Effects of salt loading on the fractional volume of atria-specific granules in dahl salt-sensitive and salt-resistant rats.код для вставкиСкачать
THE ANATOMICAL RECORD 218:157-161(1987) Effects of Salt Loading on the Fractional Volume of Atria-Specific Granules in Dahl Salt-Sensitive and Salt-Resistant Rats JOHN T. HANSEN, J.R. HAYWOOD, AND ANDREW HOWELL Department of Neurobiology and Anatomy, University of Rochester School of Medicine and Dentistry, Rochester, NY 14642 (2. TH., A.H.); Department of Pharmacology, The University of Texas Health Science Center, S a n Antonio, T X 78284 (J.R.H.) ABSTRACT The cardiac atria are known to play a role in blood volume homeostasis, secreting a peptide that induces a potent natriuresis and diuresis. This peptide is atrial natriuretic factor (ANF), and its primary site of storage is within atria-specific granules found in atrial cardiocytes. Since salt loading results in a n increase in circulating levels of ANF, our aim was to determine if the atria-specific granule population in the cardiocytes of Dahl rats would decrease accordingly. To this end, the fractional volume of the atria-specific granules was determined by ultrastructural morphometric analysis in the Dahl salt model of hypertension. This analysis was performed on the right atria of Dahl saltresistant (DR) and salt-sensitive (DS) rats fed either a low-salt (0.4%) or high-salt (8%)diet for 12 weeks prior to sacrifice. DR and DS rats fed a low-salt diet had significantly reduced plasma sodium levels and osmolalities, and a significantly lower mean arterial blood pressure than did rats fed a high-salt diet. The fractional volume of atria-specific granules was significantly lower in salt-loaded DR (P< 0.01) and DS (P<O.O25) rats than in their respective low-salt controls. This significant decrease in atrial granules corresponds to the reported decrease in the storage of atrial ANF in salt-loaded rats, and provides a morphological verification of the biochemical studies. Moreover, these results, in combination with a growing body of physiological data, lend support to the hypothesized role of ANF in the regulation of water-electrolyte balance, which may play a n important role in cardiovascular pathophysiological states related to hypertension. Mammalian atrial cardiocytes contain secretory granules characteristic of endocrine cells. These atria-specific granules were first discovered by Kisch (1956) in guinea pig heart, and later in other mammalian species (Bompiani et al., 1959; Jamieson and Palade, 1964; Hibbs and Ferrans, 19691, including humans (Battig and Low, 1961). The granules possess a homogenous, electrondense core, are surrounded by a limiting membrane, and measure 250-500 nm in diameter (Jamieson and Palade, 1964; Hibbs and Ferrans, 1969). Often, the granules are more concentrated in the central sarcoplasmic core, associated with a n extensive Golgi complex located at each nuclear pole. Rough endoplasmic reticulum, glycogen, and numerous mitochondria commonly are associated with the granule accumulations. Atria-specific granules incorporate 3H-leucine (Yunge et al., 1980) with kinetics similar to that found in endocrine cells that produce polypeptide hormones. Moreover, the cardiac atria are known to play a role in blood volume homeostasis, and the number of atria-specific granules is altered by experimental paradigms that affect water and electrolyte balance (Marie et al., 1976; de Bold, 1979). On the basis of these and other histochemical studies (see de Bold, 1985,1986), it has been hypoth0 1987 ALAN R. LISS, INC esized that the atria-specific granules contain basic polypeptides involved in water-electrolyte balance, and that mammalian atrial cardiocytes synthesize, store, and release these substances in a manner common to other polypeptide hormone-producing cells (de Bold, 1986).Recent immunocytochemical (Cantin et al., 1984; Chapeau et al., 1985; Maldonado et al., 1986) and biochemical studies have established that the atria-specific granules store a group of peptides with molecular weights ranging from about 2,500 to 13,000 that are commonly referred to a s atrial natriuretic factor (ANF) (Seidah et al., 1984; de Bold, 1986). Both large-molecule and smallmolecule forms of atrial natriuretic peptides exist, but the most abundant species in atrial extracts is a 126amino acid peptide called cardionatrin IV,or y-atrial natriuretic peptide (de Bold, 1986). The most common low-molecular-weight natriuretic peptide isolated from the rat atria is the 28-amino acid peptide called cardion- Received October 14, 1986; accepted January 12, 1987. Address correspondence to: John T. Hansen, Ph.D., Department of Neurobiology and Anatomy, Box 603, University of Rochester School of Medicine, 601 Elmwood Avenue, Rochester, NY 14642. 158 J.T. HANSEN, J.R. HAYWOOD, AND A. HOWELL 159 ATRIAL SPECIFIC GRANULES IN DAHL RATS TABLE 1. Comparative data from Dahl rats fed low- or high-salt diets Group Mean arterial pressure Plasma Na Plasma K Osmolality (mm Hg) (mEqlliter) (mEq/liter) (mOsm) DR-low DR-high DS-low ____ DS-high 112.6 k 3.3 124.9 f 2.01 137.6 f 1.7 179.0 f 13.5l 135.0 f 1.8 141.9 f 0.8’ 137.0 f 1.6 144.6 f 0.8’ 6.0 4.8 6.0 4.6 f 0.3 f 0.l2 & 0.2 f 0.l2 276 290 285 294 k2 k l2 &3 k Z3 Fractional volume ANF granules (%I 1.74 k 0.17 1.08 f 0.17’ 1.78 f 0.18 1.18 f 0.14’ All values are means f SEM (N = 4 animals per group). ‘P <: 0.025 compared to its respective low-salt group. ‘P < 0.01 compared to its respective low-salt group. 3P i0.05 compared to its respective low-salt group. atrin I (Flynn et al., 1983).Extracts of mammalian atria containing ANF induce a potent natriuresis and diuresis in the intact rat (de Bold et al., 1981), which is accompanied by hypotension, often bradycardia, and a reduction in aldosterone and renin secretion. Additionally, ANF may be a vasoactive substance that counteracts the effects of endogenous vasoconstrictors, such as norepinephrine and angiotensin 11, thereby leading to the observed natriuresis (Kleinert et al., 1984). Previous studies on ANF have been directed at elucidating the structure and normal physiological properties of this hormone. Recently, attention has been redirected toward understanding the role of ANF in cardiovascular pathophysiological states related to hypertension. Of special interest is the relationship of ANF to volume-overload hypertension. Hirata and co-workers (1984) have investigated the role of ANF in the Dahl salt model of hypertension, where the hypertension is hypothesized to be due in part to a humoral factor that affects sodium excretion, blood pressure, and vascular reactivity. The present study was undertaken to determine if volume-overload hypertension in the Dahl salt model affected the atrial cardiocyte granule population, a s it is reported to do in other experimental paradigms involving water-electrolyte balance (Marie et al., 1976; de Bold, 1979). divided into two groups of 4 rats each and fed a diet low in salt (0.4% NaC1; Teklad, Madison, WI) (this group referred to hereafter as DR-L) or a high-salt diet (8% NaC1; Teklad) (group DR-H). The DS rats likewise were divided into two groups of 4 rats each and fed the identical low-salt (DS-L)and high-salt diet (DS-H).All groups were fed their respective diets for 12 weeks; the animals on a low-salt diet received distilled water ad lib, and those on a high-salt diet received tapwater ad lib. At the end of the 12-week period, the rats were anesthetized with methoxyflurane gaseous anesthesia and prepared with femoral arterial and venous catheters. A venous blood sample was taken for plasma osmolality determinations using freezing-point depression, and sodium and potassium determinations were made by flame photometry. In addition, the mean arterial pressures were measured directly in conscious animals via the femoral artery catheter. Electron Microscopy Following the physiologicaI measurements, the rats were anesthetized and sacrificed by intracardiac perfusion of a fixative solution containing 3% glutaraldehyde, 1%paraformaldehyde in 0.1 M sodium phosphate buffer (pH 7.2, room temperature). The fixative was preceded by a saline flush. Following a 10-min perfusion, the heart was removed and immediately immersed in fresh MATERIALS AND METHODS fixative for a n additional 4-6 h r a t 4°C. The heart then Male Dahl salt-resistant (DR) rats and salt-sensitive was washed in 0.1 M phosphate buffer containing 5% (DS) rats were obtained from the Brookhaven National sucrose (pH 7.2). Prior to postfixation, the auricular porLaboratories, Upton, New York. The DR animals were tion of the right atrium was removed from each heart, and sectioned with a razor blade into small pieces (2 mm x 2 mm). Then the samples were postfixed in 1%osmium tetroxide, 1.5% potassium ferrocyanide in buffer at 4°C for 16 hr. Following a maleate buffer wash, the tissues were stained en bloc in 0.5% uranyl acetate in 0.1 M maleate buffer (pH 6) for 3 hr. The samples then Fig. 1. Electron micrograph of atrial cardiocyte showing atria-spewere rinsed in maleate buffer, dehydrated in a graded cific granules (ag) distributed preferentially in a juxtanuclear location. Golgi complexes and mitochondria (m) are common features in these series of acetones, infiltrated, and embedded in Spurr’s cardiocytes. This cardiocyte is sectioned longitudinally and clearly resin. demonstrates the preferential distribution of granules near the center of the cell. x 15,000. Fig. 2. Electron micrograph that shows the accumulation of atriaspecific granules (ag) ad,jacent to a Golgi complex. Note the variable size and density of the granules. m, mitochondrion. ~ 2 6 , 2 5 0 . Fig. 3. Electron micrograph showing granule profiles that appear to be “budding” off (arrows) from the cisternae of a Golgi complex. m, mitochondrion. ~76,000. Morphometric Analysis Sections 1 pm thick were cut with glass knives and stained with toluidine blue, and sections from blocks with atrial cardiocytes in a cross-sectional orientation were selected for thin sectioning. Thin sections of crosssectional cardiocytes were chosen because of the anisotropic nature of atrial granule distribution when viewed 160 J.T.HANSEN, J.R. HAYWOOD, AND A. HOWELL in longitudinal sections (see Fig. 11, i.e., atria-specific granules resided primarily in a juxtanuclear position. While sampling longitudinally oriented cardiocytes that contain a nucleus would increase the granule counts, such a n analysis would not be random. since only atrial granules in the approximate “center” of the cardiocyte would be sampled (Cantin et al., 1979). Cross-sectional analysis, on the other hand, randomly samples all portions of the cardiocyte without preferentially focusing on the center of the cell. For each rat in each of the four groups, 5-10 electron micrographs were collected at a n initial magnification of ~ 4 , 0 0 0 The . micrographs were collected by systematic random sampling (Weibel, 1979)from the first tissue encountered over a n open grid square. The electron microscope was calibrated with a ruled diffi-action grating containing 2,160 lines per 1 mm. Each electron micrograph was photographically enlarged on 8-in. x 10-in. , yielded a paper to a magnification of ~ 1 4 , 0 0 0which total area of 234 pm of tissue per print. The fractional volume of atria-specific granules in the cardiocyte was determined by point counting, using a transparent overlay containing 837 points. All prints were coded so that the observer collecting the data was unaware of group identity. Repeat counts demonstrated a reliability of about 2%, which was deemed acceptable. The fractional volume occupied by the granules was expressed as the mean of all prints from each group (N = 20-30) plus or minus the standard error of the mean. Comparisons of all groups were performed using a two-way analysis of variance in a 2 x 2 factorial design, with strain (resistant or sensitive) and diet (low- or high-salt) as independent variables. Statistical comparisons of the DR-L and DR-H, and the DS-L and DS-H groups (diet a s the only variable) were performed by means of a Student’s t test. A probability of P<O.O5 was considered significant. RESULTS The mean arterial blood pressure, plasma sodium and potassium levels, and plasma osmolality of all groups of animals just prior to sacrifice are shown in Table 1.All values were significantly different between DR-L and DR-H groups, and between DS-L and DS-H groups. The mean arterial blood pressure between the DR-L and DSL groups was also significant (P< 0.002). Additionally, the mean arterial blood pressure was considerably elevated in DS-H rats (range: 143-218 mm Hg). Estimates of the fractional volume of the atria-specific granules in the cardiocytes is shown in Table 1. The fractional volume of granules was significantly elevated (P< 0.01) in DR-L rats compared to the DR-H group. The DS-L rats also exhibited a significant increase (P< 0.025) in cardiocyte granules compared to the DS-H group. Two-way ANOVA showed a significant effect of diet, but no significant effect of strain or straiddiet interaction on the fractional volume of atria-specific granules. Atria-specific granules often were distributed preferentially in a juxtanuclear location (Fig. l),although scattered granules also were observed throughout the cardiocytes and along the sarcolemma. Most atria-specific granules displayed a homogeneous core that was surrounded by a unit membrane (Figs. 1-3). The granules were especially abundant in the area around the Golgi complex (Fig. 2), and they often varied in size and in the density of their granule core. In some instances, profiles were observed that resembled granules “budding” off from the cisternae of the Golgi complex (Fig. 3). With the exception of the documented estimates of atria-specific granule distribution, no other differences in morphology were observed among any of the groups. DISCUSSION Immunocytochemical and radioimmunoassay techniques have shown that ANF is stored within atriaspecific granules (Cantin et al., 1984; de Bold, 1985). Water deprivation decreases the atrial content of messenger RNA for precursor ANF (Nakayama et al., 1984) and increases the cardiocyte content of atria-specific granules (de Bold, 1979). Therefore, one might hypothesize that during water deprivation a high atrial content of ANF is the result of a decrease in demand, a n increase in storage, and a decrease in de novo synthesis of ANF. Conversely, the number of atria-specific granules decreases after saline treatment (de Bold, 1979). Apparently, salt loading increases the demand for ANF, which results in a n increase in blood levels of ANF and a decrease in the atrial content. In this instance, one might hypothesize that salt loading reflects a state of rapid synthesis and release of ANF, and concomitant decrease in storage (Snajdar and Rapp, 1985).Therefore, salt loading has the opposite effects of water deprivation on ANF synthesis, storage, and release. If one assumes that the number of atria-specific granules in a cardiocyte is related to the content of ANF in that cardiocyte, our results in Dahl rats support these two hypotheses with respect to the atrial storage of ANF. Previous studies support the basic assumption that atria-specific granule numbers and the levels of ANF in the atria are related, suggesting that changes observed in atria-specific granule density under different experimental paradigms is a reflection of relative ANF content (Marie et al., 1976; de Bold, 1979). In the DR-H and DS-H rats, atria-specific granules are significantly reduced compared to their respective low-salt controls. Plasma sodium and potassium levels and osmolality of the salt-fed groups are normal, but the mean arterial blood pressure is significantly elevated. The low-sodium-fed groups not only had more atria-specific granules, but plasma sodium and osmolality also are significantly reduced. Therefore, the fractional volume of atria-specific granules in Dahl rats subjected to high- and to low-salt intake follows the expected and hypothesized pattern. Hirata et al. (1984) have demonstrated that DS rats have more ANF than DR rats. If atria-specific granule concentration and ANF content can be correlated, as suggested by others (see de Bold, 1985), it is interesting that our DS rats showed a tendency to possess a greater fractional volume of atria-specific granules than did DR rats. However, these differences are slight, not significant, and probably simply represent variability within our sample population. The analysis of variance shows no significant interaction of strain and diet. Only diet appears to have a significant effect on atria-specific granule fractional volume estimates. Our inability to demonstrate a correspondingly higher granule fractional volume in the DS strain than in the DR strain could be the result of several factors. First, with the ATRIAL SPECIFIC GRANULES IN DAHL RATS electron microscope we have quantitated only the fractional volume of atria-specific granules. Nothing can be concluded about the amount of ANF in each granule (compare Figures 2 and 3) (Marie et al., 1976; Cantin et al., 1979). Second, DS rats appear to need more ANF because their kidneys have about a 50% lower natriuretic response to ANF compared to DR rats (Hirata et al., 1984). Moreover, Snajdar and Rapp (1985) have shown that DS rats not only have kidneys hyporesponsive to the effects of ANF but also tend to release less ANF from their atria. Hence, DS rats usually show a n increased amount of ANF in their atria. Apparently, our morphometric approach is not sensitive enough to detect this strain difference. Alternatively, diet appears to play a significant role in the granularity of the atrial cardiocytes, and this effect may simply mask any strain differences that might otherwise appear. This possibility could be tested by simply comparing DR and DS strains without the complicating influence of a low- or high-salt diet. Finally, the absolute number of atria-specific granules per atrial cardiocyte may be larger in DS rats. This would be true if the atrial cardiocytes hypertrophied secondary to a n increased blood pressure. Based upon cross-sectional area, the atrial cardiocytes do not appear to be larger in DS rats, although their longitudinal size may be increased. In fact, the wet weights of the hearts in the DS-L rats is 24% greater than in the DR-L animals, and the DS-H hearts weigh 57% more than the DR-H hearts. How much of this increase in the wet weight of the heart is attributable to a n increase in the ventricle was not determined in this study. However, if DS rats do possess larger atrial cardiocytes, the fact that the relative fractional volume data are similar for both groups on the same diet reflects that the absolute amount of atria-specific granules may be increased in the DS strain. This study provides a morphological verification of the reported decrease of atrial ANF and plasma volume regulation in Dahl rats during high-salt intake. These results, in combinat,ion with a growing body of physiological data, lend further support to the hypothesized role of ANF in the regulation of sodium and water balance. Future discoveries that support such a role for ANF may provide the basis for new therapies for hypertension and congestive heart failure (de Bold, 1985; Ballermann and Brenner, 1986; Cantin and Genest, 1986). ACKNOWLEDGMENTS The authors would like to thank Ms. Nancy A. Ball for her excellent technical assistance. This research was supported by U S . Public Health Service grants HL 36038 (J.T.H.) and HL 33937 (J.R.H.). Dr. Hansen is the recipient of a National Institute of Health Research Career Development Award. 161 LITERATURE CITED Bailermann, B.J., and B.M. Brenner (1986) Role of atrial peptides in body fluid homeostasis. Circ. Res., 58519-630. Battig, C.G., and F.N. Low (1961)The ultrastructure of human cardiac muscle and its associated tissue space. Am. J. Anat., 108:199-229. Bompiani, C.D., C. Rouiller, and P. Hatt (1959) Le tissu de conduction du coeur chez le rat. Etude au microscope electronique. Arch. Mal. Coeur., 52:1257-1274. Cantin, M., and J. Genest (1986)The heart as an endocrine gland. Sci. Am., 254:76-81. Cantin, M., M. Timm-Kennedy, E. El-Khatib, M. Huet, and L. Yunge (1979) Ultrastructural cytochemistry of atrial muscle cells. Comparative study of specific granules in right and left atrium of various animal species. Anat. Rec., 193:55-70. Cantin, M., J. Gutkowska, G. Thibault, R.W. Milne, S. Ledoux, S. Min Li, C. Chapeau, R. Garcia, P. Hamet, and J. Genest (1984) Immunocytochemical localization of atrial natriuretic factor in the heart and salivary glands. Histochemistry, 80:113-127. Chapeau, C., J. Gutkowska, P.W. Schiller, R.W. Milne, G. Thibault, R. Garcia, J. Genest, and M. Cantin (1985) Localization of immunoreactive synthetic atrial natriuretic factor (ANF) in the heart of various animal species. J. Histochem. Cytochem., 33:541-550. de Bold, A.J. (1979)Heart atria granularity effects of changes in waterelectrolyte balance. Proc. SOC.Exp. Biol. Med., 161508-511. de Bold, A.J. (1985) Atrial natriuretic factor: A hormone produced by the heart. Science. 230367-770. de Bold, A.J. (1986) Atrial natriuretic factor: An overview. Federation Proc., 45:2081-2085. de Bold, A.J., H.B. Borenstein, A.T. Veress, and H. Sonnenberg (1981) A rapid and potent natriuretic response to intravenous injection of atrial myocardial extract in rats. Life Sci., 28.49-94. Flynn, T.G., M.L. de Bold, and A.J. de Bold (1983) The amino acid sequence of a n atrial peptide with potent diuretic and natriuretic properties. Biochem. Biophys. Res. Commun., 117r859-865. Hibbs, R.G., and V.J. Ferrans (1969)An ultrastructural and histochemical study of rat atrial myocardium. Am. J. Anat., 124:251-280. Hirata, Y., M. Ganguli, L. Tobian, and J. Iwai (1984) Dahl S rats have increased natriuretic factor in atria but are markedly hyporesponsive to it. Hypertension, 6(SuppZ. 1);1148-1155; Jamieson, J.D., and G.E. Palade (1964) Specific granules in atrial muscle cells. J. Cell Biol., 23:151-172. Kisch, B. (1956) Electron microscopy of the atrium of the heart. I. Guinea pig. Exp. Med. Surg., 14t99-112. Kleinert, H.D., T. Maack, S.A. Atlas, A. Januszewicz, J.E. Sealy, and J.H. Laragh (1984) Atrial natriuretic factor inhibits angiotensin-, norepinephrine-, and potassium-inducedvascular contractility. Hypertension, G(Suppl. 1):1143-1147. Maldonado, C.A., W. Saggau, and W.G. Forssmann (1986) Cardiodilatin-immunoreactivity in specific atrial granules of human heart revealed by the immunogold stain. Anat. Embryol., 173:295-298. Marie, J.-P., H. Guiilemot, and P.-Y. Hatt (1976) Le degr6 de granulation des cardiocytes auriculaires. Etude planimetrique au cours de differents apports d e a u et de sodium chez ie rat. Pathol. Biol. (Paris), 24t549-554. Seidah, N.G., C. Lazure, M. Chretien, G. Thibauit, R. Garcia, M. Cantin, J. Genest, S.F. Brady, R.F. Nutt, T.A. Lyle, W.J. Palevada, C.D. Colton, T.M. Ciccerone, and D.F. Veber (1984) Amino acid sequence of homologous rat atrial peptides: Natriuretic activity of native and synthetic forms. Proc. Natl. Acad. Sci. USA, 81:26402644. Snajdar, R.M., and J.P. Rapp (1985) Atrial natriuretic factor in Dahl rats. Atrial content and renal aortic responses. Hypertension, 7t775-782. Weibel, E.R. (1979) Stereological Methods, Vol. 1: Practical Methods for Biological Morphometry. Academic, New York. Yunge, L., S. Benchimol, and M. Cantin (1980) Ultrastructural cytochemistry of atrial muscle cells. VIII. Radioautographic study of synthesis and migration of proteins. Cell Tissue Res., 207:l-11.