Cardiac expression and distribution of nitric oxide synthases in the ventricle of the cold-adapted Antarctic teleosts, the hemoglobinless Chionodraco hamatus and the red-blooded Trematomus bernacchii

Nitric Oxide 15 (2006) 190–198 www.elsevier.com/locate/yniox

Cardiac expression and distribution of nitric oxide synthases in the ventricle of the cold-adapted Antarctic teleosts, the hemoglobinless Chionodraco hamatus and the red-blooded Trematomus bernacchii Daniela Amelio a,1, Filippo Garofalo a,1, Daniela Pellegrino a,b, Francesca Giordano a, Bruno Tota a,¤, Maria Carmela Cerra a,b a b

Department of Cellular Biology, University of Calabria, 87030 Arcavacata di Rende, CS, Italy Department of Pharmaco-Biology, University of Calabria, 87030 Arcavacata di Rende, CS, Italy Received 15 September 2005; revised 14 December 2005 Available online 25 January 2006

Abstract The presence of nitric oxide synthase (NOS) was investigated in the ventricle of two Antarctic teleosts, the hemoglobinless iceWsh Chionodraco hamatus and its red-blooded counterpart, Trematomus bernacchii. Under unstimulated conditions, in both teleosts, NADPH–diaphorase localised NOS activity in the endocardial–endothelial cells (EEc) and in the myocardiocytes. Application of antimammalian endothelial and inducible NOS (eNOS and iNOS, respectively) primary antibodies for immunoXuorescence revealed a comparable tissue-speciWc basal expression of the two isoforms in the two species. eNOS strongly localised at the level of the EEc and, in T. bernacchii, of the vascular endothelium (VE). The enzyme is also localised, albeit to less extent, within the myocardiocytes, and in the epicardium. In contrast, iNOS immunostaining only labels the cytoplasm of the ventricular myocytes. Western blotting analysis identiWed two peptides with molecular masses of about 135 and 130 kDa, similar to those of the mammalian eNOS and iNOS. To verify whether this NOS system is susceptible to septic stimulation, C. hamatus and T. bernacchii were exposed to bacterial lipopolysaccharide (LPS). The treatment did not modify the distribution pattern of the two isoenzymes while it increased the amount of NADPH–diaphorasedependent reaction product and the expression of both eNOS and iNOS. These results indicate a high phylogenetic conservation of the intracardiac NOS system, emphasizing its importance in the control of the vertebrate heart and its relevance as a general mechanism of defense against pathogens. © 2005 Elsevier Inc. All rights reserved. Keywords: Cardiac nitric oxide synthase; Antarctic teleosts; NADPH–diaphorase; ImmunoXuorescence; Western blotting

The Antarctic teleosts belonging to the Perciform suborder Notothenioidei have evolved during the last 25 million years in geographic isolation within the near-zero °C waters of the Antarctic sea [1]. Their unique cold-adapted physiology is characterised by a number of structural and functional modiWcations, including, in some cases, the disaptive loss of traits, which in turn call for homeostatic compensation. Examples of this are: (i) the loss of the heatshock response leading to the consequent constitutive *

1

Corresponding author. Fax: +0984 492906. E-mail address: [email protected] (B. Tota). These two authors contributed equally to this study.

1089-8603/$ - see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.niox.2005.12.007

expression of heat-shock proteins [2]; (ii) the absence, in the Chaennichthyidae family (iceWsh), of the oxygen-binding proteins hemoglobin (Hb) and, in some species, cardiac myoglobin (Mb) ([3,4], respectively). The hemoglobinless condition, which is not lethal thanks to the high oxygen solubility of the cold Antarctic waters and the generally sluggish locomotory habits of these Wshes ([5] and references therein), has stimulated relevant compensatory changes at the circulatory level, including a severe hematocrit decrement, high blood volume, low peripheral resistance, and large heart ventricle with remarkably high cardiac output ([6] and references therein). The heart operates as a volume pump at low frequencies, high Xow rate, and low pressures

D. Amelio et al. / Nitric Oxide 15 (2006) 190–198

([7] and references therein). In addition, the cardiac muscle has developed an increased mitochondria compartment to enhance maximum aerobic power and/or facilitate oxygen diVusion [8–10]. We have recently shown that the nitric oxide (NO) biochemical pathway has been retained in the hemoglobinless C. hamatus [6]. In particular, the presence of iNOS, in parallel with a tonic NO–cGMP modulation of myocardial performance, has been documented in the heart of this iceWsh [6]. Studies in mammals have established that, within the heart, NO is generated in various cell types (i.e., myocardiocytes, endothelial and endocardial cells, interstitial cells, and intracardiac neurons) by three NOS isoforms (NOSs), namely NOS1 or neuronal NOS (nNOS), NOS3 or endothelial NOS (eNOS), and NOS2 or inducible NOS (iNOS) [11]. The constitutive isoforms, eNOS and nNOS, are calcium activated low-output enzymes, while iNOS is a high-output and calcium independent enzyme. It is generally believed that iNOS expression is enhanced during sepsis or after exposure to cytokines or bacterial products such as the Gram-negative endotoxin LPS [12,13]. However, this concept has been recently questioned by the identiWcation of a constitutively expressed iNOS in mammalian cardiac cells [14,15]. More recently, the existence of a mitochondrial NOS (mtNOS; [47]), corresponding to a variant of nNOS, has been reported in various cell types [16], including cardiomyocytes [17]. Being the most regulated among the known enzymes, NOSs activities, which are modulated by cofactors and calcium availability, post-transductional modiWcations, and protein–protein interactions [18], crucially depend from their subcellular localization. In fact, the reactive and diVusive nature of NO requires the isoenzymes to be located in proximity of their targets [11] as epitomised by the caveolar localization of eNOS associated with the caveolar protein, caveolin [19]. This complex intracardiac NOSs spatial conWnement (at tissue, cellular, and subcellular levels) is responsible for the diverse, and sometimes opposite, inXuences on myocardial performance [11,12]. An additional route to modulate NO bioactivity is achieved by controlling its rate of consumption. An important NO disarming mechanism is represented by Hb. This protein, in its oxygenated form, rapidly reacts with NO to produce methemoglobin and nitrate [20]. Moreover, Hb has been postulated to be involved in the systemic transport and delivery of NO to tissues and in the facilitation of oxygen release [21,22], as well as to function as a nitrite reductase [23]. Therefore, given the key role of Hb in NO homeostasis, Antarctic iceWshes provide exclusive opportunities to investigate, in naturally occurring genetic knockouts for Hb, whether the pigment disaptive loss may have elicited modiWcations of the intracardiac NOS enzymatic pattern. To this aim, we have comparatively analysed the NOS system in the heart of C. hamatus and T. bernacchii, the latter representing the red-blooded counterpart of the iceWsh. In fact, apart from the iceWsh cardiovascular adjustments, the two species share phylogenetic and ecophysio-

191

logical traits. Both belong to the sub-order Notothenioidei, have evolved for between 14 and 25 million years as endemic sluggish bottom dwellers in the sub-zero Antarctic waters, and have developed extreme stenothermia [1]. By NADPH–diaphorase, immunoXuorescence, and Western blotting, we have demonstrated that the NOS system is present in the ventricle of C. hamatus and T. bernacchii. We found that, regardless of the presence or absence of Hb and their diVerent cardiovascular adjustments, both teleosts, under unstimulated conditions, exhibit in their hearts similar tissue-speciWc expression of the two NOS isoforms eNOS and iNOS. We have also shown that the ability to up-regulate this enzymatic system in response to LPSdependent septic challenges has been retained in the heart of both species. Taken together, these results highlight the ubiquitous presence of the intracardiac NOS system and its adaptive importance in the Antarctic cold-adapted teleosts. Materials and methods Animals This research was carried out on 27 C. hamatus Lönnberg of both sexes (animal weight: 414 § 83 g, mean § SEM) and 27 T. bernacchii of both sexes (animal weight: 219 § 51 g, mean § SEM). Fishes were caught by net in the Terranova Bay, Ross Sea, Antarctica, during the XIX Italian Antarctic Expedition (December 2003–February 2004). After capture, C. hamatus and T. bernacchii were separated and maintained in tanks containing aerated, running seawater at temperatures between 0 and 2 °C. Animal care, sacriWce, and experiments were supervised under the European Community guiding principles in the care and use of animals and the projects supervised by the Local Ethical Committee. Each group of Wsh was further divided into two groups which were kept in separate tanks until sacriWce. Fishes belonging to the Wrst group received both intraperitoneal and rectal injection of 2 g/L of bacterial LPS while the second group did not receive any treatment. For comparison, a third group of both C. hamatus and T. bernacchii was treated like the LPS-treated group but with equivalent volumes of saline solution alone. Fifteen days after treatment, Wshes were anaesthetised in benzocaine (0.2 g/L) for 15 min and sacriWced; the hearts were rapidly removed and processed for the speciWc protocol. Both treatment and sacriWce of the Wsh were carried out in the Italian Antarctic Base laboratories, Terranova Bay (74°42⬘S, 164°06⬘E), Antarctica. Histochemistry, immunoXuorescence, and Western blotting were performed in Italy. Morphological determinations NADPH–diaphorase histochemistry Histochemical staining with NADPH–diaphorase was utilised to visualise total NOS activity by light microscopy at the level of the ventricle of C. hamatus and T. bernacchii of untreated, LPS-, and saline-treated animals. This

192

D. Amelio et al. / Nitric Oxide 15 (2006) 190–198

histochemical method is considered highly selective for NOS activity [24]. BrieXy, the hearts of nine (three untreated, three LPS-treated, and three saline-treated) C. hamatus and nine (three untreated, three LPS-treated, and three saline-treated) T. bernacchii were Wxed in 4% paraformaldehyde solution for 5–7 h. Tissues were cryoprotected in an optimal cutting temperature compound (OCT, IMEB INC) and rapidly frozen in liquid nitrogen. Transverse ventricular cryostat (Microm HM505E) sections (5 m) were incubated for 1 h at 37 °C in Tris–HCl (0.1 M, pH 7.5) containing Triton X-100 (0.3%), nitro blue tetrazolium (NBT, 0.6 mM), -NADPH (1 mM), and sodium azide (1 mM). The reaction was stopped by replacing the incubation medium with Tris–HCl buVer. The sections were then washed in distilled water, dehydrated in graded alcohols, cleared in xylene, mounted with Eukitt (O. Kindler GmbH & Co), observed under a light microscope (ZEISS: Axioscope), and digitalised by Olympus ZC200 camera. Parallel control sections were treated as above except that NADPH or NBT was omitted from the incubation medium. ImmunoXuorescence For the immunodetection of NOSs, the hearts of C. hamatus and T. bernacchii (three for each group of treatment) were embedded in OCT, Wxed in liquid nitrogen, and stored at ¡80 °C until use. Cryostat ventricular sections (7 m) were postWxed with acetone for 10 min and stored at ¡20 °C until use. Immunolocalization of NOSs was carried out on ventricular sections by using a primary polyclonal antibody against eNOS (Sigma, St. Louis, Missouri; 1:100) and a FITC-conjugated primary monoclonal antibody against iNOS (BD Transduction Laboratories; 1:100). Anti-eNOS was developed in rabbit against polypeptides corresponding to eNOS epitope of bovine origin at the Nterminal region (amino acids: 1185–1205). Anti-iNOS was generated from mouse iNOS. Anti-rabbit IgG FITC-conjugated (Sigma, St. Louis, Missouri; 1:100) was utilised to detect eNOS signal. Before immunostaining, slides were airdried, washed with Tris–HCl buVer saline (TBS), and incubated with anti-eNOS or anti-iNOS antibodies overnight at 4 °C. The reaction was stopped by rinsing the sections with TBS. Subsequently, the sections were incubated with propidium iodide (1:100) for 5 min to obtain nuclear counterstaining. Slides were then mounted with mounting medium (Vectashield, Vector Laboratories) and observed under a confocal laser microscope (TCS-SP2, Leika). Negative controls were obtained on parallel ventricular sections treated in the same manner, excluding primary antibodies (both direct and indirect immunoXuorescence). Biochemical determinations For both NADPH–diaphorase determination on homogenates and Western blotting analysis, hearts of C. hamatus and T. bernacchii (three for each group of treatment) were rapidly immersed in liquid nitrogen and stored

at ¡80 °C. According to Barroso et al. [25], the ventricles were suspended in ice-cold Tris–HCl buVer (30 mM; pH 7.4) containing EGTA (15 M), EDTA (10 M), dithiothreitiol (5 M), pepstatin-A (0.01 M), PMSF (1 M), leupeptin-A (0.02 M), benzamidine (0.1 M), and BH4 (0.1 M). They were then homogenised with Ultra Turrax at 22,000 rpm. The homogenates were centrifuged at 105,000g for 60 min at 4 °C and the supernatant was collected for both NADPH–diaphorase and Western blotting analysis. Protein concentration was calculated using Bradford’s dye-binding method [26] with albumin as standard. Measurement of NADPH–diaphorase activity According to Virgili et al. [27], 20 L of ventricular supernatant was incubated for 15 min at 37 °C with paraformaldehyde 4%, plus 1 ml of ice-cold Tris–HCl buVer (50 mM), pH 7.5, containing 0.1% Triton X-100, EDTA (5 mM), NBT (0.5 mM), and -NADPH (0.5 mM). The reaction product was detected spectrophotometrically at 585 nm. Control samples were obtained with the same protocol deprived of -NADPH. Absorbance values were then normalised for the protein content of each sample. Western blotting and densitometric analysis Samples of supernatants containing 100 g of proteins were heated for Wve minutes in sample buVer according to Laemmli (Fluka), separated by SDS–PAGE using 8% gel in a Bio-Rad Mini Protean III, and then electroblotted onto polyvinylidene diXuoride membrane (Hybond-P Amersham) using a mini trans-blot (Bio-Rad). The membrane was blocked with TBS-T buVer containing 5% non-fat dry milk. For immunodetection, the blots were incubated overnight at 4 °C with monoclonal mouse anti-iNOS (Transduction Laboratories) and polyclonal rabbit anti-eNOS (Sigma, St. Louis) antibodies diluted 1:1000 in TBS-T. The secondary antibodies peroxidase linked (Amersham) were diluted 1:5000 in TBS-T. The immunodetection was performed using an enhanced chemiluminescence kit (ECL Plus, Amersham). Autoradiographs were obtained by exposure to X-ray Wlms (HyperWlm ECL, Amersham). Immunoblots were digitalised and the densitometric analysis of the bands was carried out using NIH Image 1.61 for Macintosh computer based on 256 grey values. QuantiWcation of the bands was obtained by measuring the mean optical density of a square area (Wve times on each band), after the background has been subtracted. Statistics The results of absorbance measurements and the grey values obtained from the densitometric analysis were expressed as means § SE of Wve determinations for each sample. Comparisons were made using two-way analysis of variance (ANOVA). Statistical signiWcance of the diVerences was calculated by using Duncan’s multiple-range test (p < 0.05).

D. Amelio et al. / Nitric Oxide 15 (2006) 190–198

Results By means of both immunological and biochemical techniques, we described the basal NOS enzymatic pattern in the ventricle of the two Antarctic teleosts, the Hb-free C. hamatus and the red-blooded T. bernacchii. We also quantiWed the eVects of LPS treatment on both NOS expression and activity. The analysis, extended to the saline-treated Wshes for comparison purposes, showed no diVerences with respect to the untreated group. Thus, the results obtained from the saline-treated Wshes have not been reported here. NADPH–diaphorase histochemistry NADPH–diaphorase histochemistry revealed a comparable localization of the total NOS activity in the ventricular sections of untreated C. hamatus and T. bernacchii (Figs. 1A and B). In both teleosts, the formazan dark blue reaction product, indicative of the NADPH–diaphorase activity, is densely localised at the level of the endocardial-endothelium (EE) which lines the lacunary spaces. A lower but speciWc NADPH–diaphorase positivity was also detected in cardiac muscle Wbres. No background was observed and the speciWcity of the reaction was conWrmed by the absence of staining in control sections. Evaluation of NOS localization in the ventricle of LPS-treated C. hamatus and T. bernacchii revealed an increased staining at the level of both EE and myocardial cells (data not shown). ImmunoXuorescence Confocal observation of cardiac sections immunolabeled with polyclonal anti-eNOS and monoclonal antiiNOS speciWc antibodies, revealed the presence of the two NOS isoforms in the spongy-type ventricle of untreated C. hamatus and T. bernacchii. The absence of Xuorescence in the parallel control sections treated without primary antibodies conWrmed the speciWcity of the labelling. In the two teleosts, eNOS positive staining was detected at the level of

193

the EE envelope of the trabeculae of the spongiosa and, in a lesser extent, of the myocardiocytes (C. hamatus, Figs. 2A and B; T. bernacchii, Fig. 2C). In both tissues, the enzyme was prevalently associated with the membrane, however, especially in some EEc, a slight and diVuse eNOS-dependent immunoXuorescence is detectable also within the cytoplasm (Figs. 2a and b). In T. bernacchii, eNOS was also localized at the level of the endothelium of the subepicardial vessels (data not shown). An intense eNOS staining was also identiWed in the ventricular epicardial region of the two teleosts (C. hamatus, Fig. 2B; T. bernacchii, Fig. 2C). Contrary to eNOS, ventricular iNOS immunostaining was exclusively present in the myocardiocytes. In fact, no immunoXuorescence was detected at the level of the EE and the epicardium (C. hamatus and T. bernacchii), and of the VE (T. bernacchii). In transversely cut myocardiocytes, the Xuorescent signal appears uniformly distributed within the cytoplasm, while in longitudinally cut cells, it follows the myoWbrils (Figs. 3A and B). Treatment with LPS did not change the distribution pattern of both NOS isoforms (data not shown). Measurement of NADPH–diaphorase activity The application of a variant of the NADPH–diaphorase protocol [27] to the ventricular extracts of C. hamatus and T. bernacchii provided a quantitative evaluation of the total activity of all NOS isoforms. As shown in Fig. 4, under basal conditions, the hearts of both teleosts showed comparable levels of NOS activity. After LPS treatment, these levels signiWcantly increase reaching 34.6% in C. hamatus and 101.7% in T. bernacchii (Fig. 4). Western blotting and densitometric analysis The presence of eNOS and iNOS in the ventricle of both untreated and LPS-treated C. hamatus and T. bernacchii was conWrmed by the application of eNOS and iNOS

Fig. 1. NADPH–diaphorase staining on the ventricular myocardium of C. hamatus (A) and T. bernacchii (B) under basal conditions. Note the dark-blue formazan precipitate at the level of the EE (red arrow) and the myocardiocytes (black arrow). (For interpretation of the references to color in this Wgure legend, the reader is referred to the web version of this paper.)

194

D. Amelio et al. / Nitric Oxide 15 (2006) 190–198

Fig. 2. Representative confocal images of eNOS immunolabelling on ventricular sections of C. hamatus (A and B) and T. bernacchii (C). (a–c) Digital magniWcation of a particular of the eNOS ventricular localization in C. hamatus (a and b) and T. bernacchii (c). For descriptive purposes, in a1, b1, and c1 are shown the plot proWles (in 256 grey values) obtained at the level of the yellow line of the corresponding inset. Note in (c1) the higher values indicating zones of intense EE labelling. In (A and B), the nuclei (red) are counterstained with propidium iodide. Yellow arrow: EE; white arrow: myocardiocytes; blue arrow: epicardium. (For interpretation of the references to color in this Wgure legend, the reader is referred to the web version of this paper.)

antibodies for Western blotting followed by densitometric analysis. Immunoreactive bands of approximately 135– 130 kDa were observed (Figs. 5A and B), corresponding to the known eNOS and iNOS molecular masses, respectively [28]. Densitometric quantiWcation of the blots (Figs. 5A1 and B1) revealed that in C. hamatus and T. bernacchii, exposure to LPS increases both eNOS (C. hamatus, 7.2%; T. bernacchii, 36.8%), and iNOS levels (C. hamatus, 4.7%; T. bernacchii, 22.8%). Discussion This study shows that eNOS and iNOS isoforms are present and functionally expressed, both under basal conditions and after LPS stimulation, in the heart ventricle of the cold-adapted Antarctic teleosts, the hemoglobinless C. hamatus and the red-blooded T. bernacchii. These Notothenioid species share a common history in thermal adaptation, habitat, and life style. The distribution pattern of the total NOS activity, detected by NADPH–diaphorase, provides an indirect evidence of the intraventricular sources of NO production in the two Notothenioids, suggesting the presence of NOdependent autocrine–paracrine regulation of cardiac activity. NADPH–diaphorase technique is successfully

utilised as a selective marker of overall (i.e., isozyme-independent) NOS activity in the heart [29–31]. In fact, in the heart there are no other diaphorases to confound NOSdependent signal. Moreover, possible interferences due to elevated number of respiratory chain enzymes in the myocardium are prevented by addition of sodium azide in the incubation medium. This reduces non-speciWc background staining, and enhances the speciWcity of the reaction [30]. Light microscope observation of the ventricular sections revealed the same distribution pattern of NOS enzymatic activity in the two Notothenioids. Under unstimulated conditions, in both teleosts NADPH–diaphorase staining densely localizes, with similar intensity, in the EEc lining the extense lacunary spaces of the trabeculate (spongiosa) ventricle, and, in a lesser extent, in the myocardiocytes of the trabeculae. This prevalent EE localization agrees with the pattern of NADPH–diaphorase localization, described in the mammalian heart. In fact, in rat and guinea pig hearts, the dark blue reaction product denoting NOS activity was found along every vessel (arteries, veins, and capillaries) and throughout the EE including that of the valves [29]. QuantiWcation of NADPH–diaphorase activity on ventricular homogenates conWrmed that the ventricles of the two teleosts possess equivalent amounts of total NOS activity.

D. Amelio et al. / Nitric Oxide 15 (2006) 190–198

Fig. 3. Representative confocal images of iNOS immunolabelling on ventricular sections of C. hamatus (A) and T. bernacchii (B). (a and b) illustrate the digital magniWcation of a particular of the iNOS ventricular localization in C. hamatus (a) and T. bernacchii (b). For descriptive purposes, in (a1 and b1) are shown the plot proWles (in 256 grey values) obtained at the level of the yellow line of the corresponding inset. Note the peaks of the myocardial labelling and the absence of EE signal. In (A and B), the nuclei (red) are counterstained with propidium iodide. Yellow arrow, EE; white arrow, myocardiocytes; blue arrow, epicardium. (For interpretation of the references to color in this Wgure legend, the reader is referred to the web version of this paper.)

Fig. 4. Spectrophotometric quantiWcation of NADPH–diaphorase reaction on ventricular extracts of C. hamatus and T. bernacchii under basal conditions (CTRL) and after LPS stimulation (LPS). Data are means § SEM of three determinations for each animal (n D 3). Statistical diVerences were evaluated by ANOVA (**p < 0.05).

To evaluate the contribution of eNOS and iNOS isoforms to the NADPH–diaphorase-dependent signal, we utilised, for both immunoXuorescence and Western blot-

195

Fig. 5. Western blotting signals of eNOS (A) and iNOS (B) in the ventricular extracts of untreated (CTRL) and LPS-treated (LPS) C. hamatus and T. bernacchii. (A1 and B1) Densitometric quantiWcation of the blots. Data are means § SEM of three determinations for each animal (n D 3). Statistical diVerences were evaluated by ANOVA (*p < 0.05).

ting, commercially available antibodies directed toward mammalian eNOS and iNOS. The immunoXuorescence analysis revealed that in both Wshes the two NOS isoenzymes are detectable in the ventricular tissues. The speciWcity of this immunodetection was assessed by the lack of Xuorescent signal in control sections incubated without the primary antibody and was also validated by Western blotting analyses. In fact, when the ventricular extracts were immunoblotted with the same anti-mammalian NOS antibodies, we identiWed two peptides with molecular masses of approximately 135 and 130 kDa, corresponding to the masses of the mammalian eNOS and iNOS enzymes, respectively [28]. These results represent the Wrst evidence that in both C. hamatus and T. bernacchii eNOS and iNOS enzymes have sequence domains which are recognised by mammalian antibodies. This agrees with the results of interspecies comparisons of NOS isozyme homologs which have shown more than 90% sequence conservation for eNOS and about 80% sequence conservation for iNOS. Recently, in several teleosts iNOS has been partially sequenced revealing a high degree of homology with other vertebrate iNOSs. It has been found that the iNOS sequence of Salmo salar has 52% identity with human iNOS, 65% identity with

196

D. Amelio et al. / Nitric Oxide 15 (2006) 190–198

goldWsh iNOS, and 95% identity with trout iNOS ([32] and references therein). In view of the phylogenetic history of these two Notothenioid species, isolated in the frigid waters of the Southern Ocean and characterised by notable stenothermia, the conservation of the NOS proteins suggests that these two enzymes appeared in the Antarctic Notothenioidei before the geographic isolation of the Southern Ocean around 25 million years ago [1]. Interestingly, the immunoXuorescence data reveal that, under basal conditions in both the iceWsh and the redblooded counterpart eNOS and iNOS are constitutively expressed and have isoform-speciWc cell-type localization. eNOS is prevalently located at the level of the subepicardial vessels (T. bernacchii) and the EEc while iNOS is exclusively localized at myocardiocyte level. This NOS expression pattern is similar to that observed in the mammalian heart in which eNOS immunohistochemistry revealed the enzyme in both EE and VE and, in a lesser extent, in the myocardium ([33] and references therein). In contrast, iNOS was exclusively identiWed in cardiomyocytes also under basal conditions ([14] and references therein). In our preparations, eNOS appears prevalently associated with the plasma membrane. It is well known that eNOS, thanks to a dual acylation, is targeted to caveolar domains in plasma membranes in which it interacts with caveolin-1 in endothelial cells and with caveolin-3 in myocardiocytes [12,34]. This interaction represents a major mechanism to couple eNOS-dependent NO production to upstream signals such as shear stress [35], oxygen [36], and stretch [12]. The caveolin–eNOS association inhibits the enzyme while the inhibition is relieved by both calmodulin, which dissociates eNOS from caveolin [18], and the heat-shock protein HSP90 which facilitates the dissociation of eNOS from caveolin [37]. At the endothelial level, either endocardial or vascular, these protein–protein interactions contribute to the eNOS-NO-dependent paracrine control of the contractile function [12]. It has been recently shown that in T. bernacchii, the extreme stenothermia has modiWed the heatshock response, leading to the constitutive expression of heat-shock proteins (i.e., HSP70 [38]). It should be interesting to evaluate whether this has induced functional consequences on the molecular interactions which control eNOS activation. Moreover, because of the trabeculated myocardium, the ventricles of both C. hamatus and T. bernacchii possess a very extensive EE which is continuously exposed to the superfusing turbulent lacunary blood Xow, being subjected to a remarkable shear stress. Conceivably, as in the mammalian endocardium, the intraluminal shear stress may up-regulate eNOS [35,33]. Notably, in the ventricle of both Wshes, several EEc appear immunolabeled by eNOS at the cytoplasm level. It has been shown in mammalian EE that, in addition to the membrane targeting, low eNOS amounts also associate with cytoplasmic components such as the Golgi apparatus [33]. Further biochemical and immunological studies will be required to assess if this is also the case in the two Notothenioids.

An intriguing feature of the ventricular immunolabelling is the eNOS localization in the visceral pericardium. In mammals, this tissue may aVect cardiac performance by releasing cardiomodulatory substances (i.e., atrial natriuretic peptides, endothelin-1, and growth factors) into the pericardial Xuid [39]. Also in C. hamatus and T. bernacchii, the visceral pericardium shows morpho-functional traits indicative of autocrine–paracrine activity, as indicated by the presence of receptors for atrial natriuretic peptide [40]. Moreover, in T. bernacchii, the subepicardium appears organised as a germinal centre well suited to produce humoral immune response [41]. Together with the above observations, the presence of eNOS in the epicardium of the two teleosts suggests that the locally produced NO may contribute to the autocrine–paracrine role of this cardiac region. By immunoXuorescence, we found that the ventricles of both C. hamatus and T. bernacchii express iNOS under unstimulated conditions. Thus, in the myocytes of these teleosts iNOS may act in concert with eNOS contributing to the tonic generation of the NO pool. Such basal expression of iNOS is in accordance with the recent literature showing the constitutive expression of this enzyme in mammalian cells, including myocardiocytes ([14] and references therein). In both teleosts, iNOS is exclusively localised within the cytoplasm of the myocardiocytes, being absent in the EE, in the epicardium, and in the VE of the subepicardial vessels (T. bernacchii). This localization agrees with iNOS location in the ventricular myocardiocytes of C. hamatus [6]. The Wndings that in mammalian cardiac and skeletal muscle cells, iNOS appears to be associated with particulate intracellular components such as mitochondria and contractile Wbres [42,14], has suggested that the iNOSproduced NO may control both cell respiration and contractile function of the cardiac muscle [43]. NO produced in proximity of the mitochondria reversibly inhibits cytochrome c oxidase [16]. Using computer modeling, it has been suggested that coupling of NO/O2 diVusion substantially extends the zone of adequate cellular oxygenation away from the blood channels [44]. Therefore, we believe that the maintenance of NOS expression in the two Notothenioids, rather than being a vestigial character, could indicate selective retention of the molecule for aiding intracellular oxygen transport in the hearts of these organisms. Notably, the sarcoplasmic localization of iNOS in the myocardiocytes of the two teleosts may partially reXect its association with intracellular membranes such as those of the Golgi apparatus, the nuclear envelope, and the endoplasmic reticulum. If this is the case, immunodetection may have located the enzyme during the process of synthesis and maturation. The iNOS constitutive expression observed in the two teleosts does not prevent the enzyme to be sensitive to stimulatory conditions. In fact, when C. hamatus and T. bernacchii were treated with LPS, a signiWcant increase of NOS activity was observed in the ventricular homogenates which resulted particularly high in T. bernacchii (Fig. 4). As

D. Amelio et al. / Nitric Oxide 15 (2006) 190–198

expected, immunoblotting revealed that in both teleosts this increment is attributable to the enhanced expression of iNOS (Fig. 5B1). Also in this case, comparison between the two Wshes indicates a higher increment of the enzyme in the ventricular extracts of LPS-treated T. bernacchii in which iNOS levels enhance up to 22.7% with respect to untreated Wshes. It is well known in mammalian cells that exposure to LPS causes an increased iNOS expression with a resultant NO overproduction which may lead to depressed myocardial contractility [45]. Unexpectedly, our analyses revealed that LPS stimulation also induces a signiWcant hyperexpression of eNOS which reaches 36.8% in the ventricular extracts of T. bernacchii (Fig. 5A1). Very recent data on mammals have highlighted a novel pro-inXammatory role for eNOS showing, for example, that the LPS-dependent septic condition is characterised by an initial eNOS activation, with the resultant NO acting as a co-stimulus for the iNOS expression [46]. Our results do not indicate whether this cascade system is present and/or functioning, nor whether eNOS and/or iNOS induction may be beneWcial or detrimental for the cardiac functionality of the two Antarctic teleosts. However, the shift from inducible to constitutive iNOS gene expression did not aVect the sensitivity of the iNOS system to septic challenges (i.e., LPS), indicating the maintenance of the general defence behaviour. An interesting point is the lower eNOS and iNOS inducibility shown by C. hamatus with respect to T. bernacchii. We hypothesise that this may be linked to the absence of the NO scavenger Hb. In fact, in LPS-stimulated T. bernacchii, NO overproduction may be buVered by Hb, in contrast to the iceWsh in which the control of the potentially deleterious eVects of NO can be achieved by limiting synthases inducibility. In summary, the present study has demonstrated the basal expression of eNOS and iNOS in the ventricle of the Hb-free C. hamatus and its red-blooded counterpart, T. bernacchii. Comparison between the two Wshes showed that they share the same heart ventricle distribution of the two isoforms. The Wnding that both NOSs are susceptible of stimulation by LPS indicates that neither the evolutionary geographic isolation nor the cold-elicited disaptations (i.e., constitutive expression of HSPs as well as presence or absence of the Hb-dependent NO scavenging) have aVected the ventricular NOS patterns in these stenotherm teleosts. Accordingly, the NOS system even more appears as an ubiquitous and crucial mechanism which controls vertebrate cardiac function. Acknowledgment The work was supported by PNRA (Programma Nazionale di Ricerche in Antartide). References [1] J.T. Eastman, Antarctic Notothenioid Wshes as subjects for research in evolutionary biology, Antarct. Sci. 12 (2000) 276–287.

197

[2] B.A. Buckley, S.P. Place, G.E. Hofmann, Regulation of heat shock genes in isolated hepatocytes from an Antarctic Wsh, Trematomus bernacchii, J. Exp. Biol. 207 (2004) 3649–3656. [3] E. Cocca, M. Ratnayakelecamwasam, S.K. Parker, L. Camardella, M. Ciaramella, G. Di Prisco, H.W. Detrich, Do the hemoglobinless iceWshes have globin genes? Comp. Biochem. Physiol. 118 (1997) 1027–1030. [4] B.D. Sidell, M.E. Vayda, D.J. Small, T.J. Moylan, R.L. Londraville, M.L. Yuan, K.J. Rodnick, Z.A. Eppley, L. Costello, Variable expression of myoglobin among the hemoglobinless Antarctic iceWshes, Proc. Natl. Acad. Sci. USA 94 (1997) 3420–3424. [5] T.J. Moylan, B.D. Sidell, Concentrations of myoglobin and myoglobin mRNA in heart ventricles from Antarctic Wshes, J. Exp. Biol. 203 (2000) 1277–1286. [6] D. Pellegrino, C.A. Palmerini, B. Tota, No haemoglobin but NO: the iceWsh (Chionodraco hamatus) heart as a paradigm, J. Exp. Biol. 207 (2004) 3855–3864. [7] B. Tota, A. Gattuso, Heart ventricle pumps in teleosts and elasmobranchs. A morphodynamic approach, J. Exp. Zool. 275 (1996) 162– 171. [8] I.A. Johnston, P. Harrison, Morphometrics and ultrastructure of myocardial tissue in Notothenioid Wsh, Fish Physiol. Biochem. 3 (1987) 1–6. [9] B. Tota, R. Acierno, C. Agnisola, Mechanical performance of the isolated and perfused heart of the haemoglobinless Antarctic iceWsh Chionodraco hamatus (Lönnberg): eVects of loading conditions and temperature, Phil. Trans. R. Soc. Lond. B 332 (1991) 191–198. [10] K.M. O’Brien, B.D. Sidell, The interplay among cardiac ultrastructure, metabolism and the expression of oxygen-binding proteins in Antarctic Wshes, J. Exp. Biol. 203 (2000) 1287–1297. [11] L.A. Barouch, R.W. Harrison, M.W. Skaf, G.O. Rosas, T.P. Cappola, Z.A. Kobeissi, I.A. Hobai, C.A. Lemmon, A.L. Burnett, B. O’Rourke, E.R. Rodriguez, P.L. Huang, J.A. Lima, D.E. Berkowitz, J.M. Hare, Nitric oxide regulates the heart by spatial conWnement of nitric oxide synthase isoforms, Nature (Lond.) 416 (2002) 337–339. [12] P.B. Massion, J.L. Balligand, Modulation of cardiac contraction, relaxation and rate by the endothelial nitric oxide synthase (eNOS): lessons from genetically modiWed mice, J. Physiol. 546 (2003) 63–75. [13] P.A.R. Juliet, T. Hayashi, A. Iguchi, L.J. Ignarro, Concomitant production of nitric oxide and superoxide in human macrophages, Biochem. Biophys. Res. Commun. 310 (2003) 367–370. [14] I.B. Buchwalow, W. Schulze, P. Karczewski, M.M. Kostic, G. Wallukat, R. Morwinski, E.G. Krause, J. Muller, M. Paul, J. Slezak, F.C. Luft, H. Haller, Inducible nitric oxide synthase in the myocard, Mol. Cell Biochem. 217 (2001) 73–82. [15] R.I. Cohen, A.M. Hassell, X. Ye, K. Marzouk, S.F. Liu, Lipopolysaccharide down-regulates inducible nitric oxide synthase expression in swine heart in vivo, Biochem. Biophys. Res. Commun. 307 (2003) 451–458. [16] C. Giulivi, Characterization and function of mitochondrial nitricoxide, Free Radic. Biol. Med. 34 (2003) 397–408. [17] A.J. Kanai, L.L. Pearce, P.R. Clemens, L.A. Birder, M.M. Van-Bibber, S.Y. Choi, W.C. De Groat, J. Peterson, IdentiWcation of a neuronal nitric oxide synthase in isolated cardiac mitochondria using electrochemical detection, Proc. Natl. Acad. Sci. USA 98 (2001) 14126–14131. [18] D.S. Bredt, Nitric oxide signaling speciWcity—the heart of the problem, J. Cell Sci. 116 (2003) 9–15. [19] O. Feron, L. Belhassen, L. Kobzik, T.W. Smith, R.A. Kelly, T. Michel, Endothelial nitric oxide synthase targeting to caveolae. SpeciWc interactions with caveolin isoforms in cardiac myocytes and endothelial cells, J. Biol. Chem. 271 (1996) 22810–22814. [20] M.S. Joshi, T.B. Ferguson Jr., T.H. Han, D.R. Hyduke, J.C. Liao, T. Rassaf, N. Bryan, M. Feelisch, J.R. Lancaster Jr., Nitric oxide is consumed, rather than conserved, by reaction with oxyhemoglobin under physiological conditions, Proc. Natl. Acad. Sci. 99 (2002) 10341– 10346. [21] A.J. Gow, B.P. Luchsinger, J.R. Pawloski, D.J. Singel, J.S. Stamler, The oxyhemoglobin reaction of nitric oxide, Proc. Natl. Acad. Sci. 96 (1999) 9027–9032.

198

D. Amelio et al. / Nitric Oxide 15 (2006) 190–198

[22] S.S. Gross, P. Lane, Physiological reactions of nitric oxide and hemoglobin: a radical rethink, Proc. Natl. Acad. Sci. 96 (1999) 9967–9969. [23] M.T. Gladwin, J.H. Crawford, R.P. Patel, The biochemistry of nitric oxide, nitrite, and hemoglobin: role in blood Xow regulation, Free Radic. Biol. Med. 36 (2004) 707–717. [24] P.L. Gabbott, S.J. Bacon, Histochemical localization of NADPHdependent diaphorase (nitric oxide synthase) activity in vascular endothelial cells in the rat brain, Neuroscience 57 (1993) 79–95. [25] J.B. Barroso, A. Carreras, F.J. Esteban, M.A. Peinado, E. MartinezLara, R. Valderrama, A. Jimenez, J. Rodrigo, J.A. Lupianez, Molecular and kinetic characterization and cell type location of inducible nitric oxide synthase in Wsh, Am. J. Physiol. Regul. Integr. Comp. Physiol. 279 (2000) 650–656. [26] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248–254. [27] M. Virgili, A. Poli, A. Beraudi, A. Giuliani, L. Villani, Regional distribution of nitric oxide synthase and NADPH–diaphorase activities in the central nervous system of teleosts, Brain Res. 901 (2001) 202–207. [28] J.S. Pollock, U. Forstermann, W.R. Tracey, M. Nakane, Nitric oxide synthase isozymes antibodies, Histochemistry 27 (1995) 738–744. [29] P.C. Ursell, M. Mayes, Anatomic distribution of nitric oxide synthase in the heart, Int. J. Cardiol. 50 (1995) 217–223. [30] P.C. Ursell, M. Mayes, Endothelial isoform of nitric oxide synthase in rat heart increases during development, Anat. Rec. 246 (1996) 465–472. [31] M. de los Angeles Costa, R. Elesgaray, A. Loria, A.M. Balaszczuk, C. Arranz, Atrial natriuretic peptide inXuence on nitric oxide system in kidney and heart, Regul. Pept. 118 (2004) 151–157. [32] A.M. Oyan, F. Nilsen, A. Goksoyr, B. Holmqvist, Partial cloning of constitutive and inducible nitric oxide synthases and detailed neuronal expression of NOS mRNA in the cerebellum and optic tectum of adult Atlantic salmon (Salmo salar), Brain Res. Mol. Brain Res. 78 (2000) 38–49. [33] L.J. Andries, D.L. Brutsaert, S.U. Sys, Nonuniformity of endothelial constitutive nitric oxide synthase distribution in cardiac endothelium, Circ. Res. 82 (1998) 195–203. [34] A.W. Cohen, R. Hnasko, W. Schubert, M.P. Lisanti, Role of caveolae and caveolins in health and disease, Physiol. Rev. 84 (2004) 1341– 1379. [35] M. Uematsu, Y. Ohara, J.P. Navas, K. Nishida, T.J. Murphy, R.W. Alexander, R.M. Nerem, D.G. Harrison, Regulation of endothelial cell nitric oxide synthase mRNA expression by shear stress, Am. J. Physiol. 269 (1995) 1371–1378. [36] A.J. North, K.S. Lau, T.S. Brannon, L.C. Wu, L.B. Wells, Z. German, P.W. Shaul, Oxygen upregulates nitric oxide synthase gene expression in ovine fetal pulmonary artery endothelial cells, Am. J. Physiol. 270 (1996) 643–649.

[37] J.P. Gratton, J. Fontana, D.S. O’Connor, G. Garcia-Cardena, T.J. McCabe, W.C. Sessa, Reconstitution of an endothelial nitric-oxide synthase (eNOS), hsp90, and caveolin-1 complex in vitro. Evidence that hsp90 facilitates calmodulin stimulated displacement of eNOS from caveolin-1, J. Biol. Chem. 275 (2000) 22268–22272. [38] S.P. Place, M.L. Zippay, G.E. Hofmann, Constitutive roles for inducible genes: evidence for the alteration in expression of the inducible hsp70 gene in Antarctic Notothenioid Wshes, Am. J. Physiol. Regul. Integr. Comp. Physiol. 287 (2004) 429–436. [39] A. Mebazaa, R.C. Wetzel, J.M. Dodd-o, E.M. Redmond, A.M. Shah, K. Maeda, G. Maistre, E.G. Lakatta, J.L. Robotham, Potential paracrine role of the pericardium in the regulation of cardiac function, Cardiovasc. Res. 40 (1998) 332–342. [40] M.C. Cerra, M. Canonaco, R. Acierno, B. Tota, DiVerent binding activity of A- and B-type natriuretic hormones in the heart of two Antarctic teleosts, the red blooded Trematomus bernacchii and the haemoglobinless Chionodraco hamatus, Comp. Biochem. Physiol. 118 (1997) 993–999. [41] J.M. Icardo, E. Colvee, M.C. Cerra, B. Tota, Bulbus arteriosus of the Antarctic teleosts II the red-blooded Trematus Bernacchii, Anat. Rec. 256 (1999) 116–126. [42] I. Gath, E.I. Closs, U. Godtel-Armbrust, S. Schmitt, M. Nakane, I. Wessler, U. Forstermann, Inducible NO synthase II and neuronal NO synthase I are constitutively expressed in diVerent structures of guinea pig skeletal muscle: implications for contractile function, FASEB J. 10 (1996) 1614–1620. [43] T. Tatsumi, S. Matoba, A. Kawahara, N. Keira, J. Shiraishi, K. Akashi, M. Kobara, T. Tanaka, M. Katamura, C. Nakagawa, B. Ohta, T. Shirayama, K. Takeda, J. Asayama, H. Fliss, M. Nakagawa, Cytokine-induced nitric oxide production inhibits mitochondrial energy production and impairs contractile function in rat cardiac myocytes, J. Am. Coll. Cardiol. 35 (2000) 1338–1346. [44] D.D. Thomas, X. Liu, S.P. Kantrow, J.R. Lancaster Jr., The biological lifetime of nitric oxide: implications for the perivascular dynamics of NO and O2, Proc. Natl. Acad. Sci. USA 98 (2001) 355–360. [45] M.T. Ziolo, H. Katoh, D.M. Bers, Expression of inducible nitric oxide synthase depresses beta-adrenergic-stimulated calcium release from the sarcoplasmic reticulum in intact ventricular myocytes, Circulation 104 (2001) 2961–2966. [46] L. Connelly, M. Madhani, A.J. Hobbs, Resistance to endotoxic shock in endothelial nitric-oxide synthase (eNOS) knock-out mice: a proinXammatory role for eNOS-derived no in vivo, J. Biol. Chem. 280 (2005) 10040–10046. [47] Z. Lacza, E. Pankotai, A. Csordas, D. Gero, L. Kiss, E.M. Horvath, M. Kollai, D.W. Busija, C. Szabo, Mitochondrial NO and reactive nitrogen species production: Does mtNOS exist? Nitric Oxide (2005) [Epub ahead of print].