Study of the Interaction between Different Phosphodiesterases and

794

Chem. Res. Toxicol. 2006, 19, 794-800

Study of the Interaction between Different Phosphodiesterases and Yessotoxin Using a Resonant Mirror Biosensor Marıá-Jose´ Pazos,†,‡ Amparo Alfonso,† Mercedes R. Vieytes,§ Takeshi Yasumoto,| and Luis M. Botana*,†,⊥ Departamentos de Farmacologıá y Fisiologıá, and AÄ rea de Apoyo a la InVestigacio´ n Agrobiolo´ gica, Facultad de Veterinaria, UniVersidad de Santiago, 27002 Lugo, Spain, Japan Food Research Laboratories, Tama, Tokyo 206-0025, Japan, and Community Reference Laboratory for Marine Biotoxins, Vigo, Spain ReceiVed NoVember 24, 2005

Yessotoxins (YTXs) are disulfated polyether toxins that were first isolated from scallops in Japan. It has been proposed that these toxins activate cellular phosphodiesterases (PDEs). The interaction between YTX and PDEs was confirmed by resonant biosensor and fluorescence polarization studies. The aim of this work is to study the specificity of different PDEs for YTX binding. Association measurements are done in a resonant mirror biosensor. The instrument detects changes in the refractive index and/or thickness occurring within a few hundred nanometers from the sensor surface where the association PDEs-YTX takes place. We use aminosilane cuvettes, where exonuclease Phosphodiesterase I from Crotalus atrox (PDE I), exonuclease Phosphodiesterase II from bovine spleen (PDE II), or phosphodiesterase 3′,5′cyclic-nucleotide-specific from bovine brain (PDEs) are immobilized. Over immobilized exonuclease PDE I and exonuclease PDE II are added different amounts of YTX, and typical association curve profiles are observed. These association curves fit a pseudo-first-order kinetic equation where the apparent association rate constant (kon) can be calculated. The value of this constant increases with YTX concentration. From the representation of kon versus YTX concentration, the association rate constant (kass) and the dissociation rate constant (kdiss) are obtained. From these values, the kinetic equilibrium dissociation constant (KD) of the YTX-PDE association can be calculated, indicating the affinity between them. The specificity of cyclic nucleotide PDE families is studied using different inhibitors that are added over immobilized cyclic nucleotide PDEs. In these conditions, changes in the association PDEsYTX curves are detected. The results show YTX affinity by cyclic nucleotide PDE 1, PDE 3, PDE 4, and exonuclease PDE I. Introduction Yessotoxins (YTXs) are lipophilic polyether algal toxins first isolated from the scallop Patinopecten yessoensis. Its planar structure was determined in 1987 (1), and the absolute configuration was recently determined (2, 3). The structures of the YTXs differ from those of toxins in the diarrhetic shellfish poisoning (DSP) toxins group and the pectenotoxins but are similar to the brevetoxins and ciguatoxins in having a laddershaped polycyclic ether skeleton (2). The YTX molecule, Scheme 1, has 11 contiguously ether rings and an unsaturated side chain with different radicals. Significant research progress made in recent years about the occurrence of YTX in shellfish from different marine areas indicates that YTX contamination of shellfish can be more significant that previously thought. In addition, more than 50 different YTX analogues have been recently identified (4). YTX was initially classified among the diarrhetic toxins because it often coexists with okadaic acid (OA), the major DSP toxin, and its derivatives. YTX gives positive results when tested * To whom correspondence should be addressed. Phone/fax: +34 982 252 242. E-mail: [email protected]. † Departamento de Farmacologıá, Universidad de Santiago. ‡A Ä rea de Apoyo a la Investigacioń Agrobiolo´gica, Universidad de Santiago. § Departamento de Fisiologıá, Universidad de Santiago. | Japan Food Research Laboratories. ⊥ Community Reference Laboratory for Marine Biotoxins.

by the conventional mouse bioassay method for detecting DSP toxins in shellfish (5). However, the few existing studies showed that, unlike OA and its derivatives, YTX does not cause diarrhea both in adult and in suckling mice (1, 5). Anyway, the potential human toxicity of these compounds is not completely defined: only few toxicological data are available for YTX. The analogues 45-hydroxyyessotoxin (45-OH-YTX) and 1-desulfoyessotoxin have lower toxicity than YTX (6). On the other hand, no clear toxic effect has been demonstrated in mice after oral exposure of very high doses of YTX (7). The current EU regulatory limit of 1 mg/kg of YTXs equivalents was introduced in March 2002 (8). The mechanism of action of YTX is not completely known. It was reported that YTX decreases adenosine 3′,5′ cyclic monophosphate (cAMP) by activation of cellular phosphodiesterases (PDEs) (9). Furthermore, YTX is unable to affect the protein phosphatase 2A (5), thus having a mechanism of action different from that of DSP toxins. Other studies indicated that YTX increases calcium influx through voltage-dependent channels and inhibits activated calcium entry in human lymphocytes (10). In addition, YTX increases, in the same cell line, the elevation of cytosolic calcium induced by maitotoxin through a different mechanism of capacitative calcium entry (11). These results show that the in vitro mechanism of action for YTX is specific and related to calcium modulation. cAMP is a second messenger related to early activation pathways in intracellular signaling. It is generated at the cytosol

10.1021/tx0503303 CCC: $33.50 © 2006 American Chemical Society Published on Web 06/03/2006

Interaction between Yessotoxin and Phosphodiesterases

Chem. Res. Toxicol., Vol. 19, No. 6, 2006 795

Scheme 1. Structure of Yessotoxin and Various Analogues

side of the plasma membrane through the action of adenylate cyclases. Cells regulate cAMP levels by a balance between adenylyl cyclases (synthesis) and PDEs (hydrolysis). There are 11 different 3′,5′ cyclic nucleotide PDE families (PDEs 1-11) with different substrate specificity, affinity, sensitivity to inhibitors, and tissue localization. PDE families 5, 6, and 9 hydrolyze guanosine 3′,5′ cyclic monophosphate (cGMP) (12), PDE families 1, 3, 4, 7, and 8 hydrolyze cAMP (13, 14), and PDEs 2, 10, and 11 hydrolyze both cAMP and cGMP (15, 16). All cAMP PDEs have been localized in brain, even though not all of them are there highly expressed (13, 15, 17, 18). In addition, there are exonucleases such as oligonucleate 5′ nucleotidohydrolase (PDE I) and oligonucleate 3′ nucleotidohydrolase (PDE II) that hydrolyze cyclic nucleotides. PDE I hydrolyzes 5′ mononucleotides from 3′-hydroxy-terminated riboand deoxyribo-oligonucleotides (19). This enzyme has been utilized as a tool for structural and sequence studies of nucleic acids (20). PDE II hydrolyzes 3′ phosphomononucleotides from

oligonucleotides containing a 5′-hydroxyl terminus (21). This PDE is very important in the characterization of a polynucleotide chain length, base composition, and identity of terminal nucleotide (20, 22-24). Recently, we have described a new method to detect and quantify YTX-like activities. This method is based on the specific interaction of YTX and immobilized cyclic nucleotide PDEs on a biosensor surface (25). The kinetic values for the association of cyclic nucleotide PDEs and different YTX analogues were studied using a resonant mirror biosensor, and the results indicate a structure-activity relationship. The presence of the radicals in the side chain decreases the effect of YTX (26) and the cyclic nucleotide PDEs-YTX association, indicating that the side chain plays an important role in the YTX activity on cyclic nucleotide PDEs (27). To further understand the interaction between YTX and PDEs, in this paper, we focus our study on the specificity different PDEs families for YTX by using the optical biosensor. This

796 Chem. Res. Toxicol., Vol. 19, No. 6, 2006

equipment uses a sensitive waveguide-based sensor, the resonant mirror, where specific interaction between molecules can be followed in real time.

Materials and Methods Chemicals. YTX was purified by Dr. T. Yasumoto. Bovine serum albumin (BSA) was from Biomedicals, Inc. (USA). IAsys aminosilane cuvettes were from Thermo-Labsystems (UK). Etazolate and milrinone were from Alexis Corp. (USA). Rolipram, erythro-9-(2-hydroxy-3-nonyl)adenine hydrochloride (EHNA hydrochloride), and cilostamide were from Biogen Cientı´fica (Madrid, Spain). Phosphodiesterase 3′,5′-cyclic-nucleotide-specific from bovine brain (PDEs), venom exonuclease from Crotalus atrox (PDE I), spleen exonuclease from bovine spleen (PDE II), calmodulin, 3-isobutyl-1-methylxanthine (IBMX), glutaraldehyde solution 25% in water, and all of the other chemicals were from Sigma Chemical Co. (Madrid, Spain). The composition of phosphate buffer (PB) was (in mM): Na+ 20; HPO4 2- 10. The final pH was adjusted to 7.7. Deionized water (Milli-Q) was used for all experiments. Operation of the Optical Biosensor System. Association measurements were made by using an IAsys Affinity Sensor (Labsystem, UK), following the procedures recommended by the manufacturer. The instrument detects changes in refractive index and/or thickness occurring within a few hundred nanometers from the sensor surface. The sensor relies upon a biomolecule being attached to the sensor surface. Ligand immobilization can be followed in “real-time” with the instrument producing a plot of response, measured in arc seconds (arc sec). This was performed using ligands coupled to aminosilane cuvettes in a total volume of 60 µL. The stirred cuvette system used in this instrument ensures that mass-transport effects during binding are minimized. The datasampling interval was 0.3 s controlled by the IAsys software. The temperature was constant at 22 °C, and the stirring rate was 85 rpm. The optical biosensor has dual cuvette surfaces having simultaneous dual-channel monitoring. Aminosilane Surfaces Activation and Ligand Immobilization. The ligands, cyclic nucleotide PDEs, exonuclease PDE I or PDE II, were immobilized directly to an aminosilane surface. This direct immobilization was performed as follows. After initial preequilibration of an aminosilane cuvette with 10 mM phosphate sodium buffer, pH 7.7, the silanized surface was activated by 30 min incubation with glutaraldehyde (4.2% v/v) and later washed with PB. After surface activation, 0.5 mg/mL PDE solved in PB was added and allowed to bind for 30 min. Remaining activated sites were then blocked (nonspecific binding) with BSA (2 mg/mL) for 10 min, after which the BSA solution was replaced with PB. The residual noncovalently bound ligand was released from the surface by pretreatment with HCl 0.01 N for 2 min, after which the regeneration solution was replaced by PB. Monitoring the Association/Dissociation of Toxins to Immobilized Ligand. Cuvette Regeneration. A range of YTX concentrations in PB pH 7.7 were allowed to bind to the immobilized ligand, and the association curves were monitored for 20-25 min. Subsequently, the cuvette was washed with PB, and the dissociation phase was monitored for 5 min. The cuvette was finally regenerated by 2 min washing with 0.01 M HCl. In these conditions, and after 5 min equilibration with PB, a new experiment can be done. When PDE inhibitors were used, a 10 min preincubation of PDE with drugs was done before toxin addition. Because the instrument has a dual optical path, with two cuvettes, all experiments were performed in one cuvette, and, simultaneously, the second cuvette was used as a control to normalized data by subtracting this control cuvette from the response of toxin cuvette. With the subtraction of the control channel (without YTX), the responses caused by nonspecific binding and bulk refractive index change could be corrected. Rate Analysis for the Binding of Toxin to Immobilized Ligand. One of the molecules is immobilized and the other is in

Pazos et al. free solution, so the assumption made is that the concentration of molecule in solution is in excess over the concentration molecule on the surface, and therefore considered constant. This is termed pseudo-first-order conditions. If A is the immobilized species (ligand) and B is the species in solution (ligate), the rate of complex [AB] formation is d[AB]/dt ) kass[A][B] - kdiss[AB]

The result of complex (ligand-ligate) formation is a change in response (Rt), which is directly proportional to the degree of complex formation. Also, [A] is Rmax - Rt. Therefore, the above equation can be written as dRt/dt ) kass(Rmax - Rt)(B) - kdissRt

By integration, the equation can be transformed into a form suitable for application of nonlinear regression: Rt ) Req(1 - e-(kass[B]+kdiss)t)

This is a single-exponential equation, where Rt is the instrument response time t, and Req is the equilibrium response. Using the above equation, binding data can be fitted at different concentrations to determine the pseudo-first-order rate constant, where kon ) (kass[B] + kdiss)

B can be [YTX], and the equation can written as kon ) (kass[YTX] + kdiss)

kon, with units of s-1, was obtained for each concentration of YTX. Using linear least-squares regression, the association rate constant, kass, with units of M-1 s-1, was derived from the gradient of the plot of kon against YTX concentration, and the dissociation rate constant, kdiss, with units of s-1, was derived from the intercept of the plot. At equilibrium, by definition, the rates of the forward and reverse reactions are equal, so: kass[A][B] ) kdiss[AB]

rearranging to kdiss/kass ) [AB]/[A][B] ) KD

KD (M) is the dissociation equilibrium constant. If KD is high, the affinity of the interaction is low. Statistical Analysis. All of the experiments were carried out at least three times. Data were normalized. Results were analyzed using the Student’s t-test for unpaired data and the ANOVA test. A probability level of 0.05 or smaller was used for statistical significance. Results were expressed as the mean ( SEM.

Results The interaction of YTX and different PDE families was checked. Initially, a mixture of cyclic nucleotide PDEs was attached to the cuvette. The PDE specificity was studied by using different PDEs inhibitors or PDE families and by measuring the signal change at equilibrium (Req) after YTX addition (25). First, 0.5 mg/mL cyclic nucleotide PDEs from bovine brain were added to an aminosilane cuvette. As Figure 1 shows, the biosensor detected a sustained response in 10 min. The surface was then washed, and no fall in the signal was observed, indicating that the PDEs were immobilized on the cuvette sensor surface. Following, the cuvette was washed and the remaining



Figure 3. Effect of cyclic nucleotide PDE 2 inhibition on cyclic nucleotide PDEs-YTX association. The plot shows the association curves after addition of 7 µM YTX to immobilized cyclic nucleotide PDEs. Cyclic nucleotide PDEs were preincubated in the presence of 50 or 10 µM EHNA for 10 min, and then YTX was added (first arrow). Representative experiment of N ) 3.

Figure 1. Cyclic nucleotide PDEs immobilization and YTX association. Addition of 0.5 mg/mL PDEs (first arrow) from bovine brain onto an activated aminosilane cuvette surface. After 30 min, the cuvette was washed and remaining activated sites were blocked with 2 mg/mL BSA. This solution was replaced by PB (third arrow), and finally 0.01 M HCl was added (fourth arrow).

Figure 4. Effect of cyclic nucleotide PDE 3 inhibition on cyclic nucleotide PDEs-YTX association. The plot shows the association curves after addition of 7 µM YTX to immobilized cyclic nucleotide PDEs. Cyclic nucleotide PDEs were preincubated in the presence of 50 µM milrinone for 10 min, and then YTX was added (first arrow). Representative experiment of N ) 3.

Figure 2. Cyclic nucleotide PDEs-YTX association. Association curve after addition of 7 µM YTX to immobilized cyclic nucleotide PDEs.

activated sites were blocked with 2 mg/mL BSA (nonspecific binding). Finally, this solution was replaced with PB. After PDEs immobilization, 7 µM YTX was added. As Figure 2 shows, the response after YTX addition increases and follows a typical association curve profile. Next, the enzyme specificity for YTX was checked using inhibitors of different cyclic nucleotide PDEs. As Figure 3 shows, when cyclic nucleotide cGMP-stimulated PDE (PDE 2) was inhibited in the presence of EHNA (a specific inhibitor of this PDE family, with an IC50 of 1 µM (15)), no significant modifications to the binding signal for 7 µM YTX were observed; the signal at equilibrium was 28 ( 2 arc sec in the presence of 7 µM YTX, and 28 ( 1 or 32 ( 6 arc sec in the presence of 50 or 10 µM EHNA plus 7 µM YTX, respectively. When cyclic nucleotide cGMP-inhibited PDE (PDE 3) was

inhibited in the presence of 10 µM milrinone, IC50 0.30 µM (15), no modifications to the signal were observed (data not shown). However, when this concentration was increased to 50 µM, Figure 4, the equilibrium signal after YTX addition significantly decreases from 25 ( 2 to 19 ( 1. When the PDE inhibited was cyclic nucleotide cAMP-specific PDE (PDE 4), Figure 5, in the presence of 10 and 50 µM rolipram (IC50 2 µM (15)), the amount of toxin bound to the PDEs was significantly lower. The signal decreased from 25 ( 2 to 20 ( 2, for 10 µM, and down to 15 ( 2 for 50 µM rolipram plus YTX. Finally, the unspecific inhibitor 3-isobutyl-1-methylxanthine (IBMX) (IC50 2-59 µM (28)) was used. As shown in Figure 6, the signal falls from 27 ( 1 arc sec in the presence of 7 µM YTX to 20 ( 1 with 10 µM IBMX plus YTX, and down to 10 ( 3 in the presence of 50 µM IBMX plus YTX. Next, we checked the cyclic nucleotide calomodulin-dependent PDE (PDE 1) specificity. This family is highly expressed in brain, and no direct inhibitors were reported (15). Calmodulin has been extensively described as a cyclic nucleotide PDE 1 activator, so competition between YTX and calmodulin was studied. As Figure 7 shows, addition of 18 µg/mL calmodulin to PDEs mixture induced an association curve with a equilibrium response of 6 ( 2 arc sec in the plateau. Addition of 7 µM


Pazos et al.

Figure 5. Effect of cyclic nucleotide PDE 4 inhibition on cyclic nucleotide PDEs-YTX association. The plot shows the association curves after addition of 7 µM YTX to immobilized cyclic nucleotide PDEs. Cyclic nucleotide PDEs were preincubated in the presence of 50 or 10 µM rolipram for 10 min, and then YTX was added (first arrow). Representative experiment of N ) 4.

Figure 6. Effect of cyclic nucleotide PDE unspecific inhibition on cyclic nucleotide PDEs-YTX association. The plot shows the association curves after addition of 7 µM YTX to immobilized cyclic nucleotide PDEs. Cyclic nucleotide PDEs were preincubated with 10 or 50 µM IBMX for 10 min, and then YTX was added (first arrow). Representative experiment of N ) 4.

Figure 7. Effect of cyclic nucleotide PDE 1 activation on cyclic nucleotide PDEs-YTX association. Association curves after addition of 18 µg/mL calmodulin or 7 µM YTX to immobilized cyclic nucleotide PDEs. The arrow indicates substance addition. Representative experiment of N ) 4.

YTX decreased the affinity binding curve from 28 ( 3 to 18 ( 3 arc sec, indicating a probable competition between YTX and calmodulin. In another batch of experiments, the purified venom exonuclease (PDE I) was directly attached to the cuvette and YTX

Figure 8. Exonuclease PDE I-YTX association. (A) Association curves after addition of different amounts of YTX to immobilized exonuclease PDE I. (B) Analysis of ligand binding. Kinetic plot of kon obtained from plot 8A versus YTX concentration. Representative experiment of N ) 3.

association was checked. Initially, 0.5 mg/mL PDE I from Crotalux atrox was added to the aminosilane cuvette. After PDE I immobilization, different concentrations of YTX were added. As Figure 8A shows, the response after YTX addition increases with the toxin dose and follows a typical association curve profile. In the presence of 1 µM YTX, the response is 4.31 arc sec, while in the presence of 15 µM YTX the response reaches 31.67 arc sec. The individual binding curves from Figure 8A were analyzed to determine the kinetic constants of YTX-PDE binding, that is, the apparent rate constant (kon), the association rate constant (kass), the dissociation rate constant (kdis), and the kinetic equilibrium dissociation constant (KD). The association curves fit a pseudo-first-order kinetic equation where kon can be calculated. The value of this constant increased with YTX concentration. Figure 8B shows a representation of each kon against the corresponding YTX concentration. This plot follows a linear regression with a correlation coefficient of r ) 0.98. From this representation, the kass (slope of the curve) and the kdiss (y-axis intersection) were obtained with values of 210 ( 20 M-1 s-1 and 6 × 10-4 ( 2 × 10-4 s-1 respectively. From these values, the kinetic equilibrium dissociation constant (KD) for the YTX-PDE I binding, KD ) kdiss/kass, was calculated. The value of this last constant was 3 ( 1 µM for YTX. Finally, the association between YTX and exonuclease from bovine spleen (PDE II) was studied. The purified enzyme was directly attached to the cuvette surface, and YTX was added. 0.5 mg/mL PDE II was immobilized on an aminosilane surface. Different amounts of YTX were then added, and, as Figure 9A shows, typical dose-dependent association curves were obtained.


Figure 9. Exonuclease PDE II-YTX association. (A) Association curves after addition of different amounts of YTX to immobilized exonuclease PDE II. (B) Analysis of ligand binding. Kinetic plot of kon obtained from plot (A) versus YTX concentration. Representative experiment of N ) 3.

These results show YTX affinity by PDE II. The plot of kon against the corresponding YTX concentration was linear (Figure 9B), and the kinetic constants were kass ) 280 ( 90 M-1 s-1, kdiss ) 2.4 × 10-3 ( 0.8 × 10-3 s-1, and KD ) 8 ( 3 µM.

Discussion The secondary messengers cAMP and cGMP, synthesized by the enzymes adenylate cyclase and guanylate cyclase, respectively, are degraded by PDE enzymes (29-31). In this sense, it has been reported that YTX decreases cAMP levels by activation of cellular PDEs, which points to the PDE system as one of the YTX intracellular targets (9). In addition, the interaction between YTX and cyclic nucleotide PDEs was confirmed by resonant biosensor studies where biomolecular interactions were studied in real time without labeling of interacting species. The YTX binding to cyclic nucleotide immobilized PDEs is significantly increased with YTX concentration, and the kinetic equilibrium dissociation constant of the PDEs-YTX association indicates a high affinity between these two components. The value of this constant, 3.74 µM, is in the range of KD values for binding between active biological species (10-11 and 10-4 M) (25, 27). The interaction YTXPDEs was also confirmed by fluorescence polarization studies (32). The specificity of different PDEs families for YTX was studied using a resonant mirror biosensor in this paper. The


use of optical biosensors to show molecular interactions is a recent but well-accepted technique (33, 34). There are 11 cyclic nucleotide PDEs families with different localizations and nucleotide specificity (35). Cyclic nucleotide PDE isoform families 1, 2, 3, 4, 7, 8, 10, and 11 hydrolyze cAMP and can be considered as camp-specific PDE, while PDEs 10 and 11 also hydrolyze cGMP (16). All of these families, in higher or lower expression, have been described in brain (15, 17, 18). Initially, the results with PDE 2 inhibitors show no interference with YTX signal. Two possibilities can explain these results; YTX does not bind with PDE 2, or the mixture of cyclic nucleotide PDEs used had no cyclic nucleotide PDE 2. The specificity of EHNA as a cyclic nucleotide PDE 2 inhibitor has been extensively reported (36). Therefore, cyclic nucleotide PDE 2 was probably not present in the mixture of cyclic nucleotide PDEs used, because the direct modulation with drugs of cyclic nucleotide PDE 1, PDE 3, and PDE 4 did modify YTX binding. In this sense, cyclic nucleotide PDE 3 and basically PDE 4 inhibitors strongly interfere with PDE-YTX association, indicating that both cyclic nucleotide PDEs families are implicated in the YTX effect. To our knowledge, there are no commercially available direct cyclic nucleotide PDE 1 inhibitors, so to confirm the involvement of this cyclic nucleotide PDE family we used a cyclic nucleotide PDE 1 activator. The results are rather useful to clarify the interaction of YTX with these enzymes. First, the enzyme activation in the presence of calmodulin significantly decreases response to YTX. This effect could indicate that both YTX and calmodulin induce the activation effect at the same site. This proposition is further supported by the fact that the effect of YTX on cAMP levels is calcium dependent (9), in a fashion similar to the calmodulin effect (37). However, the presence of calcium is not necessary to bind either cyclic PDE and calmodulin (37) or cyclic PDE and YTX (25, 27). To check the effect of this ion, we used buffers with and without calcium, and no differences in association curves were found with YTX and calmodulin (data not shown). Therefore, in both cases, calcium is necessary for the enzyme activation but not for binding. The similarity between calmodulin and YTX should be further studied to completely clarify YTX effect. Results obtained in the presence of IBMX again show a significant decrease in the signal. The small response observed in the presence of 50 µM could probably be blocked with a higher drug concentration. Unfortunately, when we tried 200 µM IBMX at 20 °C, we observed drug precipitation. In summary, these data show that YTX is associated least to cyclic nucleotide PDE 1, PDE 3, and PDE 4, which is in agreement with previous functional results that we have reported earlier on the basis of changes of cAMP levels in human lymphocytes (9). On the other hand, we checked the association of exonucleases PDE I and PDE II with YTX. When the exonuclease PDE I from Crotalux atrox was used, the KD value obtained (3 µM) was almost the same as for mammalian (3.7 µM) cyclic nucleotide PDEs (25), indicating that the affinity between exonuclease PDE I and YTX is similar to that of cyclic nucleotide PDEs. Next, pure exonuclease PDE II from bovine spleen was utilized as a ligand. In this case, the KD value was 8 µM, higher than that for cyclic nucleotide PDEs, indicating that the affinity between PDE II and YTX is lower than that for cyclic nucleotide PDEs. These results with nucleases of known specificity, such as venom exonuclease (PDE I) or spleen exonuclease (PDE II), can provide a very useful approach to understanding the correlation between structure and function


in nucleic acids in general, and transfer RNA in particular (20, 21). This could be useful in further studies of activation of these enzymes by YTX. In summary, all of these results show that YTX binds to cyclic nucleotide PDEs 1, 3, and 4; it has high affinity to exonuclease PDE I, while the affinity to exonuclease PDE II was lower. Acknowledgment. This work was funded by grants SAF200308765-C03-02, REN2001-2959-C04-03, REN2003-06598-C0201, AGL2004-08268-02-O2/ALI, INIA CAL01-068, Xunta PGIDT99INN26101 and PGIDIT03AL26101PR, FISS REMAG03-007, EU VIth Frame Program FOOD-CT-2004-06988 (BIOCOP), and FOOD-CT-2004-514055 (DETECTOX).

Pazos et al.

(17)

(18)

(19) (20) (21)

References (22) (1) Murata, M., Kumagai, M., Lee, J. S., and Yasumoto, T. (1987) Isolation and structure of Yessotoxin, a novel polyether compound implicated in diarrhetic shellfish poisoning. Tetrahedron Lett. 28, 5869-5872. (2) Satake, M., Terasawa, K., Kadowaki, Y., and Yasumoto, T. (1996) Relative configuration of yessotoxin and isolation of two new analogues from toxic scallops. Tetrahedron Lett. 37, 5955-5958. (3) Takahashi, H., Kusumi, T., Kan, Y., Satake, M., and Yasumoto, T. (1996) Determination of the absolute configuration of yessotoxin, a polyether compound implicated in diarrhetic shellfish poisoning, by NMR spectroscopic method using a chiral anisotropic reagent, methoxy-(2-naphthyl)acetic acid. Tetrahedron Lett. 37, 7087-7090. (4) Finch, S. C., Wilkins, A. L., Hawkes, A. D., Jensen, D. J., MacKenzie, A. L., Beuzenberg, V., Quilliam, M. A., Olseng, C. D., Samdal, I. A., Aasen, J., Selwood, A. I., Cooney, J. M., Sandvik, M., and Miles, C. O. (2005) Isolation and identification of (44-R,S)-44,55-dihydroxyyessotoxin from Protoceratium reticulatum, and its occurrence in extracts of shellfish from New Zealand, Norway and Canada. Toxicon 46, 160-170. (5) Ogino, H., Kumagai, M., and Yasumoto, T. (1997) Toxicologic evaluation of yessotoxin. Nat. Toxins 5, 225-229. (6) Daiguji, M., Satake, M., Ramstad, H., Aune, T., Naoki, H., and Takeshi, Y. (1998) Structure and fluorometric HPLC determination of 1-desulfoyessotoxin, a new yessotoxin analogue isolated from mussels from Norway. Nat. Toxins 6, 235-239. (7) Aune, T., Sorby, R., Yasumoto, T., Ramstad, H., and Landsverk, T. (2002) Comparison of oral and intraperitoneal toxicity of yessotoxin toward mice. Toxicon 40, 77-82. (8) EC. (2002) Commission decision of 15 March 2002 laying down detailed rules for the implementation of Council Directive 91/492/ EEC as regards the maximum levels and the methods of analysis of certain marine biotoxins in bivalve molluscs, echinoderms, tunicates and marine gastropods (2002/225/EC). Off. J. Eur. Commun. L75, 6264. (9) Alfonso, A., de la Rosa, L. A., Vieytes, M. R., Yasumoto, T., and Botana, L. M. (2003) Yessotoxin a novel phycotoxin, activates phosphodiesterase activity. Effect of yessotoxin on cAMP levels in human lymphocytes. Biochem. Pharmacol. 65, 193-208. (10) de la Rosa, L. A., Alfonso, A., Vilarinõ, N., Vieytes, M. R., Yasumoto, T., and Botana, L. M. (2001) Modulation of cytosolic calcium levels of human lymphocytes by yessotoxin, a novel marine phycotoxin. Biochem. Pharmacol. 61, 827-833. (11) de la Rosa, L. A., Alfonso, A., Vilarinõ, N., M. R., V., Yasumoto, T., and Botana, L. M. (2001) Maitotoxin-induced calcium entry in human lymphocytes-modulation by yessotoxin, Ca2+ channel blockers and kinases. Cell. Signalling 13, 711-716. (12) Fisher, D. A., Smith, J. F., Pillar, J. S., St. Denis, S. H., and Cheng, J. B. (1998) Isolation and characterization of PDE9A, a novel human cGMP-specific phosphodiesterase. J. Biol. Chem. 273, 15559-15564. (13) Hetman, J. M., Soderling, S. H., Glavas, N. A., and Beavo, J. A. (2000) Cloning and characterization of PDE7B, a cAMP-specific phosphodiesterase. Proc. Natl. Acad. Sci. U.S.A. 97, 472-476. (14) Mukai, J., Asai, T., Naka, M., and Tanaka, T. (1994) Separation and characterization of a novel isoenzyme of cyclic nucleotide phosphodiesterase from rat cerebrum. Br. J. Pharmacol. 111, 389-390. (15) Beavo, J. A. (1995) Cyclic nucleotide phosphodiesterases: functional implications of multiple isoforms. Physiol. ReV. 75, 725-748. (16) Fujishige, K., Kotera, J., Yuasa, K., and Omori, K. (2000) The human phosphodiesterase PDE10A gene genomic organization and evolution-

(23) (24) (25) (26)

(27)

(28)

(29) (30)

(31) (32)

(33) (34) (35) (36)

(37)

ary relatedness with other PDEs containing GAF domains. Eur. J. Biochem. 267, 5943-5951. Cho, C. H., Cho, D. H., Seo, M. R., and Juhnn, Y. S. (2000) Differential changes in the expression of cyclic nucleotide phosphodiesterase isoforms in rat brains by chronic treatment with electroconvulsive shock. Exp. Mol. Med. 32, 110-114. Zhao, A. Z., Huan, J. N., Gupta, S., Pal, R., and Sahu, A. (2002) A phosphatidylinositol 3-kinase phosphodiesterase 3B-cyclic AMP pathway in hypothalamic action of leptin on feeding. Nat. Neurosci. 5, 727-728. Laskowski, M. (1971) in Venom Exonuclease, The Enzymes (Boyer, P., Ed.) 3rd ed., Vol. 4, p 313, Academic Press, New York. Ho, N., and Gilham, P. (1973) The analysis of polydeoxyribonucleotides by digestion with phosphatase and phosphodiesterases. Biochem. Biophys. Acta 308, 53. Bernardi, A., and Bernardi, G. (1971) in Spleen Acid Exonuclease, The Enzymes (Boyer, P., Ed.) 3rd ed., Vol. 4, p 329, Academic Press, New York. Philippsen, P., and Zachau, H. (1971) Fragments of yeast tRNA Phe and tRNA Ser prepared by partial digestion with spleen phosphodiesterase. FEBS Lett. 15, 69. Roychoudhury, R., Fischer, D., and Ko¨ssel, H. (1971) A new method for the sequence analysis of oligodeoxynucelotides. Biochem. Biophys. Res. Commun. 45, 430. Sneider, T. (1971) Inactivation of E. coli alkaline phosphatase prior to phosphodiesterase digestion of oligodeoxyribonucleotides. Anal. Biochem. 44, 658. Pazos, M., Alfonso, A., Vieytes, M., Yasumoto, T., and Botana, L. (2004) Resonant mirror biosensor detection method based on yessotoxin-phosphodiesterase interactions. Anal. Biochem. 335, 112-118. Ferrari, S., Ciminiello, P., Dell’Aversano, C., Forino, M., Malaguti, C., Tubaro, A., Poletti, R., Yasumoto, T., Fattorusso, E., and Rossini, G. P. (2004) Structure-activity relationships of yessotoxins in cultured cells. Chem. Res. Toxicol. 17, 1251-1257. Pazos, M., Alfonso, A., Vieytes, M., Yasumoto, T., and Botana, L. (2005) Kinetic analysis of the interaction between yessotoxin and analogues and immobilized phosphodiesterases using a resonant mirror optical biosensor. Chem. Res. Toxicol., in press. Fawcett, L., Baxendale, R., Stacey, P., McGrouther, C., Harrow, I., Soderling, S., Hetman, J., Beavo, J. A., and Phillips, S. C. (2000) Molecular cloning and characterization of a distinct human phosphodiesterase gene family: PDE11A. Proc. Natl. Acad. Sci. U.S.A. 97, 3702-3707. Cheung, W. (1971) Cylic 3′,5′-nucleotide phosphodiesterase. Evidence for and properties of a protein activator. J. Biol. Chem. 246, 28592869. Kakiuchi, S., Yamazaki, R., Teshima, Y., and Uenishi, K. (1973) Regulation of nucleotide cyclic 3′,5′-monophosphate phosphodiesterase activity from rat brain by a modular and Ca2+. Proc. Natl. Acad. Sci. U.S.A. 70, 3526-3530. Sutherland, E., and Roll, T. (1958) Fractionation and characterization of a cyclic adenosine ribonucleotide formed by tissue particles. J. Biol. Chem. 243, 1077-1091. Alfonso, C., Alfonso, A., Vieytes, M. R., Yasumoto, T., and Botana, L. M. (2005) Quantification of yessotoxin using the fluorescence polarization technique, and study of the adequate extraction procedure. Anal. Biochem., in press. Hirmo, S., Artursson, E., Puu, G., Wadstrom, T., and Nilsson, B. (1999) Helicobacter pylori interactions with human gastric mucin studied with a resonant mirror biosensor. J. Microbiol. Methods 37, 177-182. Catimel, B., Weinstock, J., Nerrie, M., Domagala, T., and Nice, E. C. (2000) Micropreparative ligand fishing with a cuvette-based optical mirror resonance biosensor. J. Chromatogr., A 869, 261-273. Houslay, M. D., and Adams, D. R. (2003) PDE4 cAMP phosphodiesterases: modular enzymes that orchestrate signaling cross-talk, desentization and compartimentalization. Biochem. J. 370, 1-18. van Staveren, W. C., Markerink-van Ittersum, M., Steinbusch, H. W., and de Vente, J. (2001) The effects of phosphodiesterase inhibition on cyclic GMP and cyclic AMP accumulation in the hippocampus of the rat. Brain Res. 888, 275-286. Ohki, S., Ikura, M., and Zhang, M. (1997) Identification of Mg2+binding sites and the role of Mg2+ on target recognition by calmodulin. Biochemistry 36, 4309-4316.

TX0503303

Study of the Interaction between Different Phosphodiesterases and

Recommend Documents