**5. Pharmacological Activities Associated to the Multifunctional Fish Cytolysins** *5.1. Cardiovascular Activity*

The most potent and best-studied pharmacological activity associated with the venoms of Scorpaeniformes is that on the cardiovascular system [9,23,28]. The profound cardiovascular alterations induced by these venoms are attributed to the action of cytolysins, being closely related to the lethality of these toxins [15,19,22].

The molecular mechanisms underlying the cardiovascular effects induced in vivo and in vitro by cytolysins from the venoms of stonefish and scorpionfish have been fairly well documented. Despite some variability, which could simply reflect the fact that different studies employed different doses and experimental models, it is a consensus that these toxins cause drastic changes in blood pressure, which are the result of their effects on the blood vessels and heart.

When the purification of VTX from the venom of *S. verrucosa* was first described, it was observed that the injection of this cytolysin (∼16 μg/kg, i.v.) into anesthetized rats induced an immediate hypotensive effect that evolved to cardiac arrest [19]. A drastic, irrecoverable fall in the blood pressure of anesthetized rats has also been described for SNTX (20 μg/kg, i.v.) and Sp-CTx (70 μg/kg, i.v.), although a biphasic response characterized by an increase followed by a decrease on arterial pressure was observed in these instances [15,22].

Consistent with the hypotension observed in vivo, a significant vasorelaxation has been associated with the action of fish cytolysins on vascular preparations [22,29,30,41]. In addition, a biphasic response on rat aortic rings pre-contracted with phenylephrine, characterized by an endothelium- and dose-dependent relaxation followed by a contractile phase, was associated with both SNTX and Sp-CTx, albeit with different intensities. Regardless, the fact that both toxins induced relaxation of isolated aortic rings in the picomolar range (68 pM and 1 pM for SNTX and Sp-CTx, respectively), attests to the high hypotensive potency of these molecules [22,29,41].

The vasorelaxant phase of the effect induced by SNTX and Sp-CTx was irreversibly abolished by 25 and 10 μM of the nitric oxide (NO) synthesis inhibitor NG-nitro-L-arginine methyl ester (L-NAME), respectively [22,41]. In addition, the relaxant response associated with SNTX was reversibly inhibited by the guanylate cyclase inhibitors methylene blue (10 μM) and hemoglobin (5 mM) [22]. It was also reduced by the H2S inhibitors DLpropargylglycine (PAG; 1 mM) and β-cyano-L-alanine (BCA; 1 mM), being potentiated when these inhibitors were individually combined with L-NAME (1 μM) [30]. The K<sup>+</sup> channel blocker tetraethylammonium (TEA; 3 mM) and the substance P (SP) receptor antagonist N-acetyl-L-tryptophan-3,5-bis(trifluoromethyl)-benzyl ester (NATB; 0.5 mM) also reduced the SNTX-induced vasorelaxation, which was completely blocked by the iNOS inhibitor AMT-HCl (0.5 mM) [29].

Based on these results, it has been proposed that SNTX acts by binding directly to the SP receptor in the endothelial cell or by promoting SP release. The consequent increase in intracellular [Ca2+] stimulates iNOS and increases the production of NO, which will be released by the endothelial cell and absorbed by the smooth muscle cell. There, it increases the levels of cGMP, which will then positively modulate K+ channels, hyperpolarizing the membrane and causing relaxation [29]. The SNTX-induced relaxation also involves the synergistic action of H2S and NO [30]. The high sequence identity and structural similarities between fish cytolysins, as well as the antigenic cross-reactivity displayed by these toxins [14,40,41], suggest that the aforementioned mechanism of action can be extended to cytolysins other than SNTX.

The effect of fish cytolysins on the heart itself has also been explored to a certain extent. In 1997, the cardiotoxicity of the freshly purified but unstable VTX (1 μg/mL) and an impure but more stable fraction named p-VTX (0.001–1 μg/mL) was assessed [36]. Both exerted negative inotropic and chronotropic effects on the peak contraction and accelerated the relaxation phase in frog atrial fibers [36]. The negative inotropic effect induced by p-VTX was reduced in higher external [Ca2+], while the chronotropic effect remained unaffected in these conditions. Moreover, p-VTX hyperpolarized the fibers and decreased the duration of both the plateau and the hyperpolarizing phase of action potentials recorded in these fibers. Both muscular and electrical activities of the atrial fibers was counteracted by glibenclamide, an ATP-sensitive K<sup>+</sup> channel (KATP) blocker. The nature of these responses suggests p-VTX acts by inhibiting Ca2+ influx into the fibers, possibly by negatively modulating Ca2+ channels, and by positively modulating KATP channels.

On the other hand, studies conducted in guinea pig ventricular myocytes, employing electrophysiology and various pharmacological strategies, showed that, unlike what had been observed in atrial fibers, VTX inhibits KATP currents via the activation of the muscarinic M3 receptor-PKC pathway and increases L-type Ca2+ currents by activating the β-adrenergic-cAMP-PKA pathway [37,38]. As pointed out by [37], this discrepancy regarding the effects of VTX on frog atrial fibers and guinea pig ventricular cells could be simply the consequence of different species having different cardiac regulatory mechanisms. Accordingly, TLY (1 μg/mL) acted much like VTX in frog atrial fibers, hyperpolarizing the membrane, shortening the action potentials, and inducing a negative inotropic effect, in addition to a contracture due to acetylcholine release. However, this negative inotropic effect associated with TLY, although Ca2+-dependent, was not blocked by Cd2+, indicating it does not involve voltage-dependent Ca2+ channels [33].

In addition to its effects in vivo and on isolated vessels, Sp-CTx had its cardiotoxic activity assessed in rat isolated hearts and papillary muscles [15]. The toxin (10−<sup>9</sup> to <sup>5</sup> × <sup>10</sup>−<sup>6</sup> M) induced a transient, concentration-dependent positive inotropic effect, characterized by a drastic increase in left ventricular pressure. It also induced vasoconstriction on the coronary bed by increasing vascular perfusion pressure. The biphasic pattern described in anesthetized rats and rat aortic rings [15,41] was not observed in isolated hearts.

In isolated papillary muscle, Sp-CTx (0.1 μM) increased myocardial contractility through a pathway involving the activation of β-adrenoreceptors, as the pre-treatment with the β-blocker propranolol and the catecholamine releasing agent tyramine inhibited this response [15]. Post-rest contraction experiments suggested that the cardiac response induced by Sp-CTx involves an increase in sarcolemmal Ca2+ influx, which was corroborated by the reversible 18% increase in L-type Ca2+ currents induced by 1 nM of this toxin in isolated ventricular myocytes [15]. Taken together, these results point to Sp-CTx acting by increasing Ca2+ influx into cardiac cells via modulation of L-type Ca2+ channels through the activation of the β-adrenergic signaling pathway, much like what had been described for VTX in guinea pig ventricular myocytes [38].

It has been suggested that the pore-forming mechanism underlying the hemolytic activity of fish cytolysins might also account for some of the other pharmacological activities associated with these toxins [22,55,57,66]. However, the aforementioned studies showing that SNTX, TLY, VTX, and Sp-CTx mediate their cardiovascular effects through complex pathways involving the release of vasorelaxant mediators and the modulation of membrane receptors and/or ion channels [15,33,36–38], point to pore-formation not being the underlying cause of these responses. It is also important to stress that, although seemingly conflicting depending on the experimental conditions, these pathways could actually contribute in a synergistic way to potentiate the cardiovascular toxicity attributed to these cytolysins.

#### *5.2. Neuromuscular Activity*

In addition to the cardiovascular effects induced by fish venoms, symptoms observed after envenomation, such as weakness, muscle spasms, and even paralysis, are an evident sign of neurotoxicity. Indeed, under experimental conditions, the crude venoms of *S. verrucosa, S. horrida (trachynis), P. volitans*, and *T. draco* have had their neurotoxic effects demonstrated [13,70–72].

At least for the estuarine stonefish venom, the neuromuscular activity could be associated directly to its cytolysins—SNTX and TLY. In 1994, the effects of SNTX (8 to 50 μg/mL) on the neuromuscular function were investigated in mouse and chick skeletal muscles in vitro and rat skeletal muscle in vivo [31]. SNTX produced an irreversible, concentrationand time-dependent block of nerve- and muscle-evoked twitches of the mouse hemidiaphragm. Similar effects were observed in the chick biventer cervicis muscle, in which nerve-evoked twitches and the contractures induced by acetylcholine (200 μM), carbachol (8 μM), and KCl (40 mM) were blocked by SNTX (22 and 44 μg/mL). The ability of SNTX to block muscle-evoked twitches, as well as the marked blockade of this effect by dantrolene sodium (6 μM)—a muscle relaxant that inhibits Ca2+ release from intracellular stores—but not by the nicotinic receptor antagonist tubocurarine (15 μM), suggested that SNTX acts directly on the muscle to inhibit neuromuscular function, inducing contractures via a Ca2+-dependent rather than an acetylcholine-dependent mechanism. In addition, SNTX did not block conduction in the frog sciatic nerve, which also points to the conclusion that the toxin affects neuromuscular function via myotoxicity [31].

On the other hand, TLY—which had its neuromuscular activities rather well explored was reported to act presynaptically by causing the release and depletion of neurotransmitters from nerve terminals [32], as had been previously shown for the crude venom itself [71]. This cytolysin (2 to 20 μg/mL) irreversibly increased the frequency of miniature end plate potentials in frog cutaneous pectoris preparations, indicating it stimulated the quantal release of acetylcholine [32]. A similar outcome was found in *Torpedo marmorata* neuromuscular junction, in which TLY (30 to 60 nM) increased the spontaneous quantal neurotransmitter release in an irreversible way [34].

This neurosecretory role attributed to TLY was supported by the observation that botulinum toxin—a neurotoxin that disturbs exocytosis—inhibits the release of catecholamines induced by this cytolysin (60 nM) in bovine adrenal chromaffin cells [35]. However, unlike in motor nerve endings, where TLY depleted the number of small vesicles almost completely (86%) without affecting that of large-dense core vesicles [32], in chromaffin cells, this cytolysin mediated a soluble N-ethylmaleimide-sensitive fusion protein attachment protein receptor (SNARE)-dependent release of catecholamines from large core vesicles [35].

As expected, this neurosecretory effect was Ca2+-dependent, although voltage-gated Ca2+ channels do not seem to participate in the process, as the specific inhibition of various Ca2+ channel isoforms failed to affect this response in both neuromuscular junction and chromaffin cells [34,35]. Nevertheless, fluorescence-based evidence showed that TLY does promote a transient influx of Ca2+ into chromaffin cells in some other way, and its effect on neurosecretion was in fact blocked by the non-specific inhibitor La3+. In addition, the toxin induced a more sustained, localized increase in intracellular Ca2+ levels through release from caffeine-sensitive intracellular stores [35].

Further evidence as to how TLY affects membrane permeability came to light in 2002, when the voltage-clamp technique was used to investigate the mechanism of action underlying the neurosecretory effects induced by this cytolysin [57]. The perfusion of neuroblastoma/glioma cells (NG108-15) with TLY (12 nM) for ∼1 min increases membrane conductance, by means of an inward cationic current that was inhibited by anti-TLY antibodies and La3+. A lack of ionic selectivity was confirmed by the current reversing around 0 mV. The rather complex nature of the macroscopic current induced by TLY, which comprises both distinct steps and low fluctuations, and the unusually high single channel conductance associated to the current steps, point to the target of TLY being other than pre-existing channels. In addition, TLY-induced currents were also recorded in membrane patches free from channels. Taken together, these results, and the fact that, once bound, the toxin cannot be washed out, indicate pore-formation as the most likely mechanism underlying the increase in conductance promoted by TLY in neuroblastoma/glioma cells [57]. The fact that the membrane must be exposed to a certain amount of toxin for some time in order to suffer permeability changes supports this conclusion, being in fact a requirement predicted by the pore-forming model proposed for SNTX [55,66].
