Inhibition of Human Peptidyl Dipeptidase(Angiotensin I Converting Enzyme: Kininase II)by Human Serum Albumin and its Fragments

RAINER J. KLAUSER, P H . D . , CAROL J. G. ROBINSON, M.S., D. V. MARINKOVIC, P H . D .

AND ERVIN G. ERDOS, M.D.

SUMMARY Purified peptidyl dipeptidase (angiotensln I converting enzyme or kininase II) from humanlung or hog kidney is inhibited by commercially prepared plasma protein preparations, by human serumalbumin and by the additive albumin stabilizer, acetyltryptophan. After the initial steps of purification,albumin was detected by iramunodiffusion as a component in human lung peptidyl dipeptidase preparation.Fragment C of albumin (sequence 124-298) is a more potent inhibitor than the parent molecule(K, = 1.7 X 10 8 M). Reduction and carboxymethylation of five of the six S-S bridges in Fragment C yield themost potent noncompetitive inhibitor (K, = 3 X 10 ' M). Reduction of the sixth bridge raises the K,. This in-dicates that maintenance of the tertiary structure in Fragment C is of importance for the inhibition. Neitheralbumin nor Fragment C are substrates of the enzyme. Fragment C and its derivative also inhibit the inactiva-tion of bradykinin by the purified human enzyme and by the peptidyl dipeptidase on the surface of intactcultured human endotbeliai cells. (Hypertension 1: 281-286, 1979)

KEY WORDS • transfusion • shock • protein inhibitor • albumin • kininase inhibitionangiotensin I conversion • bradykinin • converting enzyme * peptidyl dipeptidase

P EPTIDYL DIPEPTIDASE (angiotensin Iconverting enzyme or kininase II, E.C.3.4.15.1; CE) cleaves dipeptides from the C-

terminal end of peptides such as angiotensin I orbradykinin.1's In addition converting enzyme (CE)hydrolyzes other biologically active peptides such asenkephalins.3 The enzyme is present in plasma of manand animals,4"* but plasma also contains an inhibitoror inhibitors of CE.2t 7

Because of the obvious importance of CE in themetabolism of vasoactive peptides,8'" we investigatedthe inhibition of the enzyme by protein preparationsand by albumin and its fragments in vitro.

Methods and Materials

Human serum albumin, essentially free of fattyacid, was obtained from Sigma Chemicals (St. Louis,

From the Departments of Pharmacology and Internal Medicine,University of Texas Health Science Center at Dallas, 5323 HarryHines Boulevard, Dallas, Texas.

Supported in part by the Department of the Navy contractN0O014-75-C-O807, and by NIH USPHS Grants HL16320 andHL14187. Dr. Rainer J. Klauser received a travel grant fromDeutsche Forschungsgemeinschaft.

Address for reprints: Ervin G. Erdos, M.D., Department of Phar-macology, University of Texas, Southwestern Medical School, 5323Harry Hines Boulevard, Dallas, Texas 75235.

MO); cyanogen bromide (87%) and iodoacetic acidwere purchased from Aldrich Chemical Co.(Milwaukee, WI). Goat antiserum to human albuminwas obtained from Miles Laboratories (Elkhart, IN).Hippuryl-glycyl-glycine (Hip-Gly-Gly) was fromVega-Fox Biochemicals (Tucson, AZ). Plasma pro-tein preparations (Plasmatein, Protenate and Plas-manate) were obtained from Abbott Laboratories,Hyland Laboratories and Cutter Laboratories, respec-tively. Converting enzyme was prepared from swinekidney by the method of Oshima et al.7 Human lungCE was prepared by gel filtration of a Sephadex G-200column followed by chromatography on DEAE-cellulose and hydroxyapatite columns. Culturedhuman endothelial cells10 were also used as a source ofenzyme.

Assay of Converting Enzyme

We assayed CE using te/7-butyloxycarbonyl-Phe(NO,)-Phe-Gly (BPPG) or Hip-Gly-Gly as sub-strate.2 With the former substrate the hydrolysis wasfollowed at 310 nm and with the latter, at 254 nm in aCary Model 118 spectrophotometer. In typical ex-periments, 0.1 ml enzyme solution in 5 mM tris pH7.4 was added to 0.8 ml of 0.2 M tris buffer, pH 7.4,containing 0.2 M NaCl and preincubated at 37°C for

281

by guest on July 15, 2018http://hyper.ahajournals.org/

Dow

nloaded from

282 HYPERTENSION VOL 1, No 3, MAY-JUNE 1979

10 2 0 3 0 4 0 5 0 6 0 7 0 80 9 0 1 0 0

PLASMA FRACTION

FIGURE I. Inhibition of purified hog kidney peptidyldipeptidase by commercially prepared 5% plasma proteinpreparations. Four different batches of three manufacturers(see Methods section) were used. Substrate: t-Boc-Phe-p-(NOt)-Phe-Gly. Abscissa: \d plasma fraction. Ordinate:changes in absorption at 310 nm.

10 minutes. Then 0.1 ml 10'2 M Hip-Gly-Gly wasadded and the increase in absorbance was registered at254 nm. One unit of enzyme cleaves 1 /imoles of sub-strate per minute. When inhibitors were used, theywere preincubated for 10-15 minutes with the enzyme.The best lung CE preparation contained 8 U/mg andshowed a single protein band in polyacrylamide gelelectrophoresis.

Bioassay

The inactivation of bradykinin by CE was followedby measuring contractions of isolated rat uterus.1-7

Gel Electrophoresis

Polyacrylamide gel electrophoresis was carried outin 7.5% polyacrylamide gels (120 X 6 mm) at pH 8.9with a constant current of 3 mA per tube at 4°C for 3hours. The sample was heated in 5% mercaptoethanoland 3% sodium dodecyl sulfate (SDS) for 1 hour at60°C and the molecular weight was estimated by plot-ting log mobility against molecular weights of humanserum albumin, ovalbumin, chymotrypsinogen andribonuclease A protein standards.

Amino Acid Analysis

Amino acid analysis was performed in a DurrumAmino Acid Analyzer after hydrolyzing the peptidesat 110°C for 16 hours with 6 N HC1 in sealedevacuated tubes. Mean values of three determinationswere used. Values for residues of amino acids per molfragment were calculated with amino acid determina-tions of McMenamy et al.11 and amino acid sequencedata of Meloun et al. for comparison.12 A Beckman121 automatic amino acid analyzer was used whenattempts were made to release dipeptides fromalbumin and Fragment C with CE.

Protein Determination

The concentration of protein in the solutions waseither determined by the method of Lowry et al." orby measuring the absorption at 280 nm with FragmentC as a standard.

Preparation of Albumin Fragments

Human serum albumin was cleaved to FragmentsA, B and C with cyenogen bromide and the fragmentswere separated and purified by the sequential use ofgel filtration and column chromatography accordingto McMenamy et al.11

The reduction and carboxymethylation of the di-sulfide bridges of the fragment were performed ac-cording to Crestfield et al.14 with slight modifications.Fragment C (55.5 mg) was dissolved in 5 ml of 0.5 Mtris buffer, pH 8.6. The solution (6.2 X 10"4 M Frag-ment C) was stirred under nitrogen for 1 hour. Themercaptoethanol was added in a concentrationsufficient to give a 25-fold excess over the 12 possiblesulfhydryl (SH) residues of Fragment C. The reactionwas allowed to proceed for 90 minutes at roomtemperature. Iodoacetic acid (1.73 g) was then addeddropwise, and the pH was simultaneously adjusted to8.7 with 0.2 N NaOH. After stirring for 1 hour, thesolution was dialyzed for 48 hours against distilledwater and then lyophilized (Method 1). Alternatively,the reduction was carried out in 8 M urea (Method 2)or with only a fourfold excess of mercaptoethanolover the possible SH-groups (Method 3).

ImmunodifTusion

Immunodiffusion was carried out on Ouchterlonyplates using antiserum to human albumin and topurified human lung CE. The latter was produced inthe goat by injecting purified enzyme.

Results

Commercially prepared plasma protein prepara-tions inhibited purified CE of hog kidney. Fourdifferent preparations of three manufacturers weretested. Thirty-five n\ of the 5% protein solution in-hibited the hydrolysis of BPPG by 50% (fig. 1).

Because commercial plasma protein preparationscontain 4 X 10~a M acetyltryptophan added asstabilizer, we tested this compound for inhibition ofCE. The IM with Hip-Gly-Gly substrate and witheither human or hog CE was 5 X 10"4 M. The otheradditive, sodium caprylate, had no effect on the ac-tivity of the enzyme. Related compounds such as tryp-tophan, acetylhistidine and histidine were also inactiveat 10 s M concentration.

Presence of Albumin in CE

During purification of human CE7> 16 crude prepara-tions have low enzymic activity, so we examinedhuman lung CE preparations during the various stagesof purification for the presence of albumin. (The

by guest on July 15, 2018http://hyper.ahajournals.org/

Dow

nloaded from

INHIBITION OF ANGIOTENSIN I CONVERTING ENZYME/Klauser et al. 283

details of purification of human CE will be publishedelsewhere.) We found that CE of homogenized humanlung extracted with detergent and partially purified bygel filtration on Sephadex G-200 column still con-tained albumin in the active peak (fig. 2), as shown byimmunodiffusion. Figure 2 indicates that the prepara-tion of CE reacted with antibody to human lung CEand also with antibody to human serum albumin.Albumin, which has a molecular weight of 66,500,eluted in gel filtration on Sephadex G-200 columnwith CE, which has a much higher molecular weight.In gel filtration, CE frequently shows a molecularweight of 200,000 or even higher,8 thus CE probablyforms a complex with albumin.

After additional steps of purification on DEAE-cellulose and hydroxyapatite columns, CE. did notreact with antibody to human albumin, indicating thatalbumin was removed from the preparation duringpurification (T. Stewart, to be published). CE,however, still formed a precipitin band with antibodyto the human lung enzyme.

Albumin

Because commercial plasma protein preparationscontain over 83% albumin,1" we studied the effect ofpurified, fat-free albumin on the CE. Purified CE ofhuman lung and hog kidney are equally inhibited byalbumin. The IM values with Hip-Gly-Gly substratewere 2 X 10^ M and 1.9 X 10^ M. (The concentra-tion of human enzyme was 42 mU/ml).

The inhibition of human lung enzyme was non-competitive. The Ki determined in the Dixon plot(fig. 3) was 3 X 10^ M.

Fragments of Albumin

Since the commercial plasma protein solutions in-hibited CE relatively more than accounted for byalbumin content, which is an order lower than the IM

value of purified albumin, we investigated whetherfragments of human albumin may be more potent in-hibitors than the native protein. Human albumin wascleaved to three fragments at the methionine residueswith CNBr and the fragments were separated bycolumn chromatography.11'1J Following one of theseveral conflicting nomenclatures, we called FragmentB the residues 1-123 of albumin, Fragment C residues124-298 and Fragment A residues 299-585. All threefragments inhibited CE. The IM values were 2.2 X 10^M, 7.4 X 10"* M, 2.5 X 10"6 M for Fragments A, Band C, respectively.

Fragment C

Fragment C has most of the ligand binding proper-ties of human plasma albumin,17-18 thus we studied theinhibition of CE by this protein. In electrophoresis,Fragment C in presence of SDS migrated as a singleband of a molecular weight of 18,000 ± 800. Thearnino acid composition of Fragment C is shown intable 1. The values for single amino acids obtained arein good agreement with the theoretical value

LP

A A

FIGURE 2. Immunodiffusion of peptidyl dipeptidase ofhuman lung obtained after gel filtration (LP) and after sub-sequent column chromatographies (HL). AB = antiserumto human lung enzyme, A A = antiserum to human albumin,SA = human serum albumin. After Sephadex G-200 filtra-tion human lung enzyme (LP) reacts both with antibody tothe enzyme and with antibody to human albumin, indicat-ing the presence of albumin. After further purification of en-zyme, HL does not cross-react with antibody to albumin.

calculated from the available data,11'12 indicating ahigh degree of purity of Fragment C.

To establish whether the inhibition depends on thesecondary and tertiary structure of the fragment or onfree SH groups, we reduced and carboxymethylatedthe disulfide bridges in Fragment C (CMFC) formed

0 1 2 3 4 5 I[M]«1O~*

FIGURE 3. Dixon plot of inhibition of purified human lungpeptidyl dipeptidase by human serum albumin. Solidcircles = 2.5 X 10~' M Hip-Gly-Gly, solid triangles =1 X 10' M Hip-Gly-Gly, AT, = 3 X 1Q-* M.

by guest on July 15, 2018http://hyper.ahajournals.org/

Dow

nloaded from

284 HYPERTENSION VOL 1, No 3, MAY-JUNE 1979

TABLE 1. Amino Acid Composition of Fragment C of Human Serum Albumin and its Reduced and Carboxy-melhylated Derivatives

Aminoacid

ASP

Thr

Ser

Glu

Pro

Gly

Ala

Cys

Val

He

Leu

Tyr

Phe

His

Lys

Arg

Trp

H-Ser

CM-Cys

FragmentC

15.8

5.8

8.7

30.6

3.9

3.3

31.7*

4.6

2.7

18.6

5.0

8.8

3.8

20.0

7.8+

1.0

—

CMFC(Method

16.7

5.3

9.1

30.0

2.6

3.0

23.3

—t4.8

3.0

17.7

5.5

9.8

3.7

18.4

7.8

- t1.0

12.lt

1 CMFC 21) (Method 2)

CMFC 3(Method 3)

Residues per mol Fragment C

17.5

5.6

8.8

29.2

2.9

3.0

23.3

~t5.0

2.9

17.6

5.9

10.2

3.9

18.7

7.8

—t1.0

12.8

15.15.4

8.5

27.8

4.3

3.1

25.3*

5.6

3.2

18.4

6.2

10.54.2

19.6

11.0

—t1.0

9.6

Expectedvalues

14

6

9

24

5

3

23

12

5

4

19

6

10

4

19

9

1

1

*Cys and Ala could not be separated under conditions applied.jTrp is destroyed upon acid hydrolysis.JUpon reduction and carboxymethylation of Fragment C (GMFC) cysteine residues are completely or

partially converted into CM-Cys, depending on the conditions of reduction.

by 12 cysteine residues (CM-cysteinc). Three differentconditions were used for reduction. Method 1: mer-captoethanol was used in 25-fold excess over thetheoretically available SH groups to obtain FragmentCMFC 1. Method 2: mercaptoethanol was used in 25-fold excess with the addition of 8 M urea (CMFC 2).Method 3: mercaptoethanol was used in fourfold ex-cess (CMFC 3). Thus, with each method, the con-ditions of denaturing Fragment C differed. Differentamounts of iodoacetic acid were incorporated into the

fragments when 4- or 25-fold excess of mercapto-ethanol was applied for reduction of the disulfidebridges. With Methods 1 and 2 all disulfide bridges ofFragment C were derivatized in CMFC 1 and CMFC2. Method 3 left one of the six bridges intact in CMFC3 as shown in table 2. Accordingly, amino acidanalysis of Fragment C indicated 12.1 and 12.8 CM-cysteine residues obtained with Methods 1 or 2 butonly 9.6 with Method 3, since two CM-cysteines areformed after reduction of each bridge. All three

TABLE 2. Inhibition of Human Lung Converting Enzyme by Fragment C and its Reduced and Carboxymelhylated(CM) Derivatives

Inhibitor

Fragment C

CM Fragment C(Method 1; CMFC 1)

CM Fragment C(Method 2; CMFC 2)

CM Fragment C(Method 3; CMFC 3)

Cysteine content(mol/mol)*

8.7

_

_

2.3

CM-cysteine(mol/mol)

—

12.1

12.8

9.6

Ki(M)

1.7 X 10-'

7.5 X 10-'

8 X 10-'

3 X 10-«

'Determined by subtracting the theoretical value for alanine from the sum of alanine and cysteine obtainedby amino acid analysis.

by guest on July 15, 2018http://hyper.ahajournals.org/

Dow

nloaded from

INHIBITION OF ANGIOTENSIN I CONVERTING ENZYME/Klauser et al. 285

TABLE 3. Inhibilion by Albumin Fragments of the Inacti-valion of Rradykinin by Cultured Human Endolhelial Cells

Concentration(M)

Inhibition

Fragment C 10-"

Reduced and carboxy-methylated Fragment C(CMFC 3) 10-'

49

58

Activity: 9.3 — 14.1 nmoles of bradykinin/hr/10"' cells.

reduced fragments inhibited purified human CE non-competitively, but to a different degree. The calculatedK, values for noncompetitive inhibition of CE byFragment C and CMFC 1, CMFC 2 and CMFC 3 areshown in Table 2. Mild reduction, which leaves onedisulfide bridge intact (Method 3), lowered the K] ofFragment C from 1.7 X 10"8 M to 3 X 10"8 M. Car-boxy methylation of all cysteine residues (Method 1)decreased the inhibition by raising the K! to 7.5 X10-* M.

When Fragment C was completely denatured byurea and by reducing all the disulfide bridges (Method2), the resulting protein inhibited less than FragmentC, and the K, was 8 X 108 M.

The inhibition of Fragment C and CMFC 3 wasnot due to binding of a metal co-factor of human CE,because addition of 2 X 10~* M CoClj did not reversethe inhibition.

Bioassay

Fragment C inhibited the inactivation of bradykininby human lung CE 60% at 7 X 108 M and CMFC 3inhibited 73% at 4 X 10^ M.

Fragment C also inhibited the enzyme on the sur-face of intact suspended cultured human endothelialcells as shown in Table 3. Here also CMFC 3 was abetter inhibitor of the endothelial enzyme than wasFragment C.

Albumin as Substrate of CE

Purified human lung CE (0.01 U) was incubatedwith human serum albumin (4 X 10"* M) and Frag-ment C (3.6 X 108 M) at pH 7.4 for 1 and 6 hours,respectively. After the reaction was stopped by acidprecipitation of proteins there was no increase inninhydrin-positive material over the zero time blank.Similarly, when the reaction mixture was examined bythe amino acid analyzer, no peaks corresponding todipeptides could be detected.

These experiments demonstrated that neitheralbumin nor Fragment C were substrates for CE. Inaddition, the experiments indicate that albumin andFragment C, but not their breakdown products, arethe inhibitors of CE.

Discussion

Various reports have described severe hypotensivereactions in patients, which followed the administra-

tion of plasma protein fractions manufactured byseveral companies."• ""M The reactions are par-ticularly severe in patients with cardiopulmonarybypass. The plasma proteins are prepared by methodswhich include alcohol precipitation and heating over10 hours at 60 o C. M ' u They contain over 83%albumin, plus acetyltryptophan and other additives.The plasma kallikrein-kinin system is believed to beinvolved in the hypotensive reaction since the prepara-tions are frequently contaminated by bradykinin26

and by a prekallikrein activator, Hageman factorfragment.21

Inhibition of CE prolongs the hypotensive effects ofkinins and abolishes the vasoconstrictor effect of reninby blocking the release of angiotensin I I . 8 ' '

Bradykinin is inactivated mainly by two enzymes,kininase I and kininase II. The former enzyme is car-boxypeptidase N, which is mainly responsible forcleaving the peptide in blood plasma, while the latterenzyme is identical with the angiotensin I convertingenzyme.1' M It is present in blood plasma although itsmore important location is on the surface of endo-thelial27 and epithelial cells.8

The low activity of CE in plasma may be due in partto its inhibition by albumin. Because the concentra-tion of albumin is 5 to 8 X 10~* M in plasma,hypothetically it is high enough to inhibit sufficientlymuch of the activity of CE. Furthermore, albuminmay interfere with assaying of CE activity in humanplasma. Although in gel filtration CE exhibits amolecular weight about three times higher thanalbumin, albumin was eluted in the peak containingCE, indicating binding. Albumin present in partiallypurified human CE preparation may account for thelow CE activity of these preparations. Digests ofserum proteins can also inhibit other enzymes that in-activate bradykinin such as pancreatic carboxypep-tidase B,28 a very potent kininase.

Our experiments have shown that albumin and itsfragments inhibit CE noncompetitively, without beingsubstrates of the enzyme. This inhibition may beenhanced by the presence of acetyltryptophan, an ad-ditive in commercial plasma preparations. Thesepreparations inhibit CE more than indicated by theiralbumin content alone. They also potentiate the effectof bradykinin on the isolated rat uterus (ErdSs andRobinson, unpublished data). A convenient way to ob-tain large fragments of proteins is to cleave them withCNBr at the methionine residues. All three fragmentsobtained by this method from human serum albumininhibit CE more than albumin itself. The IM valueswere at least three times lower for the fragments thanfor serum albumin. Fragment C, which consists of theresidues 124-298 of the polypeptide chain of albumin,is of particular interest because it has most of theligand-binding potency of albumin.17 In addition to in-hibiting soluble, purified human lung enzyme, Frag-ment C and its derivative also inhibit bradykininhydrolysis by CE present on the surface of intactcultured human endothelial cells.

Changes in the structure of Fragment C affect its in-hibitory potency as shown by reduction and car-

by guest on July 15, 2018http://hyper.ahajournals.org/

Dow

nloaded from

286 HYPERTENSION VOL 1, No 3, MAY-JUNE 1979

boxymethylation of this fragment. The Kf of CMFC3, which has one disulfide bridge intact, is one sixth ofthat of Fragment C. Another protein, CMFC 1, is theproduct of reaction with a great excess of mercap-toethanol and all of the cysteine residues arederivatized (Method 1). Its K, is one-half of the valuefor underivatized Fragment C. These data suggest thatintroduction of carboxyl groups enhances the in-hibitory properties of Fragment C, particularly whenone intact disulfide bridge preserves the original con-formation of a certain region within the polypeptidechain. As yet this bridge in the peptide chain has notbeen identified. In contrast, when derivatization iscarried out under denaturing conditions, the CMFC 2fragment obtained inhibits less than Fragment Citself. Since the CM-cysteine content of fragments ob-tained with Methods 1 and 2 is identical and CMFC 2has a ten times higher Kf than CMFC 1, preservationof the original secondary structure seems to havegreater influence on the inhibition constant than theincorporation of carboxyl groups.

In conclusion, by inhibiting CE on vascular ert-dothelial cells and in plasma, albumin, its fragments,and other plasma proteins may potentiate thehypotensive effects of bradykinin, which is present inor liberated by infused plasma protein prepara-tions.21- 26 The presence of kinins, prekallikrein ac-tivator and kininase II inhibitors may explain thesevere hypotension encountered in some cases after in-fusion of such preparations, especially when the lungis excluded from the circulation, since the pulmonaryvascular bed is very active in cleaving kinins." Thisstudy of the inhibition of CE by albumin fragmentsshould help to clarify the nature of complexing of theenzyme by albumin and possibly by other plasmaproteins.

Acknowledgments

We are grateful for the advice of Dr. C. M. Herman of theUniversity of Washington, Seattle, Washington, and for theassistance of Tess Stewart and Katy Hammon of the University ofTexas Health Science Center. Endothelial cells were supplied by Dr.A. R. Johnson. Amino acid analysis was done by Dr. J. Capra andMiss P. Frank.

References1. Yang HYT, ErdSs EG, Levin Y: A dipeptidyl carboxypeptidase

that converts angiotensin I and inactivates bradykinin. BiochimBiophys Acta 214: 374, 1970

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

3. Erdos EG, Johnson AR, Boyden NT: Hydrolysis of enkephalinby cultured human endothelial cells and by purified peptidyldipeptidase. Biochem Pharmacol 27: 843, 1978

4. Skeggs LT, Kahn JR, Shumway NP: Preparation and functionof the hypertensin-converting enzyme. J Exp Med 103: 295,1956

5. Yang HYT, Erdos EG: Second kininase in human bloodplasma. Nature 215: 1402, 1967

6. Lanzillo JJ, Fanburg BL: Angiotensin I converting enzymefrom human plasma. Biochemistry 16: 5491, 1977

7. Oshima G, Geese A, Erdos EG: Angiotensin I converting en-zyme of the kidney cortex. Biochim Biophys Acta 350:26, 1974

8. ErdOs EG: The angiotensin I converting enzyme. Fed Proc 36:1760, 1977

9. Oparil S, Katholi R: Angiotensin I conversion. In Renin, Vol.2, edited by Hoirobin DF, Montreal, 1977, p 72

10. Johnson AR, Erdfls EG: Metabolism of vasoactivc peptides byhuman endothelial cells in culture: angiotensin I converting en-zyme (kininase II) and angiotensinase. J Clin Invest 59: 684,1977

11. McMenamy RH, Dintzis HM, Watson F: Cyanogen bromidefragments of human serum albumin. J Clin Invest 246: 4744,1971

12. Meloun B, Moravek L, Kostka V: Complete amino acid se-quence of human serum albumin. FEBS Lett 58: 134, 1975

13. Lowry OH, Rosebrough NJ, Fair AL, Randall RJ: Proteinmeasurement with the Folin phenol reagent. J Biol Chem 193:265, 1951

14. Crestfield AM, Moore S, Stein WH: The preparation and en-zymatic hydrolysis of reduced and S-carboxymethylatedproteins. J Biol Chem 238: 622, 1963

15. Nishimura K, Yoshida N, Hiwada K, Ucda E, Kokubu T:Purification of angiotensin I-converting enzyme from humanlung. Biochim Biophys Acta 483: 398, 1977

16. F.D.A. Bulletin: Adverse reactions to plasma protein fraction.Rockville, Md. 7: 20, 1977

17. Gambhir KK, McMenamy RH: Location of the indole bindingsite in human serum albumin. Characterization of majorcyanogen bromide fragments with respect to affinity labelingpositions. J Biol Chem 248: 1956, 1973

18. SjOdin T, Hansson R, Sjoholm I: Isolation and identification ofa trypsin-resistant fragment of human serum albumin withbilimbin- and drug-binding properties. Biochim Biophys Acta494: 61, 1977

19. Harrison GA, Robinson M, Stacey RV, McCulloch CH, TordaTA, Wright JS: Hypotensive effects of stable plasma proteinsolution (SPPS): a preliminary communication. Med J Aust 2:1040, 1971

20. Bland JHL, Laver MB, Lowenstein E: Vasodilator effect ofcommercial 5% plasma protein fraction solutions. JAMA 224:1721, 1973

21. Alving BM, Hojima Y, Pisano JJ, Mason BL, Buckingham REJr, Mozen MM, Finlayson JS: Hypotension associated withprekallikrein activator (Hageman-factor fragments) in plasmaprotein fraction. N Engl J Med 299: 66, 1978

22. Colman RW: Paradoxical hypotension after volume expansionwith plasma protein fraction. N Engl J Med 299: 97, 1978

23. Mulford DJ, Mealey EH, Welton LD: Preparation of a stablehuman plasma protein solution. J Clin Invest 34: 983, 1955

24. Hink JH Jr, Hidalgo J, Seeberg VP, Johnson FF: Preparationand properties of a heat-treated human plasma protein fraction.Vox Sanguinis 2: 174, 1957

25. Izaka K, Tsutsui E, Mima Y, Hasegawa E: A bradykinin-likesubstance in heat-treated human plasma protein solution.Transfusion 14: 242, 1974

26. Igic R, Erdos EG, Yeh HSJ, SorTells K, Nakajima T:Angiotensin I converting enzyme of the lung. Circ Res 31 (supplII): 11-51, 1972

27. Bakhle YS, Vane JR: Pharmacokinetic function of thepulmonary circulation. Physiol Rev 54: 1007, 1974

28. Hamberg U, Elg P, Stelwagen P: On the mechanism ofbradykinin potcntiation. In Pharmacology of HormonalPolypeptides and Proteins. Advances in Experimental Medicineand Biology, Vol. 2, edited by Back N, Martini L, Paoletti R,New York, Plenum Press, 1968, p 626

by guest on July 15, 2018http://hyper.ahajournals.org/

Dow

nloaded from

R J Klauser, C J Robinson, D V Marinkovic and E G ErdösII) by human serum albumin and its fragments.

Inhibition of human peptidyl dipeptidase (angiotensin I converting enzyme: kininase

Print ISSN: 0194-911X. Online ISSN: 1524-4563 Copyright © 1979 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.1.3.281

1979;1:281-286Hypertension. 

http://hyper.ahajournals.org/content/1/3/281the World Wide Web at:

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

  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 Further information about this process is available in therequested is located, click Request Permissions in the middle column of the Web page under Services.not the Editorial Office. Once the online version of the published article for which permission is being

can be obtained via RightsLink, a service of the Copyright Clearance Center,Hypertensionpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:

by guest on July 15, 2018http://hyper.ahajournals.org/

Dow

nloaded from