Degradation of thymopentin by human lymphocytes: evidence for aminopeptidase activity

Biochimica et Biophysica Acta, 955 (1988) 164-174

164

Elsevier BBA 33156

Degradation of thymopentin by human iymphocytes: evidence for aminopepfidase activity Andrew A. Amoscato a,b, A. Balasubramaniam a, j. Wesley Alexander a,b and George F. P~abcock a,b a Department of Surgery, The University of Cincinnati College of Medicine and b the Shriners Burns Institute, Cincinnati, OH (U.S.A.)

(Received 11 January 1988)

Key words: Thymopentin; Lymphocyte; Aminopeptidase; (Human)

Thymopentin (Arg-Lys.Asp-Val-Tyr) was shown to be degraded in vitro by human lymphocytes into two main fragments; the tetrapeptide Lys.Asp-Val-Tyr and the tripeptide Asp-Vai-Tyr. Degradation products were identified by HPLC and amino-acid analysis. Analysis of the time-course of degradation revealed a 'stepwise' degmdative event beginning at the N-terminal. The degradation of thymopentin after the first |0 ~nin, as well as the formation of the tetrapoptide (5-30 rain) were essentially curvi|inear. Degradation of the tripeptide, was linear. Upon screening a panel of compounds that inhibit enzymatic activity, bestatin, amastatin and 1,10-phenanthroline were shown to be the most effective. Bestatin and amastatin caused an 85-90~ inhibition of thymopentin degrading activity with |Cso values of 7.1 • 10 -6 M and 4.5 • 10 -9 M, respectively. 1,10-Phenan~hroline completely inhibited the dewadative process with an ICso of 2 • 10-4 M. When the tetrapeptida Lys-Asp-Vai-Tyr was used as the starting substrate, similar ICso values were seen for amastatin, bestatin and 1,10-phenanthmline. The importance of divalent metal ions in the degradative event was demonstrated not only by the effect of i,10-phenanthroline, but also by the ability of Zn2+ and CoZ+ to reverse the inhibition of 1,10-pbenanthroline (at its ICso) to activities near control values (no inhibitor). These data strongly suggest that an aminopeptidase(s) is responsible for the degradative activity.

lntrodaetion The immune system consists of a complex array of cell types whose functions depend upon the proper orchestration of extracellular, intercellular Abbreviations: PMSF, phenylmethylsulfonyl fluoride; TLCK, N-p-tosyI-L-lysine chlorom,:~yl ketone; t-Boc, tert-butyloxy. carbonyl; PAM, phenylacetamidomethyl; Tos, tosyl; 2CIZ, chlorobenzyloxycarbonyl; OBzl, benzyl ester; 2BrZ, bromobenzyloxycarbonyl; RT, retention time. Correspondence: Andrew A. Amoscato, Department of Surgery, University of Cincinnati Medical Center, 231 Bethesda Avenue, Cincinnati, OH 45267-0558, U.S.A.

and intracellular signals. Once evoked, these signals (i.e., peptide, hormones, lymphokines, etc.) rely on properly functioning mechanisms (specific or non-s:r,ecific) to effectively truncate or terminate their action. One such specific mechanism may involve receptor-mediated endocytosis, resulting in the degradation of ligand and/or receptor, thus achieving the appropriate state of desensitization [1-4]. Nc,~i-specific mechanisms, on the other hand, have received somewhat less attention. These mechanisms include the roles that cellular ectopeptidases play in hydrolyzing regulatory peptides at the surface of a variety of cell types. These include endopeptidases, aminopeptidases, carb-

0167-4838/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)

165

oxypeptidases, as well as dipeptidases [5]. These enzymatic activities are classified as non-specific terminating mechanisms for two reasons: (1) many of these enzymes will act on a variety of peptides and (2) the occurrence of some of these enzymes is fairly widespread and possibly ubiquitous [5]. Basic to the understanding of cellular activation and regulation is the delineation of the specific regulatory molecules involved. A further extension of this goal lies in the isolation, characterization and synthesis of the 'active portion' of these molecules, thus allowing greater manipulation of the molecules for basic studies on their mechanisms of action, as well as the potential for less harmful side effects in the clinical situation [6,7]. One exemplary peptide is thymopentin. Thymopentin (LArg-LLys-LAsp-LVaI-LTyr) is an immunomodulating pentapeptide and represents the 'active site' (residues 32-36) of thymopoietin, a 49 amino-acid polypeptide hormone secreted by the thymic epithelium [8-10]. The active-site sequence has been co,,',f~med in both bovine and human thymopoietins [8-11]. The biological activities of thymopentin include its effect on T-cell differentiation and immune regulation of mature T-cells, with the latter effect being mediated via changes in intracellular cyclic GMP (cGMP) concentrations [12-20]. As a result, thymopentin has been used in numerous clinical studies as a therapeutic agent against diseases with known or postulated immune dysfunctions. These studies, however, were only partially successful [21]. In dealing with small biologically active synthetic peptides, the achievement of a more stable product is usually desirable in order to avoid unwanted degradation and inactivation. Degradation of thymopentin has been shown to occur in plasma [22] and its potency was shown to vary greatly depending on the route and rate of administration [23]. However, no evidence for thymopentin degradation by intact cells has been presented, although it has been suggested and implied from various biological/biochemical assays performed by ourselves and others [24]. As a result, we assayed for thymopentin degradation products in cellular supernatants to determine whether degradation occurred, and screened a panel of enzyme inhibitors for potential inhibition, which in turn suggested the type of enzyme(s) involved.

The data support the hypothesis that thymopentin is degraded by human lymphocytes, with inhibition studies suggesting arrfinopeptidase activity. Materials and Methofis

Reagents A portion of the experiments employed synthetic thymopentin that was kindly provided by Dr. G. Goldstein (Ortho Pharmaceutical Corp., garitan, N J). The remainder of the experiments employed thymopentin that was synthesized in our lab. Dulbecco's phosphate-buffered saline and fetal calf serum were obtained from Gibco (Grand Island, NY). Dith_iothreitol was obtained through Bio-Rad (Riclunond, CA) and 2-mercaptoethanol was purchased from Mall/ncrodt (St. Louis, MO). The rern~firfing compounds used for enzyme inhibition studies were obtained from Sigma Chemical Co. (St. Louis, MO). Dextran I"-500 and Ficoll-Paque were purchased from Pharmacia (Piscataway, N J). Preservative-free heparin was obtained through Squibb (Cherry Hill, N J). The resin used in peptide synthesis was purchased from Applied Biosystems (Foster City, CA). Protected amino acids were obtained from Peninsula Labs (Belmont, CA). The dipeptide, Val-Tyr, and amino acid, tyrosine, was obtained from U.S. Biochemical Corp. (Cleveland, OH). Solvents used in the peptide synthesis were obtained from Burdick and Jackson (Muskegon, MI). Chemicals required for peptide synthesis were purchased through Chemical Dynamics (South Plainfield, N J). All of the other remaining chemicals were purchased through Fisher Chemical Co. The water used for all experiments was type I grade with a resistance of 15 Mfi/cm or better.

Peptide synthesis The pentapeptide was synthesized by the standard solid phase method [25] according to the previously described procedure [26]. Briefly tBoc-tyrosine (2BrZ)-PAM resin (2 g, 0.65 retool/g) was taken in the reaction vessel of a Vega Model 50 Peptide synthesizer and the peptide was synthesized by coupling the protected amino-acid derivatives sequentially. The protection scheme for the pentapeptide was as follows: Boc-Arg(Tos)Lys(2C1-Z)-Asp(OBzl)-VaI-Tyr(2Br-Z)-PAM-Resin.

166

Coupling was effected by the preformed symmetrical anhydride [27]. In the case of Boc-Arg (TOO, however, it was effected by the standard dicyclohexylc~bodiimide method to avoid side reactions [28]. Completion of coupling was determined by the qualitative ninhydrin test [29]. During the synthesis, aliquots of resin (0.5 g) containing the desired intermediary sequences were removed. The peptide was deprotected and released from the resin (0.5 g) by treatment with hydrogen fluoride gas (HF, 5 ml, 1 h at 0 ° C) containing 5~ p-cresol. All peptides were then purified by HPLC as shown in Fig. 1. Positive identification of the pentapeptide and its fragments was determined by comparing their respective retention times (see Table I) to those reported in the literature. In addition, amino-acid analysis for the pemapeptide and its fragments yielded the following ratios: pentapeptide, Arg (1.01), Lys (1), Asp (0.99), Val (0.90), Tyr (0.90); tetrapeptide, Lys (0.98), Asp (1), Val (0.96), Tyr (0.93); and tripeptide, Asp (1), Val (0.97), Tyr (0.95). For the pentapeptide, the HPLC and amino-acid analysis data was essentially identical to that of the pentapeptide supplied by the Ortho Chemical Co.

Cell preparation Heparinized venous blood was obtained from healthy donors, then mixed with 6~ Dextran T-500 (I mi Dextran/10 ml blood) and allowed to sediment for 60--90 min at 37°C. The upper fraction was washed in Dulbecco's phosphate-buffered s ~ n e and centrifuged on Ficoll-Paque according to Boyum [30]. The mononuclear cells were collected and the monocytes were removed by their characteristic adherence to p~astic. Cells were washed and resuspended in phosphate buffer consisting of the fol!~wing: NaCI (140 mM), Na2HPO47H20 (9 raM), NaH2PO4 (9 mM), KCI (3 mM), KH2PO4 (1.5 mM) (assay buffer, pH 7.4). Buffers contained 1 mM CaCI 2 unless stated otherwise. Cells were adjusted to a concentration of I . 107 per nd and viability was usually greater than 97~ as determined by trypan blue exclusion. Purity of lymphocyte preparations was typically greater than 90~ as determined by morphology and by fluorescent antibody staining in conjunction with flow cytometry.

Cell disruption Lymphocytes were adjusted to a concentration of 1.107 cells per ml in assay buffer without calcium containing 2.5 mM MgCI 2 for organdie stabilization and were placed in the cell disruption bomb (Parr Instrument Co., Moline, IL) at 4 ° C. Cells were equilibrated at 25 atm N 2 for 20 rain. After release from the cavitation bomb, the suspension was used for ligand degradation assay for certain experiments. A portion of the suspension was sonicated once (15 s at 4°C, setting 2, Heat Systems Co., Melville, NY) and the resulting suspension used for further assay.

Ligand degradation assay Lymphocytes (5.106 cells/ml) were incubated at 37 °C in assay buffer for 30 min with or without inhibitors at various conee~trations in 1.5 ml polypropylene conical centrifuge tubes. Assay buffer contained 1 mM Ca 2+. In those cases where chelating agents w e r e 11sed~ C~ 2+ was omitted. After this pre-hlcubation period, 37 nrnol thymopentin was added and the iacubation continued for an additional 60 min (total volume, 500/~1). Lymphocytes were rapidly removed from the suspension by centrifugation in a microfuge for 10 s, (Fisher Model 235 B). A 400 pl aliquot of supernatant was immediately analyzed for degradation products by high-performance liquid chromatography Experiments were run in duplicate or triplicate. Each experimental point is the average of three determinations.

High-performance liquid chromatography (HPLC) High-performance liquid chromatography was performed on a Waters HPLC gradient system. Separations were carried out on a Waters Cls reversed-phase analytical column (3.9 nun x 30 cm, 10 m packing). An isocratic solvent system was used and consisted of 0.08 M triethylammonium phosphate: acetonitrile (96 : 4, v/v), pH 4.0 as previously described [31]. The flow rate was 1.0 ml/min and detection was routinely monitored at 280 nm at a range of 0.1 absorbance units full scale. Fractions containing thymopentin degradation products were quantitated by HPLC, collected, lyophilized and analyzed for their amino-acid composition. In all cases, greater than 90% of the sample applied was recovered as intact

167

peptide or fragments, as determined by comparison with chromatographic separations of synthetic peptide fragments quantitated by HFLC and amino-acid analysis.

TABLE I

Amino-acid analysis

Synthetic fragments

Peak

RT (rain) s~ar=datds

RT (rain) degradation products

Arg-Lys-Asp-Val-Tyr Lys-Asp-Val-Tyr Asp-Val-Tyr VaI-Tyr Tyr

c d e b a

7.9 + 0.10 10.14.0.13 18.8 4.0.17 6.04.0.10 4.34-0.19

8.0 ~ 0.12 10.4:t:0.15 19.1 +0.20 - a -

Samples were concentrated, dissolved in 1.0 ml of constant boding HCI, and hydrolyzed in vacuo at 105 °C for 2~ h. The acid was removed in vacuo and the samples diluted in sodium citrate buffer prior to analysi~ ~mino-acid analysis was performed on a Beckman model 6300 amino-acid analyzer. Amino acids were separated on an ion-exchange resin and eluted with sodium citrate buffers. Results

HPLC: Characterization of thymopentin degradation products In order to determine whether degradation of thymopentin occurred, cellular supernatants incubated with the pentapeptide were analyzed by HPLC. The peaks eluted were compared to a series of tyrosine containing synthesized fragments of thymopentin (Fig. 1). The fragments and their retention times are show~ in Table I. These times correlate well with those previously reported [31]. Fig. 2 depicts a chromatographic separation of thymopentin and its degradation products after incubation with 5.106 lymphocytes for 60 m/n at I

I

d

I

RETENTION TIMES OF SYNTHETIC THYMOPENTIN FRAGMENTS AND DEGRADATION PRODUCTS FROM A 60 MIN INCUBATION WITH LYMPHOCYTES

= These degradation products were either not detected or in too small a quantity for accurate determination above baseline noise.

37°C. In addition to the thymopentin peak (c), two major fragments (peaks d and e) are evident. The retention times for peaks c, d and e are virtually identical with standards for thymopentir~, Lys-Asp-Val-Tyr and Asp-Val°Tyr, respectively (Table I). These peaks are not evident upon HPLC analysis of supernatants from lymphocytes incubated in the absence of thymopentin (not shown). In addition, no cleavage of thymopentin was evident upon incubation with buffer alone for 60 rain at 37°C (not shown). Further evidence that peaks c, d and e (Fig. 2) represent thymopentin and its degradation products is given by analyzing the am/no-acid content of the products isolated by HPLC. Table II displays these results. Peak c contains typical uni-

I

I

I

I

1

C d

t

t

0

<

c

2

Y I

i

I

I

2O

15

10

5

Retention Time (mini

Fig. 1. Reversed-phase HPLC of synthetic thymopentin and its fragments. The peak identification is as follows: a, Tyr; b, Val-Tyr; c, Arg-Lys-Asp-Val-Tyr (thyrnopentin); d, Lys-AspVal-Tyr; and e, Asp-Val-Tyr. Conditions for separation are described in Materials and Methods.

i~

e <

I

i

i

I

20

15

10

5

Retention Time (min)

Fig. 2. Degradation of thymopentin by human lymphocytes. Thymopentin (37 nmol) was incubated with 5.10 6 lymphocytes for 60 rain at 37 ° C in assay buffer. At the endpoint of the assay, cells were centrifuged and an aliquot of supernatant was analyzed by HDLC for degradation products as described . l Materials and Methods.

168 TABLE II AMINO-ACID ANALYSIS OF GRADATION PRODUCTS Peak

c (thymopentin) d •

THYMOPENTIN

DE-

Amino-acid analysis ratio Arg

Lys

Asp

Val

Tyr

1.02 -

1 0.99 -

0.97 1 1

0.91 0.98 0.99

0.90 1.08 1.05

molecular ratios of all the amino acids present in thymopentin. Likewise, peak d and e have unimolecular ratios representative of the tetrapeptide Lys-Asp-Val-Tyr and tripeptide Asp-Val-Tyr, respectively. Thus, these data, in conjunction with the retention times from HPLC, are evidence for two major degradation products of thymopentin effected by its interaction with intact lymphocytes. Arginine and Arg-Lys cleavage products are not seen due to their lack of absorbance at 280 run. The possibility exists that the enzyme(s) responsible for thymopentin's degradation are secreted by resting lymphocytes or prompted by thymopentin's interaction with the lymphocyte cell surface. With regard to the former possibility, thymopentin (37 nmol) was incubated in assay buffer for 60 rain at 37°C that was pre-conditioned for 60 rain (37 ° C) with lymphocytes alone. Only very small amounts of Lys-Asp-Val-Tyr (4.9~, 1.8 nmol) and Asp-Val-Tyr (4.6%, 1.7 nmol) were detected by HPLC, as compared to those degradation products released in the presence of lymphceytes (34 and 195, respectively). However, no degradation was evident if the pre-conditioned buffer was heated to 100 °C for 2 rain. Although small amounts of enzyme(s) may be secreted, the presence of a small percentage of damaged cells in the suspension could account for the degradation. This is supported by the fact that N2 bomb cavirated lymphocytes and N 2 cavitated/sonicated lymphocyte preparations contain only Val-Tyr and Tyr degradation products after incubation with thymopentin for 60 rain at 37 ° C. In dealing with the question of whether the interaction of thymopentin with lymphocytes causes secretion of enzymes into the medium, the following experiment was designed. Lymphocytes were incubated with thymopentin (37 nmol), for 20 rain at 37°C in the standard assay buffer. In

one set of tubes, the cells were removed and the supernatants incubated for an additional 20 and 40 rain. The amount of thymopentin remaining, as quantitated by HPLC, was compared to samples containing lymphocytes for the time-points in question. The results are shown in Fig. 3. Two points can be made. First, the possibility that thymopentin stimulates secretion of its degrading enzymes can be ruled out, since degradation should continue to occur when the cells are removed. This is not the case. In fact, the value remains constant even up to 70 rain after the cells are removed. These data also support the fact that a negligible amount of degradation can occur during the time between the endpoint of the assay and the HPLC injection (total time 30 s). Second, it can be seen from the control samples (lymphocyte-containing) that the degradation assay appears linear up to 60 rain at 37°C with a high negative correlation (r = -0.98).

Time-course of the formation of thymopentin degradation products Thymopentin was incubated at 37 °C for varying times with lymphocytes. For each incubation point, the degradation products were separated by HPLC and quantitated. The results are shown in SO

'E

4o

| E "e

so

i|

I

I

I

I

il"hi:::;:., ................

............... - ..... 1

'I ................ 1 Q

"6 E c

lo,

o

2'o

4'o

6'o

n

0o

Time (minutes) Fig. 3. Degradation of thymopentin in the presence or absence of lymphocytes. Thymopentin (37 nmol) was incubated in the presence (®) of 5.106 iymphocytes for 20 min at 37 o C. At this time, the cells in half the tubes were removed by centrifugation, and the incubation was continued for an additional 20 or 40 rain (o). At the indicated time points, cell supernatants were analyzed for degradation by HPLC as described in Materials and Methods.

Fig. 4. Time-course of thymopentin degradation and product formation. lymphocytes at 37 * C for varying times. For each time pin thymogentin degradation and the formation of Lys-As described i

170

100

c o

l

t

gg

i g

l

|

2o

o 10 "10

10 .9

10 .0

10 .7

10 .6

10 -5

10 -4

10 -3

lO-2

[inhibitor] (M)

Fig. 5. Inhibition of thymopentin degradation by amasmtin, bestatin and 1,10-phenanthroline. Lymphocytes (5-10 ~) were pre-incubated in assay buffer (control) or in the presence of varying concentrations of amastatin (o), bestafin (O) or 1,10-phenanthroline ( × ) for 30 rain at 37 o C. At this point, 37 nmol of thymopentin was added and the incubation was continued for an additional 60 min (37 ° C). At the assay endpoint, the cell supernatants were processed for quantitation of thymopentin degradation by HPLC as described in Materials and Methods. The amount of thymopentin degradation was compared to controls (without inhibitor) and the results expressed as percent inhibition as described in Table Ill.

at the end of the incubation period in the presence of all the inhibitors, except N-ethylmaleimide, where the viability dropped to 56~; as determined by trypan blue exclusion. The non-viable cells probably become 'leaky', resulting in the release of intracellular enzymes into the extracellular medium. This possibility is supported by the earlier results stating that disrupted cell suspensions contain a high degree of thymopentin degrading activity.

Inhibition of thymopentin degradation by amastatin, bestatin and 1,l O.phenanthroline Of the peptide inhibitors listed in Table III, amastatin and bestatin were found to be the most effective. In addition, the chelator 1,10phenanthroline was also found to be effective. Therefore, these inhibitors were further examined through a series of inhibition curves. Fig. 5 displays these results. It is clearly seen that amastatin was the most potent peptide inhibitor of thymopentin degradation with an ICso of 4.5-10 -9 M.

Bestatin was much less potent with an ICs0 of 7.1.10 - 4 M . However, it must be noted that complete inhibition was never seen with either amastatin or bestatin even at concentrations greater than 1 mM. A small amount (approx. 3 nmol, peak e) of Asp-Val-Tyr always remained. No Lys-Asp-Val-Tyr was detected, although the possibility exists that it was formed and rapidly degraded to the tripeptide. Complete inhibition was seen with 1,10-phenanthroline. It exhibited and ICs0 value of 2- 10 -4 M. its effectiveness of inhibition (i.e., from 0 to 100~) however, was elicited over a very narrow concentration range. The ability of 1,10-phenanthroline to cause an inhibition correlates well with the fact that EDTA is a partial inhibitor of thymopentin degradation. The differences in the effect of the two chelators may be due to their ability to complex with the particular metal associated with the enzyme. From the data presented thus far, there is strong evidence for the 'stepwise' degradation of thymopentin from its N-terminal. It seems probable,

171 TABLE III EFFECT OF VARIOUS COMPOUNDS ON THYMOPENTIN DEGRADATION BY HUMAN LYMPHOCYTES a Compound b

inhibition of thymopentin degradation

Antipain Leupeptin Bestatin Chymostatin Phosphoramidon Amastatin

7 12 85 6 7 90

Dithiothreitol 2-Mercaptoethanol Cysteine lodoacetamide p-Mercuriphenyl sulfonic acid 5,5'-Dithiobis(2-nhrobermoic acid) N-Ethylmaleimide

11 8 27 1 2 10 27

TLCK PMSF Aprotinin Soybean trypsin inhibitor Trypsin inhibitor from chicken egg white EGTA EDTA 1,10-Phenanthroline 2,2"-Bipyridyl Imidazole

-

10 5 0 0 1 17 48 100 2 0

a Percent inhibition was determined by the following:

Inhibition of thymopentin degradation by 1,10phenanthroline: effect of metal ions The crucial importance of a divalent metal ion in the degradation of thymopentin by lymphocytes was demonstrated by the inhibitory effects of chelating agents such as EDTA and 1,10phenanthroline. In order to further demonstrate the role of divalent metal ions in thymopentin degrading activity, lymphocytes were incubated in the presence of 2.10-'* M 1,10-phenanthroEne either alone or in conjunction with various divalent metal ions. After this pre-incubation, thymopentin was added (37 nmol) and the incubation carried out for an additional 60 rain. The amount of thymopentin degraded was quantitated and the results expressed in terms of percent activation as compared to samples without the inlfibitor (Fig.

100

80

60

40m

1 - [ amount thymopentin degraded with inhibitor ~ x 100 amount thymopentin degraded w / o inhibitor / b Soybean and chicken egg white trypsin inhibitors were used at a concentration of 1.00 lag/ml. All remaining inhibitors were at a final concentration of 1 raM.

20 u

! I 10 "s

then, that the second phase of the degradation process which removes the Lys from the tetrapeptide, is effected by the same enzyme(s). When the tetrapeptide Lys-Asp-Val-Tyr was used as the starting substrate, the lymphocytes were able to cleave approx. 25% of the tetrapeptide to form the tripeptide Asp-Val-Tyr. The degradation of the tetrapeptide was able to be inhibited in a similar manner as thymopentin itself, with IC50 values of 6.3- 10 -9, 4.5" 10 -6 and 1.3- 10 -4 M for amastatin, bestatin and 1,10-phenanthroline, respectively (not shown).

I I 10

I I -4

10 .3

Fig. 6. Inhibition of thymopentin degradation by 1,10phenanthrofine: effect of metal ions. Lymphocytes (5.106) were pre-incubated in assay buffer (control - without calcium) or in the presence of 2.10 -4 M 1,10-phenanthroline with or without the addition of varying concentrations of divalent cations for 30 rain at 37 o C. After the pre-incubation, 37 nmol of thymopentin was added and the incubation continued for an additional 60 rain (37 o C). Cellular supematants were then processed and analyzed for thymopentin degradation as described in Materials and Methods and compared to controls. Results are expressed in terms of percent activation of thymopentin degradation. Divalent cations were as follows: Zn 2÷ (@), Cu 2+ (X), Co 2+ (0) and Hg 2+ (U).

172

6). It can be seen that Cu 2+ and Hg 2+ were essentially without effect on the reversal of inhibition of thymopentin degradation by 1,10phenanthroline at the concentrations tested. However, Zn2+ produced an increase in activation from 50 to 79~ at a concentration of 1-10 `.5 M, and then a further increase to 85~ at a concentration of 1-10 -4 M. Higher concentrations of Zn 2+ (110 -3 M) started to inhibit the enzymatic activity, decreasing the activation to 70~;. Cobalt displayed a similar biphasic t~pe of response. Its greatest effect (an increase to 86~ activation) was also seen at a concentration of 1.10 -4 M. Other ions tested included Ca 2+ and Mn2+ which were essentially without effect, Mg 2+ which produced a slight activation at a concentration of 1.10 -4 M. and Pb2+ which was found to be inhibitory at its highest concentration (1.10 -3 M, not shown). Discussion

This study, dealing with the fate of thymopentin upon interaction with human lymphocytes, revealed extensive degradation of the molecule. This was determined by chromatographic separation of the cleavage products from cellular supematants. Analysis revealed two major degradation products from thymopentin, the tetrapeptide Lys-Asp-Val-Tyr and the tripeptide AspVal-Tyr. These degradation products were positively identified based on their retention times and amino-acid composition as compared to synthesized standards. Most, if not all of the degradation, took place in the presence of intact lymphocytes. If lymphocytes were removed, the degradation ceased, even if incubation of the cellular supematant was continued. This argues against a thymopentin-stimulated secretory enzymatic event. Only a small amount (approx. 10~;) of thymopentin degradation was evident upon incubation of the pentapeptide with conditioned media. This affect, however, could be abrogated upon heating. This could be explained on the basis of a small percentage of enzymes from cellular darnage/!ysis in the preparation. Complete lysis by N 2 bomb cavitation and sonication revealed extensive and rapid hydrolysis of thymopentin and supports this explanation. Analysis of the time-course of degradation revealed a "stepwise' degradative event

beginning at the N-terminal. The degradation of thymopentin, as well as the formation of the tetrapeptide were essentially curvilinear. Degradation of the tripeptide was essentially linear. Althou~, tuis study does not rule out the possibifity of thymopentin uptake, degradation and release, the evidence for ectoenzyme(s) participating in the degradation of thymopentin is strong. First, the degradation products are located in the extracellular medium even at the earfiest timepoints without evidence of a 'lag' period in the formation of any of the products. Thus, the appearance of these products in the supernatants at the earlier time points suggests that uptake, de,gradation and release of thymopentin does not occur, an event which would require a longer period of time. Second, the percent recovery of thymopentin and its degradation products was always greater than 90~ at all time-points tested. If uptake, degradation and release were occurring, a decrease in the percent recovery should have been seen at some point during the time-course of events, thus reflecting the cells' ability to sequester the pentapeptide for further processing. The use of various enzyme inhibitors and compounds allowed insight into the possible types of enzyme(s) involved in thymopentin degradation. Of the inhibitors used, conventional inhibitors of serine proteinases were without effect. Likewise, various thiol reagents displayed the same results. Some thiol compounds, however, showed slight inhibitory effects. These effects were attributed to this class of compound's chelating abilities. Of all the compounds tested, amastatin, bestatin and 1,10-phenanthroline were the most effective inhibitors. Bestatin is a known inhibitor of leucine aminopeptidase and aminopeptidase B [33]. Leucine aminopeptidase is found on the surface and in the cytosol of a variety of mammalian cell types including lymphoid cells [34-35]. Amastatin, on the other hand, is an inhibitor of leucine aminopeptidase but not aminopeptidase B [36]. This evidence strongly suggests an aminopeptidase as the enzyme responsible for thymopentin degradation. Further evidence in support of an andnopeptidase or an aminopeptidase-like enzyme is shown by inhibition studies using chelating agents.

173

Thymopentin degradation was fully inhibited by 1,10-phenanthroline and partially inhibited by EDTA. The inhibition by 1,10-phenanthroline was able to be reversed in the presence of Zn 2+ and Co2+, but not with other divalent cations. The reversal of inhibition exhibited by Zn 2+ and Co2+ is most likely due to the ability of these ions to complex with this reagent, thus preventing its interaction with the metal-containing enzyme(s). The ability of higher concentrations of Zn 2+ and Co2+ (i.e., 1-10 -3 M) to effect an inhibition of enzyme activity is consistent with data concerning leucine aminopeptidase activity [33]. Heavy metals such as Zn 2+ are known to modify the cell surface in many ways [37-39]. These results establish the role of a metallo-enzyme(s) in thymopentin degradation. Leucine aminopeptidase is a known metallo-enzyme. Yet another cell surface enzyme, aminopeptidase N, is also an established metalloenzyme able to be inlfibited by amastatin [5]. Although it has been shown to be present on murine peritoneal macrophages, its existence on murine granulocytes and lymphocytes was not evident [40]. Whether a sirrfilarsituation exists on human lymphoid cells is unknown at this time. It is interesting to note that the inhibition of thymopentin degradation produced by amastatin and bestatin was never complete. There was always a small amount (approx. 3 nmol) of the tripeptide remaining. However, the inhibition was complete with amastatin and bestatin when the tetrapeptide was used as the starting substrate. This raises the possibility that another enzyme(s) may be involved in a direct thymopentin to tfipeptide cleavage. Since the inhibition of thymopentin degradation was only complete in the presence of 1,10-phenanthroline, another metallo-enzyme would be likely. Recently, one study has shown that in vivo administration of thymopentin was effective in improving the PHA-induced IL-2 production in aging humans [41]. Yet, these findings were at variance with in vitro studies where thymopentin was without effect on enhancing IL-2 production by lymphocytes from normal adults [42]. Besides age-related differences, the current data on thymopentin degradation may offer at least a partial explanation for inconsistencies seen in in vivo and in in vitro studies. Judging from its short

half-fife in plasma and now, with additional evidence presented in tiffs study for its degradation by human lymphocytes, it seems most appropriate to focus on N-terminal modifications and/or combination immunotherapy with arninopeptidase inhibitors for future in vitro/in vivo studies using thymopentin. Acknowledgement The authors would like to thank Freda Pilder and Kathy Storer for their excellent secretarial assistance. This work was supported in part by ~he Shriners of North America~ BRSG 507RR05408-24 awarded by the Biomedical Research Support Grant Program, Division of Research Resources~ NIH and USPHS Grant No. AI12936. References 1 Steinman, R.M., Mellman, I.S., Muller, W.A. and Colin, Z.A. (1983) J. Cell Biol. 96, 1-27. 2 Pastan, I.H. and Willingham, M.C. (1981) Annu. Rev. Physiol. 43, 239-250. 3 King, A., Christie and Cuatrecasas, P. (1981) New Engl. J. Med. 305, 77-88. 4 Brodsky, F.M. (1984) Immunology Today 5, 350-375. 5 Kenny, A.J., Schwartz, J.C. et al., Muller, W.E.G. et al and ga'eisel, W. et al. (1986) in Cellular Biology of Ectoenzymes (Kxeutzberg, G.W. et al., ed.), pp. 257-302, Springer-Verlag, Berlin. 6 Antoni, G., Presentini, R., Perin, F., Tagliabue, A., Ghiara, P., Censini, S., Volpin], G., Villa, L. and Boraschi, D. (1986) J. Immunol. 137, 3201-3204. 7 Steward, M.W. and Howard, C.R. (1987) ]mmunol. Today 8~ 51-5g 8 Goldstein, G., Scheid, M.P., Boyse, E.A., Schlesinger, D.H. and Van Wauwe, J. (1979) Science 204, 1309-1310. 9 Schlesinger, D.H. and Goldstein, G. (1975) Cell 5, 361-365. 10 Audhya, T., Scldesinger, D.H. and Goldstein, G. (1981) Biochemistry 20, 6195-6200. 11 Audhya, T., Schlesinger, D.H. and Goidstein, G. (1987) Proc. Natl. Acad. Sci. USA 84, 3545-3549. i2 Goldstein, G. (1974) Nature (London) 247, 11-14. 13 Goldstein, G. and Schlesinger, D.H. (1975) Lancet ii, 256-259. 14 Audhya, T.K., Scheid, M.P. and Goldstein, G. (1984) Proc. Natl. Acad. Sci. USA 81, 2847-2849. 15 Rasch, R.S. and Goldstein, G. (1974) Proc. Natl. Acad. Sci. USA 71, 1474-1478. 16 Scheid, M.P., Goldstein, G. and Boyse, E.A. (1978) J. Exp. Med. 147, 1727-1743. 17 Brand, A., Gilmore, D.G. and Goldstein, G. (1976) Science 193, 319-321.

174 18 Scheid, M.P., Goldstein, G. and Boyse, E.A. (1975) Science 190, 1211-1213. 19 Range, G.E., Scheid, M.P., Goldstein, G. and Boyse, E.A. (1982) J. Exp. Med. 156, 1057-1064. 20 Sunshine, G.H., Basch, R.S., Coffey, R.G., Cohen, K.W., Goldstein, G. and Hadden, J.W. (1978) Immunology 120, 1594-1599. 21 Sundal, E. (ed.) (1985) Thymopentin in experimental and clinical medicine, Surveys in Immunology Research, Vol. 4, Suppl. 1, pp. 1-154. 22 Tischio, J.P,, Patrick, J.E., Weintraub, H.S., Chasin. ![. and Goldstein, G. (1979) Intern. J. Peptide Protein Res. 14, 479-484. 23 Audhya, T. and Ooldstein, G. (1983) Intern. J. Peptide protein Res. 22, 187-193. 24 Audhya, T., Talle, M.A. and Ooldstein, G. (1984) Arch. Biochem. Biophys. 234, 167-177. 25 Bah'any, G, and Merrifield, R.B. (1980) in The Peptides, Analysis, Synthesis and Biology, Vol. 2, Special Methods in l~ptide Synthesis, Pan A (Gross, E. and Meienhofer, J., eds.), pp, 1-284, Academic Press, New York. 26 Balasubramaniam, A., Grupp, I., Srivastava, L., Tatemoto, K., Murphy, R.F,, Joffe, S.N. and Fischer, J.E. (1987) Intern. J. Peptide Protein Res. 29, 78-83. 26 Lemaire, S., Yamashiro, D., Behevens, C. and Li, C.H. (1977) J. Am. Chem. Soc. 99, 1577-1580. 28 Merrifield, R.B., Vizioli, B.D. and Boman, H.G. (1982) Biochemistry 21, 5020-5031. 29 Kaiser, E., Cole.w~tt, R.L., Bossinger, C.B. and Cook, P.L (1970) Anal. Biochem. 34, 595-598.

30 Boyum, A. (1976) Scand. J. Immunol., Suppl. 5, 9-15. 31 Tischio, J.P. and Hetyei, N. (1982) J. Chromatogr. 236, 237-243. 32 Mellor, D.P. (1979) in The Chelation of Heavy Metals (Levine, W.G., ed.), pp. 1-106, Pergamon Press, Oxford. 33 Soda, H., Takaaki, A., Takeuchi, T. and Umezawa, H. (1976) Arch. Biochem. Wmphys. 177, 196-200. 34 Aoyagi, T., Soda, H., Nagai, M., Ogawa, K., Suzuki, J. and Takeuchi, T. (1976) Biochim. Biophys. Acta 452, 131. 35 Leyhausen, G., Schuster, D.K., Vaith, P., Zahn, R.K., Umezawa, H., Falke, D. and Muller, W.E.G. (1983) Biochem. Pharm. 32, 1051-57. 36 Umezawa, H. (1981) in Small Molecular Immunomodifiers of Microbial Origin (Umezawa, H., ed.), Pergamon Press, Oxford. 37 Malave, i. and Behnam, I.R. (1984) Cell Immunol. 89, 322-30. 38 Maro, B., Avrameas, S. and Bornes, M. (1979) Exp. Cell Res. 118, 85. 39 Chvapil, M. (1976) Med. Clin. North Am. 60, 799. 40 Wachsmuth, E.D. and Staber, F.G. (1977) Exp. Cell Res. 109, 269-276. 41 Meroni, P.L., Barcellini, W., Frasca, D., Squetti, C., Borghi, M.O., DeBartolo, G., Doria, G. and Zanussi, C. (1987) Clin. Immunol. Immunopath 42, 151-159. 42 Duchateau, J., Servais, G., Cooman, R., Collet, H. and Bolla, K. (1985) Surv. lmmunol. Res. 4, suppl. 1, 116-124.