.

THE PHD FINGER OF SP140: A STRUCTURAL AND FUNCTIONAL STUDY

C. Zucchelli
2010

Abstract

Sp140 is an IFNg-inducible leukocyte-specific member of the Sp100 family proteins. Along with the PML tumor suppressor, Sp100 family proteins are major components of PML-nuclear bodies (PML-NBs), they interact with DNA and various regulatory factors and may be involved in chromatin-dependent transcriptional activation or repression. Sp140 is expressed in all human mature B cells and plasma cell lines, as well as some T cells, suggesting an important role in cellular functions that are peculiar to these cells. In mammalian cells Sp140 behaves as a transcription co-activator for a reporter gene (Bloch et al, 2000). Despite the involvement in B-cell Chronic Lymphocytic Leukaemia (CLL, OMIM151400) (Di Bernardo et al, 2008) and HIV-1 replication (Guldner et al, 1992), until now Sp140 physiological and pathological role is unknown and unexplored, both at cellular and molecular level. The predicted Sp140 amino acid sequence, 867 residues in length, indicates a modular structure similar to other Sp100 family members (Bloch et al, 1996). The N-terminus HSR domain (amino acids 36-157) is responsible both for PML-NBs targeting and homo/hetero-dimerization. The region between residues 306-404 is strongly negatively charged, while the central portion contains a putative bipartite nuclear localization signal, a SAND domain (residues 588-661), a PHD finger (residues 692-734) and a bromodomain (residues 756-859). In this thesis we investigated Sp140 PHD finger both at functional and structural level. Structural determination was performed in solution by means of NMR spectroscopy. Analysis of 1H-15N HSQC spectra of Sp140 15N PHD finger wild type and 15N PHD finger P45A mutant revealed peptidyl-prolyl cis trans isomerization of the T44-P45 peptide bond and a ratio between cis and trans conformers of 1:2. As the NetPhos 2.0 server predicts phosphorylation of the T44 residue by the p38 mitogen-activated protein kinase (p38 MAPK), we suppose that the phosphorylated T44-P45 bond might be a site of regulation of the domain structure through the activity of PIN-1, an enzyme that efficiently and specifically bind to and isomerizes the phosphorylated S/T-P motifs in proteins (Wulf et al, 2005). Using triple resonance NMR experiments we assigned 95% of backbone resonances, 96% of side chain 1H resonances and 73% of side chain non-1H resonances (side chain resonances of OH, SH, CO, NH and NH2 groups were not assigned) of the Sp140 PHD finger trans conformer. The tautomeric state of the two histidines was determined, analyzing a 1H-15N long range correlation HSQC spectrum. The 3J(HNHa) coupling constants and the corresponding phi dihedral angles were calculated by analysis of the 3D HNHA spectrum and were then compared to those predicted by TALOS+ software. We manually assigned more than 850 NOE cross-peaks of the 2D and 3D NOESY spectra. NOEs restraints and 12 dihedral angles were employed for structure calculation of the Sp140 trans conformer by means of ARIA 2.1.3 software. The best NMR ensemble achieved up to now showed a compact globular fold with two short a-helices (stretches 12-EVCR-15 and 50-FCRM-53) as the sole elements of secondary structure. Analysis of the electrostatic surface potential revealed the strong negative charged character of the Sp140 PHD finger trans conformer, with a positive patch only on one side of the domain. Heteronuclear NOE experiments revealed that loop 2, containing the T44-P45 bond which undergoes cis trans isomerization is flexible from E40 to N47. Also the N-terminal tail of the domain is flexible up to residue L8. Flexibility caused a limited number of available NOE restraints in these two regions, especially in loop 2, and consequently an overall high backbone RMSD of the NMR ensemble. In order to increase the number of NOE restraints to calculate the structure we have recently acquired NOESY spectra at higher magnetic fields (900 MHz 1H frequency) that will be soon analyzed. We next investigated the functional role of Sp140 PHD finger and verified its ability to work as histone binding modules. According to sequence alignments, Sp140 PHD finger belongs to the PHD finger subclass specifically recognizing the H3K4me0 epigenetic mark. Indeed, it has an N-terminal conserved aspartic acid which should be involved in electrostatic interactions with the unmodified histone K4 side-chain, as previously demonstrated by both AIRE PHD1 (Chignola et al, 2009) and BHC80 PHD finger (Lan et al, 2007) structures in complex with an unmodified H3 peptide. Unexpectedly tryptophan fluorescence and NMR titrations of Sp140 PHD finger with a 15mer H3K4me0 peptide showed no binding, indicating that the presence of the N-terminal acidic hallmark is not sufficient to predict binding to the H3K4me0 epigenetic mark. We next explored the possibility that SP140 PHD finger might recognize other epigenetic modifications, we therefore extended our binding assays to other synthetic peptides carrying different types of epigenetic marks (such as methylation or acetylation). We also applied a large screening approach using the MODifiedTM Histone Peptide Arrays, which allows the screening of 384 unique modification combinations on the N-terminal tails of histone H3 (up to residue 45), H4 (up to residue 30), H2A and H2B (up to residue 19). However, until now we identified no specific binding of Sp140 PHD finger to the unmodified histone tails and to any combination of histone epigenetic marks spotted on the array (acetylation, methylation, phosphorylation and citrullination). Taken together these data suggest that Sp140 PHD finger is not a classical epigenetic reader and other functions might be attributed to this domain, such as a SUMO E3 ligase activity towards the adjacent bromodomain. This hypothesis was supported by the presence of a typical KxE SUMOylation site at the N-terminus of the Sp140 bromodomain (761-LKCE-764) and by recent data on KAP1 PHD finger, which functions as SUMO E3 ligase for the adjacent bromodomain (Ivanov et al, 2007; Zeng et al, 2008). In support to this hypothesis NMR titration experiments revealed binding of recombinant SUMO E2 ligase Ubc9 to a small, well defined surface of Sp140 PHD finger. We next expressed and purified in E.coli Sp140 PHD finger – bromodomain (PB) tandem and through in vitro SUMOylation test and mass spectrometry analysis we found that the construct was SUMOylated on lysine K837. This residue constitutes an atypical SUMOylation site, located at the bromodomain BC loop. Importantly, sequence alignment reveals that this position corresponds to one of the four SUMOylation sites indentified in KAP1 bromodomain. The intramolecular SUMO E3 ligase activity of Sp140 PHD finger is also supported by the observation that Sp140 PHD finger and bromodomain interact with each other, as deduced by superposition of the 1H-15N HSQC spectra of Sp140 PHD finger alone and PB tandem. Interaction could occur through the hydrophobic core made by stretch 765-FLLLKV-770 in the bromodomain and V695, F712, F718, F732 residues in the PHD finger. Indeed, this is the same core that mediates KAP1 PHD finger and bromodomain interaction, leading to a structural and functional unit whose integrity is fundamental for the KAP1 PHD finger SUMO E3 liagse activity (Zeng et al, 2008). Through NMR titrations we also demonstrated binding of SUMO-1 to both Sp140 PHD finger alone and PB tandem. The binding surface mapped on the homology model of the PB tandem is in agreement with the covalent addition of one SUMO-1 moiety to both the bromodomain K873 and K762 (in the typical KxE SUMOylation site at the bromodomain N-terminus). We are currently performing mutagenesis experiments to confirm the two Sp140 PB tandem SUMOylation sites by means of in vitro SUMOylation tests and mass spectrometry analysis on the Sp140 PB tandem single mutants K762R and K837R. To further validate the intramolecular SUMO E3 ligase activity of the PHD finger we will perform in vitro SUMOylation tests on a PB tandem mutant, in which the function of the PHD finger is inhibited trough an unfolding mutation. In this thesis we have collected strong evidences that in vitro Sp140 PHD finger has SUMO E3 ligase activity for the adjacent bromodomain. This finding might have an important biological relevance in the context of the full length protein: first of all the SUMO E3 ligase activity correlates well with Sp140 localization in leukocytes PML-NBs. PML-NBs are nuclear sub-compartments in which 48% of their protein components show one or more SUMOylation sites which, once they are SUMOylated, work as PML-NBs recruitment signal for these proteins (Van Damme et al, 2010). A similar mechanism could be hypothesized for Sp140 protein, whereby the PHD finger mediates the SUMOylation of the adjacent bromodomain in order to enable the protein recruitment to PML-NBs. SUMOylation might also have a role in the Sp140 transcription co-activating function, other than localization. It is conceivable that Sp140 SUMOylation might recruit proteins able to interact non-covalently with the conjugated SUMO-1 through their SIMs (SUMO Interaction Motifs). Hereby SUMOylated Sp140 protein might serve as a platform for the assembly of a leukocyte-specific transcription complex, whose target genes are still to be identified. Importantly, a similar mechanism has been observed for the co-repressor KAP1; SUMOylation is required for KAP1-mediated gene silencing, because the SUMOylated bromodomain serves as a scaffold and recruits the SETDB1 histone methyltransferase and the CHD3/Mi2 component of the NuRD complex through recognition of the SUMO moieties by the SIMs of these proteins (Ivanov et al, 2007). Through NMR and ITC titrations we found that Sp140 PB tandem is able to bind to a 15mer H3K9ac synthetic peptide. We excluded any PHD finger involvement in this interaction, because when we titrated the H3K9ac peptide to Sp140 PHD finger alone we did not see any binding. Therefore Sp140 bromodomain seems to retain the characteristic ability of bromodomains to bind to acetylated histone tails, despite it does not show the conserved Tyr, Tyr and Asn residues employed by bromodomains in order to bind to acetylated lysines. If confirmed, these results suggest that Sp140 bromodomain is characterized by a new, peculiar mode of recognition of acetylated histone tails. In conclusion, we are collecting important functional and structural data on two domains (PHD finger and bromodomain) of Sp140, a leukocyte-specific nuclear protein involved in B-cell Chronic Lymphocytic Leukaemia and HIV-1 replication, but unexplored up to now. Our functional data strongly indicate that the two domains interact with each other constituting a structural and functional unit in which the PHD finger mediates SUMOylation of the bromodomain. We are solving the NMR solution structure of the Sp140 PHD finger trans conformer. With the exception of the T44-P45 peptidyl-prolyl cis trans isomerization, we are not observing great differences with the solved structures of PHD fingers recognizing histone H3 epigenetic marks, on the contrary of Sp140 PHD finger. Therefore, our data further support the concept that PHD fingers are versatile domains able to perform different activities according subtle but significant structural differences.
9-dic-2010
Settore BIO/10 - Biochimica
Sp140 ; PHD finger ; SUMOylation
DURANTI, MARCELLO MARIA
BONOMI, FRANCESCO
Doctoral Thesis
THE PHD FINGER OF SP140: A STRUCTURAL AND FUNCTIONAL STUDY / C. Zucchelli ; tutor : Giovanna Musco ; Marcello Maria Duranti; coordinatore del dottorato: Francesco Bonomi. Universita' degli Studi di Milano, 2010 Dec 09. 23. ciclo, Anno Accademico 2010.
File in questo prodotto:
File Dimensione Formato  
PhD_unimi_ZUCCHELLI_CHIARA_R07804_1.pdf

accesso riservato

Tipologia: Tesi di dottorato completa
Dimensione 5.7 MB
Formato Adobe PDF
5.7 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
PhD_unimi_ZUCCHELLI_CHIARA_R07804_2.pdf

accesso riservato

Tipologia: Tesi di dottorato completa
Dimensione 93.11 kB
Formato Adobe PDF
93.11 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
PhD_unimi_ZUCCHELLI_CHIARA_R07804_3.pdf

accesso riservato

Tipologia: Tesi di dottorato completa
Dimensione 249.85 kB
Formato Adobe PDF
249.85 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
PhD_unimi_ZUCCHELLI_CHIARA_R07804_4.pdf

accesso riservato

Tipologia: Tesi di dottorato completa
Dimensione 32.62 kB
Formato Adobe PDF
32.62 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2434/148885
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact