| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Hypertension Division CHUV, CH-1011 Lausanne, Switzerland.
a Author for correspondence. Fax 41-21-314 0761; e-mail
juerg.nussberger{at}chuv.hospvd.ch.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2-globulin angiotensinogen angiotensin-(1–10)
decapeptide [Ang-(1–10)], which is further metabolized to smaller
peptides that help maintain cardiovascular homeostasis. The Ang-(1–7)
heptapeptide has been reported to have several physiological effects,
including natriuresis, diuresis, vasodilation, and release of
vasopressin and prostaglandins. Methods: To investigate Ang-(1–7) in clinical settings, we developed a method to measure immunoreactive (ir-) Ang-(1–7) in 2 mL of human blood and to estimate plasma concentrations by correcting for the hematocrit. A sensitive and specific antiserum against Ang-(1–7) was raised in a rabbit. Human blood was collected in the presence of an inhibitor mixture including a renin inhibitor to prevent peptide generation in vitro. Ang-(1–7) was extracted into ethanol and purified on phenylsilylsilica. The peptide was quantified by radioimmunoassay. Increasing doses of Ang-(1–7) were infused into volunteers, and plasma concentrations of the peptide were measured.
Results: The detection limit for plasma ir-Ang-(1–7) was 1 pmol/L. CVs for high and low blood concentrations were 4% and 20%, respectively, and between-assay CVs were 8% and 13%, respectively. Reference values for human plasma concentrations of ir-Ang-(1–7) were 1.0–9.5 pmol/L (median, 4.7 pmol/L) and increased linearly during infusion of increasing doses of Ang-(1–7).
Conclusions: Reliable measurement of plasma ir-Ang-(1–7) is achieved with efficient inhibition of enzymes that generate or metabolize Ang-(1–7) after blood sampling, extraction in ethanol, and purification on phenylsilylsilica, and by use of a specific antiserum.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Antiserum against Ang-(1–7) heptapeptide was raised in a New Zealand White rabbit by immunization with synthesized Ang-(1–7) coupled to bovine thyroglobulin by the carbodiimide method at a molar incorporation ratio of 63 ± 24 (2). The antiserum was used in a final dilution of 1:41 000 in the Tris-albumin buffer. Radioimmunoassay calibration curves with different angiotensin peptides were established in the Tris-albumin buffer, and the sensitivity of the antiserum against a given peptide was defined as the reciprocal of the concentration of unlabeled peptide that displaced 50% of the 125I-labeled Ang-(1–7) bound to the antibody in the absence of any inhibitor. The specificity of the antiserum in discriminating between two peptides was defined as the quotient of their respective sensitivities (2). Taking the reactivity of the antiserum with Ang-(1–7) as 1.00, the cross-reactivities were as follows: Ang-(1–14), 0.00023; Ang-(1–10), 0.00031; Ang-(1–8), 0.004; Ang-(2–10), 0.00026; Ang-(2–8), 0.0008; Ang-(3–8), 0.0001; and Ang-(4–8), 0.000005. The detection limit of the antiserum (2 SD from zero binding) was, under optimal conditions, 0.5 fmol of Ang-(1–7).
The polyethylene glycol solution contained 200 g/L polyethylene glycol
(mean molecular weight, 8000; Sigma) dissolved in water containing 0.2
g/L sodium azide. Bovine 
-globulin (1.5 g; Sigma)
was dissolved in 100 mL of the Tris-albumin buffer.
blood sampling
Venous blood (5 mL) was drawn into a syringe containing 0.3 mL of
inhibitor solution and rapidly transferred to a polypropylene tube in
an ice bath.
liquid-phase extraction
Two 2-mL aliquots of the inhibitor-containing blood were
immediately pipetted into polypropylene tubes, on ice, containing 40 mL
of cold absolute ethanol (998 mL/L) and 4 mL of the barbitone
buffer. After mixing, the tubes were centrifuged at 3000g
and 2 °C for 25 min. The supernatant extract solution was decanted
into a Tris-albumin-buffer-coated polypropylene tube containing 0.25 mL
of 100 mL/L glycerol to enhance recoveries, dried under an air stream
at 37 °C, and then redissolved in 1 mL of bovine serum albumin (5
g/L) solution by sonication (40 W) and vortex-mixing. The latter
procedure optimized recoveries and prevented irreversible adsorption of
the peptide to the polypropylene. Samples were stored at -70 °C
until analyzed.
solid-phase extraction
Phenylsilylsilica cartridges were primed with 1 mL of methanol
followed by 1 mL of water and 2 mL of Tris-albumin buffer. The
extracted sample aliquots (1 mL) were thawed, kept on ice, and diluted
with 3 mL of cold Tris-albumin buffer. After centrifugation at
3000g and 2 °C for 5 min, supernatants were drawn through
the activated cartridges by a vacuum manifold. Columns were rinsed
twice with 3 mL of water. Retained angiotensins were eluted with 0.8 mL
of methanol into polypropylene tubes coated with the Tris-albumin
buffer. The methanol was evaporated under an air stream at 37 °C,
the dry extract was redissolved in 0.25 mL of ice-cold Tris-albumin
buffer, and Ang-(1–7) was quantified by subsequent radioimmunoassay.
labeling of ang-(1–7) heptapeptide
Ang-(1–7) heptapeptide was labeled with
125I by the chloramine-T method (3).
Specific activity of the detection angiotensin was 1500–2200 pCi/fmol,
i.e., 
1–2 fmol of angiotensin provided 2000 cpm.
radioimmunoassay
Anti-Ang-(1–7) antiserum (0.25 mL) and the extract solution (0.25
mL) were added together with 0.025 mL of detection-labeled Ang-(1–7)
heptapeptide (2000 cpm). For calibration curves, increasing amounts of
Ang-(1–7) heptapeptide (0–1000 fmol) were added to 0.25 mL of
Tris-albumin buffer, rather than the extract solution, and 0.025 mL of
0.04 mol/L o-phenanthroline in 30 mL/L ethanol was added to
correct for extracted o-phenanthroline. For volume
correction, 0.025 mL of 30 mL/L ethanol was added to the unknown
samples. Calibration and unknown samples were incubated for 72 h
at 4 °C. Antibody-bound and free Ang-(1–7) heptapeptides were
separated by polyethylene glycol precipitation, and bovine -globulin
solution (0.075 mL) was added to all samples to facilitate subsequent
precipitation of the antibody-bound peptide by mixing with 0.75 mL of
200 mL/L polyethylene glycol. After centrifugation for 20 min at
4000g and 4 °C, both the precipitate and supernatant were
counted in a well-type gamma counter. The hormone concentration of the
blood extract was read from the calibration curve, and plasma
concentrations were calculated taking the hematocrit into account.
assay validation
Recovery.
Recoveries were determined by supplementing
the blood of three different volunteers with 5, 10, or 15 fmol of
Ang-(1–7) heptapeptide, respectively (n = 10 each). The
endogenous concentrations (mean ± SD of triplicate measurements)
of this peptide were 2.7 ± 0.2, 6.1 ± 0.7, and 5.0 ±
0.2 pmol/L, respectively. Before ethanol extraction, 5, 10, or 15 fmol
of exogenous Ang-(1–7) in 0.05 mL of Tris-albumin buffer was added to
10 2-mL aliquots of the blood of one volunteer, and total Ang-(1–7)
was then measured. The difference between total and endogenous peptide
was expressed as a percentage of the added (exogenous) Ang-(1–7).
Subsequent results were not corrected for recoveries.
Imprecision.
Imprecision was defined by CVs.
Between-assay imprecision was determined by measuring the Ang-(1–7)
heptapeptide concentrations of three blood samples in six consecutive
independent assays (n = 6). Within-assay imprecision was
determined by repeated measurements of three different blood samples in
a single assay (n = 10).
Accuracy.
The accuracy and the absence of
nonspecific interferences in the radioimmunoassay were determined by
serially diluting pooled final-blood extracts with the Tris-albumin
buffer (1:2, 1:4, and 1:8 by volume). In another experiment, we tested
for possible interferences caused by the extraction procedure by
extracting water samples (water blank; n = 6) and analyzing them
for Ang-(1–7).
Healthy human subjects.
Eight healthy male volunteers
(ages, 21–44 years) came to the hospital after an overnight fast. They
were placed in a supine position, and intravenous cannulas were
inserted into an antecubital vein of each arm: one for infusion and one
for blood sampling. Five consecutive infusions of saline and Ang-(1–7)
(Clinalfa) at increasing doses (0, 2, 4, 8, 20
pmol·kg-1·min-1)
were administered for 30 min each (0.5 mL/min). The doses were
separated by 30-min washout infusions with saline. Blood samples were
drawn during the final 3 min of each dosing.
To obtain more reference values, we also collected venous blood from healthy hospital staff members in a supine position (ages, 24–62 years). Results were not corrected for recovery losses.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Precision.
Between-assay CVs for three different blood samples
were 12% (9.1 ± 1.1 pmol/L; n = 6), 10% (9.4
± 0.98 pmol/L; n = 6), and 7.6% (20.6 ± 1.8 pmol/L; n
= 6). Within-assay CVs for three different blood samples were
20% (3.9 ± 0.77 pmol/L; n = 10), 7.7% (18.3 ± 1.4 pmol/L;
n = 10), and 3.5% (26.8 ± 0.95 pmol/L; n = 10).
Accuracy.
Assay accuracy was demonstrated by the linearity of
Ang-(1–7) concentrations measured in serially diluted blood extracts
(Fig. 1A
). Water blanks were consistently below the detection limit of
the assay.
|
Detection limit.
Despite the somewhat better sensitivity
of the antiserum, the lowest point of the calibration curve was set at
an Ang-(1–7) concentration of 1 pmol/L, which is well beyond 2 SD of
the zero binding [125I-labeled Ang-(1–7) bound
to antibody in the absence of unlabeled peptide].
Reference values.
Plasma Ang-(1–7) concentrations in
healthy subjects were 5.1 ± 2.0 pmol/L (mean ± SD; n =
28; range, 1.0–9.5 pmol/L). During infusion of Ang-(1–7), plasma
Ang-(1–7) concentrations increased linearly in relation to the infused
dose from a baseline of 4.7 ± 0.9 pmol/L to 10.4 ± 0.7,
20.6 ± 3.0, 37.9 ± 4.5, and 79.1 ± 8.4 pmol/L
(mean ± SE) at 2, 4, 8, and 20
pmol·kg-1·min-1,
respectively (Fig. 1B
).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods for the measurement of the Ang-(1–7) in plasma and tissues have been established both without (8) and with specific HPLC isolation of this heptapeptide (9)(10). The generation and metabolization of angiotensin peptides after specimen sampling is a major problem, and enzyme inhibition before peptide measurement is crucial (11)(12)(13). We have therefore established a simple method for the measurement of ir-Ang-(1–7) in blood, based on an optimized enzyme-inhibitor cocktail in the sampling syringe, immediate ethanol precipitation of proteins, and subsequent solid-phase extraction of angiotensin peptides on phenylsilylsilica before radioimmunoassay.
The sensitivity of our antiserum and its low cross-reactivity of <1% with other angiotensin peptides made it possible to reliably measure low endogenous Ang-(1–7) concentrations in blood even without additional HPLC. Our reference values for ir-Ang-(1–7) in human subjects (1.0–9.5 pmol/L of plasma) are only slightly higher than the 1.0 ± 0.7 pmol/L measured in human plasma after HPLC by Lawrence et al. (9). We also found lower plasma concentrations of Ang-(1–7) in supine volunteers (1.9 ± 1.6 pmol/L; range, 0.3–4.8 pmol/L; n = 9) after solid-phase extraction and HPLC as described previously for the specific measurement of angiotensin peptides (11), but modified by blood collection on additional renin inhibitor (12). Ferrario et al. (14) found plasma Ang-(1–7) concentrations of 22.9 ± 8.8 pmol/L in healthy volunteers. Botelho et al. (8) found plasma ir-Ang-(1–7) concentrations of 33 ± 20 ng/L in healthy male Wistar rats, which appear to be considerably higher than our human results. However, Campbell et al. (15) found Ang-(1–7) concentrations in plasma from male Sprague–Dawley rats that were fivefold higher than those found by Lawrence et al. (9) for human plasma. In addition to the species differences, sampling techniques and nonspecific cross-reactions may account for the higher values obtained by Botelho et al. (8). The specific cross-reactivities of the antisera appear comparable, with even slightly better specificity for the Brazilian antiserum (8).
Precision validation for the described new assay was performed at
low Ang-(1–7) concentrations and provided results similar to or
markedly better than established angiotensin measurements
(11). Within the physiological range, the CV
decreased from 20% to 4% when blood Ang-(1–7) concentrations
increased from 4 to 27 fmol in 2-mL samples. The good precision
is based on both a reliable extraction procedure with constant
recoveries and efficient and rapid enzyme-inhibiting procedures that do
not interfere with the radioimmunoassay by sensitive and specific
antibodies. A clear linearity of Ang-(1–7) concentrations measured in
serially diluted blood extracts confirmed the accuracy of the new
method (Fig. 1A
). Another indication of the accuracy of our method is
provided by the dose–response curve of infused Ang-(1–7) in healthy
volunteers (Fig. 1B
), whose blood pressure and heart rates remained
unchanged throughout the infusions. Very similar plasma Ang II
concentrations were obtained when Ang-(1–8) was infused at the same
rates in humans (16) and dogs (17).
Interestingly, plasma Ang-(1–7) increased 2.8, 4.0, 4.2, and 3.7
pmol/L per pmol of Ang-(1–7) infused per kilogram per minute for
infusion rates of 2, 4, 8, and 20
pmol·kg-1·min-1,
respectively [i.e., the smallest increase in plasma Ang-(1–7)
occurred at the lowest infusion rate]. This would be compatible with a
feedback suppression by Ang-(1–7) of renin secretion and, hence, a
decrease of endogenous Ang-(1–7) during infusion of exogenous
Ang-(1–7). Such a phenomenon was documented previously for endogenous
and exogenous Ang-(1–8) (16), but not for Ang-(1–7).
However, such an interpretation of our infusion results remains
hypothetical because no renin measurements were performed in the
present study.
In conclusion, we present a reliable assay for ir-Ang-(1–7) in human plasma based on efficient inhibition of enzymes that generate or degrade Ang-(1–7) after blood sampling, extraction in ethanol, and a sensitive and specific antiserum.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  B. Gurzu, M. Costuleanu, S. M. Slatineanu, A. Ciobanu, and G. Petrescu Are Multiple Angiotensin Receptor Types Involved in Angiotensin (1-7) Actions on Isolated Rat Portal Vein? Journal of Renin-Angiotensin-Aldosterone System, June 1, 2005; 6(2): 90 - 95. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
