New insights about the monomer and homodimer structures of the human AOX1.

Human aldehyde oxidase (hAOX1) is a molybdenum dependent enzyme that plays an important role in the metabolism of various compounds either endogenous or xenobiotics. Due to its promiscuity, hAOX1 plays a major role in the pharmacokinetics of many drugs and therefore has gathered a lot of attention from the scientific community and, particularly, from the pharmaceutical industry. In this work, homology modelling, molecular docking and molecular dynamics simulations were used to study the structure of the monomer and dimer of human AOX. The results with the monomer of hAOX1 allowed to shed some light on the role played by thioridazine and two malonate ions that are co-crystalized in the recent X-ray structure of hAOX1. The results show that these molecules endorse several conformational rearrangements in the binding pocket of the enzyme and these changes have an impact in the active site topology as well as in the stability of the substrate (phthalazine). The results show that the presence of both molecules open two gates located at the entrance of the binding pocket, from which results the flooding of the active site. They also endorse several modifications in the shape of the binding pocket (namely the position of Lys893) that, together with the presence of the solvent molecules, favour the release of the substrate to the solvent. Further insights were also obtained with the assembled homodimer of hAOX1. The allosteric inhibitor (THI) binds closely to the region where the dimerization of both monomers occur. These findings suggest that THI can interfere with protein dimerization.

Crystal structure of the zinc-, cobalt-, and iron-containing adenylate kinase from Desulfovibrio gigas: a novel metal-containing adenylate kinase from Gram-negative bacteria.

Adenylate kinases (AK) from Gram-negative bacteria are generally devoid of metal ions in their LID domain. However, three metal ions, zinc, cobalt, and iron, have been found in AK from Gram-negative bacteria. Crystal structures of substrate-free AK from Desulfovibrio gigas with three different metal ions (Zn(2+), Zn-AK; Co(2+), Co-AK; and Fe(2+), Fe-AK) bound in its LID domain have been determined by X-ray crystallography to resolutions 1.8, 2.0, and 3.0 Å, respectively. The zinc and iron forms of the enzyme were crystallized in space group I222, whereas the cobalt-form crystals were C2. The presence of the metals was confirmed by calculation of anomalous difference maps and by X-ray fluorescence scans. The work presented here is the first report of a structure of a metal-containing AK from a Gram-negative bacterium. The native enzyme was crystallized, and only zinc was detected in the LID domain. Co-AK and Fe-AK were obtained by overexpressing the protein in Escherichia coli. Zn-AK and Fe-AK crystallized as monomers in the asymmetric unit, whereas Co-AK crystallized as a dimer. Nevertheless, all three crystal structures are very similar to each other, with the same LID domain topology, the only change being the presence of the different metal atoms. In the absence of any substrate, the LID domain of all holoforms of AK was present in a fully open conformational state. Normal mode analysis was performed to predict fluctuations of the LID domain along the catalytic pathway.

Protein Crystallization in a Microfluidic Contactor with Nafion®117 Membranes.

Protein crystallization still remains mostly an empirical science, as the production of crystals with the required quality for X-ray analysis is dependent on the intensive screening of the best protein crystallization and crystal's derivatization conditions. Herein, this demanding step was addressed by the development of a high-throughput and low-budget microfluidic platform consisting of an ion exchange membrane (117 Nafion® membrane) sandwiched between a channel layer (stripping phase compartment) and a wells layer (feed phase compartment) forming 75 independent micro-contactors. This microfluidic device allows for a simultaneous and independent screening of multiple protein crystallization and crystal derivatization conditions, using Hen Egg White Lysozyme (HEWL) as the model protein and Hg2+ as the derivatizing agent. This microdevice offers well-regulated crystallization and subsequent crystal derivatization processes based on the controlled transport of water and ions provided by the 117 Nafion® membrane. Diffusion coefficients of water and the derivatizing agent (Hg2+) were evaluated, showing the positive influence of the protein drop volume on the number of crystals and crystal size. This microfluidic system allowed for crystals with good structural stability and high X-ray diffraction quality and, thus, it is regarded as an efficient tool that may contribute to the enhancement of the proteins' crystals structural resolution.

The effect of the sixth sulfur ligand in the catalytic mechanism of periplasmic nitrate reductase.

The catalytic mechanism of nitrate reduction by periplasmic nitrate reductases has been investigated using theoretical and computational means. We have found that the nitrate molecule binds to the active site with the Mo ion in the +6 oxidation state. Electron transfer to the active site occurs only in the proton-electron transfer stage, where the Mo(V) species plays an important role in catalysis. The presence of the sulfur atom in the molybdenum coordination sphere creates a pseudo-dithiolene ligand that protects it from any direct attack from the solvent. Upon the nitrate binding there is a conformational rearrangement of this ring that allows the direct contact of the nitrate with Mo(VI) ion. This rearrangement is stabilized by the conserved methionines Met141 and Met308. The reduction of nitrate into nitrite occurs in the second step of the mechanism where the two dimethyl-dithiolene ligands have a key role in spreading the excess of negative charge near the Mo atom to make it available for the chemical reaction. The reaction involves the oxidation of the sulfur atoms and not of the molybdenum as previously suggested. The mechanism involves a molybdenum and sulfur-based redox chemistry instead of the currently accepted redox chemistry based only on the Mo ion. The second part of the mechanism involves two protonation steps that are promoted by the presence of Mo(V) species. Mo(VI) intermediates might also be present in this stage depending on the availability of protons and electrons. Once the water molecule is generated only the Mo(VI) species allow water molecule dissociation, and, the concomitant enzymatic turnover.

Crystal structure of the xanthine oxidase-related aldehyde oxido-reductase from D. gigas.

The crystal structure of the aldehyde oxido-reductase (Mop) from the sulfate reducing anaerobic Gram-negative bacterium Desulfovibrio gigas has been determined at 2.25 A resolution by multiple isomorphous replacement and refined. The protein, a homodimer of 907 amino acid residues subunits, is a member of the xanthine oxidase family. The protein contains a molybdopterin cofactor (Mo-co) and two different [2Fe-2S] centers. It is folded into four domains of which the first two bind the iron sulfur centers and the last two are involved in Mo-co binding. Mo-co is a molybdenum molybdopterin cytosine dinucleotide. Molybdopterin forms a tricyclic system with the pterin bicycle annealed to a pyran ring. The molybdopterin dinucleotide is deeply buried in the protein. The cis-dithiolene group of the pyran ring binds the molybdenum, which is coordinated by three more (oxygen) ligands.

Cobalt-, zinc- and iron-bound forms of adenylate kinase (AK) from the sulfate-reducing bacterium Desulfovibrio gigas: purification, crystallization and preliminary X-ray diffraction analysis.

Adenylate kinase (AK; ATP:AMP phosphotransferase; EC 2.7.4.3) is involved in the reversible transfer of the terminal phosphate group from ATP to AMP. AKs contribute to the maintenance of a constant level of cellular adenine nucleotides, which is necessary for the energetic metabolism of the cell. Three metal ions, cobalt, zinc and iron(II), have been reported to be present in AKs from some Gram-negative bacteria. Native zinc-containing AK from Desulfovibrio gigas was purified to homogeneity and crystallized. The crystals diffracted to beyond 1.8 A resolution. Furthermore, cobalt- and iron-containing crystal forms of recombinant AK were also obtained and diffracted to 2.0 and 3.0 A resolution, respectively. Zn(2+)-AK and Fe(2+)-AK crystallized in space group I222 with similar unit-cell parameters, whereas Co(2+)-AK crystallized in space group C2; a monomer was present in the asymmetric unit for both the Zn(2+)-AK and Fe(2+)-AK forms and a dimer was present for the Co(2+)-AK form. The structures of the three metal-bound forms of AK will provide new insights into the role and selectivity of the metal in these enzymes.

Crystal structure of the first dissimilatory nitrate reductase at 1.9 A solved by MAD methods.

BACKGROUND: The periplasmic nitrate reductase (NAP) from the sulphate reducing bacterium Desulfovibrio desulfuricans ATCC 27774 is induced by growth on nitrate and catalyses the reduction of nitrate to nitrite for respiration. NAP is a molybdenum-containing enzyme with one bis-molybdopterin guanine dinucleotide (MGD) cofactor and one [4Fe-4S] cluster in a single polypeptide chain of 723 amino acid residues. To date, there is no crystal structure of a nitrate reductase. RESULTS: The first crystal structure of a dissimilatory (respiratory) nitrate reductase was determined at 1.9 A resolution by multiwavelength anomalous diffraction (MAD) methods. The structure is folded into four domains with an alpha/beta-type topology and all four domains are involved in cofactor binding. The [4Fe-4S] centre is located near the periphery of the molecule, whereas the MGD cofactor extends across the interior of the molecule interacting with residues from all four domains. The molybdenum atom is located at the bottom of a 15 A deep crevice, and is positioned 12 A from the [4Fe-4S] cluster. The structure of NAP reveals the details of the catalytic molybdenum site, which is coordinated to two MGD cofactors, Cys140, and a water/hydroxo ligand. A facile electron-transfer pathway through bonds connects the molybdenum and the [4Fe-4S] cluster. CONCLUSIONS: The polypeptide fold of NAP and the arrangement of the cofactors is related to that of Escherichia coli formate dehydrogenase (FDH) and distantly resembles dimethylsulphoxide reductase. The close structural homology of NAP and FDH shows how small changes in the vicinity of the molybdenum catalytic site are sufficient for the substrate specificity.

Structure and function of molybdopterin containing enzymes.

Molybdopterin containing enzymes are present in a wide range of living systems and have been known for several decades. However, only in the past two years have the first crystal structures been reported for this type of enzyme. This has represented a major breakthrough in this field. The enzymes share common structural features, but reveal different polypeptide folding topologies. In this review we give an account of the related spectroscopic information and the crystallographic results, with emphasis on structure-function studies.

Crystallographic and fluorescence studies of ligand binding to N-carbamoylsarcosine amidohydrolase from Arthrobacter sp.

Crystal structures of N-carbamoylsarcosine amidohydrolase (CSHase; EC 3.5.1.59) have been analyzed by X-ray diffraction methods with two different inhibitors bound to the active site at 2.28 A and 2.37 A resolution. The catalytic center of the enzyme could be identified on the basis of these structures. The four substrate binding sites are situated at the intersubunit interfaces of the compact dimers AB and CD of the homotetrameric enzyme. Both inhibitors inactivate the enzyme irreversibly through covalent binding of their aldehyde groups to the thiol group of the active-site cysteine residue Cys177. Within the identified substrate binding sites a number of residues from different subunits are involved in hydrogen bonding of the inhibitors. Two residues (Ala172 and Thr173) that form an unusual cis-peptide bond at the binding site are important components in fixing the examined inhibitors by hydrogen bonds. An electrochemical enzyme assay for CSHase was used to test the effect of inhibitors and substrate analogs on the enzyme's activity, revealing the high substrate specificity of CSHase. The intrinsic tryptophan fluorescence of CSHase increases strongly upon substrate and inhibitor binding. As most of the tryptophyl residues are located at the active sites, they are thus considerably affected by ligand binding. Fluorescence-detected stopped-flow measurements have been used to study the kinetics of glyoxylate and substrate binding to CSHase. Substrate and inhibitor binding could clearly be distinguished in the stopped-flow experiments. Inhibitor binding reveals at least three different elementary processes, whereas substrate binding is much faster and contains phases with different signs in amplitude.

Altered specificity mutations define residues essential for substrate positioning in xanthine dehydrogenase.

We describe the sequence changes of a number of mutations of the Aspergillus nidulans xanthine dehydrogenase (XDH). We have located the amino acids affected by these changes in the three-dimensional (3D) structure of aldehyde oxido-reductase (MOP) from Desulfovibrio gigas, related to eukaryotic XDHs. Of these, two are loss of function mutations, mapping, respectively, in the molybdenum-pterin co-factor (MoCo) domain and in the domain involved in substrate recognition. Changes in two amino acids result in resistance to the irreversible inhibitor allopurinol. In Arg911 two different changes, conserved among all XDHs and MOP but not in other aldehyde oxidases (AO), change the position of hydroxylation of the analogue 2-hydroxypurine from C-8 to C-6. A number of changes affect residues adjacent to the molybdenum or its ligands. Arg911 is positioned in the substrate pocket in a way that it can account for the positioning of purine substrates in relation to the MoCo reactive center, together with a glutamate residue, universally conserved among the XDHs (Glu833).

Crystal structure analysis, refinement and enzymatic reaction mechanism of N-carbamoylsarcosine amidohydrolase from Arthrobacter sp. at 2.0 A resolution.

N-carbamoylsarcosine amidohydrolase from Arthrobacter sp., a tetramer of polypeptides with 264 amino acid residues each, has been crystallized and its structure solved and refined at 2.0 A resolution, to a crystallographic R-factor of 18.6%. The crystals employed in the analysis contain one tetramer of 116,000 M(r) in the asymmetric unit. The structure determination proceeded by multiple isomorphous replacement, followed by solvent-flattening and density averaging about the local diads within the tetramer. In the final refined model, the root-mean-square deviation from ideality is 0.01 A for bond distances and 2.7 degrees for bond angles. The asymmetric unit consists of 7853 protein atoms, 431 water molecules and four sulfate ions bound into the putative active site clefts in each subunit. One subunit contains a central six-stranded parallel beta-pleated sheet packed by helices on both sides. On one side, two helices face the solvent, while two of the helices on the other side are buried in the tight intersubunit contacts. The catalytic center of the enzyme, tentatively identified by inhibitor binding, is located at the interface between two subunits and involves residues from both. It is suggested that the nucleophilic group involved in hydrolysis of the substrate is the thiol group of Cys117 and a nucleophilic addition-elimination mechanism is proposed.

The 2.4 A resolution crystal structure of boar seminal plasma PSP-I/PSP-II: a zona pellucida-binding glycoprotein heterodimer of the spermadhesin family built by a CUB domain architecture.

The crystal structure of porcine seminal plasma spermadhesin PSP-I/PSP-II heterodimer has been determined in two crystal forms by multiple isomorphous replacement in an hexagonal crystal (space group P6(1)22) and molecular replacement in a trigonal crystal of space group P3(2)21. The crystal structure has been refined at 2.4 A resolution to an R-factor of 20.0% (Rfree = 25.9%) for 14,809 independent reflections with intensities greater than 2 sigma (I), with root-mean-square deviations of 0.009 A and 1.657 degrees from ideal bond lengths and bond angles, respectively. The final model includes 1688 non-hydrogen protein atoms of 221 amino acids and 79 water molecules. PSP-I/PSP-II represents the first crystal structure of a mammalian zona pellucida-binding protein. PSP-II displays a putative carbohydrate-recognition site located around its Asn50. This region shares structural features with sugar binding sites of known lectin structures of the leguminous and galectin families. PSP-I and PSP-II are N-glycosylated at asparagine residues 50 and 98, respectively, and show site heterogeneity. Only the innermost N-acetylglucosamine of PSP-I is defined in the crystal structure. Both subunits of the PSP-I/PSP-II heterodimer are built by a single CUB domain architecture. The CUB domain displays a novel fold, which consists of a compact ellipsoidal beta-sandwich structure (42 A x 27 A x 23 A) organized into two 5-stranded beta-sheets. Each sheet contains parallel and antiparallel beta-strands. Two disulphide bridges, which are conserved in all spermadhesin molecules and many CUB domains, crosslink loop LA and strand beta 4 and loops LE and LG, respectively, at opposite edges of the same face of the domain. The four highly conserved aromatic residues and 15 out of 17 invariant hydrophobic residues, which define the CUB domain signature, display an interior location, suggesting that this hydrophobic core may be essential for maintaining the overall folding of the domain. Most of the hydrophobic core residue characteristics are conserved in the jellyroll topology of certain icosahedral virus capsid proteins, indicating that the CUB domain and the viral proteins share a minimal structural core.

Crystal structure of desulforedoxin from Desulfovibrio gigas determined at 1.8 A resolution: a novel non-heme iron protein structure.

The crystal structure of desulforedoxin from Desulfovibrio gigas, a new homo-dimeric (2 x 36 amino acids) non-heme iron protein, has been solved by the SIRAS method using the indium-substituted protein as the single derivative. The structure was refined to a crystallographic R-factor of 16.9% at 1.8 A resolution. Native desulforedoxin crystals were grown from either PEG 4K or lithium sulfate, with cell constants a = b = 42.18 A, c = 72.22 A (for crystals grown from PEG 4K), and they belong to space group P3(2)21. The indium-substituted protein crystallized isomorphously under the same conditions. The 2-fold symmetric dimer is firmly hydrogen bonded and folds as an incomplete beta-barrel with the two iron centers placed on opposite poles of the molecule. Each iron atom is coordinated to four cysteinyl residues in a distorted tetrahedral arrangement. Both iron atoms are 16 A apart but connected across the 2-fold axis by 14 covalent bonds along the polypeptide chain plus two hydrogen bonds. Desulforedoxin and rubredoxin share some structural features but show significant differences in terms of metal environment and water structure, which account for the known spectroscopic differences between rubredoxin and desulforedoxin.

Crystal structure of acidic seminal fluid protein (aSFP) at 1.9 A resolution: a bovine polypeptide of the spermadhesin family.

We report the three-dimensional crystal structure of acidic seminal fluid protein (aSFP), a 12.9 kDa polypeptide of the spermadhesin family isolated from bovine seminal plasma, solved by the multiple isomorphous replacement method and refined with data to 1.9 A resolution with a final R-factor of 17.3%. aSFP is built by a single CUB domain architecture, a 100 to 110 amino-acid-residue extracellular module found in 16 functionally diverse proteins. The structure of aSFP reveals that the CUB domain displays a beta-sandwich topology organised into two 5-stranded beta-sheets, each of which contain two parallel and four antiparallel strands. The structure of aSFP is almost identical to that of porcine spermadhesins PSP-I and PSP-II, which in turn show limited structural similarity with jellyroll topologies of certain virus capsid proteins. Essentially, topologically conserved residues in these proteins are those internal amino acids forming the hydrophobic core of the CUB and the jellyroll domains, suggesting their importance in maintaining the integrity of these protein folds. On the other hand, the structure of aSFP shows structural features that are unique to this protein and which may provide a structural ground for understanding the distinct biological properties of different members of the spermadhesin protein family.

Gene sequence and crystal structure of the aldehyde oxidoreductase from Desulfovibrio desulfuricans ATCC 27774.

The aldehyde oxidoreductase (MOD) isolated from the sulfate reducer Desulfovibrio desulfuricans (ATCC 27774) is a member of the xanthine oxidase family of molybdenum-containing enzymes. It has substrate specificity similar to that of the homologous enzyme from Desulfovibrio gigas (MOP) and the primary sequences from both enzymes show 68 % identity. The enzyme was crystallized in space group P6(1)22, with unit cell dimensions of a=b=156.4 A and c=177.1 A, and diffraction data were obtained to beyond 2.8 A. The crystal structure was solved by Patterson search techniques using the coordinates of the D. gigas enzyme. The overall fold of the D. desulfuricans enzyme is very similar to MOP and the few differences are mapped to exposed regions of the molecule. This is reflected in the electrostatic potential surfaces of both homologous enzymes, one exception being the surface potential in a region identifiable as the putative docking site of the physiological electron acceptor. Other essential features of the MOP structure, such as residues of the active-site cavity, are basically conserved in MOD. Two mutations are located in the pocket bearing a chain of catalytically relevant water molecules. As deduced from this work, both these enzymes are very closely related in terms of their sequences as well as 3D structures. The comparison allowed confirmation and establishment of features that are essential for their function; namely, conserved residues in the active-site, catalytically relevant water molecules and recognition of the physiological electron acceptor docking site.

Crystallization and preliminary X-ray diffraction studies of aSFP, a bovine seminal plasma protein with a single CUB domain architecture.

Bovine acidic seminal fluid protein (aSFP) is a 1.29 kDa polypeptide of the spermadhesin family built by a single CUB domain architecture. The CUB domain is an extracellular module present in 16 functionally diverse proteins. To determine the three-dimensional structure of aSFP, the protein was crystallized at 21 degrees C by vapor diffusion in hanging drops, using ammonium sulfate, pH 4.7, and polyethyleneglycol 4,000 as precipitants, containing 10% dioxane to avoid the formation of clustered crystals. Elongated prismatic crystals with maximal size of 0.6 x 0.3 x 0.2 mm3 diffract to beyond 1.9 A resolution and belong to space group P2(1)2(1)2(1), with cell parameters a = 52.4 A, b = 41.5 A, c = 48.2 A. There is one aSFP molecule per asymmetric unit, which corresponds to a crystal volume per unit molecular mass of 2.04 A3/Da, and analytical ultracentrifugation analysis show that aSFP is a monomeric protein.

Structural studies by X-ray diffraction on metal substituted desulforedoxin, a rubredoxin-type protein.

Desulforedoxin (Dx), isolated from the sulfate reducing bacterium Desulfovibrio gigas, is a small homodimeric (2 x 36 amino acids) protein. Each subunit contains a high-spin iron atom tetrahedrally bound to four cysteinyl sulfur atoms, a metal center similar to that found in rubredoxin (Rd) type proteins. The simplicity of the active center in Dx and the possibility of replacing the iron by other metals make this protein an attractive case for the crystallographic analysis of metal-substituted derivatives. This study extends the relevance of Dx to the bioinorganic chemistry field and is important to obtain model compounds that can mimic the four sulfur coordination of metals in biology. Metal replacement experiments were carried out by reconstituting the apoprotein with In3+, Ga3+, Cd2+, Hg2+, and Ni2+ salts. The In3+ and Ga3+ derivatives are isomorphous with the iron native protein; whereas Cd2+, Hg2+, and Ni2+ substituted Dx crystallized under different experimental conditions, yielding two additional crystal morphologies; their structures were determined by the molecular replacement method. A comparison of the three-dimensional structures for all metal derivatives shows that the overall secondary and tertiary structures are maintained, while some differences in metal coordination geometry occur, namely, bond lengths and angles of the metal with the sulfur ligands. These data are discussed in terms of the entatic state theory.

New insights into the chemistry of fac-[Ru(CO)3]²âº fragments in biologically relevant conditions: the CO releasing activity of [Ru(CO)3Cl2(1,3-thiazole)], and the X-ray crystal structure of its adduct with lysozyme.

Complexes of the general formula fac-[Ru(CO)(3)L(3)](2+), namely CORM-2 and CORM-3, have been successfully used as experimental CO releasing molecules (CO-RMs) but their mechanism of action and delivery of CO remain unclear. The well characterized complex [Ru(CO)(3)Cl(2)(1,3-thiazole)] (1) is now studied as a potential model CO-RM of the same family of complexes using LC-MS, FTIR, and UV-vis spectroscopy, together with X-ray crystallography. The chemistry of [Ru(CO)(3)Cl(2)(1,3-thiazole)] is very similar to that of CORM-3: it only releases residual amounts of CO to the headspace of a solution in PBS7.4 and produces marginal increase of COHb after long incubation in whole blood. 1 also reacts with lysozyme to form Ru adducts. The crystallographic model of the lysozyme-Ru adducts shows only mono-carbonyl Ru species. [Ru(H(2)O)(4)(CO)] is found covalently bound to a histidine (His15) and to two aspartates (Asp18 and Asp119) at the protein surface. The CO release silence of both 1 and CORM-3 and their rapid formation of protein-Ru(CO)(x)(H(2)O)(y) (x=1,2) adducts, support our hypothesis that fac-[Ru(CO)(3)L(3)] CO-RMs deliver CO in vivo through the decay of their adducts with plasma proteins.

Towards improved therapeutic CORMs: understanding the reactivity of CORM-3 with proteins.

The biological role of carbon monoxide (CO) has completely changed in the last decade. Beyond its widely feared toxicity, CO has revealed a very important biological activity as a signaling molecule with marked protective actions namely against inflammation, apoptosis and endothelial oxidative damage. Its direct use as a therapeutic gas showed significant and consistent positive results but also intrinsic severe limitations. The possibility of replacing the gas by pro-drugs acting as CO-Releasing Molecules (CO-RMs) has clearly been demonstrated with several experimental compounds. Transition metal carbonyls complexes have proven to be the most versatile experimental CO-RMs so far. Presently, the challenge is to equip them with drug-like properties to turn them into useful pharmaceuticals. This requires studying their interactions with biological molecules namely those that control their pharmacokinetic and ADME profiles like the plasma proteins. In this account we analyze these questions and review the existing interactions between Metal Carbonyls and proteins. The recently explored case of CORM-3 is revisited to exemplify the methodologies involved and the importance of the results for the understanding of the mode of action of such pro-drugs.

Real-time monitoring of molecular dynamics of ethylene glycol dimethacrylate glass former.

The isothermal cold-crystallization of the glass-former low-molecular-weight compound, ethylene glycol dimethacrylate (EGDMA), was monitored by real-time dielectric relaxation spectroscopy (DRS) and differential scanning calorimetry (DSC). The alpha-relaxation associated with the dynamic glass transition as detected by DRS was followed at different crystallization temperatures, T(cr), nearly above the glass transition temperature, 176 K (1.06 < or = T(cr)/T(g) < or = 1.12). It was found that the alpha-process depletes upon cold-crystallization with no significant changes in either shape or location. At advanced crystallization states, a new relaxation, alpha'-process, evolves that was assigned to the mobility of molecules lying adjacent to crystalline surfaces. From the time evolution of the normalized permittivity, it was possible to get kinetic information that was complemented with the calorimetric data. From DSC measurements that were also carried out under melt-crystallization, an enlarged temperature range was covered (up to T(cr)/T(g) = 1.24), allowing us to draw a diagram of time-temperature crystallization for this system. Dielectric relaxation spectroscopy proved to be a sensitive tool to probe the mobility in the remaining amorphous regions even at high crystallinities.