Modification of the 5' position of purine nucleosides. 2. Synthesis and some cardiovascular properties of adenosine-5'-(N-substituted)carboxamides

A2A adenosine receptor agonists, antagonists, inverse agonists and partial agonists

The Gs-coupled A2A adenosine receptor (A2AAR) has been explored extensively as a pharmaceutical target, which has led to numerous clinical trials. However, only one selective A2AAR agonist (regadenoson, Lexiscan) and one selective A2AAR antagonist (istradefylline, Nouriast) have been approved by the FDA, as a pharmacological agent for myocardial perfusion imaging (MPI) and as a cotherapy for Parkinson's disease (PD), respectively. Adenosine is widely used in MPI, as Adenoscan. Despite numerous unsuccessful clinical trials, medicinal chemical activity around A2AAR ligands has accelerated recently, particularly through structure-based drug design. New drug-like A2AAR antagonists for PD and cancer immunotherapy have been identified, and many clinical trials have ensued. For example, imaradenant (AZD4635), a compound that was designed computationally, based on A2AAR X-ray structures and biophysical mapping. Mixed A2AAR/A2BAR antagonists are also hopeful for cancer treatment. A2AAR antagonists may also have potential as neuroprotective agents for treatment of Alzheimer's disease.

Exploring the Role of N6-Substituents in Potent Dual Acting 5'-C-Ethyltetrazolyladenosine Derivatives: Synthesis, Binding, Functional Assays, and Antinociceptive Effects in Mice ∇

Structural determinants of affinity of N⁶-substituted-5'-C-(ethyltetrazol-2-yl)adenosine and 2-chloroadenosine derivatives at adenosine receptor (AR) subtypes were studied with binding and molecular modeling. Small N⁶-cycloalkyl and 3-halobenzyl groups furnished potent dual acting A₁AR agonists and A₃AR antagonists. 4 was the most potent dual acting human (h) A₁AR agonist (K_i = 0.45 nM) and A₃AR antagonist (K_i = 0.31 nM) and highly selective versus A_2A; 11 and 26 were most potent at both h and rat (r) A₃AR. All N⁶-substituted-5'-C-(ethyltetrazol-2-yl)adenosine derivatives proved to be antagonists at hA₃AR but agonists at the rA₃AR. Analgesia of 11, 22, and 26 was evaluated in the mouse formalin test (A₃AR antagonist blocked and A₃AR agonist strongly potentiated). N⁶-Methyl-5'-C-(ethyltetrazol-2-yl)adenosine (22) was most potent, inhibiting both phases, as observed combining A₁AR and A₃AR agonists. This study demonstrated for the first time the advantages of a single molecule activating two AR pathways both leading to benefit in this acute pain model.

N6-Substituted 5'-N-Methylcarbamoyl-4'-selenoadenosines as Potent and Selective A3 Adenosine Receptor Agonists with Unusual Sugar Puckering and Nucleobase Orientation

Potent and selective A₃ adenosine receptor (AR) agonists were identified by the replacement of 4'-oxo- or 4'-thionucleosides with bioisosteric selenium. Unlike previous agonists, 4'-seleno analogues preferred a glycosidic syn conformation and South sugar puckering, as shown in the X-ray crystal structure of 5'-N-methylcarbamoyl derivative 3p. Among the compounds tested, N⁶-3-iodobenzyl analogue 3d was found to be the most potent A₃AR full agonist (K_i = 0.57 nM), which was ≥800- and 1900-fold selective for A₁AR and A_2AAR, respectively. In the N⁶-cycloalkyl series, 2-Cl analogues generally exhibited better hA₃AR affinity than 2-H analogues, whereas 2-H > 2-Cl in the N⁶-3-halobenzyl series. N⁷ isomers 3t and 3u were much weaker in binding than corresponding N⁹ isomers, but compound 3t lacked A₃AR activation, appearing to be a weak antagonist. 2-Cl-N⁶-3-iodobenzyl analogue 3p inhibited chemoattractant-induced migration of microglia/monocytes without inducing cell death at ≤50 μM. This suggests the potential for the development of 4'-selenonucleoside A₃AR agonists as novel antistroke agents.

Synthesis and general biological activity of a small adenosine-5'-(carboxamide and sulfanilamide) library

A small library of fifty-five adenosine peptide analogs was synthesized, under the Pilot Scale Library (PSL) Program of the NIH Roadmap initiative, from 2',3'-O-isopropylideneadenosine-5'-carboxylic acid 2. The coupling of amine or sulfanilamide reactants to the free 5'-carboxylic acid moiety of 2, in automated solution-phase fashion, led after acid-mediated hydrolysis to target compounds 3-57 in good yields and high purity. No marked anticancer or antimalarial activity was noted on preliminary cellular testing. Initial screening through the MLPCN program, however, indicates that these analogs may show diverse and interesting biological activities.

Synthesis of novel peptidyl adenosine antibiotic analogs

A small library of peptidyl adenosine antibiotic analogs was synthesized, under the Pilot Scale Library Program of the NIH Roadmap initiative, from 2',3'-O-isoproylideneadenosine-5'-carboxylic acid 2 in excellent yield. The coupling of the amino terminus of L-2-aminophenylbutyric methyl ester to a free 5'-carboxylic acid moiety of 2 followed by sodium hydroxide treatment led to carboxylic acid analog 4. Hydrolysis of this latter gave unprotected nucleoside analog 5. Intermediate 4 served as the precursor for the preparation of novel peptidyl adenosine analogs 6-18 in good yields and high purity through peptide coupling reactions to diverse amine derivatives. No marked anticancer and antimalaria activity was noted on preliminary cellular testing; however these analogs should be useful candidates for other types of biological activity.

Recent developments in adenosine A2A receptor ligands

The development of potent and selective agonists and antagonists of adenosine receptors (ARs) has been a target of medicinal chemistry research for several decades, and recently the US Food and Drug Administration has approved Lexiscan, an adenosine derivative substituted at the 2 position, for use as a pharmacologic stress agent in radionuclide myocardial perfusion imaging. Currently, some other adenosine A(2A) receptor (A(2A)AR) agonists and antagonists are undergoing preclinical testing and clinical trials. While agonists are potent antiinflammatory agents also showing hypotensive effects, antagonists are being developed for the treatment of Parkinson's disease.However, since there are still major problems in this field, including side effects, low brain penetration (for the targeting of CNS diseases), short half-life, or lack of in vivo effects, the design and development of new AR ligands is a hot research topic.This review presents an update on the medicinal chemistry of A(2A)AR agonists and antagonists, and stresses the strong need for more selective ligands at the human A(2A)AR subtype, in particular in the case of agonists.

Highlights on the development of A(2A) adenosine receptor agonists and antagonists

Although significant progress has been made in the past few decades demonstrating that adenosine modulates a variety of physiological and pathophysiological processes through the interaction with four subtypes of a family of cell-surface G-protein-coupled receptors, clinical evaluation of some adenosine receptor ligands has been discontinued. Major problems include side effects due to the wide distribution of adenosine receptors, low brain penetration (which is important for the targeting of CNS diseases), short half-life of compounds, or a lack of effects, in some cases perhaps due to receptor desensitization or to low receptor density in the targeted tissue. Currently, three A(2A) adenosine receptor agonists have begun phase III studies. Two of them are therapeutically evaluated as pharmacologic stress agents and the third proved to be effective in the treatment of acute spinal cord injury (SCI), while avoiding the adverse effects of steroid agents. On the other hand, the great interest in the field of A(2A) adenosine receptor antagonists is related to their application in neurodegenerative disorders, in particular, Parkinson's disease, and some of them are currently in various stages of evaluation. This review presents an update of medicinal chemistry and molecular recognition of A(2A) adenosine receptor agonists and antagonists, and stresses the strong need for more selective ligands at the A(2A) human subtype.

Conjugation of adenosine and hexa-(D-arginine) leads to a nanomolar bisubstrate-analog inhibitor of basophilic protein kinases

Conjugates of oligoarginine peptides with adenine, adenosine, adenosine-5'-carboxylic acid, and 5-isoquinolinesulfonic acid were synthesized and characterized as bisubstrate-analog inhibitors of cAMP-dependent protein kinase. Adenosine and adenine derivatives were connected to the N- or C-terminus of peptides containing four to six L- or D-arginine residues via a linker with a length that had been optimized in structure-activity studies. The orientation of the peptide chain strongly affected the activity of compounds incorporating D-arginines. The biligand inhibitor containing Hidaka's H9 isoquinolinesulfonamide connected to the L-peptide had 65 times higher potency than the corresponding adenosine-containing conjugate, while both types of the conjugate comprising D-peptides had similar low nanomolar activity. Two of the most active adenosine- and H9-peptide conjugates were tested in the panel of 52 different kinases. At 1 microM concentration, both compounds showed strong (more than 95%) inhibition of several basophilic AGC kinases, including pharmaceutically important kinases ROCK II and PKB/Akt.

N(6)-[(hetero)aryl/(cyclo)alkyl-carbamoyl-methoxy-phenyl]-(2-chloro)-5'-N-ethylcarboxamido-adenosines: the first example of adenosine-related structures with potent agonist activity at the human A(2B) adenosine receptor

A new series of N(6)-[(hetero)aryl/(cyclo)alkyl-carbamoyl-methoxy-phenyl]-(2-chloro)-5'-N-ethylcarboxamido-adenosines (24-43) has been synthesised and tested in binding assays at hA(1), hA(2A) and hA(3) adenosine receptors, and in a functional assay at the hA(2B) subtype. The examined compounds displayed high potency in activating A(2B) receptors with good selectivity versus A(2A) subtypes. The introduction of an unsubstituted 4-[(phenylcarbamoyl)-methoxy]-phenyl chain at the N(6) position of 5'-N-ethylcarboxamido-adenosine led us to the recognition of compound 24 as a full agonist displaying the highest efficacy of the series (EC(50) hA(2B)=7.3 nM). These compounds represent the first report about adenosine-related structures capable of activating hA(2B) subtype in the low nanomolar range.

2-Alkynyl-8-aryl-9-methyladenines as novel adenosine receptor antagonists: their synthesis and structure-activity relationships toward hepatic glucose production induced via agonism of the A(2B) receptor

Novel adenosine antagonists, 2-alkynyl-8-aryl-9-methyladenine derivatives, were synthesized as candidate hypoglycemic agents. These analogues were evaluated for inhibitory activity on N-ethylcarboxamidoadenosine (NECA)-induced glucose production in primary cultured rat hepatocytes. In general, aromatic moieties at the 8-position and alkynyl groups at the 2-position had significantly increased activity compared to unsubstituted compounds. The preferred substituents at the 8-position of adenine were the 2-furyl and 3-fluorophenyl groups. In modifying the alkynyl side chain, change of the ring size, cleavage of the ring, and removal of the hydroxyl group were well tolerated. The order of the stimulatory effects of adenosine agonists on rat hepatocytes was NECA > CPA > CGS21680, which is consistent with involvement of the A(2B) receptor. In Chinese hamster ovary cells stably transfected with human A(2B) receptor cDNA, one of the compounds potent in hepatocytes, 15o (IC(50) = 0.42 microM), antagonized NECA-induced stimulation of cyclic AMP production (IC(50) = 0.063 microM). This inhibitory effect was much more potent than those of FK453, KF17837, and L249313 which have been reported to be respectively A(1), A(2A), and A(3) selective antagonists. These findings agree very well with the result that, compared to 15o, these selective antagonists for each receptor subtype showed only marginal effects in rat hepatocytes. These results suggest that adenosine agonist-induced glucose production in rat hepatocytes is mediated through the A(2B) receptor. Furthermore, 15o showed hypoglycemic activity in an animal model of noninsulin-dependent diabetes mellitus, the KK-A(y) mice. It is possible that inhibition of hepatic glucose production via the A(2B) receptor could be at least one of the mechanisms by which 15o exerts its in vivo effects. Further elaboration of this group of compounds may afford novel antidiabetic agents.

Nucleosides and nucleotides. 200. Reinvestigation of 5'-N-ethylcarboxamidoadenosine derivatives: structure-activity relationships for P(3) purinoceptor-like proteins

The non-P(1) and non-P(2) muscle relaxant effect of ATP in rabbit thoracic aorta has recently been attributed to a putative P(3) purinoceptor, which is activated by either adenosine or ATP. Since the physiological roles of this putative P(3) purinoceptor and of a new [(3)H]-5'-N-ethylcarboxamidoadenosine (NECA)-binding protein from rat brain membranes called P(3) purinoceptor-like protein (P(3)LP), due to its ligand specificity, have not been fully elucidated, we needed a specific ligand to obtain further information about the receptor. We examined the structure-activity relationship (SAR) of various 5'-N-substituted-carboxamidoadenosine derivatives toward P(3)LP and discovered a hydrophobic binding region near the 5'-N-substituted-carboxamide group. From the linear alkyl N-substituted derivatives, the N-n-pentyl derivative 10 was found to be the most potent ligand with a K(i) value of 12 nM. In the series of the N-cycloalkyl derivatives, the N-cyclohexyl derivative 27 was the strongest ligand with a K(i) value of 18 nM. On the other hand, the N-substituents having branched alkyl side chains and bulky cycloalkyl groups did not show any potent affinities for P(3)LP. Therefore, the hydrophobic pocket accommodates approximately a 10-carbon-atom-long linear alkyl side chain, while a considerably stronger hydrophobic binding region of about a 5-carbon-atom-long depth exists near the nitrogen atom of the amide group. This pocket also allows substitution with bulky hydrophobic groups since the 5'-N-cycloalkyl derivatives have high affinities. We also examined the receptor selectivity for the selected nucleosides 10 and 27 with 1 [9-(6,7-dideoxy-beta-D-allo-hept-5-ynofuranosyl)adenine, HAK2701] and NECA versus P(1) purinoceptor subtypes, such as adenosine A(1), A(2A), A(2B), and A(3) receptors, and found that 27 is the most selective ligand for P(3)LP.

Adenosine-derived non-phosphate antagonists for P2Y(1) purinoceptors

Novel type antagonists for P2Y(1) adenine nucleotide receptors were synthesized by coupling of adenosine 5'-OH group with oligo-aspartate chain via a carbonyl linker. All these conjugates (AdoOC(O)Asp(n), n = 1-4) inhibited the 2MeSADP-stimulated synthesis of inositol phosphates in 1321N1 human astrocytoma cells stably expressing human P2Y(1) receptors. This inhibitory effect followed the rank order AdoOC(O)Asp(2)> AdoOC(O)Asp(3)> AdoOC(O)Asp(1)> AdoOC(O)Asp(4) with antagonistic constant pA(2) = 5.4 for AdoOC(O)Asp(2). Potency of this non-phosphate inhibitor was comparable with the previously known adenosine 3',5'- and 2', 5'-bisphosphates. Chemical and biological stabilities of these novel adenosine derived antagonists of the nucleotide receptor provide perspectives of their pharmacological implication.

5'-N-substituted carboxamidoadenosines as agonists for adenosine receptors

Novel as well as known 5'-N-substituted carboxamidoadenosines were prepared via new routes that provided shorter reaction times and good yields. Binding affinities were determined for rat A1 and A2A receptors and human A3 receptors. EC50 values were determined for cyclic AMP production in CHO cells expressing human A2B receptors. On all receptor subtypes relatively small substituents on the carboxamido moiety were optimal. Selectivity for the A3 receptor was found for several analogues (1a, 1d, 1h, and 1k). On A1 receptors a number of compounds, but not 5'-N-ethylcarboxamidoadenosine (NECA, 1b), showed small GTP shifts, which could be indicative of lower intrinsic activities at the A1 receptor. At the A2B receptor, derivatives 1i-k with modified ethyl substituents had reduced activities compared to the A2B reference agonist NECA (1b). Thiocarboxamido derivatives (8b and 8c) displayed considerable although decreased A2B receptor activity. The X-ray structure determination of compound 8b was carried out. Due to intramolecular hydrogen bonding between the carboxamido NH and the purine N3 in the crystal structure, the ribose moiety of this compound is in a syn conformation. However, theoretical calculations support that NECA (1b), and less so 8b, can readily adopt both the syn and the anti conformation, therefore not excluding the proposed anti mode of binding to the receptor.

2'-C-Methyl analogues of selective adenosine receptor agonists: synthesis and binding studies

2'-C-Methyl analogues of selective adenosine receptor agonists such as (R)-PIA, CPA, CCPA, NECA, and IB-MECA were synthesized in order to further investigate the subdomain that binds the ribose moiety. Binding affinities of these new compounds at A1 and A2A receptors in bovine brain membranes and at A3 in rat testis membranes were determined and compared. It was found that the 2'-C-methyl modification resulted in a decrease of the affinity, particularly at A2A and A3 receptors. When such modification was combined with N6-substitutions with groups which induce high potency and selectivity at A1 receptors, the high affinity was retained and the selectivity was increased. Thus, 2-chloro-2'-C-methyl-N6-cyclopentyladenosine (2'-Me-CCPA), which displayed a Ki value of 1.8 nM at A1 receptors, was selective for A1 vs A2A and A3 receptors by 2166- and 2777-fold, respectively, resulting in one of the most potent and A1-selective agonists so far known. In functional assay, this compound inhibited forskolin-stimulated adenylyl cyclase activity with an IC50 value of 13.1 nM, acting as a full agonist.

Arteriopathy induced by an adenosine agonist-antihypertensive in monkeys

An adenosine agonist, designated chemically as (R)-N-(2,3-dihydro-1H-inden-1-yl) adenosine or CI-947, was administered orally to 2 males and 2 female cynomolgus monkeys each at 5, 10, 20, and 50 mg/kg of body weight for 2 wk. One male and 1 female given 50 mg/kg were euthanatized on days 10 and 8, respectively, because of poor clinical condition. Emesis was present at 10, 20, and 50 mg/kg. Decreased heart rate and QT prolongation were present at 50 mg/kg. Extramural coronary arterial lesions consisting of medial necrosis with cellular debris and mixed inflammatory cell response in the intima, media, and adventitia were present in 1 male at 20 mg/kg and 1 male at 50 mg/kg at study termination. Similar arterial lesions were present in the small and large intestines and testis of the male at 50 mg/kg. Colonic mucosal erosions with mixed inflammatory cell infiltrates in the lamina propria were seen in this male and in all CI-947 treated females at 10, 20, and 50 mg/kg. Myocardial degeneration and necrosis of myocardial fibers with mononuclear cell infiltrates in the interstititum were noted in the left ventricle of 1 female at 20 mg/kg and in all animals at 50 mg/kg. Renal cortical tubular dilatation with increases in serum creatinine and/or blood urea nitrogen were noted in a control female and animals at 10 and 50 mg/kg. Plasma CI-947 concentration increased with increasing dose. Coronary vascular injury in the monkey was similar to the arterial lesion in CI-947-treated dogs and may relate to the pharmacologic/hemodynamic effects induced by CI-947. When compared with the dog, the monkey appears to be less sensitive to development of arteriopathy, as indicated by lower incidence, at similar systemic exposure levels.

5'-substituted adenosine analogs as new high-affinity partial agonists for the adenosine A1 receptor

5'-(Alkylthio)-, 5'-(methylseleno)-, and 5'-(alkylamino)-substituted analogues of N6-cyclopen-tyladenosine (CPA) were synthesized in 30-50% overall yields. The affinities of these compounds for the adenosine A1 and A2A receptors were determined in rat brain membranes. The 5'-substituted CPA analogues proved selective for the adenosine A1 receptors, displaying affinities in the nanomolar range. The compounds were also evaluated for their ability to stimulate [35S]GTP gamma S binding, also in rat brain membranes. The Ki values in receptor binding studies corresponded well to the EC50 values thus obtained. Intrinsic activities of the compounds were tested in vitro by determining the GTP shift in receptor binding studies as well as the maximal binding of [35S]GTP gamma S. It appeared that the 5'-thio and 5'-seleno derivatives in particular behaved as partial agonists.

Adenosine receptor agonists: synthesis and biological evaluation of the diastereoisomers of 2-(3-hydroxy-3-phenyl-1-propyn-1-yl)NECA

Among the recently reported 2-(ar)alkynyl derivatives of 5'-N-ethylcarboxamidoadenosine (NECA), the (R,S)-2-(3-hydroxy-3-phenyl-1-propyn-1-yl)NECA [(R,S)-PHPNECA or SCH 59761] was found to be a very potent agonist at A1 and A2A receptor subtypes, with a Ki of 2.5 nM and 0.9 nM, respectively. Furthermore, this compound showed an inhibitory activity on platelet aggregation 16-fold higher than NECA, being the most potent anti-aggregatory nucleoside reported so far. Since this compound bears a chiral carbon in the side chain, the diastereoisomer separation was undertaken both by chiral HPLC and by a stereospecific synthetic method. Binding assays have shown that the (S)-diastereomer is about fivefold more potent and selective than the (R)-diastereomer as agonist of the A2A receptor subtype [(S)-PHPNECA, KiA2A = 0.5 nM; (R)-PHPNECA, KiA2A = 2.6 nM]. Functional studies indicated that (S)-PHPNECA possesses marked vasodilating activity and produces a relevant decrease in heart rate. Moreover, the (S)-diastereomer proved to be about ten times more potent than the (R)-diastereomer in inducing cardiovascular effects, in in vivo hemodynamic studies. However, the greatest difference between these two enantiomers resulted in the platelet aggregation test: in fact, the (R)-diastereomer displayed an inhibitory activity similar to that of NECA, whereas the (S)-diastereomer was 37-fold more active than NECA as an inhibitor of rabbit platelet aggregation, induced by ADP. These data suggest that (S)-PHPNECA could be a useful tool to investigate the mode of binding of agonists to the platelet adenosine receptor subtype.

Inactivation of S-adenosyl-L-homocysteine hydrolase by amide and ester derivatives of adenosine-5'-carboxylic acid

S-Adenosyl-L-homocysteine (AdoHcy) hydrolase has been shown to have (5'/6') hydrolytic activity with vinyl (5') or homovinyl (6') halides derived from adenosine (Ado). This hydrolytic activity is independent of its 3'-oxidative activity. The vinyl (or homovinyl) halides are converted into 5'(or 6')-carboxaldehydes by the hydrolytic activity of the enzyme, and inactivation occurs via the oxidative activity. Amide and ester derivatives of Ado-5'-carboxylic acid were prepared to further probe the hydrolytic capability of AdoHcy hydrolase. The oxidative activity (but not the hydrolytic activity) is involved in the mechanism of inhibition of the enzyme by the ester and amide derivatives of Ado-5'-carboxylic acid, in contrast to the inactivation of this enzyme by adenosine-derived vinyl or homovinyl halide analogues during which both activities are manifested.

Inhibition of platelet aggregation by adenosine receptor agonists

2-(Ar)alkoxyadenosines, which are agonists selective for the A2AAR in PC 12 cell and rat striatum membranes, are also agonists at the A2AR coupled to adenylate cyclase (AC) that mediates the inhibition of platelet aggregation. A panel of twelve well-characterized adenosine analogues stimulated human platelet AC and inhibited ADP-induced platelet aggregation at sub- to low-micromolar concentrations with a potency ranking CGS 21680 > adenosine > R-PIA. There were significant correlations between the EC50 of anti-aggregatory activity and either the EC50 of stimulation of platelet and PC 12 cell AC (r2 = 0.66 and 0.67, respectively) or the Ki of inhibition of [3H]NECA binding to the rat striatum membranes (r2 = 0.75). Likewise, platelet AC stimulation correlated well with stimulation of PC 12 cell AC and with [3H]NECA binding (r2 = 0.94 and 0.91, respectively). Ten 2-(ar)alkoxyadenosines stimulated platelet AC at EC50s ranging between 0.16 and 2.3 microM and inhibited platelet aggregation at EC50s ranging between 2 and 30 microM. There were no correlations between the EC50s of anti-aggregatory activity and either the EC50s of the stimulation of platelet or PC 12 AC (r2 = 0.08 and 0.06, respectively) or with the Ki of the inhibition of [3H]NECA binding to the A2aAR in rat striatum (r2 = 0.02). The EC50s of the stimulation of platelet AC correlated with those of the stimulation of PC 12 AC (r2 = 0.48), and also with the Ki of [3H]NECA binding (r2 = 0.71). Each of the 23 adenosines completely inhibited platelet aggregation and thus, functionally, all behaved as full agonists.(ABSTRACT TRUNCATED AT 250 WORDS)

Preclinical toxicity studies of an adenosine agonist, N-(2,2-diphenylethyl) adenosine

N-(2,2-Diphenylethyl)adenosine (DPEA) has been identified as a potential antipsychotic agent acting via stimulation of adenosine receptors. The projected human therapeutic dose, based on animal studies, is 2-3 mg/kg. DPEA has been tested for potential toxicity in mice, rats, dogs and monkeys. Following single oral doses, median lethal dose values were approximately 10-fold greater in rats than in mice, although similar clinical signs including reduced activity, prostration, and necrosis of the tail were seen in both species. DPEA was well tolerated at daily doses up to 40 mg/kg in rats for 2 weeks. A no observed effect level (NOEL) was not identified in the dog or monkey studies. Reduced activity, dacryorrhea, ptosis, hypothermia, necrosis of the tail, and death occurred in rats given 120 and 160 mg/kg. Pathologic changes consisted of pancreatitis, gastric erosion/ulceration, lymphocyte depletion of the thymus, and pulmonary congestion and hemorrhage at 80 mg/kg or greater. In dogs, sporadic emesis was noted at 12.5 mg/kg and greater, and significant pathologic changes consisted of coronary arteritis associated with myocardial lesions and lymphocyte depletion at 25 and 50 mg/kg, pancreatic acinar necrosis at 50 mg/kg, and renal tubular degeneration at 12.5 mg/kg and greater. Emesis and depression were noted at 25 and 50 mg/kg in monkeys. Renal tubular dilatation and degeneration at 25 and 50 mg/kg were noted in the monkeys. These studies demonstrated that DPEA produced a range of adverse effects in common laboratory animal species.