Leads for Development of New Naphthalenesulfonate Derivatives with Enhanced Antiangiogenic Activity
2003; Elsevier BV; Volume: 278; Issue: 24 Linguagem: Inglês
10.1074/jbc.m212833200
ISSN1083-351X
AutoresCarlos Fernández‐Tornero, Rosa M. Lozano, Mariano Redondo‐Horcajo, Ana M. Gómez, J. Cristóbal López, Ernesto Quesada, Clara Uriel, Serafı́n Valverde, Pedro Cuevas, Antonio Romero, Guillermo Giménez‐Gallego,
Tópico(s)Fibroblast Growth Factor Research
ResumoInhibition of angiogenesis-promoting factors such as fibroblast growth factors is considered to be a potential procedure for inhibiting solid tumor growth. Although several peptide-based inhibitors are currently under study, the development of antiangiogenic compounds of small molecular size is a pharmacological goal of considerable interest. We have already shown that certain naphthalene sulfonates constitute minimal functional substitutes of the antiangiogenic compounds of the suramin and suradista family. Using those data as a lead, we have carried out a rational search for new angiogenesis inhibitors that could provide new pharmacological insights for the development of antiangiogenic treatments. The results of the study strongly underline the relevance of the stereochemistry for an efficient inhibition of acidic fibroblast growth factor mitogenic activity by the naphthalene sulfonate family and allow us to formulate rules to aid in searching for new inhibitors and pharmaceutical developments. To provide further leads for such developments and acquire a detailed insight into the basis of the inhibitory activity of the naphthalene sulfonate derivatives, we solved the three-dimensional structure of acidic fibroblast growth factor complexed to 5-amino-2-naphthalenesulfonate, the most pharmacologically promising of the identified inhibitors. The structure shows that binding of this compound would hamper the interaction of acidic fibroblast growth factor with the different components of the cell membrane mitogenesis-triggering complex. Inhibition of angiogenesis-promoting factors such as fibroblast growth factors is considered to be a potential procedure for inhibiting solid tumor growth. Although several peptide-based inhibitors are currently under study, the development of antiangiogenic compounds of small molecular size is a pharmacological goal of considerable interest. We have already shown that certain naphthalene sulfonates constitute minimal functional substitutes of the antiangiogenic compounds of the suramin and suradista family. Using those data as a lead, we have carried out a rational search for new angiogenesis inhibitors that could provide new pharmacological insights for the development of antiangiogenic treatments. The results of the study strongly underline the relevance of the stereochemistry for an efficient inhibition of acidic fibroblast growth factor mitogenic activity by the naphthalene sulfonate family and allow us to formulate rules to aid in searching for new inhibitors and pharmaceutical developments. To provide further leads for such developments and acquire a detailed insight into the basis of the inhibitory activity of the naphthalene sulfonate derivatives, we solved the three-dimensional structure of acidic fibroblast growth factor complexed to 5-amino-2-naphthalenesulfonate, the most pharmacologically promising of the identified inhibitors. The structure shows that binding of this compound would hamper the interaction of acidic fibroblast growth factor with the different components of the cell membrane mitogenesis-triggering complex. The development and maintenance of blood vessel networks for the supply and withdrawal of metabolites is a critical event in the development of solid tumors when they reach a critical size (1Yancopoulos G.D. Davis S. Gale N.W. Rudge J.S. Wiegand S.J. Holash J. Nature. 2000; 407: 242-248Crossref PubMed Scopus (3307) Google Scholar). In addition, these vascular networks strongly contribute to the dissemination of malignant cells to distant sites (2Carmeliet P. Jain R.K. Nature. 2000; 407: 249-257Crossref PubMed Scopus (7603) Google Scholar). Inhibition of angiogenesis was proposed as a strategy for the treatment of tumors about three decades ago (3Folkman J. N. Engl. J. Med. 1971; 285: 1182-1186Crossref PubMed Scopus (220) Google Scholar). This now commonly accepted idea is at the forefront of one of the most promising fields in cancer research, i.e. the blockage of angiogenesis-promoting factors (2Carmeliet P. Jain R.K. Nature. 2000; 407: 249-257Crossref PubMed Scopus (7603) Google Scholar, 4Rak J. Yu J.L. Kerbel R.S. Coomber B.L. Cancer Res. 2002; 62: 1931-1934PubMed Google Scholar). At present there are more than 20 antiangiogenic agents in different stages of clinical development for oncology (5Kruger E.A. Duray P.H. Price D.K. Pluda J.M. Figg W.D. Semina. Oncol. 2001; 28: 570-576Crossref PubMed Scopus (42) Google Scholar), a considerable number of which are polypeptides that the organism uses to counterbalance the effect of the positive angiogenesis regulators (6Hagedorn M. Bikfalvi A. Crit. Rev. Oncol. Hematol. 2000; 34: 89-110Crossref PubMed Scopus (117) Google Scholar). Acidic and basic fibroblast growth factors (aFGF and bFGF) 1The abbreviations used are: aFGF, acidic FGF; bFGF, basic FGF; FGF, fibroblast growth factor; NTS, 1,3,6-naphthalenetrisulfonate; NMS, naphthalene monosulfonate. are two very important angiogenesis-promoting polypeptides. They belong to a family of mitogens that to date includes 23 polypeptides (7The ADHR Consortium Nat. Genet. 2000; 26: 345-348Crossref PubMed Scopus (1307) Google Scholar, 8Yamashita T. Yoshioka M. Itoh N. Biochem. Biophys. Res. Commun. 2000; 277: 494-498Crossref PubMed Scopus (453) Google Scholar). The two members of the group that were first described are aFGF and bFGF. They have very similar biochemical and biological properties and have served as paradigms for the whole family (9Baird A. Böhlen P. Sporn M.B. Roberts A.B. Handbook of Experimental Pharmacology. 95. Springer, Berli1990: 369-418Google Scholar, 10Giménez-Gallego G. Cuevas P. Neurolog. Res. 1994; 16: 313-316Crossref PubMed Scopus (70) Google Scholar). FGFs are often detected in tumors (11Folkman J. Klagsbrun M. Science. 1987; 235: 442-447Crossref PubMed Scopus (4061) Google Scholar, 12Hayek A. Culler F.L. Beattie G.M. Lopez A.D. Cuevas P. Baird A. Biochem. Biophys. Res. Commun. 1987; 147: 876-880Crossref PubMed Scopus (110) Google Scholar). Histological studies have shown that the growth of solid tumors is suppressed by monoclonal antibodies against bFGF and that this effect was due to the inhibition of bFGF-induced angiogenesis (13Hori A. Sasada R. Matsutani E. Naito K. Sakura Y. Fujita T. Kozai Y. Cancer Res. 1991; 51: 6180-6184PubMed Google Scholar). In addition, it has been reported that antisense targeting of bFGF and FGF receptor-1 in human melanomas blocks intratumoral angiogenesis and tumor growth (14Wang Y. Becker D. Nat. Med. 1997; 3: 887-893Crossref PubMed Scopus (280) Google Scholar). Finally, angiogenesis promoted by another well characterized inducer of blood vessel development, the vascular endothelial growth factor, has been reported to require endogenous expression of bFGF by endothelial cells, and it is therefore blocked by neutralizing antibodies against bFGF (15Jonca F. Ortega N. Gleizes P.E. Bertrand N. Plouet J. J. Biol. Chem. 1997; 272: 24203-24209Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar, 16Mandriota S.J. Pepper M.S. J. Cell Sci. 1997; 110: 2293-2302PubMed Google Scholar). Consequently, inhibition of FGF mitogenic activity seems a crucial target for the development of antiangiogenic cancer treatments. FGFs show a characteristically high affinity for the glycosaminoglycan heparin and the glycosidic moiety of heparan sulfate proteoglycan (9Baird A. Böhlen P. Sporn M.B. Roberts A.B. Handbook of Experimental Pharmacology. 95. Springer, Berli1990: 369-418Google Scholar, 10Giménez-Gallego G. Cuevas P. Neurolog. Res. 1994; 16: 313-316Crossref PubMed Scopus (70) Google Scholar, 17Burgess W.H. Maciag T. Annu. Rev. Biochem. 1989; 58: 575-606Crossref PubMed Google Scholar). It has been demonstrated that binding to either of these polysulfates is required for FGFs to recognize their specific tyrosine kinase receptor on the cell surface (18Rapraeger A.C. Krufka A. Olwin B.B. Science. 1991; 252: 1705-1708Crossref PubMed Scopus (1294) Google Scholar, 19Yayon A. Klagsbrun M. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 5346-5350Crossref PubMed Scopus (112) Google Scholar, 20Jaye M. Schlessinger J. Dionne C.A. Biochim. Biophys. Acta. 1992; 1135: 185-199Crossref PubMed Scopus (598) Google Scholar). In the case of aFGF-driven mitogenesis, the presence of either heparin or some specific polyanions like myo-inositol hexasulfate is, in addition, a nearly absolute requirement (21Giménez-Gallego G. Con G. Hatcher V.B. Thomas K.A. Biochem. Biophys. Res. Commun. 1986; 135: 541-548Crossref PubMed Scopus (91) Google Scholar, 22Pineda-Lucena A. Jiménez M.A. Nieto J.L. Santoro J. Rico M. Giménez-Gallego G. J. Mol. Biol. 1994; 242: 81-98Crossref PubMed Scopus (46) Google Scholar). Thus, disruption of the interaction of FGFs with heparin and heparan sulfates seems an obvious target for antiangiogenesis. The polysulfonated binaphthyl ureas known as suramins are considered potential anti-cancer agents because of their anti-angiogenic activity (23Manetti F. Corelli F. Botta M. Curr. Pharm. Des. 2000; 6: 1897-1924Crossref PubMed Scopus (67) Google Scholar). The antiangiogenic activity of suramins is based, at least in part, on their ability to disrupt the interaction of many growth factors with their membrane receptors, such as in the case of FGFs and their tyrosine kinase receptors (19Yayon A. Klagsbrun M. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 5346-5350Crossref PubMed Scopus (112) Google Scholar, 24Gagliardi A. Kassack M. Kreimeyer A. Muller G. Nickel P. Collins D.C. Cancer Chemother. Pharmacol. 1998; 41: 117-124Crossref PubMed Scopus (45) Google Scholar, 25Pesenti E. Sola F. Mongelli N. Grandi M. Spreafico F. Brit. J. Cancer. 1992; 66: 367-372Crossref PubMed Scopus (123) Google Scholar, 26Takano S. Gately S. Neville M.E. Herblin W.F. Gross J.L. Engelhard H. Perricone M. Eidsvoog K. Brem S. Cancer Res. 1994; 54: 2654-2660PubMed Google Scholar). Because it has been shown that heparin disrupts aFGF·suramin complexes (27Middaugh C.R. Mach H. Burke C.J. Volkin D.B. Dabora J.M. Tsai P.K. Bruner M.W. Ryan J.A. Marfia K.E. Biochemistry. 1992; 31: 9016-9024Crossref PubMed Scopus (147) Google Scholar) and counteracts the antiangiogenic effect of these polysulfonated ureas (25Pesenti E. Sola F. Mongelli N. Grandi M. Spreafico F. Brit. J. Cancer. 1992; 66: 367-372Crossref PubMed Scopus (123) Google Scholar, 26Takano S. Gately S. Neville M.E. Herblin W.F. Gross J.L. Engelhard H. Perricone M. Eidsvoog K. Brem S. Cancer Res. 1994; 54: 2654-2660PubMed Google Scholar), suramins are considered to act by blocking the heparin-binding sites of FGFs (28Stein C.A. Cancer Res. 1993; 53: 2239-2248PubMed Google Scholar). Another group of antiangiogenic and anti-tumoral compounds is comprised by suradistas, a type of non-cytotoxic synthetic binaphthalene sulfonic distamycin-A derivatives. These compounds tightly interact with FGFs, inhibit the binding of these polypeptides to the tyrosine kinase cell membrane receptors, and suppress FGF-induced angiogenesis and neovascularization in vivo (29Mariani M. Paio A. Ciomei M. Pastori W. Franzetti C. Melegaro G. Grandi M. Mongelli N. Experientia (Basel). 1992; 61: 455-458Google Scholar, 30Ciomei M. Pastori W. Mariani M. Sola F. Grandi M. Mongelli N. Biochem. Pharmacol. 1994; 47: 295-302Crossref PubMed Scopus (51) Google Scholar, 31Zamai M. Parola A.H. Grandi M. Mongelli N. Caiolfa V.R. Med. Chem. Res. 1997; 7: 36-44Google Scholar), probably by a mechanism similar to that of the suramins. Despite the potential relevance of the suramins and the suradistas as inhibitors of the FGF angiogenic activity, no high resolution data of their complexes with these polypeptides are available so far. Knowledge of the molecular structure of these complexes could be used to improve their pharmacological properties and lead to the design of better angiogenesis inhibitors. A major reason for the lack of data is that these two inhibitors promote the appearance of heterogeneous aggregates of the complexes (23Manetti F. Corelli F. Botta M. Curr. Pharm. Des. 2000; 6: 1897-1924Crossref PubMed Scopus (67) Google Scholar, 27Middaugh C.R. Mach H. Burke C.J. Volkin D.B. Dabora J.M. Tsai P.K. Bruner M.W. Ryan J.A. Marfia K.E. Biochemistry. 1992; 31: 9016-9024Crossref PubMed Scopus (147) Google Scholar). Of outstanding interest was the finding that 1,3,6-naphthalenetrisulfonate (NTS) constitutes a minimal model for the inhibition of aFGF mitogenic activity by suramins and suradistas (32Lozano R.M. Jiménez M. Santoro J. Rico M. Giménez-Gallego G. J. Mol. Biol. 1998; 281: 899-915Crossref PubMed Scopus (53) Google Scholar). Moreover NTS has been tested with positive results both in vitro and in vivo as an inhibitor of aFGF-induced angiogenesis and glioma proliferation (32Lozano R.M. Jiménez M. Santoro J. Rico M. Giménez-Gallego G. J. Mol. Biol. 1998; 281: 899-915Crossref PubMed Scopus (53) Google Scholar, 33Cuevas P. Lozano R.M. Giménez-Gallego G. Neurol. Res. 1999; 21: 191-194Crossref PubMed Scopus (9) Google Scholar, 34Cuevas P. Carceller F. Reimers D. Cuevas B. Lozano R.M. Giménez-Gallego G. Neurol. Res. 1999; 21: 481-487Crossref PubMed Scopus (12) Google Scholar, 35Cuevas P. Reimers D. Díaz D. Lozano R.M. Giménez-Gallego G. Neurosci. Lett. 1999; 275: 149-151Crossref PubMed Scopus (15) Google Scholar). This suggests potential new avenues for the development of new antiangiogenic compounds. The three-dimensional structure of aFGF complexed with NTS obtained by 1H NMR showed that NTS binds weakly and quite heterogeneously to the heparin-binding site of aFGF, in agreement with its low binding constant (32Lozano R.M. Jiménez M. Santoro J. Rico M. Giménez-Gallego G. J. Mol. Biol. 1998; 281: 899-915Crossref PubMed Scopus (53) Google Scholar). The studies of Lozano et al. (32Lozano R.M. Jiménez M. Santoro J. Rico M. Giménez-Gallego G. J. Mol. Biol. 1998; 281: 899-915Crossref PubMed Scopus (53) Google Scholar) also showed that naphthalene derivatives containing a reduced number of sulfonate groups per aromatic ring seem to act as better inhibitors of aFGF mitogenic activity than NTS. However, this enhancement was accompanied by the appearance of a clear toxicity of the compounds against quiescent cells at concentrations where they inhibit aFGF mitogenic activity. Based on these preliminary data, we have explored a wide window of charge, size, and relative position of substituents of the naphthalene ring in an attempt to identify new naphthalene derivatives that combine the highest inhibitory activity with the lowest toxicity. In a further step, the most pharmacologically promising of those compounds, 5-amino-2-naphthalenesulfonate, was evaluated for its antiangiogenic activity in vivo with positive results, and the three-dimensional structure of its complex with aFGF was solved by x-ray crystallography to a resolution of 2.0 Å. All of these studies allow formulation of a set of stereochemical rules that may constitute the basis for the development of new antiangiogenesis treatments. A complete list of all of the compounds used in the study reported here is shown in Table I, and their formulas are provided in the supplemental data. All of the reactions were performed in dry flasks fitted with glass stopper or rubber septa under argon, unless otherwise noted. Air- and moisture-sensitive liquid reagents were transferred via syringe or stainless steel cannula. Flash column chromatography was performed employing 230–400 mesh silica gel. Thin layer chromatography was conducted in Kiesel gel 60 F254 (Merck). Detection was first by UV (254 nm) and then by charring with a solution of 20% aqueous sulfuric acid (200 ml) in acetic acid (800 ml). Anhydrous MgSO4 or Na2SO4 was used to dry the organic solutions during work-ups, and the removal of the solvents was carried out under vacuum with a rotary evaporator. Unless otherwise noted, the materials were obtained from commercially available sources and used without further purification. Compounds prepared in our lab have had their structures and purity confirmed by NMR. The solvents were dried and purified using standard methods. 1H and 13C NMR spectra were recorded in D2O. The mass spectra were recorded on a liquid chromatography coupled mass spectroscopy HP 1100 spectrometer using either chemical or electrospray ionization.Table IList of assayed compoundsCompoundIUPAC name1-NMS seriesC-11-Naphtalenesulfonic acidC-2Sodium 2-methyl-1-naphtalenesulfonateC-3Sodium 4-amino-1-naphtalenesulfonateC-4Sodium 4-(acetylamino)-1-naphtalenesulfonateC-5Sodium 4-(benzoylamino)-1-naphtalenesulfonateC-6Sodium 5-dimethylamino-1-naphtalenesulfonateC-7Sodium 4-amino-3-hydroxy-1-naphtalenesulfonate2-NMS seriesC-82-Naphtalenesulfonic acidC-9Sodium 5-amino-2-naphtalenesulfonateC-10Sodium 5-(acetylamino)-2-naphtalenesulfonateC-11Sodium 5-(benzoylamino)-2-naphtalenesulfonateC-12Sodium 2-{[(6-sulfo-1-naphtyl)amino]carbonyl}-benzoic acidC-13Sodium 5-({6-oxo-6[(6-sulfo-1-naphtyl)amino]-hexanoyl}amino)-2-naphtalenesulfonateC-14Sodium 8-amino-2-naphtalenesulfonateC-15Sodium 8-(acetylamino)-2-naphtalenesulfonateC-16Sodium 8-(benzoylamino)-2-naphtalenesulfonateC-17Sodium 6-amino-4-hydroxy-2-naphtalenesulfonateC-18Sodium 7-amino-4-hydroxy-2-naphtalenesulfonateC-19Sodium 7-(acetylamino)-4-hydroxy-2-naphtalenesulfonateOtherC-20Sodium 4-aminobenzenesulfonateC-21Sodium 4-(aminomethyl)-benzenesulfonateC-22Sodium 4-(aminoethyl)-benzenesulfonateC-23Sodium 6-quinolinesulfonateC-24Sodium 8-quinolinesulfonateC-252-[(Dimethylamino)methyl]-1H-indole-5-sulfonic acid Open table in a new tab Sodium Salts—Sodium salts were prepared by adding the required amounts of 0.1 n sodium hydroxide to a suspension of the corresponding sulfonic acid in water. Evaporation of the solvents yielded the sodium salts. Acetyl Amide Derivatives (C-4, C-10, C-15, and C-19)—Acetyl amides were prepared by heating a suspension of the corresponding sodium sulfonate salts in acetic anhydride for 4 h at 90 °C. The solvent was evaporated, yielding the acetyl amide derivatives. Benzoyl Amide Derivatives (C-5, C-11, and C-16)—A slight excess of benzoyl chloride (1.2 equivalents) was added to a solution of the corresponding sulfonic acid in pyridine. The solvent was evaporated after 30 min, and the residue was chromatographed on a silica gel column (MeOH:CH2Cl2, 1:9 v/v). The purified product was transformed in its corresponding sodium salt (see above). Compound C-2—Concentrated H2SO4 (0.32 ml) was added to 2-methyl-naphtalene (2.8 mmol). The mixture was heated in an oil bath at 80 °C for 3 h. This solution was cooled, poured over ice, and made basic by the addition of 1 n NaOH. The product was separated in cold and isolated by filtration. Compound C-8 —Ion-exchange resin (Amberlite) (H+ form) was added to a solution of sodium 2-naphthalenesulfonate in methanol. The mixture was stirred overnight and filtered, and the solvent was evaporated. The residue was lyophilized and yields the acid. Compound C-12—Excess of phthalic anhydride (1.3 equivalents) was added to a solution of C-9 (0.6 mmol) in methanol. The mixture was stirred overnight and evaporated. The residue was chromatographed on silica gel column (MeOH:CH2Cl2, 3:7 v/v) to yield the required product. Compound C-13—Adipoyl chloride (0.5 equivalents) was added to a suspension of C-9 (1.0 mmol) in pyridine. Stirring was continued for 48 h, and the solvent was evaporated. The residue was digested with a small volume of methanol, and the solid that separated was collected and characterized. Compound C-25—Gramine (6.45 mmol) was dissolved in concentrated H2SO4 (10 ml) and stirred for 5 min at 0 °C. This solution was poured over ice and made basic by the addition of 1 n NaOH. This aqueous solution was extracted with dichloromethane, made neutral by addition of 1 n H2SO4, and evaporated. The residue was digested with hot ethanol, and the solids were separated by filtration. The ethanol solution was evaporated, and the residue was subjected to column chromatography (ethanol) to yield the desired compound. The effect of the compounds listed on Table I on aFGF-driven mitogenesis and quiescent cell viability was studied in vitro as described by Lozano et al. (32Lozano R.M. Jiménez M. Santoro J. Rico M. Giménez-Gallego G. J. Mol. Biol. 1998; 281: 899-915Crossref PubMed Scopus (53) Google Scholar) using 100-μl cultures of plated fibroblasts Balb/c 3T3. The concentration of aFGF used in the assays (0.32 ng/ml) correspond to that eliciting a half-maximum mitogenic response at myo-inositol hexasulfate concentrations that induce a nearly full activation of the mitogen (20 μg/ml; 32). The inhibitors and, when pertinent, aFGF were added in 10 μl of a solution containing myo-inositol hexasulfate (200 μg/ml), bovine serum albumin (1 mg/ml) in Dulbecco's modified Eagle's medium, once the cultures have reached quiescence (15 h after being transferred to minimal culture medium). Pathogen-free C57/BI/6 mice (Charles River) weighing 25 ± 4 g were used. The animals were housed in plastic cages in temperature- and humidity-controlled conditions; food and water were available ad libitum, and a 12-h light/dark schedule was maintained. The animal welfare guidelines of the National Institutes of Health and the European Union were carefully followed. Sterile gelatin sponge cubes of 10 mm3 (Curaspon Dental, Clinimed Holding, Zwanenburg, The Netherlands) were implanted subcutaneously in the backs of the mice after induction of intraperitoneal anesthesia as described (33Cuevas P. Lozano R.M. Giménez-Gallego G. Neurol. Res. 1999; 21: 191-194Crossref PubMed Scopus (9) Google Scholar). The animals were distributed as follows: Group A (n = 10) sponges loaded with 200 μl of phosphate-buffered saline containing 29 μg·ml-1 heparin; Group B (n = 40) sponges were embedded with the same solution containing 10 μg·ml-1 aFGF. After implantation of the sponge into the subcutaneous pouch, the skin was sutured. The mice of Group B were randomly divided in four groups (n = 10) that received 0.008, 0.08, 0.8, and 8 mg·kg-1 of 5-amino-2-NMS, respectively, in an intraperitoneal injection of 200 μl of phosphate-buffered saline, 24 h after surgery (25Pesenti E. Sola F. Mongelli N. Grandi M. Spreafico F. Brit. J. Cancer. 1992; 66: 367-372Crossref PubMed Scopus (123) Google Scholar, 33Cuevas P. Lozano R.M. Giménez-Gallego G. Neurol. Res. 1999; 21: 191-194Crossref PubMed Scopus (9) Google Scholar). All procedures were performed under sterile conditions. For angiogenesis evaluation, the mice were re-anesthetized as described, and the sponges were surgically extracted and treated for histological studies as described by Cuevas et al. (33Cuevas P. Lozano R.M. Giménez-Gallego G. Neurol. Res. 1999; 21: 191-194Crossref PubMed Scopus (9) Google Scholar) 7 days after the implants. Neovascularization was quantified with computerized morphometric software connected to a microscope. Ingrowths of neovessels into sponges was assessed by measuring surface area with erythrocyte content. Neovascularization was analyzed in four predetermined visual fields in three different sections at 10-fold magnification. The statistical analyses were performed by using Student's t test. The protein was obtained using established protocols (22Pineda-Lucena A. Jiménez M.A. Nieto J.L. Santoro J. Rico M. Giménez-Gallego G. J. Mol. Biol. 1994; 242: 81-98Crossref PubMed Scopus (46) Google Scholar). The residues are numbered according to their positions in the primary structure of the 154-amino acid aFGF (36Zazo M. Lozano R.M. Ortega S. Varela J. Díaz-Orejas R. Ramírez J.M. Giménez-Gallego G. Gene. 1992; 113: 231-238Crossref PubMed Scopus (64) Google Scholar). Crystals of the complex between aFGF and either 5-amino-2-NMS or NTS were grown using the sitting drop vapor diffusion method at 295 K by mixing 0.75 mm protein, 1.5 mm of the inhibitor and 60% sodium/potassium tartrate buffered with 5 mm sodium phosphate (pH 7.85) over a well solution of 1.3 m Li2SO4. The typical crystal size achieved after ∼1 month was about 0.7 × 0.5 × 0.2 mm. To corroborate the presence of the inhibitor inside the crystals, several crystal specimens were washed using the crystallization buffer, solubilized with 1 n NaOH, and subjected to electrospray mass spectrometry. The mass spectra were recorded on a liquid chromatography coupled mass spectroscopy HP 1100 spectrometer using electrospray ionization. For the diffraction experiments, the crystals were cryo-cooled in a stream of nitrogen gas at 100 K using the crystallization solution supplemented with 20% glycerol as cryo-protectant. X-rays from a synchrotron source at Beamline X11-DESY (Hamburg, Germany) were employed to collect data on a Marccd detector. Diffraction was visible to 1.8 Å, but only data to 2.0 Å were suitable for subsequent processing with MOSFLM (37Leslie A.G.W. Moras D. Podjarny A.D. Thierry J.C. Crystallographic Computing. 5. Oxford University Press, Oxford1991: 27-38Google Scholar) and programs of the CCP4 suite (38P4 CC Acta Crystallogr. Sect. D Biol. Crystallogr. 1994; 50: 760-763Crossref PubMed Scopus (19824) Google Scholar). The crystals belong to the monoclinic system, space group P2 (a = 96.5 Å, b = 47.2 Å, c = 97.8 Å, β = 107.0°), with six protein molecules/asymmetric unit corresponding to a 48% solvent content. The calculated VM value (39Matthews B.W. J. Mol. Biol. 1968; 33: 491-497Crossref PubMed Scopus (7935) Google Scholar) is 2.4 Å3/Da. The structure was determined by molecular replacement using AMoRe (40Navaza J. Acta Crystallogr. Sect. A. 1994; 50: 157-163Crossref Scopus (5031) Google Scholar). The searching model was the one derived from 2axm (41DiGabriele A.D. Lax I. Chen D.I. Svahn C.M. Jaye M. Schlessinger J. Hendrickson W.A. Nature. 1998; 393: 812-817Crossref PubMed Scopus (330) Google Scholar), with one molecule searched six times at 3.5 Å resolution. The positions of the six molecules in the asymmetric unit were optimized using rigid body refinement, leading to an R value of 44.9 (Rfree = 45.2). Inspection of the initial sigma-A maps showed that certain side chains were quite disordered, although some electron density was obvious. This problem was overcome by use of noncrystallographic symmetry restraints. Several steps of simulated annealing and B factor refinement were carried out with CNS (42Brünger A.T. Adams P.D. Clore G.M. DeLano W.L. Gros P. Grosse-Kunstleve R.W. Acta Crystallogr. Sect. D Biol. Crystallogr. 1998; 54: 905-921Crossref PubMed Scopus (16991) Google Scholar) until the Rwork and Rfree values dropped to 22.8 and 25.9%, respectively. Bulk solvent and anisotropic overall B factor corrections were applied through the refinement. The final model is of excellent geometry as shown in Table II.Table IIData collection, phasing, and refinementData collection statisticsSpace groupP2Cell dimensions (Å, °)a = 96.5, b = 47.2, c = 97.8, β = 107.0No. molecules/asymmetric unit6Wavelength (Å)0.9073(Resolution (Å)33.5-2.0Measurements321,819Unique reflections56,957Rsym (%)aThe values in parentheses correspond to the highest resolution shell (2.15-2.00 Å).4.4 (21.9)I/σ(I)aThe values in parentheses correspond to the highest resolution shell (2.15-2.00 Å).9.7 (2.4)Completeness (%)aThe values in parentheses correspond to the highest resolution shell (2.15-2.00 Å).99.1 (99.1)Structure determination statisticsMolecular replacement C-factor58.5Molecular replacement R-factor41.4Refinement statisticsResolution range (Å)30-2.0Reflections (work/free)bThe reflections in the test set represent 10% of the total number of reflections.51,171/5,786Rwork/Rfree (%)23.1/26.1Root mean square deviationsBond lengths (Å)0.007Bond angles (°)1.23Dihedral angles (°)24.5Number of atomsProtein61885-Amino-2-NMS30Solvent103Average B factor (Å2)34.9a The values in parentheses correspond to the highest resolution shell (2.15-2.00 Å).b The reflections in the test set represent 10% of the total number of reflections. Open table in a new tab The coordinates for the final model have been deposited in the Protein Data Bank under the code 1hkn. Inhibition of aFGF Mitogenic Activity by Naphthalene Sulfonates—To single out new naphthalene sulfonate derivatives with better potential pharmacological profiles as angiogenic inhibitors than naphthalene trisulfonate (32Lozano R.M. Jiménez M. Santoro J. Rico M. Giménez-Gallego G. J. Mol. Biol. 1998; 281: 899-915Crossref PubMed Scopus (53) Google Scholar), a rational search, by iterative trial-and-error, of the best functional group substitutions was carried out. Two starting compounds were initially used in this search: 1-naphthalene monosulfonate (1-NMS) and 2-naphthalene monosulfonate (2-NMS). 1-NMS inhibits aFGF mitogenic activity at lower concentrations than naphthalene trisulfonate (32Lozano R.M. Jiménez M. Santoro J. Rico M. Giménez-Gallego G. J. Mol. Biol. 1998; 281: 899-915Crossref PubMed Scopus (53) Google Scholar), but it becomes toxic for quiescent fibroblasts at the highest concentrations required to reach a substantial inhibition aFGF mitogenic activity (32Lozano R.M. Jiménez M. Santoro J. Rico M. Giménez-Gallego G. J. Mol. Biol. 1998; 281: 899-915Crossref PubMed Scopus (53) Google Scholar). 2-NMS inhibits aFGF mitogenic activity at still lower concentrations than 1-NMS, but it has the same toxicity effects as 1-NMS (supplemental data). The search space was later widened to other sulfonate-derivatized aromatic compounds. A complete list of all tested compounds is shown in Table I, and their formulas are provided in the supplemental data. The whole series of tested compounds covers a wide range of specific inhibitory activities illustrated in Fig. 1a, despite the sometimes significant similarities in chemical composition. This underlines the relevance of the stereochemistry in the inhibition of aFGF mitogenic activity by
Referência(s)