An Improved Fluorometric Assay of Rat Serumand Plasma Converting Enzyme

ROBSON A. S. SANTOS, EDUARDO M. KRIEGER, AND LEWIS J. GREENE

SUMMARY The most sensitive nonradiometric routine assay for angiotensin-converting enzyme(ACE) activity uses fluorometry to detect His-Leu released from Hip-His-Leu. Our results indicatethat, in contrast to human serum, rat serum and plasma contain large and variable amounts ofdipeptidase activity that lead to a subestimation of the ACE activity measured in 0.1 M potassiumphosphate buffer, pH 8.3, containing 0.3 M NaCI, the most commonly used assay for human serumand tissue ACE. We describe and validate an assay for 1 to 10 /xL rat and human serum or plasmausing 5 mM Hip-His-Leu in 500 fiL of 0.4 M sodium borate buffer, pH 8.3, containing 0.9 M NaCI at37°C that reduced the subestimation error to ^ 3% (rat serum) and ^ 0.1% (human serum) andincreased the ACE activity twofold to threefold. The Km and V,^ are reported for rat serum ACE(Hip-His-Leu) and dipeptidase (His-Leu) in borate buffer and phosphate buffer. Rat serum ACEhydrolysis of Hip-His-Leu measured by fluorometry correlated (r = 0.99, p < 0.05) with the hydroly-sis of angiotensin I measured by high-performance liquid chromatography. A direct method based onami no acid analysis is described for evaluating the dipeptidase error of complex mixtures such astissue extracts and other physiological fluids. We have found that the assay can be used to measureACE activity in 25 samples (in duplicate) in 2 hours with small intraassay (2.2%) and interassay(3.9%) coefficients of variation. It appears to be suitable for serial measurements in small animals instudies of the renin-angiotensin system in experimental hypertension and other physiopathologicalconditions. (Hypertension 7: 244-252, 1985)

KEY WORDS • rat angiotensin-converting enzyme • rat serum conversion of angiotensin Irat dipeptidase activity • fluorometric converting-enzyme assay • renin-angiotensin system

RENIN and angiotensin-converting enzyme(ACE) release the vasoconstrictor peptide an-giotensin II (ANG II) from the precursor pro-

tein angiotensinogen by sequential, specific limitedproteolysis reactions. Renin and ACE have a diffuse

From the Departments of Physiology and Biochemistry, and theInterdepartmental Protein Chemistry Center, Faculty of Medicineof Ribeirao Preto, University of Sao Paulo, Sao Paulo, Brazil.

Dr. Santos received a predoctoral fellowship from Coordena^aode Aperfeic.oamento de Pessoal de Nfvel Superior (CAPES) whileon leave from the Department of Physiology, Federal University ofMinas Gerais, Belo Horizonte, Minas Gerais, Brazil.

Parts of this study were presented by Dr. Santos in partial fulfill-ment of the requirements for the doctoral degree in Physiology tothe Faculty of Medicine of Ribeirao Preto, University of Sao Paulo,1983. Some of the data was presented at the Fifth Scientific Meetingof the Inter-American Society of Hypertension, Guarujd, SaoPaulo, Brazil, March 13-16, 1983.

Supported by Fundagao de Amparo a Pesquisa do Estado de SaoPaulo (FAPESP), Grants 78/1531, 80/240, 81/104-3, and 84/501,and by Financiadora de Estudos e Projetos (F1NEP).

Address for reprints: Dr. Lewis J. Greene, Department of Bio-chemistry, Faculty of Medicine of Ribeirao Preto, 14,100 RibeiraoPreto, Sao Paulo, Brazil.

Received March 13, 1984; revision accepted September 18,1984.

distribution in the body and have been detected invascular beds, blood, and many tissues including kid-ney and brain.u 2 Although it is generally believed thatthe ACE present in pulmonary endothelial cells plays amajor role in the conversion of angiotensin I (ANG I)to ANG II, extrapulmonary vascular and tissue ACEalso may be functionally important.1"3

Oparil and colleagues4 have suggested that plasmaACE activity may reflect pulmonary conversion andcan be altered by pulmonary disease, for which bothhigh4"6 and low4 levels of plasma ACE activity havebeen reported. Levels of plasma ACE increase in re-sponse to long-term blockade of ACE activity by cap-topril7 or MK-4218 in parallel to the increase also ob-served in levels of lung ACE, which indicates thatchanges of plasma ACE activity can reflect perturba-tions of the renin-angiotensin system.

Hip-His-Leu and, to a lesser extent, Hip-Gly-Glyare the NH2-protected tripeptide substrates used mostfrequently to measure ACE activity in blood, tissueextracts, and purified enzyme preparations.2-6 Themost commonly used routine serum assay employsHip-His-Leu in 0.1 M potassium phosphate buffer, pH8.3, containing 0.3 M NaCI (PB) and the spectropho-

244

by guest on Novem

ber 4, 2017http://hyper.ahajournals.org/

Dow

nloaded from

RAT SERUM CONVERTING-ENZYME ASSAY/Santos et al. 245

tometric determination of hippuric acid after extractioninto ethyl acetate.9 Friedland and Silverstein10 adaptedthis assay for the fluorometric determination of humanserum ACE, which is the most sensitive nonradiomet-ric detection method available, and also provided indi-rect evidence showing that human serum dipeptidaseactivity is sufficiently low so as not to interfere withthe assay.

When we attempted to use this fluorometric methodfor serial measurements of rat serum and plasma ACEactivity we found that the rat serum and plasma dipep-tidase activity were high enough to substantially inter-fere with the assay. This paper describes our solutionto the problem of dipeptidase interference: we used a0.4 M sodium borate buffer, pH 8.3, containing 0.9 MNaCl (BB) to inhibit dipeptidase activity. This bufferalso increased human and rat ACE activity 2.3-foldand threefold respectively. We found that the assayconditions proposed can be used for the measurementof ACE in as little as 1 /zL of rat serum or plasma forserial determinations in experimental models of hyper-tension or other pathophysiological states. Some of thedata have been reported previously."

Materials and MethodsHip-His-Leu, His-Leu, ANG I, and ANG II were

synthesized by Professors L. Juliano and A.CM.Paiva, Escola Paulista de Medicina, Sao Paulo, Brazil,and were homogeneous by the criteria of amino acidcomposition, high-voltage paper electrophoresis, andthin-layer chromatography. The bradykinin-potentiat-ing peptides (BPP), BPP5l (SQ 20,475), <Glu-Lys-Trp-Ala-Pro,12 and BPP,, (SQ 20,881), <Glu-Trp-Pro-Arg-Pro-Gln-Ile-Pro-Pro,12- " were purchased fromSchwarz/Mann, Orangeburg, New York. The aminoacid composition and concentration of each peptide weredetermined by amino acid analysis after acid hydrolysiswith HC1 at 110°C for 22 hours. Chemicals and aminoacid standards used for the amino acid analyzer andtrifluoroacetic acid were products of Pierce ChemicalCompany, Rockford, Illinois. The o-phthaldialdehydewas purchased from Sigma Chemical Company, St.Louis, Missouri. Methanol (Lichrosolv grade) was ob-tained from Merck & Co., Inc., Rahway, New Jersey.All other chemicals were reagent grade or equivalent. Allglassware was siliconized.

SerumBlood was collected into glass tubes from normoten-

sive male Wistar rats (200-250 g) by decapitation, andserum was separated after 2 hours at 37°C. Aliquots(500 fxh) were transferred to siliconized glass tubesand used immediately or stored at — 22°C. Rat plasmawas separated from blood collected in tubes precoatedwith heparin. Human serum was prepared in the samemanner from blood obtained by venous puncture of thearm.

Angiotensin-Converting EnzymeSolutions

One liter of a 1.25-fold concentrated buffer solution(0.5 M sodium borate buffer, pH 8.3, containing 1.125

M NaCl) was prepared with 30.5 g of H3BO4 and 65.7g of NaCl, adjusted to pH 8.30 with 13 N NaOH, andstored at room temperature for up to 1 month. Thestock substrate solution, 25 mM Hip-His-Leu, wasprepared by dissolving 448 mg in 40 ml of 25 mMNaOH and was stored at - 22°C. The concentration ofHip-His-Leu was determined by amino acid analysisand adjusted to 25 mM if necessary. The assay solutionwas prepared by mixing 4 volumes of 1.25-fold con-centrated buffer with 1 volume of substrate, with thepH checked at 25°C. When needed, 13 N NaOH orconcentrated H2SO4 was used to adjust the pH. Theassay solution was stored at — 22°C for up to 1 month.

AssaySerum (1-10 fiL) was incubated with 490 /xL of

assay solution containing 5 mM Hip-His-Leu in 0.4 Msodium borate buffer, pH 8.3, and 0.9 M NaCl for 15minutes at 37°C. The reaction was stopped by theaddition of 1.2 ml of 0.34 N NaOH. The product, His-Leu, was measured fluorometrically (365 nm excita-tion and 495 emission, Aminco Model J4-7461 fiuoro-monitor, American Instrument Co., Silver Springs,MD) after the addition of 100 /xL of o-phthaldialde-hyde (20 mg/ml) in methanol, which was followed 10minutes later by the addition of 200 /xL of 3 N HC1 andcentrifugation in a clinical centrifuge at 800 g for 5minutes at room temperature. Blanks (To) were pre-pared by reversing the order of the addition of enzymeand NaOH. All measurements were made in duplicate.

The relative fluorescence of the blank was 0.2 to 0.3compared with 10.0 to 30.00 obtained for rat serumsamples. The To blanks could be used because therelative fluorescence of serum (without substrate) orsubstrate (without serum) was constant under the con-ditions of the assay. Fifty tubes (25 samples) wereassayed at approximately the same time with a 15-second interval between additions.

Solutions of His-Leu (1-30 fiM) were prepared dai-ly as described for the substrate solution from stocksolutions of known concentration (amino acid analy-sis), and standard curves were obtained daily. Therewas a linear relationship between relative fluorescenceand His-Leu concentration. The relative fluorescenceof free histidine and free leucine was 5% and< 0.05%, respectively, of that of His-Leu. The pres-ence of 10 /AL of rat or human serum did not change theslope of the His-Leu standard curve. When assayswere carried out in potassium phosphate buffer con-taining different amounts of NaCl, standard solutionsof His-Leu were prepared in a solution appropriate forthe assay and run in parallel. The assay that used PB9-l0

was carried out as described for human serum byFriedland and Silverstein.10

The effect of BPP5l and BPP,, on ACE activity wasdetermined after preincubating 10 /nL of rat serum with10 fiL of the peptide for 7.5 minutes and then assayingthe sample in 500 /LAL of 5 mM Hip-His-Leu in BB.The concentrations of BPP5l and BPP,, in the assaywere 6.6 ^iM and 0.8 M respectively. The effect ofadded chloride ion was determined by incubating 10

by guest on Novem

ber 4, 2017http://hyper.ahajournals.org/

Dow

nloaded from

246 HYPERTENSION VOL 7, No 2, MARCH-APRIL 1985

/xL of rat serum (not dialyzed) with 0.490 ml of 5 mMHip-His-Leu in 0.4 M sodium borate buffer, pH 8.3.

Measurement of Dipeptidase ActivityThe conditions for the measurement of dipeptidase

activity in BB or PB are given in Table 2. The reactionwas stopped by adding an equal volume of 0.2 Msodium citrate buffer, pH 2.2, containing 0.4 N HC1and 30% polyethylene glycol, and the sample was fur-ther diluted to 2.0 ml with 0.2 M sodium citrate buffer,pH 2.2, containing 15% polyethylene glycol. Aliquots(0.900 ml) were applied directly to an amino acidanalyzer. Full-scale deflection of the amino acid ana-lyzer recorder was obtained with 5 nmol histidine and6 nmol leucine. Duplicate measurements varied by 8%or less. The levels of histidine (about 1 nmol/10 /xL ofplasma) and leucine (about 2 nmol/10 piL of plasma)did not increase with autolysis under the assay condi-tions.

Direct Determination of the DipeptidaseSubestimation Error

The ACE assay was carried out as described; fourtubes each were used for the assay and control (To). Atthe end of the 15-minute incubation, half the tubeswere processed for the fluorometric determination ofHis-Leu and the remaining tubes were processed foramino acid analysis as described for the measurementof dipeptidase activity. The To blanks were subtracted.The subestimation error, His/(His + His-Leu) x 100,thus provides a direct measurement of the effect ofdipeptidase activity under the actual assay conditions.

1 N HC1. The To blank was obtained by adding HC1before serum. In separate control experiments, ANG IIwas stable in the presence of serum for 60 minutesunder these conditions and neither serum nor ANG Iincubated alone for 60 minutes liberated material thateluted with the retention time of ANG II. The acidifiedreaction mixture was filtered through a MilliporeSwinnex assembly with a Millipore GSWP 0.2 \i filter(Waters Associates, Milford, MA), and 50 ^.L was

TABLE 1. Effect of Assay Conditions on Ral and Human SerumAngiotensin-Converting Enzyme Activity and the Dipeptidase Sub-estimation Error

Rat

ACE(nmolmin"1

ml"1)

serum

Error(%)

Human

ACE(nmolmin"1

ml"1)

serum

Error(%)Buffer

0.1 M Potassiumphosphate, pH 8.3,with 0.3 M NaCl 54.8 23.2 41 3.3

0.2 M Potassiumphosphate, pH 8 3,with 0.9 M NaCl 74.6 6.3 45 1.8

The substrate, 5 mM Hip-His-Leu, was incubated with 10 /xL ofrat or human serum in 0.25 ml of buffer, pH 8 3, at 37°C for 15minutes. His-Leu was measured by fluorometry and His by aminoacid analysis.

Subestimation error (%) = His/(His + His-Leu) x 100.ACE = angiotensin-converting enzyme.

Michaelis-Menten ConstantsThe concentrations of Hip-His-Leu (ACE) used for

measurements in BB were 0.2, 0.3, 0.5, 0.75, 1.0,1.5, and 2.0 mM and 0.5, 1.0, 2.0, 5.0, and 13.0 mMfor PB. His-Leu (dipeptidase activity) was used inconcentrations of 0.25, 0.5, 1.25, and 2.5 mM for BBand in concentrations of 0.0625, 1.125, 0.25, 0.5, and1.0 mM for PB. The incubation times were 5,10, and15 minutes for Hip-His-Leu in both buffers and His-Leu in PB. The assays of His-Leu in BB were carriedout for 15, 20, and 45 minutes. The hydrolysis of Hip-His-Leu was measured by fluorometry; the hydrolysisof His-Leu was measured by amino acid analysis. Theassays were carried out with 10 /iL of serum (poolderived from the serum of 5 rats, which had beenstored for 7-10 days) in a final volume of 250 fiL. Allmeasurements were carried out in duplicate, and the To

blank was subtracted. The Km and V ^ were calculatedby the method of Wilkinson14 from the velocity of thereaction, which was constant for all time points.

Determination of Conversion of Angiotensin Ito Angiotensin II

Serum (10 /xL) was incubated with 240 (xL of 0.25mM ANG I in 0.1 M potassium phosphate buffer, pH7.5, containing 30 mM NaCl at 37°C for 60 minutes.The reaction was stopped by the addition of 30 /xL of

TABLE 2. Effect of NaCl, Potassium Phosphate, and Sodium Bo-rate Concentration on Rat Serum Dipeptidase Activity at pH 8.3,2TC

NaCI(M)

0.15

0.15

0.30

0.30

0.30

0.30

0.30

0.30

0.90

0.90

0.90

0.90

0.90

0.90

Potassiumphosphate

(M)

0.10

—

0.05

0.10

0.30

—

—

—

0.10

0.30

0.40

—

—

—

Sodiumborate(M)

—

0.10

—

—

—

0.05

0.10

0.30

—

—

—

0.10

0.30

0.40

Relativeactivity

(%)

138

99

117

100

57

98

85

40

60

43

43

55

31

23

Serum, 10 fi\, was incubated with 0.25 ml of 0.25 mM His-LeuforO, 3, 6, and 9 minutes. Leucine was measured by amino acidanalysis. Initial rates of hydrolysis are reported as percentage ofactivity in 0.1 M potassium phosphate buffer containing 0.3 MNaCl: 72 nmol min"1 ml" 1 .

by guest on Novem

ber 4, 2017http://hyper.ahajournals.org/

Dow

nloaded from

RAT SERUM CONVERTING-ENZYME ASSAY/Santos et al. 247

injected into a Spherosorb (Applied Science, StateCollege, PA) ODS-C18 column (4.6 X 250 nm)equilibrated and eluted isocratically with metha-nol : 0 .1% trifluoroacetic acid in H2O (60 : 40)15 at1.5 ml/minute at 22°C (3000 psi). The effluent wasmonitored at 220 nm at 0.04 AUFS (absorbance unitsrecorder full scale). High-performance liquid chroma-tography (HPLC) was carried out with LaboratoryData Control (West Palm Beach, FL) equipment (Con-stametric II pump, Spectromonitor II, and model 3402recorder). The dipeptide His-Leu was eluted near thebreakthrough of the column, and ANG II and ANG Iwere eluted at 5.6 and 11 minutes respectively. Alinear relationship ( ± 5%) between peak height andconcentration was obtained for 0.16 to 3 nmol ANG II.

t

ISiooH

5 « -

\ I r i i i i

A 0.1 M lodiiaa bonu

I I I I I I I

B 0 J M iodhan chlorlda

„ •

i i i i i i I0 04 0J 1J

Sodium chlorkb, M

i i i r0 1 0J OJ 04

</• 1 4 0 -

t -

5 « j -

1

c

0

1 1 1

o • •• •

0

8

1 1 1

0

•

1D

i i 1

0

1

0

I0.4

I0J

I1.2

To.i 02 r

0J

I0.4

•D0UCT ONUlidv, U

FIGURE 1. Effect of NaCl, potassium phosphate, and sodiumborate concentration on rat and human serum ACE activity atpH 8.3, 37°C. Serum, 10 yL, was incubated with 250 yL of 5mM Hip-His-Leu for 15 minutes and hydrolysis was measuredby fluorometry. Activity is reported as percent of the activityobtained under the following conditions: A and B. 0.1 Msodium borate, 0.3 M NaCl, 104 (rat) and 46 (human) nmolmin'1 mt~'; C and D. 0.1 M potassium phosphate, 0.3 MNaCl, 75 (rat) and 43 (human) nmol min~' mt'. Open circles =rat serum; closed circles = human serum.

ResultsDipeptidase Problem

The fluorometric measurement of the hydrolysis ofHip-His-Leu is sensitive to a subestimation errorcaused by the dipeptidase hydrolysis of His-Leu be-cause the relative fluorescences of His and Leu are 5%and <0.05% of His-Leu. Table 1 shows that with thePB assay medium normally used to measure humanserum ACE (0.1 M potassium phosphate buffer, pH8.3, containing 0.3 M NaCl), the subestimation erroris small (3.3%) for a representative pool of humanserum but large (23.2%) for rat serum. Increasing buff-er and NaCl concentration decreased the error and in-creased the ACE activity of serum from both species.

Effect of Assay Conditions on Dipeptidase andAngiotensin-Converting Enzyme Activity

The data in Table 2, normalized to the rate of hy-drolysis obtained with PB (100%), show that the di-peptidase activity was lower in borate than in phos-phate buffer and that, of the combinations tested, 0.4M buffer plus 0.9 M NaCl gave the maximum reduc-tion of dipeptidase activity. It should be noted thatthese are initial rates of hydrolysis of His-Leu and notACE subestimation errors. Figure 1 shows the effect ofvarying NaCl and buffer concentrations on rat andhuman ACE activity, which has not been corrected forthe dipeptidase subestimation error. In the presence of0.1 M sodium borate or 0.1 M potassium phosphate,ACE activity increased with NaCl concentration andreached a maximum above 0.7 to 0.9 M NaCl with adecline demonstrable at 1.2 M NaCl in the presence of0.1 M sodium borate. In the presence of 0.3 M NaCl,increasing concentrations of borate up to 0.4 M acti-vated ACE activity of both species, whereas phosphateinhibited it at concentrations greater than 0.25 M. Re-sults for other serum pools in Table 3 complement thedata in Figure 1 and show that maximal activation of

TABLE 3. Effect of NaCl. Potassium Phosphate, and SodiumBorate Concentration on Rat and Human Angiotensin-ConvertingEnzyme Activity at pH 8.3. 37°C

NaCl(M)

0.3

0.3

0.9

0.9

0.9

0.9

0.9

0.9

Potassiumphosphate

(M)

0.10

—

0.10

—

0.30

—

0.40

—

Sodiumborate(M)

—

0.10

—

0.10—

0.30

—

0.40

Serum ACE

Rat

100

164

253

259

119

288

95

303

activity (%)

Human

100

151

178

211

ND

226

ND

228

Serum, 10/xL. was incubated with 0.25 ml of 5 mM Hip-His-Leufor 15 minutes, and hydrolysis was measured by fluorometry. Theresults are reported as a percentage of that obtained with 0.1 Mpotassium phosphate buffer containing 0.3 M NaCl, which was 57and 28.4 nmol min"1 ml"1 for rat and human serum respectively.

ND = not determined; ACE = angiotensin-converting enzyme.

by guest on Novem

ber 4, 2017http://hyper.ahajournals.org/

Dow

nloaded from

248 HYPERTENSION VOL. 7, No 2, MARCH-APRIL 1985

ACE activity was obtained when the assay solutioncontained 0.9 M NaCl and 0.4 M sodium borate buff-er, pH 8.3. The maximal increases for rat and humanACE were threefold and 2.5-fold respectively.

Comparison of the Effects of PB and BBFigure 2 (top panel) shows that the BB assay medi-

um (speckled bars) inhibited rat serum dipeptidase ac-tivity 80 to 85% in comparison to the PB assay medium(open bars) for both Pools 1 and 2, which were selectedfor low and high ACE activity. Dipeptidase activitydecreased about 50% after the first day of storage at— 20°C and was essentially constant for the following24 days.

Figure 2 (middle panel) shows that the subestima-tion error was constant and about 3 to 5% under BBconditions. In contrast, for the PB medium the errorwas 16% (Pool 1) and 40% (Pool 2) on Day 0 and waspool and storage-time dependent. Figure 2 (bottompanel) compares the measured uncorrected ACE activ-ity with the corrected activity (black area on top ofeach bar) as a function of storage time. Rat ACE activ-ity was about three times higher in BB than in PB andwas essentially constant when measured on Days 0, 6,and 30 in both assay media.

The Michaelis-Menten constants (Table 4) showedthat the reduction of dipeptidase activity by BB wasachieved by increasing Km threefold and decreasingVmu by 30%. The increase of the dipeptidase activityKm from 0.34 to 1.06 mM is as important as the reduc-tion of V ^ because His-Leu reaches a concentrationof only about 10 to 40 (JLM at the end of a typical assay.Table 4 also shows that the increase of ACE activitywas due to a 3.2-fold increase in V ^ . The decrease ofKm from 3.0 to 1.10 mM contributed much less to theincrease in measured activity because 5 mM substrateis used in the assay.

As the concentration of His-Leu in the assay is wellbelow the Km of the dipeptidase activity, carrying outthe assay in a larger volume but with the same amountof serum and same substrate concentration should re-duce the subestimation error in proportion to the vol-ume increase. Indeed, increasing the incubation vol-ume of BB from 250 fiL to 500 /xL reduced the errorfor Pools 1 and 2 by a little less than 50% (from 3.0%and 4.3% to 1.8% and 2.3% respectively). As expect-ed, the corrected ACE activity was the same in bothassay volumes. The error in PB was also reduced, from14.8% and 40.0% to 8.9% and 24% respectively,which, however, is unacceptably high. A surprisingand as yet unexplained result was that the correctedACE activity in PB was reduced by about 22% in thelarger reaction volume.

Properties and Specificity of the Assay of Rat SerumAngiotensin-Converting Enzyme Borate Buffer

The pH optimum for the hydrolysis of 5 mM Hip-His-Leu by rat serum in BB was near 8.3. Hydrolysiswas linear with time for at least 30 minutes, and therewas a linear relationship between the velocity of hy-

. .

S ..

M<

l [

Oil

1 n

r r- I

•i

ro

1m

f

Oil

1

1

1

1

1

10 * 3 *

DAY

• *

DAY

FIGURE 2. Effect of assay conditions and storage on rat di-peptidase activity, ACE subestimation error, and ACE activity.Assays were carried out atpH 8.3 at 37°C in BB (speckled bars)or PB (open bars). Top panel: Dipeptidase activity was mea-sured as described in Methods and in Table 2. Middle panel:ACE activity subestimation error was determined as describedin Methods under the conditions given in Table I. Bottompanel: ACE activity was measured as described in Methodsunder the conditions described in Table 3. The dark area on topof each bar corresponds to the correction of measured ACEactivity for the dipeptidase subestimation error. Pools I and 2each were prepared from the serum of three rats selected for lowand high ACE activity respectively.

by guest on Novem

ber 4, 2017http://hyper.ahajournals.org/

Dow

nloaded from

RAT SERUM CONVERTING-ENZYME ASSAY/Santos et al. 249

TABLE 4. Michaelis-Menten Constants for Rat Serum Dipeptidase Activity onEnzyme Activity on Hip-His-Leu at pH 8.3, 37°C

Buffer

0.1 M Potassium phosphate with 0.3 M NaCl

0.4 M Sodium borate with 0.9 M NaCl

Dipeptidase activity

Km(mM)

0.34±0.071.06±0.12

max

(nmolmin~' ml"1)181.5±12.6

132 + 6.6

His-Leu and Angiotensin-Converting

ACE:

(mM)

3.0±0.31

1.10±0.07

activity

max

(nmolmin"1 ml"1)

68.6 ±1.7

221.2±6.5

All values are means ± SD.ACE = angiotensin-converting enzyme.

drolysis and 1 to 10 /xL of rat serum. Under the condi-tions proposed for the assay (i.e., up to 10 yu.L of ratserum in 500 fxL of incubation volume for 15 minutesat 37°C) most sera liberated 10 to 40 nmol His-Leu, theTo blank was < 3% of the product obtained in a typicalassay, and the dipeptidase error was < 3%. The inter-assay and intraassay coefficients of variation were3.9% and 2.2% respectively.

The specificity of the hydrolysis of Hip-His-Leu as ameasure of ACE activity was indirectly demonstrableby the 96% inhibition of activity when chloride wasnot added to the substrate assay solution and by the97% inhibition obtained by the Bothrops venom ACEinhibitors BPP5s (6.6 yM) and BPP^ (0.8 fiM).

Quantitative Comparison of Serum Angiotensin-Converting Enzyme Activity Measured withHip-His-Leu and Angiotensin I

We measured the conversion of ANG I by serum byusing HPLC to determine the levels of ANG II, as

illustrated in Figure 3 (p. 250), and by fluorescencemeasurement of His-Leu. The HPLC elution diagramobtained for a rat serum hydrolysate of ANG I is shownin Figure 3 (bottom). The peak eluted at 5.6 minutescorresponds to synthetic ANG II (Figure 3, top). Thepeak height of ANG II was used to determine theextent of hydrolysis of ANG I because His-Leu waseluted near Vo and other ultraviolet-positive materialeluting in this position interfered with the quantitationof His-Leu.

Table 5 shows the data for human and rat serumacting on ANG II and Hip-His-Leu under assay condi-tions appropriate for each substrate. Both products ofhydrolysis of ANG I were measured: ANG II by HPLCand His-Leu by fluorescence. The hydrolysis of Hip-His-Leu was measured by fluorescence (His-Leu). Thedata show that when human serum acted on ANG Ithere was a stoichiometric release of both products asindicated by the fact that His-Leu/ANG II =« 0.95(Column 1). Although the rate of hydrolysis of Hip-

TABLE 5. Comparison of Human and Rat Serum Angiotensin-Converting Enzyme Activity on Hip-His-Leu and Angiotensin I Measured byFluorometry and High-Performance Liquid Chromatography

Serum

Human

Mean±SD

Rat

Mean±SD

Sample

1

2

3

4

5

1

2

3

Converting-enzyme activity

Angiotensin I

Fluorometry (F)(nmol His-Leumin"1 ml"1)

2.55

3.61

2.61

5.56

3.50

3.56 ±1.22

3.76

3.24

3.53

3.51+0.26

HPLC(nmol ANG IImin"1 ml"1)

2.66

3.70

2.72

6.04

3.75

3.77±1.37

4.74

4.31

4.68

4.57±0.23

Hip-His-Leu

Fluorometry (F)(nmol His-Leumin"' ml"1)

36.1

50.4

32.1

79.0

44.0

48.3± 18.5

90.0

75.2

88.3

84.5±8.1

Ratios of rates of hydrolysis

1ANG I (F)

ANG I (HPLC)

0.96

0.98

0.96

0.92

0.93

0.95 ±0.02

0.79

0.75

0.75 •

0.76 ±0.02

2Hip-His-Leu (F)ANG I (HPLC)

13.6 '

13.6

11.8

13.1

11.7

12.7±0.94

18.98

17.40

18.86

18.4±,0.88

3Hip-His-Leu (F)

ANG 1 (F)

14.1

14.0

12.3

14.2

12.6

13.4 + 0.91

24.0

23.2

25.0

24.0 + 0.90

Ten /AL of serum was incubated with 0.25 mM ANG I in 0.24 ml of 0.1 M sodium phosphate buffer, pH 7.5, containing 30 mM NaCl for 60minutes at 37°C. When Hip-His-Leu was the substrate, 10 y.L of serum was incubated with 0.490/iL of 5 mM Hip-His-Leu in 0.4 M sodiumborate buffer, pH 8.3, containing 0.9 M NaCl for 15 minutes at 37°C.

The ratios of rates of hydrolysis by serum-converting enzyme are identified in Columns 1 to 3 by the substrate, with the method of detectiongiven in parentheses. Fluorometry (F) was used to detect His-Leu released from Hip-His-Leu and angiotensin I (ANG I), and high-performanceliquid chromatography (HPLC) was used to detect angiotensin II (ANG II) released from ANG I.

by guest on Novem

ber 4, 2017http://hyper.ahajournals.org/

Dow

nloaded from

250 HYPERTENSION VOL 7, No 2, MARCH-APRIL 1985

0.012-

0.008-

0.004-

s0.016-

0.012-

0.008-

0.004 -

IS

TIME, mil

FIGURE 3. Measurement of the conversion of ANG I to ANG II

by rat serum. The HPLC was carried out with a SpherosorbODS-CI8 column as described in Methods. Top: The samplecontained 0.6 nmol ANG II in 50 fiL of the same solution asused for the enzyme hydrotysate. Bottom: Rat serum hydroly-sate of ANG I was prepared as described in Methods. Thevolume of sample injected, 50 fxL, corresponded to 17.9% of theenzyme incubation mixture after the reaction was stopped.Peaks were identified on the basis of the elution positions ofknown peptides.

His-Leu was much higher than that of ANG II, and theserum samples were different, the ratios of the rates ofhydrolysis of the substrates were essentially constantwhen calculated on the basis of the hydrolysis of ANGI with either ANG II (12.7 ± 0.91, r = 0.98, p <

0.05, Column 2) or His-Leu (13.4 ± 0.91, r = 0.99,p < 0.05, Column 3). Essentially constant ratios of therates of hydrolysis of the two substrates also wereobtained for rat serum when ANG I hydrolysis wasmeasured by HPLC (18.4 ± 0.88, r = 0.99, p <0.05, Column 2). Stoichiometry for the release of bothANG I products was not demonstrable for rat serum(0.76 ± 0.03, Column 1), presumably because ratdipeptidase activity hydrolyzed about 20% of the His-Leu product. This finding can be explained by the factthat the assay was carried out in 0.1 M sodium phos-phate buffer, pH 7.5, containing 30 mM NaCl, whichis usually employed for ANG I hydrolysis16 but whichalso had high levels of dipeptidase activity because ofthe low concentrations of salt and buffer (data notpresented). These data show the excellent correlationbetween the hydrolysis of both substrates by humanand rat serum and provides quantitative evidence forthe validity of using Hip-His-Leu to measure ACEactivity in this complex mixture.

Comparison of the Measurement of Angiotensin-Converting Enzyme Activity in Rat Serumand Plasma

Serum was chosen for detailed study because pre-liminary experiments indicated that the level of dipep-tidase activity in serum was up to 30% higher than inplasma. After defining the appropriate assay condi-tions for rat serum, the behavior of the assay for ratplasma was checked. As documented here for rat se-rum, rat plasma ACE activity was increased in BB andhad a constant velocity that was linear with enzymeconcentration. The K,,, for Hip-His-Leu was the samefor both plasma and serum activity, as was the inhibi-tion demonstrable for BPP5a, BPP9a, and in the absenceof added chloride ion. When the ACE activity andsubestimation error for pools of plasma and serumprepared from the blood of the same rats were assayedin BB and PB, the specific activity of plasma wasapproximately 10% lower than serum for both buffersand the subestimation error was lower in plasma(7.3%) than in serum (11.7%) when PB was used, butwas still of unacceptably high levels. In BB plasmaand serum had a low (< 2.5%) subestimation error.These results indicate that the BB assay conditions aresuitable for the measurement of ACE activity in ratplasma as well as rat serum.

DiscussionThe most serious limitation of the fluorometric

method for the measurement of ACE in serum, plas-ma, or tissue extracts is the possible subestimationerror caused by dipeptidase hydrolysis of His-Leuor other dipeptides released from amino-terminal-blocked tripeptide substrates. Indeed, when we mea-sured rat serum or plasma ACE activity using Hip-His-Leu in 0.1 M potassium phosphate, 0.3 M NaCl,which is one of the most commonly used media formeasuring ACE activity in serum and tissue,5 6-9 10 l6

direct measurement showed that the rat serum dipepti-dase activity led to subestimation errors of 10% to 40%

by guest on Novem

ber 4, 2017http://hyper.ahajournals.org/

Dow

nloaded from

RAT SERUM CONVERTING-ENZYME ASSAY/Santos et al. 251

depending on serum sample, storage time, and theextent of hydrolysis. This problem does not arise whenhippuric acid is measured by spectrophotometry orHPLC after extraction or by radioactivity. Because ofthe sensitivity, accuracy, reproducibility, speed, andeconomy of the fluorometric method, however, wesought to minimize this error in order to apply themethod for the serial measurement of rat serum andplasma ACE in experimental models of hypertension.The difference between the behavior of rat and humanserum indicates that species differences — in this casethe levels of serum and plasma dipeptidase activity —must be considered when applying assays developedand validated for other species.

The extent of the ACE activation by chloride de-pends on pH, substrate, and buffer.1720 At pH 8.3,ACE activity is maximum at about 1.0 M NaCl, whichreflects the pH-dependent increase of the dissociationconstant, Ka, of chloride ion, thereby increasing V ^and decreasing Km.17-"

The higher activity of ACE in BB compared withthat in PB has been observed.21 n Figure 1 shows thatthere was a moderate concentration-dependent activa-tion by borate that was demonstrable even in the pres-ence of a near optimal NaCl concentration (0.9 M) atpH 8.3 (see Table 3). Ionic strength has been suggest-ed as an explanation for activation by sodium sulfate18;however, Bunning and Riordan17 and Shapiro and col-leagues'9 have shown that the activation achieved byincreasing the concentration of some anions appears tobe a specific effect and is not due only to ionicstrength. In contrast to borate, increasing concentra-tions of phosphate in the presence of 0.3 M NaClincreased and then decreased both human and rat se-rum ACE activity (Figure 1). The data in Table 3 showthat even at maximal NaCl concentration (0.9 M) therewas a 52% reduction of ACE activity when phosphateconcentration was increased from 0.1 M to 0.3 M and63% from 0.1 M to 0.4 M. Detailed studies of phos-phate inhibition of ACE activity were first carried outby Dorer and co-workers18 and Oshima and Na-gasawa,23 although other investigators had previouslyobserved the effect.24-25 It is possible that, due to thechelating properties of ortho- and polyphosphates,phosphates may inhibit ACE by interacting with zinc.23

Interestingly, the V ^ of rat serum ACE increasedtwofold when 0.1 M sodium borate with 0.3 M NaClwas compared with PB, with essentially no change inKm (data not presented). The biphasic effect of phos-phate (Figure 1) and the data given in Table 3 showingthat increasing phosphate concentration overcame thechloride activation may be interpreted in terms ofphosphate competing with chloride at one or morelysyl-e-amino groups.26-27

The inhibition of serum dipeptidase activity by in-creasing NaCl concentration, for which borate wasmore effective than phosphate (Tables 2 and 4), ap-pears to be an ionic strength effect that cannot beanalyzed in more detail because of the scarcity of datain the literature.

The use of synthetic substrates, such as Hip-His-

Leu, rather than the physiological substrate ANG I forthe measurement of ACE activity in complex mixturessuch as serum, plasma, or tissue extracts has beenvalidated with purified enzyme and with inhibitiondata for the complex mixtures of the type providedhere (BPP5a, BPP^, minus chloride). Friedland andSilverstein28 demonstrated parallelism between His-Leu release from ANG I and Hip-His-Leu by humanserum, which was determined fluorometrically. In thepresent study we have addressed this question by pro-viding both quantitative data and the positive identifi-cation of ANG II. We demonstrated the stoichiometricrelease of ANG II and His-Leu by human serum. Thelower recovery of His-Leu in rat serum hydrolysateswas due to dipeptidase activity in 0.1 M sodium phos-phate buffer, pH 7.5, 30 mM NaCl (conditions select-ed for optimal ANG I hydrolysis).16 Furthermore, ourdata showed a close correlation between ANG I andHip-His-Leu hydrolysis for several samples of rat andhuman serum. These data provide a quantitative andspecific validation for the use of Hip-His-Leu to mea-sure ACE activity in complex mixtures such as rat andhuman serum.

In spite of the almost tenfold differences in the rateof hydrolysis (nmol min"1 ml"1) of Hip-His-Leu by ratand human serum, the hydrolysis of the physiological-ly important substrate ANG I was essentially the same,4.57 ± 0.25 versus 3.77 ± 1.37 nmol min"' ml"1 forrat and human serum respectively (see Table 5). Con-roy and colleagues29 showed that purified rabbit anddog lung ACE have different Km and V ^ for Hip-His-Leu hydrolysis but that these constants are essentiallythe same for ANG I. These considerations emphasizethe problem of comparing levels of ACE activity be-tween species when Hip-His-Leu or other syntheticsubstrates are used.29 The conservation of specific ac-tivity for the physiologically important substrate, incontrast to differences observed for synthetic sub-strates, suggests subtle differences among ACE fromdifferent species, and this should be considered wheninterpreting differences in the activity of synthetic ornatural ACE inhibitors.

Our results suggest the efficacy of the use of 0.4 Msodium borate buffer, pH 8.3, containing 0.9 M NaClfor the fluorometric determination of rat serum andplasma and human ACE. It is most likely that thisassay solution can be used for tissue as well, but beforebeing applied to other species and tissues the extent ofthe possible interference by dipeptidase activity mustbe evaluated. We found that this assay, which requiresfive additions to the same tube, centrifugation, and afluorometric measurement, can be used to process 25samples in duplicate in less than 2 hours. It also ap-peared to be highly sensitive, simple, rapid, and muchless complicated than other methods in current use,which require organic solvent extraction, isotopecounting30 or HPLC31 for monitoring Hip-His-Leu hy-drolysis. We found that the sensitivity of the assaycould be increased tenfold by increasing the incubationtime and reducing the volumes of reagents used for thefluorometric assay.

by guest on Novem

ber 4, 2017http://hyper.ahajournals.org/

Dow

nloaded from

252 HYPERTENSION VOL 7, No 2, MARCH-APRIL 1985

The high sensitivity of the assay in its present formfacilitates serial measurements of ACE in blood fromsmall animals and is currently being used in our labora-tory to measure circulating ACE in different rat modelsof experimental hypertension."

AcknowledgmentsThe authors thank Wanda Perussi de Jesus and Gilberto J. Pado-

van for their expert technical assistance and Elettra Greene forskillful preparation and typing of the manuscript. The Interdepart-mental Protein Chemistry Center, Faculty of Medicine of RibeiraoPreto, gratefully acknowledges Laboratory Data Control, Divisionof Milton Roy, West Palm Beach, Florida, for donating part of theHPLC equipment used in this study.

References1. Peach MJ. Renin-angiotensin system: biochemistry and

mechanisms of action. Physiol Rev 1977;57:313—3702. SofTer RL. Angiotensin-converting enzyme. In: Soffer RL, ed.

Biochemical regulation of blood pressure. New York: JohnWiley & Sons, 1981:124-164

3. Oparil S, Koerner T, O'Donoghue JK. Mechanism of angio-tensin converting enzyme inhibition by SQ 20,881 (<Gln-Trp-Pro-Arg-Pro-Gln-Ue-Pro-Pro) in vivo: further evidence for ex-trapulmonary conversion. Hypertension 1979; 1:13—22

4. Oparil S, Low J, Koerner TJ Altered angiotensin I conversionin pulmonary disease. Clin Sci Mol Med 1976;51:537-543

5. Lieberman J. Elevation of serum angiotensin-converting-en-zyme (ACE) levels in sarcoidosis Am J Med 1975,59.365-372

6. Weaver LJ, Solliday NH, Celic L, Gugell D. Serial observa-tions of angiotensin-converting enzyme and pulmonary func-tions in sarcoidosis. Arch Intern Med 1981 ;141 931—934

7. Kokubo T, Ueda E, Ono M, Kawabe T, Hayashi Y, Kan T.Effects of captopril (SQ 14,225) on the remn-angiotensin-al-dosterone system in normal rats. Eur J Pharmacol 1980;62:269-275

8. Forslund T\ Fyhrquist F, Gronhagen Riska C, Tikkanen I.Induction of angiotensin-converting enzyme with the ACE in-hibitory compound MK-421 in rat lung. Eur J Pharmacol1982;80:121-125

9. Cushman DW, Cheung HS. Spectrophotometric assay andproperties of the angiotensin converting enzyme of rabbit lung.Biochem Pharmacol 1971 ;20:1637-1648

10. Friedland J, Silverstein E. A sensitive fluorimetric assay forserum angiotensin-converting enzyme. Am J Clin Pathol1976;66:416-^24

11. Santos RAS. Krieger EM, Greene LJ Plasma angiotensin con-verting enzyme (kininase II) activity before and after renalartery unclipping of chronic one-kidney, one clip renal hyper-tensive rats [Abstract P-4]. Fifth Scientific Meeting of theInter-American Society of Hypertension, Guaruja\ Sao Paulo,Brazil, March 13-16, 1983

12. Ferreira SH, Bartelt DC, Greene LJ. Isolation of bradykinin-potentiating peptides from Bothrops jararaca venom. Bio-chemistry 1970;9:2583-2593

13. Ondetti MA, Williams NJ, Sabo EF, Plus^ec J, Weaver ER,Kocy O. Angiotensin-converting enzyme inhibitors from thevenom of Bothrops jararaca: isolation, elucidation of struc-ture, and synthesis. Biochemistry 1971;10:4O33-4O39

14. Wilkinson GN. Statistical estimations in enzyme kinetics. Bio-chem J 1961;80:324-332

15. Husain A, Bumpus FM, Smeby RR, et al. Evidence for theexistence of a family of biologically active angiotensin I-likepeptides in the dog central nervous system. Circ Res 1983;52:460-464

16. Cushman DW, Cheung HS. Studies in vitro of angiotensin-converting enzyme of lung and other tissues. In: Genest J, ed.Hypertension. Berlin, Heidelberg, New York: Springer-Ver-lag, 1972:532-541

17. Biinning P, Riordan JF. Activation of angiotensin convertingenzyme by monovalent anions. Biochemistry 1983;22:110-116

18. Dorer FE, Kahn JR, Lentz KE, Levine M, Skeggs LT. Kineticproperties of pulmonary angiotensin-converting enzyme: hy-drolysis of hippurylglycylglycine. Biochim Biophys Acta1976;429:220-228

19. Shapiro R, Holmquist B, Riordan JF. Anion inactivation ofangiotensin converting enzyme: dependence on nature of sub-strate. Biochemistry 1983;22:2850-2857

20. Inokuchi J, Nagamatsu A. Effects of halides on dipeptidyl andtripeptidyl carboxypeptidase activities of kininase II. ChemPharm Bull (Tokyo) 1984;32:237-243

21. Boomsma F, de. Bruyn JHB, Derkx FHM, SchalekampMADH Opposite effects of captopril on angiotensin I-con vert-ing enzyme "activity" and "concentration"; relation betweenenzyme inhibition and long-term blood pressure response. ClinSci 1981;60:491-498

22. Hurst PL, Lovell-Smith CJ. Optimized assay for serum angio-tensin-converting enzyme activity. Clin Chem 1981 ;27:2048-2052

23. Oshima G, Nagasawa K. Some enzymatic properties of pepti-dyl dipeptide hydrolase (angiotensin I-converting enzyme). JBiochem (Tokyo) 1977;81:57-63

24. Yang HYT, Erdos EG, Levin Y. Characterization of a dipep-tide hydrolase (kininase II: angiotensin I converting enzyme). JPharmacol Exp Ther 1971; 177:291-300

25. Igic R, Erdos EG, Yeh HSJ, Sorrells K, Nakajima T. Angio-tensin I converting enzyme of the lung. Circ Res 1972;3O,31(Suppl II):II-51-II-61

26. Weare JA. Activation/inactivation of human angiotensin I con-verting enzyme following chemical modifications of aminogroups near the active site. Biochem Biophys Res Commun1982;104:13I9-1326

27. Shapiro R, Riordan JF. Critical lysine residue at the chloridebinding site of angiotensin converting enzyme. Biochemistry1983;22:5315-5321

28. Friedland J, Silverstein E. Sensitive fluonmetric assay for sc-rum angiotensin converting enzyme with the natural substrateangiotensin. Am J Clin Pathol 1977;58:225-228

29. Conroy JM, Hartley JL, Soffer RL. Canine pulmonary angio-tensin-converting enzyme: physicochemical, catalytic and im-munological properties. Biochim Biophys Acta 1978;524:403-^12

30. Ryan JW, Chung A, Ammons C, Carlton ML. A simple ra-dioassay for angiotensin-converting enzyme. Biochem J 1977;167:501-504

31. Chiknas SG. A liquid chromatography-assisted assay for an-giotensin I-converting enzyme (peptidyl dipeptidase) in se-rum. Clin Chem 1979;25:1259-1262

by guest on Novem

ber 4, 2017http://hyper.ahajournals.org/

Dow

nloaded from

R A Santos, E M Krieger and L J GreeneAn improved fluorometric assay of rat serum and plasma converting enzyme.

Print ISSN: 0194-911X. Online ISSN: 1524-4563 Copyright © 1985 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Hypertension doi: 10.1161/01.HYP.7.2.244

1985;7:244-252Hypertension. 

http://hyper.ahajournals.org/content/7/2/244World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://hyper.ahajournals.org//subscriptions/

is online at: Hypertension Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer process is available in the

Request Permissions in the middle column of the Web page under Services. Further information about thisOffice. Once the online version of the published article for which permission is being requested is located, click

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialHypertension Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on Novem

ber 4, 2017http://hyper.ahajournals.org/

Dow

nloaded from