Vol.
175,
March
No.
29,
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS Pages
1991
DISTRIBUTION IMMUNOREACTIVE
AND PORCINE
CHARACTERIZATION C-TYPE
Shuzo UEDA1j2, Naoto MINAMIN03, Kenji KANGAWAl, Shigeru MATSUKURA2
759-767
OF
NATRIURETIC
PEPTIDE
Masahito ABURAYAl, and Hisayuki MATSU03*
2 ‘Department of Biochemistry and Department of Internal Medicine, Miyazaki Medical College, Kihara, Kiyotake, Miyazaki 889-16, Japan 3National Cardiovascular
Received
February
12,
Center Research Institute, Fujishirodai,
Suita, Osaka 565, Japan
1991
SUMMARY: C-type natriuretic peptide (CNP) is a new member of the natriuretic peptide family recently identified in porcine brain (1). We raisedan antiserumagainstporcine CNP and set up a radioimmunoassay (RIA) for CNP. Using this RIA system, distribution of immunoreactive (ir-) CNP in porcine tissuewas measuredand comparedwith that of ir-atria1 natriuretic peptide (ANP) and ir-brain natriuretic peptide (BNP). Tissue concentration of ir-CNP in brain was the highest of the three natriuretic peptidesat about 0.79 pmol/g wet wt. CNP was present in medulla-pons in high concentration, with a significant concentration detected in cerebellum. In contrast, ir-CNP was not detected in peripheral tissue, including heart, in a significant concentration. Thesedata demonstratedsharpcontrastsin the distribution of the three natriuretic peptides, suggestingthat CNP is a natriuretic peptide functioning in the central nervous system. 0 1991 Academic Press, Inc.
In addition to atria1 natriuretic peptide (A-type natriuretic peptide, ANP) and brain natriuretic peptide (B-type natriuretic peptide, BNP) , we have recently identified in porcine brain a third member of the natriuretic peptide family, designatedC-type natriuretic peptide (CNP) (l-3) . CNP shows a high sequencehomology to ANP and BNP, especially in its 17-residue ring structure formed by disulfide linkage, although it lacks the C-terminal tail commonly found in ANP and BNP.
CNP also exerts pharmacological effects, such as
natriuresis and vasorelaxation, in a manner similar to ANP and BNP, but relative potency of CNP in each assaysystem was different from that of ANP and BNP. These data suggestthat
*To whom correspondenceshouldbe addressed. Abbreviations: CNP, C-type natriuretic peptide; ANP, atria1(A-type) natriuretic peptide; BNP, brain (B-type) natriuretic peptide; RIA, radioimmunoassay; ir, immunoreactive; TFA, trifluoroacetic acid; HPLC, high performanceliquid chromatography; MW, molecular weight.
759
0006-291X/91 $1.50 Copyright 0 1991 by Academic Press. Inc. All rights of reproduction in any form resenaed.
Vol.
175,
No.
3, 1991
BIOCHEMICAL
CNP has receptors and physiological though these three natriuretic
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
functions different from those of ANP and BNP, even
peptides are thought to participate
volume, electrolyte and blood pressure homeostasis cooperatively.
in regulating body fluid In order to help elucidate
the physiological significance of CNP, we established an RIA system for CNP and measured its regional distribution
in porcine central nervous sytsem and peripheral tissue.
MATERIALS
AND METHODS
Peptides: CNP and [Tyr’] -CNP were synthesized in our laboratory by the solid phase method with a peptide synthesizer (Applied Biosystems, 430A), and purified by reverse phase high performance liquid chromatography (HPLC). Homogeneity of synthetic peptides was confirmed by reverse phase and CM ion exchange HPLC as well as amino acid analysis and sequencing. Conjugation and immunization of CNP: CNP (6 mg) and bovine thyroglobulin (15 mg) were conjugated in 2 ml of O.lM sodium phosphate buffer (pH 7.4) by action of glutaraldehyde (4). The reaction mixture was dialyzed against saline and then 50mM sodium phosphate buffer (pH 7.4) /O.O8M NaCl, and used for immunization. The conjugates were proven to contain 60-70 CNP molecules per thyroglobulin molecule by amino acid analysis. The antigenic conjugate solution was emulsified with an equal volume of Freund’s complete adjuvant and used for immunizing 3 male New Zealand White rabbits by subcutaneous injection in the interscapulovertebral region. Rabb’ s were boostered every 20 days and bled 10 days after each booster. Radioiodination: [Tyr 8 ]-CNP was radioiodinated by the lactoperoxidase method (5). The resulting reaction mixture was subjected to rev rse phase HPLC (TSK gel ODS-120A column, 4.6 x 150 mm, Tosoh), and monoiodinated [Tyr &]-CNP was collected as a tracer for RIA. RIA procedures for CNP: RIA for CNP was performed by the same method as that reported for ANP and BNP (.5,6). The incubation buffer for RIA was 50mM sodium phosphate buffer (pH 7.4), containing 0.25% N-ethylmaleimide-treated BSA, 0.1% Triton X-100, 80mM NaCl, 25mM EDTA*2Na, 0.05% NaN3 and 2% dextran T-40. Standard CNP or unknown sample (100 ,u 1) was incubated with antiserum diluent (100 fi 1) for 24 hr, then tracer solution (17,000 cpm in 100 ,ul) was added. After additional incubation for 36 hr, free and bound tracers were separated by the polyethyleneglycol method. Radioactivity of the pellet was counted with a gamma counter (Aloka ARC-600), and assay was performed in duplicate at 4°C. RIAs for ANP and BNP: RIAs for ANP and BNP were performed as reported previously (5,6). CNP showed less than 0.001% crossreactivity in each RIA for ANP and BNP. Tissue extraction: Porcine brain, spinal cord and peripheral tissue were collected at a local slaughter house soon after killing. Brains were dissected on ice into nine regions according to the rat brain dissection method of Glowinski and Iversen (7). Striatum consisted of caudate nucleus, putamen and globus pallidurn. Spinal cords were dissected into six regions, i.e., ventral and dorsal halves of upper, middle and lower segments. For peripheral organs, about 1 g of each tissue was collected from each of at least five animals, washed with saline and used for extraction. After weighing, tissue of each brain region or peripheral tissue was pooled and extracted, since tissue concentration of ir-CNP was very low. Pooled tissue was diced and boiled in 10 volumes of water for 10 min to inactivate intrinsic proteases. After cooling, acetic acid was added (final concentration = IM), and boiled tissue was homogenized with a Polytron mixer for 4 min. Homogenates were centrifuged at 18,000 x g for 30 min, and supernatants were stored. In the case of central nervous system, about 30 g wet wt equivalents of the supernatant were desalted and condensed with a reverse phase C-18 column (90 ml, LC-SORB ODS, Chemco). The extracts of peripheral tissues were also treated with Sep-pak C-18 cartridge (2.5 ml, Waters). Adsorbed materials were eluted with a 60% CH3CN-0.1% trifluoroacetic acid (TFA) solution, and portions of the eluates were used for measurement of ir-CNP, ir-BNP and ir-ANP concentrations. Concentration of ir-CNP, ir-BNP and ir-ANP in each region or tissue meas red at least three times. Recovery yield of ir-CNP was estimated by adding EY I-[Tyr 8 ]-CNP, and more than 95% of radioactivity was always recovered. Characterization of ir-CNP: Porcine whole brain (740 g wet wt from 9 animals), spinal cord (252 g wet wt from 7 animals) and adrenal medulla (14.4 g from 54 animals) were extracted by 760
Vol.
175, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
the method describedabove. The C-18 column adsorbedfraction was separatedby Sephadex G-50 fine column (1.8 x 135 cm, brain and spinal cord) or SephadexG-75 column (3.0 x 151 cm, adrenal medulla) using 1M acetic acid as a solvent. An aliquot of each fraction was submitted to RIA for CNP. The fractions containing ir-CNP were further separatedby reverse phaseHPLC and CM ion exchange HPLC. ReversephaseHPLC was performed on a Hi-Pore RP-318 column (4.6 x 250 mm, Bio-Rad) with a linear gradient elution of CH CN from 10% to 60% in a solution of 0.1% TFA. CM ion exchange HPLC was carried ou? on a TSK gel CM-2SW column (4.6 x 250 mm, Tosoh) with a linear gradient elution of HCOONH4 (pH 6.6) from 1OmM to l.OM each containing 10% CH3CN. Characterization of ir-CNP in each brain region and spinal cord was also performed by the methods described above, including gel filtration and reversephaseHPLC. RESULTS AND DISCUSSION Specificity of antiserumagainstporcine CNP: Antiserum #171-4 recognizesporcine CNP with high avidity and was usable at a final dilution of 1:60,000 for RIA.
As shown in Fig. 1,
half-maximum inhibition of tracer binding by CNP-22 was observed at 10 fmol/tube and peptideswere measurableat a range of l-50 fmol/tube. Porcine a-ANP and BNP-26 showed 0.015% and 0.46% crossreactivity in this RIA system. On the other hand, recently identified CNP-53 (8) an N-terminally elongated form of CNP-22, as well as chicken CNP showed an affinity comparable to porcine CNP-22, while crossreactivity of frog CNP was only 2.3% (9,lO).
Basedon theseresults, the antiserumwas found to mainly recognize the ring structure
of CNP-22, especially its C-terminal half. In this RIA system, the intra- and inter-assay coefficients of variation were lessthan 3.0% and 10.4% (at 10 fmol/tube), respectively. Characterization of ir-CNP in porcine brain and spinal cord: Ir-CNP in whole brain and spinal cord extracts were each subjectedto SephadexG-50 gel filtration. As shown in Fig. 2A, four peaksof ir-CNP were observed in brain extracts. The third and fourth peaks,which constituted 75% of the total ir-CNP, were eluted at molecular weight (MW) regions correspondingto
100
I
I 1
I
10
,
I
102 103 104 Peptide (fmolltube)
I
105
I
106
Figure 1. Inhibition of 12.5I-CNP binding to antiserum #171-4 against porcine CNP at a final dilution of 1:60,000. l ----- porcine CNP-22, + ----gporcine BNP-26, A ----- porcine (human) a -ANP, V ----- CGRP, somatostatin, [Arg ]-vasopressin, bombesin, neurotensin and enkephalin.
761
Vol.
BIOCHEMICAL
175, No. 3, 1991
I
I
I
20
30
AND BIOPHYSICAL
I
I 40 Fraction
RESEARCH COMMUNICATIONS
50 number
60
I 70
80
Fieure 2. Sephadex G-50 gel filtration of (A) whole brain and (B) whole spinal cord extracts. Sample: Acid extracts of porcine (A) whole brain (20 g wet wt equivalents) and (B) whole spinal cord (50 g wet wt equivalents) pre-treated with reverse phase C-18 column. Column: 1.8 x 135 cm, fine, Pharmacia. Flow rate: 8 ml/hr. Fraction size: 5 ml/tube. Solvent: 1M acetic acid. Arrows indicate elution positions of 1) Vo, 2) CNP-53, 3) CNP-22 and 4) Vt.
CNP-53 (5K) and CNP-22 (3K). CNP immunoreactivity in spinal cord extracts were separated into two peaksof MW 5K and 3K, and no other peak of ir-CNP was detected(Fig. 2B). In both tissues,5K CNP was the major component,with the ratio of 5K CNP to 3K CNP being 9:l in brain and 5:l in spinal cord. 5K and 3K ir-CNP were further analyzed by reverse phaseand CM ion exchange HPLC.
As shown in Figs. 3A-3D, 3K CNP in brain was exclusively
composedof CNP-22, while 5K CNP consistedmainly of CNP-53 along with a less basic CNP-53 related peptide. In ir-CNP in spinal cord (Figs. 4A-4D), 3K CNP was also shown to be CNP-22, but 5K CNP was composedof CNP-53 and its related peptide with higher basicity. Basedon theseresults, CNP is presentmainly as CNP-53 and its related peptidesof MW 5K along with smaller amountsof CNP-22. The first peak (MWs20K) and secondpeak (MW+ 12K) of ir-CNP observed in the gel filtration of brain extracts were also analysed by reverse phase HPLC. The first peak was highly hydrophobic in nature and was eluted asa broad peak in HPLC, suggestingnon-specific interaction with the antiserum. On the other hand, 12K CNP was eluted as a sharppeak around the elution times of CNP-22 and CNP-53. Since the MW of this ir-CNP is comparable to recently elucidated pro-CNP (ll), 12K CNP might be a pro-CNP of 103 residues. Since ir-CNP in brain was shown to compriseat least 3 componentsand 25% of the total ir-CNP remained unidentified, ir-CNP in each brain and spinal cord region was characterized
Vol.
175,
No.
3,
1991
BIOCHEMICAL
LJ
1
1
1
I
I
c
0
10
20
30
40
50
60
Time
(mln)
AND
BIOPHYSICAL
0
10
20
RESEARCH
30
40
Tlme
50
COMMUNICATIONS
60
70
8(
(min)
Figure 3. Reverse phase HPLC (A and B) and CM ion exchange HPLC (C and D) of ir-CNP in porcine whole brain extracts. Sample: (A$) 2 ml aliquots of fractions #40-43, (B,D) 2 ml aliquots of fractions #49-52 in Fig. 2A. Column: (A,B) Hi-Pore RP-318 (4.6 x 250 mm, Bio-Rad), (C,D) TSK gel CM-2SW (4.6 x 250 mm, Tosoh). Flow rate: (A,B) 1.5 ml/min, (C,D) 1.0 ml/min. Solvent system: (A,B) Linear gradient elution of CH CN from 10% to 60% in 0.1% TFA over 80 min, (C,D) linear gradient elution of HCOO&H4 (pH 6.6) from 1OmM to l.OM in the presence of 10% CH3CN over 100 min. Arrows indicate elution times of 1) methionine sulfoxide form of CNP-22, 2) CNP-22 and 3) CNP-53.
Ia 0
I
1
1
1
1
L
10
20
30
40
50
60
Time
0
(mln)
10
20
30
40
50 60
70
80
Time (mln)
Figure 4. Reverse phase HPLC (A and B) and CM ion exchange HPLC (C and D) of ir-CNP in porcine whole spinal cord extracts. Sample: (A,C) 1 ml aliquot of fractions #40-43, (B,D) 0.5 ml aliquot of fractions #49-52 in Fig. 2B. Chromatographic conditions and arrows for (A,B) and (CD) were identical to those used in Figs. 3A and 3C, respectively. 763
Vol. 175, No. 3, 1991
BIOCHEMICAL
Al
11 0
I 10 Fraction
41 Cl
23
1 20
Q 30 number
AND BIOPHYSICAL
I 40
II 0
23
I
1
4(E
I
I
II
10 20 30 40 0 Fraction
RESEARCH COMMUNICATIONS
number
123
I 10 Fraction
I 20
4
I 30 number
I 40
Fieure5. SephadexG-50 gel filtration of extractsof six brainregions. Sample:10g wet wt equivalentsof (A) medulla-pons, (B) hypothalamus, (C) septum, (D) midbrain-thalamus, (E) cerebellumand(F) striatum,eachpre-treatedwith reversephaseC-18 column. Column:2.2x 41 cm,fine, Pharmacia. Fractionsize: 5 ml/tube. Solvent:1M aceticacid. Flow rate: 12ml/k. Arrows indicateelutionpositionsof 1) Vo, 2) CNP-53, 3) CNP-22and 4) Vt.
separately. As shown in Fig. 5, 5K CNP was the major form in all regions, and CNP-22 was not detected in cerebellum. Based on these results, net concentrationsof 5K and 3K ir-CNP were calculated and are listed in the extreme right column in Table 1. Regional distribution of immunoreactive CNP in porcine central nervous system: Table 1 summarizestissueconcentration of ir-CNP, along with those of ir-ANP and ir-BNP, in porcine brain and spinal cord. In brain, concentration of ir-CNP was slightly higher than that of ir-BNP and about 10 times higher than that of ir-ANP (6). Ir-CNP was present in spinal cord in a concentration l/4 that of ir-BNP but 10 times higher than that of ir-ANP. Among the regions examined, the highest concentration of ir-CNP was observedin medulla-pons, and the second highest in hypothalamus. Dorsal halves of spinal cord also contained ir-CNP in a high concentration, comparable to that of hypothalamus. Interestingly, ir-CNP was observed at a relatively high concentration in cerebellum,where neuropeptidesare generally undetectable(4). Thus, ir-CNP is widely distributed in porcine brain at relatively small concentration differences, with its regional distribution being similar to that of ir-BNP except for cerebellum and striatum, but distinct from that of ANP. Regional distribution and characterization of ir-CNP in porcine peripheral tissue: In porcine peripheral tissue,a significant concentration of ir-CNP was found in heart and adrenalmedulla. In SephadexG-75 gel filtration of cardiac atrium extracts, ir-CNP was observedonly in a high MW region concomitantly with 7 -ANP and 7 -BNP. Since high concentrationsof ir-ANP 764
Vol.
175,
No.
3, 1991
Table
1. Regional
weight
@
brain*
82.20
olfactory
bulb
hippocampus
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
of immunoreactive ANP, BNP and CNP brain and spinal cord ir- ANP (pmolfg)
ir- BNP (pm01 / 9)
ir- CNP (pm01 I g)
ir- CNP** (pmol/ $9
0.06
0.52
0.79
0.58
2.82
k
0.18
0.67
0.54
0.64
0.23
3.32
t
0.55
175,
March
No.
29,
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS Pages
1991
DISTRIBUTION IMMUNOREACTIVE
AND PORCINE
CHARACTERIZATION C-TYPE
Shuzo UEDA1j2, Naoto MINAMIN03, Kenji KANGAWAl, Shigeru MATSUKURA2
759-767
OF
NATRIURETIC
PEPTIDE
Masahito ABURAYAl, and Hisayuki MATSU03*
2 ‘Department of Biochemistry and Department of Internal Medicine, Miyazaki Medical College, Kihara, Kiyotake, Miyazaki 889-16, Japan 3National Cardiovascular
Received
February
12,
Center Research Institute, Fujishirodai,
Suita, Osaka 565, Japan
1991
SUMMARY: C-type natriuretic peptide (CNP) is a new member of the natriuretic peptide family recently identified in porcine brain (1). We raisedan antiserumagainstporcine CNP and set up a radioimmunoassay (RIA) for CNP. Using this RIA system, distribution of immunoreactive (ir-) CNP in porcine tissuewas measuredand comparedwith that of ir-atria1 natriuretic peptide (ANP) and ir-brain natriuretic peptide (BNP). Tissue concentration of ir-CNP in brain was the highest of the three natriuretic peptidesat about 0.79 pmol/g wet wt. CNP was present in medulla-pons in high concentration, with a significant concentration detected in cerebellum. In contrast, ir-CNP was not detected in peripheral tissue, including heart, in a significant concentration. Thesedata demonstratedsharpcontrastsin the distribution of the three natriuretic peptides, suggestingthat CNP is a natriuretic peptide functioning in the central nervous system. 0 1991 Academic Press, Inc.
In addition to atria1 natriuretic peptide (A-type natriuretic peptide, ANP) and brain natriuretic peptide (B-type natriuretic peptide, BNP) , we have recently identified in porcine brain a third member of the natriuretic peptide family, designatedC-type natriuretic peptide (CNP) (l-3) . CNP shows a high sequencehomology to ANP and BNP, especially in its 17-residue ring structure formed by disulfide linkage, although it lacks the C-terminal tail commonly found in ANP and BNP.
CNP also exerts pharmacological effects, such as
natriuresis and vasorelaxation, in a manner similar to ANP and BNP, but relative potency of CNP in each assaysystem was different from that of ANP and BNP. These data suggestthat
*To whom correspondenceshouldbe addressed. Abbreviations: CNP, C-type natriuretic peptide; ANP, atria1(A-type) natriuretic peptide; BNP, brain (B-type) natriuretic peptide; RIA, radioimmunoassay; ir, immunoreactive; TFA, trifluoroacetic acid; HPLC, high performanceliquid chromatography; MW, molecular weight.
759
0006-291X/91 $1.50 Copyright 0 1991 by Academic Press. Inc. All rights of reproduction in any form resenaed.
Vol.
175,
No.
3, 1991
BIOCHEMICAL
CNP has receptors and physiological though these three natriuretic
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
functions different from those of ANP and BNP, even
peptides are thought to participate
volume, electrolyte and blood pressure homeostasis cooperatively.
in regulating body fluid In order to help elucidate
the physiological significance of CNP, we established an RIA system for CNP and measured its regional distribution
in porcine central nervous sytsem and peripheral tissue.
MATERIALS
AND METHODS
Peptides: CNP and [Tyr’] -CNP were synthesized in our laboratory by the solid phase method with a peptide synthesizer (Applied Biosystems, 430A), and purified by reverse phase high performance liquid chromatography (HPLC). Homogeneity of synthetic peptides was confirmed by reverse phase and CM ion exchange HPLC as well as amino acid analysis and sequencing. Conjugation and immunization of CNP: CNP (6 mg) and bovine thyroglobulin (15 mg) were conjugated in 2 ml of O.lM sodium phosphate buffer (pH 7.4) by action of glutaraldehyde (4). The reaction mixture was dialyzed against saline and then 50mM sodium phosphate buffer (pH 7.4) /O.O8M NaCl, and used for immunization. The conjugates were proven to contain 60-70 CNP molecules per thyroglobulin molecule by amino acid analysis. The antigenic conjugate solution was emulsified with an equal volume of Freund’s complete adjuvant and used for immunizing 3 male New Zealand White rabbits by subcutaneous injection in the interscapulovertebral region. Rabb’ s were boostered every 20 days and bled 10 days after each booster. Radioiodination: [Tyr 8 ]-CNP was radioiodinated by the lactoperoxidase method (5). The resulting reaction mixture was subjected to rev rse phase HPLC (TSK gel ODS-120A column, 4.6 x 150 mm, Tosoh), and monoiodinated [Tyr &]-CNP was collected as a tracer for RIA. RIA procedures for CNP: RIA for CNP was performed by the same method as that reported for ANP and BNP (.5,6). The incubation buffer for RIA was 50mM sodium phosphate buffer (pH 7.4), containing 0.25% N-ethylmaleimide-treated BSA, 0.1% Triton X-100, 80mM NaCl, 25mM EDTA*2Na, 0.05% NaN3 and 2% dextran T-40. Standard CNP or unknown sample (100 ,u 1) was incubated with antiserum diluent (100 fi 1) for 24 hr, then tracer solution (17,000 cpm in 100 ,ul) was added. After additional incubation for 36 hr, free and bound tracers were separated by the polyethyleneglycol method. Radioactivity of the pellet was counted with a gamma counter (Aloka ARC-600), and assay was performed in duplicate at 4°C. RIAs for ANP and BNP: RIAs for ANP and BNP were performed as reported previously (5,6). CNP showed less than 0.001% crossreactivity in each RIA for ANP and BNP. Tissue extraction: Porcine brain, spinal cord and peripheral tissue were collected at a local slaughter house soon after killing. Brains were dissected on ice into nine regions according to the rat brain dissection method of Glowinski and Iversen (7). Striatum consisted of caudate nucleus, putamen and globus pallidurn. Spinal cords were dissected into six regions, i.e., ventral and dorsal halves of upper, middle and lower segments. For peripheral organs, about 1 g of each tissue was collected from each of at least five animals, washed with saline and used for extraction. After weighing, tissue of each brain region or peripheral tissue was pooled and extracted, since tissue concentration of ir-CNP was very low. Pooled tissue was diced and boiled in 10 volumes of water for 10 min to inactivate intrinsic proteases. After cooling, acetic acid was added (final concentration = IM), and boiled tissue was homogenized with a Polytron mixer for 4 min. Homogenates were centrifuged at 18,000 x g for 30 min, and supernatants were stored. In the case of central nervous system, about 30 g wet wt equivalents of the supernatant were desalted and condensed with a reverse phase C-18 column (90 ml, LC-SORB ODS, Chemco). The extracts of peripheral tissues were also treated with Sep-pak C-18 cartridge (2.5 ml, Waters). Adsorbed materials were eluted with a 60% CH3CN-0.1% trifluoroacetic acid (TFA) solution, and portions of the eluates were used for measurement of ir-CNP, ir-BNP and ir-ANP concentrations. Concentration of ir-CNP, ir-BNP and ir-ANP in each region or tissue meas red at least three times. Recovery yield of ir-CNP was estimated by adding EY I-[Tyr 8 ]-CNP, and more than 95% of radioactivity was always recovered. Characterization of ir-CNP: Porcine whole brain (740 g wet wt from 9 animals), spinal cord (252 g wet wt from 7 animals) and adrenal medulla (14.4 g from 54 animals) were extracted by 760
Vol.
175, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
the method describedabove. The C-18 column adsorbedfraction was separatedby Sephadex G-50 fine column (1.8 x 135 cm, brain and spinal cord) or SephadexG-75 column (3.0 x 151 cm, adrenal medulla) using 1M acetic acid as a solvent. An aliquot of each fraction was submitted to RIA for CNP. The fractions containing ir-CNP were further separatedby reverse phaseHPLC and CM ion exchange HPLC. ReversephaseHPLC was performed on a Hi-Pore RP-318 column (4.6 x 250 mm, Bio-Rad) with a linear gradient elution of CH CN from 10% to 60% in a solution of 0.1% TFA. CM ion exchange HPLC was carried ou? on a TSK gel CM-2SW column (4.6 x 250 mm, Tosoh) with a linear gradient elution of HCOONH4 (pH 6.6) from 1OmM to l.OM each containing 10% CH3CN. Characterization of ir-CNP in each brain region and spinal cord was also performed by the methods described above, including gel filtration and reversephaseHPLC. RESULTS AND DISCUSSION Specificity of antiserumagainstporcine CNP: Antiserum #171-4 recognizesporcine CNP with high avidity and was usable at a final dilution of 1:60,000 for RIA.
As shown in Fig. 1,
half-maximum inhibition of tracer binding by CNP-22 was observed at 10 fmol/tube and peptideswere measurableat a range of l-50 fmol/tube. Porcine a-ANP and BNP-26 showed 0.015% and 0.46% crossreactivity in this RIA system. On the other hand, recently identified CNP-53 (8) an N-terminally elongated form of CNP-22, as well as chicken CNP showed an affinity comparable to porcine CNP-22, while crossreactivity of frog CNP was only 2.3% (9,lO).
Basedon theseresults, the antiserumwas found to mainly recognize the ring structure
of CNP-22, especially its C-terminal half. In this RIA system, the intra- and inter-assay coefficients of variation were lessthan 3.0% and 10.4% (at 10 fmol/tube), respectively. Characterization of ir-CNP in porcine brain and spinal cord: Ir-CNP in whole brain and spinal cord extracts were each subjectedto SephadexG-50 gel filtration. As shown in Fig. 2A, four peaksof ir-CNP were observed in brain extracts. The third and fourth peaks,which constituted 75% of the total ir-CNP, were eluted at molecular weight (MW) regions correspondingto
100
I
I 1
I
10
,
I
102 103 104 Peptide (fmolltube)
I
105
I
106
Figure 1. Inhibition of 12.5I-CNP binding to antiserum #171-4 against porcine CNP at a final dilution of 1:60,000. l ----- porcine CNP-22, + ----gporcine BNP-26, A ----- porcine (human) a -ANP, V ----- CGRP, somatostatin, [Arg ]-vasopressin, bombesin, neurotensin and enkephalin.
761
Vol.
BIOCHEMICAL
175, No. 3, 1991
I
I
I
20
30
AND BIOPHYSICAL
I
I 40 Fraction
RESEARCH COMMUNICATIONS
50 number
60
I 70
80
Fieure 2. Sephadex G-50 gel filtration of (A) whole brain and (B) whole spinal cord extracts. Sample: Acid extracts of porcine (A) whole brain (20 g wet wt equivalents) and (B) whole spinal cord (50 g wet wt equivalents) pre-treated with reverse phase C-18 column. Column: 1.8 x 135 cm, fine, Pharmacia. Flow rate: 8 ml/hr. Fraction size: 5 ml/tube. Solvent: 1M acetic acid. Arrows indicate elution positions of 1) Vo, 2) CNP-53, 3) CNP-22 and 4) Vt.
CNP-53 (5K) and CNP-22 (3K). CNP immunoreactivity in spinal cord extracts were separated into two peaksof MW 5K and 3K, and no other peak of ir-CNP was detected(Fig. 2B). In both tissues,5K CNP was the major component,with the ratio of 5K CNP to 3K CNP being 9:l in brain and 5:l in spinal cord. 5K and 3K ir-CNP were further analyzed by reverse phaseand CM ion exchange HPLC.
As shown in Figs. 3A-3D, 3K CNP in brain was exclusively
composedof CNP-22, while 5K CNP consistedmainly of CNP-53 along with a less basic CNP-53 related peptide. In ir-CNP in spinal cord (Figs. 4A-4D), 3K CNP was also shown to be CNP-22, but 5K CNP was composedof CNP-53 and its related peptide with higher basicity. Basedon theseresults, CNP is presentmainly as CNP-53 and its related peptidesof MW 5K along with smaller amountsof CNP-22. The first peak (MWs20K) and secondpeak (MW+ 12K) of ir-CNP observed in the gel filtration of brain extracts were also analysed by reverse phase HPLC. The first peak was highly hydrophobic in nature and was eluted asa broad peak in HPLC, suggestingnon-specific interaction with the antiserum. On the other hand, 12K CNP was eluted as a sharppeak around the elution times of CNP-22 and CNP-53. Since the MW of this ir-CNP is comparable to recently elucidated pro-CNP (ll), 12K CNP might be a pro-CNP of 103 residues. Since ir-CNP in brain was shown to compriseat least 3 componentsand 25% of the total ir-CNP remained unidentified, ir-CNP in each brain and spinal cord region was characterized
Vol.
175,
No.
3,
1991
BIOCHEMICAL
LJ
1
1
1
I
I
c
0
10
20
30
40
50
60
Time
(mln)
AND
BIOPHYSICAL
0
10
20
RESEARCH
30
40
Tlme
50
COMMUNICATIONS
60
70
8(
(min)
Figure 3. Reverse phase HPLC (A and B) and CM ion exchange HPLC (C and D) of ir-CNP in porcine whole brain extracts. Sample: (A$) 2 ml aliquots of fractions #40-43, (B,D) 2 ml aliquots of fractions #49-52 in Fig. 2A. Column: (A,B) Hi-Pore RP-318 (4.6 x 250 mm, Bio-Rad), (C,D) TSK gel CM-2SW (4.6 x 250 mm, Tosoh). Flow rate: (A,B) 1.5 ml/min, (C,D) 1.0 ml/min. Solvent system: (A,B) Linear gradient elution of CH CN from 10% to 60% in 0.1% TFA over 80 min, (C,D) linear gradient elution of HCOO&H4 (pH 6.6) from 1OmM to l.OM in the presence of 10% CH3CN over 100 min. Arrows indicate elution times of 1) methionine sulfoxide form of CNP-22, 2) CNP-22 and 3) CNP-53.
Ia 0
I
1
1
1
1
L
10
20
30
40
50
60
Time
0
(mln)
10
20
30
40
50 60
70
80
Time (mln)
Figure 4. Reverse phase HPLC (A and B) and CM ion exchange HPLC (C and D) of ir-CNP in porcine whole spinal cord extracts. Sample: (A,C) 1 ml aliquot of fractions #40-43, (B,D) 0.5 ml aliquot of fractions #49-52 in Fig. 2B. Chromatographic conditions and arrows for (A,B) and (CD) were identical to those used in Figs. 3A and 3C, respectively. 763
Vol. 175, No. 3, 1991
BIOCHEMICAL
Al
11 0
I 10 Fraction
41 Cl
23
1 20
Q 30 number
AND BIOPHYSICAL
I 40
II 0
23
I
1
4(E
I
I
II
10 20 30 40 0 Fraction
RESEARCH COMMUNICATIONS
number
123
I 10 Fraction
I 20
4
I 30 number
I 40
Fieure5. SephadexG-50 gel filtration of extractsof six brainregions. Sample:10g wet wt equivalentsof (A) medulla-pons, (B) hypothalamus, (C) septum, (D) midbrain-thalamus, (E) cerebellumand(F) striatum,eachpre-treatedwith reversephaseC-18 column. Column:2.2x 41 cm,fine, Pharmacia. Fractionsize: 5 ml/tube. Solvent:1M aceticacid. Flow rate: 12ml/k. Arrows indicateelutionpositionsof 1) Vo, 2) CNP-53, 3) CNP-22and 4) Vt.
separately. As shown in Fig. 5, 5K CNP was the major form in all regions, and CNP-22 was not detected in cerebellum. Based on these results, net concentrationsof 5K and 3K ir-CNP were calculated and are listed in the extreme right column in Table 1. Regional distribution of immunoreactive CNP in porcine central nervous system: Table 1 summarizestissueconcentration of ir-CNP, along with those of ir-ANP and ir-BNP, in porcine brain and spinal cord. In brain, concentration of ir-CNP was slightly higher than that of ir-BNP and about 10 times higher than that of ir-ANP (6). Ir-CNP was present in spinal cord in a concentration l/4 that of ir-BNP but 10 times higher than that of ir-ANP. Among the regions examined, the highest concentration of ir-CNP was observedin medulla-pons, and the second highest in hypothalamus. Dorsal halves of spinal cord also contained ir-CNP in a high concentration, comparable to that of hypothalamus. Interestingly, ir-CNP was observed at a relatively high concentration in cerebellum,where neuropeptidesare generally undetectable(4). Thus, ir-CNP is widely distributed in porcine brain at relatively small concentration differences, with its regional distribution being similar to that of ir-BNP except for cerebellum and striatum, but distinct from that of ANP. Regional distribution and characterization of ir-CNP in porcine peripheral tissue: In porcine peripheral tissue,a significant concentration of ir-CNP was found in heart and adrenalmedulla. In SephadexG-75 gel filtration of cardiac atrium extracts, ir-CNP was observedonly in a high MW region concomitantly with 7 -ANP and 7 -BNP. Since high concentrationsof ir-ANP 764
Vol.
175,
No.
3, 1991
Table
1. Regional
weight
@
brain*
82.20
olfactory
bulb
hippocampus
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
of immunoreactive ANP, BNP and CNP brain and spinal cord ir- ANP (pmolfg)
ir- BNP (pm01 / 9)
ir- CNP (pm01 I g)
ir- CNP** (pmol/ $9
0.06
0.52
0.79
0.58
2.82
k
0.18
0.67
0.54
0.64
0.23
3.32
t
0.55
