Nitroxyl gets to the heart of the matter

Martin Feelisch* Louisiana State University Health Sciences Center, Department of Molecular and Cellular Physiology, 1501 Kings Highway, Shreveport, LA 71130

eart failure (HF) is a progres- arterial dilatation, allowing to stimulate oxygen consumption and energetics. sively disabling and ultimately cardiac output without increasing oxy- Taken together, these results suggest fatal disease, which is charac- gen consumption. Although conceptually that nitroxyl donors represent a novel terized by a decline in the ideal compounds, currently used inodila- class of inodilator with potential for the heart’sH ability to pump blood efficiently tors (compounds with positive inotropic treatment of HF. enough to meet the body’s metabolic and vasodilatior properties) tend to in- In most cases, not a single cause but demands. Despite substantial advances crease myocardial oxygen demand at a combination of systolic dysfunction in our understanding of the underlying higher doses, precipitating ischemia in (inability to contract and eject blood pathophysiology (1) and the therapeutic patients with coronary artery disease. normally) and͞or diastolic dysfunction management of acute and chronic HF Clearly, there is room for improvement (inability to relax and fill normally), en- (2) in recent years, the outlook of pa- in HF management, in particular with ergetic and vascular loading factors, tients with these conditions remains regard to quality of life and survival. contributes to the manifestation of HF. poor. Not only are mortality and mor- In this issue of PNAS, Paolocci et al. The type of cardiac dysfunction prevail- bidity discouragingly high, but also the (9) describe the beneficial cardiovascular ing and the accompanying hemodynamic patients’ quality of life remains impaired effects of nitroxyl (HNO)† in the failing situation of the patient have an obvious because of a substantial symptom bur- heart. A well characterized canine impact on the choice of pharmacological den. In the United States alone, HF is model of chronic heart failure was used treatment. The analysis of pressure- responsible for almost 1 million hospital in which cardiac dysfunction is produced volume loops obtained at different load- admissions (more than for all forms of by rapid ventricular pacing over a period ing conditions is among the best of all cancer combined) and Ͼ50,000 deaths of weeks. The authors used sophisti- current approaches to assess the con- each year, with estimated annual costs cated hemodynamic analyses suited to tractile behavior of the heart in vivo. exceeding $20 billion (3, 4). Despite im- discriminate direct cardiac effects from This approach, which also formed the proved patient information, beneficial indirect effects secondary to changes in basis of the present studies by Paolocci changes in lifestyle and better treatment preload and afterload to demonstrate et al. (9), has a long-standing history in options, HF remains to be a major pub- that nitroxyl increases myocardial con- experimental physiology, but its diagnos- lic health problem in industrialized na- tractility and enhances relaxation (posi- tic power in animal studies and clinical tions and the leading cause of hospital- tive lusitropic effect) in failing hearts. investigations has only been realized in ization in people older than 65 years. At These effects were accompanied by ar- the last two decades. In the past, effects a time when other cardiovascular dis- terial and venous dilation. Paolocci’s on myocardial contractility have been eases are on the decline, HF is rising finding that the cardiotonic action of difficult to evaluate because of the load ␤ and likely to further escalate over the HNO was unaffected by -receptor dependence of conventional measures of coming decades because of an aging blockade and additive to that of dobut- ventricular function. Since a couple of population and increased survival from amine is therapeutically significant not years ago, left ventricular end-systolic the underlying causes such as coronary only because the action of dopaminergic pressure-volume relationships can be heart disease and hypertension. A broad agonists and phosphodiesterase inhibi- assessed in the setting of a routine car- spectrum of different drugs and various tors are often attenuated in HF, but diac catheterization procedure. Using a guidelines for the treatment of HF exist also in view of the recent advent of ␤ conductance catheter with a microma- (5, 6). In the past, HF was mainly -blockers and their negative inotropic nometer tip for continuous measure- viewed as a problem of diminished car- effects in certain clinical settings. In ment of intraventricular volume and diac output. Maximization of the latter contrast to nitric oxide (NO)-generating pressure in combination with an occlud- with positive inotropic (contractility en- compounds, HNO production was not ing device to rapidly vary venous inflow, hancing) agents led to therapies that associated with increased plasma levels a largely load-independent measure of initially improved functional capacity, of the second messenger, cGMP. In- cardiac contractility can be obtained but increased mortality. Today, the ther- stead, enhanced concentrations of calci- without altering the status of the heart apeutic focus is on reducing elevated tonin gene-related peptide (CGRP) (11). In addition to the assessment of filling pressures that lead to the symp- were detected during HNO, but not NO systolic and diastolic function, additional toms of congestion (7). Although most administration, suggesting that the parameters allow estimation of the rela- recommendations agree on the major former may exert its favorable action, at tive effects of vasodilation on cardiac drug classes for the first-line and ad- least in part, via this endogenous neu- performance. In their studies, Paolocci junct therapy of HF, there is consider- ropeptide. Although the same group et al. (9) used a nitrate (nitroglycerin) able controversy about the role of posi- had previously observed positive inotro- and a NONOate (DEA͞NO) to gener- tive inotropic agents (8). Despite a pic effects of HNO in healthy hearts ate NO and Angelis’ salt to generate documented negative impact on survival, (10), the study outcome with this com- HNO. Although all three compounds these agents are still widely used, often pound in the setting of HF was not were used at equieffective doses as combined with vasodilators, to limit se- obvious. Numerous experimental and vere episodes of HF or as a bridge to clinical studies in the past have demon- transplantation. The rationale for com- strated that the same pharmacological See companion article on page 5537. bining vasodilatation with positive ino- principle capable of increasing contrac- *E-mail: [email protected]. tropic intervention lies in the possibility tility in the normal heart can produce †Not to be confused with nitroxyl radicals, which are to ‘‘unload’’ the heart, i.e., to reduce its negative inotropy in the failing heart spin traps for the detection of radical species by EPR preload and afterload by venous and due to unfavorable effects on cardiac spectroscopy.

4978–4980 ͉ PNAS ͉ April 29, 2003 ͉ vol. 100 ͉ no. 9 www.pnas.org͞cgi͞doi͞10.1073͞pnas.1031571100 Downloaded by guest on September 27, 2021 COMMENTARY

judged by the degree of reduction in vertable species (24), NOϪ may not be ing ischemia, nicotine, capsaicin (35), systolic pressure, the cardiac effects of readily oxidized to NO under all condi- nitroglycerin, and a nitroxyl donor (but NO were dramatically different from tions. Select nitroxyl donors, but not not other NO generating compounds) those of HNO. This is most likely due Angeli’s salt, have been shown to un- (36). An alternative way to increase to differences in the chemical properties dergo facile oxidative conversion to re- plasma CGRP is to slow down its enzy- of these two species, which dictates their lax vascular tissue and inhibit platelet matic breakdown by inhibiting neutral reactivity with endogenous biomolecules aggregation in a manner indistinguish- endopeptidase, but this is bound to af- and the signaling pathways affected. able from NO (25). The clear-cut di- fect numerous other pathways. It ap- NO is a ubiquitous endogenous mes- chotomy between the pharmacological pears fair to assume that the HNO- senger and modulator of cell function, profile of Angeli’s salt and that of NO induced CGRP increase in Paolocci’s which is produced from L-arginine by a donors observed by Paolocci et al. (9), study (9) originated from perivascular family of isoenzymes, the NO synthases however, indicates that HNO to NO nerves. It would be interesting to see (NOS) (12). HF is associated with re- conversion does not take place in every whether the inotropic effects of HNO duced expression of endothelial NOS tissue. This conclusion is further sup- are blunted in dogs depleted of endoge- and increased vascular oxidative stress, ported by their cGMP and CGRP mea- nous CGRP by prolonged infusion of which translates into diminished NO surements, which suggest that the car- capsaicin (37). Further insight may also availability, endothelial dysfunction and diac effects of HNO and NO are be gained from studies in CGRP-defi- reduced vasodilator capacity (13). NO is mediated by different signaling path- cient mice (33). Such investigations also the pharmacological principle of a ways. Besides the significance of these would not only confirm the cardiotonic number of drugs collectively termed ni- findings for HF, Paolocci’s study also mechanism proposed for nitroxyl, but trovasodilators, which are used clinically offers a novel pharmacological avenue may also shed new light on the role of to control hypertensive crises, protect for the modulation of CGRP levels. CGRP in HF. Both increased (38) and patients from attacks of angina pectoris CGRP is a 37-aa peptide that is syn- reduced (39) plasma CGRP concen- and to unload the heart during acute thesized by alternative splicing of the trations have been reported in HF. HF. Numerous other compounds, in- primary RNA transcript of the calcito- Whether these changes are reflections cluding NONOates are available to ex- nin gene in sensory neurons. Blood ves- of a counterregulatory mechanism or perimentally generate NO (14). Not- sels of all vascular beds are surrounded causally involved in disease progression withstanding the principal difference by a dense network of CGRP-containing is not known. Physiologically, it would that nitroglycerin requires tissue metab- nerve fibers, and most of the CGRP make sense to increase CGRP early dur- olism to generate NO whereas DEA͞NO circulating in plasma is thought to origi- ing HF as it would complement other releases it spontaneously, the cardiac nate from perivascular nerves (26). compensatory systems aimed at improv- effects of both compounds were similar CGRP is the most potent vasodilator ing cardiac efficiency. Prolonged stimu- (9), indicating that their action was me- known to date and thought to be in- lation of CGRP release may lead to diated by the same signaling mechanism. volved in the regulation of resting blood gradual peptide depletion, offering an Nitroxyl anion (NOϪ) is the one-elec- pressure and regional blood flow (27), explanation for the lower plasma levels tron reduction product of NO. Its chem- particularly in the coronary circulation observed in this and other studies istry is not very well understood and (28). In addition, it is a cardiotonic (9, 39). complicated by the fact that it exists in agent with positive inotropic and, in CGRP-related immunoreactivity is two electronic forms, a singlet and a normal subjects, positive chronotropic often found either together with NOS in triplet state (15). A recent reevaluation (heart rate increasing) effects (26). the same neuronal structures or in close of its pKa value revealed that at physio- CGRP interacts with specific cellular proximity to NOS-containing nerves. logical pH it exists largely in its proton- receptors that are coupled via G pro- Interestingly, NOS activity appears to be ated form, HNO (16), which can readily teins to adenylyl cyclase. The consecu- involved in capsaicin-induced CGRP cross cell membranes. Whether nitroxyl tive increase in cAMP is considered the release (40), and there is mounting evi- is formed in vivo is currently unclear. principal mechanism responsible for dence to believe that NOS is capable of Nevertheless, it may be formed from CGRP-mediated smooth muscle relax- producing NOϪ under certain conditions nitrosothiols (17), which are found to be ation, although NO-dependent effects (41, 42). Although admittedly specula- present in a variety of biological systems (29) and opening of ATP-sensitive po- tive, the mechanism of CGRP stimula- (18). In experimental settings, nitroxyl tassium channels (30) have been de- tion by HNO may not just be a peculiar can be conveniently generated by using scribed as well. In addition to cAMP, pharmacological phenomenon, but in Angelis’s salt (14). In fact, it was the phospholipase C may be involved in the fact represent an endogenous pathway spontaneous decomposition of this inor- stimulation of intracellular Ca2ϩ concen- involved in the fine-tuning of CGRP ganic salt that led chemists to postulate trations in the heart (31). In patients release. Whether all cardiovascular ef- the existence of HNO at the turn of the suffering from HF, CGRP has been fects of HNO are due to CGRP remains century (19). Angeli’s salt has been shown to reduce pulmonary and sys- to be investigated. In addition to the use shown to induce vasorelaxation and to temic pressures and increase cardiac of knockout animals this issue could be lower blood pressure (20, 21). Unlike performance without producing tachy- addressed by administration of a CGRP NO, which does not directly react with cardia (32). This is consistent with the receptor antagonist. Should the response sulfhydryl groups, HNO is a potent thiol absence of a change in heart rate with to HNO be only partially blunted by oxidant (22) and possesses a high affin- HNO in Paolocci’s study and suggests a CGRP receptor blockade it might be of ity for ferric heme proteins (23). The possible inhibitory modulation of sym- interest to investigate whether inhibition physiological significance of these or- pathetic nervous activity (33) at the si- of Naϩ͞Kϩ-ATPase (digitalis-like activ- thogonal properties of NO and HNO noatrial and͞or the arterial baroreceptor ity) or a Ca2ϩ-sensitizing component are are not entirely clear, but may offer an reflex level, which clearly distinguishes involved in addition. Considering the explanation for the discrete effects of nitroxyl donors from other positive ino- reactivity of HNO with thiols one possi- these two redox congeners in the failing tropes, including levosimendan (34). ble site of action might reside at the heart. Although it has been suggested A number of conditions and stimuli level of the ryanodine receptor, which that NOϪ and NO are redox-intercon- can cause the release of CGRP, includ- plays an important role in the regulation

Feelisch PNAS ͉ April 29, 2003 ͉ vol. 100 ͉ no. 9 ͉ 4979 Downloaded by guest on September 27, 2021 of intracellular Ca2ϩ transients (hence of inodilator. Additional studies will be there is no single, straightforward ap- contraction) by controlling Ca2ϩ release required to address whether nitroxyl proach for the therapeutic management from the sarcoplasmic reticulum and donors are subject to tolerance develop- of all patients. Hence, we will continue whose channel opening probability is ment, which often limits the effective- to require several individual agents with modulated by oxidation (43) and nitro- ness of organic nitrates. Tolerance to distinct pharmacological profiles to cor- sation (44) of critical thiols. Alterna- HNO might develop as a result of rect specific hemodynamic abnormali- tively, protein oxidation may shift the CGRP depletion from peripheral nerves ties. CGRP has shown potential for HF association͞dissociation equilibrium of or desensitization of signaling pathways management in clinical studies, but lacks the regulatory protein FKBP12 with the downstream of CGRP receptor activa- oral availability, is rapidly metabolized channel (45). Interestingly, the CGRP- tion, albeit there is no indication for this and has thus to be given by continuous related peptide, adrenomedullin has to occur from infusion studies with infusion. With no selective CGRP- been shown to enhance cardiac contrac- CGRP in man (47). To come up with a mimetic on the horizon and a recently tility via cAMP-independent mecha- drug candidate for commercial develop- renewed interest in inodilators (34), this nisms including Ca2ϩ release from ryan- ment that was sufficiently stable, orally may be a unique chance for nitroxyl do- odine-sensitive stores (46). available, and amenable to optimization nors. As with any new pharmacological Clearly, there is much more to learn of its pharmacokinetic properties, struc- principle at this early discovery stage, about the biological chemistry of HNO͞ tures are required that offer a variety of many obstacles have to be overcome NOϪ. Nevertheless, it looks as if there is possibilities for chemical derivatization. before a new lead compound can even- clear potential for therapeutic exploita- Further aspects that demand investiga- tually enter the developmental phase. tion of nitroxyl donors, and it appears tion include the frequency of unwanted Should nitroxyl donors pass these hur- timely to consider intensifying research side effects of nitroxyl donors such as dles in the next couple of years, HF may efforts in this relatively new field. Not- hypotension, headache, and gastrointes- become the key indication for such com- withstanding the fact that Angelis’ salt is tinal symptoms, which limits the useful- pounds in the future and HNO-based nothing more than an investigational ness of other vasodilators, and the risk inodilators a potentially useful addition tool, the studies by Paolocci et al. (9) of triggering ventricular arrhytmias. to the therapeutic arsenal available are nothing less than a proof-of-princi- Because of the overall hemodynamic for treatment of this life-threatening ple for a potentially promising new class complexity of the different forms of HF syndrome.

1. Braunwald, E. & Bristow, M. R. (2000) Circulation 16. Bartberger, M. D., Liu, W., Ford, E., Miranda, 32. Anand, I. S., Gurden, J., Wander, G. S., O’Gara, 102, IV14–IV23. K. M., Switzer, C., Fukuto, J. M., Farmer, P. J., P., Harding, S. E., Ferrari, R., Cornacchiari, A., 2. McMurray, J. & Pfeffer, M. A. (2002) Circulation Wink, D. A. & Houk, K. N. (2002) Proc. Natl. Panzali, A., Wahi, P. L. & Poole-Wilson, P. A. 105, 2099–2106 and 2223–2228. Acad. Sci. USA 99, 10958–10963. (1991) J. Am. Coll. Cardiol. 17, 208–217. 3. American Heart Association (2002) Heart Disease 17. Wong, P. S., Hyun, J., Fukuto, J. M., Shirota, F. N., 33. Oh-hashi, Y., Shindo, T., Kurihara, Y., Imai, T., and Stroke Statistics–2003 Update (American DeMaster, E. G., Shoeman, D. W. & Nagasawa, Wang, Y., Morita, H., Imai, Y., Kayaba, Y., Nishi- Heart Association, Dallas). H. T. (1998) Biochemistry 37, 5362–5371. matsu, H., Suematsu, Y., et al. (2001) Circ. Res. 89, 4. Rich, M. W. & Nease, R. F. (1999) Arch. Intern. 18. Stamler, J. S. (1995) Curr. Top. Microbiol. Immu- 983–990. nol. 196, Curr. Opin. Med. 159, 1690–1700. 19–36. 34. Cleland, J. G. & McGowan, J. (2002) 19. Angeli, A., Angelico, F. & Scurti, F. (1902) Chem. Cardiol. 17, 257–265. 5. Remme, W. J. & Swedberg, K. (2001) Eur. Heart J. Zentralbl. 73, 691–693. 35. Lundberg, J. M., Franco-Cereceda, A., Alving, K., 22, 1527–1560. 20. Zamora-Pino, R. & Feelisch, M. (1994) Biochem. Delay-Goyet, P. & Lou, Y. P. (1992) Ann. N.Y. 6. Hunt, S. A., Baker, D. W., Chin, M. H., Cinque- Biophys. Res. Commun. 201, 54–62. Acad. Sci. 657, 187–193. grani, M. P., Feldman, A. M., Francis, G. S., 21. Vanin, A. F., Vedernikov, Y. I., Galagan, M. E., 36. Booth, B. P., Tabrizi-Fard, M. A. & Fung, H. Ganiats, T. G., Goldstein, S., Gregoratos, G., Kubrina, L. N., Kuzmanis, Y. A., Kalvin’sh, I. & (2000) Biochem. Pharmacol. 59, 1603–1609. Jessup, M. L., et al. (2001) J. Am. Coll. Cardiol. 38, Mordvintsev, P. I. (1990) Biokhimiia 55, 1408–1413. 37. Wharton, J., Gulbenkian, S., Mulderry, P. K., 2101–2113. 22. Cook, N. M., Shinyashiki, M., Jackson, M. I., Leal, Ghatei, M. A., McGregor, G. P., Bloom, S. R. & 7. Nohria, A., Lewis, E. & Stevenson, L. W. (2002) F. A. & Fukuto, J. M. (2003) Arch. Biochem. Polak, J. M. (1986) J. Auton. Nerv. Syst. 16, 289–309. J. Am. Med. Assoc. 287, 628–640. Biophys. 410, 89–95. 38. Ferrari, R., Panzali, A. F., Poole-Wilson, P. A. & 8. Felker, G. M. & O’Connor, C. M. (2001) Am. 23. Miranda, K. M., Nims, R. W., Thomas, D. D., Anand, I. S. (1991) Lancet 338, 1084. Heart J. 142, 393–401. Espey, M. G., Citrin, D., Bartberger, M. D., Pa- 39. Taquet, H., Komajda, M., Grenier, O., Belas, F., 9. Paolocci, N., Katori, T., Champion, H. C., St. olocci, N., Fukuto, J. M., Feelisch, M. & Wink, Landault, C., Carayon, A., Lechat, P., Grosgogeat, John, M. E., Miranda, K. M, Fukuto, J. M., Wink, D. A. (2003) J. Inorg. Biochem. 93, 52–60. Y. & Legrand, J. C. (1992) Eur. Heart J. 13, D. A. & Kass, D. A. (2003) Proc. Natl. Acad. Sci. 24. Murphy, M. E. & Sies, H. (1991) Proc. Natl. Acad. 1473–1476. USA 100, 5537–5542. Sci. USA 88, 10860–10864. 40. Hughes, S. R. & Brain, S. D. (1994) Br. J. Phar- 10. Paolocci, N., Saavedra, W. F., Miranda, K. M., 25. Zamora, R., Grzesiok, A., Weber, H. & Feelisch, macol. 111, 425–430. Martignani, C., Isoda, T., Hare, J. M., Espey, M. (1995) Biochem. J. 312, 333–339. 41. Schmidt, H. H., Hofmann, H., Schindler, U., M. G., Fukuto, J. M., Feelisch, M., Wink, D. A. & 26. Wimalawansa, S. J. (1996) Endocr. Rev. 17, 533–585. Shutenko, Z. S., Cunningham, D. D. & Feelisch, M. Kass, D. A. (2001) Proc. Natl. Acad. Sci. USA 98, 27. Gangula, P. R., Zhao, H., Supowit, S. C., Wimala- (1996) Proc. Natl. Acad. Sci. USA 93, 14492–14497. J. Biol. 10463–10468. wansa, S. J., Dipette, D. J., Westlund, K. N., Gagel, 42. Adak, S., Wang, Q. & Stuehr, D. J. (2000) R. F. & Yallampalli, C. (2000) Hypertension 35, Chem. 275, 33554–33561. 11. Kass, D. A. & Maughan, W. L. (1988) Circulation 470–475. 43. Abramson, J. J. & Salama, G. (1989) J. Bioenerg. 77, 1203–1212. 28. DiPette, D. J., Schwarzenberger, K., Kerr, N. & Biomembr. 21, 283–294. 12. Moncada, S. & Higgs, A. (1993) N. Engl. J. Med. Holland, O. B. (1987) Hypertension 9, III142– 44. Eu, J. P., Sun, J., Xu, L., Stamler, J. S. & Meissner, 329, 2002–2012. III146. G. (2000) Cell 102, 499–509. 13. Sharma, R. & Davidoff, M. N. (2002) Congest. 29. Brain, S. D. & Cambridge, H. (1996) Gen. Phar- 45. Hasenfuss, G. & Seidler, T. (2003) Circulation 107, Heart Fail. 8, 165–172. macol. 27, 607–611. 378–380. 14. Feelisch, M. & Stamler, J. S. S. (1996) in Methods In 30. Nelson, M. T., Huang, Y., Brayden, J. E., Hes- 46. Szokodi, I., Kinnunen, P., Tavi, P., Weckstrom, Nitric Oxide Research, eds. Feelisch, M. & Stamler, cheler, J. & Standen, N. B. (1990) Nature 344, M., Toth, M. & Ruskoaho, H. (1998) Circulation J. S. S. (Wiley, Chichester, U.K.), pp. 71–115. 770–773. 97, 1062–1070. 15. Hughes, M. N. (1999) Biochim. Biophys. Acta. 31. Aiyar, N., Disa, J., Stadel, J. M. & Lysko, P. G. 47. Stevenson, R. N., Roberts, R. H. & Timmis, A. D. 1411, 263–272. (1999) Mol. Cell. Biochem. 197, 179–185. (1992) Int. J. Cardiol. 37, 407–414.

4980 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.1031571100 Feelisch Downloaded by guest on September 27, 2021