Biochemical, biophysical and molecular dynamics studies on the proteoglycan-like domain of carbonic anhydrase IX
- 744 Downloads
Human carbonic anhydrase IX (hCA IX) is a tumour-associated enzyme present in a limited number of normal tissues, but overexpressed in several malignant human tumours. It is a transmembrane protein, where the extracellular region consists of a greatly investigated catalytic CA domain and a much less investigated proteoglycan-like (PG) domain. Considering its important role in tumour biology, here, we report for the first time the full characterization of the PG domain, providing insights into its structural and functional features. In particular, this domain has been produced at high yields in bacterial cells and characterized by means of biochemical, biophysical and molecular dynamics studies. Results show that it belongs to the family of intrinsically disordered proteins, being globally unfolded with only some local residual polyproline II secondary structure. The observed conformational flexibility may have several important roles in tumour progression, facilitating interactions of hCA IX with partner proteins assisting tumour spreading and progression.
KeywordsTumour Polyproline II (PPII) Intrinsically disordered protein (IDP) Natively unfolded Cancer Hypoxia Metalloenzyme
Carbonic anhydrases (CAs) are ubiquitous metallo-enzymes catalysing the reversible hydration of CO2 to HCO3− and H+ . In humans, among the 12 catalytically active isoforms (CA I–IV, VA–VB, VI–VII, IX, and XII–XIV), CA IX has been recognized as a tumour-associated protein [2, 3, 4, 5]. In fact, apart its expression in a limited number of normal tissues with an almost total exclusivity in the gastrointestinal tract epithelium [6, 7], CA IX is overexpressed in the cell membrane of several malignant tumour cells, where it is generally associated with the hypoxic phenotype mediated by the transcription factor HIF-1 . In tumours, CA IX modulates growth, survival, proliferation, adhesion, and invasion of malignant cells  by means of several mechanisms, such as tumour pH regulation, interference with the Rho/ROCK signaling pathway  and interaction of the enzyme with β-catenin, which causes destabilization of intercellular adhesion [3, 5, 10, 11]. Among these mechanisms, the most investigated one regards the pH regulation of cancer cells here summarized. Upon hypoxia, HIF-1 transcription factor activates several specific genes which lead to up-regulation of glycolysis and, therefore, to an over-production of lactate and protons. To maintain a normal intracellular pH (pHi) [7, 12], these ions are extruded by means of monocarboxylate transporters (MCTs), pumps such as the V-type H+ ATPase (V-ATPase) and H+ exchangers as the Na+/H+ exchanger (NHE). Alternatively, the formed H+ ions are titrated by HCO3−, which enters the cell through HCO3− transporters, as Na+/bicarbonate cotransporters (NBCs) and anion exchangers (AEs) [4, 12]. In this case, the newly formed CO2 spreads out through the cell membrane. CA IX catalytic domain, expressed on the extracellular membrane of the cell, subtracts the newly spread CO2 transforming it into protons and bicarbonate ions. As a whole, this process allows the maintenance of a physiologic pHi crucial for the proliferation and survival of cancer cells and an acidification of the extracellular pH (pHe 6.9–7.0), which affects cancer progression by promoting invasion and metastasis .
Recent studies opened a completely new scenario on this enzyme, demonstrating that it can undergo nuclear translocation through the interaction with proteins involved in nucleocytoplasmic traffic . Furthermore, it has also been shown that it can interact with cullin-associated NEDD8 dissociated protein 1 (CAND1), a protein involved in gene transcription and assembly of SCF ubiquitin ligase complexes. Notably, lower CA IX levels were observed in cells where CAND1 expression is downregulated via shRNA-mediated interference, suggesting that CAND1/CA IX interaction could be required for the enzyme stabilization [14, 15].
The presence of the PG domain makes hCA IX one of the most active enzymes among hCAs [17, 18]. Indeed, kinetic measurements showed that the catalytic activity of the entire extracellular domain was greater than that of the catalytic domain alone (kcat/KM = 1.5 × 108 vs 5.4 × 107 M−1 s−1, respectively) . The PG domain was also reported to influence the optimal working pH of the enzyme; indeed, whereas the CA domain alone had an optimal activity at pH 7.0, the entire extracellular domain presented an optimal activity in acidic environment at pH 6.5 [18, 19]. It is worth noting that the slightly acidic pH value of 6.5 is within the typical pH range of solid and hypoxic tumours, where CA IX is generally overexpressed. Thus, it was suggested that the PG domain could be an evolutionarily evolved feature, unique to CA IX, which contributes to the improvement of its catalytic activity at the slightly acidic pH values [3, 4, 5].
Due to its role in tumour biology, hCA IX has become an interesting target for the drug design of new diagnostic and therapeutic tools in cancer treatment. Therefore, many studies have been dedicated to the elucidation of its biochemical and structural features. In particular, biochemical studies showed that the enzyme has both an intramolecular (C156–C336) and a symmetric intermolecular (through C174) disulphide bond, with the latter making the protein a dimer on the cell surface [17, 18]. Moreover, two glycosylation sites were identified: an O-linked glycosylation in the region immediately flanking the PG domain (T115), and an N-linked glycosylation localized on the catalytic domain (N346) . Finally, three phosphorylation sites, namely, T443, S448, and Y449, were recognized on the IC tail [20, 21].
Notably, structural information is only available for the catalytic domain and for the C-terminal part of the protein (residues 418–459). In particular, the catalytic domain was crystallized by our group in 2009, showing the typical α-CA fold with a unique dimeric arrangement , whereas information on the secondary structure of the C-terminus has been recently obtained, indicating a predominant helical content for this region . The absence of structural data on the full-length protein or on the PG domain is quite surprising, considering its important role in tumour biology, mediating cell adhesion and intercellular communications [22, 23] in addition to assisting catalysis mediated by the CA domain. Indeed, all attempts to obtain crystallographic structure of the PG domain failed, due to its high propensity to undergo protease degradation . To fill this gap, we hereby report the first detailed investigation on the N-terminal part of the hCA IX protein (residues 38–136), hereafter referred as PG(38–136) (Scheme 1), by means of a multidisciplinary approach including biochemical, biophysical and molecular dynamics (MD) studies.
Materials and methods
Expression host strain E. coli BL21(DE3) and engineered plasmid pET28a/SUMO were a kind gift from EMBL, Heidelberg. E. coli strain TOP10F’ was obtained from Invitrogen (San Diego, CA, USA). QIAprep spin miniprep kit and PCR Clean-Up DNA Purification System were from Qiagen (Germantown, MD, USA). Enzymes and other reagents for DNA manipulation were from New England Biolabs (Ipswich, MA, USA). All other chemicals were from Sigma-Aldrich (Milano, Italy).
The primary sequence of the PG(38–136) protein was analyzed using the program Composition Profiler (http://www.cprofiler.org/) . The query sample, analyzed for its intrinsic disorder, was compared with a reference sample which is a standard amino acid dataset (Swissprot) . In the graphical output, the less abundant amino acids have negative values, whereas those more abundant have positive values.
The Charge/Hydrophobicity (CH) relation for PG(38–136) was obtained as described by Uversky . The CH plot is divided into two regions by a line, which corresponds to the equation H = (|R| + 1.15)/2.782, where R is the mean net charge and H is the mean hydrophobicity . Proteins that fall in the left part of the diagram where H < (R + 1.151)/2.785 are predicted as disordered, whereas they are predicted as ordered if they fall in the right part. Data regarding the intrinsically disordered proteins were partially taken from Uversky et al. , whereas those regarding natively folded proteins were randomly taken from PDB.
Cloning, expression and purification of PG(38–136)
pET28a/SUMO vector containing SenP2 protease recognition site was chosen for E. Coli expression of the PG(38–136). Briefly, pg cDNA was amplified and cloned in the Age I and XhoI site of pET28a/SUMO using the following site-specific primers:
The plasmid generated was verified by appropriate digestion with restriction enzymes and sequencing. The recombinant construct was expressed in E. coli BL21(DE3) cells for 16 h at 22 °C with 0.1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG). After centrifugation, the supernatant was resuspended in lysis buffer (20 mM Tris–HCl, 20 mM imidazole, 500 mM NaCl, pH 8.0), in the presence of 1 mM phenylmethanesulfonyl fluoride, 5 mg/ml DNaseI, 0.1 mg/ml lysozyme and 1 µg/ml Aprotinin, 1 µg/ml Leupeptin, 1 µg/ml Pepstatin protease inhibitors and left for 30 min at room temperature before sonication. After centrifugation, the supernatant was loaded onto a nickel-immobilized affinity chromatography column (5 ml His Trap FF column, GE Healthcare) and purified by FPLC according to manufacturer’s instruction (GE Healthcare). Fractions containing PG(38–136) were pooled and dialysed in 20 mM Tris–HCl, 250 mM NaCl, pH 8.0, with a membrane cutoff (MWCO) of 3.500. Tag removal was performed by digesting PG(38–136) sample with protease enzyme SenP2 in a ratio SenP2/PG 1:25 (w/w) for 3 h at 20 °C and loading the mixture on an affinity HisTrap column according to manufacturer’s instruction (GE Healthcare). Purity level was assessed by 15% SDS-PAGE and LC–ESI–MS.
Quaternary structure investigations of PG(38–136)
The quaternary structure of PG(38–136) was investigated using SEC–MALS−QELS (Size Exclusion Chromatography–Multi-angle Light-Scattering–Quasi-Elastic Light Scattering) as previously reported [28, 29]. In particular, 50 µl of 1.5 mg/ml protein was loaded onto a Wyatt technology corporation column (WTC 015S5), equilibrated in PBS 1× (10 mM phosphate, 2.7 mM KCl, 137 mM NaCl, pH 7.4) and connected to FPLC ÅKTA, coupled to a light-scattering detector (mini-DAWN TREOS, Wyatt Technology) and a refractive index detector (Shodex RI-101). Data were analyzed using the program ASTRA 188.8.131.52 (Wyatt Technology Corporation).
Dynamic light-scattering (DLS) measurements were carried out using a Malvern nanozetasizer (Malvern, UK) following a procedure previously described . 37 µM PG(38–136) in 20 mM Tris–HCl, pH 7.5 was placed in a disposable cuvette and held at 25 °C during analysis. Spectra were recorded six times with 11 sub-runs using the multimodal mode. Only monodisperse peaks (% polydispersity lower than 20%) were considered. The Z-average diameter of the monodisperse peak was calculated from the correlation function using the Malvern technology software.
Effect of PG(38–136) on CA IX catalytic activity
Measurements of the catalytic activity of CA IX were performed by stopped-flow spectrophotometric measurements (Applied Photophysics Model SX.18MV). The solutions containing CO2 in a concentration range between 1.7 and 17 mM were obtained by bubbling CO2 in water. The maximum absorbance was observed at 557 nm, using 0.2 mM red phenol as pH indicator. CA domain was used at a concentration of 1 × 10−7 M in 10 mM Hepes, 10 mM Tris–HCl and 100 mM Na2SO4, pH 7.5. PG(38–136) was added at different concentrations ranging from 1 × 10−9 to 1 × 10−6 M with an incubation time of 10 min before reading the absorbance.
Circular dichroism (CD)
CD measurements were performed on a Jasco J815 spectropolarimeter (Jasco, Essex, UK), equipped with a temperature control system, using a 1-mm quartz cell in the far UV range 190–260 nm (20 nm/min scan speed). Each spectrum was the average of three scans with the background of the buffer solution subtracted. Measurements were performed at 20 °C at a protein concentration of 14 μM in buffers such as 10 mM Tris–HCl pH 8.0 or 10 mM phosphate buffer, pH 7.5. The effect of temperature on the secondary structure content of PG(38–136) was investigated using a 18 μM protein in 10 mM Tris–HCl buffer at pH 8.0. CD spectra ranging from 5 to 90 °C were taken every 5 °C, keeping temperature, set manually, within ± 0.1 °C by a peltier device. For the pH titration, PG domain was diluted in 10 mM Citrate phosphate buffer at different pHs (pH 2.6, 3.0, 4.0, 5.0, 6.0, 7.0, and 7.6) and the resulting curves were obtained keeping a fixed temperature of 5 °C. The effect of urea on the secondary structure of the protein was evaluated recording the spectra at 20 °C with different concentrations of the denaturant (0, 2, 4, 6, and 8 M) in 10 mM sodium phosphate pH 7.4. The wavelength range was set from 250 to 215 nm due to the absorbance properties of urea which prevented the acquisition of spectra below 215 nm . For all the CD experiments, raw spectra were corrected for buffer contribution and converted to mean molar ellipticity per residue (Ɵ) (deg cm2 dmol−1) .
NMR spectra were recorded at 303 K on a Varian Unity Inova 600 MHz spectrometer provided with a cold probe. To prepare the NMR sample, PG(38–136) was dissolved in 600 µl (concentration equal to 0.7 mg/ml) of a buffer containing 20 mM sodium phosphate pH 6.6, 100 mM NaCl, 40 µl D2O (98% D, Armar Chemicals, Dottingen, Switzerland) and 0.01% sodium azide. The following NMR experiments were collected: 1D [1H], 2D [1H, 1H] TOCSY  (70 ms mixing time), 2D [1H, 1H] NOESY  (300 ms mixing time). The 1D [1H] spectrum was acquired with a relaxation delay d1 of 1.5 s and 128 scans; 2D experiments were acquired with 32 scans, 128–256 FIDs in t1, 1024 or 2048 data points in t2. Water suppression was achieved through Excitation Sculpting . Chemical shifts were referenced to the water signal (4.75 ppm). The software VNMRJ (Varian by Agilent Technologies, Italy) was used for spectra processing, whereas NEASY , that is comprised in Computer Aided Resonance Assignment (CARA) package (http://cara.nmr.ch/doku.php), was implemented for spectra analysis.
Protease sensitivity of PG(38–136)
Protease sensitivity of PG(38–136) was evaluated by incubating the protein with TPCK-treated trypsin (Sigma-Aldrich, Milan) protease at different ratios such as 1:100 and 1:200 (w/w) at 26 °C. The reaction was monitored by 15% SDS-PAGE after incubation for 1, 3, 6, and 16 h with the proteolytic enzyme. hCA II as standard was incubated with trypsin in the same conditions.
Modelling and molecular dynamics studies
Among the five good quality models built by I-TASSER, the fifth one (C-score = − 2.20) was chosen for subsequent studies, being the only one compatible with the X-ray dimeric structure of CA IX. Indeed, in the other models, PG(38–136) partially occupied the dimeric interface. The quality of the selected I-TASSER model was further assessed by means of PROSA [37, 38] and PROCHECK  software. According to these analyses the I-TASSER model shows 73% of residues in the allowed regions of the Ramachandran plot and an energetic Z-score of − 7.42 indicating the good quality of the model. Subsequently, the hCA IX (38–391) dimeric model was built by superimposing two identical monomeric models obtained by I-TASSER to the crystallographic dimer of the catalytic domain. The final dimeric model was energy minimized by 1000 steps of Conjugate Gradient using Discover module of InsightII package.
The obtained dimeric model was subjected to all-atom MD simulations using the GROMACS simulation package . CHARMM22* force field  was used for simulations, since it was proven to be accurate for the simulation of IDPs, producing conformational ensemble consistent with experimental data [42, 43]. The model was solvated in a dodecahedral box filled with TIP3P water molecules with at least 12 Å distance to the border adding counterions to neutralize the system (reaching a concentration of 0.1 M). The simulations were run under NPT conditions (300 K and P = 1 bar) using the V-rescale thermostat  and Berendsen barostat, respectively. Periodic boundary conditions were employed and the LINCS algorithm  was used to constrain bond lengths. The particle mesh Ewald method was applied to treat electrostatic interactions  and a non-bonded cutoff of 1.4 nm was used for the Lennard–Jones potential. Water molecules were relaxed by energy minimization, followed by 50 ps of simulations at 300 K, restraining the protein atomic positions with a harmonic potential. Then, the system was heated up gradually to 300 K and equilibrated as described elsewhere . After equilibration, the system was simulated in NPT standard conditions for 100 ns using positional restraints for backbone atoms of the core-structured part of the CA domains, whereas the rest of the system was free to move. The analysis of the MD trajectory was carried out using GROMACS tools as well as MOLMOL  and DSSP  program. PROSS server was used for assignment of Polyproline II conformation .
Results and discussion
The significant presence of Pro and Glu within the PG(38–136) sequence, as well as its putative belonging to the IDP family, prompted us to investigate whether this domain contained PEST motifs. These sequences, enriched in Pro (P), Glu (E), Ser (S) and Thr (T), frequently located within unfolded protein regions [56, 59], serve as specific degradation signals [56, 59]; therefore, they play an important role in rapid turnover of regulatory proteins involved in signaling pathways that control cell growth, differentiation, stress responses, and physiological cell death [59, 60]. Using the Pestfindalgorithm , a PEST sequence (residues 43–72) was identified with a very high score of + 18.55. Studies will be carried out in our lab to explore the exact role of the PEST sequence in PG domain and how it might affect CA IX stability.
Expression and biochemical characterization of PG(38–136)
Further structural insights into PG(38–136) were obtained by means of 1D [1H] and 2D [1H, 1H] NMR spectroscopy . The 1D [1H] NMR spectrum (Fig. S3a) together with the 2D [1H, 1H] TOCSY (Total Correlation Spectroscopy)  and 2D [1H, 1H] NOESY (Nuclear Overhauser Enhancement Spectroscopy)  experiments (Fig. S3b) appear typical of IDPs. In particular, the 1D [1H] (Fig. S3a) and 2D [1H, 1H] TOCSY (Fig. S3b left panel) spectra present low chemical shift dispersion with the backbone amide HN protons resonating in the narrow random coil range between 8 and 8.6 ppm. Moreover, methyl protons from side chains of Leu and Val residues give rise to a strong peak at the random coil chemical shift (i.e., 0.9 ppm) (Fig. S3a), thus highlighting the absence of the hydrophobic core of a folded protein. The NOESY spectrum (Fig. S3c) contains a few very weak inter-residue HN–HN contacts pointing out the presence of a rather small population of more ordered conformations. Thus, in agreement with the above-described results from other biophysical techniques, NMR spectroscopy further shows the largely disordered nature of PG(38–136).
Due to the lack of a hydrophobic packed core and to the wide solvent accessibility, IDPs are prone to be easily degraded in the presence of a protease with broad substrate specificity such as trypsin . This is one of the main differences compared to structured proteins with well-defined secondary structure elements, which are preferentially cleaved at exposed and flexible loops . The incubation of PG(38–136) with trypsin protease at different ratios showed a complete cleavage in the early hours of the reaction (Fig. S4a). The same experiment performed on hCA II, a structured globular protein, showed a strong resistance to proteolysis even after 16 h of incubation (Fig. S4b). These data are in agreement with those obtained by CD, SEC, and LS confirming that PG(38–136) is a largely disordered and flexible protein.
Molecular modelling and molecular dynamics simulations
Root mean square fluctuations (RMSF) of Cα atom positions were evaluated during simulation time, showing high values for residues 38–136 in both monomeric units, indicative of the high flexibility of this region (Fig. 4b). Interestingly, the RMSF curves of the two monomers in this region are diverse, since the two PG(38–136) behave differently due to their inherent conformational plasticity.
Within each monomer, PG(38–136) conformations are stabilized by polar interactions with the aqueous solvent, as well as by intra- and inter-domain interactions (between PG(38–136) and CA), mainly through the formation of salt-bridges and hydrogen bonds (Table S1). Interestingly, couples of residues involved into hydrogen bonds are different in the two monomers, further indicating the flexibility of PG(38–136). Indeed, this region possesses many polar and charged residues within its six repeats, which can be alternately involved into stabilizing interactions according to the adopted conformation.
Finally, since far-UV-CD analysis indicated that PG(38–136) possesses some content of PPII conformation, PPII occurrence along MD trajectory was investigated. To this aim, PROSS server was employed, since differently from most commonly used secondary structure assignment methods, it can assign PPII structures. For PROSS analysis, 20 structures of the two most populated clusters for each monomer were selected and obtained data were reported in terms of frequency of occurrence of PPII structure vs residue number. The results indicate the presence of at least four regions having a significant preference for PPII conformation (frequency > 50%) (Fig. S7). The four short regions (3–5 residues in length) are wide-spread along the PG(38–136) sequence and roughly correspond to regions 58–60, 75–77, 98–100, and 119–121 (Fig. S7). As a consequence, the computed data are consistent with far-UV-CD analysis showing the presence of a residual PPII structure in PG(38–136).
Despite the great amount of studies on tumour-associated protein hCA IX, to date, very little information concerning the biochemical and structural features of its N-terminal region containing the PG domain is available. By means of a multidisciplinary approach, we hereby report for the first time a comprehensive study on PG(38–136), showing that it belongs to the family of IDPs, being natively highly flexible and mainly unfolded with only local tendencies to assume PPII conformations. Furthermore, the obtained data indicate that N-terminal residues (38–87) show a more extended conformation, being probably involved into partner recognition, whereas C-terminal residues (88–136) adopt a slightly more compact conformation and could have a role in modulating the catalytic activity of the CA domain. These results further extend our previous studies on the structural features of CA IX protein and provides new pieces in the complicated puzzle of CA IX functions in tumour biology.
We thank Immacolata Ventotto for her help during protein purification; we are grateful to Luca De Luca, Maurizio Amendola and Giosuè Sorrentino for their technical assistance. This work was supported by a Grant from CNR-DSB ProgettoBandiera “InterOmics”.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflicts of interest with the contents of this article.
- 6.Ilardi G, Zambrano N, Merolla F, Siano M, Varricchio S, Vecchione M, De Rosa G, Mascolo M, Staibano S (2014) Histopathological determinants of tumor resistance: a special look to the immunohistochemical expression of carbonic anhydrase IX in human cancers. Curr Med Chem 21:1569–1582CrossRefPubMedPubMedCentralGoogle Scholar
- 13.Estrella V, Chen T, Lloyd M, Wojtkowiak J, Cornnell HH, Ibrahim-Hashim A, Bailey K, Balagurunathan Y, Rothberg JM, Sloane BF, Johnson J, Gatenby RA, Gillies RJ (2013) Acidity generated by the tumor microenvironment drives local invasion. Cancer Res 73:1524–1535CrossRefPubMedPubMedCentralGoogle Scholar
- 14.Buanne P, Renzone G, Monteleone F, Vitale M, Monti SM, Sandomenico A, Garbi C, Montanaro D, Accardo M, Troncone G, Zatovicova M, Csaderova L, Supuran CT, Pastorekova S, Scaloni A, De Simone G, Zambrano N (2013) Characterization of carbonic anhydrase IX interactome reveals proteins assisting its nuclear localization in hypoxic cells. J Proteome Res 12:282–292CrossRefPubMedGoogle Scholar
- 15.Buonanno M, Langella E, Zambrano N, Succoio M, Sasso E, Alterio V, Di Fiore A, Sandomenico A, Supuran CT, Scaloni A, Monti SM, De Simone G (2017) Disclosing the interaction of carbonic anhydrase IX with Cullin-associated NEDD8-dissociated protein 1 by molecular modeling and integrated binding measurements. ACS Chem Biol 12:1460–1465CrossRefPubMedGoogle Scholar
- 17.Hilvo M, Baranauskiene L, Salzano AM, Scaloni A, Matulis D, Innocenti A, Scozzafava A, Monti SM, Di Fiore A, De Simone G, Lindfors M, Janis J, Valjakka J, Pastorekova S, Pastorek J, Kulomaa MS, Nordlund HR, Supuran CT, Parkkila S (2008) Biochemical characterization of CA IX, one of the most active carbonic anhydrase isozymes. J Biol Chem 283:27799–27809CrossRefPubMedGoogle Scholar
- 18.Alterio V, Hilvo M, Di Fiore A, Supuran CT, Pan P, Parkkila S, Scaloni A, Pastorek J, Pastorekova S, Pedone C, Scozzafava A, Monti SM, De Simone G (2009) Crystal structure of the catalytic domain of the tumor-associated human carbonic anhydrase IX. Proc Natl Acad Sci USA 106:16233–16238CrossRefPubMedGoogle Scholar
- 19.Innocenti A, Pastorekova S, Pastorek J, Scozzafava A, De Simone G, Supuran CT (2009) The proteoglycan region of the tumor-associated carbonic anhydrase isoform IX acts as anintrinsic buffer optimizing CO2 hydration at acidic pH values characteristic of solid tumors. Bioorg Med Chem Lett 19:5825–5828CrossRefPubMedGoogle Scholar
- 20.Ditte P, Dequiedt F, Svastova E, Hulikova A, Ohradanova-Repic A, Zatovicova M, Csaderova L, Kopacek J, Supuran CT, Pastorekova S, Pastorek J (2011) Phosphorylation of carbonic anhydrase IX controls its ability to mediate extracellular acidification in hypoxic tumors. Cancer Res 71:7558–7567CrossRefPubMedGoogle Scholar
- 28.D’Ambrosio K, Lopez M, Dathan NA, Ouahrani-Bettache S, Kohler S, Ascione G, Monti SM, Winum JY, De Simone G (2014) Structural basis for the rational design of new anti-Brucella agents: the crystal structure of the C366S mutant of l-histidinol dehydrogenase from Brucella suis. Biochimie 97:114–120CrossRefPubMedGoogle Scholar
- 30.Di Lelio I, Caccia S, Coppola M, Buonanno M, Di Prisco G, Varricchio P, Franzetti E, Corrado G, Monti SM, Rao R, Casartelli M, Pennacchio F (2014) A virulence factor encoded by a polydnavirus confers tolerance to transgenic tobacco plants against lepidopteran larvae, by impairing nutrient absorption. PLoS One 9:e113988CrossRefPubMedPubMedCentralGoogle Scholar
- 34.Hwang T-L, Shaka AJ (1995) Water suppression that works. Excitation sculpting using arbitrary waveforms and pulsed field gradients. J Magn Reson A112:5Google Scholar
- 48.Koradi R, Billeter M, Wuthrich K (1996) MOLMOL: a program for display and analysis of macromolecular structures. J Mol Graph 14(51–55):29–32Google Scholar
- 53.Dunker AK, Lawson JD, Brown CJ, Williams RM, Romero P, Oh JS, Oldfield CJ, Campen AM, Ratliff CM, Hipps KW, Ausio J, Nissen MS, Reeves R, Kang C, Kissinger CR, Bailey RW, Griswold MD, Chiu W, Garner EC, Obradovic Z (2001) Intrinsically disordered protein. J Mol Graph Model 19:26–59CrossRefPubMedGoogle Scholar
- 67.Delak K, Harcup C, Lakshminarayanan R, Sun Z, Fan Y, Moradian-Oldak J, Evans JS (2009) The tooth enamel protein, porcine amelogenin, is an intrinsically disordered protein with an extended molecular configuration in the monomeric form. Biochemistry 48:2272–2281CrossRefPubMedPubMedCentralGoogle Scholar
- 69.Hotta K, Ranganathan S, Liu R, Wu F, Machiyama H, Gao R, Hirata H, Soni N, Ohe T, Hogue CW, Madhusudhan MS, Sawada Y (2014) Biophysical properties of intrinsically disordered p130Cas substrate domain—implication in mechanosensing. PLoS Comput Biol 10:e1003532CrossRefPubMedPubMedCentralGoogle Scholar