The N-Terminal Domain of Human Centrin 2 Has a Closed Structure, Binds Calcium with a Very Low Affinity, and Plays a Role in the Protein Self-Assembly†,‡

Centrins are well-conserved calcium binding proteins from the EF-hand superfamily implicated in various cellular functions, such as centrosome duplication, DNA repair, and nuclear mRNA export. The intrinsic molecular flexibility and the self-association tendency make difficult the structural characterization of the integral protein. In this paper we report the solution structure, the Ca2+ binding properties, and the intermolecular interactions of the N-terminal domain of two human centrin isoforms, HsCen1 and HsCen2. In the absence of Ca2+, the N-terminal construct of HsCen2 revealed a compact core conformation including four almost antiparallel alpha-helices and a short antiparallel beta-sheet, very similar to the apo state structure of other calcium regulatory EF-hand domains. The first 25 residues show a highly irregular and dynamic structure. The three-dimensional model for the N-terminal domain of HsCen1, based on the high sequence conservation and NMR spectroscopic data, shows very close structural properties. Ca2+ titration of the apo-N-terminal domain of HsCen1 and HsCen2, monitored by NMR spectroscopy, revealed a very weak affinity (10(2)-10(3) M(-1)), suggesting that the cellular role of this domain is not calcium dependent. Isothermal calorimetric titrations showed that an 18-residue peptide, derived from the N-terminal unstructured fragment, has a significant affinity (approximately 10(5) M(-1)) for the isolated C-terminal domain, suggesting an active role in the self-assembly of centrin molecules.

Solution structure and internal dynamics of NSCP, a compact calcium-binding protein

The solution structure of Nereis diversicolor sarcoplasmic calcium-binding protein (NSCP) in the calcium-bound form was determined by NMR spectroscopy, distance geometry and simulated annealing. Based on 1859 NOE restraints and 262 angular restraints, 17 structures were generated with a rmsd of 0.87 A from the mean structure. The solution structure, which is highly similar to the structure obtained by X-ray crystallography, includes two open EF-hand domains, which are in close contact through their hydrophobic surfaces. The internal dynamics of the protein backbone were determined by studying amide hydrogen/deuterium exchange rates and 15N nuclear relaxation. The two methods revealed a highly compact and rigid structure, with greatly restricted mobility at the two termini. For most of the amide protons, the free energy of exchange-compatible structural opening is similar to the free energy of structural stability, suggesting that isotope exchange of these protons takes place through global unfolding of the protein. Enhanced conformational flexibility was noted in the unoccupied Ca2+-binding site II, as well as the neighbouring helices. Analysis of the experimental nuclear relaxation and the molecular dynamics simulations give very similar profiles for the backbone generalized order parameter (S2), a parameter related to the amplitude of fast (picosecond to nanosecond) movements of N(H)-H vectors. We also noted a significant correlation between this parameter, the exchange rate, and the crystallographic B factor along the sequence.

Calcium and Magnesium Binding to Human Centrin 3 and Interaction with Target Peptides

There are four isoforms of centrin in mammals, with variable sequence, tissue expression, and functional properties. We have recently characterized a number of structural, ion, and target binding properties of human centrin isoform HsCen2. This paper reports a similar characterization of HsCen3, overexpressed in Escherichia coli and purified by phase-reversed chromatography. Equilibrium and dynamic binding studies revealed that HsCen3 has one mixed Ca(2+)/Mg(2+) binding site of high affinity (K(d) = 3 and 10 microM for Ca(2+) and Mg(2+), respectively) and two Ca(2+)-specific sites of low affinity (K(d) = 140 microM). The metal-free protein is fragmented by an unidentified protease into a polypeptide segment of 11 kDa, which was purified by HPLC, and identified by mass spectrometry as the segment of residues 21-112. Similarly, controlled trypsinolysis on Ca(2+)-bound HsCen3 yielded a mixture of segments of residues 1-124 and 1-125. The Ca(2+)/Mg(2+) site could be assigned to this segment and thus resides in the N-terminal half of HsCen3. Temperature denaturation experiments, circular dichroism, and utilization of fluorescence hydrophobic probes allowed us to propose that the metal-free protein has molten globule characteristics and that the dication-bound forms are compact with a polar surface for the Mg(2+) form and a hydrophobic exposed surface for the Ca(2+) form. Thus, HsCen3 could be classified as a Ca(2+) sensor protein. In addition, it is able to bind strongly to a model target peptide (melittin), as well as to peptides derived from the protein XPC and Kar1p, with a moderate Ca(2+) dependence.

SAXS study of the PIR domain from the Grb14 molecular adaptor: a natively unfolded protein with a transient structure primer?

Grb14 belongs to the Grb7 family of adapters and was identified as a negative regulator of insulin signal transduction. Between the PH (pleckstrin homology) and SH2 (Src homology 2) domains is a new binding domain implicated in the interaction with receptor tyrosine kinases called PIR (phosphorylated insulin receptor interaction region). Both PIR and SH2 domains interact with the insulin receptor, but their relative role varies considering the member of the Grb7 family and the tyrosine kinase receptor. In the case of Grb14, PIR is the main binding domain and is sufficient to inhibit the insulin receptor kinase activity. We have proposed, on the basis of NMR measurements, that PIR lacks ordered structure and presents a high flexibility, although remaining fully active. To complement this first study, we have used small-angle x-ray scattering in solution together with a modeling approach representing the PIR domain as a chain of pseudo residues. Circular dichroism experiments were also performed in the presence of variable amounts of trifluoroethanol. These observations, together with an ensemble of sequence analyses and previous NMR results, all support the view of PIR as essentially unstructured but with a potentially structured short stretch encompassing residues 399-407. This stretch, which may be only structured transiently in the isolated molecule, could play a major role in Grb14 PIR binding to a biological partner by undergoing a structural transition.

Calcium-dependent self-assembly of human centrin 2

Human centrin 2 (HsCen2) is a member of the EF-hand superfamily of calcium-binding proteins, often associated with the centrosomes and basal bodies. These organelles exhibit different morphological aspects, including a variety of centrin-containing fibers that connect the two centrioles or other structural elements of the pericentriolar space. The molecular basis of the Ca(2+)-sensitive fibers and their precise role in centrosome duplication are not known. To explore the possible structural role of HsCen2, we initiated a physicochemical study of the self-assembly properties of the purified protein in vitro. Using light scattering experiments, we investigated the temporal evolution of the assembly process and characterized the dependence on various chemical and physical factors, including temperature, di-cation concentration, ionic strength, protein concentration, and pH. The reversible self-assembly revealed many features of a large-size protein polymerization, with nucleation and elongation steps. Kinetic and equilibrium experiments show that a hydrophobic fluorescent probe (ANS) inhibits the polymerization by interfering with the nucleation step, probably through interactions with the apolar exposed sites on the protein surface. A truncated form of HsCen2, lacking the first 25 residues (Delta25HsCen2), shows no detectable self-assembly, pointing to the critical role played by the N-terminal fragment in the supermolecular organization of HsCen2. As revealed by isothermal titration experiments, the isolated N-terminal domains bind with a significant affinity (2 x 10(5) m(-1)) to preformed oligomers of Delta25HsCen2 through an entropy-driven mechanism.

Structural and dynamic studies on ligand-free adenylate kinase from Mycobacterium tuberculosis revealed a closed conformation that can be related to the reduced catalytic activity

Tuberculosis is the leading cause of death worldwide from a single infectious disease. Search of new therapeutic tools requires the discovery and biochemical characterization of new potential targets among the bacterial proteins essential for the survival and virulence. Among them are the nucleoside monophosphate kinases, involved in the nucleotide biosynthesis. In this work, we determined the solution structure of adenylate kinase (AK) from Mycobacterium tuberculosis (AKmt), a protein of 181 residues that was found to be essential for bacterial survival. The structure was calculated by a simulated annealing protocol and energy minimization using experimental restraints, collected by nuclear magnetic resonance spectroscopy. The final, well-defined 20 NMR structures show an average root-mean-square deviation of 0.77 A for the backbone atoms in regular secondary structure segments. The protein has a central CORE domain, composed of a five-stranded parallel beta-sheet surrounded by seven alpha-helices, and two peripheral domains, AMPbd and LID. As compared to other crystallographic structures of free form AKs, AKmt is more compact, with the AMP(bd) domain closer to the CORE of the protein. Analysis of the (15)N relaxation data enabled us to obtain the global rotational correlation time (9.19 ns) and the generalized order parameters (S(2)) of amide vectors along the polypeptide sequence. The protein exhibits restricted movements on a picosecond to nanosecond time scale in the secondary structural regions with amplitudes characterized by an average S(2)() value of 0.87. The loops beta1/alpha1, beta2/alpha2, alpha2/alpha3, alpha3/alpha4, alpha4/beta3, beta3/alpha5, alpha6/alpha7 (LID), alpha7/alpha8, and beta5/alpha9 exhibit rapid fluctuations with enhanced amplitudes. These structural and dynamic features of AKmt may be related to its low catalytic activity that is 10-fold lower than in their eukaryote counterparts.

Deciphering the function of an ORF: Salmonella enterica DeoM protein is a new mutarotase specific for deoxyribose

We identified in Salmonella enterica serovar Typhi a cluster of four genes encoding a deoxyribokinase (DeoK), a putative permease (DeoP), a repressor (DeoQ), and an open reading frame encoding a 337 amino acid residues protein of unknown function. We show that the latter protein, called DeoM, is a hexamer whose synthesis is increased by a factor over 5 after induction with deoxyribose. The CD spectrum of the purified recombinant protein indicated a dominant contribution of betatype secondary structure and a small content of alpha-helix. Temperature and guanidinium hydrochloride induced denaturation of DeoM indicated that the hexamer dissociation and monomer unfolding are coupled processes. DeoM exhibits 12.5% and 15% sequence identity with galactose mutarotase from Lactococcus lactis and respectively Escherichia coli, which suggested that these three proteins share similar functions. Polarimetric experiments demonstrated that DeoM is a mutarotase with high specificity for deoxyribose. Site-directed mutagenesis of His183 in DeoM, corresponding to a catalytically active residue in GalM, yielded an almost inactive deoxyribose mutarotase. DeoM was crystallized and diffraction data collected for two crystal systems, confirmed its hexameric state. The possible role of the protein and of the entire gene cluster is discussed in connection with the energy metabolism of S. enterica under particular growth conditions.

Xeroderma pigmentosum group C protein possesses a high affinity binding site to human centrin 2 and calmodulin

Human centrin 2 (HsCen2), a member of the EF-hand superfamily of Ca2+-binding proteins, is commonly associated with centrosome-related structures. The protein is organized in two domains, each containing two EF-hand motifs, but only the C-terminal half exhibits Ca2+ sensor properties. A significant fraction of HsCen2 is localized in the nucleus, where it was recently found associated with the xeroderma pigmentosum group C protein (XPC), a component of the nuclear excision repair pathway. Analysis of the XPC sequence (940 residues), using a calmodulin target recognition software, enabled us to predict two putative binding sites. The binding properties of the two corresponding peptides were investigated by isothermal titration calorimetry. Only one of the peptides (P1-XPC) interacts strongly (Ka = 2.2 x 10(8) m-1, stoichiometry 1:1) with HsCen2 in a Ca2+-dependent manner. This peptide also binds, with a similar affinity (Ka = 1.1 x 10(8) m-1) to a C-terminal construct of HsCen2, indicating that the interaction with the integral protein is mainly the result of the contribution of the C-terminal half. The second peptide (P2-XPC) failed to show any detectable binding either to HsCen2 or to its C-terminal lobe. The two peptides interact with different affinities and mechanisms with calmodulin. Circular dichroism and nuclear magnetic resonance were used to structurally characterize the complex formed by the C-terminal domain of HsCen2 with P1-XPC.

C-terminal half of human centrin 2 behaves like a regulatory EF-hand domain

Human centrin 2 (HsCen2) is an EF-hand protein that plays a critical role in the centrosome duplication and separation during cell division. We studied the structural and Ca(2+)-binding properties of two C-terminal fragments of this protein: SC-HsCen2 (T94-Y172), covering two EF-hands, and LC-HsCen2 (M84-Y172), having 10 additional residues. Both fragments are highly disordered in the apo state but become better structured (although not conformationally homogeneous) in the presence of Ca(2+) and depending on the nature of the cations (K(+) or Na(+)) in the buffer. Only the longer C-terminal domain, in the Ca(2+)-saturated state and in the presence of Na(+) ions, was amenable to structure determination by nuclear magnetic resonance. The solution structure of LC-HsCen2 reveals an open two EF-hand structure, similar to the conformation of related Ca(2+)-saturated regulatory domains. Unexpectedly, the N-terminal helix segment (F86-T94) lies over the exposed hydrophobic cavity. This unusual intramolecular interaction increases considerably the Ca(2+) affinity and constitutes a useful model for the target binding.

Structural and nucleotide-binding properties of YajQ and YnaF, two Escherichia coli proteins of unknown function

Structural genomics is a new approach in functional assignment of proteins identified via whole-genome sequencing programs. Its rationale is that nonhomologous proteins performing similar or related biological functions might have similar tertiary structure. We used dye pseudoaffinity chromatography, two-dimensional gel electrophoresis, and mass spectrometry to identify two novel Escherichia coli nucleotide-binding proteins, YnaF and YajQ. YnaF exhibited significant sequence identity with MJ0577, an ATP-binding protein from a hyperthermophile (Methanococcus jannaschii), and with UspA, a protein from Haemophilus influenzae that belongs to the Universal Stress Protein family. YnaF conserves the ATP-binding site and the dimeric structure observed in the crystal of MJ0577. The protein YajQ, present in many bacterial genomes, is missing in eukaryotes. In the absence of significant similarities of YajQ to any solved structure, we determined its structural and ligand-binding properties by NMR and isothermal titration calorimetry. We demonstrate that YajQ is composed of two domains, each centered on a beta-sheet, that are connected by two helical segments. NMR studies, corroborated with local sequence conservation among YajQ homologs in various bacteria, indicate that one of the beta-sheets is mostly involved in biological activity.

Solution structure and backbone dynamics of the defunct domain of calcium vector protein

CaVP (calcium vector protein) is a Ca(2+) sensor of the EF-hand protein family which is highly abundant in the muscle of Amphioxus. Its three-dimensional structure is not known, but according to the sequence analysis, the protein is composed of two domains, each containing a pair of EF-hand motifs. We determined recently the solution structure of the C-terminal domain (Trp81-Ser161) and characterized the large conformational and dynamic changes induced by Ca(2+) binding. In contrast, the N-terminal domain (Ala1-Asp86) has lost the capacity to bind the metal ion due to critical mutations and insertions in the two calcium loops. In this paper, we report the solution structure of the N-terminal domain and its backbone dynamics based on NMR spectroscopy, nuclear relaxation, and molecular modeling. The well-resolved three-dimensional structure is typical of a pair of EF-hand motifs, joined together by a short antiparallel beta-sheet. The tertiary arrangement of the two EF-hands results in a closed-type conformation, with near-antiparallel alpha-helices, similar to other EF-hand pairs in the absence of calcium ions. To characterize the internal dynamics of the protein, we measured the (15)N nuclear relaxation rates and the heteronuclear NOE effect in (15)N-labeled N-CaVP at a magnetic field of 11.74 T and 298 K. The domain is mainly monomeric in solution and undergoes an isotropic Brownian rotational diffusion with a correlation time of 7.1 ns, in good agreement with the fluorescence anisotropy decay measurements. Data analysis using a model-free procedure showed that the amide backbone groups in the alpha-helices and beta-strands undergo highly restricted movements on a picosecond to nanosecond time scale. The amide groups in Ca(2+) binding loops and in the linker fragment also display rapid fluctuations with slightly increased amplitudes.

Backbone dynamics of the regulatory domain of calcium vector protein, studied by (15)N relaxation at four fields, reveals unique mobility characteristics of the intermotif linker

CaVP is a calcium-binding protein from amphioxus. It has a modular composition with two domains, but only the two EF-hand motifs localized in the C-terminal domain are functional. We recently determined the solution structure of this regulatory half (C-CaVP) in the Ca(2+)-saturated form and characterized the stepwise ion binding. This paper reports the (15)N nuclear relaxation rates of the Ca(2+)-saturated C-CaVP, measured at four different NMR fields (9.39, 11.74, 14.1, and 18.7 T), which were used to map the spectral density function for the majority of the amide H(N)-N vectors. Fitting the spectral density values at eight frequencies by a model-free approach, we obtained the microdynamic parameters characterizing the global and internal movements of the polypeptide backbone. The two EF-hand motifs, including the ion binding loops, behave like compact structural units with restricted mobility as reflected in the quite uniform order parameter and short internal correlation time (< 20 nsec). Comparative analysis of the two Ca(2+) binding sites shows that site III, having a larger affinity for the metal ion, is generally more rigid, and the amide vector in the second residue of each loop is significantly less restricted. The linker fragment is animated simultaneously by a larger amplitude fast motion and a slow conformational exchange on a microsecond to millisecond time scale. The backbone dynamics of C-CaVP characterized here is discussed in relation with other well-characterized Ca(2+)-binding proteins.

SUPPLEMENTAL MATERIAL

See www.proteinscience.org

Ion-induced conformational and stability changes in Nereis sarcoplasmic calcium binding protein: evidence that the APO state is a molten globule

Nereis sarcoplasmic Ca(2+)-binding protein (NSCP) is a calcium buffer protein that binds Ca(2+) ions with high affinity but is also able to bind Mg(2+) ions with high positive cooperativity. We investigated the conformational and stability changes induced by the two metal ions. The thermal reversible unfolding, monitored by circular dichroism spectroscopy, shows that the thermal stability is maximum at neutral pH and increases in the order apo < Mg(2+) < Ca(2+). The stability against chemical denaturation (urea, guanidinium chloride) studied by circular dichroism or intrinsic fluorescence was found to have a similar ion dependence. To explore in more detail the structural basis of stability, we used the fluorescent probes to evaluate the hydrophobic surface exposure in the different ligation states. The apo-NSCP exhibits accessible hydrophobic surfaces, able to bind fluorescent probes, in clear contrast with denatured or Ca(2+)/Mg(2+)-bound states. Gel filtration experiments showed that, although the metal-bound NSCP has a hydrodynamic volume in agreement with the molecular mass, the volume of the apo form is considerably larger. The present results demonstrate that the apo state has many properties in common with the molten globule. The possible factors of the metal-dependent structural changes and stability are discussed.

Sequential calcium binding to the regulatory domain of calcium vector protein reveals functional asymmetry and a novel mode of structural rearrangement

Calcium vector protein (CaVP) from amphioxus is a two-domain, calcium-binding protein (18.3 kDa) of the calmodulin superfamily. Only two of the four EF-hand motifs (sites III and IV) have a significant binding affinity for calcium ions. We determined the solution structure of the domain containing these active sites (C-CaVP: W81-S161), in the Ca(2+)-saturated state, using NMR spectroscopy and restrained molecular dynamics. The tertiary structure is similar to other Ca(2+)-binding domains containing a pair of EF-hand motifs. The apo state has spectroscopic and thermodynamic characteristics of a molten globule, with conserved secondary structure but highly fluctuating tertiary organization. Titration of C-CaVP with Ca(2+) revealed a stepwise ion binding, with a stable equilibrium intermediate in which only site III binds a calcium ion. Despite a highly fluctuating structure of the free site IV, the calcium-bound site III has a persistent structure, with similar secondary elements but different interhelix angle and hydrophobic packing relative to the fully calcium-saturated state.

A phosphoglycerate kinase mutant (PGK Herlev; D285V) in a Danish patient with isolated chronic hemolytic anemia: mechanism of mutation and structure-function relationships

Phosphoglycerate kinase (PGK) is a X-linked enzyme that plays a key role in the glycolytic pathway. Twelve different variants have already been reported. We describe a new PGK variant, PGK Herlev (Asp 285-->Val), in a 69-year-old Danish patient with isolated chronic hemolysis but who had no neurological or muscular disorders. The description of the mutation is based upon PCR amplification of specific regions of the PGK gene, followed by direct sequencing. Although observed in a male patient, this mutated X-linked gene is expressed partially, i.e., both normal and substituted nucleotides are present at the same position in a ratio of approximately 1:9. The most likely explanation for this observation is based on the occurrence of a somatic mutation of the PGK gene. The relationship of structure to function in PGK Herlev, as well as in all known variants, was examined by the use of a computer model based on the known spatial structure of the yeast and horse enzymes. Such an approach can be generalized to any other protein that has been crystallized and for which x-ray diffraction data are available in a species closely related to man.

Cation- and peptide-binding properties of human centrin 2

Centrin and calmodulin (CaM) are closely related four-EF-hand Ca(2+)-binding proteins. While CaM is monomeric, centrin 2 is dimeric and binds only two Ca(2+) per dimer, likely to site IV in each monomer. Ca(2+) binding to centrin 2 displays pronounced negative cooperativity and a [Ca(2+)](0.5) of 30 microM. As in CaM, Ca(2+) binding leads to the exposure of a hydrophobic probe-accessible patch on the surface of centrin 2. Provided Ca(2+) is present, centrin 2 forms a 1:1 peptide:monomer complex with melittin with an affinity of 100 nM. The complex binds four instead of two Ca(2+). Our data point to surprising differences in the mode of activation of these homologous proteins.

Multinuclear magnetic resonance studies of Escherichia coli adenylate kinase in free and bound forms. Resonance assignment, secondary structure and ligand binding

The crystal structure of Escherichia coli adenylate kinase (AKe) revealed three main components: a CORE domain, composed of a five-stranded parallel beta-sheet surrounded by alpha-helices, and two peripheral domains involved in covering the ATP in the active site (LID) and binding of the AMP (NMPbind). We initiated a long-term NMR study aiming to characterize the solution structure, binding mechanism and internal dynamics of the various domains. Using single (15N) and double-labeled (13C and 15N) samples and double- and triple-resonance NMR experiments we assigned 97% of the 1H, 13C and 15N backbone resonances, and proton and 13Cbeta resonances for more than 40% of the side chains in the free protein. Analysis of a 15N-labeled enzyme in complex with the bi-substrate analogue [P1,P5-bis(5'-adenosine)-pentaphosphate] (Ap5A) resulted in the assignment of 90% of the backbone 1H and 15N resonances and 42% of the side chain resonances. Based on short-range NOEs and 1H and 13C secondary chemical shifts, we identified the elements of secondary structure and the topology of the beta-strands in the unliganded form. The alpha-helices and the beta-strands of the parallel beta-sheet in solution have the same limits (+/- 1 residue) as those observed in the crystal. The first helix (alpha1) appears to have a frayed N-terminal side. Significant differences relative to the crystal were noticed in the LID domain, which in solution exhibits four antiparallel beta-strands. The secondary structure of the nucleoside-bound form, as deduced from intramolecular NOEs and the 1Halpha chemical shifts, is similar to that of the free enzyme. The largest chemical shift differences allowed us to map the regions of protein-ligand contacts. 1H/2H exchange experiments performed on free and Ap5A-bound enzymes showed a general decrease of the structural flexibility in the complex which is accompanied by a local increased flexibility on the N-side of the parallel beta-sheet.

A new subfamily of short bacterial adenylate kinases with the Mycobacterium tuberculosis enzyme as a model: A predictive and experimental study

The adk gene from Mycobacterium tuberculosis codes for an enzyme of 181 amino acids. A sequence comparison with 52 different forms of adenylate kinases (AK) suggests that the enzyme from M. tuberculosis belongs to a new subfamily of "short" bacterial AKs. The recombinant protein, overexpressed in Escherichia coli, exhibits a low catalytic activity and an unexpectedly high thermal stability (Tm = 64.8 degrees C). Based on various spectroscopic data, on the known three-dimensional structure of the AK from E. coli and on secondary structure predictions for various sequenced AKs, we propose a structural model for AK from M. tuberculosis (AKmt). Proteins 1999;36:238-248.