ny-yns'niwufluh Jag“; .., iv v anak. \TT‘TTxiTTwT T\T\T\\T\\\\\\\\\T\T This is to certify that the dissertation entitled L-GLUTAMATE INDUCED CONTRACTIONS IN ISOLATED SCHISTOSOMA MANSON! FIBERS: EVIDENCE FOR A GLUTAMATE TRANSPORTER presented by Cynt/u'a <5ng ”75/4:- has been accepted towards fulfillment of the requirements for PhD. degree in Pharmacologl & Toxicology Major professor . Date Mg! 1. 1996 MS U is an Affirmative Action/Equal Opportunity Institution 0- 12771 LIERARY Michigan state University PLACE II RETURN BOX to move this chockom from your record. TO AVOID FINES Mum on or before date duo. DATE DUE DATE DUE DATE DUE * usu iaAnAifim-mo ActionIEquI oppomnmmuion Wan-9.1 L-GLUTAMATE INDUCED CONTRACTIONS IN ISOLATED SCHISTOSOMA WSONI MUSCLE FIBERS: EVIDENCE FOR A GLUTAMATE TRANSPORTER by Cynthia Lynn Miller A DISSERTATION Submitted to Michigan State University in partial fulfillment of requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pharmacology and Toxicology 1996 ABSTRACT L-GLUTAMATE INDUCED CONTRACTIONS IN ISOLATED SCHISTOSOMA MANSONI MUSCLE FIBERS: EVIDENCE FOR A GLUTAMATE TRANSPORTER by Cynthia Lynn Miller Schistosoma mansoni muscle fibers contracted in response to L- glutamate in a dose-dependent manner (1045-IO‘3M). D-Glutamate, L-aspartate and D-aspartate also caused contraction of the fibers. The glutamate receptor agonists NMDA, ibotenate, kainate, AMPA, quisqualate, ACPD, and L-AP4 produced little or no contraction at concentrations as high as 1 mM. The glutamate receptor antagonists, MK-SOI, CNQX, AP-S, and MCPG, did not block glutamate responses. However, other amino acids, L-aspartate, L-cysteate, and cysteine sulfinate, were found to elicit contraction of the muscle fibers. Contraction induced by L-glutamatc is dependent on extracellular Ca” and is blocked by the voltage- gated Ca” channel blocker nicardipine (10 and 1 HM). [3H]-L-Glutamate, incubated with the muscle fiber preparation, was taken up in a dose-dependent manner, which is also time- and temperature-dependent. Both the L-glutamate induced contractile response of the fibers and [3H]-L- glutamate uptake are Na*- dependent, and can be blocked by specific inhibitors of the hi gh-affinity transporter, DL-threo-B-hydroxyaspartate, and L-trans-pyrollidine-2,4- dicarboxylic acid (TI-IA, PDC). This pharmacology suggests that there may be an electrogenic glutamate transporter on the muscle fibers. It is possible that the electrogenic nature of the transporter is causing the fiber membrane to depolarize, thereby opening voltage- gated Ca‘“+ channels, and raising intracellular Ca++ concentrations leading to contraction. This experimental evidence supports the hypothesis that there is a Na+-dependent hi gh-affinity glutamate transporter on the schistosome muscle membrane. To my David iv ACKNOWLEDGMENTS I would like to express my appreciation to Dr. Ralph A. Pax (Department of Zoology) for providing me the opportunity to work in his laboratory. I thank you for the time, patience and advice you provided me over the course of my graduate career. A special thanks to Dr. James Bennett (Department of Pharmacology & Toxicology) my major professor. I would also like to thank Dr. James Galligan (Department of Pharmacology & Toxicology) for his unwavering confidence in me as a scientist, I will not soon forget your support. In addition, I would like to express my thanks to Dr. Peggy Contreras for her support of myself and the graduate students as a whole. Best wishes for the future to my lab mates Kati Loeffler, Tim Day, Eunjoon Kim, and Ming Tien. I also would like to acknowledge contributions made by Mary Thomas, George Chen, Helen Cirrito and Mary Lou Pax. TABLE OF CONTENTS LIST OF TABLES ................................................ viii LIST OF FIGURES ................................................ ix ABBREVIATIONS ................................................ xi INTRODUCTION ................................................. 1 I. Glutamate Background ...................................... 1 A. Mammalian Glutamate Receptor Subtypes ................ 2 B. Mammalian Glutamate hi gh-Affinity NaT-Dependent Transport 7 C. Glutamate and Invertebrate Muscle ..................... 18 D. Glutamate in Flatworms .............................. 20 OBJECTIVES ................................................... 25 MATERIALS AND METHODS ..................................... 27 1. Muscle Fiber Isolation Procedure ............................. 27 II. Microperfusion Procedure .................................. 28 III. Glutamate Uptake Experiments ............................. 31 RESULTS ...................................................... 34 I. Microperfusion Experiments ................................ 34 A. L-Glutamate Dose Response Curve ..................... 34 B. Stereospecificity .................................... 37 C. Glutamate Receptor Pharmacology ..................... 4O 1. Agonists ..................................... 40 2. Antagonists .................................. 43 D. Hi gh-Affinity Glutamate Transporter Pharmacology ....... 47 l. Transporter Substrates .......................... 47 2. Transport Inhibitors ............................ 50 E. NET-Dependence .................................... 53 F. CaH-Dependence ................................... 59 vi II. Glutamate Uptake Experiments ............................. 65 A. Time-Dependence .................................. 65 B. Dose-Dependence .................................. 65 C. Temperature-Dependence ............................ 70 D. Transport Pharmacology--Transport Inhibitors ............ 73 E. Na*-Dependence .................................... 76 DISCUSSION ................................................... 82 I. Pharmacology ............................................ 82 A. Glutamate ......................................... 82 B. Stereospecificity .................................... 85 C. Receptor Agonists .................................. 86 D. Receptor antagonists ................................ 87 E. Transporter substrates ............................... 88 F. Transport Inhibitors ................................. 90 II. Ion Specificity ........................................... 91 A. Na*-Dependence ................................... 91 B. CaH-Dependence ................................... 93 III. Glutamate Transport ..................................... 94 IV. Transporter Subtype ..................................... 95 V. Transporter Function ...................................... 97 A. Modulation of Membrane Potential ..................... 97 B. Glial Cell Theory ................................... 99 C. Post-Junctional Transport ........................... 100 D. Metabolic Theory SUMMARY .................................................... 105 BIBLIOGRAPHY ............................................... 107 vii IABLEE Table 1. Table 2. Table 3. Table 4. Table 5. LIST OF TABLES The mammalian glutamate receptor subtypes according to pharmacological characterization. . EAAC- like high-affinity Na dependent glutamate transporter clones. . . GLAST-like high-affinity Na dependent glutamate transporter clones. . GLT-like high-affinity NaTdependent glutamate transporter clones. . . Media employed in the microperfusion and glutamate uptake experiments. . . viii ll 14 16 3O Figure 1. Figure 2. Figure 3. Figure 4. Figure 5. Figure 6. Figure 7. Figure 8. LIST OF FIGURES Schistosoma mansoni muscle fibers contract in response to microperfused L-glutamate in a dose- dependent manner. . The enantiomers of glutamate and aspartate elicit different percentages of contraction in Schistosoma mansoni muscle fibers. Relative ineffectiveness of mammalian glutamate- receptor agonists in eliciting contractions in S. mansoni muscle fibers. The glutamate receptor antagonists tested do not inhibit the contraction produced by microperfusion of 100 ,uM L-glutamate. Several amino acids could elicit contractions of S. mansoni muscle fibers. Two inhibitors of the mammalian excitatory amino acid transporter were effective at reducing the percentage of fibers contracting in response to 100 uM L- glutamate. . . Replacing Na+ with N-methyl-D-glucamine diminished the percentage of fibers contracting in response to microperfusion of 1 mM L-glutamate. By replacing Na+ with Li“, fewer fibers contracted in response to microperfusion with 100 ,uM L-glutamate. . ix HALTER 35 39 42 45 49 52 56 58 LIST OF FIGURES (cont’d) Figure 9. Figure 10. Figure 11. Figure 12. Figure 13. Figure 14. Figure 15. Figure 16. Figure 17. The contraction elicited by L-glutamate is Ca”- dependent. [3H]-L-Glutamate is taken up in a time-dependent manner. . [3H]-L-Glutamate is transported in a dose-dependent manner. [3H]-L-Glutamate transport is temperature-dependent. . trans-Pyrollidine-Z,4-dicarboxylic acid (PDC), a high-affinity glutamate transport inhibitor, inhibits uptake of [3H]-L-glutarnate in a dose-dependent manner. The addition of 100 uM PDC or THA inhibits the transport of [3H]-L-glutamate. . . [3H]-L-Glutamate uptake is Na+ dependent. Proposed mechanistic model of the L-glutamate induced contraction is the isolated S. mansoni isolated muscle fiber. Proposed metabolic pathways in S. mansoni. 61 67 69 72 75 78 80 84 103 ACPD AMPA L-AP-4 D-AP-S CA CNQX CSA DHK DMEM EAAC EAAT EDTA EGTA GLAST GLT HEPES HCA I-DIVIEM MCPG MK-80 l NMDA PDC ABBREVIATIONS DL-a-Aminoadipic acid trans-(iH-Amino-1,3-cyclopentanedicarboxylic acid (:t)-a-Amino-3-hydroxy-5-methylisoxazole-4-propionic acid hydrobromide L-amino-4-phosphonobutanoate D-amino-S-phosphonopentanoate Cysteaic acid 6-Cyano-7-nitroquinoxalinc-2,3-dione Cysteine sulfinic acid Dihydrokainic acid Dulbecco's Modified Eagle's Medium Excititory amino acid carrier Excititory amino acid transporter Ethylenediamine tetraacetic acid Ethyleneglycol-bis-(B-aminoethyl ether)N,N,N,N'- tetraacetic acid L-Glutamate/L-aspartate transporter L-Glutamate transporter 4-(2-hydroxyethyl)-l-piperazine ethanesulfonic acid L-Homocysteic acid Inorganic Dulbecco's Modified Eagle's Medium Kainic acid a-methyl-4~carboxyphenylglycine (5R,IOS)-(+)-5-Methyl-10,l 1-dihydro-5H-dibenzo[a,d]cyclohepten- 5,10-imine hydrogen maleate N-methyl-D—aspartic acid L-trans-pyrollidine-Z,4-dicarboxylic acid DL-threo-B-hydroxyaspartic acid INTRODUCTION 1. Glutamate is an Excitatory Chemical Messenger L-Glutamate is the principle excitatory neurotransmitter in the mammalian central nervous system (Hediger et al., 1995). In the past it was thought that an amino acid widely involved in protein synthesis and metabolism could not also fimction as a neurotransmitter. Now it is known that glutamate meets the criteria set for a neurotransmitter candidate. Glutamate is stored in synaptic vesicles, and it’s release is CaH-dependent. Glutamate binds receptor subtypes with hi gh- affinity, and termination of its action is thought to occur by high-affinity transport out of the synapse (Chamberlin & Bridges, 1993). Glutamate neurotransmission is responsible for a broad spectrum of activities, ranging from neuronal plasticity to neurotoxicity. Long-term potentiation (LTP) is a process implicated in learning and memory acquisition (Seeburg, 1993). LTP is characterized by a sustained increase in synaptic efficacy (Riedel & Reymann, 1993), in which various glutamate receptors play a central role. Glutamate has been implicated in several pathogenic processes. Over- stimulation of glutamate receptors, most notably the Ca++ permeable N-methyl-D- l 2 aspartate (NMDA) receptor, has been shown to lead to neuronal degeneration during ischemia (Nakanishi, 1992). When rat astrocytes are placed in a hypoxic environment, glutamate uptake is reduced by 35-45% (Swanson et al., 1995). It is speculated that glutamate is also involved in several neurodegenerative diseases. Amyotrophic lateral sclerosis is a neurode generative disorder characterized by death of motor neurons. It is thought to be associated with abnormal metabolism of glutamate involving the hi gh-affinity glutamate transporter (Rothstein et al., 1992). In the brains of humans with Alzheimer disease, it appears that different cortical regions have distinct glutamate transporter pharmacology when compared to normal controls at autopsy (Scott et al., 1995). The study of glutamate neurotransmission and the involvement of glutamate in neurode generative disease are rapidly-growing fields. A. Mammalian Glutamate Receptor Subtypes The mammalian glutamate receptors have been thoroughly characterized, both pharmacologically and by molecular biological techniques. The mammalian glutamate receptors are more recently referred to as the excitatory amino acid receptors, and can be divided into two broad categories, the ionotropic and metabotropic receptors (Table l). The ionotropic receptors are multimeric and contain an intrinsic cation-specific ion channel (Seeburg, 1993), while Table l. The mammalian glutamate receptor subtypes according to their pharmacological characterization. Those agonists which are underlined were employed in the characterization of the glutamate contractile response of the isolated Schistosome muscle fiber. Although this categorization is oversimplified and incomplete, it provides a basic outline of the mammalian glutamate receptor subtypes. ACPD, trans-(3:)-l-Amino—1,3-cyclopentanedicarboxylic acid; AMPA, (:t)-a-Amino-3-hydroxy-5-methylisoxazole-4-propionic acid hydrobromide; L- AP-4, L-amino-4-phosphonobutanoate; AP-5, D-amino-5-phosphonopentanoate; CNQX, 6-Cyano-7-nitroquinoxaline-2,3-dione; DCG-IV, (25,1'R,2'R,3'R)-2-(2',3'- dicarboxycyclopropyl)glycine; DI-IPG, 3,5-dihydropheylglycine; DNQX, 6,7- dintr'oquinoxaline-2,3-dione; MAP4, a-methyl-L-AP4; MCCG 25,1's,2's-2-methyl- 2-(2'carboxycylopropyl)glycine; MCPG, a-methyl-4-carboxyphenylglycine; MK- 801, (5R,IOS)-(+)—5-Methyl-10, 1 1-dihydro-5H-dibenzo[a,d]cyclohepten-5, 10- mine hydrogen maleate; NBQX, 6-nitro-7-sulphamobenzo(f)quinoxaline—2,3- dione; NMDA, N-methyl-D-aspartic acid. IONOTROPIC METABOTROPIC NMDA AMPA Kainate L-AP4 ACPD Gene nmdal glul-glu4 glu5—glu7, mglu4, mglul-mglu7 nmda2A-2D kal, ka2 mg1u6, mglu7 Agonbts NMDA AMPA Kainat: EARS A9212 Animate Wm Domoate A9212 DCG-IV mm DHPG LcAPA Antagonists APfi CNQX CNQX MAP4 MCPG MK:8.Q1 DNQX DNQX MAP4 PCP NBQX MCCG Mechanism ion channel ion channel ion channel Gi, (-) AD Gi, (—) AD Ca+*,Na*.K* NaflK“ Na”,K+ Gs (+) AD GQ9 (+) IP3 G (+) AA G (+) PLD 5 metabotropic receptors are coupled to G proteins, and contain seven membrane spanning regions and modulate intracellular messengers (Schoepp & Conn, 1993). The ionotropic receptor subtypes include the NIVHDA and (:t)-cc-amino-3- hydroxy-S-methylisoxazole-4-propionic acid hydrobromide (AMPA)/kainate sensitive receptors. N-Methyl-D-aspartate (NMDA) selectively gates an intrinsic cationic channel that has a 10:1 preference for Ca++ to Na+ or K”. This Ca” permeability is implicated in both LTP and neurotoxicity (Monaghan et al., 1989). The NMDA receptor has several unique properties, including voltage-dependent block by Mg“ (Sprengel & Seeburg, 1993). The NMDA receptor cation channel is open when glutamate is bound, however at negative potentials, Mg” blocks current flow by binding inside the channel. The functional consequence of the Mg“ block is that the membrane in which the NMDA receptor resides must be depolarized for Mg++ to be dislodged and current to flow through the channel. This provides an important method of regulating the predominately Ca“ current. In addition, NMDA receptors are modulated by glycine, and without glycine the channel will not open even when glutamate is bound (Barnard, 1992). It has been found that the NMDA subunit composition and alternative splicing are responsible for altering ion specificity, size of the current, degree of susceptibility to Mg++ blockade, ability of glycine to stimulate, and affinity for agonists (Nakanishi, 1992). 6 The ionotropic receptor subtypes, alpha-amino-3 -hydroxy-5-methyl-4- isoxazole propionic acid (AMPA) and kainate, were first described pharmacologically (Monaghan et al., 1989). It is now known that their sensitivity to agonists, and their electrical properties are derived from their molecular subunit composition (Seeburg, 1993). AMPA and kainate subtmits may combine with themselves, or with each other, to form unique receptors regarding pharmacological profile, ion selectivity and kinetic parameters (Seeburg, 1993). In addition to subunit composition, alternate splicing and RNA editing both play a role to further diversify these receptors (Nakanishi, 1992). Both AMPA and kainate receptor channels are predominately permeable to Na” and K+ (Sprengel & Seeburg, 1993). AMPA receptor subtypes display fast kinetics, and are responsible for the majority of fast excitatory neurotransmission in the mammalian central nervous system (Seeburg, 1993). Usually AMPA type receptors are characterized by low Ca++ permeability, however, subunit assembly, RNA editing, and alternative splicing can alter the exact Ca“ permeability of these channels (Mishina et al., 1991). Metabotropic glutamate receptor subtypes produce their effects through the activation of G-proteins and subsequent modulation of intracellular messengers such as inositol 1,4,5-trisphosphate, cyclic adenosine monophosphate (CAMP), and phosphlipase D (PLD) (Schoepp & Conn, 1993; Boss & Conn, 1992). Recently 7 specific agonists and antagonists have become available to aid in characterizing the metabotropic receptors. At least seven metabotropic glutamate receptors have been cloned (Winder & Conn, 1995). B. Mammalian Glutamate High-Affinity Na+-Dependent Transport The hi gh-affinity glutamate transporter is thought to terminate the action of glutamate in the CNS by rapidly removing it from the synapse (Hedi ger et al., 1995). This makes the transporter crucial for normal synaptic firnction. These high-affinity transporters have also been implicated in such processes as excitotoxicity, epilepsy, and neurodegenerative diseases. The first studies of glutamate transport employed synaptosomes and epithelial membrane vesicles (Kanai et al. , 1993). From this work, two types of glutamate transporters were characterized, the high-affinity (Km=2 to 50 uM), and the low-affinity transporter (Km>100 ,uM). Because hi gh-affinity glutamate transporters are thought to be involved in termination of glutamate fast excitatory synaptic transmission, the majority of information collected to date, concerns the hi gh-afiinity uptake systems. Historically, amino acid transporters have been characterized by substrate specificity and ionic dependence. Initial pharmacological studies in the vertebrate CNS indicated that there is probably more than one type of hi gh-afiinity 8 transporter for excitatory amino acids (Kanai et al., 1993). Recently, several groups set out to clone the glutamate transporters (Bouvier etal., 1994). However, traditional cloning strategies proved unsuccessful, because they relied on sequence similarity to other cloned transporters which belong to the superfamily which includes GABA, glycine, norepinephrine, dopamine, and serotonin transporters (Kanai et al., 1993). In 1992 it was independently discovered, by several groups, that the glutamate transporters do not belong in this superfamily of transporters, but are part of a new family, which includes gltP a Na+-independent glutamate-proton transporter fi'om E. coli, gltT the Na+/proton glutamate transporter from B. stearothermophilus, dctA the C4-dicarboxylate transporter, and ASCTl which is a human neutral amino acid transporter (Arriza et al., 1993; Bouvier et al., 1994). This new family of transporters is diverse in tissue distribution, kinetics, and pharmacological profile (Bouvier et al., 1994). The hi gh-affinity glutamate transporters are electrogenic. It is currently hypothesized that two Na“ ions are co-transported with each molecule of glutamate and that one K” and one OH' are countertransported (Bouvier et al. , 1992; Kanai et al., 1993). However unpublished results from the laboratory of M. Kavanaugh reveal that there may be three Na" ions co-transported with each molecule of glutamate (personal communication). Additional studies of the glutamate 9 transporters have shown that certain subtypes also mediate a chloride current (W adiche et al., 1995a). Regardless of the exact stoichiometry, inward transport of glutamate will produce a depolarizing current. Presently the mammalian glutamate transporters can be grouped into three categories, the EAAC-like transporters, the GLAST-like transporters, and the GLT-like transporters. These categories have been named according to the name of the gene first cloned for each subtype. This categorization of the cloned hi gh- affinity glutamate transporters is most likely an oversimplification of the actual number of transporter categories that exist (Arriza et al., 1994). As more transporters are cloned, the relationships between the subtypes will become more clear. The first EAAC-like transporter to be cloned was from rabbit small intestine (EAACl) using Xenopus oocyte expression methods (Table 2) (Kanai & Hediger, 1992). From this sequence, the analogous human and mouse genes have been cloned using PCR techniques (EAAT3 and MEAACI). The EAAC-like transporters are 523-525 amino acids in length, are all found in the neuronal tissue in the brain, and have Km values ranging fi'om 12 to 28 uM for glutamate. It is difficult to compare the pharmacology of the transporters, because several different methods have been used to generate the pharmacological profiles. In general, the EAAC-like transporters appear to transport L- glutamate, and L-and 10 Table 2. The EAAC-like high-affinity glutamate transporter clones. EAAT3, the human transporter, has 92% identity with the rabbit transporter EAAC 1, and MEAACI, the mouse transporter, has 89.3% identity with EAAC l. AAD, arninoadipate; L-asp, L-aspartate; D-asp, D-aspartate; CA, cysteate; CSA, cysteine sulfinate; DHK, dihydrokainate; EAAC, excitatory amino acid carrier; EAAT, excitatory amino acid transporter; L-glut, L-glutamate; D-glut, D-glutamate; HCA, homocysteate, KA, kainate; PDC, trans-pyrollidine-2,4—decarboxylate, THA, DL- threo-B-hydroxyaspartate 11 EAAC-like clone EAAC] (rabbit) EAAT3 (human) MEAACl (mouse) immunohisto— brain (neurons) brain brain localization intestine kidney kidney kidney placenta lung liver lung muscle heart muscle size (amino acids) 524 525 523 L-glutamate K1n = 12 12M K“I = 24i2 11M na oocyte (current) cos cell (transport) KIn = 28:1:6 uM oocyte (current) other agonists mm W na W Knxalnes THA=6.9/.4M THA=3 7:1:1 1.4M DHK>lmM DHK (no current) AAD=201 12M KA (no current) PDC=27i5 [2M L-asp=24i2 uM D-asp=47t8 [2M D-glut =1 .78mM antagonists 195W Whit na Lzalmamate 11311512911 91mm PDC 61:14 11M THA 7.1;.1M THA 25:15 [1M DHK >lmM DHK > 3mM AAD 165/.1M KA > 3mM CA 19i9 CSA 17i2 references Kauai & Hediger, Arriza et aI. , 1994; Freund et al., 1995 1992 Shashidharan et al. , 1994; Wadiche et aL,1995b 12 D-aspartate, and both cysteate and cysteine sulfinate can block the uptake of L- [3H]glutamate. The specific transport inhibitors trans-pyrollidine-2,4- decarboxylate (PDC), and DL-threo-B-hydroxyaspartate (THA) block [3H]-L- glutamate uptake and inhibit the glutamate-induced depolarizing current in patch clamped oocytes injected with EAACl message (Kanai & Hediger, 1992; Wadiche et al., 1995b). However, dihydrokainate (DHK), kainate, and aminoadipate (AAD) were not effective inhibitors. The first GLAST-like transporter to be cloned was from a rat cDNA library which was screened with an oligonucleotide derived from the partial sequence of a purified protein responsible for transporter activity (Storck et al., 1992). Based on this sequence, Inoue et a1. (1995) cloned a transporter from the bovine retina, BNGLUAS; Tanaka (1993) cloned a transporter from mouse brain, mGLuT-l; and Arriza et al. (1994) cloned a human GLAST-like transporter, EAATl. These transporter proteins are 542-543 amino acids in length and the K"1 values range from 20 to 77 11M (Table 3). D-glutamate was not a good substrate or inhibitor of the GLAST-like transporters. In general, DHK and kainic acid were not able to block the GLAST-like transporters, with the exception of the transporter cloned from the mouse (MGlut-l) (Tanaka, 1993). GLT-like transporters have been cloned from the rat, human and mouse. The rat transporter, GLT-l, was the first to be cloned by Kanner et al. (1992). 13 Table 3. The GLAST-like high-affinity glutamate transporter clones. The GLAST-like transporters have sequence similarity with the prokaryotic glutamate transporters GTLP (Kanai et al., 1993). AAD, aminoadipate; L-asp, L-aspartate; D-asp, D-aspartate; CA, cysteate; CSA, cysteine sulfinate; DHK, dihydrokainate; GLAST, L-glutamate/L-aspartate transporter; L- glut, L-glutarnate; D- glut, D- glutamate; HCA, homocysteate, KA, kainate; PDC, trans-pyrollidine-2,4- decarboxylate, THA, DL-threo-B-hydroxyaspartate 14 GLAST-like clone GLAST BNGLUAS MGLuT-l EAAT] (rat) (cow) (mouse) (human) immunohisto- brain retina brain brain localization glial cells 11a lung heart (bergrnann) muscle placenta spleen muscle testes size 543 542 543 542 (amino acids) L-glutamate K, = 77uM K... = 38.1il4uM K... = 72 11M K... = 20 1.1M oocyte (current) oocyte (current) oocyte oocyte (current) K,n = 12 11M (transport) K, = 4811mm! oocyte (current) cos-7 (transport) Pharmacology KW W W W in: glutamate cfluMlfl: W glutamate 929W glutamate W W L-glut 701M naming QJmMmld CA lOfluM L-aSP 651M LQQuMJnhibitct substrate CSA 14:71:14 CSA 8014M THA 97% L-glut 77t8% THA 32i8uM THA 65pM CA 87% D-glut 30% PDC 79i7pM HCA 2.7mM L-glut 80% L-asp 75i9% DHK >3mM KA 3mM L-asp 70% D-asp 66=l:5% KA >3mM DHK 3.1mM PDC 41% AAD IOmM D-asp 8% KA 5% “Allisnlammt mm W Wham WW (current) mam W L-asp 16=tluM THA 9mM=90% MM D—asp 23t2uM DHK 35:1:5% D-glut THA 57t4% 595:1:50uM THA 33:1:3uM PDC 28:1:2uM KA no current DHK no current references Storck et al., Inoue et al., 1995 Tanaka, 1993 Arriza et al., 1992; Klockner et 1994; Wadiche et al., 1993, 1994; al., 1995b Tanaka, 1994 15 Table 4. The GLT-like transporters. AAD, aminoadipate; L-asp, L-aspartate; D-asp, D-aspartate; CA, cysteate; CSA, cysteine sulfinate; DHK, dihydrokainate; GLT, glutamate transporter; L- glut, L- glutamate; D- glut, D- glutamate; HCA, homocysteate, KA, kainate; PDC, trans-pyrollidine-Z,4-decarboxylate, THA, DL- threa- B-hydroxyaspartate 16 GLT-like clone GLT-l (rat) EAAT2 (human) mGLT-l (mouse) immunohisto- brain brain brain localization astrocytes placenta size 573 574 572 (amino acids) L-glutamate K“1 = 2 uM K,n = 971-4 uM hela cells cos cell (transport) (transport) Km = 18:3 oocyte (current) Phafinacology Ksl’lflslutamate Enigma transport L-asp 7*] ,uM L-asp Ki=0.2uM D-asp 13:1:1uM D-asp Ki=0.6 uM D-glut 5.4:L-0.4 mM CSA Ki=1.7uM PDC 7d:0 uM PDC Ki=0.73uM THA 10:1:1 uM THA Ki=1.0 uM o . . . . . ' . _ WIDE II'I'I' Wm I AAD=81% KA=59il8pM DHK = 97% DHK = 23:1:6 EM PDC = 8:2 ,uM CA = 10:1:2 11M CSA = 611 references Pines et al., 1992 Arriza et al., 1994; Freund et al., 1995 Wadiche et al. , 1995b 17 This group used an antibody to the purified glial transporter protein to screen a An, library from rat brain. Again, using an oligonucleotide based on sequence similarity, the human EAAT2, and mouse mGLT-l transporters were cloned (Arriza et al., 1994; Freund et al., 1995). The GLT-like transporters are 572-574 amino acids in length, and the K, values for glutamate range from 2 to 97 ,uM (Table 4). The GLT-like transporters are pharmacologically different fiom the other transporters because DHK inhibits glutamate transport and is itself transported. Also, AAD inhibits glutamate transport in GLT-1. Otherwise, the pharmacology is quite similar to the previously-described glutamate transporter subtypes. Although several of the transporters have been irnmunohistolocalized to tissues other than the brain (including intestine, kidney, heart, and muscle), little is known regarding their physiological function in these tissues. Hediger et. al. (1995), suggest that the transporters found in the intestine and kidney are involved in trans-epithelial glutamate transport. Glutamate transporters are also thought to be involved in cellular amino acid nitrogen metabolism (Arriza et al., 1994). Presumably, glutamate transporters could also serve to transport glutamate in to cells to be incorporated into protein. While glutamate receptors and transporters have been widely studied in the mammalian central nervous system, less information is available about glutamate receptors or transporters in invertebrates. 18 C. Glutamate and Invertebrate Muscle Glutamate is hypothesized to be a neurotransmitter in several invertebrates. It has been suggested that glutamate is a primordial neurotransmitter, which developed as a signal molecule before the evolution of specialized neurotransmitters (Schuster et al., 1991). In the primitive Phylum Coelenterata, glutamate is the most abundant fiee amino acid in the sphincter muscle of the sea anemone Actina equina, and glutamate inhibits electrically-induced contraction of isolated sphincter muscle preparations (Carlyle, 1974; Walker & Holden-Dye, 1989). This work was done before mammalian glutamate receptor subtypes had been well-characterized, and the authors were hesitant to suggest that glutamate was functioning as a neurotransmitter. Glutamate receptors are present on crayfish muscle at the neuromuscular junction, and glutamate serves as the excitatory neuromuscular transmitter in these animals (Takeuchi & Takeuchi, 1964). Solubilization and purification of a glutamate receptor from crustacean muscle yielded a protein which binds glutamate and quisqualate with high affinity (Gray et al., 1991). Outside-out patches of crayfish muscle revealed glutamate-activated cation channels (Dudel et al., 1990). In addition, Shinozaki et al. described an excitatory glutamate receptor that is found extra-junctionally on the crayfish muscle, and that is sensitive to kainate (Shinozaki & Ishida, 1992). This same group also characterized a 19 presynaptic metabotropic glutamate receptor which is sensitive to 28,3 S,4S-2- (carboxycyclopropyl) glycine (L-CCG- 1 ). Insect muscle membrane also contains glutamate receptors. Locust muscle membrane has glutamate-sensitive currents, and patch clamp studies by MacDonald et al., revealed the presence of a quisqualate-sensitive receptor (Cull- Candy & Parker, 1982; Bates et al., 1990; MacDonald et al., 1992). A glutamate receptor subunit from Drosophila muscle has now been cloned, and it exhibits sequence sirrrilarity with ionotropic mammalian glutamate receptor subunits (Schuster et al., 1991). Glutamate and aspartate elicit a current across the recombinant protein expressed in Xenopus oocytes. Based on sequence divergence, the authors concluded that this protein is an evolutionarilly-distant subtype of excitatory glutamate receptor. In the smooth muscle of Aplysia anterior aorta, glutamate is a potential excitatory neurotransmitter (Sawada et al., 1984). L-glutamate depolarizes the muscle and this response is blocked by 2-APB (a glutamate receptor antagonist) and mimicked by L-aspartate. This glutamate effect was not modulated by glycine. It is difficult to relate information about possible invertebrate glutamate subtypes to the well-characterized mammalian glutamate receptor subtypes, because there is little evidence that invertebrate glutamate receptors have the same 20 pharmacological profiles or physiological roles as their mammalian counterparts. It is necessary to study the invertebrate glutamate-induced effects without the bias of mammalian subtypes. D. Glutamate in Flatworms Within the phylum platyhelminthes, glutamate has been hypothesized to be an excitatory neurotransmitter. Glutamate is the most abundant amino acid in S. mansoni proteins and is also an abundant fiee amino acid, second only to alanine (Chappell, & Walker, 1982; Webb, 1986). Several techniques have been used to visualize areas of concentration of the free amino acid glutamate. Immunocytocherrrical techniques have been performed on Trichibilharzia ocellata and S. mansoni using fi'eeze-drying-paraformaldehyde fixation. The results showed localization of irnmunoreactivity in both species in the main commissure and longitudinal nerve trunks of the cercaria (Solis-Soto & Brink, 1994). Glutamate-like irnmunoreactivity has also been shown in the longitudinal nerve cords and in sites of sacroneural intervention of muscle in the cestode Hymenolepis diminuta (Webb & Eklove, 1989). Sacroneural innervation is a term used to describe the cestode neural-muscular relationship, where the muscle contains a cytoplasmic extension which contacts the nerve cord (Webb, 1987). Webb and Eklove (1989) used a primary antibody directed toward a glutamate- 21 glutaraldehyde-protein conjugate, and frxed the flatworm tissue with glutaraldehyde. In addition, using histofluorecence methods, Keenan and Koopowitz (1982) have shown the presence of glutamate in the longitudinal nerve cords of Gyrocotyle fimbriata. Because glutamate may play a metabolic role, as well as that of a neurotransmitter, there is commonly high background staining in each of these procedures. Tissues with a high level of metabolic activity may tend to stain more intensely for glutamate (Webb & Eklove, 1989). Therefore, distinguishing between transmitter pools and metabolic pools of glutamate is problematic. Nevertheless, each of these studies suggests that glutamate is relatively concentrated in the longitudinal nerve cord, which in turn supports the hypothesis that L- glutamate is a neurotransmitter in the Platyhelminths. However, this information does not necessarily suggest a physiological role for glutamate. Studies employing in vitro flatworm preparations of selected neuronal tissues have shown that L- glutamate evokes excitatory responses. The primitive cestode G. fimbrr'ata has longitudinal nerve cords that have increased spontaneous activity in response to applied glutamate and aspartate, and this activity can be blocked by 2-arnino-4-phosphonobutyrate (APB), a non-specific glutamate receptor antagonist (Keenan & Koopowitz, 1982). In support of the hypothesis that glutamate is a neurotransmitter in 22 flatworms, Webb et al. demonstrated glutamate hi gh-affmity uptake into flatworm tissue. When [3H]-L-glutamate is incubated with tissue slices of H. diminuta, it is transported into the tissue and can be released by K+ depolarization (Webb, 1988). This release of glutamate is CaH-dependent and is enhanced by the presence of 5- HT. Webb et al. (1986) also measured the kinetics of glutamate uptake in H. diminuta tissue slices and described both a hi gh-affinity glutamate transport system K.=l 8 HM, and a low affinity glutamate transport system K,=220 1.1M. In addition, glutamate is taken up by intact S. mansoni adults through the tegument (Asch & Read, 197 5; Comford & Oldendorf, 1979; Chappell & Walker, 1982; Comford, 1985). Thompson & Mettrick (1989), demonstrated Ca”-dependent stimulated release of glutamate and specific glutamate binding sites. The authors suggested that glutamate is released from nervous tissue and may serve as a neurotransmitter. Early experiments tested putative neuromuscular transmitters on intact flatworms, and the effects on the musculature were measured by force transduction. These methods failed to reveal any effect for glutamate on the schistosome musculature. However, glutamate elicited powerfirl rhythmic contractions when applied to longitudinal muscle preparations of H. diminuta (Thompson & Mettrick, 1989; Webb, 1988). The contractile response of the H. diminuta muscle preparation is concentration-dependent and L-glutamate has a 23 greater effect than D- glutamate. The anatomy of the flatworms makes it difficult to isolate individual tissues. Therefore, these longitudinal muscle preparations most likely contain a variety of tissues, including neuronal tissue. It is therefore not possible to pinpoint the site of action of glutamate to a particular tissue on the basis of these studies. The nricroanatomy of the schistosome musculature was studied by Silk and Spence in 1969. They described unstriated longitudinal, circular, and radial muscle containing both thick and thin myofilaments (Silk & Spence, 1969). The schistosome muscle also contains glycogen granules and poorly-defined sarcoplasmic reticulum. The neuromuscular relationship of the schistosome has not been well-defined. However, in the cestodes, the musculature sends a cytoplasmic arm to contact the nervous system, which has been termed the sarconeural arm (Webb, 1987). This neuromuscular anatomy may also be present in the schistosomes. The schistosomes are parasites that are able to infect humans and cause the disease complex schistosomiasis. The World Health Organization ranks schistosomiasis second only to malaria in terms of socioeconomic importance (World Health Organization, 1995). The adult S. mansoni parasites reside in the mesenteric veins of the host and the female produces approximately 300 eggs per day. The eggs normally pass through the mesenteric veins and into the intestine to 24 be released with the feces. However, 50% of the eggs become trapped in the liver. In the liver the host immune system forms granulomas around the eggs; fibrous tissue replaces these granulomas, and the resulting hepatic scarring leads to portal hypertension, esophageal varices and death. Because several antiparasitic drugs produce marked effects on the schistosome musculature, our laboratory has strived to further understand schistosome neuromuscular physiology. Our laboratory has developed a procedure for isolating individual muscle fibers from the flatworm S. mansoni (Blair et al. , 1991). This preparation permits the direct application of neurotransmitters onto the individual muscle fibers, without other tissues present to confound results. Herein lies the first evidence of an effect produced by glutamate on the S. mansoni muscle fibers. OBJECTIVES The development of the procedure to isolate muscle fibers from the schistosome has changed the way our laboratory has studied the muscle physiology of this flatworm. It is now possible to apply neurotransmitters directly to the muscle fibers without experimental results being confounded by other tissue types. We have found that the excitatory amino acid, L- glutamate, produces contraction of the isolated muscle fibers. The main goal of this study was to characterize the glutamate-induced contraction of isolated Schistosoma mansoni muscle fibers to gain a greater understanding of the underlying mechanism. First, the pharmacology of the contractile response was characterized by using the microperfusion contraction assay. In addition, the ionic dependence of this glutamate-induced contractile response was characterized. Based on the resulting data, it was determined that the glutamate-induced contractile response may be mediated by a high-affinity glutamate transporter. This putative transporter was characterized by measuring radiolabeled glutamate transport into the schistosome muscle fiber preparation. From these experiments the kinetics of the transport were determined. This response was further analyzed 25 26 in terms of ion dependence and transport pharmacology, employing the same tools used to describe the glutamate-induced contractile response in the isolated fibers. This now provides a way to compare the observations of glutamate-induced contraction and radiolabeled glutamate transport, and to understand if the electrogenic transport of glutamate could be responsible for the contraction observed in response to microperfusion of glutamate onto the isolated fi'ayed muscle fibers. MATERIALS AND METHODS 1. Muscle Fiber Isolation Procedure S. mansoni muscle fibers were isolated using a modified version of the procedure previously published (Day et al., 1994a). In short, Puerto Rican strain S. mansoni were surgically removed from mesenteric and portal veins of female ICR mice (Harlan Sprague-Dawley) 40-60 day post-infection. Adult parasites (3 545 pairs) were cut into approximately 2 mm pieces, and suspended in modified Dulbecco's Modified Eagle's Medium (DMEM) at 35-37°C. This medium has been described by Day et al.(1993), and consists of powdered DMEM stock dissolved in water to 67% of it's normal volume with the addition of 2.2 mM CaClz, 2.7 mM MgSO,, 0.04 mM NaZI-IPO4, 61.1 mM glucose, 1.0 mM dithiothreitol (DTT), 1011M serotonin, and 0.1 mg/ml gentarnicin replacing 10 mg/ml Pen-strep (pH 7.4) (Day et al., 1994a). The resultant worm pieces were digested three times at 35-37 °C for 10 nrinutes with gentle agitation in a solution of DMEM that contains 0.75 mg/ml papain, 1 mM EGTA and 1 mM EDTA. The pieces were then rinsed with enzyme-free DMEM containing 0.1% bovine serum albumin (BSA) for 10 minutes, and then washed three times with DMEM. The 27 28 individual fibers were released from the worm pieces by forcing the suspension through a Pasteur pipet approximately 30-60 times. This muscle fiber suspension was plated onto 35 mM petri dishes and left at room temperature for 30 minutes while fibers attached to the surface of the petri dishes. The media of the plated fibers was then replaced with an inorganic version of DMEM (I-DMEM) that contains 82.5 mM Na“, 4.1 mM K‘, 3.6 mM Ca“, 3.3 mM Mg“, 100.4 mM Cl’, 79.9 mM glucose, 15.0 mM 4-(2-hydroxyethyl)-l-piperazine ethanesulfonic acid (HEPES), 1.0 mM DTT, 10 1.1M serotonin, and 0.1 mg/ml gentarnicin (pH 7.4). Fibers were stored at 18 °C in I-DMEM until experimental procedure. This primary preparation was performed on the morning of each day in which data were collected, and the entire procedure takes approximately three and a half hours. 11. Microperfusion Procedure Prior to microperfusion, the schistosome muscle fibers were incubated at 35 °C for ten minutes and kept at 35-37°C throughout the microperfirsion procedure by means of a heated microscope stage. Neurotransmitters and drugs were dissolved in I-DMEM, loaded into borosilicate glass nricropipets (W -P Instruments, New Haven, Conn.) and perfused in a constant manner by applied positive pressure. The individual fibers were exposed to drug solutions by bringing the microperfusion pipet into the field of the fiber and applying the drug 29 solution directly onto the fiber. The data collected were visual observations of muscle fiber contraction. All observations were recorded on VHS tape, by means of a video camera and monitor. (Model CCD72, Dage MTI, Michigan City, IN, USA). A fiber was considered to have contracted if there was a detectable change in fiber length when viewed on the monitor. Each fiber which was microperfused was tallied as either contracting or not contracting in response to the applied drug solution. From these data, percentages of fibers responding to the drug solution were tabulated and averaged with at least three other trials. Only fibers described as the "frayed" type by Blair, et al. (1991), were tested in this study. Frayed fibers are easily distinguishable by their bifurcated endings and average length of 20pm length. Approximately 15-30 fibers in each petri dish were microperfused. Microperfirsion of I-DMEM, which is the same media in which the fibers were bathed, served as a negative control. In some cases, microperfusion of 25 mM K“ served as a positive control, to asses preparation viability. Often microperfusion of 1.0 and/or 0.1 mM L-glutamate was the positive control, depending on the concentration of other agonists or antagonists to be tested. Drugs and neurotransmitters were microperfused at designated concentrations and antagonists were normally added to the fiber's bath before the 37°C ten-minute incubation and remained in the dish through out the experiment. Where indicated, the inhibitors were also added to the microperfirsion pipet at the same concentration. When 3O Table 5. Media employed in the microperfusion and glutamate uptake experiments (mM) modified I-DMEM 25 mM Calcium Sodium DMEM K+ free Free Na‘ 4.1 4.1 25.0 4.1 4.1 Ca2‘ 82.6 82.5 82.5 82.6 - M g2‘ 3.6 3.6 3.6 - 3.6 CI‘ 3.3 3.3 3.3 3.3 3.3 _ 93.7 100.4 16.4 93.2 17.9 SO. , 3.3 - - - - P04 Glucose 0'04 ' ' ' ' Phenol R ed 79.: 79.9 : 79.9 79.9 HEPES ' ' L-Arginine HCl 13.17,) 15.0 15.0 15.0 15.0 L-Cystine 2HC1 0'2 _ - _ _ L-Glutamine 3'0 _ _ _ _ Glycine 0'5 _ _ _ _ L-Histidine HCl ' . 0.2 - - - - L-Isoleucme 0 6 - _ - L-Leucine ' ' . 0.6 - - - - L-Lysme 0 8 - _ _ L-Methionine 0'2 _ - _ : L-Phenylalanine 0'3 - _ _ _ L-Serine 0'7 _ _ _ _ L-Threonine 0' 6 _ _ _ _ L-Tryptophan ' . 0.1 - - - - L-Tyrosme 0 4 - - - _ L-Valine 0' 6 _ _ _ - D-Ca- 3'0 _ _ _ _ Pantothenate _' _ Choline ' ' ’ Chloride 3'3 : I : ' Folic Acid 5'4 - - _ : Myo-Inositol 3'0 _ _ _ - Niacinamide 37.5 - _ _ - Orotic Acid 3'0 _ _ _ _ Pyroxidine-HCI 0'3 - _ - _ Riboflavin ' . . 3.0 - - - - Thramrne Hcl Sucrose - - 65.0 - - mum j 1 101-9 j : N-meflryl-D- gluconate : : _ 0 5 82_'6 EGTA . . _ .. ‘ .. S-HT 10 10 10 10 31 ionic concentrations of the I-DMEM was altered this modified I-DMEM was used both to bathe the fibers and to dissolve the drug to be tested (Table 5). Statistical comparisons were conducted using the two-tailed Mann-Whitney U-test (P<0.05). III. Glutamate Uptake Experiments L-[2,3,4-3H]glutamate was incubated with the fiber preparation to assess the possibility that glutamate was being taken up by the muscle fibers. The preparation used in these studies was as described by the muscle fiber isolation procedure in this methods section, with the following modifications. At least 45- 60 worm pairs were used for each preparation. The digested pieces were separated into two parts and the fibers in each were released by pipetting into approximately 1.7 ml of I-DMEM to produce a concentrated preparation. This concentrated fiber suspension was then allowed to rest for two minutes by which time the large unbroken worm pieces settled to the bottom and the fiber suspension could be drawn off and placed in microcentrifirge tubes (200 all tube). One 200 pl aliquot was frozen to be assayed for protein content by the Albro method (Albro, 1975). Each sample was preheated to 37°C for ten minutes prior to the addition of [3H]- L-glutarnate. During this time, heat-shocked samples were submerged in boiling water for two minutes, indicated samples were sonicated, and specified samples received inhibitors. Unless otherwise indicated, each sample was incubated for 30 32 minutes at 37°C with 1 uCi of L-[2,3,4-3H]glutamate, to yield a final concentration of 84 nM glutamate. After incubation, the samples were microcentrifirged (Reliable Scientific, Quick Spin-18, 16,000g) for 45-60 seconds. The supernatant was discarded and the pellet rcsuspended in I-DMEM. This suspension was again microcentrifuged for 45-60 seconds, and the rinse procedure repeated. Each sample was resuspended with its own Pasteur pipet, because the fibers tended to adhere to the side of the pipets. After the last rinse, the pellets were resuspended in 0.5 ml of I-DMEM and transferred to scintillation vials, and the microcentrifuge tube and pipet were rinsed with scintillation fluid. The samples were dissolved in 5 ml scintillation fluid, and deteriorations per minute were measured (BetaTrac 6895, TmAnalytic). Variations of this experiment include a time-dependant experiment in which samples are incubated for increasing time periods in the presence of [3H]-L- glutamate (1 uCi). During these experiments, the incubation and pre-incubation was preformed at 37°C. For the temperature-dependence experiment, the samples were held at the designated temperature fiom pro-incubation to the end of the 30-minute incubation, at which time the samples were pelleted and rinsed with I-DMEM at room temperature. In experiments where normal I-DMEM was replaced with low Nam-DMEM, or I- DMEM containing no Na*, the preparation samples were treated with an additional 60-second microcentrifugation step. The resultant pellet was then resuspended in 33 the appropriate buffer. Control samples were resuspended in normal DMEM-I. For complete media contents listing see Table 5. The effects of glutamate uptake inhibitors were tested by adding them to the fiber suspension prior to the 10-minute pre-incubation. This experimental paradigm is similar to the microperfusion experiments, where the inhibitor was added prior to the 10-minute incubation preceding nricroperfusion. The inhibitors remained present until the end of the 30-minute incubation with [3H]-L-glutamate. RESULTS 1. Microperfusion Experiments A. L-Glutamate dose response curve L-Glutamate microperfitsed onto S. mansoni frayed muscle fibers elicited contractions in a dose-dependent manner (Figure l). The EC50 for this effect was approximately 113:3 uM, as calculated by the sigrnoidal curve fit to these data. The negative control in this series of experiments was microperfirsion of I-DMEM. I-DMEM is the same medium in which the fibers were bathed. An average of 12i1% of the fibers contracted in response to control medium. This contraction represents non-specific effects of microperfusion and may be accounted for by the fiayed fibers' known mechanosensitivity and the ability of the fibers to spontaneously contract (Day et al. , 1994a). Often glutamate contractions were compared on corresponding days to the contraction elicited by 25 mM K“. This elevated K+ solution is a positive control, which has served as an indicator of muscle preparation viability (Day et al., 1994a). Muscle fibers contracted 75:l:2% in response to microperfusion of 25 mM K+ solution. This has been considered to represent maximal contraction according to the dose-response relationship of increasing amounts of K+ (Day at al., 1994a). The percent contraction produced by microperfusion of 1 mM L-glutamate (76i1%) was not significantly different 34 35 Figure l. Schistosoma mansoni muscle fibers contract in response to microperfused L-glutamate in a dose-dependent manner. Frayed fibers were nricroperfused with DMEM without or with various concentrations of L- glutamate. The ECSO value calculated from this curve is 11:l:3 11M. *Significantly different from I-DMEM negative control medium (P<0.01). The upper dashed line represents the % fibers contracting in response to 25 mM K*, and the lower dashed line represents the % fibers contracting in response to control medium (I-DMEM). In each petri dish, 15-30 fibers were microperfirsed, and each data point represents the average percentage (i 1 S.E.M.) of fibers contracting from at least 8 dishes. Mann Whitney-U Test. 36 OZHHUEHZOU mmmmE .x. . 1. M w I. a .0“. _ t. . ._. 100 1000 10000 L-GLUTAMATE (uM) 10 0.1 37 from that of 25 mM K”. The contraction in response to microperfused glutamate was qualitatively different fi'om the contractions previously described for both FMRFamide and 25 mM Ki The contraction of fi'ayed fibers induced by FMRF amide is a slow, smooth contraction, and the contraction elicited by 25 mM K+ is a rapid, twitching contraction (Day et al., 1994a, b). In contrast, L-glutamate caused the fibers to contract rapidly and smoothly, without twitching, or shortening beyond approximately half of their original length. B. Stereospecificity To further characterize the contraction produced by glutamate, both enantiomers of glutamate and aspartate were microperfused onto frayed fibers (Figure 2). Each enantiomer elicited contraction in a significantly greater percentage of fibers than microperfirsion of negative I-DMEM control. The percentage of fibers contracting in response to D-glutamate was significantly lower than that for L-glutamate, demonstrating that the contractile effect of glutamate is stereospecific. 38 Figure 2. The enantiomers of glutamate and aspartate elicit different percentages of contraction in Schistosoma mansoni muscle fibers. Each enantiomer was tested at the concentration of lmM, and produced levels of contraction which were significantly different from I-DMEM control values, which are represented by the dashed line (9il%). *Significantly different from L- glutamate (P<0.05). Data are represented as :tl S.E.M. Each bar represents N2 8. 39 L-GLUTAMATE i 4 D-GLUTAMATE -l * L-ASPARTATE 5 l D-ASPARTATE 5 -l I J I I I 0 20 40 60 80 % FIBERS CONTRACTING 40 C. Glutamate Receptor Pharmacology 1. Agonists Agonists of the well characterized mammalian glutamate receptor subtypes were microperfused onto S. mansoni muscle fibers to determine if the contractile response was mediated by a glutamate receptor that fits the described subtypes (Figure 3). All agonists were tested at 1 mM, a concentration of L-glutamate which elicited the maximal percentage of contraction of the muscle fibers. NMDA microperfused onto S. mansoni muscle fibers produced no significant amount of contraction above I-DMEM control values. Because the NIVHDA receptor in the mammalian system requires glycine in order to function, 100 uM glycine was included in the bath and the microperfusion pipet. This concentration of glycine did not increase the percentage of fibers contracting in response to 1 mM NMDA. In addition, MgH is known to block the NMDA receptor at negative membrane potentials. It was necessary to address the possibility that MgM could be blocking a NMDA type channel in the isolated fibers. The resting membrane potential of the muscle tissue of the schistosome has been estimated to be approximately -39 mV (Fetterer et al., 1981). If this is true, then the NMDA receptor may be partially blocked by Mg”. The NMDA current is augmented by reducing the concentration of extracellular Mg4+ (Ascher et al., 1988; Shannon & Sawyer, 1989). Consequently, the fibers were microperfused with NMDA without MgH in 41 Figure 3. Relative ineffectiveness of mammalian glutamate-receptor agonists in eliciting contractions in Schistosoma mansoni muscle fibers. Fibers were rrricroperfused without (dashed line) or with various glutamate receptor agonists at the concentration of 1 mM. Both metabotropic agonists tested, ACPD and L-AP-4 elicited contraction of the frayed muscle fibers at values significantly above control. *Significantly different fiom I-DMEM control values (P<0.05). Each value represents the mean :l:1 S.E.M. for at least 7 determinations. GLUTAMATE lmM NMDA lmM IBOTENATE lmM AMPA lmM KAINATE lmM QUISQUALATE lmM ACPD lmM L-AP-4 lmM 42 * ‘I— + 'I- t + + "l— +- l J l a l 20 40 60 '/o FIBERS CONTRACTING 100 43 the microperfusate or bath solution. Alteration of the MgH concentration did not significantly affect the percentage of fibers contracting in response to 1 mM NMDA. In addition, ibotenate, a less specific agonist at the mammalian NMDA receptor, also did not elicit contraction of the muscle fibers. Non-NMDA ionotropic glutamate agonists, AMPA and kainate, also were ineffective at producing contraction in the frayed fibers (Figure 3). Quisqualate, a fairly non-specific glutamate receptor agonist, did not produce contraction of the S. mansoni frayed muscle fibers. ACPD is a non-specific agonist for all cloned mammalian metabotropic receptor subtypes, and L-AP-4 is an agonist at a subset of these receptors which negatively modulate cAMP levels. ACPD and L-AP-4 microperfused at the concentration of 1 mM both produced levels of contraction that were statistically significantly different from control values, 24:1:4% and 35i6% respectively. However, none of the tested glutamate receptor agonists were as effective as L-glutamate in eliciting contraction of the isolated fiayed muscle fibers. 2. Antagonists Antagonists for the mammalian glutamate receptor subtypes were also employed in the characterization of the glutamate contractile response (Figure 4). For these experiments each antagonist was present in the bath during the 10 minute 44 Figure 4. The glutamate receptor antagonists tested do not inhibit the contraction produced by microperfusion of 100 uM L-glutamate. The response of the fibers to 100 [JM L- glutamate (a concentration which produces sub-maximal percentage of contractions) in the presence of the antagonists tested was not significantly less than the response to 100 1.1M L-glutamate alone. Data are represented as :l: 1 S.E.M. Each bar represents N27. 45 NO INHIBITOR 4 100nM MK-801 r-I 100uM AP-5 ~—-l 100uM CNQX -—I 100uM MCPG 4 OHHZTOHHZOHHSOIHIBIOI % FIBERS CONTRACTING IN RESPONSE TO 100uM L-GLUTAMATE 46 pre-incubation, included in the 100 uM L- glutamate microperfirsion medium, and remained present throughout the experiment. None of the antagonists tested were able to block the contraction induced by 100 pM L-glutamate. MK-801, an antagonist of the NMDA receptor subtype which blocks the intrinsic cation channel of the receptor, did not significantly inhibit the response of the fiayed muscle fibers to 100 1.1M L-glutarnate. The competitive NMDA antagonist AP-5 also did not produce a significant decrease in the percentage of fibers responding to 100 [2M L-glutamate. CNQX, a non-specific antagonist at the AMPA/kainate mammalian subtype glutamate receptor, did not produce a decrease in the percentage of fibers responding to glutamate, nor did the general metabotropic receptor antagonist a-methyl-4-carboxyphenyl glycine (MCPG). Because no specific receptor agonists or antagonists were particularly effective, a receptor subtype could not be assigned to the S. mansoni contractile response, and little could be extrapolated fiom these data regarding the mechanism that results in contraction. Therefore, it was necessary to explore mechanisms other than a normal receptor-mediated response by which L-glutamate might be causing contraction in the schistosome muscle fibers. Hi gh-affinity glutamate transporters are known to be electrogenic, producing a depolarizing current when the transporter is actively taking up glutamate. If such a transporter exists on the schistosome muscle membrane, then activation of this transporter might cause 47 sufficient depolarization to result in contraction of the muscle fiber, much like depolarization of the fibers by microperfusion of elevated K+ causes contraction. D. High-Affinity Glutamate Transporter Pharmacology 1. Transporter Substrates If an electrogenic transporter is responsible for the contraction of the S. mansoni muscle fibers in response to glutamate application, then other amino acids known to be substrates of the transporter would be expected to produce a comparable contractile response. Most hi gh-affrnity amino acid transporters are known to transport L-glutamate, L- and D-aspartate, L-cysteate, and L-cysteine sulfinate quite efficiently (Kanai et al., 1993). It is interesting to point out that the high-affinity glutamate transporters do not transport D-glutamate as well as L- glutamate, which is consistent with experimental observations described in section B of the results. Selected amino acids were microperfused at a concentration of 1 mM (Figure 5). Microperfusion of L-aspartate, and L-cysteate resulted in maximal levels of contraction, which were not significantly different from the percent of fibers contracting in response to 1 mM L- glutamate. L-cysteine sulfmate, a transporter substrate and an agonist at the NMDA-type receptor, elicited 74i3% contraction. 48 Figure 5. Several amino acids could elicit contractions of Schistosoma mansoni muscle fibers. All amino acids were tested at the concentration of 1 mM. Microperfirsion of L-aspartate or L-cysteate resulted in maximal percentage contraction, which was not significantly different from the percent contraction elicited by 1 mM L-glutamate. *Significantly different from I-DMEM control values, which are represent by the dashed line (P<0.05). All amino acids tested were levorotatory. Each bar symbolizes N26. Error bars are :l: 1 S.E.M. 49 HOMOCYSTEATE LYSIN E all GLYCINE --4 T GLUTAMINE CYSTEINE CYSTEINE SULFINATE --1 ---.- h--- pi CYSTEATE ASPARTATE GLUTAMATE d --d .ib--— b-- b. I j l l 20 40 60 80 °/o FIBERS CONTRACTING 50 L-Cysteine and L-glutamine both produced contraction percentages that were significantly different from the I-DMEM negative controls, 48:1:7% and 272t4% respectively. L-Lysine, L-glycine, and L-homocysteate produced no significant percentage of contraction above control levels. 2. Transport Inhibitors To further explore the hypothesis that a glutamate transporter is involved in the contractile response of S. mansoni muscle fibers, several inhibitors of the mammalian glutamate transporter were employed. All inhibitors were placed in the bath and microperfirsion pipet at the concentration of 100 11M, and tested against 100 ”M L-glutamate. The inhibitors L-trans-pyrollidine-Z,4-dicarboxylic acid (PDC) and DL-threo-B-hydroxyaspartic acid (THA), both significantly inhibited contractions induced by 100 11M L- glutamate (Figure 6A). The percent fibers contracting in response to L- glutamate in the presence of PDC was not significantly different fi'om the response of the fibers to microperfusion of I- DMEM control medium. The microperfusion of L-glutamate in the presence of THA produced a slightly higher percentage of contraction from I-DMEM control, 16i1% and 12:1% respectively (P=0.036). The inhibition produced by PDC and THA was reversible, because fibers which had been treated with the inhibitors and then rinsed with normal I-DMEM contracted in response to the application of 100 51 Figure 6. Two inhibitors of the mammalian excitatory amino acid transporter were effective at reducing the percentage of fibers contracting in response to 100 uM L-glutamate. A. *The contraction produced by 100 uM L-glutamate in the presence of both THA and PDC was significantly lower than that produced by L-glutamate alone. The contractile response induced by either FMRFamide or 25 mM K+ was not affected by the presence of these inhibitors, suggesting that the inhibition produced by THA and PDC is specific to the L-glutamate contraction. B. Other classic inhibitors of the high-affinity excitatory amino acid uptake were not effective in blocking the L-glutamate-induced contraction. The contraction produced by L- glutamate, and the contraction produced by L-glutamate in the presence of the inhibitors AAD or DHK, was not significantly different. The dashed line represents the percentage of fibers contracting in response to microperfusion of I-DMEM control medium. Each bar represents N25. Error bars depict i1 S.E.M. AAD, arrrinoadipic acid; DHK, dihydrokainic acid; F MRF, FMRFamide; GLUT, L-glutamate; PDC, L-trans-pyrollidine-Z,4-dicarboxylic acid; THA, DL-threo-B-hydroxyaspartic acid; 25 mM K, elevated K“. 52 A. GLUT 100uM THA 100uM GLUT 100uM FMRF 0.1IIM THA 100uM FMRF 0.1uM ZSmM K THA 100uM 25mM K GLUT 100uM PDC 100uM GLUT 100uM FMRF 0.1nM PDC 100uM FMRF 0.1nM ZSmM K PDC 100uM ZSmM K 0 20 40 60 80 100 % FIBERS CONTRACTING B. GLUT 100uM GLUT 100uM * DHK 100uM GLUT 100uM AAD 100uM l J 0 20 40 60 80 % FIBERS CONTRACTING 53 ,uM L-glutamate. The inhibitor arninoadipic acid (AAD) was not effective at blocking the contraction elicited by 100 12M L-glutamate in the S. mansoni muscle fiber (Figure 6B). DHK was also ineffective at blocking the L-glutarnate-induced contraction. The percent fibers contracting in response to L- glutamate in the presence of either AAD or DHK was not significantly different than that of 100 ,uM L-glutamate alone. To determine if the inhibition produced by THA and PDC was specific to the L- glutamate induced contraction, the fiayed muscle fibers were perfused with 25 mM K+ or FMRFamide in the presence and absence of the inhibitors PDC and THA (Figure 6A). Neither the contractions produced by 25 mM K+ nor FMRFamide were significantly inhibited by the presence of these inhibitors at the concentration of 100 12M. From these data it appears that the inhibition produced by PDC and THA is specific for the contraction induced by L- glutamate. E. Na*-Dependence If the contraction in response to L-glutamate is mediated by a high-affinity excitatory amino acid transporter, it would be expected to be dependent on extracellular Na", because the transporters are highly selective for Na“. When Na+ was replaced with N-methyl-D-glucamine (Table 5), fewer fibers contracted in 54 response to 100 “M L-glutamate (Figure 7). In fact, the contraction produced by L- glutamate in the presence of N-methyl-D-glucamine was not significantly different from the contraction produced by rrricroperfusion of I-DMEM negative control medium, with or without Na”. Replacing Na‘ with N-methyl-D-glucamine appeared to have no deleterious effects on the fibers, as measured by their continued response to 25 mM K+ positive control medium. In addition, the perfusion of I-DMEM containing no Na+ was not significantly different from that of I-DMEM containing Na”. This result clearly shows the marked dependence of the contractile effect of glutamate on the presence of extracellular sodium. Unfortunately, Na+ dependence alone cannot be used to distinguish between an ionotropic receptor-mediated effect and a transporter-mediated mechanism. When NaCl was replaced with LiCl, fewer fibers contracted in response to microperfusion of 100 uM L-glutamate (26i6% in the presence of Li” , as opposed to 70i6% with Na“ present (Figure 8)). The response of fibers to 100 [2M L- glutamate in the presence of Li+ was not significantly different from fibers microperfused with control I-DMEM containing Li“ (26i6% and 21:l:3% respectively). When the fibers were microperfirsed with Li+ I-DMEM control medium, a significantly higher percentage of fibers contracted than fibers 55 Figure 7. Replacing Na+ with N-methyI-D-glncamine diminished the percentage of fibers contracting in response to microperfusion of 1 mM L- glutamate. The response of the fibers to 100 [2M glutamate in the presence of N- methyl-D-glucamine was not significantly different from fibers perfused with I- DMEM containing sodium or N-methyl-D-glucamine. Fibers microperfused with elevated K+ were not affected by replacing Na+ with N-methyl-D-glucamine; in fact, slightly more fibers contracted in response to microperfusion of elevated K+ in the presence of N-methyl-D-glucamine. The dark bars represent samples where Na+ has been replaced with N-methyl-D-glucamine. *Significantly different from fibers rrricroperfused with L-glutarnate in the presence of Na“. The error bars represent i1 S.E.M. Each bar represents N23. % FIBERS CONTRACTING 100 80 60 40 20 56 [CI I-DMEM normal 111111] I-DMEM no Na+1 - t _ .TTTT GLUTAMATE 100uhd I-DMEM + J k ; i— 25 mMK 57 Figure 8. By replacing Na“ with Li”, fewer fibers contracted in response to microperfusion with 100 uM L-glutamate. However, both positive (25 mM K“) and negative (I-DMEM) controls containing Li” responded with a significantly higher percentage of fibers contracting than did their Na"-containing counterparts. Each bar represents N25. Error bars are :l:1 S.E.M. Dark bars represent samples in which Li+ has replaced Na*. % FIBERS CONTRACTING 100 80 60 4O 20 58 1:1 I-DMEM normal E I-DMEM Li+ GLUTAMATE I-DMEM 25 mM K 100uhl 59 microperfirsed with I-DMEM control containing Na+ (21i3% and l3:l:1% respectively). Both positive (25 mM K“) and negative (I-DMEM) controls containing Li+ responded with a significantly higher percentage of fibers contacting than their Na+-containin g counterparts. F. Ca“ Dependence If a L-glutamate tansporter is responsible for the contraction of the muscle fibers due to the depolarizing electogenic current of the tansporter, then this depolarization must somehow ti gger an increase in intacellular Ca” needed to initiate contaction. This source of Ca++ could be from release of internal stores of Ca“, or influx of extracellular Ca“. To determine if influx of extacellular Ca“ was responsible for a rise in intacellular Ca++ leading to contaction, normal I- DMEM was replaced with CaH-fi'ee I-DMEM containing 0.5 mM EGTA directly preceding the 10 minute incubation (Table 5). EGTA remained present throughout the experiment, and all agonists tested were dissolved in the same I-DMEM containing 0.5 mM EGTA. When L-glutamate was nricroperfused onto the fibers in the presence of EGTA, the muscle fibers contacted significantly less than contol fibers, 3d:l% and 7 5:1:4% respectively (Figure 9). The presence of EGTA in the bath lowered the amount of spontaneous contactions observed when fibers were microperfused with negative contol I-DMEM. 60 Figure 9. The contraction elicited by L-glntamate is CaH-dependent. A. By using medium containing no Ca++ and 0.5 mM EGTA, less fibers contact in response to L-glutamate. B. CaH appears to be flowing through a channel that can be blocked by nicardipine, a L-type voltage- gated Ca“ channel blocker *Significantly different from glutamate positive contol values. Each bar depicts N24. The error bars represent :l:l S.E.M. 61 60- 40 1 20- %F1 BERS CONTRACTING * l GLUTAMATE GLUTAMATE 100 uM 100 uM 0.5 mM EGTA B 100 uM GLUTAMATE 100 uM GLUTAMATE 0.5mM COBALT 1 mM GLUTAMATE 1 mM GLUTAMATE 1 uM NICARDIPINE 1 mM GLUTAMATE 10 11M NICARDIPINE 0 20 4O 60 80 % FIBERS CONTRACTING 62 The loss of the ability of the fibers to contact in the presence of EGTA did not appear to be due to irreversible damage, because the EGTA-teated fibers retained their ability to contact in response to elevated K+ solution containing the normal amount of Ca“. The percent contaction produced by elevated K+ in the presence of 0.5 mM EGTA was not significantly different from that of elevated K” microperfused onto normally teated fibers, suggesting that 0.5 mM EGTA is not permanently damaging the fibers. If 0.5 mM EGTA is present in the bath, and the microperfusion pipet contains normal amounts of Ca”, then an increased percentage of fibers will contact in response to L- glutamate. Fibers contacted 3:l: l % in response to glutamate with EGTA in the bath and the microperfusion pipet, while fibers contacted 30i9% when EGTA was present in just the bath and absent in the microperfusion pipet. The Ca“ present in the microperfusion pipet alone was enough to allow contaction in response to glutamate, even when 0.5 mM EGTA was present in the bath. It was not possible to asses the effect of medium made with no Ca”, because this medium independently produced spontaneous contactions in the frayed fibers. Consequently, the measurement of contaction could no longer be employed. This phenomenon may be due to Ca“ leaching fi'om internal stores, causing tansient increases in the intacellular Ca” levels. 63 If CaH is flowing into the fibers by way of a Ca“ channel, it would be possible to block this action by the addition of cobalt chloride, a non-specific competitive Ca++ channel blocker. When normal I-DMEM was replaced with I- DMEM containing 0.5 mM cobalt in the bath and in the microperfusion pipet, 100 11M glutamate had a significantly reduced ability to contact the fibers compared to contol values, 21i2% and 78:1:1% respectively. The ability of Co++ to block contaction in response to 100 12M L-glutamate was lost when the microperfusion pipet did not contain cobalt. Also, microperfusion of elevated K+ medium containing no Co‘H caused the fibers bathed in medium containing cobalt to contact in a normal fashion; again suggesting that the effect of Co++ was both rapidly reversible and not damaging to the fiber's ability to contact. In the S. mansoni frayed muscle fibers, the Ca++ needed to produce contaction in response to microperfusion of elevated K+ appears to be extacellular (Day et al., 1994a). This was demonstated by blocking 25 mM K” contactions with the dihydropyridine voltage- gated Ca++ channel blocker, nicardipine. Nicardipine significantly reduced the percentage of fibers contacting in response to 25 mM K” microperfusion, at the concentation of 1 and 10 ,uM (Day et al., 1994a). To examine if Ca++ was passing through voltage-gated Ca” channels to cause contaction in response to glutamate, nicardipine was placed in the bath of frayed fibers rrricroperfused with 1 mM L-glutamate. Nicardipine at 64 the concentations of l and 10 NM significantly reduced the percentage of fibers contacting in response to 1 mM L-glutamate (Figure 9B). L-Glutamate at the concentation of 1 mM produced 70i% contaction, and 1 mM L-glutamate with 10 [2M nicardipine produced 23:l:9% contaction. This is consistent with the hypothesis that CaH may be flowing through voltage gated Ca” channels, to produce contaction. Although the concentations of nicardipine used in these experiments may seem high, few toxic effects were observed. FMRFamide, a platyhelminth peptide which causes schistosome muscle fibers to contact, is not inhibited by these same concentations of nicardipine (Day et al., 1994b). However, this concentation of nicardipine may not be specifically blocking L-type voltage-gated Ca“+ channels. Verapamil, a phenylalkylamine voltage-gated Ca++ channel blocker, at the concentation of 10 11M produced no significant inhibition of the L-glutamate induced contaction. Fibers contacted 7 5:l:3% in response to 100 uM L- glutamate in the presence of 10 12M verapamil, and 77i1% in response to 100 uM L- glutamate alone. It is interesting to note that veraparnil also had no effect on elevated K“ and F MRFamide-induced contactions. These data imply that the Ca++ needed for contaction in response to L- glutamate is extacellular, and may be passing through a voltage-gated Cafichannel. Therefore, if an electogenic glutamate tansporter is mediating this effect, then the current produced is 65 sufficient to cause depolarization of the fiber membrane leading to opening of voltage-gated Ca++ channels. To support this hypothesis, additional evidence of glutamate tansport is needed. II. Glutamate Uptake Experiments A. Time-Dependence If a high-affinity glutamate tansporter is mediating the S. mansoni muscle fiber contaction, it should be possible to observe the uptake of [3H]-L-glutamate into the isolated fiber preparation. To demonstate the presence of an excitatory amino acid tansporter, the preparation was incubated with [3H]-L- glutamate, (1 uCi/60 mM) at 37 °C. The samples were all pro-incubated for 10 minutes at 37°C to acclimate the muscle fibers to this temperature. [3H]-L-Glutamate was taken up in a time-dependent manner, reaching maximal rate of uptake between 20 and 30 nrinutes (Figure 10). In subsequent experiments, all samples were incubated for 30 minutes. B. Dose-Dependence Increasing concentations of unlabeled L-glutamate inhibited uptake of 100 nM [3H]-L-glutamate (1 uCi/60 mM) in a dose-dependent manner (Figure 11). Maximal inhibition of the uptake of [3H]-L-glutamate was observed in the 66 Figure 10. [3H]-L-Glutamate is taken up in a time-dependent manner. The rate of tansport appears to be linear between 0 and 20 min, and reaches a maximum between 20 and 30 minutes. Error bars represent in] S.E.M. Each data point represents an average of three or more samples. 67 .1- Io 20- 0 s . .m. m m w 3.55:3 7553:2282. azzfianutann £23 120 - TIME (min) 68 Figure 11. FELL-Glutamate is transported in a dose-dependent manner. Increasing concentrations of unlabeled L- glutamate and 100 nM [3H]-L-glutamate were used to define the transport affinity. The 15le0 value estimated from these data is 52 11M using the least squares analysis. 1 mM of unlabeled L-glutamate completely inhibited the transport of 100 nM [3H]-L-glutamate, and this value was not significantly different from samples which were sonicated. Each data point depicts an average of four samples, and error bars are i1 S.E.M.. % MAXIMAL INHIBITION 69 100- ........................ . 20i oi go ' 460 ' 660 ' 850 ' 1o'oo UNLABELED L-GLUTAMATE (uM) 7O presence of 1000 pM unlabeled glutamate. This level of uptake was equivalent to that associated with samples which were sonicated prior to incubation with [3H]-L- glutamate, (P=0.69). When samples are sonicated, there is no intracellular space available into which [3H]-L- glutamate can be sequestered by a transport mechanism. Therefore, sonicated samples represent binding of [3H]-L- glutamate to the muscle fiber preparation and not transport. These data were analyzed using non-linear regression analysis, and the ICSO was determined to be 52119 “M. The normalized data plotted in Figure 11 is fit with a rectangular hyperbola which predicts the inhibition at 50% to be 48 uM. These estimated values of 52 and 48 uM are on the upper limit of what has been traditionally described for the hi gh-affinity, Na*-dependent excitatory amino acid transporter, which is defined by a K, below 50 [1M (Cox et al., 1977). However, these traditional defining characteristics are of less importance now that molecular sequence data are available. From this new cloned family of receptors, it has been shown that the Km's of the expressed transporter range from 2.0 to 97.0 11M, depending on the specific clone, the expression system, and method of measurement (Kanai et al., 1993; Arriza et al., 1994). C. Temperature-Dependence The uptake of [3H]-L-glutamate was temperature-dependent (Figure 12). 71 Figure 12. FELL-Glutamate transport is temperature-dependent. Samples were pre-incubated for ten minutes at the designated temperatures, 1 uCi [3H]-L- glutamate was then added and incubation continued at the respective temperatures for 30 minutes. Each data point is an average of at least 3 samples and error bars are :tl S.E.M. 72 ”av—<23 mh<2:.<..—m~— TEMPERATURE (degrees celcius) 73 Frayed fiber preparation samples tested at 22 °C most efficiently transported [3H]- L-glutamate. Because these data were highly variable, the uptake observed at 0°C was normalized to 1. The observed variability is inherent to the difficulties of keeping samples at a designated temperature throughout the experiment. [3H]-L- Glutamate transport was most efficient at 22°C, however, 37°C was chosen for the standard temperature to allow comparison of these data to the microperfusion experiments which were also performed at 37°C. The Q", value of glutamate uptake was measured to be 3.3 using the equation Q.O=(K,/K2)'°’("“2) where K, and K2 represent the velocity constants of the transport of glutamate (dpm/mg/hr), and t, and t2 represent temperature in Celsius at 10 and 22 degrees. The Q", value is a measure of the increase in reaction velocity over a temperature rise of 10 °C (Prosser, 1973). D. Transport Pharmacology-Transport Inhibitors If L- glutamate transport is mediating the contraction observed when fibers are microperfused with L- glutamate, then it would be expected that the pharmacology observed for the uptake of [3H]-L-glutamate would be similar to the pharmacology of the contractile response. Since the inhibitors PDC and THA effectively lowered the percentage of fibers which contracted in response to microperfusion of L-glutamate, then it would be expected that these inhibitors 74 would also block uptake, which was observed. PDC inhibited the transport of [3H]-L-glutamate into the preparation in a dose dependent manner (Figure 13). The logistic sigmoidal curve fit to these data predicts the ICSO to be 3.2 [2M PDC, which indicates that PDC is quite potent and maximally effective. Both PDC and THA completely block uptake of [3H]-L-glutamate at the concentration of 100 uM. When 100 uM PDC or THA was added to the samples, the percentage of [3H]-L-glutamate uptake was 11d:1%, and l6i2%, respectively (Figure 14). These percentages are not significantly different from the percentage of [3H]-L-glutamate measured in samples that were sonicated, representing non- specific binding of [3H]-L-glutamate. This indicates that these inhibitors at the concentration of 100 uM were completely blocking [3H]-L-glutamate uptake. Because these inhibitors are thought to be quite specific for the hi gh-affinity glutamate transporters, these data provide further support of the hypothesis that there is a glutamate transporter in S. mansoni. E. Na*-Dependence An inherent quality of high-affinity glutamate transport is Na+ dependence. Therefore, an integral piece of data, in addition to the observed Na+ dependence of glutamate-induced contraction, is Na+ dependence of [3H]-L- glutamate uptake. To demonstrate Na+-dependence, samples were incubated with the normal amount of 75 Figure 13. trans-Pyrollidine-2,4-dicarboxylic acid (PDC), a high-affinity glutamate transport inhibitor, inhibits uptake of [’H]-L-glutamate in a dose- dependent manner. The estimated 1C5,0 is 3.2 ,uM. [3H]-L-Glutamate transport was inhibited 100% by 100 ,uM PDC. Error bars depict il S.E.M. 76 I Inwfl 100 'l 10 Will _ ..-_ 1. T 1 .10. m0 1 T :1 . _ . . . . . q m. 0 0 m 0 O O m 8 4 2 ZOE—BIZ— A<2~X<2 .x. PDC (uM) 77 Figure 14. The addition of 100 [2M PDC or THA inhibits the transport of [3H]-L-glutamate. *Significantly different from samples which were incubated without inhibitor. The percent [3H]-L- glutamate in samples incubated with [3H]-L- glutamate and 100 uM inhibitor was not significantly different from samples that were sonicated prior to incubation with 83.5 nM [3H]-L- glutamate. Samples which have been sonicated represent non-specific binding of [3H}L-glutamate (9i1% of total uptake), and serve as the negative control represented by the dashed line. Error bars are :tl S.E.M.. Each bar symbolizes N24. PDC, trans-pyrollidine-2,4- dicarboxylic acid; THA, DL-threo-B-hydroxyaspartic acid % GLUTAMATE UPTAKE 120 100 78 * l i 100uM PDC * No Inhibitor 79 Na+ (82 mM), 41 mM Na+, and zero Na“ (Figure 15). N—methyl-D-glucamine was used to replace the Na+ in a equimolar fashion. Reducing the Na” concentration reduced the transport of [3H]-L-glutamate. Samples containing zero Na“ were not significantly different fi'om samples which were sonicated, demonstrating non- specific binding 8:1:l% and 9:1:1%, respectively. These results support the hypothesis that there is a high-affinity, Na*-dependent glutamate transporter on the membrane of the S. mansoni muscle fiber. 80 Figure 15. [3H]-L-Glutamate uptake is Na”—dependent. *Samples with no Na“, and those with ‘/2 the normal amount of Na+, were significantly different from the samples containing 82 mM Na+ (P<0.05). When Na+ was completely replaced with N-methyl-D-glucamine, the amount of [3H]-L- glutamate uptake was not significantly different from the amount of [3H]-L-glutamate in samples which were sonicated to represent non-specific binding (8.2:bl .3% and 8.8:t1.3% respectively). Each bar represents an average (:tl S.E.M.) of 6 samples. 120 .s C O a: C DPMIPROTEIN(mg)/HOUR # a: O O N O 81 SODIUM DEPENDENCE .L 0 mM Na+ 41 mM Na+ 82 mM NH DISCUSSION 1. Pharmacology A. Glutamate The results from both the microperfilsion experiments and the [3H]-L- glutamate uptake experiments support the hypothesis that there is a transporter on the isolated S. mansoni muscle fiber membrane, and that it is most likely responsible for the observed contraction of the isolated muscle fibers in response to microperfusion of L-glutamate. The model proposed to describe the mechanism is depicted in Figure 16. In this model, L-glutamate and other transport substrates are taken up by an electrogenic, high-affinity glutamate transporter into the fi'ayed fiber. This Na+-dependent transport produces a depolarizing current, which in turn opens nicardipine-sensitive voltage-gated Ca2+ channels allowing [Ca2+], to enter the fiber, initiating contraction. The estimated ECSO of transport for [3H]-L- glutamate into S. mansoni muscle fibers is approximately 52 uM, which is similar to the ECSO of the contractile response which was estimated to be 11 uM. Even if the transport of glutamate was solely responsible for the contraction induced by microperfusion of glutamate, the EC50 values of these processes may not be equal for several reasons. For instance, the mechanism leading to contraction may 82 83 Figure 16. Proposed mechanistic model of the contraction induced by L- glutamate in the S. mansoni muscle fiber. When glutamate is microperfused onto the frayed muscle fiber, it may be transported by a Na+-dependent high-affinity glutamate transporter. The question mark indicates that the exact stoichiometry of the transporter is unknown. Currently it is thought that two Na“ ions are co- transported with each molecule of glutamate (Kanai et al,. 1993). However, unpublished research fiom the laboratory of M. Kavanaugh suggests that there may be three Na+ ions co-transported with each molecule of glutamate (personal communication). It is thought that one K+ ion is counter-transported with each molecular of glutamate, as measured by K*-sensitive electrodes (Bouvier et al,. 1994). In addition, a pH-changing ion is transported; it is still unknown, however, whether this effect is mediated by co-transport of a H“, or counter-transport of a ' OH'. In retinal cells evidence fi'om anion substitutions supports counter-transport of a OH‘ or HCO3‘ (Bouvier et al,. 1994). Regardless of the exact stoichiometry, inward transport of substrate through a hi gh-affinity glutamate transporter produces a depolarizing current. It is possible that this depolarization is sufficient to open nicardipine-sensitive voltage-gated CaH channels. The flow of Ca” down its electrochemical gradient into the fiber may be sufficient to initiate contraction of the S. mansoni muscle fiber. 84 GLUTAMATE VOLTAGE GATED TRANSPORTER Ca++ CHANNEL MICROPERF USED GLUT,Na+ Ca-H _ - v + + + + v - - DEPOLARIZATION ? CONTRACTION 85 consist of one or more amplification steps. This would result in an ECSO value for contraction that is lower than the ECSO value for glutamate transport, which is consistent with the results of this study. In addition, the amount of depolarization required to produce a detectable contraction of the muscle fiber in our assay is unknown. The transporter may not need to be maximally activated to observe maximal percentage contraction in our assay. Because the ECSO values for glutamate transport and the glutamate-induced contraction are similar, it suggests that glutamate transport could account for the contractile effect. However, it would be premature to conclude that a glutamate receptor is not also present on the schistosome muscle membrane. B. Stereospecificity When isolated fibers are microperfused with enantiomers of both glutamate and aspartate, the resulting stereospecificity pattern is characteristic of the preferences of the hi gh-affinity excitatory amino-acid transporter (Kanner & Schuldiner, 1987; Arriza et al., 1994; Klockner et al., 1994). L-glutamate, and both L- and D-aspartate, produce maximal contraction of the isolated muscle fibers; however D- glutamate is not as potent. D-isomers are not generally potent agonists for glutamate receptors. In fact, the crayfish neuromuscular glutamate receptor is 250-fold more sensitive to L-glutamate than D—glutamate (Bishop et al., 86 1987). Because D-aspartate elicited maximal contraction of the isolated muscle fibers, this effect is probably not mediated by a glutamate receptor, which is consistent with the hypothesis supporting a glutamate transporter-mediated mechanism. C. Receptor Agonists Microperfusion of selected glutamate receptor agonists onto the muscle fibers produced little or no contraction of the isolated muscle fibers, which provides firrther evidence that the contractile response induced by glutamate is not being mediated by a mammalian-like glutamate receptor. However, it is difficult to characterize an invertebrate glutamate receptor using mammalian receptor agonists, because often the pharmacological profile of invertebrate receptors is different from that of the well-described and cloned mammalian subtypes (Shinozaki & Ishida, 1992). Although both mammalian metabotropic receptor agonists produced statistically significant contraction of the isolated muscle fibers, this response may not represent a specific action on a metabotropic receptor. However, the low efficacy and potency of these metabotropic agonists could be due to inherent differences between a schistosome glutamate receptor and the mammalian metabou'opic subtypes. It is possible that there may be a subset of the frayed fibers that contract in response to the metabotropic agonists. If this is true, 87 then the contractile response induced by L- glutamate may be mediated by more than one mechanism. It is also possible that the metabotropic agonists, L-AP-4 and ACPD, are functioning as substrates for a glutamate electrogenic transporter. Presently, a thorough study of the mammalian glutamate receptor agonists that are also substrates for the mammalian hi gh-affinity transporter is incomplete. However, it is known that NMDA is either poorly transported or not transported at all (Johnston et al., 1979; Garthwaite, 1985; Rosenberg et al., 1992; Wadiche et al., 1995b). This is consistent with data presented in this study which show that NMDA does not cause the muscle fibers to contract. In addition, kainate is not transported by the mammalian hi gh-affinity transporters, and actually functions as an antagonist in some transporter subtypes (W adiche et al., 1995b). Kainate did not elicit contraction in the flayed fibers, suggesting that it is neither transported, nor does it bind to an excitatory kainate receptor. D. Receptor antagonists Glutamate receptor antagonists have been insu'umental in the early process of defining the mammalian glutamate receptor subtypes. However, in invertebrate systems few of these inhibitors are effective (Walker & Holden-Dye, 1989; Shinozaki & Ishida, 1992). Therefore it is difficult to use the antagonists as a tool 88 to define the nature of an invertebrate receptor, especially in the evolutionarilly distant schistosome. The antagonists tested did not inhibit the contraction of the muscle fibers induced by L- glutamate, even at relatively high concentrations known to block mammalian receptor responses (Hoehn & White, 1990). This result is consistent with the hypothesis that an electrogenic transporter is mediating contraction. However, it is not reasonable to eliminate the possibility that there may be a glutamate receptor on the schistosome muscle because an active mammalian glutamate receptor antagonist has not yet been identified. E. Transporter substrates All commonly-transported amino acids tested ( L-glutamate, L-and D- aspartate, L-cysteate, and L-cysteine sulfinate) caused contraction of the S. mansoni muscle fibers. If an electrogenic transporter is depolarizing the membrane, then presumably the amount of current produced by the transporter is the factor which determines if the fibers contract. The amount of current measured in response to a transport substrate is related to the amount of substrate transported, although it is not a direct or straight-forward relationship (W adiche et al. , 1995a). These transporter substrate data are consistent with the proposed model for glutamate-induced contraction mediated by a hi gh-affinity glutamate transporter. 89 L-Homocysteate did not produce contraction of the S. mansoni flayed muscle fibers. It is thought that L-homocysteate is transported by the low-affinity glutamate transporter and is a poor substrate for the hi gh-affinity glutamate transporter (Cox et al., 1977; Tanaka, 1994). In addition, L-homocysteate is an agonist for the NMDA mammalian glutamate receptor. Since homocysteate did not produce a significant percentage of contraction in the flayed fibers, it seems unlikely that the contractile response is mediated by a NMDA receptor subtype, or by a low-affinity transporter. The amino acid glycine also did not produce a significant percentage of contraction in the isolated fibers, which suggests that glycine is not transported into the isolated muscle fibers. Like homocysteate, glycine is a relatively selective substrate for the low-affinity glutamate transporter (Webb, 1986). Recently, it has been shown that L-cysteine is a substrate of the hi gh-affinity EAAT3 transporter subtype (K, = 190 uM), although it does not produce maximal current when transported (Zerangue & Kavanaugh, in press). This is consistent with the finding that 1 mM L-cysteine did not cause maximal contraction of the fibers. These data support the proposed model where the contractile effect is produced by electrogenic transport in the isolated muscle fibers. 90 F. Transport Inhibitors trans-Pyrollidine dicarboxylic acid (PDC) and threo-hydroxyaspartic acid (THA) are a potent inhibitors of glutamate high-affinity transport. PDC is specific for the high-affinity transporters, and it does not bind mammalian glutamate receptors (Freund et al., 1995). PDC inhibits [3H]-L-g1utamate uptake into the muscle fiber preparation in a dose-dependent manner. PDC and THA (100 uM) completely blocked [3H]-L- glutamate uptake, and inhibited the contraction elicited by 100 [1M L-glutamate in the isolated muscle fibers. Although 100 pM of inhibitor may seem quite high, these concentrations and higher are commonly employed (Tanaka, 1993; Arriza et al., 1994). Most inhibitors of glutamate hi gh-affrnity transport are competitive, and are themselves transported. When PDC (1 mM) was microperfused as a substrate, it caused 43:1:7% of the fibers tested to contract (N=4). This suggests that the transport of PDC is producing a depolarizing current. PDC also produces a depolarizing current in patch-clamped oocytes that were injected with cloned glutamate transporter message (Arriza et al. , 1994). However, the current produced by PDC is only 34-52% of the current produced by L- glutamate (Arriza et al., 1994). This explains why PDC did not produce maximal levels of contraction in S. mansoni muscle fibers (even at the high concentration of 1 mM), and why PDC can also serve as an inhibitor of glutamate-induced contraction. 91 If the S. mansoni muscle fiber contained an excitatory glutamate receptor that was responsible for mediating contraction in addition to a high-affinity glutamate transporter, then the co-application of PDC and glutamate should cause more fibers to contract than application of glutamate alone. If the transporter is blocked by PDC, then less glutamate is sequestered which in turn would increase the concentration of glutamate available to bind the excitatory glutamate receptor. This paradigm is typically observed in brain slice preparations which contain both transporters and receptors for glutamate. However, fewer S. mansoni fibers contracted in response to co-applied of PDC and glutamate, suggesting that the contractile effect is not mediated by a glutamate receptor. 11. Ion Specificity A. Na*-Dependence Both the contractile response of the muscle fibers elicited by L-glutamate, and the uptake of [3H]-L- glutamate by the muscle fiber preparation were found to be Na*-dependent processes. In the absence of Na*, glutamate-induced conu'action and [3H]-L- glutamate uptake were reduced to control levels, suggesting that these processes are highly Na+-dependent. With this information, those glutamate- mediated processes that do not involve a Na+ current can be eliminated, such as glutamate metabotropic receptor-mediated 1P3 release of Ca“+ flom internal stores, 92 or other such mechanisms. When Na+ was replaced with Li+, no significant contraction of the isolated muscle fibers was observed in response to microperfusion of L- glutamate. The hi gh-affinity transporters are very selective for Na“, and Li“ does not substitute for Na+ in the transport of glutamate. However, Li+ is a suitable ion candidate for the NMDA and non-NMDA glutamate receptor channels and other Na*-conductin g channels (Schwartz & Tachibana, 1990; Yamaguchi & Ohmori, 1990; Barbour et al., 1991; Wyllie et al., 1991). Since Li+ did not functionally substitute for Na", the contraction was probably not mediated by a glutamate ionotropic receptor. These data are, however, consistent with the hypothesis that contraction is being mediated via a high-affinity, Na+-dependent excitatory amino acid transporter. It is interesting to note that both positive (25 mM K“) and negative (I- DMEM) controls containing Li+ responded with a significantly higher percentage of fibers contracting than did their Na*-containing counterparts. This could be explained by an increased amount of spontaneously-contracting fibers (although this was not apparent during the time of data collection), or the membrane could be more permeable to Li+, causing the fibers to become depolarized. It is also possible that Li+ cannot adequately substitute for Na+ in the Na"/K+ ATPase, which would result in depolarization of the membrane and lead to its increased excitability. It would be interesting to know if ouabain would mimic this effect. 93 B. Ca“-Dependence The L-glutamate-induced contraction appears to be dependent on the presence of extracellular Ca“ and is blocked by the dihydropyridine L-type voltage- gated Ca++ channel blocker nicardipine; however the response was not inhibited by the phenylalkylamine veraparnil. Although veraparnil blocks the same type of voltage-gated CaH channels, it is not as potent as nicardipine at relaxing vascular smooth muscle. In addition, dihydropyridines produce a state- dependent block of voltage- gated Ca” channels, and are thought to bind best to the inactivated state of the Ca“ channel. As a result, dihydropyridines are more potent at depolarized potentials. From the results of early microelectrode studies, it is thought that the resting potential of schistosome muscle is approximately -28 mV (Thompson et al., 1982). This relatively depolarized value would increase the ability of dihydropyridines to block voltage-gated Ca++ channels. In addition, dihydropyridines are known to inhibit cyclic nucleotide phosphodiesterases, which may increase cyclic nucleotide concentrations. These factors may play a role in the ability of the nicardipine, but not veraparnil to inhibit the contractile response. The contraction induced by 25 mM K+ is also blocked by nicardipine and not verapamil (Day at al., 1994a). This suggests that depolarization and Ca++ entry through voltage-gated Ca“ channels are common mechanisms to both the 25 mM 94 K+-induced contraction and the glutamate-induced contraction. This is consistent with the hypothesis that the Ca” needed to initiate contraction is extracellular and may be flowing through nicardipine-sensitive voltage- gated Ca++ channels. III. Glutamate Transport The observation that [3H]-L-glutamate is incorporated into the muscle fiber preparation is strong evidence for the existence of a glutamate transporter. The estimated ECso of [3H]-L-glutamate transport into the muscle fiber preparation is 51.7 pM. This is a typical value for a hi gh-affinity glutamate transporter. S. mansoni glutamate transport saturates at approximately 30 minutes. The flayed fibers are part of a primary preparation, and are known to be quite temperature- sensitive. Fibers heated to 37°C retain their ability to contract only for a limited amount of time. Therefore, at time points of 30 minutes and greater, the muscle fiber preparation is deteriorating, adding to the observed effect of transport saturation. [3H]-L-Glutamate is taken up in a time-dependent manner which is similar to that of other transporters (Pines et al., 1992). Transport of [3H]-L-glutamate into the flayed fiber preparation is also temperature-dependent. Temperature dependence is quite common in amino acid transporters (Lerner, 1978). The calculated Q10 value of 3.3 for the transport of glutamate into the fiber preparation 95 is higher than that expected for metabolic processes which typically have Qw values between 2 and 2.5 (Prosser, 1973). This is additional evidence for a transport-mediated process, rather than a ionotropic receptor-mediated response which would be less temperature-dependent. IV. Transporter Subtype Based on the data as a whole, it is difficult to assign the putative S. mansoni glutamate transporter to a particular transporter category. It is important to note that the categorization constructed for this dissertation for the cloned hi gh-affinity glutamate uansporters into EAAC-like, GLAST-like and GLT-like groups is most likely an oversimplification of the actual number of categories of these transporters (Arriza et al., 1994). The transport ECSO of the schistosome transporter has been estimated at approximately 52 uM, and the ECSO of contraction in response to L- glutamate is approximately 11 MM. However, it does not seem possible to delineate between transporter categories based on these values, because the K, values for the cloned transporters range flom 2.0 to 97.0 [2M (Tables 2-4). This variation is due to the specific transporter clone being characterized, the expression system used, and method of measurement (either transport of [3H]-L-glutamate or current produced by L-glutamate). 96 It would seem reasonable that the schistosome transporter would be similar to a subtype that is found in several tissues, and not a subtype that is restricted to the brain. In this case, the EAAC-like transporters seem a likely choice, because they have been found in broad range of tissues including the heart and skeletal muscle (Kanai & Hediger, 1992; Arriza et al., 1994; Freund et al., 1995). Some of the GLAST-like transporters have also been immunohistologically localized to muscle tissue (Tanaka, 1993; Arriza et al., 1994; Wadiche et al., 1995b), but the GLT-like transporters, to date, have only been found in the brain (Pines et al., 1992; Arriza et al., 1994; Freund et al., 1995). The pharmacology of the different hi gh-affinity glutamate transporter categories is overlapping, but some generalizations can be made. Aminoadipic acid (AAD) and dihydrokainic acid (DHK) are effective inhibitors of the GLT-like transporters, but do not block uptake by the GLAST-like or EAAC-like transporters. Neither AAD or DHK were able to inhibit the glutamate-induced contraction of the muscle fibers. In addition, the amino acid cysteine, which produced contraction of the isolated muscle fibers, is a transport substrate for the EAAT3 transporter which is a GLAST-like transporter. If these generalizations hold true, then the putative transporter on the S. mansoni muscle membrane is most likely not a GLT-like transporter, but could be a GLAST-like or EAAC-like transporter. Of course, this question could be addressed directly by cloning and 97 expressing the schistosome glutamate transporter. Preliminary sequence alignment of cloned transporters reveals that it would be possible to design PCR primers or an oligonucleotide probe. V. Transporter Function A. Modulation of Membrane Potential The data collected in this study support the hypothesis that there is a glutamate transporter on the S. mansoni muscle fiber, and that it is most likely responsible for the glutamate-induced conu'action observed in the isolated muscle fibers. Glutamate transporter activation has never before been associated with a contractile effect. However, glutamate transporters are known to cause depolarization and increased levels of [Ca2+]i. From literature concerning glutamate's actions on rat brain synaptosomes, it was observed that externally- applied L- glutamate was causing the synaptosomes to release adenosine (Hoehn & White, 1990). It was shown that a high-affinity glutamate transporter was causing the synaptosomes to depolarize, and that this depolarization was opening voltage- gated Ca“ channels normally involved in transmitter release. The authors suggest that the glutamate transporter on the synaptosome may be involved in modulation of a feedback mechanism involving glutamate and adenosine by means of altering the membrane potential. This ability of glutamate transporters to modulate the 98 membrane potential appears to be a recurring theme to proposals of their physiological relevance in viva. Glutamate transport has also been shown to raise cytosolic CaH in GH3 pituitary cells via a high-affinity transporter (V illalobos & Garcia-Sancho, 1995). GH, cells are responsible for hormone release, which is controlled by oscillations of inuacellular Ca” concentrations. The authors suggest that the ability of the glutamate transporter to modulate the GH, cell membrane potential may cooperate with other regulatory mechanisms, such as electrical activity and hypothalarnic releasing factors, to alter hormone secretion. Glutamate transporters play an important role in the visual system of several vertebrates. The transporters are located in rods and cones of the goldfish retina (Marc & Lam, 1981), and salamander retinal glial cells (Mtlller cells) (Bouvier et al., 1992; Eliasof & Werblin, 1993). It has been suggested that the cellular acidification caused by the counter-transport of OH' may modulate intracellular messengers (Bouvier et al., 1992). It is known that activation of the hi gh-affmity glutamate transporter in Muller cells raises [K"]°. When light activates photoreceptors, glutamate release is suppressed, which in turn reduces the K+ efflux flom the Muller cells, and may contribute to shaping the C-wave of the electroretinograrn (Amato et al., 1994). As the mammalian hi gh-affinity glutamate transporters were cloned, it was 99 found that several of the sequences were expressed in areas outside the central nervous system, such as intestine, kidney, heart, placenta, and skeletal muscle. In fact, the rabbit EAAC] transporter was cloned on the premise that the intestinal transporters, presumably being employed for nutrient absorption, were similar to the transporters in the CNS which are responsible for terminating the action of glutamate by reuptake. Although the immunohistochemistry was explored, surprisingly little was said about the appearance of these transporters and their function outside the central nervous system. For instance, a high-affinity glutamate transporter has been described in flog red blood cells; however no attempt to formulate a physiological relevance was made (Gallardo et al. , 1994). There is evidence for hi gh-affinity glutamate transporters in muscle tissue (Arriza et al., 1994; Shashidharan et al., 1994), but little is known about their function. The physiological function of the schistosome transporter is also unknown. B. Glial Cell Theory Because flatworms are not reported to have glial cells, It has been proposed that other cell types might play the role of a glial cell by transporting glutamate out of the extracellular space (Webb, 1986). Glial cells in the mammalian CNS contain hi gh-affrnity glutamate transporters and keep levels of glutamate in the extracellular space below toxic levels and provide a concentration gradient to draw 100 glutamate out of the synapse. It is possible that the transporter on the schistosome muscle membrane is present to transport excess glutamate released flom the nervous system in a fashion similar to the extraneuronal norepinephrine uptake sites found in guinea pig tracheal smooth muscle. These transporters have a lower affinity (Km = 156 uM) for glutamate and have been named uptake-2 (O’Donnel & Saar, 1978). C. Post-Junctional Transport In the flatworms, glutamate irnmunofluorescence is concentrated in the neural tissue (Solis-Soto & Brink, 1994), and glutamate is proposed to be an excitatory neuromuscular transmitter (Webb & Eklove, 1989). However, the studies presented here with isolated muscle fibers revealed little evidence of glutamate receptors on the muscle fibers. There are several reasons why this conclusion may be flawed. First, only one type of muscle fiber in the preparation was studied (flayed fibers), and it is possible that other schistosome fiber types contain glutamate receptors. It may be possible that only a subset of the muscle fibers contact the neural tissue, and the remaining fibers are controlled by electrical coupling (Thompson etal., 1982). It is possible that the flayed fibers studied contain glutamate receptors that were functionally damaged by the enzymatic portion of the isolation procedure, 101 yielding receptors that produce no excitatory activity upon application of glutamate. It is also important to consider that the flayed fibers have lost their nucleus and possibly their sarconeural arms. The schistosome muscle cell nucleus is located on a cytoplasmic stalk. Presumably, it is sheared during the isolation procedure, because they are not normally observed to be attached to the flayed fibers in our preparation. In addition, the sacroneural arm, which is an extension of the muscle fiber that contacts the nerve, may also have been sheared in the isolation procedure. Therefore, if the receptors for glutamate are localized to one of these structures, then no receptor-mediated response for glutamate would be observed in the flayed fibers. Mammalian hi gh-affrnity transporters are present at glutamatergic synapses, and are thought to quickly sequester glutamate to terminate its action. If an excitatory glutamate receptor is normally present on the flayed fibers, then the role of the glutamate transporter on the flayed fiber membrane may be to sequester glutamate released flom the neuronal tissue to terminate its action. C. Metabolic Theory L-glutamate is known to be involved in energy metabolism by fimctioning as a metabolic intermediate in invertebrates (Webb, 1986). The glutamate transport detected in the S. mansoni muscle fibers may serve the function of 102 transporting amino acids for metabolic purposes (Figure 17). In support of this theory, whole schistosomes have been shown to take up radiolabeled L-glutamate and L-glutamine flom bath medium in vitra, and metabolites of ["C]-L-glutamine have been traced. Results reveal that L-glutamine is metabolized to L- glutamate, a-ketoglutaramate and a-ketoglutarate (Foster et al. , 1989). a-Ketoglutarate can be directly incorporated into the citric acid cycle, an aerobic metabolic pathway (Figure 17). Although it is relatively well-accepted that schistosomes are predominately lactate producers, it has been shown that they have the ability to use an aerobic metabolic pathway, especially when in an in-vitro environment (Tielens & VanDenBergh, 1987). This is supported by the observation that worms incubated in vitro release relatively large amounts of alanine (Foster et al., 1989). When glutamate is converted to either a-ketoglutaramate or a-ketoglutarate, an amino group is lost, pyruvate may serve as the amino recipient in this transamination. The product of this reaction is alanine. In addition, it has been shown that schistosomes metabolize ["C]-L-glutamine and ["C]-L-glutamate to I‘COZ, which is the product expected if metabolism occurs via the citric acid cycle (Foster et al. , 1989). It is possible that glutamate metabolism in the schistosome muscle is a back-up energy pathway, employed predominantly during times of stress. 103 Figure 17. Proposed metabolic pathways of the schistosome. Normally schistosomes produce ATP in the muscle through glycogenolysis of glycogen storage granules, which are converted to pyruvate by glycolysis. The parasites release lactate as a waste product. Recently, it has been shown that S. mansoni will take up radiolabeled glutamine and metabolize it to glutamate, alpha- ketoglutarate, and C02. These in-vitro experiments provide evidence for aerobic metabolism. It may be that glutamine is taken up through the tegument and converted to glutamate, which is transported by the hi gh-affinity glutamate transporter into the muscle. This glutamate could then be transaminated to alpha- ketoglutarate which can be shunted into the citric acid cycle and used as a source of energy for the muscle. The NH. group produced by the transamination reactions could be combined with pyruvate to be converted to alanine. This may explain why parasites incubated in vitro (whose glycogen stores have been depleted) produce large amounts of alanine. 104 GLUTAMINE GLYCOGEN v 1 GLUCOSE N+H4 """""""""""""" PYRUVATE GLUTAMATE ALANINE / $ \ LACTATE ACETYL-CoA v '1'”4 “““““ “X CITRIC ALPHA-KETO > ACID GLUTARATE CYCLE SUMMARY 1. Isolated flayed fibers contract in a dose-dependent manner in response to L- glutamate microperfusion, and pharmacological characterization of this response suggests that it is not mediated by a glutamate receptor. 2. The flayed muscle fiber preparation transports [3H]-L-glutamate in a dose-dependent manner, which is also temperature- and time-dependent. 3. Both the contractile response of the flayed fibers and the [3H]-L- glutamate transport was Na"-dependent, and could be blocked by hi gh-affinity glutamate transport inhibitors. 4. Other substrates of the hi gh-affrnity glutamate transporters could mimic the contractile response in isolated muscle fibers. 5. The presence of a hi gh-affinity glutamate transporter on the membrane of the S. mansoni isolated flayed muscle fiber may be responsible for the contraction observed in response to microperfusion of glutamate. 6. The high-affinity glutamate transporter on the S. mansoni muscle fiber may play a role in modulating the membrane potential of the muscle. 105 106 7. Because flatworms do not contain glial cells, it is possible that the hi gh-affinity transporter on the flayed fiber serves this purpose. 8. The S. mansoni hi gh-affinity glutamate transporter could transport glutamate and other amino acids to be used in energy metabolism for the muscle. BIBLIOGRAPHY Albro, P. W. (1975). Determination of protein in preparations of microsomes. Anahcticalfiicchcmim 64 485-493 Amato, A., Barbour, B., Szatkowski, M. & Attwell, D. (1994). Countertransport of potassium by the glutamate uptake carrier in glial cells isolated flom the tiger salamander retina. Wm 479:37 1-3 80. Arriza, J .L., Kavanaugh, M.P., Fairman, W.A., Wu, Y., Murdoch, G.H., North, R. A. & Amara, S. G. (1993). Cloning and expression of a human neutral amino acid transporter with structural similarity to the glutamate transporter gene family MW 268 15329-15332 Arriza, J. L. Fairman, W..,A Wadiche, J. I., Murdoch, G..,H Kavanaugh, MP. & Amara, S. G. (1994). Functional comparisons of three glutamate transporter subtypes cloned flom human motor cortex. W 14: 5559-5569. Asch, H.L. & Read, C.P. (1975). Membrane transport in Schistosoma mansoni: Transport of amino acids by adult males. BMW 38: 123-135. Ascher, P., Bregestovski, P. & Nowak, L. (1988). N-Methyl-D-aspartate-activated channels of mouse central neurones in magnesium-flee solutions. .[QnmaLQf thsiglm 399:207-226. Barbour, B., Brew, H. & Attwell, D. (1991). Electrogenic uptake of glutamate and aspartate into glial cells isolated flom the salamander (Ambystoma) retina. Inumalflhxsiulm 4361169493. Barnard, EA. (1992). Receptor classes and the transmitter-gated ion channel. IjES Octoberz368-374. 107 108 Bates, S.E., Sansom, M.S.P., Ball, F.G., Ramsey, R.L. & Usherwood, P.N.R. (1990). Glutamate receptor-channel gating. Bimhysigafloumal 58:219- 229. Bishop, C.A., Wine, J.J., Nagy, F. & O'Shea, MR. (1987). Physiological consequences of a peptide co-transmitter in a crayfish nerve-muscle preparation. WW 7 :1769-1779. Blair, K.L., Day, T.A., Lewis, M.C., Bennett, J .L. & Pax, R.A. (1991). Studies on muscle cells isolated flom Schistosoma mansoni: a Ca“-dependant K+ channel. W 102251-258. Boss, V., Conn, R]. (1992). Metabotropic excitatory amino acid receptor activation stimulates phospholipase D in hippocampal slices. W W 59:2340-2343. Bouvier, M., Szatkowski, M., Amato, A. & Attwell, D. (1992). The glial cell glutamate uptake carrier counter transports pH-changin g anions. Nam 360:471-474. Bouvier, M, Amato, A. & Attwell, D. (1994). Molecular biology and physiology of glutamate transporter. Wm: 6: 117-125. Carlyle, RF. (1974). The occurrence in and actions of amino acids on isolated supra oral sphincter preparations of the sea anemone Actina equina. Journal mm 236:635-652. Chamberlin, R. & Bridges, R. (1993). Conformationally constrained acidic amino acids as probes of glutamate receptors and transporters. In: W W Ed: Kozikowski, A.P., Raven Press, New York, pp. 231-259. Chappell, L. H. & Walker, E (1982). Schistosoma mansoni: Incorporation and metabolism of protein amino acids in vitro. Comparamflimhemimm Physiology 73B. 701 70.7 Cornford, EM. (1985). Schistosoma mansoni, S. japonicum, and S. haematobium: Permeability to acidic amino acids and effect of separated and unseparated adults. WW 59:355-363. 109 Comford, E. M. & Oldendorf, W. H. (1979). Transintegumental uptake of metabolic substrates 1n male and female Schistosoma mansoni. W98! 65. 357-363. Cox, D.W.G., Headley, M.H. & Watkins, J .C. (1977). Actions of L- and D- homocysteate in rat CNS: A correlation between low-affinity uptake and the time courses of excitation by microelectrophoretically applied L- glutamate analogues. 1W 29:579-588. Cull-Candy, S.G. & Parker, 1. (1982). Rapid kinetics of single receptor channels. Name 295:410-412. Day, T.A., Bennett, J .L. & Pax, R.A. (1994a). Serotonin and its requirements for maintenance of contractility in muscle fibres isolated flom Schistosoma mansoni. mm 108:425-432. Day, T.A., Maule, A.G., Shaw, C., Halton, .W., Moore, S., Bennett, J .L. & Pax, R.A. (1994b). Platyhelrninth F MRFamide-related peptides (F aRPs) contract Schistosoma mansoni (T rematoda: Di genea) muscle fibers in vitro. Parasitology 109:455-459. Dudel, J., Franke, C. & Hatt, H. (1990). A family of glutamatergic, excitatory channel types at the crayfish neuromuscular junction. mm Comparatiyalihxsiologam 166: 757-768. Eliasof, S. & Werblin, F. (1993). Characterization of the glutamate transporter in retinal cones of the Tiger Salamander. WW 13:402- 411. Fetterer, R.H., Pax, R.A. & Bennett, J .L. (1981). Na*-K+ transport, motility and tegumental membrane potential in adult male Schistosoma mansoni. mm 82:97-109. Foster, L.A., Chen, G.Z., VanDewaa, E.A., Pax, R.A. & BennetLJ.L. (1989). Glutamine- vs Glucose-Supported motor activity in Schistosoma mansoni: physiological relevance of aerobic metabolism. Experimentalfiamsitology 69:44-53. 110 F reund, W.D., Grieshop, B. & Neumann, SR. (1995). Glutamate-induced calcium responses in rat primary cortical cultures are potentiated by co- administration of glutamate transport inhibitors. W 188:61-64. Gallardo, M I Ferret, M I & Sanchez, J. (1994). Presence of an X-AG Carrier 1n flog (Rana esculenta) red blood cells. WM 139: 97-102. Garthwaite, J. (1985). Cellular uptake disguises action of L- glutamate on N- methyl-D-aspartate receptors. W 85:297-307. Gray, S.R., Batstone, F.R., Santiapillai, .F. & Richardson, R]. (1991). Solubilization and purification of a putative quisqualate sensitive glutamate receptor flom crustacean muscle. Biochemicaflmrmal 273:165-171. Hediger, M.A., Kanai, Y., You, G. & Nussberger, S. (1995). Mammalian ion- coupled solute transporters. MW 482:7S-l7S. Hoehn, K. & White, TD. (1990). Glutamate evoked release of endogenous adenosine flom rat cortical synaptosomes is mediated by glutamate uptake and not by receptors. Lugmhgm, 54:1716-1724. Inoue, K., Sakaitani, M., Shimada, S. & Tohyaama, M. (1995). Cloning and expression of a bovine glutamate transporter. MW 28: 343-348. Johnston, G.A.R., Kennedy, S.M.E. & Twitchin, B. (1979). Action of the neurotoxin kainic acid an high affinity uptake of L-glutamic acid in rat brain slices. Lflemghem. 32:121-127. Kanai, Y. & Hediger, MA. (1992). Primary structure and functional characterization of a hi gh-affinity glutamate transporter. Nam 360:467- 471. Kanai, Y., Smith, C.P. & Hediger, MA. (1993). The elusive transporters with a high affinity for glutamate. WW 16:365-370. 111 Kanner, B I & Schuldiner, S. (1987). Mechanism of transport and storage of neurotransmitters WM 22: l-38 Keenan, L. & Koopowitz, H. (1982). Physiology and in situ identification of putative aminergic neurotransmitters in the nervous system of Gyrocotyle fimbriata, a parasitic flatworm. leumalefNeumhiolegy 13:9-21. Klockner, U, Storck, T., Conradt, M. & Stoffel, W. (1993). Electrogenic L- glutamate uptake in Xenopus laevis oocytes expressing a cloned rat brain L- glutamate / L- aspartate transporter (GLAST-l). Wed Chemistry 268: 14594-14596. Klockner, U., Storck, T., Conradt, M. & Stoffel, W. (1994). Functional properties and substrate specificity of the cloned L- glutamate / L- aspartate transporter GLAST-1 flom rat brain expressed in Xenopus oocytes. Ihe loumalnfhleumssience 14:5759-5765. Lerner, J. (1978): A review of amino acid transport processes in animal cells and tissues. University of Maine at Orono Press, Orono, MN, p 72. MacDonald, A.G., Ramsey, R.L., Shelton, C.J. & Usherwood, P.N.R. (1992). Single channel analysis of ketamine interaction with a quisqualate receptor. WWW 210223-229- Marc, R.E. & Lam, D.M.K. (1981). Uptake of aspartic and glutamic acid by photoreceptors in goldfish retina. PEAS 78:7185-7189. Mishina, M., Sakimura, K., Mori, H., Kushiya, E., Harabayashi, M., Uchino,S. & Nagahari, K. (1991). A single amino acid residue determines the Ca2+ permeability of AMPA-selective glutamate receptor channels. Bmhemieal andBiaphxsicamesearchficmmum'caticm 180: 813- 821 Monaghan, D. T., Bridges, RA. & Cotman, C. W. (1989). The excitatory amino acid receptors Annuallieximflfihannamlmdloximlm 29: 365- 402. Nakanishi, S. (1992). Molecular diversity of glutamate receptors and implications for brain firnction. Seienee 258:597-603. 112 O’Donnel, S.R., Saar, N. (1978). The uptake kinetics and metabolism of extraneural noradrenaline in guinea-pi g trachea as studied with quantitative fluorescence microphotometry. BfitishleumaLQflmanaeelggy 62(2):23 5- 239. Prosser, C.L. (1973). Temperature. In: Cgmpamn'xg Animal thrsicl D . 1 mm Ed: Prosser, C.L. W.B. Saunders Company, Philadelphia, PA, pp, 363-364. Pines, G., Danbolt, N.C., Bjoras, M., Zhang, Y., Bendahan, A., Eide, E., Koepsell, H., Storm-Mathisen, J ., Seeberg, E. & Kanner, 3.1. (1992). Cloning and expression of a rat brain L- glutamate transporter. Nature 360:464-467. Riedel, G. & Reymann, K. (1993). An antagonist of the metabotropic glutamate receptor prevents LTP in the dentate gyrus of fleely moving rats. Neumphaunaeelegy 32:929-931. Rosenberg, P.A., Amin,S . & Leitner, M. (1992). Glutamate uptake disguises neurotoxic potency of glutamate agonists in cerebral cortex in dissociated cell culture. WW 12:56-61. Rotlrstein, J .D., Martin, L.J. & Kunci, R.W. (1992). Decreased glutamate transport by the brain and spinal cord in amyotrophic lateral sclerosis. Them EnglandlonmalnfMedicine 326:1464-1468. Thompson, D.P., Pax, R.A., Bennett, J .L. (1982). Microelectrode studies of the tegument and sub-tegumental compartments of the male Schistosoma mansoni: an analysis of electrophysiological properties. 2mm 85: 163-178. Sawada, M., Gibson, D. & McAdoo, D .J. (1984). L-glutamic acid, a possible neurotransmitter to anterior aorta of Aplysra MW 51: 375-3 86. Schoepp, D.D. & Conn, P.J. (1993). Metabotropic glutamate receptors in brain function and pathology. IiES 14:13-20. 113 Schuster, C.M., Ultsch, A., Schloss, P., Cox, J .A., Schmitt, B. & Betz, H. (1991). Molecular cloning of an Invertebrate glutamate receptor subunit expressed in Drosophila muscle. Seienee 254:1 12-114. Schwartz, E.A. & Tachibana, M. (1990). Electrophysiology of glutamate and sodium co-transport in a glial cell of the salamander retina. lmrmaLof Ehyeielegy 426:43-80. Scott, H.L., Tannenberg, A.E.G. & Dodd, PR. (1995). Variant forms of neuronal glutamate transporter sites in Alzheimer disease cerebral cortex. 1mm Neumehem. 64:2193-2202. Seeburg, PH. (1993). The TiPS/T'IN S lecture: The molecular biology of mammalian glutamate receptor channels. IiES 14:297-303. Shannon, H. E. & Sawyer, B D. (1989). Glutamate receptors of the N-methyl-D- aspartate subtype 1n the myenteric plexus of the guinea pig ileum. The . ; . . .z - ' 251: 518-523. Shashidharan, P., Huntley, G.W., Meyer, T., Morrison, H. & Plaitakis, A. (1994). Neuron-specific human glutamate transporter: molecular cloning, characterization and expression inhuman brain. Brainfieseareh 662:245- 250. Shinozaki, H. & Ishida, M. (1992). A metabotropic L- glutamate receptor agonist: Pharmacological difference between rat central neurons and crayfish neuromuscular junctions Compamixefiiochemimmmsiom 103C: 13-17. Solis-Soto, J .M. & Brink, M.D.J. (1994). Immunocytochemical study on biologically active neurosubstrates in daughter sporocysts and cercariae of T richobilharzia ocellata and Schistosoma mansoni. W 108:301- 31 1. Sprengel, R. & Seeburg, P. H. (1993). The unique properties of glutamate receptor channels Eederationnffiumncanfliochemicamociefies 325: 90-94. Storck, T., Schulte, S., Hofrnann, K. & Stoffel, W. (1992). Structure, expression, and functional analysis of a Na+ dependant glutamate laspartate transporter 114 from rat brain. W 89: 10955-10959. Swanson, R.A., Farrell, K. & Simon, RP. (1995). Acidosis causes failure of astrocyte glutamate uptake during hypoxia. WWW amLMemhelism 15:417-424. Takeuchi, A. & Takeuchi, N. (1964). The effect on crayfish muscle of iontophoretically applied glutamate. Wang): 170:296-317. Tanaka, K. (1993). Cloning and expression of a glutamate transporter flom mouse brain Neumscrencelattm 159: 183-186 Tanaka, K. (1994). Pharmacological characterization of a cloned rat glutamate transporter (GluT-l). MeleeulaLBrainReseanm 21:167-170. Thompson, C.S. & Mettrick, DP. (1989). The effects of 5-HT and glutamate on muscle contraction in Hymenolepis diminuta. W 67:1257-1262. Tielens, A.G.M. & VanDenBergh, SC. (1987). Glycogen metabolism in Schistosoma mansoni worms after their isolation flom the host. Meleeular andflochemieaLRarasitoloax 24:247-254- Villalobos, C. & Garcia-Sancho (1995). Glutamate increases cytosolic calcium in GH3 pituitary cells acting via a high-affinity glutamate transporter. EederationoffinmpeanfliochemicaLSmietiesm 815- 819 Wadiche, J .I., Amara, S.G. & Kavanaugh, M.P. (1995a). Ion fluxes associated with excitatory amino acid transport. Neuron 15:721-728. Wadiche, J .I., Arriza, J .L., Amara, S.G. & Kavanaugh, M.P. (1995b). Kinetics of a human glutamate transporter. Neumn 14:1019-1027. Walker, R. J. & Holden-Dye (1989). Commentary on the evolution of transmitters, receptors and 1on channels 1n vertebrates. WWW Ehysielegy 93A. 25- 39. 115 Webb, R.A. (1986). The uptake and metabolism of L-glutamate by tissue slices of the cestode Hymenalepis diminuta. W Bhysielegy 85C. 151- 162. Webb, RA. (1987). Innervation of muscle in the cestode Hymenolepis microstoma. W 65. 928-935. Webb, R.A. (1988). Release of exogenously supplied H3-glutamate and endogenous glutamate flom tissue slices of the cestode Hymenolepis diminuta. Canadiauloumalnffihxsinlmanifihaunmlogx 66:889-894- Webb, R.A. & Eklove, H. (1989). Demonstration of intense glutamate-like immunoreactivity in the longitudinal nerve cords of the cestode Hymenolepis diminuta. W 75:545-548. Winder, D.G., Conn, P.J. (1995). Metabotropic glutamate receptor (mGlu-R)- mediated potentiation of cyclic AMP responses does not require phosphoinositide hydrolysis: Mediation by a group II-like mGluR. Journal omeurochemlm 64:592-599- World Health Organization. WWW Regen UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases. (1995) Geneva, Switzerland. Wyllie, D.J.A., Mathie, A., Symonds, C.J. & Cull-Candy, SC. (1991). Activation of glutamate receptors and glutamate uptake in identified macroglial cells in rat cerebellar cultures. lmrmalgflfihysiglpgy 432:235-258. Yamaguchi, K. & Ohmori, H. (1990). Voltage-gated and chemically-gated ionic channels in the cultured cochlear ganglion neurone of the chick. lmrmalgf Physiology 420: 1 85-206. Zerangue, N., Kavanaugh, M. P. (in press). Interaction of L-cysteine with a human excitatory amino acid transporter. WWW.