复旦大学：《药物设计学》课程教学资源（虚拟实验室课外实践）Exercises for Drug Design Courses_Practice-IV TOXICITY SCREENING OF THERAPEUTIC DRUGS

version date: 1 December 2006 EXERCISEⅣV1 TOXICITY SCREENING OF THERAPEUTIC DRUGS Mario Suwalsky Faculty of Chemical Sciences, University of Concepcion, Casilla 160-C, Concepcion, Chile Fax: 56-41 245 974: E-mail: msuwalsk@udec cl INTRODUCTION In the course of in vitro systems search for the toxicity screening of therapeutic drugs (TDs), different cellular models have been applied to examine their adverse effects in isolated organs. This article describes a simple method to determine effects of TDs at the cell membrane level. The cell membrane is an assembly of proteins and lipids that separate inside from outside, protecting the cell interior. The membrane is also involved in a variety of indispensable cellular functions. It is responsible for the selective transport of molecules and ions into and out of the cell in the extensive network responsible for the traffic between organelles. Without exception, these activities depend on, and are influenced by the physical milieu provided by the molecules making up the membrane bilayers. Changes in the physical and chemical environment of the cell membranes have a direct effect on the membrane structure with serious effects on the cell functions [1-2]. Most biological membranes possess an asymmetric trans-bilayer distribution of phospholipids [3]. Thus, for instance, most eukaryotic plasma membranes present a high percentage of the phospholipid sphingomyelins and phosphatidylcholines in the outer monolayer, whereas the inner one is enerally richer phosphatidylethanolamines, phosphatidylserines phosphatidylinositols. However, the existence of asymmetric plasma membranes is less certain in bacteria than in eukaryotes. Studies of the phospholipid distribution of a gram positive bacteria revealed that the outer monolayer is rich in phosphatidylglycerols, the inner one in phosphatidylinositols, while cardiolipins are symmetrically distributed between both monolayers. Studies on gram-negative bacteria such as Escherichia coli have detected phosphatidylethanolamines in the outer membrane, whereas the cytoplasmic membrane has been reported to be rich in phosphatidylglycerols and cardiolipins With the aim to better understand the molecular mechanisms of the interaction of tds with cell membranes. we utilize a well-established model consisting of intact human

1 EXERCISE IV.1 TOXICITY SCREENING OF THERAPEUTIC DRUGS Mario Suwalsky Faculty of Chemical Sciences, University of Concepcion, Casilla 160-C, Concepcion, Chile Fax: 56-41 245 974; E-mail: msuwalsk@udec.cl INTRODUCTION In the course of in vitro systems search for the toxicity screening of therapeutic drugs (TDs), different cellular models have been applied to examine their adverse effects in isolated organs. This article describes a simple method to determine effects of TDs at the cell membrane level. The cell membrane is an assembly of proteins and lipids that separate inside from outside, protecting the cell interior. The membrane is also involved in a variety of indispensable cellular functions. It is responsible for the selective transport of molecules and ions into and out of the cell in the extensive network responsible for the traffic between organelles. Without exception, these activities depend on, and are influenced by the physical milieu provided by the molecules making up the membrane bilayers. Changes in the physical and chemical environment of the cell membranes have a direct effect on the membrane structure with serious effects on the cell functions [1–2]. Most biological membranes possess an asymmetric trans-bilayer distribution of phospholipids [3]. Thus, for instance, most eukaryotic plasma membranes present a high percentage of the phospholipids sphingomyelins and phosphatidylcholines in the outer monolayer, whereas the inner one is generally richer in phosphatidylethanolamines, phosphatidylserines, and phosphatidylinositols. However, the existence of asymmetric plasma membranes is less certain in bacteria than in eukaryotes. Studies of the phospholipid distribution of a gram￾positive bacteria revealed that the outer monolayer is rich in phosphatidylglycerols, the inner one in phosphatidylinositols, while cardiolipins are symmetrically distributed between both monolayers. Studies on gram-negative bacteria such as Escherichia coli have detected phosphatidylethanolamines in the outer membrane, whereas the cytoplasmic membrane has been reported to be rich in phosphatidylglycerols and cardiolipins. With the aim to better understand the molecular mechanisms of the interaction of TDs with cell membranes, we utilize a well-established model consisting of intact human version date: 1 December 2006

version date: 1 December 2006 erythrocytes and molecular models of its membrane. The latter consist of bilayers of dimyristoylphosphatidylcholine (DMPC) and dimyristoylphosphatidylethanolamine (DMPE), representative of phospholipid classes respectively located in the outer and inner monolayers of erythrocytes and other cell membranes. Erythrocytes were chosen because although less specialized than many other cell membranes, they carry on enough functions in common with them such as active and passive transport, and the production of ionic and electric gradients, to be considered representative of the plasma membrane in general. The capacity of TDs to interact with the erythrocyte membrane can be determined by scanning electron microscopy(SEM), whereas the interaction with the bilayer structures of DMPC and DMPE can be defined by X-ray diffracti These techniques have been used in our laboratories to determine the interaction with and the membrane-perturbing effects of local anesthetics [4-5], antiarrhythmic [6], and anticancer drugs [7-8] MATERIAL AND METHODS X-ray diffraction analysis of phospholipid bilayers The capacity of TDs to perturb the structures of dmPC and dmpe bilayers is to be determined by X-ray diffraction. For this purpose, about 1 mg of each phospholipid (Sigma or Polar Avanti, USA) is introduced into 2-mm-diameter special glass capillaries(Glas-Technik Konstruktion, Berlin, Germany), which are then filled with about 200 HL of(a) distilled water and(b)aqueous solutions of the td in a range of concentrations. The experiments must be performed at 17+ 2C, which is below the main phase transition temperature of both DMPC and DMPE SEM studies on human erythrocytes Blood samples can be taken from clinically healthy adult donors by puncture of the ear lobe disinfected with ethanol and aspiration into a tuberculin syringe without a needle containing 50 units/mL heparin in saline solution (0.9% NaCl). Centrifuge the red blood cells for 10 min at 1000 rpm, wash twice in saline solution, resuspend in salin solutions containing the TD in adequate concentrations, and incubate for 1 h at 37C Control are cells resuspended in saline solution without TD. Fix the specimens overnight at 5C by adding one drop of each sample to plastic tubes containing l mL of 5% glutaraldehyde, wash twice in distilled water, place them on siliconized Al stubs

2 erythrocytes and molecular models of its membrane. The latter consist of bilayers of dimyristoylphosphatidylcholine (DMPC) and dimyristoylphosphatidylethanolamine (DMPE), representative of phospholipid classes respectively located in the outer and inner monolayers of erythrocytes and other cell membranes. Erythrocytes were chosen because although less specialized than many other cell membranes, they carry on enough functions in common with them such as active and passive transport, and the production of ionic and electric gradients, to be considered representative of the plasma membrane in general. The capacity of TDs to interact with the erythrocyte membrane can be determined by scanning electron microscopy (SEM), whereas the interaction with the bilayer structures of DMPC and DMPE can be defined by X-ray diffraction. These techniques have been used in our laboratories to determine the interaction with and the membrane-perturbing effects of local anesthetics [4–5], antiarrhythmic [6], and anticancer drugs [7–8]. MATERIAL AND METHODS X-ray diffraction analysis of phospholipid bilayers The capacity of TDs to perturb the structures of DMPC and DMPE bilayers is to be determined by X-ray diffraction. For this purpose, about 1 mg of each phospholipid (Sigma or Polar Avanti, USA) is introduced into 2-mm-diameter special glass capillaries (Glas-Technik & Konstruktion, Berlin, Germany), which are then filled with about 200 µL of (a) distilled water and (b) aqueous solutions of the TD in a range of concentrations. The experiments must be performed at 17 ± 2°C, which is below the main phase transition temperature of both DMPC and DMPE. SEM studies on human erythrocytes Blood samples can be taken from clinically healthy adult donors by puncture of the ear lobe disinfected with ethanol and aspiration into a tuberculin syringe without a needle containing 50 units/mL heparin in saline solution (0.9 % NaCl). Centrifuge the red blood cells for 10 min at 1000 rpm, wash twice in saline solution, resuspend in saline solutions containing the TD in adequate concentrations, and incubate for 1 h at 37 ºC. Control are cells resuspended in saline solution without TD. Fix the specimens overnight at 5 ºC by adding one drop of each sample to plastic tubes containing 1 mL of 2.5 % glutaraldehyde, wash twice in distilled water, place them on siliconized Al stubs version date: 1 December 2006

version date: 1 December 2006 and air-dry at 37C for 30 min. Gold-coat the Al stubs for 3 min at 13.3 pascal in a sputter device and examine the samples in an SEm DESCRIPTION OF RESULTS n order to understand the possible results to be obtained, a description of the experimental results observed with the local anesthetic bupivacaine is presented X-ray diffraction studies Figure lA shows a comparison of the diffraction patterns of DMPC alone immersed in excess water and those of dMPC incubated with bupivacaine in the range of 0.1 mM up to 15 mM. The reflections labeled(a) correspond to the 64 a distance between DMPC polar groups(see Fig. 2a), whereas the strong reflection of 4.2 A labeled(b) corresponds to the average distance between DMPc fully extended acyl chains organized with rotational disorder in hexagonal packing. It is noticeable that addition of 0. 1 mM bupivacaine caused only a very slight decrease in the phospholipid reflection intensities, but 1 mM induced a marked decrease of the (a) intensities, whereas the intensity of 4.2 A reflection was essentially unchanged. However, 8 mM bupivacaine induced a marked decrease of the 4.2 A reflection intensity and the complete disappearance of the(a)reflections, which were replaced by a diffuse halo. This pattern remained practically unchanged after exposure to 15 mM bupivacaine. These results imply that bupivacaine induced serious molecular disorder in the dmPc bilayer, especially in the region of the polar head groups Figure 1B shows the results of the interaction of bupivacaine with DMPE. The perturbing effect of this compound upon the structure of dmpe bilayers was practically negligible even at a 23 mM concentration. As a matter of fact, these two phospholipids differ only in their terminal amino groups, these being N(CH3 )3 in DMPC and NH3 in DMPE. Moreover, both molecular conformations are very similar with the hydrocarbon chains mostly parallel and extended, and the polar groups lying perpendicular to them (Fig. 2b ). However, the hydration of DMPC results in water filling the highly polar interbilayer spaces, a phenomenon that allows the incorporation of bupivacaine into DMPC bilayers, producing its structural perturbation and almost complete destruction at a 8 mM concentration. On the other hand. dmPe molecules pack tighter than those of dmpc due to their smaller polar group and higher effective charge, resulting in a

3 and air-dry at 37 ºC for 30 min. Gold-coat the Al stubs for 3 min at 13.3 pascal in a sputter device and examine the samples in an SEM. DESCRIPTION OF RESULTS In order to understand the possible results to be obtained, a description of the experimental results observed with the local anesthetic bupivacaine is presented. X-ray diffraction studies Figure 1A shows a comparison of the diffraction patterns of DMPC alone immersed in excess water and those of DMPC incubated with bupivacaine in the range of 0.1 mM up to 15 mM. The reflections labeled (a) correspond to the 64 Å distance between DMPC polar groups (see Fig. 2a), whereas the strong reflection of 4.2 Å labeled (b) corresponds to the average distance between DMPC fully extended acyl chains organized with rotational disorder in hexagonal packing. It is noticeable that addition of 0.1 mM bupivacaine caused only a very slight decrease in the phospholipid reflection intensities, but 1 mM induced a marked decrease of the (a) intensities, whereas the intensity of 4.2 Å reflection was essentially unchanged. However, 8 mM bupivacaine induced a marked decrease of the 4.2 Å reflection intensity and the complete disappearance of the (a) reflections, which were replaced by a diffuse halo. This pattern remained practically unchanged after exposure to 15 mM bupivacaine. These results imply that bupivacaine induced serious molecular disorder in the DMPC bilayer, especially in the region of the polar head groups. Figure 1B shows the results of the interaction of bupivacaine with DMPE. The perturbing effect of this compound upon the structure of DMPE bilayers was practically negligible even at a 23 mM concentration. As a matter of fact, these two phospholipids differ only in their terminal amino groups, these being + N(CH3)3 in DMPC and + NH3 in DMPE. Moreover, both molecular conformations are very similar with the hydrocarbon chains mostly parallel and extended, and the polar groups lying perpendicular to them (Fig. 2b). However, the hydration of DMPC results in water filling the highly polar interbilayer spaces, a phenomenon that allows the incorporation of bupivacaine into DMPC bilayers, producing its structural perturbation and almost complete destruction at a 8 mM concentration. On the other hand, DMPE molecules pack tighter than those of DMPC due to their smaller polar group and higher effective charge, resulting in a version date: 1 December 2006

n date 1 December 2006 very stable bilayer system that is not significantly affected by water nor by a number of compounds A DMPC: DMPE: HO 0.1mM 入+1 mM Bupi 8 mM Bupin 15 mM Bupivacaine 入t+10 mMBupivacaine A+23 mM Bupivacaine H2o H2o 64.5 4.2 4.0 Observed Spacing(A) Fig. 1 X-ray diffraction diagrams of DMPC(A)and DMPE (B)in water and aqueous solutions of bupivacaine;(a) reflections arising from the polar group and(b) from the acyl chain regions

4 very stable bilayer system that is not significantly affected by water nor by a number of compounds. Fig. 1 X-ray diffraction diagrams of DMPC (A) and DMPE (B) in water and aqueous solutions of bupivacaine; (a) reflections arising from the polar group and (b) from the acyl chain regions. (a) (b) A 64.5 4.2 H2O + 8 mM Bupivacaine + 15 mM Bupivacaine + 1 mM Bupivacaine + 0.1 mM Bupivacaine +H2O DMPC: O bserved Relative Intensity (a) (b) A 64.5 4.2 H2O + 8 mM Bupivacaine + 15 mM Bupivacaine + 1 mM Bupivacaine + 0.1 mM Bupivacaine +H2O DMPC: O bserved Relative Intensity Spacing (Å) (a) (b) B 51.4 4.0 H2O + 23 mM Bupivacaine + 10 mM Bupivacaine +H2O DMPE: version date: 1 December 2006

version date 1 December 2006 爹游 Fig 2 Two-dimensional packing arrangements of (a) DMPC and(b) DMPe bilayer SEM studies on human erythrocytes Human red blood cells were incubated with 3 mM bupivacaine. The phase contrast and SEM observations indicated that bupivacaine induced a significant change in the shape Fig. 3 Effect of bupivacaine on the morphology of human erythrocytes. SEM images of (A)control(2500X) and(B)erythrocytes incubated with 3 mM bupivacaine 2400X) <www.iupac.org/publications/cd/medicinalchemistry

5 Fig. 2 Two-dimensional packing arrangements of (a) DMPC and (b) DMPE bilayers. SEM studies on human erythrocytes Human red blood cells were incubated with 3 mM bupivacaine. The phase contrast and SEM observations indicated that bupivacaine induced a significant change in the shape A B Fig. 3 Effect of bupivacaine on the morphology of human erythrocytes. SEM images of (A) control (2500X) and (B) erythrocytes incubated with 3 mM bupivacaine (2400X). version date: 1 December 2006

version date: 1 December 2006 of the erythrocytes. In fact, the erythrocytes underwent a morphological alteration as they changed their discoid shape(Fig. 3A)to spheroechinocytes(Fig. 3B). According to the bilayer couple hypothesis [9], the shape changes induced in erythrocytes by foreign molecules are due to differential expansion of their two monolayers. Thus, spiculated shapes(echinocyte) are induced when the added compound is inserted in the outer monolayer, whereas cup shapes(stomatocytes) arise when the compound accumulates in the inner monolayer. The fact that bupivacaine produced pheroechinocytes would indicate that the anesthetic was located in the outer moiety of the red cell membrane n conclusion, the experimental results indicate that bupivacaine interacts with the human erythrocyte membrane, the anesthetic being located in the outer moiety of the red cell membrane. The X-ray experiments, performed on bilayers made up of classes of the major phospholipids present in either the outer and inner sides of the erythrocyte membrane, confirmed this result. In fact, they showed that 1 mM bupivacaine slightly disordered the polar head region of DMPC (major class of lipid present in the outer monolayer of the erythrocyte membrane) and 8 mM completely perturbed it, whereas 23 mM bupivacaine produced negligible effects on DMPE ( which preferentially locates in the erythrocyte inner monolayer) REFERENCES [1] Membrane Physiology, T.E. Andreoli, J. F. Hoffman, DD. Fanestil, eds, Plenum, 1980 [2] In Search of a New Biomembrane Model, O.G. Mouritsen, O.S. Anderen, eds Munksgaard, 1998 [3]J M. Boon, B D. Smith, Chemical Control of Phospholipid Distribution Across Bilayer Membranes, Med. Res Rev. 22(2002), 251-281 14 M. Suwalsky, C. Schneider, F. Villena, B. Norris, H. Cardenas, F. Cuevas, C P Sotomayor, Structural Effects of the Local Anesthetic Bupivacaine Hydrochloride on the Human Erythrocyte Membrane and Molecular Models, Blood Cells, Mol. Dis (2002),29,14-23 5 M. Suwalsky, C. Schneider, F. Villena, B. Norris, H. Cardenas, F. Cuevas, C P Sotomayor, Effects of the Local Anesthetic Benzocaine on the Human Erythrocyte Membrane and Molecular Models, Biophys. Chem. (2004), 109, 189-199

6 of the erythrocytes. In fact, the erythrocytes underwent a morphological alteration as they changed their discoid shape (Fig. 3A) to spheroechinocytes (Fig. 3B). According to the bilayer couple hypothesis [9], the shape changes induced in erythrocytes by foreign molecules are due to differential expansion of their two monolayers. Thus, spiculated shapes (echinocytes) are induced when the added compound is inserted in the outer monolayer, whereas cup shapes (stomatocytes) arise when the compound accumulates in the inner monolayer. The fact that bupivacaine produced spheroechinocytes would indicate that the anesthetic was located in the outer moiety of the red cell membrane. In conclusion, the experimental results indicate that bupivacaine interacts with the human erythrocyte membrane, the anesthetic being located in the outer moiety of the red cell membrane. The X-ray experiments, performed on bilayers made up of classes of the major phospholipids present in either the outer and inner sides of the erythrocyte membrane, confirmed this result. In fact, they showed that 1 mM bupivacaine slightly disordered the polar head region of DMPC (major class of lipid present in the outer monolayer of the erythrocyte membrane) and 8 mM completely perturbed it, whereas 23 mM bupivacaine produced negligible effects on DMPE (which preferentially locates in the erythrocyte inner monolayer). REFERENCES [1] Membrane Physiology, T.E. Andreoli, J. F. Hoffman, D.D. Fanestil, eds., Plenum, 1980. [2] In Search of a New Biomembrane Model, O.G. Mouritsen, O.S. Anderen, eds., Munksgaard, 1998 [3] J.M. Boon, B.D. Smith, Chemical Control of Phospholipid Distribution Across Bilayer Membranes, Med. Res. Rev. 22 (2002), 251-281. [4] M. Suwalsky, C. Schneider, F. Villena, B. Norris, H. Cárdenas, F. Cuevas, C.P. Sotomayor, Structural Effects of the Local Anesthetic Bupivacaine Hydrochloride on the Human Erythrocyte Membrane and Molecular Models, Blood Cells, Mol. Dis. (2002), 29, 14-23. [5] M. Suwalsky, C. Schneider, F. Villena, B. Norris, H. Cárdenas, F. Cuevas, C.P. Sotomayor, Effects of the Local Anesthetic Benzocaine on the Human Erythrocyte Membrane and Molecular Models, Biophys. Chem. (2004), 109, 189-199. version date: 1 December 2006

version date 1 December 2006 [6] M. Suwalsky, I. Sanchez, M. Bagnara, C P. Sotomayor, Interaction of Antiarrhythmic Drugs with Model Membranes, Biochim. Biophys. Acta(1994)1195, 189-196 [7 M. Suwalsky, P. Hernandez, F. Villena, F. Aguilar, C.P. Sotomayor, The Anticancer Drug Adriamycin Interacts with the Human Erythrocyte Membrane, Z Naturforsch (1999,54c,271-277 8 M. Suwalsky, P. Hernandez, F. Villena, C P. Sotomayor, The Anticancer Drug Chlorambucil Interacts with the Human Erythrocyte Membrane and Model Phospholipid Bilayers, Z. Naturforsch. (1999)54c, 1089-1095 [9 M.P. Sheetz, S.J. Singer. Biological Membranes as Bilayer Couples. A Molecular Mechanism of Drug- Erythrocyte Induced Interactions. Proc. Natl. Acad. Sci. USA (1974)71,4457-4461 Dr. Mario Suwalsky nsuwalsk@udec cl High standards in safety measures should be maintained in all work carried out in Medicinal Chemistry laboratories The handling of electrical instruments, heating eleme glass materials dissolvents and other inflammable materials does not present a problem if the supervisor s instructions are carefully followed This document has been supervised by Prof. Mario Suwalsk (msuwalsk@udec cl) who has informed that no special risk (regarding toxicity, inflammability, explosions), outside of the standard risks pertaining to a Medicinal Chemistry laboratory exist when performing this exercise If your exercise involves any special risks, please inform the editor

7 [6] M. Suwalsky, I. Sánchez, M. Bagnara, C.P. Sotomayor, Interaction of Antiarrhythmic Drugs with Model Membranes, Biochim. Biophys. Acta (1994) 1195, 189-196. [7] M. Suwalsky, P. Hernández, F. Villena, F. Aguilar, C.P. Sotomayor, The Anticancer Drug Adriamycin Interacts with the Human Erythrocyte Membrane, Z. Naturforsch. (1999), 54c, 271-277. [8] M. Suwalsky, P. Hernández, F. Villena, C.P. Sotomayor, The Anticancer Drug Chlorambucil Interacts with the Human Erythrocyte Membrane and Model Phospholipid Bilayers, Z. Naturforsch. (1999) 54c, 1089-1095. [9] M.P. Sheetz, S.J. Singer. Biological Membranes as Bilayer Couples. A Molecular Mechanism of Drug-Erythrocyte Induced Interactions. Proc. Natl. Acad. Sci. USA (1974) 71, 4457-4461. Dr. Mario Suwalsky msuwalsk@udec.cl. High standards in safety measures should be maintained in all work carried out in Medicinal Chemistry Laboratories. The handling of electrical instruments, heating elements, glass materials, dissolvents and other inflammable materials does not present a problem if the supervisor’s instructions are carefully followed. This document has been supervised by Prof. Mario Suwalsky (msuwalsk@udec.cl) who has informed that no special risk (regarding toxicity, inflammability, explosions), outside of the standard risks pertaining to a Medicinal Chemistry laboratory exist when performing this exercise. If your exercise involves any “special” risks, please inform the editor. version date: 1 December 2006