THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 251, No. 11, Issue of June 10, pp. 3206-3212, 1976

Printed in U.S.A.

Rat Intestinal Brush Border Membrane Peptidases II. ENZYMATIC PROPERTIES, IMMUNOCHEMISTRY, AND INTERACTIONS WITH LECTINS OF TWO

DIFFERENT-FORMS OF THE ENZYME*

(Received for publication, October 17, 1975)

YOUNG S. KIM, EMMET J. BROPHY, AND J. ALEX NICHOLSON

From the Gastrointestinal Research Laboratory, Veterans Administration Hospital, San Francisco, California 94121, and the Department of Medicine, University of California School of Medicine, San Francisco, California 94143

The properties of two purified peptidases derived from the intestinal brush border membrane of the rat have been investigated. The pH optima, heat stabilities, substrate specificities, and metal ion requirements of the two enzymes and the effects of inhibitors on their activities were nearly identical. The isoenzymes catalyzed the hydrolysis of a wide range of peptides containing from 2 to 8 amino acid residues. The enzymes are aminopeptidases; no evidence for carboxypeptidase or endopeptidase activity was found. For hydrolysis, there appears to be an absolute requirement for an L-amino acid at the NH,-terminus of the peptide substrate. There was a similar but less absolute requirement for the penultimate NH,-terminal amino acid. Thus, although peptides of the type L-aminoacyl-L-proline, L-aminoacyl-L-prolyl-(L-amino acid),, or L-aminoacyl-n-amino acid were not hydrolyzed, L-leucyl-P-

naphthylamide could be utilized as a substrate. The enzymes appeared to be metalloenzymes in that metal ion-chelating agents could inhibit their activities. Co’+ partially restored the activities lost by chelation. Immunodiffusion studies showed that the two enzymes were immunologically identical. The antipeptidase antisera were specific for the enzymes and did not react with other constituents of the intestinal cell. Both enzymes have binding sites for the lectin phytohemagglutinin which recognizes N-acetylgalactosamine residues located at or near the terminal positions of glycoprotein carbohydrate chains. Both the lectin and the antibodies inhibited enzyme activities, but the mechanisms of inhibition appeared to be different.

Previous studies performed with crude or partially purified enzyme preparations have given much useful information

concerning the substrate specificities of the peptidases bound to the intestinal brush border membrane (l-6). A precise description of the kinetics and metal ion requirements of these enzymes, however, is not available since such information has been dependent on the isolation of the enzymes in pure form. In the companion paper we described the purification to a state of homogeneity of two enzymes, designated peptidase F and peptidase S, derived from the brush border membranes of the intestinal epithelium of the rat. The molecular weights and chemical compositions of the two enzymes were found to be

very similar (7). In this paper we compare and contrast the substrate specificities, kinetic analyses, metal ion require- ments, and immunochemical properties of the two enzymes. In addition, some interactions between various sugar-specific lectins and the enzymes, both of which are glycoproteins, are examined.

*This work was supported by Grant AM 17938 from the United States Public Health Service.

3206

EXPERIMENTAL PROCEDURE

Materials-~. phenylalanyl L valyl L. glutaminyl L trypto phyl - L - leucyl L methionyl L asparagyl L threonine, the B chain of bovine insulin, and casein were purchased from Schwa& Mann and bradykinin from Sigma. Other peptides and amino acids were purchased either from Cycle Chemical Corp. or from Sigma. The chemical purity of all peptides and amino acids was established by ion exchange chromatography.

Phenylmethylsulfonyl fluoride (PhCH,S02F),’ L-l-tosylamido- 2-phenylethyl chloromethyl ketone (TosPheCH,Cl), and o-phenan- throline were obtained from Sigma. m-Phenanthroline was a gift from Dr. Richard Levy of Syracuse University. EDTA was purchased from Fischer Scientific Co. and EGTA from Sigma.

o-Phenanthroline and m-phenanthroline were assayed for their chelating abilities and to assess possible contamination of m-phenan- throline by o-phenanthroline. This was determined by measuring a color complex formed between o-phenanthroline and ferrous ions. These studies showed that m-phenanthroline did not form a chelation complex with ferrous cations and contained no detectable o-phenan- throline.

‘The abbreviations used are: PhCH,SO,F, phenylmethylsulfonyl fluoride; TosPheCh,Cl, L-l-tosylamido-2.phenylethyl chloromethyl ketone; EGTA, ethylene glycol-bis (P-aminoethyl ether)l\i,Wtetraa- cetic acid; HOHgPhCOO, p-hydroxymercuribenzoate.

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

Properties of Intestinal Brush Border Membrane Peptidases

The sources of the lectins were as follows. Phytohemagglutinin (type P) was obtained from Difco and “chromatographically homogeneous” wheat germ agglutinin from Calbiochem. Ricinus communis lectin (RCA I) was a gift from Dr. M. Etzler, University of California, Davis. Concanavalin A was purified from jack bean meal by the method of Agrawal and Goldstein (8).

at 37” are illustrated in Fig. 2. The stabilities of the two enzymes are similar and for each enzyme the activity curve is the same for both the dipeptide substrate and the tripeptide substrate.

Methods-Purification of the two peptidases is described in the accompanying paper (7). The studies described in this paper were performed with the-enzymes obtained after the final step in the purification scheme and shown to be homogeneous by analytical isoelectric focusing.

Enzyme Assays-Unless otherwise stated, rates of hydrolysis of the peptide substrates were determined by use of a Beckman 120C amino acid analyzer. The standard assay mixture contained 50 rnM Tris-HCl buffer, pH 8.0, the peptide substrate at various concentrations, and 0.14 pg of enzyme protein in a volume of 200 ~1. The reaction mixture was incubated for 20 min at 37” and the reaction terminated by the addition of 400 ~1 of 7.5 M sulfosalicylic acid. For analysis of the products of the reaction, the mixture was diluted as necessary with sodium citrate (0.2 M, pH 2.2).

For some studies a one-step L-amino acid oxidase assay was employed. Details of this method have been described (9).

In all experiments where rates of hydrolysis were to be measured the conditions of the assays allowed initial rates of hydrolysis to be determined.

Preparation of Peptidase Antisera-Four hundred microliters (100 pg of enzyme protein) of the purified peptidase fraction were emulsi- fied with 3 volumes of Freund’s complete adjuvant. The emulsion was injected subcutaneously into the back of the neck of a New Zealand rabbit. After a period of 2 weeks and again after 4 weeks, 50 kg of enzyme protein were injected. The rabbit was bled 2 weeks after the final injection and the peptidase antiserum separated. To maintain high titers of antipeptidase antibody, 25 fig of enzyme in complete Freund’s adjuvant were injected every other week; rabbits were bled every week.

RESULTS

Substrate Specificit?/-The relative rates of hydrolysis of a variety of peptide substrates by the two peptidases were determined. The results are presented in Table I. In the case of dipeptide hydrolysis, the presence of proline or the D isomer of an amino acid at either the NH, terminus or the COOH terminus inhibits hydrolysis. For other dipeptides, those with NH,-terminal hydrophobic amino acids appear to be the preferred substrates, whereas those with NH,-terminal glycine were poorly hydrolyzed. When the rate of hydrolysis of the series of peptides L-Ala-(Gly), was investigated, the tripeptide was most rapidly hydrolyzed, and the dipeptide least rapidly. Similarly, L-leucyl-diglycine and L-phenylalanyl-diglycine were more rapidly hydrolyzed than the dipeptides, L-leucylgly- tine and L-phenylalanylglycine. The substrate L-leucyl-P- naphthylamide was hydrolyzed by both peptidases. When hippuryl-L-arginine was used as substrate there was no evi- dence of hydrolysis even after 22 hours of incubation, indicat- ing that the enzymes had no carboxypeptidase B activity. No amino acids were released from the large polypeptides bradyki- nin and the B chain of insulin or from casein even after prolonged periods of incubation.

Hydrolysis of an Octapeptide-When the release of amino acids from an octapeptide was followed with time, a sequential pattern of release was observed. As shown in Fig. 3, the NH,-terminal amino acid was released first, followed by the sequential release of other amino acids from the NHs terminus

to the COOH terminus. Properties of Two Enzymes

pH Optimum-The effects of pH on the rate of hydrolysis of L-phenylalanylglycine by the two peptidases F and S are shown in Fig. 1. For both enzymes the rate of hydrolysis was greatest at pH 8.0.

Kinetic Studies-The K, and V,,,,, values for the two peptidases with several peptides as substrates are shown in

TABLE I

Heat Stabilit,y-The effects of preincubation of the enzymes at various temperatures before measurement of their activities

Relative rates of hydrolysis of peptides by the peptidases

The reactions were carried out under the standard conditions described under “Experimental Procedure,” using 5 mM solutions of the indicated substrates except for bradykinin (0.3 mM), insulin (0.3 mM), and casein (4 mg/ml). Hydrolysis of the peptides was measured by the use of an amino acid analyzer as described in the text. When hippuryl-L-arginine, hradykinin, insulin, and casein were the sub- strates, the incubation time was 22 hours. The data shown are for peptidase F. Very similar results were obtained with peptidase S.

PH TEMPERATURE c

FIG. 1 (left). Effect of pH on peptidase activity. The enzyme activity of peptidase F (0-O) and S (O---O) was measured at 37” for 10 min in 50 rnM Tris-HCl between pH 7 and 9 using 5 mM L-phenyl- alanylglycine as substrate. An amino acid analyzer was used to deter- mine enzyme activity. Enzyme units are micromoles of substrate hydrolyzed per 20 min.

FIG. 2 (right). Heat stability of purified peptidases F and S for two substrates (L-Phe-Gly and L-Phe-Gly-Gly). The peptidases were treated for 5 min at various temperatures, between 40 and 75”, rapidly cooled to 4”, and then incubated at 37” with 5 mM-solutions of each substrate in 50 mM Tris-HCl buffer, pH 8.0. Enzyme activity was measured by the L-amino acid oxidase assay (9). O-0, Peptidase F with L-Phe-Gly; O---O, peptidase S with L-Phe-Gly; O---O, peptidase F with L-Phe-Gly-Gly; 0- - -0, peptidase S with L-Phe-Gly- Gly.

3207

Activity related to L-Phe-Gly

= 100” Substrates

200-270 L-Ala(Gly),, L-Phe(Gly), 150-200 L-Ala(Gly), 100-150 L-Ala(Gly),, L-Leu(Gly),, L-Leu-L-Met, L-Leu-L-Val

50-100 L-Ala-Gly, L-Leu-L-Ala, L-Leu-L-Leu 10-50 L-Leu-Gly, Gly-Gly, L-Phe-L-Glu, L-Ala-L-Glu, L-Glu-

L-Ala, L-Arg-L-Phe, L-Met-L-Ser, L-Leu-BNA* I-10 Gly-L-Phe, Gly-L-Leu, L-Phe-L-Asp, L-Asp-L-Phe,

L-Lys-r-Phe, L-Ser.L-Phe, L-Ser-L-Met, L-Val-L-Leu 0 L-Pro-r-Leu, L-Pro-L-Phe, L-Leu-L-Pro, L-Phe-L-Pro,

L-Met-L-Pro; n-Leu-n-Leu, L-Leu-n-Leu, n-Leu-L- Leu, o-Leu(Gly),; hippuryl-L-Arg; bradykinin,’ in- sulin B chain, Casein

“Values are initial rates of release of AH,-terminal amino acids relative to the initial rate of hydrolysis of L-Phe-Gly.

b L-Leu-BNA: L-Leucyl-@-naphthylamide. ’ Bradykinin: L-Arginyl-L-prolyl-~-prolylglycyl- L-phenylalanyl-n-

seryl-L-prolyl-L-phenylalanyl-L-arginine.

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

3208 Properties of Intestinal Brush Border Membrane Peptidases

Table II. For each substrate the K, values of the two peptidases were quite similar, except in the case of L-alanyl-

triglycine, where peptidase S shows a lower K, for this substrate than does peptidase F. When V,,,,. values are examined there is again a similarity between the two pepti-

dases, particularly when the relative rates of hydrolysis of the peptides in the series L-Ala-(Gly), are considered.

Shown in Fig. 4 are double-reciprocal plots of the initial rate of hydrolysis as a function of the concentration of L-leucyl-L- alanine (A) and L-phenylalanylglycine (B). No inhibition of enzyme activity was seen at concentrations of L-leucyl-L-ala- nine as high as 225 mM. In contrast, in the case of L-

phenylalanylglycine, inhibition of enzyme activity occurred when the concentration of the peptide was greater than 5 mM.

Similar substrate inhibition at concentrations of peptide

greater than 5 mM was observed during kinetic studies with L-phenylalanyl-L-isoleucine, L-leucyl-L-methionine, and L-

leucylglycine. The effects of variations in the concentration of one dipep-

tide, L-alanylglycine, on the rates of hydrolysis of two other dipeptides and a tripeptide were investigated. The results of this study are shown in Fig. 5. Hydrolysis of all three peptides was inhibited by L-alanylglycine. In each case the K, was

increased without alteration of the V,,,, hence L-alanylglycine

s /y- - Phe Val Gin-Trp~Leu~Met-Asn-Thr

5l

v, 100 6 LL =I w n u 6 5o 9 5 a Am Thr

2

K z 0 10 20 30 40 50 60 120 180 240 300 a

TIME (minutes)

FIG. 3. Effect of purified peptidase F on hydrolysis of an octapep- tide (~-phenylalanyl-~-valyl-~-glutaminyl-~-tryptophyl-~-leucyl-~- methionyl-L-asparagyl-L-threonine. The peptidase (0.75 pg) was incu- bated with the octapeptide (0.34 rn~ in 50 rn~ Tris-HCl buffer, pH 8.0) at 37” for various time periods. Analysis of the hydrolysis products was performed with an amino acid analyzer.

TABLE II

Kinetic parameters of peptidases F and S

The initial rates of hydrolysis of various peptides by peptidases F and S were determined under standard assay conditions (50 mM Tris-HCl buffer, pH 8.0; 37”; 20.min incubation) at various concentra- tions of the substrates using 0.1 pg of enzyme protein. An amino acid analyzer was used to measure the hydrolysis products as described in the text. The data are the mean values of three experiments.

Substrate Peptidase F Peptidase S

K”, V “Tax Km V “‘XX

mM nmol/min/~g

protein mM nmollminlpg protein

L-Phe-Gly 4.1 37.8 3.0 21.6 L-Leu-L-Ala 0.2 9.2 0.2 7.4 L-Ala-Gly 2.8 29.7 1.8 16.9 L-Ala-(Gly), 6.2 102.0 3.5 74.3 L-Ala-Gly), 4.2 118.8 0.8 72.8 L-Ala-(Gly), 2.1 50.1 1.4 25.9

inhibits hydrolysis of the other three peptides in a competitive manner.

Effects of Metal Ions, Chelating Agents, and Various In- hibitors-Aliquots of the two peptidases were preincubated with the potential inhibitor for 1 hour at 25” prior to assaying for activity with L-phenylalanylglycine as substrate. Aliquots incubated under the same conditions in the absence of the inhibitor served as controls.

As indicated in Table III none of the metal ions tested increased the activity of either peptidase. On the contrary, particularly in the case of peptidase F, there were substantial reductions in the rates of peptide hydrolysis by several metal ions. Zn2+ completely inhibited the activities of both pepti- dases.

Table III also shows the results of the effect of chelating

agents on enzyme activity. During preparation of the purified enzymes the initial homogenization of mucosal tissue was performed in EDTA buffer. Studies performed on tissue homogenized in the presence or absence of 5 InM EDTA indicated that no inhibition of brush border peptidase resulted from this procedure. As shown in Table III, the purified enzymes are not inhibited by this chelator. However, EGTA, a more effective chelating agent than EDTA, produced a moder- ate inhibition of the activities of both peptidases. o-Phenan-

throline, an even stronger chelating agent, produced severe inhibition of the peptidases.

The results of studies performed in an attempt to determine which metal ions might restore activity after treatment with o-phenanthroline are also shown in Table III. After preincuba- tion of the enzyme with the chelating agent, the chloride salt of the metal under study was added and the incubation continued further for 1 hour at 25” before measurement of enzyme activity. Of the metal ions studied, only Co2+ restored enzyme activity and this restoration was partial.

The sulfhydryl binding reagent HOHgPhCOO had no effect on the activities of the purified enzymes. The serine protease

inhibitor phenylmethylsulfonyl fluoride was also without effect on enzyme activities, while the histidine site-specific reagent, TosPheCH,Cl, produced a moderate inhibition of activities.

Immunological Studies

Zmmunodiffusion-The results of immunodiffusion studies

with antisera prepared against the two peptidases are shown in Fig. 6. The two peptidases appear to be immunologically identical. No precipitin line was formed between either anti- serum and the soluble cytoplasmic fraction of the intestinal mucosa or a purified fraction of intestinal goblet cell mucin

(10). Quantitative Immunoprecipitation-The results of quanti-

tative precipitation of peptidases F and S by their respective antisera are shown in Fig. 7. Each peptidase could be com- pletely precipitated by its antiserum.

Interactions of Peptidases with Lectins

Precipitation Reactions-Of the lectins studied, only phy- tohemagglutinin gave a line of precipitation when diffused against the two peptidases on immunodiffusion plates. No precipitin line was formed with concanavalin A, wheat germ agglutinin, or ricin. Studies of the quantitative precipitation of the peptidases by phytohemagglutinin were therefore per- formed. As shown in Fig. 8, incubation of the enzymes with increasing amounts of the lectin was associated with increased losses of enzyme activity from the supernatant until a plateau

was reached at 0.1 mg of phytohemagglutinin. As the amount

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

Properties of Intestinal Brush Border Membrane Peptidases 3209

FIG. 4. Double-reciprocal plots of ini- tial velocity of hydrolysis uers’sus sub- strate concentration for two peptides, L-leucyl-L-alanine (A) and L-

phenylalanylglycine (B). Peptidase F (0.08 fig of protein) was assayed in 50 mM Tris-HCl buffer, pH 8.0, at 37O. Products of hydrolysis were measured with an amino acid analyzer.

FIG. 5. Competition between L-

alanylglycine and r.-leucyl-L-alanine (L- Leu-L-Ala), L-phenylalanylglycine (L- Phe-Gly), and L-phenylalanylglycylgly- tine (L-Phe-Gly-Gly) for peptidase F. Various concentrations of the latter three peptides were assayed in the pres- ence of fixed concentrations of L- alanylglycine. The three graphs are dou- ble-reciprocal plots of the initial veloci- ties of hydrolysis LX~.S’SUS substrate con- centration for the three peptides in the presence of four concentrations of L- alanylglycine: a = 0 mM, b = 0.5 mhr, c = 1.0 mM, d = 5.0 mM. Enzyme protein, 0.2 /~g, was incubated at 37” in 50 mM Tris-HCl buffer, pH 8.0. Activity was measured by the L-amino acid oxidase assay (9).

4 L-Leu-L-Ala

$ (mM)-’

TABLE III

Effect of metal ions, chelating agents, and other substances on activity of peptidases F and S

The potential inhibitor was preincubated with aliquots of the enzymes as described in the text. Enzyme activity was then deter- mined under the standard assay conditions described in the text (50 mM Tris-HCl buffer, pH 8.0; 37”; 20-min incubation) with L-

phenylalanylglycine (5 mM) as substrate. The hydrolysis products were quantitated by means of an amino acid analyzer.

Substances added

None CoCl, (1 mM) MnCl, (1 mM) MgCl, (1 mM) HgCl, (1 mM) ZnCl, (1 mM) EDTA (1 mM) EGTA (1 mM) o-Phenanthro’ke (1 mM) Al o-Phenanthroline (1 mru)

and MnCl, (5 mM)

o-Phenanthroline (1 mM) and MgCl, (5 mM)

o-Phenanthroline (1 mM) and CoCl, (5 mM)

m-Phenanthroline (1 mM) Cysteine (1 msi) p-Hydroxymercuribenzoate

(0.5 rnM)

Phenylmethylsulfonyl fluoride (1 mM)

TosPhe CH,Cl (1 mM)

Percentage of activity

Peptidase F Peptidase S

100 100 34 88 54 98 65 110 10 24 0 0

99 102 52 65

6 5 5 6

3 4

73 59

78 68 95 94 99 98

107 100

12 65

I-Phe-Gly d

c

b 0

I

0.5 1

f (I-VIM)“ $ (mM)-’

FIG. 6. Immunodiffusion of peptidase antisera against purified peptidases. A purified intestinal mucin (10) and the soluble cytoplas- mic fraction of the intestinal mucosa. The center wells contained either peptidase F antisera (Anti-F) or peptidase S antisera (Anti-S). Peripheral wells contained peptidase F (2.5 pg of protein), peptidase S (2.5 pg of protein), the purified mucin, and the soluble cytoplasmic fraction (cytosol).

of added lectin was increased above 0.8 mg, progressively more enzyme activity appeared in the supernatant. There was no detectable peptidase activity in the phytohemagglutinin prep- aration used in these studies. A greater percentage of peptidase S could be precipitated by the lectin than of peptidase F.

Effects of Phytohemagglutinin and of Peptidase Antisera on Peptidase Activities-Varying amounts of phytohemagglutinin were incubated with aliquots of the enzyme and the uncen- trifuged mixtures were assayed for enzyme activity. As shown in Table IV, maximal inhibitions of 30% with peptidase F and 40% with peptidase S were found.

In another study, several peptide substrates were examined for inhibition of peptidase activities by the lectin or by the antisera; the results are shown in Table V for concentrations of

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

3210 Properties of Intestinal Brush Border Membrane Peptidases

si 22 I *-)-)------* _____ ------*

6 5 z ? g 1 ++-------+------ ----- ~ 1

0

AMOUNT 0:” ANTISERA (pi) 100

FIG. 7. Quantitative precipitation of purified peptidases F and S by antipeptidase antisera. Rabbit antipeptidase F antisera or rabbit antipeptidase S, 5 to 100 ~1, was allowed to react with their respective peptidase (1 pg of protein) 1 hour at 37” and then overnight at 4” in a volume of 200 ~1. The tubes were centrifuged at 1100 x g for 15 min and the enzyme activity was measured in the supernatant fraction. A control incubation of enzymes with nonimmune rabbit serum was also carried out. The enzyme assay was performed by the L-amino acid oxidase method (9) under standard assay conditions (50 mM Tris-HCl buffer, pH 8.0; 37”) with 5 rnM L-phenylalanylglycine as substrate. O-0, Antisera F and peptidase F; O-0, antisera S and peptidase S; O---O, nonimmunized rabbit sera and peptidase F; 0- - -0, nonimmunized rabbit sera and peptidase S.

> e > i= 100 U Q

2

2 50 u.l

PHYTOHEMAGGLUTININ (mg)

FIG. 8. Quantitative precipitation of purified peptidases F and S by phytohemagglutinin. Phytohemagglutinin (13 pg to 1.25 mg) was reacted with peptidases F (11.8 Kg of protein) and S (11.8wg of protein) in a final volume of 200 ~1 at 37” for 20 min and then at 25” for 3 hours. The tubes were centrifuged at 43,000 x g for 20 min, and the supernatant fraction was assayed for enzyme activity using the L-amino acid oxidase method (9) under standard assay conditions (50 mM Tris-HCl buffer, pH 8.0; 37”) with 5 mM L-phenylalanylglycine as substrate. 04, Peptidase F and phytohemagglutinin; O---O, peptidase S and phytohemagglutinin.

lectin or antiserum previously shown to result in a 50% precipitation of the enzymes using L-phenylalanylglycine as substrate. For each agglutinin a similar inhibition of peptidase activity occurs irrespective of the substrate used.

DISCUSSION

Previous studies of the peptidase activity associated with the intestinal brush border of both rat and hamster had suggested the presence of more than one peptidase in this membrane

TABLE IV

Effect ofuarious concentrations ofphytohemagglutinin (PHA) on rate of hydrolysis of L-Phe-Gly bypeptidase F or S

Varying amounts of PHA were mixed with 1.18 pg each of peptidases F or S in 200 ~1 of Tris-HCl, pH 8.0 for 30 min at 37” and then allowed to stand for 3 hours at 25”. Enzyme assays were then performed under standard assay conditions (50 mM Tris-HCl buffer, pH 8.0; 37”; 20-min incubation) with 5 rnM L-phenylalanylglycine as substrate, using the L-amino acid oxidase method (9).

Amount ofPHA added

Pi?

None 4

22 73

146 218 327 437

Per cent inhihition of enzyme activity

Peptidase F Peptidase S

0 0 14 20 20 25 23 30 30 42 28 39 31 39 30 41

TABLE V

Effects of rabbit antipeptidase antisera and of phytohemagglutinin (PHA) on rate of hydrolysis of variouspeptides bypeptidases F and S

An aqueous solution (80 ~1) containing 146 pg of PHA or 80 ~1 each of rabbit antipeptidase F or S antisera at a concentration known to produce 50% precipitation of the enzyme were mixed with 0.59 pg each of peptidases F or S in 250 ~1 of Tris-HCl, pH 8.0, at 37” for 30 min and then allowed to stand for 3 hours at 25”. Enzyme assays were then performed under standard conditions (50 rnM Tris-HCl buffer, pH 8.0, 37”; 20.min incubation) with 5-mM solutions of each substrate using the L-amino acid oxidase method (9).”

Per cent inhibition of enzyme activity

Substrate

L-Phe-Gly L-Phe-(Gly), Gly-r-Phe L-Leu-Gly L-Leu-(Gly),

Peptidase F Peptidase S

AntI-Fb PHA Anti-S PHA

65 33 76 26 69 30 71 32 60 38 72 28 65 22 74 16 67 28 80 32

a The addition of 146 pg of PHA to the incubation mixtures did not affect the L-amino acid oxidase method.

‘Anti-F and anti-S: rabbit antipeptidase F and antipeptidase S antisera.

(2-4, 11). This has been confirmed by the isolation of two distinct enzymes, both with peptidase activity, from the rat intestinal brush border (7).

The physical and chemical properties of the two enzymes were found to be very similar. The present studies indicate that they are also similar with respect to several additional proper-

ties. Both enzymes catalyze the hydrolysis of several dipep- tides and of L-leucyl-P-naphthylamide, but. do not hydrolyze dipeptides containing proline or n-amino acids at either the COOH-terminus or the NH,-terminus. The specificity of the purified enzymes is not restricted to dipeptides; tripeptides, a tetrapeptide, a pentapeptide, and an octapeptide were all readily hydrolyzed. At equimolar concentrations of substrate the larger peptides were hydrolyzed more rapidly than were dipeptides. When kinetic analyses were performed the V,,, values obtained with tripeptide and tetrapeptide substrates were approximately 4 times greater than that obtained for the

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

Properties of Intestinal Brush Border Membrane Peptidases 3211

dipeptide with the same NH,-terminal amino acid. Competi- tion studies suggested that the same active site was involved in the hydrolysis of dipeptides and tripeptides. Studies with an octapeptide showed that both enzymes act only at the NH,-ter- minus of this polypeptide, and analysis of the products of tri-, tetra-, and pentapeptides gave similar evidence of NH,- terminal specificity.

These results are in agreement with previously published data obtained from studies with intact rat brush borders and with heterogeneous enzyme systems solubilized from purified rat brush borders (l-6). It cannot be assumed, however, that the two purified peptidases are the only enzymes with pepti- dase activity bound to the membrane. Fujita et al. (2) found evidence for an endopeptidase in intact brush borders which catalyzed the hydrolysis of the B chain of insulin with release of a mixture of amino acids and peptides. In the present study neither of the two purified enzymes could hydrolyze the B chain of insulin, bradykinin, or casein. It is possible, bowever, that the endopeptidase activity found in the intact brush border membrane might not be intrinsic to the membrane but might be due to adsorbed pancreatic enzymes (12).

In addition to their apparently identical substrate specifici- ties, the two enzymes have several other properties in common. They have similar pH activity curves and thermostabilities. While both enzymes were not inhibited by sulfhydryl binding reagents, their activities were reduced by some metal ion- chelating agents. Although no increase in activity was’observed when Co2+ was added to the native enzymes, addition of cobaltous ions partially restored the loss of enzyme activities produced by incubation with o-phenanthroline. It is of interest that m-phenanthroline, an isomer of o-phenanthroline, but which is not a chelating agent is capable of a modest inhibition of the activities of the two peptidases. Thus, part of the peptidase inhibition produced by o-phenanthroline may be unrelated to its chelating property. This may explain the less than total restoration of activity by Co’+ in the presence of o-phenanthroline.

The double immunodiffusion studies further emphasize the close relationship between the two purified enzymes in that the enzymes appear to be immunologically identical. No immuno- logical cross-reactivity was found between the enzymes and the soluble cytoplasmic fraction of the intestine or with a purified fraction of intestinal goblet cell mucin. Further evidence for the immunospecificity of the antipeptidase antisera has subse- quently been obtained. In studies to be reported, several other glycoprotein brush border enzymes (sucrase, maltase, and alkaline phosphatase) did not react with the peptidase antisera.2

The effect on peptidase activity was measured in the presence of phytohemagglutinin and antisera to the enzymes. In the case of lectin inhibition, when the concentration of phytohemagglutinin was increased lo-fold over the amount required for precipitation (Fig. 7) the enzyme activity began to reappear in the supernatant. At higher lectin concentrations nearly complete peptidase activity could be observed. Thus, although the lectin was probably saturating the binding sites on the enzyme, the enzyme was not markedly inhibited. In contrast, although not shown, when the antipeptidase antisera was added in concentrations up to loo-fold over that required for 90% precipitation, the enzymes remained 95% inhibited. These findings suggest that the inhibition at low lectin and

‘Y. S. Kim, and Y. W. Kim, unpublished experiments.

antisera concentrations is accomplished by two mechanisms. In the case of the lectin it is likely that aggregation may prevent adequate accessibility of substrate to the active site of the enzyme, since binding of lectin alone does not inhibit the enzyme activity. However, even at high concentration of antibody, when presumably no aggregation is occurring but the antigenic sites are saturated with antibody, enzyme activity is not regained. Thus, binding alone between antibody and enzyme in some way blocks enzyme activity. These findings suggest that the binding site of the lectin is topographically remote from the active site while the antibody combining site may be nearer the active center.

The partial precipitation and inhibition of the enzyme activities by phytohemagglutinin is of interest. The agglutinat- ing property of this lectin is inhibited by N-acetylgalactosa- mine and it is therefore inferred that this lectin recognizes and binds to N-acetylgalactosamine residues of glycoproteins and particularly to residues of this hexosamine which are located at or near the terminal, nonreducing positions on the carbohy- drate chains of a glycoprotein. It is therefore possible that the greater precipitation of peptidase S than of peptidase F by this lectin is a correlative of the greater molar concentration of N-acetylgalactosamine found in peptidase S or a function of the accessibility of this sugar residue in the two enzymes.

The two isolated enzymes are so similar in their properties that they may properly be regarded as isoenzymes. It is of interest that two peptidase isoenzymes have previously been isolated from a particulate fraction of the rat kidney (13). Histochemical evidence suggested that the kidney peptidases were derived from the brush border of the proximal tubules. There is a close similarity between these rat kidney brush border enzymes and the intestinal brush border enzymes characterized in the present study. Although the physical and chemical properties of the kidney peptidases were not estab- lished, their substrate specificities and metal ion requirements and the effects of inhibitors on their activities were qualita- tively similar, with respect to these parameters, to the intesti- nal peptidases. We have previously shown, however, that the electrophoretic mobilities of kidney plasma membrane pepti- dases differ considerably from those of the intestinal brush border peptidases (4). It will be of interest to determine whether the kidney brush border peptidases are also immuno- logically and chemically distinct from the intestinal brush border peptidases.

The existence of peptidase isoenzymes in the mammalian intestinal brush border membrane has not been unequivocally demonstrated hitherto. It is not known whether this phenome- non occurs only in the rat and not in other mammals. Only single enzymes with peptidase activity have been isolated from hog and rabbit brush borders (14, 15) and no information is available concerning the number of peptidases in the human intestinal brush border membrane.

The biological implications of the existence of more than one peptidase in the intestinal brush border are uncertain. In contrast to the clinical syndromes associated with congenital deficiencies of various intestinal brush border disaccharidases, no congenital disease due to intestinal brush border peptidase deficiency has as yet been described. It is possible that there are two or more peptidases in the human intestinal brush border which have overlapping substrate specificities but which are under separate genetic controls. This might be the reason why a congenital absence of intestinal brush border peptidase activity rarely if ever occurs.

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

3212 Properties of Intestinal Brush

Acknowledgments-We would like to thank the following for their generous gifts of compounds used in this study: Ricinus communis lectin (RCA I), Dr. M. Etzler; m-phenanthroline, Dr. Richard Levy. We are indebted to Dr. James Whitehead for critically reviewing the manuscript. We gratefully acknowledge

the secretarial aid of Ms. Linda Luhman.

REFERENCES

1. Rhodes, J. B., Eicholz, A., and Crane, R. K. (1967) Biochim. Biophys. Acta 135, 959-965

2. Fujita, M., Parsons, D. S., and Wojnarowska, F. (1972) J. Physiol. (London) 227, 377-394

3. Auricchio, S., Pierro, M., Generoso, A., and DeRitis, G. (1972) Biochim. Biophys. Acta 274, 420-425

4. Kim, Y. S., Birtwhistle, W., and Kim, Y. W. (1972) J. Clin. hat.

Border Membrane Peptidases

5. 6.

7. 8.

9.

10.

11. 12. 13.

14.

15.

51, 1419-1430 Peters, T. J. (1973) Clin. Sci. Mol. Med. 45, 803-816 Kim. Y. S.. Kim. Y. W.. and Sleiseneer. M. H. (1974) Biochim.

Biophys.‘Acta 370, 283-296 ~’ Kim, Y. S., and Broohv, E. (1976) J. Biol. Chem. 251.3199-3205 Agrawal, B. B. L., and Goldstein, I. J. (1967) Biochim. Biophys.

Acta 133, 376-379 Nicholson, J. A., and Kim, Y. S. (1975) Anal. Biochem. 63,

110-117 Bella, A., and Kim, Y. S. (1972) Arch. Biochem. Biophys. 150,

679-689 Eichholz, A. (1968) Biochim. Biophys. Acta 163,101-107 Woodlev, J. F., and Kenny, A. J. (1969) &o&em. J. 115, 18-19P Felgenhauer, K., and Glenner, G. G. (1966) J. H&o&en. Cyto-

them. 14, 401-413 Maroux, S., Louvard, D., and Baratti, J. (1973) Biochim. Biophys.

Acta 321, 282-295 Takesue, Y. (1975) J. Biochem. (Tokyo) 77, 103-115

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

Y S Kim, E J Brophy and J A Nicholsonenzyme.

immunochemistry, and interactions with lectins of two different forms of the Rat intestinal brush border membrane peptidases. II. Enzymatic properties,

1976, 251:3206-3212.J. Biol. Chem. 

  http://www.jbc.org/content/251/11/3206Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/251/11/3206.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from