0021-972X/90/7106-1611$02.00/0 Journal of Clinical Endocrinology fnd Metabolism Copyright © 1990 by The Endocrine Society
Vol. 71, No. 6 Printed in U.S.A.
Endothelin Family in Human Plasma and Cerebrospinal Fluid TOHRU YAMAJI, HIHOO JOHSHITA, MIYUKI ISHIBASHI, FUMIMARO TAKAKU, HIDETO OHNO, NOBUHIRO SUZUKI, HIROKAZU MATSUMOTO, AND MASAHIKO FUJINO Third Department of Internal Medicine, Faculty of Medicine, University of Tokyo, Tokyo; the Department of Neurosurgery, Saitama Medical College (H.J.), Saitama; the Department of Medicine, Teikyo University School of Medicine (M.I.), Kanagawa; Biochemicals Section, Central Research Laboratory, Mitsubishi Petrochemical Co., Ltd. (H.O.), Ibaraki; and Tsukuba Research Laboratories, Takeda Chemical Industries, Ltd. (N.S., H.M., M.F.), Ibaraki, Japan
ABSTRACT. To clarify whether endothelin may be present in human cerebrospinal fluid (CSF) and, if it exists, to compare its molecular forms with those of endothelin in human plasma, we analyzed pooled human CSF and plasma by high performance liquid chromatography with specific enzyme immunoassays for each endothelin peptide. Of the four human endothelin peptides hitherto identified, big endothelin-1 was the major molecular form of endothelin present in human CSF. In addition, there was a small but significant amount of endothelin-1 and endothelin-3 in CSF, while endothelin-2 was not detectable. Similarly, big endothelin-1^ endothelin-1, and endothelin-3 were identified in human plasma. Although big endothelin-1 was the most abundant, a substantial amount of endothelin-1 and endothelin-3 was found ifa plasma with a resultant lower molar ratio of big endothelin-1 to endothelin-1 than in CSF. In all CSF
A
POTENT vasoconstrictor peptide, endothelin, has been isolated from the culture medium of aortic endothelial cells, sequeticed, and synthesized (1). Infusion of endothelin into experimental animals causes profound cardiovascular, r$nal, and endocrine alterations (2-4). We and others demonstrated that endothelin is present in human plasmft and that the levels are elevated in certain pathological Conditions, suggesting a possible role for endothelin in the pathophysiology of human diseases (5-10). While human endothelin consists of three distinct isopeptide$ with similar but quantitatively different pharmacological activities (11), the molecular forms of endothelin in human blood are not well understood. Recently, endothelin has been shown in nonvascular tissues of rats and pigs, including brain and spinal cord (12-15). This raises a possibility that in addition to the Received April 3, 1990. Address requests for reprints to: Tohru Yamaji, M.D., Third Department of Internal Medicifte, Faculty of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo^ku, Tokyo 113, Japan.
samples from 17 patients requiring diagnostic lumbar puncture or lumbar anesthesia, endothelin was detectable, with a preponderance of big endothelin-1 relative to endothelin-1 and endothelin-3. The mean concentrations of endothelin-1 and endothelin-3 in simultaneously collected plasma were significantly (P < 0.01) higher than those in CSF, while there was no significant difference between the mean big endothelin-1 concentration in plasma and that in CSF. There was no significant correlation among concentrations of big endothelin-1, endothelin-1, and endothelin-3 in each paired sample. These results indicate that endothelin is present in human CSF that is differently processed from endothelin in vascular endothelial cells and suggest a possible role for endothelin as a modulator of neuronal functions. (J Clin Endocrinol Metab 7 1 : 1611-1615,1990)
regulation of vascular tonus, endothelin may have a function as a neuropeptide. If so, then endothelin may exist in cerebrospinal fluid (CSF), since accumulating evidence indicates that many neuropeptides, including hypothalamic releasing hormones, are released directly or indirectly into the brain ventricles. CSF may be a conduit of comuunication between hypothalamus and brain, between pituitary and brain, and between different regions of the central nervous system (16, 17). The present study was undertaken to prove the presence of endothelin in human CSF and, if it exists, to compare its molecular forms with those of endothelin in human plasma. Subjects and Methods Subjects Blood was collected from six healthy volunteers (four men and two women) into chilled heparinized tubes. Plasma was quickly separated by centrifugation at 4 C and stored at -20 C until analyzed. In two patients with subarachnoid hemorrhage, ventricular CSF from cannulation of the lateral ventricle per1611
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 22:39 For personal use only. No other uses without permission. . All rights reserved.
YAMAJI ET AL.
1612
formed for a therapeutic purpose was pooled in chilled plastic containers. Judging from the blood cell count in CSF, the amount of hemorrhage was estimated to be less than 1% of the total CSF volume in both patients. Pooled CSF was centrifuged, and the supernatant was stored at —20 C before the characterization of endothelin. To measure CSF levels of endothelin in individual patients, CSF was collected from 17 patients (7 men and 10 women) with various proven or suspected diseases who required diagnostic lumbar puncture or lumbar anesthesia. Their ages ranged from 29-77 yr. Three of the patients complaining of headache were subsequently found not to be suffering from any central nervous system disease. Peripheral blood samples also were taken into chilled heparinized syringes before lumbar puncture. Each sample was centrifuged, and plasma or the supernatant of CSF was stored at —20 C until assayed. Extraction of endothelin from plasma or CSF Endothelin was extracted from plasma or CSF by means of affinity chromatography on endothelin-1 monoclonal antibodycoupled agarose. The endothelin-1 monoclonal antibody was prepared by immunization of BALB/c mice with endothelin-1porcine thyroglobulin conjugates. Splenocytes of the mice were taken, and cell fusions were performed using mouse myeloma cells. A monoclonal antibody was selected, separated from ascites fluid by ammonium sulfate precipitation, and coupled to CH-Sepharose 4B (Pharmacia Fine Chemicals, Uppsala, Sweden). The resulting gel was packed in a small column (total volume, 0.5 mL) and equilibrated with 0.01 M phosphate-0.14 M sodium chloride, pH 7.0 [phosphate-buffered saline (PBS)]. Pooled plasma (160 mL) from healthy volunteers or pooled CSF (710 mL), to which an equal volume of PBS and 200 U/ mL aprotinin (Hoechst Japan Ltd., Tokyo, Japan) were previously added, was applied to the antiendothelin-1-Sepharose column. The column was washed with 10 mL PBS and eluted with 5 mL 1 M acetic acid. The eluates were lyophilized and subjected to reverse phase high performance liquid chromatography (HPLC). When endothelin concentrations in plasma or CSF from individual patients were measured, endothelin in 3-5 mL plasma or CSF was extracted with a smaller column (total volume, 0.1 mL) of antiendothelin-1-Sepharose. The column was eluted with 0.8 mL 1 M acetic acid containing 1 mg/mL BSA, lyophilized, and subjected to enzyme immunoassays (EIAs). The recoveries of 125I-labeled big endothelin-1, endothelin-1, endothelin-2, and endothelin-3 were 75.3 ± 4.1% (mean ± SD; n = 8), 78.3 ± 2.6%, 75.8 ± 4.2% and 60.2 ± 3.0%, respectively, when 3-5 mL plasma or CSF containing 47 fmol synthetic big endothelin-1 and 8 fmol each of endothelin-1, endothelin-2, and endothelin-3 was filtered through the column. Results were not corrected for recoveries. Synthetic big endothelin-1, endothelin-1, endothelin-2, and endothelin-3 used in this study were purchased from the Peptide Institute, Inc. (Osaka, Japan). Methionine sulfoxide forms of big endothelin-1 and endothelin-1 were prepared by incubating synthetic big endothelin-1 and endothelin-1, respectively, in 1 M acetic acid containing 1% hydrogen peroxide for 10 min at room temperature.
JCE & M • 1990
HPLC of plasma and CSF extracts Reverse phase HPLC of plasma and CSF extracts was performed on a 4.6 X 250-mm column of octadecylsilane, as described in detail previously (18). The solvents used were H2Oacetonitrile-10% trifluoroacetic acid (90:10:1; solvent A) and H2O-acetonitrile-10% trifluoroacetic acid (40:60:1; solvent B). A linear gradient elution system was employed from solvent A to B over 80 min. The flow rate was 1.0 mL/min and each 1mL fraction was collected. Fractions were lyophilized, reconstituted with the assay buffer, and assayed for endothelin by EIAs. The recoveries of big endothelin-1, endothelin-1, and endothelin-3 from reverse phase HPLC in both experiments averaged 88.1%, 85.0%, and 78.2%, respectively. EIAs of endothelin Three different sandwich-type EIAs were used to measure big endothelin-1, endothelin-1, endothelin-2, and endothelin-3. The details of the EIAs were previously described (6, 14, 19, 20). The El A for big endothelin-1 (EIA-A) and that for endothelin-3 (EIA-C) did not show a significant cross-reactivity with other endothelin peptides, respectively (cross-reactivities,
Vol. 71, No. 6 Printed in U.S.A.
Endothelin Family in Human Plasma and Cerebrospinal Fluid TOHRU YAMAJI, HIHOO JOHSHITA, MIYUKI ISHIBASHI, FUMIMARO TAKAKU, HIDETO OHNO, NOBUHIRO SUZUKI, HIROKAZU MATSUMOTO, AND MASAHIKO FUJINO Third Department of Internal Medicine, Faculty of Medicine, University of Tokyo, Tokyo; the Department of Neurosurgery, Saitama Medical College (H.J.), Saitama; the Department of Medicine, Teikyo University School of Medicine (M.I.), Kanagawa; Biochemicals Section, Central Research Laboratory, Mitsubishi Petrochemical Co., Ltd. (H.O.), Ibaraki; and Tsukuba Research Laboratories, Takeda Chemical Industries, Ltd. (N.S., H.M., M.F.), Ibaraki, Japan
ABSTRACT. To clarify whether endothelin may be present in human cerebrospinal fluid (CSF) and, if it exists, to compare its molecular forms with those of endothelin in human plasma, we analyzed pooled human CSF and plasma by high performance liquid chromatography with specific enzyme immunoassays for each endothelin peptide. Of the four human endothelin peptides hitherto identified, big endothelin-1 was the major molecular form of endothelin present in human CSF. In addition, there was a small but significant amount of endothelin-1 and endothelin-3 in CSF, while endothelin-2 was not detectable. Similarly, big endothelin-1^ endothelin-1, and endothelin-3 were identified in human plasma. Although big endothelin-1 was the most abundant, a substantial amount of endothelin-1 and endothelin-3 was found ifa plasma with a resultant lower molar ratio of big endothelin-1 to endothelin-1 than in CSF. In all CSF
A
POTENT vasoconstrictor peptide, endothelin, has been isolated from the culture medium of aortic endothelial cells, sequeticed, and synthesized (1). Infusion of endothelin into experimental animals causes profound cardiovascular, r$nal, and endocrine alterations (2-4). We and others demonstrated that endothelin is present in human plasmft and that the levels are elevated in certain pathological Conditions, suggesting a possible role for endothelin in the pathophysiology of human diseases (5-10). While human endothelin consists of three distinct isopeptide$ with similar but quantitatively different pharmacological activities (11), the molecular forms of endothelin in human blood are not well understood. Recently, endothelin has been shown in nonvascular tissues of rats and pigs, including brain and spinal cord (12-15). This raises a possibility that in addition to the Received April 3, 1990. Address requests for reprints to: Tohru Yamaji, M.D., Third Department of Internal Medicifte, Faculty of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo^ku, Tokyo 113, Japan.
samples from 17 patients requiring diagnostic lumbar puncture or lumbar anesthesia, endothelin was detectable, with a preponderance of big endothelin-1 relative to endothelin-1 and endothelin-3. The mean concentrations of endothelin-1 and endothelin-3 in simultaneously collected plasma were significantly (P < 0.01) higher than those in CSF, while there was no significant difference between the mean big endothelin-1 concentration in plasma and that in CSF. There was no significant correlation among concentrations of big endothelin-1, endothelin-1, and endothelin-3 in each paired sample. These results indicate that endothelin is present in human CSF that is differently processed from endothelin in vascular endothelial cells and suggest a possible role for endothelin as a modulator of neuronal functions. (J Clin Endocrinol Metab 7 1 : 1611-1615,1990)
regulation of vascular tonus, endothelin may have a function as a neuropeptide. If so, then endothelin may exist in cerebrospinal fluid (CSF), since accumulating evidence indicates that many neuropeptides, including hypothalamic releasing hormones, are released directly or indirectly into the brain ventricles. CSF may be a conduit of comuunication between hypothalamus and brain, between pituitary and brain, and between different regions of the central nervous system (16, 17). The present study was undertaken to prove the presence of endothelin in human CSF and, if it exists, to compare its molecular forms with those of endothelin in human plasma. Subjects and Methods Subjects Blood was collected from six healthy volunteers (four men and two women) into chilled heparinized tubes. Plasma was quickly separated by centrifugation at 4 C and stored at -20 C until analyzed. In two patients with subarachnoid hemorrhage, ventricular CSF from cannulation of the lateral ventricle per1611
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 22:39 For personal use only. No other uses without permission. . All rights reserved.
YAMAJI ET AL.
1612
formed for a therapeutic purpose was pooled in chilled plastic containers. Judging from the blood cell count in CSF, the amount of hemorrhage was estimated to be less than 1% of the total CSF volume in both patients. Pooled CSF was centrifuged, and the supernatant was stored at —20 C before the characterization of endothelin. To measure CSF levels of endothelin in individual patients, CSF was collected from 17 patients (7 men and 10 women) with various proven or suspected diseases who required diagnostic lumbar puncture or lumbar anesthesia. Their ages ranged from 29-77 yr. Three of the patients complaining of headache were subsequently found not to be suffering from any central nervous system disease. Peripheral blood samples also were taken into chilled heparinized syringes before lumbar puncture. Each sample was centrifuged, and plasma or the supernatant of CSF was stored at —20 C until assayed. Extraction of endothelin from plasma or CSF Endothelin was extracted from plasma or CSF by means of affinity chromatography on endothelin-1 monoclonal antibodycoupled agarose. The endothelin-1 monoclonal antibody was prepared by immunization of BALB/c mice with endothelin-1porcine thyroglobulin conjugates. Splenocytes of the mice were taken, and cell fusions were performed using mouse myeloma cells. A monoclonal antibody was selected, separated from ascites fluid by ammonium sulfate precipitation, and coupled to CH-Sepharose 4B (Pharmacia Fine Chemicals, Uppsala, Sweden). The resulting gel was packed in a small column (total volume, 0.5 mL) and equilibrated with 0.01 M phosphate-0.14 M sodium chloride, pH 7.0 [phosphate-buffered saline (PBS)]. Pooled plasma (160 mL) from healthy volunteers or pooled CSF (710 mL), to which an equal volume of PBS and 200 U/ mL aprotinin (Hoechst Japan Ltd., Tokyo, Japan) were previously added, was applied to the antiendothelin-1-Sepharose column. The column was washed with 10 mL PBS and eluted with 5 mL 1 M acetic acid. The eluates were lyophilized and subjected to reverse phase high performance liquid chromatography (HPLC). When endothelin concentrations in plasma or CSF from individual patients were measured, endothelin in 3-5 mL plasma or CSF was extracted with a smaller column (total volume, 0.1 mL) of antiendothelin-1-Sepharose. The column was eluted with 0.8 mL 1 M acetic acid containing 1 mg/mL BSA, lyophilized, and subjected to enzyme immunoassays (EIAs). The recoveries of 125I-labeled big endothelin-1, endothelin-1, endothelin-2, and endothelin-3 were 75.3 ± 4.1% (mean ± SD; n = 8), 78.3 ± 2.6%, 75.8 ± 4.2% and 60.2 ± 3.0%, respectively, when 3-5 mL plasma or CSF containing 47 fmol synthetic big endothelin-1 and 8 fmol each of endothelin-1, endothelin-2, and endothelin-3 was filtered through the column. Results were not corrected for recoveries. Synthetic big endothelin-1, endothelin-1, endothelin-2, and endothelin-3 used in this study were purchased from the Peptide Institute, Inc. (Osaka, Japan). Methionine sulfoxide forms of big endothelin-1 and endothelin-1 were prepared by incubating synthetic big endothelin-1 and endothelin-1, respectively, in 1 M acetic acid containing 1% hydrogen peroxide for 10 min at room temperature.
JCE & M • 1990
HPLC of plasma and CSF extracts Reverse phase HPLC of plasma and CSF extracts was performed on a 4.6 X 250-mm column of octadecylsilane, as described in detail previously (18). The solvents used were H2Oacetonitrile-10% trifluoroacetic acid (90:10:1; solvent A) and H2O-acetonitrile-10% trifluoroacetic acid (40:60:1; solvent B). A linear gradient elution system was employed from solvent A to B over 80 min. The flow rate was 1.0 mL/min and each 1mL fraction was collected. Fractions were lyophilized, reconstituted with the assay buffer, and assayed for endothelin by EIAs. The recoveries of big endothelin-1, endothelin-1, and endothelin-3 from reverse phase HPLC in both experiments averaged 88.1%, 85.0%, and 78.2%, respectively. EIAs of endothelin Three different sandwich-type EIAs were used to measure big endothelin-1, endothelin-1, endothelin-2, and endothelin-3. The details of the EIAs were previously described (6, 14, 19, 20). The El A for big endothelin-1 (EIA-A) and that for endothelin-3 (EIA-C) did not show a significant cross-reactivity with other endothelin peptides, respectively (cross-reactivities,
