Plasma uptake and in vivo metabolism of [Leu]enkephalin following its intraperitoneal administration to rats

Peptides, Vol. 10, pp. 913-919. ©Pergamon Press plc, 1989. Printed in the U.S.A.

0196-9781/89 $3.00 + .00

Plasma Uptake and In Vivo Metabolism of [Leu]Enkephalin Following Its Intraperitoneal Administration to Rats G E R Y S C H U L T E I S , 1'2 S U S A N B. W E I N B E R G E R A N D J O E L. M A R T I N E Z , JR. 3

Department o f Psychology, University o f California, Berkeley, CA 94720 R e c e i v e d 10 M a r c h 1989

SCHULTEIS, G., S. B. WEINBERGER AND J. L. MARTINEZ, JR. Plasma uptake and in vivo metabolism of [Leu]enkephalin following its intraperitoneal administration to rats. PEPTIDES 10(5) 913-919, 1989.--To understand better how [Leu]enkephalin (LE) acts to modulate learning and memory in rats, the plasma uptake, disappearance, and metabolism of LE were investigated following its intraperitoneal administration. Concentrations of [3H]-LE and its radioactive metabolites were determined by thin layer chromatography in plasma samples withdrawn from rats at various times after injection of peptide. As measured in rats receiving an IP injection of a dose of LE (3 v,g/kg) that impairs active avoidance conditioning, the LE was very rapidly metabolized, with greater than 95% of plasma [3H] in the form of metabolites by 1 min after injection. Despite this rapid metabolism, low but measurable quantities of intact LE were detectable in plasma at all sampling times. Consistent with a greater potency of D-Ala2[D-LeuS]enkephalin (DADLE) than of LE in modulating avoidance conditioning, DADLE was less rapidly metabolized than was LE following its IP administration. The metabolism of DADLE and LE in vivo was more rapid than it was in plasma in vitro, suggesting a role for membrane bound enzymes in the metabolism of IP-administered enkephalins. The data demonstrate that, despite a rapid hydrolysis of LE in vivo, sufficient LE is present in plasma following IP administration of a behaviorally active dose to support a role of circulating intact LE in the modulation of avoidance conditioning. [Leu]enkephalin

Plasma uptake

In vivo

Hydrolysis

[LEU]ENKEPHALIN (LE) impairs the acquisition and retention of a variety of conditioned responses when it is administered systemically in microgram quantities (10, 11, 19, 22, 32). Evidence suggests that this action of LE is mediated at a site that lies outside the blood-brain barrier, and that activation of opioid delta receptors is required for LE to exert its effects (19, 22, 32). Although the half-life for LE in rat plasma in vitro is known to be 2.0-2.5 min (6, 21, 41), the enzymatic breakdown of LE in vivo following its systemic administration is unknown. Also not yet known are the concentrations of LE achieved in the circulation following its intraperitoneal (IP) administration. Given the small doses at which LE is effective in impairing conditioning, as well as our recent finding that the rate of hydrolysis of LE in rat plasma is highly correlated with active avoidance acquisition (21), we felt it necessary to determine the plasma uptake and rate of in vivo LE degradation, in order to gain a more complete understanding of how LE acts to modulate learning and memory following its peripheral administration. In the current study we measured the uptake of [3H]-LE into rat plasma and the rate of its in vivo degradation over a 15-rain period

Rat

following IP LE administration. This period was chosen to correspond to our active avoidance paradigm, in which training begins 5 min after drug injection and lasts about 10 min. Because the dose-response function for LE's effects on avoidance conditioning is U-shaped [see (19,32) for review], we studied the uptake and degradation of LE following IP administration of both a 3 v~g/kg dose, that impairs active avoidance conditioning in rats, and a 30 wg/kg dose, that is without effect on avoidance conditioning [(39); Janak and Martinez, unpublished observations]. We also determined the uptake and in vivo metabolism of [3H]-D-Ala2[D-LeuS]enkephalin (DADLE), an enkephalin analog that is more slowly metabolized than LE, but that is more potent than LE in impairing avoidance conditioning in rats (29,30). METHOD

Subjects Male Sprague-Dawley rats (Harlan Sprague-Dawley, Indianapolis, IN) weighing between 265 and 350 g were used as subjects, Upon arrival from Harlan, the animals were housed in pairs in

1Supported by Public Health Service National Research Service Award (NRSA) 1-F31-DA05334. 2Requests for reprints should be addressed to Gery Schulteis, Department of Psychology, 3210 Tolman Hall, University of California, Berkeley, CA 94720. 3Supported by PHS grant DA04195.

913

914

SCHULTEIS, WEINBERGER AND MARTINEZ

::t/ I

I

I

I

I

I

I

8001 '0

B

OOoOl

X

|

I

I

I

I

I

100 80~~ '~ I

.E

I

I ~"

60 40 9O

I,I

I

I

I

I

i

1201

'°t

60 30

I

j I

I

Time Mter Injection (rain)

FIG. 1. Profiles of total plasma [3H] as a function of time after peptide injection.A) 3 p,g/kg [Leu]enkephalin(LE), administeredIP (n = 6); B) 30 txg/kg LE, administered IP (n=3); C) 0.6 p,g/kg D-Ala2-[D-Leu5] enkephalin, administered IP (n=5); D) 3 ~g/kg LE, administered IA (n=4).

standard laboratory cages, with food and water freely available. The rats were maintained on a standard 12:12 hr light:dark cycle, with lights on at 08:00.

Drugs [3H]-[Leu]enkephalin and [3H]-D-Ala2-[D-LeuS]enkephalin(specific activity 36.3-47.6 Ci/mmol) were purchased from New England Nuclear (Boston, MA). Unlabelled [Leu]enkephalin and D-Ala2-[D-LeuS]enkephalin were obtained from Bachem (Torrance, CA). Bestatin was purchased from Sigma (St. Louis, MO). All drugs were prepared in saline for injection in a volume of I ml/kg.

Surgery Six days after arrival, animals were anesthetized with pentobarbital (50 mg/kg IP) and implanted with left femoral artery cannulae (PE 50 tubing, Clay-Adams), as detailed in Martinez and Weinberger (21). The carmula was passed under the skin of the back to the neck region, where it was exteriorized and plugged. The rats were allowed a twenty-four hour recovery period before blood collection.

Plasma Uptake and In Vivo Metabolism Studies To study the uptake and metabolism of LE and DADLE following their IP administration, rats were injected with one of the following drugs: a) 3 ~g/kg LE (25% [aH]-LE by weight), b)

30 txg/kg LE (15% [3H]-LE by weight), or c) 0.6 ixg/kg DADLE (100% [3H]-DADLE). To determine the disappearance of LE from plasma in the absence of the influence of absorption kinetics, a group of rats was injected intraarterially (IA) with 3 ixg/kg LE (administered through the femoral artery cannula). To determine whether LE injected IP is metabolized as it crosses membranes from the site of injection into the bloodstream, or whether most metabolism occurs once LE has entered the bloodstream, rats received one of the following combinations of LE and the aminopeptidase inhibitor bestatin: a) bestatin (10 mg/kg) administered IA 50 sec prior to the IP administration of 3 Ixg/kg LE, or b) a combined IP injection of 3 la.g/kg LE and 10 mg/kg bestatin. Blood samples (0.4 ml) were collected into heparinized polypropylene tubes at 1, 2, 3 , 4 , 5, 7.5, and 15 minutes after peptide injection; blood volume removed was immediately replaced with saline. Plasma was rapidly separated by microcentrifugation (20 sec) and the enzymes in a 160-1xl aliquot of plasma were denatured in two volumes of 0.1% trifluoroacetic acid in methanol (TFA-MeOH) 70 sec after blood collection. The presence of 500 IxM bestatin (final concentration) in the collection tube minimized the degradation of LE between the time of blood collection and enzyme denaturation. Pilot studies determined the amount of LE hydrolysis that occurred in the presence of 500 txM bestatin during the time between blood collection and enzyme denaturation; all data subsequently were corrected for this amount of hydrolysis (5.56%). Following centrifugation of the denatured plasma sample, a 120-I.d aliquot of the supematant was added to 3 ml of Ready-Gel Scintillation Cocktail (Beckman, Palo Alto, CA) and total [3H] content was quantified with an LS 1800 liquid scintillation counter (Beckman). The remainder of each sample was assayed as described below (TLC Determination of Enkephalin Hydrolysis).

In Vitro Metabolism Studies To compare the rate of hydrolysis of LE in vivo with the in vitro time course of LE and DADLE metabolism in plasma, a single blood sample (1.0 ml) was collected from each rat and microcentrifuged for 20 sec. A 300-400 p~l sample of plasma was removed and incubated for 5 min at 370(2 (to allow hydrolysis of endogenous enkephalin), followed by addition of [3H]-LE or [3H]-DADLE (5-6 p~l) to the plasma sample. Aliquots were denatured in two volumes of 0.1% TFA-MeOH at 1, 2 and 7.5 min after LE addition, or at 5, 30, 60 and 150 min after DADLE addition. Samples were then centrifuged and subjected to TLC analysis as described below. To determine whether the in vitro metabolism of LE in plasma and whole blood differ, 300 ILl of plasma was treated as described in the previous paragraph. An additional 300 ILl sample of whole blood was collected from the same animals and incubated for 5 min at 37°C. [3H]-LE (6 Ixl) was added to the blood, and aliquots of the blood sample were denatured in three volumes of 0.1% TFA-MeOH at 1, 2, and 7.5 min after addition of radiolabel. Samples were then separated with TLC. In a separate study, we determined the amount of inhibition of in vitro LE hydrolysis that is seen in blood collected from animals receiving a prior IA or IP injection of bestatin (10 mg/kg). Blood samples (0.25 ml) for this study were collected into heparinized tubes containing 2 ILl [3H]-LE at times corresponding to the 1-, 5-, and 15-min blood sample collection times in the in vivo studies. The reaction was terminated after 2 min by addition of 0.1% TFA-MeOH as described above. This was again followed by TLC separation.

TLC Determination of Enkephalin Hydrolysis Details of the thin layer chromatography (TLC) assay method-

[LEU]ENKEPHALIN UPTAKE AND IN VIVO METABOLISM

915

TABLE 1 MEAN PERCENT ( + SEM) OF PLASMA [3H] IN THE FORM OF ENKEPHALINMETABOLITESAS A FUNCTIONOF TIME AFTER INJECTIONOF RADIOLABELLEDPEPTIDES Time After Injection (min) 1

LE 3 ixg/kg (IP) 95.8 ± 0.8 (n = 6) LE 30 Ixg/kg (IP) 93.8 --- 1.0 (n = 3) LE 3 Ixg/kg (IA) 80.2 + 5.2 (n = 4) LE3 ~g/kg(IP) + 75.5 __. 2.5 Bestatin 10 mg/kg (IA) (n = 5) LE 3 ~g/kg (IP) + 46.4 ___ 0.6 Bestatin 10 mg/kg (IP) (n=5) DADLE0.61.Lg/kg(IP) 12.2 ___ 4.3 (n = 5)

2

3

4

5

7.5

15

98.2 __ 0.3

98.3 -- 0.3

98.1 ± 0.4

98.1 -+ 0.3

98.2 -- 0.4

98.5 --+ 0.4

97.3 --- 0.3

97.8 --- 0.4

98.0 --- 0.6

98.3 --+ 0.1

98.5 --- 0.1

97.0 __- 1.8

97.2 --+ 0.7

95.3 --- 3.2

96.5 --- 1.7

98.4 --- 0.5

99.3 --- 0.4

99.1 __- 0.5

89.6 _ 0.9

95.1 -+ 0.4

96.2 --_ 0.3

96.0 ± 1.1

98.0 --- 0.7

98.2 ± 0.8

64.4 _.+ 1.1

76.2 --- 0.6

80.8 ± 0.9

82.9 ± 1.7

89.4 ± 1.7

91.0 __. 2.2

16.8 ± 4.1

24.0 --- 4.4

32.1 - 5.5

34.6 --- 5.4

49.4 - 4.7

63.7 __. 3.4

Note: Route of administration is indicated in parentheses after each drug. (IP) = intraperitoneal, (IA) = intraarterial.

ology are provided in Martinez and Weinberger (21). Previous studies in our laboratory showed that virtually identical patterns of plasma LE metabolism are observed using either TLC (41) or HPLC (34) separation. However, our radiometric TLC assay is more sensitive than is HPLC (34) to measurement of plasma metabolism of enkephalins at concentrations approximating physiological limits; therefore we used TLC in the present study. To review briefly the details of this method, denatured plasma samples were applied to reverse phase KC-18 TLC plates (Whatman, Clifton, N J). The plates were developed in standard glass TLC tanks containing 1-propanol and 0.1 M phosphate buffer (pH 4.1) in a ratio of 30:70 (42). This method allows the separation of LE or DADLE from their combined metabolites, with at least 95% reproducibility of results. The separated samples were then scraped from the plates and added to 5 ml Ready-Gel for LS quantification. RESULTS

Profiles of Total Plasma [~HI After LE or DADLE Injection Following IP injection of the avoidance-impairing dose of 3 Ixg/kg of LE, plasma [3H] showed its steepest rise in the first minute after injection, and levels peaked within 2--4 min (see Fig. 1A). A similar pattern was seen in animals injected IP with 30 ~g/kg of LE (Fig. 1B). By 15 min, [3H] levels in plasma fell to 60.4% and 63.0% of peak levels for the 3 and 30 ~g/kg doses of LE, respectively. Following injection of DADLE (0.6 Ixg/kg), plasma levels of [3HI again showed their steepest rise in the first minute after injection (Fig. 1C). However, in contrast to the results seen following LE injection, in animals treated with DADLE [3H] levels were still at 89.7% of peak levels at 15 min. To determine whether this result was due to a slower distribution of intact enkephalins than of their metabolic fragments, we compared the levels of plasma [3HI as a function of time in animals treated with both LE (3 I~g/kg IP) and the aminopeptidase inhibitor bestatin (10 mg/kg IP) to those in animals given LE (3 ~g/kg) alone. A mixed design ANOVA revealed that between 4 and 15 min after injection, but not between min 1 and 3, plasma [3H] levels in animals receiving the combination of LE and bestatin were significantly higher than those in animals receiving

LE alone [min 4-15: main drug effect, F(1,13)=6.60, p<0.05; min 1-3: main drug effect, F(1,13) = 0.855, NS; interactions not significant; data not shown]. These data are consisent with a slower distribution from plasma of intact LE than of metabolites. Following IA administration of 3 iJ.g/kg of LE, plasma [3H] levels fell rapidly, such that by 15 minutes only 18.6% of the radioactivity present at 1 min remained in plasma (see Fig. ld).

In Vivo Metabolism of LE and DADLE Following IP administration, LE is rapidly hydrolyzed, such that by 1 min after injection of the 3 Ixg/kg LE dose greater than 95% of plasma [3HI is in the form of metabolites (see Table 1). Similarly, almost 94% of plasma [3HI is in the form of enkephalin metabolites by I min after the IP administration of 30 ~g/kg LE. As measured by the relative levels of intact peptide and metabolites appearing in plasma, LE metabolism following its IA administration starts out more slowly than it does after IP administration. Following IA administration of 3 Ixg/kg LE, only about 80% of plasma [3H] is in metabolite form 1 min after injection, but by 2 min and thereafter the percentages of metabolites are similar to those found with IP LE administration, with over 97% metabolized. As expected, DADLE is much more slowly metabolized

TABLE 2 INHIBITIONOF IN VITRO [LEU]ENKEPHALINMETABOLISMIN BLOOD TAKEN FROMRATS INJECTEDWITH BESTATIN Time of Blood Collection After Injection (rain)

Bestatin (10 mg/kg IP) Bestatin (10 mg/kg IA)

1

5

15

64.36* 75.81

71.48 67.18

63.93 56.63

*Data represent the % inhibition of metabolism produced by bestatin relative to metabolism in blood taken from animals treated with saline. Saline n = 4, both bestatin groups n = 3.

916

SCHULTEIS, WEINBERGER AND MARTINEZ

100 90-80-70--

60-5040-

//

LE

30DADLE

2010-

31o

~o

~o

1*so

Reaction T i m e (min)

FIG. 2. Time course of in vitro plasma degradation of [Leu]enkephalin (LE; n = 7) and D-Ala2-[D-LeuS]enkephatin (DADLE; n = 3). As was seen in vivo, LE is more rapidly metabolized than DADLE; however, both peptides are more slowly degraded in rat plasma in vitro than they are following IP administration to rats (compare data with that in Table 1). than is LE, with only 12.2% of plasma [3H] in the form of DADLE metabolites 1 min after its IP injection. When bestatin (10 mg/kg) is administered IA, the hydrolysis of IP administered LE is significantly attenuated (see Table 1), in that approximately 75% (as compared to more than 95% in animals receiving only LE) of plasma [3H] is in the form of enkephalin metabolites 1 min after LE injection in these animals; this difference persists through postinjection min 5 [mixed design ANOVA: main drug effect, F(1,9)=34.71, p<0.001; drug x time interaction, F(6,54)= 57.13, p<0.001]. The IP administration of bestatin in combination with LE results in a more pronounced inhibition of hydrolysis than does IA bestatin paired with IP LE administration, with only 46.4% plasma [3HI in the form of metabolites at 1 min in the former group; this difference is maintained over the 15 rain sampling period [main drug effect, F(1,8) = 159.79, p<0.0001; drug × time interaction, F(6,48)= 31.94, p<0.001]. However, when [3H]-LE is added in vitro to blood samples withdrawn from animals pretreated with bestatin administered either IP or IA, the IA route of bestatin administration results in greater inhibition of LE degradation in blood at 1 min than does IP bestatin (Mann-Whitney U-test: U = 0 , p<0.05; see Table 2).

In Vitro Plasma and Whole Blood Metabolism of LE and DADLE As can be seen in Fig. 2, both LE and DADLE are more slowly

metabolized in plasma in vitro than they are in vivo following their IP administration to rats. In plasma in vitro, LE is 34.2% metabolized in 1 min, whereas in vivo plasma contains 95.8% LE metabolites by 1 min. DADLE is only 16.2% metabolized after 2.5 hours of incubation in plasma in vitro, whereas it is almost 64% metabolized by 15 min after IP administration. To determine whether certain factors present in whole blood that are absent in plasma may acount for some of the differences in metabolism observed in vivo and in plasma in vitro, we compared the in vitro metabolism of LE in whole blood and plasma samples taken from the same animal. There was no significant difference in the metabolism of LE in these samples [mean difference in percent metabolized (whole blood - plasma) -+ SEM: 1 min, 1.07 --_ 1.10%; 2 min, 1 . 3 9 +- 1.79%; 7.5 min, 0.48-+0.49%; n = 5 ] .

Profiles of Intact Peptides in Plasma The concentrations of intact peptide remaining in plasma as a function of time after administration are summarized in Table 3. Although LE is very rapidly metabolized following IP administration of a 3 p,g/kg dose, measurable levels of intact LE are achieved and maintained in plasma throughout the 15-min sampling period. Administration of 30 i~g/kg of LE results in plasma concentrations of intact peptide that are roughly 10 times those seen after injection of the 3 txg/kg dose at all sampling time points. In contrast to IP injections, following IA administration of 3 ~g/kg of LE plasma levels of intact LE fall rapidly, such that by 7.5 min after administration they are below the limits of detectability of our assay. Consistent with the slower in vivo metabolism of DADLE than of LE, IP administration of 0.6 Ixg/kg of DADLE results in higher plasma concentrations of intact peptide than are observed with a larger (3 Ixg/kg) dose of LE (see Table 3). DISCUSSION Our results demonstrate that LE is rapidly absorbed into the bloodstream following its IP administration, with peak plasma levels of intact peptide achieved within the fn,st min after injection (see Table 3). Following IP administration of an avoidanceimpairing dose (3 p~g/kg), low but measurable levels of LE are maintained in plasma during the entire 15-min sampling period, which encompasses the interval between drug injection and the completion of active avoidance training in our behavioral studies. In contrast, LE disappears rapidly from plasma when it is administered IA, such that it is no longer detectable by 7.5 rain after injection. This suggests that the maintenance of plasma LE levels following its IP administration may be due to the continued

TABLE 3 CONCENTRATIONSOF UNMETABOLIZEDPEPTIDEIN PLASMAAS A FUNCTIONOF TIME AFTERINJECTION Time After Injection (min) 1

LE 3 p,g/kg (IP) (n = 6) LE 30 Ixg/kg (IP) (n = 3) LE 3 p.g/kg (IA) (n = 4) DADLE 0.6 p.g/kg lIP) (n = 5)

2

3

4

5

7.5

15

78.2 ± 8.0

48.0 +-- 3.5

50.6 ± 3.0

53.5 --- 2.7

53.9 ± 1.8

43.6 - 1.2

27.7 -+ 0.8

1014 - 165

647 ± 70

568 --- 97

579 ± 94

524 ±- 12

390 ± 26

592 ± 258

52

51.8 - 4.9

70.0 - 7.6

44.5 ± 5.0

17.7 ±- 1.3

236 --- 85

339 ±- 82

356 - 66

343 - 58

354 ± 55

798 ±

Note: ND = not detectable. Data represent mean ( -+SEM) concentrations, expressed as pg/ml plasma.

ND 270 +-- 25

ND 176 ±

10

[LEU]ENKEPHALIN UPTAKE AND IN VIVO METABOLISM

917

uptake of peptide from the site of administration. Importantly, the levels of LE seen in plasma following the administration of 3 o,g/kg of LE, a dose that is known to impair avoidance conditioning in rats [(39); Janak and Martinez, unpublished observations], are within the range of circulating levels of enkephalins reported in rodent plasma (5-150 pg/ml plasma) [see (8, 22, 35, 36)]. This suggests that the impairment in avoidance conditioning seen with this dose of LE is not likely due to a pharmacological response to supraphysiological levels of circulating LE. That a lower dose of DADLE (0.6 i~g/kg) achieves plasma concentrations of intact peptide that are greater than those achieved by LE (3 i~g/kg) is consistent with the greater potency of DADLE for impairing avoidance conditioning (29,30). The present study further demonstrates that the enzymatic hydrolysis of LE following its IP administration is extremely rapid (less than 5% of plasma [3H] is in the form of intact LE by one min after injection of a behaviorally active 3 ixg/kg dose). Following IA administration, a higher percentage (almost 20%) of the [3HI in plasma is in the form of intact LE by one rain after injection, but by 2 rain and thereafter the percentages of intact LE (2-3%) are similar to those seen with IP LE administration. Interestingly, the degradation of both LE and DADLE in vivo is much more rapid than it is in plasma in vitro (see Table 1 and Fig. 2). Factors present in whole blood that are absent in plasma cannot account for this large difference, because the in vitro metabolism of LE in whole blood does not differ from that in plasma. Thus, in addition to circulating enzymes, there must be other contributions to the in vivo metabolism of enkephalins. Membranes of the peripheral vasculature are known to contain aminopeptidase M (25,26). A variety of peripheral tissues, including lung, liver, kidney, and intestine, contain this enzyme as well (7). Since LE is metabolized more rapidly when administered IA than it is when added to plasma or whole blood in vitro, noncirculating enzymes most likely contribute to the in vivo metabolism of systemically administered LE. Bestatin more effectively inhibits in vivo hydrolysis of LE given IP when it is administered in combination with the LE than when the bestatin is administered IA. Thus, it is likely that following IP LE administration a great deal of LE is metabolized as it encounters bestatin-sensitive enzymes that are found between its site of injection and the bloodstream. An alternative explanation, that bestatin is rapidly removed from or inactivated in plasma following its IA administration, is not consistent with our finding that bestatin effectively inhibits LE metabolism in blood withdrawn from animals pretreated with bestatin IA (see Table 2). In addition to aminopeptidase M, other enkephalin hydrolyzing enzymes found in peripheral tissues also may play a role in the metabolism of circulating enkephalins. For example, the dipeptidyl carboxypeptidase known as "enkephalinase" is present in many of the same peripheral tissues that contain aminopeptidase M activity (3,17). Perfused rat lungs also contain dipeptidyl carboxypeptidase activity in the form of angiotensin converting enzyme (5), which effectively cleaves enkephalins. A D-Ala2 substitution makes an enkephalin resistant to aminopeptidase M activity, and thus a useful substrate for studying dipeptidyl carboxypeptidase activity [e.g., (2, 3, 23)]. Our finding that DADLE is more rapidly metabolized in vivo than it is in plasma in vitro suggests a role for noncirculating angiotensin converting enzyme and/or enkephalinase in the hydrolysis of circulating enkephalins. Since LE is rapidly degraded following its IP administration, it is possible that its effect on conditioning may be mediated at least in part through the production of behaviorally active metabolic fragments. In line with this suggestion, two tyrosine-containing metabolites of LE (Tyr-Gly-Gly-Phe-Leu), Tyr-Gly [(12,19); Martinez, Weinberger, Janak, and Schulteis, submitted) and Tyr-Gly-Gly (19,39), produce impairments in learning that paral-

lel those produced by the parent peptide. On the other hand, considerable evidence supports the notion that intact enkephalins modulate learning and memory. For example, the effects of LE on active avoidance conditioning in mice (33) and on taste avoidance conditioning in chicks (28) are reversed by the delta opioid receptor antagonist I e I 174,864, suggesting the effects of LE are mediated through delta receptors. Since enkephalin metabolites have little or no opioid receptor affinity (4), the reversal of LE's effect by ICI 174,864 argues against the possibility that the conditioning effects of LE are mediated soleIy through production of behaviorally active metabolites. Further evidence for the importance of the intact parent compound comes from the finding that metabolically more stable analogs of LE such as D-Pen 2[D-PenS]enkephalin (DPDPE) and D-Pen2-[L-Pen5]enkephalin (DPLPE) (28, 31, 39) and DADLE (29,30) share LE's impairing actions on avoidance conditioning, These analogs also share LE's opioid agonist properties and affinity for the delta receptor, but frequently are more potent than LE in producing their behavioral effects. Finally, the present study demonstrated that, although LE is rapidly and extensively metabolized following its IP administration, the concentrations of intact LE detectable in plasma following IP administration of a behaviorally active dose are within the range of endogenous enkephalin concentrations as reported by others (8, 22, 35, 36). This suggests that sufficient intact LE may remain to exert a behavioral effect. Thus, while enkephalin fragments are known to impair conditioning, and it is possible that they may contribute to the impairment produced by LE, the plasma concentrations of LE that are attained following its IP administration in rats suggest that intact enkephalins may modulate learning and memory in a physiologically relevant manner. Findings parallel to those described here are reported for other behaviorally active opioid peptides. For example, the uptake and distribution of ot-endorphin (15) and of des-[Tyrl]-3,-endorphin and des-enkephalin-~/-endorphin (37,38) are quite rapid following their systemic administration. Consistent with the rapid in vivo degradation of LE that we observed, by one min after IV administration of des-enkephalin-~-endorphin only t2-20% of total plasma radioactivity is in the form of unmetabolized peptide, and this level declines rapidly thereafter (38). Thus, the uptake, distribution, and metabolism of behaviorally active endorphin fragments, like those of LE, are quite rapid following systemic peptide administration. Consistent with our findings in rats injected IA with LE, Honghten et al. (9) reported a rapid distribution of [3-endorphin following its intravenous administration in rats and rabbits, with a distribution half-life of 2.0 minutes. However, [3-endorphin is more slowly metabolized than is LE; approximately 50% of radioactivity in plasma is in the form of intact fl-endorphin 45 rain after IV administration to rats. In this regard it is interesting to note that [3--endorphin can alter avoidance conditioning when administered in doses as low as 100 ng/kg (10,20), a dose lower than that required to produce similar effects with LE [see (21,32) for review]. Thus, as with DADLE, a greater behavioral potency of [3--endorphin relative to that of LE is accompanied by a slower rate of in vivo degradation. In summary, the most important f'mding of the current study is that, despite the rapid hydrolysis of LE, the plasma concentrations of intact LE that are reached following IP administration of an avoidance-impairing dose are within the range of previously reported endogenous circulating enkephalin levels (8, 22, 35, 36). This suggests that the effects on avoidance conditioning we observe (18, 22, 32) following administration of exogenous LE may mimic the actions of endogenous circulating enkephalins and may be mediated through the intact enkephalin molecule. Of importance to the results presented here, Kastin and

918

SCHULTEIS, WEINBERGER AND MARTINEZ

colleagues (1, 13, 24) have discussed the controversy surrounding behavioral actions of systemically administered neuropeptides. The list of peptides with behavioral action following systemic administration and the behaviors they affect is enormous; the interested reader is directed to several recent reviews on the subject (14, 16, 18, 19, 22, 32). However, given the low doses at which many peptides exert their behavioral effects, their rapid hydrolysis, and our poor understanding of how these peptides interact with the central nervous system to exert their behavioral effects (1, 13, 24), many investigators are hesitant to study peptide action following peripheral administration. However, the results presented in the current study provide evidence that small doses of a peripherally administered peptide can achieve concentrations in plasma that are within the range reported to be present in plasma (8, 22, 35, 36) despite the rapid in vivo hydrolysis of this peptide. Although how enkephalins interact with the CNS to affect behav-

ior is not yet understood, research suggests that enkephalins do not cross the blood-brain barrier to produce behavioral actions (19, 22, 27, 32). They may, however, directly affect the CNS at areas poorly protected by the blood-brain barrier, such as circumventricular organs (13, 19, 24), or by influencing peripheral autonomic afferent activity (13, 14, 18, 19, 24). We feel the conclusion reached by Kastin, Olson, Schally and Coy (13) a decade ago is still valid: "regardless of how the peptides affect the brain, investigation of their actions need not wait for the desired [complete] understanding of their mechanism."

ACKNOWLEDGEMENTS The authors wish to thank L. Chisholm, P. H. Janak, B. E. Derrick and A. Pritchard for technical assistance.

REFERENCES

1. Banks, W. A.; Kastin, A. J. Permeability of the blood-brain barrier to neuropeptides: The case for penetration. Psychoneuroendocrinology 10:385-399; 1985. 2. Benovitz, D. E.; Spatola, A. F. Enkephalin pseudopeptides: Resistance to in vitro proteolytic degradation afforded by amide bond replacements extends to remote sites. Peptides 6:257-261; 1985. 3. De la Baume, S.; Brion, F.; Dam Trung Tuong, M.; Schwartz, J. C. Evaluation of enkephalinase inhibition in the living mouse, using [3H]acetorphan as a probe. J. Pharrnacol. Exp. Ther. 274:653-660; 1988. 4. Dewey, W. L. Structure-activity relationships. In: Maliek, J. B.; Bell, R. M. S., eds, Endorphins: chemistry, physiology, pharmacology, and clinical relevance. New York: Marcel Dekker Inc.; 1982:23-56. 5. Gillespie, M. N.; Krechniak, J. W.; Crooks, P. A.; Altiere, R. J.; Olson, J. W. Pulmonary metabolism of exogenous enkephalins in isolated perfused rat lungs. J. Pharmacol. Exp. Ther. 232:675-681; 1985. 6. Hambrook, J. M.; Morgan, B. A.; Rance, M. J.; Smith, C. F. C. Mode of deactivation of the enkephalins in rat and human plasma and rat brain homogenates. Nature 262:782-783; 1976. 7. Hersh, L. B.; Aboukhair, N.; Watson, S. Immunohistochemical localization of aminopeptidase M in rat brain and periphery: Relationship of enzyme localization and enkephalin metabolism. Peptides 8:523-532; 1987. 8. Hexum, T. D.; Russett, L. R. Plasma enkephalin-like peptide response to chronic nicotine infusion in guinea pig. Brain Res. 406: 370-372; 1987. 9. Houghten, R. A.; Swarm, R. W.; Li, C. H. 13-endorphin: Stability, clearance behavior, and entry into the central nervous system after intravenous injection of the tritiated peptide in rats and rabbits. Prec. Natl. Acad. Sci. USA 77:4588-4591; 1980. 10. Introini, I. B.; McGaugh, J. L.; Baratti, C. M. Pharmacological evidence of a central effect of naltrexone, morphine, and I~-endorphin and a peripheral effect of met- and leu-enkephalin on retention of an inhibitory response in mice. Behav, Neural Biol. 44:43 A, A.A-6;1985. 11. Izquierdo, I.; Carrasco, M. A.; Dias, R. D.; Perry, M. L. S.; NeRo, C. A.; Wofchuk, S. T.; Souza, D. O. Effects of 13-endorphin, met-, leu-, and des-tyr-met-enkephalin on learned behaviors: One or more effects? International Congress Series Proceedings--Amsterdam. Excerpta Medica 620:83-95; 1982. 12. Janak, P. H.; Valedon, A.; Schulteis, G.; Derrick, B. E.; Fett, K.; Weinberger, S. B.; Martinez, J. L., Jr. Two enkephalin metabolites, Tyr-Gly-Gly-Pbe and Tyr-Gly, impair acquisition of an active avoidance response in mice. Soc. Neurosci. Abstr. 13:845; 1987. 13. Kastin, A. J.; Olson, R. D.; Schally, A. V.; Coy, D. H, CNS effects of peripherally administered brain peptides. Life Sci. 25:401--414; 1979. 14. Koob, G. Neuropeptides and memory. In: Iversen, L. L.; Iversen, S. D.; Snyder, S. H., eds. Handbook of psychopharmacology, vol. 19. New directions in behavioral pharmacology. New York: Plenum Press; 1987:531-573. 15. Koob, G. F.; Le Moal, M.; Bloom, F. E. Enkephalin and endorphin

16.

17. 18. 19.

20. 21. 22. 23. 24. 25. 26. 27. 28.

29.

30.

influences on appetitive and aversive conditioning. In: Martinez, J. L., Jr.; Jensen, R. A.; Messing, R. B.; Rigter, H.; McGaugh, J. L., eds. Endogenous peptides and learning and memory processes. New York: Academic Press; 1981:249-267. Le Moal, M.; Bluthe, R.-M.; Dantzer, R.; Bloom, F. E.; Koob, G. F. The role of arginine vasopressin and other neuropeptides in brain-body regulation. In: Stahl, S. M.; Iversen, S. D.; Goodman, E. C., eds. Cognitive neurochemistry. New York: Oxford University Press; 1987:203-232. Llorens, C.; Schwartz, J.-C. Enkephalinase activity in rat peripheral organs. Eur. J. Pharmacol. 69:113-116; 1981. McGaugh, J. L. Involvement of hormonal and neuromodulatory systems in the regulation of memory storage. Annu. Rev. Neurosci. 12:255-287; 1989. Martinez, J. L., Jr.; Janak, P. H.; Weinberger, S. B.; Schulteis, G.; Derrick, B. E. Enkephalin influences on behavioral and neural plasticity: Mechanisms of action. In: Erinoff, L., ed. Behavior as an indicator of neuropharmacotogical events: Learning and memory. NIDA research monograph. Washington, DC; 1989:in press. Martinez, J. L., Jr.; Rigter, H. Endorphins alter acquisition and consolidation of an inhibitory avoidance response in rats. Neurosci. Lett. 18:197-201; 1980. Martinez, J. L., Jr.; Weinberger, S. B. Enkephalin hydrolysis in plasma is highly correlated with escape performance in the rat. Behav. Neurosci. 102:404--408; 1988. Martinez, J. L., Jr.; Weinberger, S. B.; Schulteis, G. Enkephalins and learning and memory: A review of evidence for a site of action outside the blood-brain barrier. Behav. Neural Biol. 49:192-221; 1988. Matsas, R.; Kenny, J.; Turner, A. Endopeptidase-24.11 "enkephalinase": Its role in the central nervous system. Prog. Clin. Biol. Res. 180:273-277; 1985. Olson, R. D.; Kastin, A. J.; Olson, G. A.; Coy, D. H. Behavioral effects after systemic injection of opiate peptides. Psychoneuroendocrinology 5:47-52; 1980. Palmieri, F. E.; Bausback, H. H.; Ward, P. E. Metabolism of vasoactive peptides by vascular endothelium and smooth muscle aminopeptidase M. Biocbem. Pharmacol. 38:173-180; 1989. Palmieri, F. E.; Petrelli, J. J.; Ward, P. E. Vascular, plasma membrane aminopeptidas¢ M. Metabolism of vasoactive peptides. Biochem. Phannacol. 34:2309-2317; 1985. Pardridge, W. M. Mechanisms of neuropeptide interaction with the blood-brain battier. Ann. NY Acad. Sci. 481:231-248; 1986. Patterson, T. A.; Schulteis, G.; Atvarado, M. C.; Martinez, J. L., Jr.; Bennett, E. L.; Rosenzweig, M. R. Influence of opioid peptides on memory formation in the chick. Behav. Neurosci. 103:429--437; 1989. Rigter, H.; Hannan, T. J.; Messing, R. B.; Martinez, J. L., Jr.; Vasquez, B. J.; Jensen, R. A.; Veliquette, J.; MeOaugh, J. L. Enkephalins interfere with acquisition of an active avoidance response. Life Sci. 26:337-345; 1980. Rigter, H.; Jensen, R. A.; Martinez, J. L., Jr.; Messing, R. B.; Vasquez, B. J.; Liang, K. C.; McGaugh, J. L. Enkephalins and

[LEU]ENKEPHALIN UPTAKE AND IN VIVO METABOLISM

919

fear-motivated behavior. Proc. Natl. Acad. Sci. USA 77:3729-3732; 1980. Schulteis, G.; Hruby, V. J.; Martinez, J. L., Jr. Stimulation and antagonism of opioid delta receptors produce opposite effects on active avoidance conditioning in mice. Behav. Neurosci. 102:678686; 1988. Schulteis, G.; Janak, P. H.; Derrick, B. E.; Martinez, J. L., Jr. Endogenous opioid peptides and learning and memory. In: Szekely, J. I., Ramabadran, K., eds. Opioid peptides, vol. 4. Boca Raton: CRC Press; 1989:in press. Schulteis, G.; Martinez, J. L., Jr. ICI 174,864, a selective delta opioid antagonist, reverses the learning impairment produced by [leu]enkephalin. Psychopharmacology (Berlin), in press; 1989. Shibanoki, S.; Weinberger, S. B.; Ishikawa, K.; Martinez, J. L., Jr. HPLC-ECD determination of the enzymatic hydrolysis of [leu]- and [met]enkephalin in rat plasma. Proc. Int. Narc. Res. Conf., in press; 1989. Singer, E. A.; Mitra, S. P.; Carraway, R. E. Plasma protein(s) yields met-enkephalin-related peptides in near-micromolar concentrations when treated with pepsin. Endocrinology 119:1527-1533; 1986. Van Loon, G. R.; Pierzchala, K.; Brown, L. V.; Bobbitt, F. A. Plasma free and cryptic met enkephalin responses to stress. In: Tucek, S., ed. Synaptic transmitters and receptors. New York: John Wiley &

Sons; 1987:355-364. 37. Verhoef, J. C.; Scholtens, H.; Vergeer, E. G.; WiRer, A. Destyrl-"/-endorphin (DTTE) and des-enkephalin-',/-endorphin (DE~/E): Plasma profile and brain uptake after systemic administration in the rat. Peptides 6:467-474; 1985. 38. Verhoef, J. C.; van den Wildenberg, H. M. Des-enkephalin",/-endorphin: bioavailability in rats following the subcutaneous and intramuscular route of administration. Regul. Peptides 14:113-124; 1986. 39. Weinberger, S. B.; Gehrig, C. A.; Martinez, J. L., Jr. [Leu]enkephalin and its delta selective analog, d-pen2-[d-penS]-enkephalin, impair avoidance conditioning in an automated shelf-jump task in rats. Soc. Neurosci. Abstr. 14:1029; 1988. 40. Weinberger, S. B.; Martinez, J. L., Jr. Differential effects on active avoidance performance and locomotor activity of two major enkephalin metabolites, Tyr-Gly-Gly and des-Tyr-[leu]enkephalin. Life Sci. 43:769-776; 1988. 41. Weinberger, S. B.; Martinez, J. L., Jr. Characterization of hydrolysis of [leu]enkephalin and D-ala2-[L-leu]enkephalin in rat plasma. J. Pharmacol. Exp. Ther. 247:129-135; 1988. 42. Ziring, B.; Sheppard, S.; Kreek, M. J. Reversed phase thin-layer chromatography for the separation of [3-endorphin, [3-1ipotropin and enkephalins. Int. J. Pept. Protein Res. 22:32-38; 1983.

31.

32.

33. 34.

35. 36.