Immunoradiometric assay for the N-terminal fragment of proatrial natriuretic peptidein human plasma

¹ Diagnostic Science Division, Shionogi & Company, Ltd., 2-5-1 Mishima, Settsu-shi, Osaka 566, Japan.

² Division of Cardiology, Kumamoto University of Medicine School, Kumamoto 860, Japan.

³ Department of Medicine and Clinical Science, Kyoto University Graduate School of Medicine, Kyoto 606, Japan.
^a Author for correspondence. Fax 06-319-4109; e-mail yoshito.numata{at}shionogi.co.jp.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Recently, the N-terminal fragment of proatrial natriuretic peptide (N-terminal proANP) has been proposed as a marker of chronic congestive heart failure. In this study, we established a two-step immunoradiometric assay using monoclonal antibodies and synthetic N-terminal proANP (1–67) as a standard. It allows us to measure plasma N-terminal proANP in only 4 h without prior extraction. The detection limit of this assay was 15 pmol/L for a 100 µL sample of plasma. Within-run CVs ranged from 1.7% to 2.9% and between-run CVs ranged from 4.2% to 5.1%. The dilution curves of plasma samples showed good linearity and analytical recovery was 89–104%. The mean (±SD) N-terminal proANP in plasma of 33 healthy subjects was 188 (±71) pmol/L and 1030 (±411) pmol/L in 25 patients with heart failure. Our immunoradiometric assay is rapid and precise enough for routine determination of N-terminal proANP in human plasma.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Human atrial natriuretic peptide (ANP) is synthesized in atrial myocytes and stored in secretory granules in the form of a 126-amino acid prohormone designated as proANP (1–126) (1)(2)(3). On release, the prohormone is split into equimolar amounts of the biologically active moiety ANP, or proANP (99–126) and the remaining part, called N-terminal proANP, or proANP (1–98) (4)(5)(6). The major known stimulus for secretion of these peptides is increased atrial wall stress, usually described as increased atrial stretch (7).

In patients with chronic congestive heart failure, increased plasma concentrations of both ANP and N-terminal proANP have been observed in proportion to the severity of the disease (6)(7)(8). However, because of the low ANP concentration in plasma, it takes 2 days for measurement by a commercially available immunoradiometric assay (IRMA) (9), which shows better sensitivity, precision, and accuracy than the RIAs, without prior extraction. Furthermore, a well-standardized protocol is required for the routine determination of plasma ANP by immunoassay because sample collection and storage can greatly affect the measurement (10). Blood samples should be transferred into chilled tubes containing EDTA and aprotinin and centrifuged immediately. The plasma should be separated and stored at -20 °C as soon as possible. Plasma samples with evident hemolysis must be discarded.

In the case of N-terminal proANP, all of the immunoassays reported previously were competitive RIAs (4)(5)(6)(7)(8)(9)(11), which involve incubation of the samples for >20 h. N-Terminal proANP has a substantially longer half-life in blood compared with ANP and is present in concentrations up to 50 times higher than the plasma concentration of ANP (12). In addition, N-terminal proANP is more stable under laboratory conditions for measurement (13)(14). These findings led us to develop a sensitive method for easier and less time-consuming N-terminal proANP measurement. Our idea was to prepare monoclonal antibodies recognizing distinct epitopes of N-terminal proANP and to use them in a sandwich immunoassay. We were successful in developing a sensitive IRMA for N-terminal proANP in plasma; here, we report on its performance.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
peptides
N-Terminal proANP (1–25), proANP (43–67), proANP (43–66)-Cys, proANP (1–67), ANP, and brain natriuretic peptide were purchased from the Peptide Institute. The purity of N-terminal proANP (1–67) as a standard was 98% as determined by HPLC analysis.

antibodies
The monoclonal antibody against N-terminal proANP (1–25), KY-ANP-lll (IgG1, ) was prepared as described previously (15). For development of a sandwich immunoassay, we established a new monoclonal antibody, 7B6 (IgG1, ), by immunization of Balb/c mice with N-terminal proANP (43–66)-Cys-bovine thyroglobulin conjugate. KY-ANP-lll and 7B6 recognize the N-terminal region and middle region of N-terminal proANP (1–98), respectively. We developed the IRMA using KY-ANP-lll as the immobilized antibody and 7B6 as the labeled antibody.

ky-anp-lll-coated polystyrene beads
Polystyrene beads (6.5 mm diameter; Immuno Chemical) were incubated in 50 mmol/L phosphate buffer, pH 7.8, (buffer A) containing 25 mg/L of KY-ANP-lll, for 3 h at 25 °C and overnight at 4 °C. The antibody solution was then removed from the beads, which were washed four times with buffer A. The beads were coated with buffer A containing 250 mL/L Block Ace (Dainippon Pharmaceutical) and incubated overnight at 4 °C. After the incubation buffer was aspirated, the beads were stored in buffer B (100 mmol/L phosphate buffer, pH 6.5, containing 1 g/L bovine serum albumin, 150 mmol/L NaCl, 1 mmol/L EDTA, and 1 g/L NaN₃).

iodination of 7b6 and n-terminal proANP (1–25)
Monoclonal antibody 7B6 and N-terminal proANP (1–25) were radiolabeled with I (Amersham) by the chloramine T method (16). I-7B6 and I-N-terminal proANP (1–25) were purified on a Superose 12 HPLC column (Pharmacia) and C₁₈ µBondapak column (Waters), respectively.

irma for n-terminal proANP
Standard solutions of N-terminal proANP (1–67) in concentrations from 0 to 6000 pmol/L were prepared with buffer B. In the typical assay procedure, a standard or plasma sample (100 µL each) and buffer B (200 µL) were incubated with KY-ANP-lll-coated polystyrene beads for 2 h at 37 °C. After removal of the incubation mixture, the beads were washed three times with 2 mL of 50 mmol/L phosphate buffer (pH 6.5) containing 75 µL/L Tween 20, 1 mmol/L EDTA, and 90 mg/L NaN₃. The beads were then incubated with I-labeled 7B6 (~200 000 cpm in 300 µL of 50 mmol/L phosphate buffer, pH 6.5, containing 1 g/L bovine serum albumin, 75 µL/L Tween 20, 1 mmol/L EDTA, and 1 g/L NaN₃) for 2 h at 37 °C. After removal of the incubation mixture, the beads were washed as described above and then the radioactivities were measured with a gamma counter, ARC-600 (Aloka). Experiments were performed in duplicate except where noted otherwise. N-terminal proANP concentrations were expressed as N-terminal proANP (1–67)-like immunoreactivities.

competitive ria for n-terminal proANP
Standard solutions of N-terminal proANP (1–67) in concentrations from 0 to 6000 pmol/L were prepared with buffer C (50 mmol/L phosphate buffer, pH 7.0, containing 1 g/L bovine serum albumin, 150 mmol/L NaCl, 1 mmol/L EDTA, and 1 g/L NaN₃). In the assay procedure, aliquots of standard or plasma sample (100 µL each) and buffer C (200 µL) were preincubated with 100 µL of KY-ANP-lll (5 ng) for 20 h at 4 °C. I-N-terminal proANP (1–25) solution (~45 000 cpm in 100 µL of buffer C) was then added, and the mixture was incubated for an additional 20 h at 4 °C. Separation of free from bound antigen was achieved by precipitation with 25 µL of 100 mL/L normal mouse serum and 1 mL of goat anti-mouse IgG solution (230 µg in 50 mmol/L phosphate buffer, pH 7.4, containing 84 g/L polyethylene glycol 6000, 150 mmol/L NaCl, and 0.2 g/L NaN₃), followed by centrifugation at 1600g for 20 min at 4 °C. The supernatants were aspirated, and the radioactivities in the pellets were counted in a gamma counter.

irma for anp
A commercial kit, Shiono RIA ANP (Shionogi & Co.) based on IRMA (9), was used.

gel filtration chromatography
Pooled plasma (from five subjects) was loaded on a Sephadex G-50 superfine column (Pharmacia, 16 x 980 mm) that had been equilibrated with buffer B and eluted with the same solution. Fractions of 2.85 mL were collected, and aliquots of each were assayed with IRMA and RIA. The column was calibrated with Blue Dextran, ribonuclease A, synthetic N-terminal proANP (1–67), and N-terminal proANP (1–25).

plasma samples
Blood samples were drawn into plastic syringes and quickly transferred to chilled tubes containing EDTA (1.5 g/L, blood) and centrifuged at 1600g at 4 °C for 20 min. The plasma samples thus obtained were kept frozen below -20 °C until determination. When both ANP and N-terminal proANP concentrations were measured, EDTA plasma samples containing aprotinin (50 000 kIU/L, blood) were used (10). Human sample acquisition was conducted in accordance with the policies and procedures of the Institutional Review Board for the use of human subjects in research at Diagnostic Science Division, Shionogi & Co., Ltd.

stability test in blood
Blood samples collected from three volunteers were quickly transferred into chilled tubes containing EDTA. These blood samples were transported on ice within 30 min and kept at 4 °C or 25 °C for 0, 3, 6, or 24 h. Aliquots (1 mL) were withdrawn and centrifuged at 1600g at 4 °C for 20 min. The plasma samples thus obtained were kept frozen below -20 °C until determination. The IRMA for N-terminal proANP was carried out as described above.

stability test in edta plasma
Blood samples collected from three volunteers were quickly transferred into chilled tubes containing EDTA. These blood samples were transported on ice within 30 min and centrifuged at 1600g at 4 °C for 20 min. The plasma samples separated were kept at 4 or -20 °C for 0, 1, 2, 3, 7, 14, 21, or 28 days, then assayed for N-terminal proANP as described above.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
analytical evaluation
Calibration curve and sensitivity.
A calibration curve was prepared for the standard assay system, using synthetic N-terminal proANP (1–67) as a standard. A representative standard curve is depicted in Fig. 1 . The lower detection limit of this IRMA, defined as the concentration at the mean 3SD of 20 determinations of the zero calibration, was 15 pmol/L.

View larger version (18K): [in this window] [in a new window]   Figure 1. Standard curve of the IRMA.

Precision.
The reproducibility of our present IRMA was estimated using clinically available plasma having different N-terminal proANP concentrations. The coefficients of variation within and between series were 1.7–2.9% (n = 5) and 4.2–5.1% (n = 10), respectively, as shown in Table 1 .

Dilution and recovery tests.
Dilution curves of three plasma samples gave good linearity (Fig. 2 ). Recoveries of exogenous added N-terminal proANP (1–67) from plasma samples containing three different concentrations of endogenous N-terminal proANP were estimated (Table 2 ). The recovery ranged from 89% to 104%.

View larger version (25K): [in this window] [in a new window]   Figure 2. Plasma dilution curves. Three plasma samples were serially diluted with buffer B.

Interferences and cross-reactivities.
Assessment of the influence of hemoglobin (<4.5 g/L) and bilirubin (0.15 g/L) showed no interference in this assay. We also examined the cross-reactivities with other natriuretic peptides. The cross-reactivities with N-terminal proANP (1–25) and (43–67) were <0.2%. In addition, this system did not react with 2 µg/L of human ANP and brain natriuretic peptide.

Method comparison.
The correlation between the values obtained by the newly developed IRMA method (Y) and those by the competitive RIA method (X) was given by the linear regression equation, Y = 1.35X - 340 (pmol/L), for which the correlation coefficient (r) was 0.95 (S_y|x = 120 pmol/L, n = 128; Fig. 3 ). The imprecision profiles in the low range of the IRMA and the RIA are compared in Fig. 4 . With the IRMA, N-terminal proANP concentrations equal to or more than ~20 pmol/L can be considered as the working range (i.e., the range of N-terminal proANP concentrations that can be measured with an imprecision of <15%). However, the lower limit of the working range of the RIA was ~200 pmol/L, which was an order of magnitude higher than that of the IRMA.

View larger version (28K): [in this window] [in a new window]   Figure 3. Comparison of the N-terminal proANP concentrations determined by the newly developed IRMA method and by RIA.

View larger version (21K): [in this window] [in a new window]   Figure 4. Imprecision profiles of the IRMA and the RIA. The precision profiles (CV, %) were calculated from five determinations of each point in one assay. The working range was arbitrarily defined as the range of N-terminal proANP concentration measured with an imprecision <15%.

Gel filtration chromatography.
To analyze the immunoreactivities measured by our IRMA method and RIA method, we conducted a gel filtration study of pooled plasma (Fig. 5 ). Total recoveries of immunoreactivities by IRMA and RIA, were 98% and 87%, respectively. N-terminal proANP immunoreactive fractions determined by the two methods both gave one major peak eluting at a position slightly before N-terminal proANP (1–67) (7.4 kDa). The approximate molecular masses of these peaks seemed to be 10 kDa, possibly matching that of N-terminal proANP (1–98).

View larger version (20K): [in this window] [in a new window]   Figure 5. Gel filtration of immunoreactive N-terminal proANP in pooled human plasma. Fractions were subjected to IRMA and RIA. The elution points for molecular markers are indicated at the top [13.8 kDa, ribonuclease A; 7.4 kDa, N-terminal proANP (1–67); and 3.1 kDa, N-terminal proANP (1–25)].

N-terminal proANP concentrations in plasma.
The N-terminal proANP concentrations obtained from 33 healthy adults were 188 ± 71 pmol/L (mean ± SD). The mean concentration of N-terminal proANP in plasma of 25 patients with heart failure, 1030 ± 411 pmol/L, was significantly higher (P <0.001) than that of the control subjects. The correlation coefficients (r) between plasma N-terminal proANP concentrations and ANP concentrations were 0.58 and 0.76 in the control subjects and the patients, respectively. In the overall group of 58 samples, a strong correlation (r = 0.90) was found (Fig. 6 ).

View larger version (20K): [in this window] [in a new window]   Figure 6. Correlation between the N-terminal proANP concentrations and the ANP concentrations in human plasma.

stability
Stability of N-terminal proANP in blood.
We evaluated N-terminal proANP in whole blood with three different samples. The N-terminal proANP concentration in blood remained mostly unchanged for 24 h at 4 °C in the presence of EDTA. When stored at 25 °C, the N-terminal proANP concentration remained stable for 6 h and was at 89% of the initial concentration after 24 h.

Effect of hemolysis.
To examine the influence of hemolysis, various amounts of hemolysate were added to three plasma samples. The N-terminal proANP concentrations in plasma samples remained at 95% and 92%, with the hemolysate corresponding to 2.8 and 11.3 g/L of hemoglobin, respectively.

Stability of N-terminal proANP in EDTA plasma.
N-Terminal proANP in EDTA plasma was stable for at least 1 month at -20 °C. When stored at 4 °C, the N-terminal proANP concentrations remained stable for 3 days and was at 81% of the initial concentration after 4 weeks. In EDTA plasma containing aprotinin, the N-terminal proANP concentration remained unchanged for 4 weeks, even at 4 °C. Ten freeze-thaw cycles had no effect on plasma N-terminal proANP.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
ANP is an important contributor to the maintenance of sodium and volume homeostasis in healthy and diseased conditions. The plasma concentration of ANP has become an important indicator of heart failure. Recently, attention has also been focused on N-terminal proANP in the diagnosis of patients with chronic heart failure. Mathisen et al. (7) demonstrated the close correlation between concentrations of N-terminal proANP and cardiopulmonary hemodynamics in patients with cardiac disease. Increased plasma concentrations of N-terminal proANP were observed with increasing severity of heart failure assessed by the New York Heart Association Classification (6)(7)(8)(9). Lerman et al. (8) reported that plasma N-terminal proANP is markedly increased in asymptomatic patients with left ventricular dysfunction but less so than in patients with overt symptoms of heart failure. Hall et al. (17) reported that the measurement of N-terminal proANP is an important independent predictor of long-term prognosis after myocardial infarction.

Although RIA methods for N-terminal proANP provide interesting information on cardiac diseases, they are time-consuming. To overcome this problem, we developed a new IRMA method for routine determination of N-terminal proANP in human plasma, using monoclonal antibodies. Our IRMA method is not only less time-consuming but also more sensitive than the previous RIA method (Fig. 4 ).

We estimated the molecular mass of plasma N-terminal proANP by gel filtration on a Sephadex G-50 column. As shown in Fig. 5 , a single peak with an approximate molecular mass of 10 kDa was found by both IRMA and RIA. These results indicate an absence of low molecular weight N-terminal proANP fragments and suggest that the major circulating form has a high molecular weight, possibly being identical with N-terminal proANP (1–98). Itoh et al. (6) detected proANP (1–126) in some patients with heart failure, but its amount seemed to be negligible compared with that of N-terminal proANP (<2%). Meleagros et al. (18) also measured the circulating N-terminal proANP in human plasma, using an antibody directed against the C-terminal region of N-terminal proANP (87–98) as well as one against the N-terminal region, N-terminal proANP (1–16). They also concluded that N-terminal proANP (1–98) was the major circulating form in healthy subjects and patients with heart failure. Similar results were obtained by other investigators (4)(5)(6). Furthermore, there was a good correlation between plasma N-terminal proANP concentrations determined by IRMA and RIA, as shown in Fig. 3 . The relationship between the two methods is not very similar to the identity line (i.e., y = x). This result suggests differences in cross-reactivities of the endogenous N-terminal proANP and standard N-terminal proANP (1–67). Because the sequence of circulating N-terminal proANP is not known and because the human N-terminal proANP (1–98) standard is currently not available, the values obtained by IRMA and RIA may not represent absolute endogenous plasma concentrations of the N-terminal proANP. To be exact, the values represent N-terminal proANP (1–67)-like immunoreactivities in each assay condition.

The N-terminal proANP and ANP concentrations showed high correlation, in agreement with studies in most laboratories (6)(7)(18). Because proANP (1–126) is split into equimolar amounts of N-terminal proANP (1–98) and ANP on release, a good correlation between N-terminal proANP and ANP concentrations might be expected. But N-terminal proANP circulates at higher concentrations because it has a substantially longer half- life in plasma compared with ANP, which is very short-lived (half-life, 2–5 min). No evidence for specific receptor sites for N-terminal proANP currently exists. Therefore, the different clearance would lead to different clinical conclusions in some situations. For example, in patients with chronic renal failure, hemodialysis results in a 30% decrease in ANP concentrations but no changes in plasma N-terminal proANP concentrations (5)(6). Kettunen et al. (19) reported the plasma concentrations of N-terminal proANP and ANP did not go hand in hand in acute myocardial infarction. Furthermore, N-terminal proANP may serve as a better indicator during therapeutic administration of ANP and inhibitors of ANP metabolism (20).

A well-standardized protocol is needed for the routine determination of plasma ANP by immunoassay because sample collection and storage can greatly affect plasma ANP concentrations (10). Blood samples should be collected in the presence of EDTA and aprotinin and immediately centrifuged at 4 °C. The resulting plasma should be kept frozen at -20 °C or below. In this study, we showed that N-terminal proANP is stable for up to 6 h in whole blood containing only EDTA, even at room temperature. N-terminal proANP in EDTA plasma is stable for 3 days at 4 °C and 4 weeks at -20 °C. Additionally, hemolysis does not seem to markedly affect the N-terminal proANP assay, whereas ANP measurements cannot be done with hemolytic blood samples.

In conclusion, the major advantages of our IRMA method for N-terminal proANP over previous RIA methods and/or measurements of ANP can be summarized as follows: (a) It is quantitative enough to satisfy the fundamental analytical criteria; (b) it is less time-consuming; and (c) no special conditions are needed for sample collection and storage as those for ANP.

	Acknowledgments

 
We are very grateful to Kazuaki Endo and Toshihiko Futana for their useful discussions.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. de Bold AJ, Borenstein HB, Veress AT, Sonnenberg H. A rapid and potent natriuretic response to intravenous injection of atrial myocardial extract in rat. Life Sci 1981;28:89-94. [ISI][Medline] [Order article via Infotrieve]
2. Miyata A, Kangawa K, Toshimori T, Hatoh T, Matsuo H. Molecular forms of atrial natriuretic polypeptides in mammalian tissues and plasma. Biochem Biophys Res Commun 1985;129:248-255. [ISI][Medline] [Order article via Infotrieve]
3. Thibault G, Garcia R, Gutkowska J, Bilodeau J, Lazure C, Seidah NG, et al. The peptide Asn1-Tyr126 is the storage form of rat atrial natriuretic factor. Biochem J 1987;241:265-272. [ISI][Medline] [Order article via Infotrieve]
4. Buckley MG, Sagnella GA, Markandu ND, Singer DRJ, MacGregor GA. Concentrations of N-terminal ProANP in human plasma: evidence for ProANP (1–98) as the circulating form. Clin Chim Acta 1990;191:1-14. [ISI][Medline] [Order article via Infotrieve]
5. Sundsfjord JA, Thibault G, Lacrochelle P, Cantin M. Identification and plasma concentrations of the N-terminal fragment of proatrial natriuretic factor in man. J Clin Endocrinol Metab 1988;66:605-610. [Abstract]
6. Itoh H, Nakao K, Sugawara A, Saito Y, Mukoyama Y, Morii N, et al. -Atrial natriuretic polypeptide (-ANP)-derived peptides in human plasma: cosecretion of N-terminal -ANP fragments and ANP. J Clin Endocrinol Metab 1988;67:429-437. [Abstract]
7. Mathisen P, Hall C, Simonsen S. Comparative study of atrial peptides. ANF (1–98) and ANF (99–126) as diagnostic markers of atrial distension in patients with cardiac disease. Scand J Clin Lab Invest 1993;53:41-49. [ISI][Medline] [Order article via Infotrieve]
8. Lerman A, Gibbons RJ, Rodeheffer RJ, Bailey KR, McKinley LJ, Heublein DM, et al. Circulating N-terminal atrial natriuretic peptide as marker for symptomless left ventricular dysfunction. Lancet 1993;341:1105-1109. [ISI][Medline] [Order article via Infotrieve]
9. Clerico A, Iervasi G, Del Chicca MG, Maffei S, Berti S, Sabatino L, et al. Analytical performance and clinical usefulness of a commercially available IRMA kit for measuring atrial natriuretic peptide in patients with heart failure. Clin Chem 1996;42:1627-1633. [Abstract/Free Full Text]
10. Tsuji T, Masuda H, Imagawa K, Haraikawa M, Shibata K, Kono M, et al. Stability of human atrial natriuretic peptide in blood sample. Clin Chim Acta 1994;225:171-177. [ISI][Medline] [Order article via Infotrieve]
11. Boomsma F, Bhaggoe UM, Veld AJM, Schalekamp MADH. Comparison of N-terminal pro-atrial natriuretic peptide and atrial natriuretic peptide in human plasma as measured with commercially available radioimmunoassay kits. Clin Chim Acta 1996;252:41-49. [ISI][Medline] [Order article via Infotrieve]
12. Wei CM, Heutlein DM, Perrella MA, Lerman A, Rodeheffer RJ, McGregor CGA, et al. Natriuretic peptide system in human heart failure. Circulation 1993;88:1004-1009. [Abstract/Free Full Text]
13. Davidson NC, Coutie WJ, Strunthers AD. N-Terminal proatrial natriuretic peptide and brain natriuretic peptide are stable for up to 6 h in whole blood in vitro [Letter]. Circulation 1995;91:1276.
14. Hall C, Aaberge L, Stokke O. In vitro stability of N-terminal proatrial natriuretic factor in unfrozen samples: an important prerequisite for its use as a biochemical parameter of atrial pressure in clinical routine [Letter]. Circulation 1995;91:911.
15. Mukoyama M, Nakao K, Yamada T, Itoh H, Hosoda K, Saito Y, et al. Preparation of monoclonal antibodies against atrial natriuretic polypeptide precursor and application to highly sensitive sandwich enzyme immunoassay. J Hypertension 1988;6:S320-S322.
16. McConahey PJ, Dixon FJ. Radioiodination of proteins by the use of the chloramine-T method. Methods Enzymol 1980;70:210.[Medline] [Order article via Infotrieve]
17. Hall C, Rouleau JL, Moye L, de Champlain J, Bichet D, Klein M, et al. N-Terminal proatrial natriuretic factor: an independent predictor of long-term prognosis after myocardial infarction. Circulation 1994;89:1934-1942. [Abstract/Free Full Text]
18. Meleagros L, Ghatei MA, Gibbs JSR, Bloom SR. Pro-atrial natriuretic peptide (1–98): the circulating cardiodilatin in man. Peptides 1989;10:545-550. [ISI][Medline] [Order article via Infotrieve]
19. Kettunen RVJ, Leppaluoto J, Jounela A, Vuolteenaho O. Plasma N-terminal atrial natriuretic peptide in acute myocardial infarction. Am Heart J 1994;127:1449-1455. [ISI][Medline] [Order article via Infotrieve]
20. Gros C, Souque A, Schwartz JC, Duchier J, Cournot A, Baumer P, et al. Protection of atrial natriuretic factor against degradation: diuretic and natriuretic responses after in vivo inhibition of enkepharinase (EC 3.4.24.11) by acetorphan. Proc Natl Acad Sci U S A 1989;86:7580-7584. [Abstract/Free Full Text]