Scientific paper
A Secreted Phospholipase A2 Binds to Calmodulin at Sub-micromolar Concentrations of Calcium
Uro{ Logonder,1 Jernej Jorga~evski,1 Gregor Anderluh,2 Uro{ Petrovi~,1
Igor Poberaj3 and Igor Križaj1'*
'Department of Molecular and Biomedical Sciences, Jožef Stefan Institute, Jamova cesta 39, SI-'000 Ljubljana, Slovenia
^Department of Biology, Biotechnical Faculty, University of Ljubljana, Večna pot ''', SI-'000 Ljubljana, Slovenia
3Department of Physics, Faculty of Mathematics and Physics, University of Ljubljana, Jadranska cesta '9, SI-'000 Ljubljana, Slovenia
* Corresponding author: E-mail: igor.krizaj@ijs.si, Tel: +386 ' 477 36 26, Fax: +386 ' 477 39 84
Received: 14-03-2008
Abstract
To determine the possibility that calmodulin, a cytosolic regulatory protein, and ammodytoxin, a neurotoxic secreted phospholipase A2, interact in vivo, we studied the dependence of their association on Ca2+. The interaction between the two proteins was positively dependent on Ca2+, and greatest at millimolar concentrations of this ion. Importantly, they interacted already in the presence of sub-micromolar concentrations of Ca2+, as demonstrated by affinity labelling, laser tweezers and surface plasmon resonance. Tight binding of the secreted phospholipase A2 to calmodulin is therefore expected on depolarization of the axolemma, when the concentration of free Ca2+ at active zones rises to 100 ^M. These results strengthen the proposal that binding of ammodytoxin to calmodulin is a step in the process of presynaptic toxicity of this and related phospholipases A2. Moreover, they suggest that some other (patho)physiological effects induced by endogenous secreted phospholipases A2 could be also due to their interaction with calmodulin in the cytosol of cells.
Keywords: Secreted phospholipase A2, Ammodytoxin, Calmodulin, Neurotoxicity, Ca2+-dependence, Cytosol
1. Introduction
Secreted phospholipases A2 (sPLA2s; EC 3.1.1.4) hydrolyse the ester bond at the sn-2 position of 1,2-diacyl-sn-3-phosphoglycerides in a Ca2+ dependent manner. In spite of considerable structural similarity, these enzymes display a remarkable variety of physiological and pathological activities.1'2 Some sPLA2s are potent presynaptic neurotoxins (ß-neurotoxins). Abundantly present in snake venoms, they inhibit the communication between nerves and muscles. The most obvious effects that ß-neurotoxins induce in the nerve terminal are a reduction in the number of synaptic vesicles, appearance of Q-shaped invagina-tions in the presynaptic membrane, damage of mitochondria, inhibition of synaptosomal protein phosphorylation and influence on certain terminal ion currents.3 Due to this complexity, the molecular basis of ß-neurotoxicity is still
largely unknown. It is clear however that the phospholipa-se activity and the specific binding of ß-neurotoxins at the nerve ending are both essential for ß-neurotoxicity.4 We demonstrated that ammodytoxin (Atx), a ß-neurotoxin from Vipera a. ammodytes venom, enters the cytosol of rat hippocampal neurons and that this multi-disulfide-bridged toxin is stable under cytosolic conditions.56 These findings strengthen the proposal that ß-neurotoxins act inside nerve cells and that the binding of sPLA2s to certain in-tracellular proteins3 is physiologically relevant. Calmodu-lin (CaM) is one of the most interesting of these proteins. Present in the cytosol of eukaryotic cells, it occupies a central position in the regulation of many cellular processes, including neurotoxic sPLA2-affected processes such as exocytosis, endocytosis, protein phosphorylation and ion channel conductivity.7 We have observed that all the
Atx mutants expressing high neurotoxicity also show high binding affinity for CaM,8 suggesting that binding of Atx to CaM is important for the ß-neurotoxicity.
CaM and sPLA2s bind Ca2+ and their activities and binding properties are influenced by its concentration. The association between Atx and CaM was established in the presence of millimolar concentration of Ca2+.9 The intracellular concentration of Ca2+ ([Ca2+]), where the interaction would have to take place physiologically, is however ~104-fold lower. We therefore decided to determine whether the [Ca2+] that exists in the cytosol of a eukaryo-tic cell could support the binding of Atx to CaM.
2. Materials and Methods
2. 1. Materials
Ammodytoxin (Atx) was purified from Vipera a. ammodytes venom.10 Native and recombinant vertebrate CaM were used in our experiments - human brain CaM was purchased from Calbiochem (USA) while mouse CaM was prepared in our laboratory by recombinant expression in E. coli as described.11 All other reagents and chemicals used were of analytical grade.
2. 2. Determination of [Ca2+]
[Ca2+] in buffers was determined using flame atomic absorption spectrometry (FAAS) on a Varian Spectra AA 110 (Mulgrave, Australia) atomic absorption spectrometer, as described.12
on 12.5% (w/v) polyacrylamide gels, which were dried and autoradiographed at -70 °C using Kodak BioMax Light Film (Sigma-Aldrich, USA).
2. 5. Immobilization of Proteins on Polystyrene Microspheres
Amino-modified polystyrene microspheres (IDC, USA), having a diameter of 0.97 pm, were coated with Atx and those with a diameter of 1.80 pm with CaM. The 4% (w/v) stock suspension of microspheres was diluted 10-fold and incubated with 10% (w/v) glutaraldehyde for 80 min with gentle mixing (120 rpm) at room temperature. Microspheres were rinsed twice with doubly distilled deionised water supplemented with 1 mM Ca2+ and resus-pended in 150 pl of this solution. Optimal surface density of each attached ligand on the spheres was found empirically. It was achieved when 1% of the amount of each protein calculated to completely cover the microspheres was added to suspensions of the appropriate microspheres, which were then incubated overnight at 4 °C with constant stirring (120 rpm). The microspheres were rinsed three times with 150 pl of Ca2+/water and resuspended. BSA was added to the suspension to a final concentration of 5 mg/ml. Samples were stirred (120 rpm) for 45 minutes at room temperature to completely react the remaining reactive groups. Samples of each size of microsphere were finally rinsed three times with 150 pl of 50 mM Tris-HCl, pH 8.4, resuspended in 150 pl of buffer and then mixed together. Derivatized microspheres were stored at 4 °C for a maximum of one week.
2. 3. Radioiodination of Atx
Atx was radioiodinated to a specific activity of 52 Ci/mmol as described.13 1251-Atx was shown to be identical to the native Atx in enzymatic, neurotoxic and immu-nological properties.
2. 4. Cross-linking of 125I-Atx to CaM
Vertebrate CaM was incubated at 20 nM for 30 min at room temperature with 10 nM 1251-Atx in the labelling buffer (75 mM Hepes, pH 7.4, 150 mM NaCl and 0.15% (w/v) Triton X-100). The [Ca2+] in the buffer was either zero (no addition of Ca2+), 50 pM, 100 pM, 1 mM or 10 mM. One experiment was performed in the labelling buffer containing 1 mM EGTA, which complexed both free and protein bound Ca2+. The control experiment was performed in the presence of a 500-fold molar excess of unla-belled Atx over the labelled toxin in the labelling buffer containing 1 mM Ca2+. Freshly dissolved disuccinimidyl suberate (DSS) (Pierce, USA) in DMSO was added to all samples to a final concentration of 100 pM and, after 5 min, the cross-linking reaction was stopped by the addition of SDS-PAGE sample buffer. Samples were analyzed
2. 6. Laser Tweezers
Laser tweezers enable microscopic dielectric particles to be brought and held together - "trapping" - by using tightly focused laser beams. The trapping mechanism relies on a strong electric field gradient in a focal region of tightly focused laser beam that attracts dielectric particles if their refractive index is higher than that of surrounding medium.14-16 The maximum laser trap holding force is proportional to the laser beam intensity.
Laser tweezers were built around an inverted microscope Axiovert 200M (Zeiss, Germany). A continuous wave infrared laser Compass 2500MN with a wavelength of 1064 nm (Coherent, USA) was introduced into the microscope through the bottom port and was tightly focused on the sample by a water immersion microscope objective Acroplan 63W - NA 0.9 (Zeiss, Germany). Because of the low water absorption at 1064 nm, the sample temperature in the trap centre rose by less than 1°C. The position and intensity of the trap were precisely controlled at sub-nanometer resolution by acousto-optic deflectors DTD 274HA6 (1ntraaction, USA) driven by a dedicated beam-steering controller BSC160 (Aresis, Slovenia). Two laser traps were generated by time-multiplexing the laser beam
between the traps with a 50 kHz switching rate. Bright field sample images were recorded with a high resolution CMOS camera PL A741 (PixeLink, Canada).
The affinity between Atx and CaM was measured using polystyrene beads of two different sizes, the smaller coated with Atx and the larger with CaM. The beads served as handles for laser tweezers. Coated beads of each type were mixed together in buffer and introduced into the sample cell, made of a microscopic slide and a cover slip glued together with UV-curing glue. All glass surfaces, as well as the polystyrene beads, were coated with a layer of BSA to prevent the beads sticking to the glass and to each other. The interaction between Atx and CaM molecules was established by bringing and keeping two corresponding beads together for a few seconds. The formation of the non-covalent bond was checked by switching off the laser tweezers and observing whether the beads drifted apart due to Brownian motion or not.
2. 7. Surface Plasmon Resonance
Surface plasmon resonance (SPR) experiments were performed using a Biacore X system (Biacore AB, Sweden). Atx was covalently immobilized on the surface of flow cell 1 located on a Sensor Chip CM5 from Biacore. Flow cell 2 was mock-immobilized with ethanolamine and used as a reference surface. HBS-EP buffer (0.01 M HEPES, pH 7.4, 0.15 M NaCl, 3 mM EDTA, 0.005% Surfactant P20) was used to immobilize approximately 750 response units (R.U.) of protein. SPR experiments were performed at 25 °C in 50 mM Tris-HCl buffer, pH 8.4 or 7.4, and different [Ca2+]s. The interaction between CaM and the chip-immobilized Atx was studied by injecting CaM (0.5, 1, 5, 10, 20 and 30 pM) over the chip at different [Ca2+]s (0, 0.1, 1 and 10 mM). Association and dissociation were followed at a flow-rate of 30 pl/min for 1 min and 3 min respectively. The chip was regenerated between successive experiments with 30 pl of 10 mM NaOH. The experimental curves were fitted according to a 1:1 interaction model using the Biacore BIAevaluation 3.2 software.
3. Results
3. 1. [Ca2+]-dependence of Association Between 125I-Atx and CaM Detected by Chemical Cross-linking
Using chemical cross-linking, the complex between Atx and CaM was formed only when Ca2+ ions were present in the incubation mixture (Figure 1, lanes 1 to 5). Removal of free Ca2+ ions by EGTA (Figure 1, lane 6) prevented its formation. Using this technique, we scanned the [Ca2+]-dependence of binding between Atx and CaM over the range of 50 pM to 10 mM Ca2+. The amount of the complex increased from 50 pM to 10 mM Ca2+. Atx and
CaM contain one and up to four bound Ca2+ ions, respectively. To verify whether just structurally bound Ca2+ suffices for complex formation or not 20 nM CaM and 10 nM Atx were mixed in the absence of externally added Ca2+, meaning that at most 90 nM Ca2+ (20 x 4 nM on CaM and 10 x 1 nM on Atx) was present during the incubation. Following cross-linking the adduct formation was clearly visible (Figure 1, lane 7) demonstrating that sub-micromolar concentration of free Ca2+ is enough for both molecules to interact. The possibility that the variation of [Ca2+] affected the yield of the cross-linking reaction and not only the affinity between Atx and CaM was excluded by the laser tweezers and SPR experiments described below.
Figure 1: 125I-Atx affinity labelling of CaM as a function of Ca2+ concentration. CaM was affinity labelled with 125I-Atx as detailed under Materials and methods in the presence of different [Ca2+]s (lane 1: 1 mM, lane 2: 50 pM, lane 3: 100 pM, lane 4: 1 mM, lane 5: 10 mM, lane 7: 0 M) or EGTA (lane 6: 1 mM). In lane 1, a 500fold molar excess of Atx over 125I-Atx was present during the incubation to define the specific signals. On autoradiogram, positions of the specific adduct and unbound 125I-Atx are shown by arrowheads.
3. 2. [Ca2+]-dependence of Association Between Atx and CaM Determined by Laser Tweezers
After the interaction between Atx and CaM was ascertained (see under laser tweezers in Methods) we measured its strength by laser tweezers. This was done by holding two beads linked by non-covalent Atx -CaM bond in two separate optical traps while the separation between the trap centres was slowly increased. The laser power was kept constant during the measurement. If the link between the beads did not break, another measurement with a higher laser power was performed until the link broke or the maximum laser power was reached. Non-breaking of the link was attributed to the formation of multiple non-covalent Atx - CaM bonds and the experiment was discarded. The laser power at which the bond broke was taken as a quantitative measure of the bond strength. The experiments were repeated at different [Ca2+] in solution while keeping other conditions constant.
Bond strengths were measured for [Ca2+] of 0 pM (0.15, P < 0.05), 1 pM (0.33, P < 0.05), 10 pM (0.44, P = 0.079) and 100 pM (0.54, P = 0.079). At least 20 measurements were made at each concentration of calcium. In Figure 2, the values are expressed relative to the value obtained in the absence of externally added Ca2+ (of 0 pM). The bond strength at 100 pM Ca2+ was 3.7 times greater than that at 0 pM Ca2+. The laser tweezers experiments thus also demonstrated that Atx and CaM interact already when only the protein-bound Ca2+ is present. Adding additional Ca2+ to the solution led to stronger non-covalent Atx - CaM bond. Reliable measurements could not be obtained above 100 pM Ca2+, the most probable reason being that the bond strength was out of the range of our measurement capabilities. Those results were therefore discarded.
Due to methodological reasons the laser tweezers experiments were carried out at pH 8.4. The SPR measurements presented below showed that the complex between Atx and CaM is equally stable at pH 7.4 and pH 8.4.
the affinity labelling and laser tweezers results. Within the limits of experimental error, the results were identical using the binding buffer with the pH of 7.4 or 8.4. Association rate constant (ka) for the interaction between Atx and CaM increased about 10-fold in the range from 0 to 10 mM Ca2+ in the buffer. On the other hand the dissociation rate constant (kd) did not change by more than 10% over the same range of [Ca2+]. Accordingly, the dissociation constant Kd for the interaction was about 10-fold higher at sub-micromolar [Ca2+] than at millimolar Ca2+.18
Figure 3: The [Ca2+]-dependence of interaction between Atx and CaM measured by surface plasmon resonance. Atx was covalently immobilized on the sensor Chip CM5. Sensorgrams were recorded using 20 pM human CaM in 50 mM Tris-HCl containing, from (a) to (d), 0 mM, 0.1 mM, 1 mM and 10 mM Ca2+. The association was followed for 1 min and the dissociation for three minutes.
4. Discussion
Figure 2: [Ca2+]-dependence of the relative bond strength between Atx and CaM molecules as measured with laser tweezers. Two beads linked together by a non-covalent Atx - CaM bond were pulled apart with laser tweezers and the force to break the interaction was measured. Experiments were performed at four different [Ca2+]s (0, 1 pM, 10 pM, 100 pM) while keeping other parameters constant. At each [Ca2+] at least 20 measurements were made and the SD is shown by error bars.
3. 3. [Ca2+]-dependence of Association Between Atx and CaM Determined by SPR
An SPR sensor chip was prepared with Atx immobilized on the surface. High concentration of 20 pM CaM, which is however in the range of the physiological concentration of CaM in the cytosol of a mammalian cell,17 was needed to observe the interaction clearly at lower [Ca2+]s. The [Ca2+]-dependence of association between Atx on the chip and CaM is shown in Figure 3. Most importantly, interaction between the two proteins is seen in the absence of added Ca2+. The extent of complex formation increased with increasing [Ca2+], in agreement with
Three techniques, affinity labelling, laser tweezers and SPR, were employed to determine the [Ca2+] dependence of Atx - CaM binding. The combined results showed that a complex (1 : 1 according to native electrophoresis performed at pH 7.4 and 8.4, results not shown) is formed as long as the structurally bound Ca2+ ions are present. The extent of complex formation depended on the [Ca2+] in solution, as is evident from the affinity labelling and SPR experiments (Figures 1 and 3). In the same way, as indicated by the laser tweezers and SPR results, the binding of Atx to CaM increased with increasing [Ca2+]. In the absence of added Ca2+, the Kd for binding of Atx to CaM was sub-micromolar, which would ensure interaction with CaM in the mammalian cytosol, where its concentration can be as high as 10 pM.17
The most obvious general effect of presynaptically toxic sPLA2s is impairment of synaptic vesicle cycling. It appears that these toxins promote Ca2+-regulated neurosecretion and block the retrieval of synaptic vesicles from the plasma membrane of the motoneuron.3 The first evidence that CaM is required in neurosecretion came from functional dissection of exocytosis in model systems such as permeabili-zed bovine chromaffin cells and in rat PC12 cells.19,20 Later,
it was found that CaM may provide one of the links between Ca2+-signalling and SNARE assembly. The binding site on the v-SNARE synaptobrevin for Ca2+-loaded CaM (Ca2+/CaM) was identified.21 Precisely the same site binds to acidic phospholipids, and lipids can be competitively displaced by Ca2+/CaM. Approaches using both mimetic peptide injection and directed mutagenesis in neuroendocrine cells indicated that CaM and/or phospholipid interactions with synaptobrevin are required for Ca2+-dependent exocyto-sis.17,22,23 In neurons some evidence points also to regulation by CaM of the refilling of readily releasable vesicle pool24 while, in adrenal chromaffin cells, it was found that CaM is the Ca2+ sensor for rapid endocytosis.25 In addition, presy-naptically toxic sPLA2s inhibit protein phosphorylation26 and reduce terminal K+-currents in motoneurons27, both processes being regulated by CaM in neuronal cells.72829
On depolarization of the presynaptic membrane, the intracellular free [Ca2+] increases locally up to 100 pM. According to our results this is the calcium ion concentration at which Atx and CaM form a tight complex. It is thus possible that, by binding to CaM, Atx affects exo-/en-docytosis, protein phosphorylation and the conductivity of certain ion channels in motoneurons. Conversely, since CaM is known as the modulator of activity of various enzymes7, it could also be expected to influence the phospholipase activity of sPLA2s.
Atx enters the cytosol, as demonstrated in hippo-campal neurons.5 Neurotoxic sPLA2s are members of group IA, IIA and IIIA sPLA2s3 and are structurally closely related to mammalian group I, II and III sPLA2s.30 Endogenous sPLA2s have been detected in cellular compartments on the way into which they should pass the cytosol. For example, group I sPLA2 was translocated from outside the normal rat uterus stromal cell (UIII cell line) to its nucleus31, group IIA sPLA2 was detected in rat brain cortex mitochondria32 and group V sPLA2 was found in nuclei of PC12 and U251 astrocytoma cells.33 Interaction between CaM and mammalian endogenous sPLA2s has not been evaluated, but some of the various effects that these enzymes produce in different types of cells2 may well be the result of their interaction with CaM in the cytosol of these cells.
In this study we demonstrated that already sub-mi-cromolar [Ca2+]s support the interaction between CaM and the snake venom sPLA2. The presented results envisage the existence of CaM-mediated (patho)physiologi-cal/pharmacological effects of sPLA2s, among them also presynaptic toxicity.
5. Acknowledgments
This work was supported by grant P1-0207 from the Slovenian Research Agency. Surface plasmon resonance experiments were performed in Infrastructural Centre for Surface Plasmon Resonance at the Department of Bio-
logy, Biotechnical Faculty, University of Ljubljana. We sincerely thank Dr. Janez Ščančar for his assistance with FAAS measurements, Dr. Dušan Babič for help with design of equipment and Natan Osterman for help with laser tweezers. We are grateful to Dr. Roger H. Pain for critical reading of the manuscript.
6. References
1.	R. M. Kini (Ed.): Venom Phospholipase Aj Enzymes: Structure, Function and Mechanism, John Wiley & Sons, Chichester, U. K., 1997.
2.	M. Murakami, I. Kudo, Biol. Pharm. Bull. 2004, 27, 11581164.
3.	J. Pungerčar, I. Križaj, Toxicon 2007, 50, 871-892.
4.	I. Križaj, F. Gubenšek, Biochimie 2000, 82, 807-814.
5.	U. Petrovič, J. Šribar, A. Pariš, M. Rupnik, M. Kržan, N. Var-djan, F. Gubenšek, R. Zorec, I. Križaj, Biochem. Biophys. Res. Commun. 2004, 324, 981-985.
6.	U. Petrovič, J. Šribar, M. Matis, G. Anderluh, J. Peter-Kata-linic, I. Križaj, F. Gubenšek, Biochem. J. 2005, 391, 383388.
7.	D. Chin, A. R. Means, Trends Cell Biol. 2000, 10, 322-328.
8.	P. Prijatelj, J. Šribar, G. Ivanovski, I. Križaj, F. Gubenšek, J. Pungerčar, Eur. J. Biochem. 2003, 270, 3018-3025.
9.	J. Šribar, A. Čopič, A. Pariš, N. E. Sherman, F. Gubenšek, J. W. Fox, I. Križaj, J. Biol. Chem. 2001, 276, 12493-12496.
10.	F. Gubenšek, A. Ritonja, J. Zupan, V. Turk, Period. Biol. 1980, 82, 443-447.
11.	T. A. Craig, D. M. Watterson, F. G. Prendergast, J. Haiech, D. M. Roberts, J. Biol. Chem. 1987, 262, 3278-3284.
12.	J. Ščančar, R. Milačič, M. Benedik, P. Bukovec, Clin. Chim. Acta 2000, 293, 187-197.
13.	I. Križaj, J. O. Dolly, F. Gubenšek, Biochemistry 1994, 33, 13938-13945.
14.	A. Ashkin, J. M. Dziedzic, J. E. Bjorkholm, S. Chu, Optics Lett. 1986, 11, 288-290.
15.	K. C. Neuman, S. M. Block, Rev. Sci. Instr. 2004, 75, 2787-2809.
16.	J. R. Moffitt, Y. R. Chemla, S. B. Smith, C. Bustamante, An-nu. Rev. Biochem. 2008, 77, doi: 10.1146/annurev.bioc-hem.77.043007.090225.
17.	L. De Haro, S. Quetglas, C. Iborra, C. Leveque, M. Seagar, Biol. Cell. 2003, 95, 459-464.
18.	T. Petan, I. Križaj, F. Gubenšek, J. Pungerčar, Biochem. J. 2002, 363, 353-358.
19.	L. H. Chamberlain, D. Roth, A. Morgan, R. D. Burgoyne, J. Cell. Biol. 1995, 130, 1063-1070.
20.	Y. A. Chen, V. Duvvuri, H. Schulman, R. H. Scheller, J. Biol. Chem. 1999, 274, 26469-26476.
21.	S. Quetglas, C. Leveque, R. Miquelis, K. Sato, M. Seagar, Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 9695-9700.
22.	S. Quetglas, C. Iborra, N. Sasakawa, L. De Haro, K. Kuma-kura, K. Sato, C. Leveque, M. Seagar, EMBO J. 2002, 21, 3970-3979.
23.	L. De Haro, G. Ferracci, S. Opi, C. Iborra, S. Quetglas, R. Miquelis, C. Leveque, M. Seagar, Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 1578-1583.
24.	T. Sakaba, E. Neher, Neuron 2001, 32, 1119-1131.
25.	C. R. Artalejo, A. Elhamdani, H. C. Palfrey, Neuron 1996, 16, 195-205.
26.	E. Ueno, P. Rosenberg, Toxicon 1996, 34, 1219-1227.
27.	E. G. Rowan, A. L. Harvey, British J. Pharmacol. 1988, 94, 839-847.
28.	S. S. Hook, A. R. Means, Annu. Rev. Pharmacol. Toxicol. 2001, 41, 471-505.
29.	Y. Saimi, C. Kung, 2002, 64, 289-311.
30.	M. Rouault, J. G. Bollinger, M. Lazdunski, M. H. Gelb, G. Lambeau, Biochemistry 2003, 42, 11494-503.
31.	O. Specty, J. F. Pageaux, M. Dauja, M. Lagarde, C. Laugier, J. M. Fayard, FEBS Lett. 2001, 490, 88-92.
32.	L. Macchioni, L. Corazzi, V. Nardicchi, R. Mannucci, C. Arcuri, S. Porcellati, T. Sposini, R. Donato, G. Goracci, J. Biol. Chem. 2004, 279, 37860-37869.
33.	V. Nardicchi, L. Macchioni, M. Feririni, G. Goracci, Bio-chim. Biophys. Acta 2007,1771, 1345-1352.
Povzetek
Z namenom demonstracije, da se amoditoksin, nevrotoksicna sekretorna fosfolipaza Aj, lahko veže na kalmodulin, re-gulatorni protein v citosolu, tudi in vivo, smo raziskali odvisnost tvorbe kompleksa med tema dvema proteinoma od Ca2+. Ugotovili smo, da je interakcija med proteinoma pozitivno odvisna od koncentracije teh ionov, afiniteta vezave pa najvišja v območju milimolarnih koncentracij. Ključnega pomena je ugotovitev, da interakcija med kalmodulinom in fosfolipazo Aj obstaja že v prisotnosti sub-mikromolarne koncentracije Ca2+, kar smo demonstrirali s pomočjo afinitet-nega označevanja, z lasersko pinceto in tehniko površinske plazmonske rezonance. Ob depolarizaciji aksoleme se lokalno, na področju aktivnih con, intracelularna koncentracija prostega Ca2+ dvigne do 100 pM, kar zagotavlja nastanek stabilnega kompleksa med kalmodulinom in nevrotoksično fosfolipazo Aj. Predstavljeni rezultati pritrjujejo hipotezi, da je vezava amoditoksina na kalmodulin korak v procesu presinaptične nevrotoksičnosti te in sorodnih nevrotoksičnih sekre-tornih fosfolipaz Aj. Kažejo pa tudi na možnost, da so nekateri, mehanistično še nepojasnjeni (pato)fiziološki procesi, ki jih povzročajo endogene sekretorne fosfolipaze Aj, posledica interakcije teh encimov s kalmodulinom v celičnem ci-tosolu.