Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

FK409, the chemical structure of which is shown in figure 1, is a compound isolated from the acid-treated fermentation broth of Streptomyces griseosporeus (Hino et al., 1989a). We demonstrated that FK409 spontaneously released NO in PB solution (pH 7.4) by chemiluminescence analysis (Kita et al., 1994). In addition, FK409 inhibited ADP-induced aggregation of human platelet and norepinephrine-induced contraction of isolated rat aorta. It was also shown by Isono et al. (1993) that the remarkable vasorelaxant effect of FK409, in particular, was due to the activation of soluble guanylyl cyclase followed by the increase in cGMP levels. From these data, it is suggested that the biological activities of FK409 are associated with spontaneous NO release from the molecule.

View larger version (K): [in this window] [in a new window]   Fig. 1.   Chemical structures of FK409 and its derivatives.

Recently, we investigated the spontaneous NO-releasing pathway of FK409 in PB solution by NMR analysis, and proposed that the essential process for decomposition of FK409 with release of NO was the deprotonation of -hydrogen atom of the nitro moiety by bases such as hydroxyl ion (Fukuyama et al., 1995a). Furthermore, Kato et al. (1996) synthesized a series of FK409 derivatives, in which chain length or substituent of FK409 at positions C-2 or C-5 were modified systematically, and clearly showed that steric and electronic factors of substituents in their structures were important for controlling the rates of decomposition. Because it was also suggested that the structural modification led to a change in the acidity of -hydrogen atom of the nitro moiety, FK409 derivatives were expected to have various NO-releasing rates.

We questioned whether NO-releasing rates of FK409 derivatives could account for their biological activities or not. For example, NO-releasing rate of FR144420 (e.g., compound 4 in fig. 1), in which the amide group of FK409 at the C2-position was substituted for methylene amide moiety, was lower than that of FK409 (Kita et al., 1995a). Thus, it is proposed that the stabilization of the hydrogen atom at the C-5 position due to the substitution at the C2-position of FK409 lead to slower NO release and weaker biological activity than FK409. However, it has been unclear whether other structural modification of FK409 at positions C-2, C-4 and C-5 affects the NO-releasing rates and reflects on the biological activities or not.

Accordingly, the spontaneous NO-releasing rate of a series of FK409 derivatives (see fig. 1) in PB solution was determined along with their in vitro vasorelaxant and antiplatelet activities. In this way, the correlations between NO-releasing rates of FK409 derivatives and their in vitro biological activities were evaluated.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Materials. FK409 and its derivatives (see fig. 1) were synthesized at Fujisawa Pharmaceutical Co. (Osaka, Japan). 2-Hydroxyimino-5-nitro-3-alkenecarboxamides (FK409, 1 and 2) and 1-acylamino-2-hydroxyimino-5-nitro-3-alkenes (3-8, 11 and 12) were prepared according to the procedures in the previous literature (Kato et al., 1996). Thioether derivatives 9 and 10 were also prepared in an analogous fashion as shown in figure 2. Stereochemistry of the hydroxyimino group was determined on the basis of NOE in the NOESY spectrum. All compounds had ¹H NMR spectra in accord with the assigned structures, and physical data for compounds prepared are given below.

View larger version (K): [in this window] [in a new window]   Fig. 2.   Synthetic scheme of FK409 derivatives 9 and 10.

Animals. All male Sprague-Dawley rats were purchased from Nihon SLC (Shizuoka, Japan).

Determination of cumulative amount of NO released. The cumulative amount of NO generated from FK409 derivatives was determined according to the method reported previously (Fukuyama et al., 1995a). The cumulative amount of NO released was measured by an X-band ESR spectrometer (JES-RE3X, JEOL, Tokyo, Japan) using carboxy-PTIO. The conditions for ESR measurement were as follows: modulation frequency, 100 KHz; modulation amplitude, 0.05 mT; scanning field, 337.3 ± 5.0 mT; response time, 0.03 sec; sweep time, 1 min; microwave power, 4 mW.

In vitro antiplatelet study. Blood from male human volunteers was collected into plastic vessels containing 3.8% sodium citrate (1/10 volume) (Nishizawa et al., 1983). Platelet-rich plasma was obtained from the supernatant fraction after centrifugation of blood at 150 × g for 10 min. The effect of each compound on platelet aggregation was determined by Born's turbidimetric method (Born and Cross, 1968) using an aggregometer (Hema Tracer 801, Niko Bioscience Co., Tokyo, Japan). FK409 or its derivative was dissolved in DMSO at several concentrations. The compounds were stable in DMSO. To 242.5 µl of platelet-rich plasma in a cuvette 2.5 µl of the drug solution or vehicle were added, and the mixture was incubated at 37°C for 2 min. After the incubation, platelet aggregation was induced by the addition of 5 µl of 125 µM ADP (final concentration: 2.5 µM). To quantify the inhibition of platelet aggregation of each compound, the maximum increase in light transmission was determined from the aggregation curve for 7 min after the addition of ADP. The effects of each compound were expressed as % inhibition of ADP-induced platelet aggregation compared with vehicle treatment.

In vitro vasorelaxant study. Male SD rats weighing 225 to 375 g were sacrificed by stunning and exsanguination. The thoracic aorta from the rat was removed and cut into helical strips after removal of excess fat and connective tissues. The strip was mounted vertically in an organ bath containing 25 ml of Tyrode solution of the following composition: 136.9 mM NaCl, 2.7 mM KCl, 1.8 mM CaCl₂, 1.0 mM MgCl₂, 11.9 mM NaHCO₃, 0.4 mM NaH₂PO₄ and 5.6 mM dextrose. Isometric tension was measured with a force-displacement transducer (UL-10GR, Shinkoh, Tokyo, Japan) connected to an amplifier (AP-630G, Nihon Koden, Tokyo, Japan) and was recorded with a polygraph (Recti-Horiz-8K, Sanei, Tokyo, Japan). The tissue bath solution was maintained at 37°C and bubbled with a 95% O₂ and 5% CO₂ gas mixture. After the resting tension was adjusted to 0.5 g, the strip was contracted with 0.25 ml of 3.2 µM ()-norepinephrine (final concentration: 32 nM) 10 min after the addition of 0.25 ml of 5.7 mM L-ascorbic acid (final concentration: 57 µM). FK409 or its derivative was dissolved in a mixture of polyethylene glycol 200 and ethanol (1:1, v/v) at several concentrations, and 0.025-ml aliquot of the solution was added to the organ bath cumulatively. The compounds were stable in the mixture. Cumulative addition of only vehicle did not cause any relaxation of the isolated rat aorta. Finally, 0.25 ml of 10 mM papaverine (final concentration: 100 µM) was added to the organ bath to obtain the maximum relaxation.

Data analysis. The data represent the means ± S.E. for the number of experiments indicated. The IC₅₀ value is expressed as the concentration of each compound required to produce 50% inhibition of 2.5 µM ADP-induced human platelet aggregation. The IC₅₀ value was computed by regression analysis logistically for the platelet-rich plasma from each human volunteer. The EC₅₀ value is expressed as the concentration of each compound required to produce 50% of 100 µM papaverine-induced relaxation. The EC₅₀ value was computed by regression analysis logistically for the isolated aorta from each rat.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

NO release. The calculated NO-releasing rates of FK409 and its derivatives are summarized in table 1. Larger substituents than methyl group (e.g., ethyl analogs 1, 5, 12 and isopropyl analogs 6, 7) at the C5-position led to delayed NO release. The compound with the slowest NO generation was 7. The NO-releasing rate of 7 was some 80-fold less than that of FK409. The replacement of amide moiety for methylene moieties at the C2-position slowed down NO generation (e.g., 3, 4, 9, 10 and 11). The cyclohexane derivative (e.g., 2) showed a similar rate of NO release as FK409. SNP and ISDN did not release NO spontaneously in PB solution as reported previously (Kita et al., 1994a, 1995b).

Antiplatelet activity. FK409 and its derivatives dose-dependently inhibited ADP-induced platelet aggregation in human platelet-rich plasma, and the maximum inhibition was almost 100%. The antiplatelet effects of FK409 and several derivatives (2, 4 and 12) on ADP-induced (2.5 µM) human platelet aggregation are shown in figure 3, and the IC₅₀ values of FK409 and its derivatives are indicated in table 1. Among the derivatives examined, the compound 2 was the most potent inhibitor of platelet aggregation. On the other hand, SNP and ISDN showed only 42 ± 6 and 34 ± 3% inhibition of platelet aggregation even at 100 µM, respectively.

View larger version (K): [in this window] [in a new window]   Fig. 3.   Antiplatelet effects of FK409 () and several FK409 derivatives [2 (), 4 () and 12 ()] on ADP-induced human platelet aggregation. The effects of each compound were expressed as % inhibition of 2.5 µM ADP-induced platelet aggregation compared with vehicle treatment. Each value represents the mean ± S.E. for three to five experiments.

View larger version (K): [in this window] [in a new window]   Fig. 5.   Correlations between (A) antiplatelet activities [IC50 values (µM)], (B) vasorelaxant activities [EC50 values (nM)], of FK409 and its derivatives, and the reciprocal of rate constants of NO release (liter/min) from the compounds. Each point represents the mean value of three to five experiments.

Vasorelaxant activity. In isolated rat aorta contracted with norepinephrine, cumulative addition of FK409 and its derivatives caused concentration-dependent relaxation, and the maximum relaxation was almost 100%. The vasorelaxant effects of FK409 and several derivatives (2, 4 and 11) on norepinephrine-induced contraction in isolated rat aorta are shown in figure 4, and the EC₅₀ values of FK409 and its derivatives are indicated in table 1. The compound 1, which exhibited slower NO release than FK409, showed the most potent vasorelaxant effect. The analogs with an amide moiety at the C-2 position (e.g., FK409, 1 and 2) were potent relative to those with methylene amide moiety at the same position (e.g., 3, 4, 6, 7, 8 and 11) even if they showed similar rates of NO liberation (e.g., 1 vs. 4, 8). SNP and ISDN also caused concentration-dependent relaxation. SNP and ISDN showed 90-fold more potent and 80-fold weaker vasorelaxant effects than FK409, respectively.

View larger version (K): [in this window] [in a new window]   Fig. 4.   Vasorelaxant effects of FK409 () and several FK409 derivatives [2 (), 4 () and 11 ()] on 32 nM norepinephrine-induced contraction in isolated rat aorta. Relaxation with 100 µM papaverine was expressed as 100%. Each value represents the mean ± S.E. for three to four experiments.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

We recently suggested that the deprotonation of the hydrogen atom at -position of the nitro group, at the C5-position, by bases such as hydroxyl ion was an essential step for both FK409 degradation and its NO release (Fukuyama et al., 1995a). Therefore, the electronic and steric factors of the substituents being located in the vicinity of the nitro group were expected to be responsible for the rate of the deprotonation of the -hydrogen atom of the nitro moiety according to the report of Kato et al. (1996). That is, the rate for NO release can be controlled by the acidity of the -hydrogen atom. As a result, the structural modification of FK409 led to an alteration of the NO-releasing rates. The substitution of functional groups at the C5- and C2-positions showed a substantial increase or decrease in NO-releasing activities. The possible scheme for NO release from FK409 derivatives are shown in figure 6. Larger substituents at the C5-position such as ethyl and isopropyl group cf. methyl group led to a delay release of NO. This is probably due to both the increase in steric effect and the decrease in the acidity at -position of the nitro moiety. Moreover, the substitution of amide groups for methylene moieties at the C2-position slowed down NO release. This could be produced by the decrease in the acidity at -position of the nitro group. From these results, it is clear that FK409 derivatives possess different spontaneous NO-releasing rates according to their individual structures.

View larger version (K): [in this window] [in a new window]   Fig. 6.   Possible scheme for NO release from FK409 derivative.

Each FK409 derivative showed different potency in in vitro biological experiments as well. A close correlation between the IC₅₀ values for inhibition of ADP-induced human platelet aggregation and NO-releasing rates of FK409 derivatives was observed. These results suggest that the antiplatelet effects of FK409 analogs are dependent on only their NO-releasing activities in PB solution. A similar relationship has been observed between EC₅₀ values for activation of isolated soluble guanylyl cyclase and the rate of NO generation from organic nitrates such as glyceryl trinitrate, or SIN-1 analogs (Feelisch and Noack, 1987; Noack and Feelisch, 1989). Additionally, it was reported that the vasorelaxant activities of amine/NO complex ions in isolated rabbit thoracic aorta also related to the velocity of NO liberation (Maragos et al., 1991). However, this report is the first to show a correlation between NO-releasing rates and the antiplatelet activities of NO donors.

However, the NO-releasing rates of FK409 analogs in PB solution did not correlate with their in vitro vasorelaxant effects in isolated rat aorta in contrast to the report of Maragos et al. (1991). The vasorelaxant effects appear to be reflected on chemical structures of FK409 analogs as well as their velocities of NO liberation. The substituent at the C2-position gave a great influence on the vasorelaxant effect of FK409 derivatives than that at the C5-position. As shown in table 1, the analogs with amide moiety at the C2-position (e.g., FK409, 1, 2) have a more potent effect rather than those with the methylene amide moiety at the C-2 position (e.g., 3, 4, 6, 7, 8, 11). For example, although compound 1, 4 and 8 showed similar NO-releasing rates, the vasorelaxant activity of compound 1 was about 35-fold potent than that of compounds 4 and 8.

Isono et al. (1993) investigated the vasorelaxant mechanism of FK409. They found that the vasorelaxant effect of FK409 was independent of the integrity of the endothelium, and was unaffected by L-N^G-monomethylarginine, an inhibitor of NO synthase, and oxyhemoglobin, a extracellular NO-quenching protein. However, the vasorelaxant effect of FK409 was accompanied by the increase in intracellular cGMP levels. Consequently, FK409 does not cause endogenous NO release, and NO released from FK409 itself in the vascular smooth muscle cells appear to be attributable to its vasorelaxant action.

From the facts described above, one possibility causing the difference in the correlation between NO-releasing rates of FK409 analogs and their biological activities is that in deliveries of the analogs into vascular smooth muscle cells and platelets. Thus, the difference between vascular smooth muscle cells in isolated aorta and platelets that are suspended cells lies in whether or not the extracellular matrix exists. Probably the vasorelaxant effects of FK409 derivatives may depends on both their NO-releasing activities and the interaction with the extracellular matrix. The structural properties of FK409 analogs determine both their NO-releasing activities and the extent of the interaction between the analogs and the extracellular matrix, giving rise to differences in the analogs' ability to pass into the vascular smooth muscle cells. Therefore, under these circumstances the vasorelaxant effects of FK409 analogs would not correlate with their NO-releasing rates. As described above, the substitution of the amide group for methylene amide moiety at the C-2 position resulted in a lowering of the vasorelaxant activities. FK409 analogs with methylene amide moiety at the C-2 position are more hydrophobic than those with amide group. Hence, the interaction between FK409 derivatives and extracellular matrix, i.e., the ability to trap the compounds on the extracellular matrix, may be influenced by hydrophobicity of the compounds. Thus, their delivery rates into vascular smooth muscle cells may differ from one another by the substituents at the C-2 or C-5 position in their structures. However, the antiplatelet potencies are unaffected by extracellular matrix, they closely correlate with the spontaneous NO-releasing activities, suggesting that delivery rates of all FK409 analogs into platelets are almost the same.

Another possibility causing the difference in the correlation is the difference between metabolic rates of FK409 derivatives in vascular smooth muscle cells and those in platelets. This is probably derived from the difference in the interaction between the compounds and sulfhydryl groups existing in the biological components. We compared the NO-releasing property and in vitro biological activities, i.e., antiplatelet and vasorelaxant activities, of FK409 with those of other NO donors, SNP and ISDN in this study and previously (Kita et al., 1994a; 1995b). Although FK409 released NO spontaneously, the NO release was accelerated in the presence of thiolate anion derived from sulfhydryl-bearing compounds such as cysteine and glutathione (Fukuyama et al., 1995b). However, SNP and ISDN released NO only in the presence of sulfhydryl-bearing compounds. As shown in table 1, SNP and ISDN had much weaker antiplatelet potencies compared with their vasorelaxant potencies. The difference between antiplatelet potencies and vasorelaxant potencies of the compounds is probably due to that between NO-releasing rates in human platelets and NO-releasing rates in isolated rat aorta. The difference in NO-releasing rates of the compounds in the biological components would be caused by that in the concentration of thiol compounds including free cysteine and glutathione or cysteine residues in many polypeptides. Therefore, the amount of sulfhydryl groups existing in isolated rat aorta seems to be greater than those in human platelets. FK409 releases NO after the deprotonation at the -position of the nitro moiety by bases such as hydroxyl ion and thiolate anion (Fukuyama et al., 1995b). As shown in figure 6, other FK409 derivatives also have the -hydrogen atom of the nitro moiety and they are considered to release NO after the deprotonation at the -position by bases such as hydroxyl ion and thiolate anion. Therefore, NO release from FK409 derivatives could be also intensively influenced by thiol compounds in isolated rat aorta in particular. The difference in the interaction of FK409 derivatives with thiol compounds probably expands the difference of vasorelaxant activities, and produces the vast difference between antiplatelet potencies (IC₅₀) and vasorelaxant potencies (EC₅₀), leading to the lack of correlation between NO-releasing rates in PB solution and vasorelaxant potencies. Thus, both cell permeabilities and the metabolic rates of FK409 derivatives differ from one another according to their individual chemical structures or biological components used, i.e., isolated rat aorta and human platelets. Therefore, these factors probably form the difference in the biological potencies.

In conclusion, the structural modification of FK409 led to production of FK409 derivatives with different NO-releasing rates. The antiplatelet potencies produced by FK409 derivatives closely correlated with their NO-releasing rates in PB solution. The vasorelaxant potencies of FK409 derivatives were dependent on not only the NO-releasing rates but also their delivery rates into vascular smooth muscle cells and metabolic rates in the cells.