Abstract
(R)-N-(2-Heptyl)-N-methylpropargylamine (R-2HMP) is a monoamine oxidase inhibitor and putative antiapoptotic agent analogous to (R)-deprenyl. In the rat, the major amine metabolites of R-2HMP have been identified as (R)-N-2-heptylmethylamine (R-2HMA), (R)-N-2-heptylpropargylamine (R-2HPA), and (R)-2-heptylamine (R-2HA). After R-2HMP was administered s.c. to male Wistar rats, it was observed that the greatest concentration was of the original drug followed in decreasing order byR-2HMA, R-2HPA, and R-2HA in brain, liver, and plasma at all times after administration. The greatest concentrations of the three metabolites were found in brain followed by liver and plasma, and the peak concentrations occurred between 15 and 30 min after administration. After oral administration, the liver contained the greatest concentrations of drug and metabolites, and, again, the peak concentrations occurred at about 15 min. In all cases, depropargylation appears to occur at a faster rate than demethylation. After s.c. administration, R-2HMP and its metabolites exhibited biexponential redistribution and elimination losses. Half-lives of the compounds in brain for the redistribution phase were: R-2HMP, 10 min;R-2HMA, 11 min; R-2HPA, 16 min; andR-2HA, 15 min.
A series of alkyl-methyl-propargylamines, exemplified by (R)-N-(2-heptyl)-N-methylpropargylamine (R-2HMP)2, which is analogous to (R)-deprenyl [also named (−)-Deprenyl, l-Deprenyl, Selegiline] and to pargyline, were originally developed as monoamine oxidase (MAO) type B inhibitors in an attempt to improve the properties of such compounds and reduce side effects (Yu et al., 1992). More recently, these alkyl-methyl-propargylamines, such asR-2HMP, have been shown to have antiapoptotic actions similar to those of R-deprenyl, by protecting against DSP-4 [N-(2-chloroethyl)-N-ethyl-2-bromobenzylamine]-induced neurodegeneration of noradrenergic axons and terminals (Yu et al., 1994a,b; Zhang et al., 1996), reducing delayed neuronal death of hippocampal pyramidal cells after a carbon monoxide ischemic insult (Paterson et al., 1997a; Paterson and Tatton, 1998), and preventing apoptosis in cerebellar granule cells induced by cytosine arabinoside (Paterson et al., 1998). Evidence also shows that, as with deprenyl, this action is independent of MAO inhibition, and may be due to the desmethyl metabolite, (R)-2-heptylpropargylamine (Paterson et al., 1997b; Durden et al., 1998). Information on the metabolism of the alkyl-methyl-propargylamines would be useful in understanding the mechanism of action of these drugs, and we hypothesize that such metabolism may be inferred from the metabolism of deprenyl or of pargyline.
Deprenyl has been shown to be dealkylated to form its major amine metabolites, amphetamine (Reynolds et al., 1978, 1979; Philips, 1981;Karoum et al., 1982; Yoshida et al., 1986; Salonen, 1990), methamphetamine (Karoum et al., 1982; Yoshida et al., 1986; Heinonen et al., 1989; Salonen, 1990), desmethyldeprenyl (N-propargylamphetamine, nordeprenyl) (Yoshida et al., 1986;Heinonen et al., 1989; Salonen, 1990; Reimer et al., 1993; La Croix et al., 1994; Szebeni et al., 1995). These compounds, in turn, are further metabolized by ring hydroxylation to p-hydroxyamphetamine (Philips, 1981), p-hydroxy-N-propargylamphetamine (Yoshida et al., 1986), and p-hydroxymethamphetamine (Lengyel et al., 1997) or by β-carbon hydroxylation to (1S,2R) norephedrine, (1R,2R) norpseudoephedrine, (1S,2R) ephedrine, (1R,2R) pseudoephedrine (Shin, 1997). It has been shown that stereochemical configuration is maintained throughout these metabolic steps (Heinonen et al., 1994; Szökö and Magyar, 1995; Shin, 1997).
The other major N-methyl-N-propargyl type of MAO-B inhibitor, pargyline, has been shown similarly to be dealkylated to benzylamine (Durden et al., 1976),N-methylbenzylamine, N-methylpropargylamine, andN-propargylbenzylamine (Weli et al., 1982). Additional metabolism may result in N-oxides such as pargylineN-oxide (Weli et al., 1982).
Because of the structural similarities between deprenyl, pargyline, and 2HMP, one would expect the major metabolites of R-2HMP to be formed by dealkylation to the desmethyl, despropargyl, desmethyl-despropargyl, and perhaps the desalkyl amines as shown in Fig. 1.
In this report, we: 1) describe the synthesis of R-2HMP via various intermediates, which also happen to be the amine metabolites proposed above; 2) demonstrate that the above amines were produced during the metabolism of R-2HMP in the rat; and 3) quantitate them in rat liver, brain, and plasma at various times after administration and in rat urine, using deuterium-labeled analogs.
Materials and Methods
Water was purified by reverse osmosis followed by ion exchange (Barnstead Nanopure) to 18 megohms. HPLC- or distilled-in-glass-grade ethyl acetate and hexane were obtained from Caledon Laboratories (Georgetown, Ontario, Canada). Other solvents and chemicals were reagent grade or better: dichloromethane, ether, methanol, and sodium hydroxide (BDH Inc., Ontario, Canada). (+/−)-2-Heptylamine,l-(+)-tartaric acid, 2-heptanone, and ethyl chloroformate were obtained from Lancaster Synthesis Inc. (Windham, NH, via Caledon), and the remainder, lithium aluminum hydride, lithium aluminum deuteride, propargyl bromide, N-methylpropargylamine, and pentafluorobenzoyl chloride (PFB-Cl) were obtained from Aldrich (Milwaukee, WI). The reagent (S)-N-trifluoroacetylprolyl chloride was synthesized in this laboratory (Durden et al., 1997) following the method of Lim et al. (1986) for (S)-N-heptafluorobutyrylprolyl chloride.
Synthesis of (R)-2-heptylamine (R-2HA), (R)-N-2-heptylmethylamine (R-2HMA), (R)-N-2-heptylpropargylamine (R-2HPA), and R-2HMP.
R-2HA.
R-2HA was prepared from the racemate by repeated recrystallizations of its l-tartrate salt from methanol (Mazur, 1970; Durden et al., 1998). The optical purity was 99.7% R-2HA 0.3%S-2HA after nine recrystallizations, as assessed by gas chromatography (GC;Durden et al., 1997). The free base was isolated by basification with sodium hydroxide, extraction with ether, and distillation of the product (b.p. 140–141°C, recovery 68%).
R-2HMA.
Without isolating the free base, the l-tartrate salt ofR-2HA was suspended in ether/water and extractively derivatized with ethyl chloroformate and aqueous sodium hydroxide. Reduction of the carbamate with lithium aluminum hydride in ether gaveR-2HMA in 80% yield (b.p. 73–80°C/30 mm; maleate salt, m.p. 76–77°C).
R-2HPA.
A solution of R-2HA (2 equivalents) and propargyl bromide was refluxed for 72 h in ether, and the solution was treated with 4 N HCl and evaporated to dryness. The residue was alkalinized with NaOH, extracted with ether, and the product was distilled in 55% yield (b.p. 88–93°C/30 mm; HCl salt m.p. 83.5–84.5°C).
R-2HMP.
A mixture of R-2HMA, anhydrous sodium carbonate, and propargyl bromide in ether was refluxed for 72 h. The filtered mixture was treated with 6 N HCl and evaporated to dryness. The free base was isolated by extracting the alkalinized residue with ether followed by distillation (b.p. 103–110°C/30 mm; HCl salt m.p. 130.5–131°C). The identities of the above were confirmed by their mass and NMR spectra and their chemical purities by elemental analysis.
Synthesis of Deuterium-Labeled Internal Standards.
1,1,1,3,3-Pentadeutero-2-heptylamine (2HA-D5)
2-Heptanone was heated with NaOD in D2O for 8 h at 100°C. After cooling, the aqueous layer was removed. Four more such exchanges were carried out, and then the ketone was extracted into ether to give crude 2-heptanone-1,1,1,3,3-D5[80% recovery; >95% D5, assessed by mass spectrometry (MS)], which was converted into the oxime by heating at 60°C with hydroxylamine hydrochloride in NaOD/D2O. The crude product was isolated by extraction with ether and then reduced with lithium aluminum hydride. The product was purified by distillation (b.p. 141–145°; HCl salt, m.p. 74–75°C).
N-trideuteromethyl-2-heptylamine (2HMA-D3), N-methyl-1,1,1,3,3-pentadeutero-2-heptylamine (2HMA-D5) and N-trideuteromethyl-1,1,1,3,3-pentadeutero-2-heptylamine (2HMA-D8).
2HA-D5 treated with ethyl chloroformate and triethylamine in dichloromethane gave the carbamate in 90% yield. Reduction with lithium aluminum hydride yielded 2HMA-D5; reduction with lithium aluminum deuteride gave 2HMA-D8 which was isolated as its oxalate salt (m.p. = 76–77°). Reaction of 2-heptylamine with ethyl chloroformate and triethylamine in dichloromethane followed by reduction with lithium aluminum deuteride gave 2HMA-D3.
N-trideuteromethyl-N-[1,1,1,3,3-pentadeutero-2-heptyl]-propargylamine (2HMP-D8).
2HMA-D8 was refluxed 24 h in ether with anhydrous sodium carbonate and propargyl bromide. The product was distilled (b.p. 170–195°; HCl salt m.p. = 127–129°).
N-[1,1,1,3,3-pentadeutero-2-heptyl]-propargylamine (2HPA-D5).
An ether solution of 2HA-D5 (2 eq.) and propargyl bromide was refluxed for 48 h, treated with ethanolic HCl, and evaporated to dryness. The residue was basified and extracted with ether. The product was isolated as its oxalate salt (m.p. = 132–133°C).
The relative amount of deuterium incorporation was assessed by MS from the molecular ion region of the amine or of the pentafluorobenzoyl (PFB) derivative.
Animals.
All procedures involving animals were performed with the approval of the University of Saskatchewan Animal Care Committee, in accordance with guidelines established by the Canadian Council on Animal Care and the National Institutes of Health. Male Wistar rats (200–250 g, Charles River, Montreal, Canada) were housed in groups under a 12-h dark/light cycle with ad libitum access to food and water.
Identification of metabolites of 2HMP.
Approximately equimolar amounts of R-2HMP · HCl and racemic 2HMP-D8 · HCl were dissolved in saline and administered (s.c.) to male Wistar rats at a total dose of 19.2 mg/kg (i.e., 9.2 and 10.0 mg/kg, respectively, calculated as the free base). Control rats were administered saline. In some experiments,R-2HMP alone was administered. After 30 min, the rats were sacrificed by stunning followed by cervical dislocation, and the brain and liver were removed. The brains were cut sagitally into halves, and livers were cut into approximately 1-g sections, weighed, and frozen on dry ice. The tissues were stored at −70°C until processed. These tissues were thawed and homogenized in 5 ml of 0.1 N perchloric acid and centrifuged at 10,000g for 30 min at 4°C. The supernatants were extracted three times with 1 ml of ethyl acetate/hexane 1:1 (v/v). The organic phase was saved as an acid extract (extract 1). To the aqueous phase was added 1 ml of 10 N NaOH and PFB-Cl (10 μl). The tubes were capped and shaken gently for 1 h at 20°C. The solutions were then extracted three times with ethyl acetate/hexane (1:1, v/v), with centrifugation at 1200 rpm, and the basic extracts were pooled (extract 2) and reduced in volume to about 200 μl in 12 × 32-mm GC auto sampler vials.
The organic layer from the acid extract 1 was dried down under nitrogen, and 100 μl of trifluoroethanol (TFEOH) and 10 μl of PFB-Cl were added with vortex mixing, and the mixture was heated for 1 h at 80°C. Ethyl acetate/hexane (1:1, v/v, 500 μl) was added and the organic layer was washed three times with phosphate buffer (pH 6). The organic layer was transferred to GC vials and reduced in volume, initially to 200 μl.
In other experiments the supernatant from the initial perchlorate step was passed through 0.2-μm syringe filters and lyophilized to dryness. The residue was treated with TFEOH-PFB-Cl as above.
Quantitation of 2HMP and basic metabolites in blood and tissue.
R-2HMP was administered to male Wistar rats at a dose of 10 mg/kg either s.c. or orally by gavage. The rats were sacrificed at time intervals ranging from 5 to 240 min after administration and, as above, samples of blood, liver, and whole brain were obtained and stored at −70°C. The samples (approximately 1 g of tissue) were homogenized (Brinkmann polytron) in 0.1 N perchloric acid (5 ml) containing known amounts of the internal standards 2-HMP-D8, 2-HA-D5, 2HMA-D3, and 2-HPA-D5. Homogenates (and controls) were centrifuged at 4°C for 15 min at 10,000g. The supernatant was removed to a new test tube and extracted three times with 2 ml of ethyl acetate/hexane (1:1, v/v), and the organic layer (extract 1) discarded. The aqueous phase was centrifuged for 15 min at 20,000g. The supernatant was derivatized as above and the basic extract 2 was transferred into GC auto sampler vials.
Standard calibration curves containing various amounts of the amines (e.g., 0–1000 ng R-2HMP for p.o. experiments, 0–5000 ng for s.c. experiments, and appropriately lower amounts for metabolites) with constant amounts of the deuterated internal standards (in perchloric acid) were run in conjunction with every experiment. It was assumed that the extraction of the drugs and deuterated internal standards occurred at equal efficiencies and that any extraction or derivatization losses were compensated for by the use of the deuterated standards. Tissue blanks were of the order of 0 to 1.5 ng/g, giving a minimum detectable quantity of R-2HMP of 3 to 5 ng/g. Metabolites gave similar values.
The location of 2HMP and its metabolites in whole blood was examined by measuring the concentration in plasma and red blood cells (RBC) and binding to either plasma or RBC by addition of Triton X-100. Eight rats received, s.c., 10 mg/kg R-2HMP. After 30 min, the rats were sacrificed and as much blood as possible was collected from each. The whole blood samples were centrifuged in a clinical centrifuge at 4°C; the separated plasma and RBC fractions from individual rats were pooled and the volumes measured. One-milliliter aliquots of plasma or RBC were then taken. To half of the aliquots was added 100 μl of 0.1% Triton X-100 in water, and to the rest 100 μl of water with mixing. Internal standards and 3 ml of 0.1 N HClO4 were then added and mixed, and the samples were centrifuged at 20,000g for 15 min. An additional 1 ml of HClO4 was added to each tube and again they were centrifuged. Supernatants were derivatized with PFB-Cl as above.
Urinary studies.
R-2HMP, in saline, was administered s.c. to male Wistar rats at a dose of 10 mg/kg. Control rats received saline. The rats were kept in metabolic cages with food and water ad libitum. Urine was collected for two 24-h periods. Each urine collection was transferred to volumetric flasks and taken to a volume of 25 ml with Nanopure water. The diluted urines were divided into 2.5-ml aliquots, which were stored at −20°C until analysis. Before analysis, the urines were thawed and 0.5-ml volumes, to which the deuterium-labeled internal standards were added, were used. The urines were mixed with 0.1 M HClO4 (3 ml), centrifuged, and then extracted with ethyl acetate/hexane 1:1 (v/v) to obtain an acid extract. The aqueous layer was then made basic (pH >10) with 10 M NaOH and reacted with PFB-Cl as above; the derivatives of the basic metabolites were then extracted into ethyl acetate/hexane and quantitated.
GC-MS Analysis.
Analyses were performed using a Hewlett-Packard (Palo Alto, CA) 5700 GC, helium carrier gas, interfaced to a VG 7070F double focusing mass spectrometer via an open split interface. For most of the analyses the GC column was a J & W DB1701 30 m, 0.32-mm i.d. with 0.25-μm film and a 1 m × 0.53 mm i.d. deactivated retention gap precolumn. A Supelco (Bellefonte, PA) SPB-1301 column was also used in later experiments to improve GC separation. Injections (1 μl) were made by an HP7673A auto-sampler directly into the 0.53-mm i.d. precolumn with no splitting. The solvent peak was removed by flushing the open split interface with increased helium flow to reduce ion source loading.
Full spectra were recorded in electron impact (EI) mode at 1000 resolution and, for quantitation, ion signals were recorded by selected ion monitoring at 3000 or 5000 resolution. The ions used for quantitation of the drug and the PFB derivatives of the metabolites were for R-2HMP and 2HMP-d8,m/z 96.0813 and 102.1190; for R-2HA and 2HA-D5, m/z 238.0291 and 241.0480; for R-2HMA and 2HMA-D8,m/z 252.0448 and 255.0636; and forR-2HPA and 2HPA-D5,m/z 276.0448 and 279.0636. Methylpropargylamine (MPA) was detected at m/z 263.0370 and its concentration at 30 min was estimated by comparing its signal with that of 2HMA-D8.
Pharmacokinetic Calculations.
Pharmacokinetic calculations of the s.c. data were performed by a program written (D.A.D., unpublished data) in Microsoft Excel using a noncompartmental triexponential model, with two loss terms (redistribution and elimination) and one increase term (absorption or metabolic formation). The program was verified using examples in various textbooks (Gabrielsson and Weiner, 1977; Welling, 1986; Shargel and Yu, 1993) and assumed extravascular administration: where t is in minutes and A and α are the intercept coefficient (nanograms per gram) and rate constant for the α (redistribution) phase, B and β are the intercept and rate constant for the β (elimination) phase, and C and γ are the intercept and rate constant for the absorption (metabolic appearance) phase, with theoretically C = A + B. These coefficients were initially determined by the method of residuals (MOR; Gabrielsson and Weiner, 1977; Welling, 1986; Shargel and Yu, 1993), and then refined for A, B, C, α, and γ using variance-weighted nonlinear least-squares regression (NLR; Gabrielsson and Weiner, 1977) with the Excel Solver iterative function (Billo, 1997). For these iterations β was not changed and C = A + B (Shargel and Yu, 1993). The s.c. results are shown in Table 1. Because a shorter time interval was used for the p.o. data, it was calculated assuming a biexponential model, i.e., with only one redistribution/elimination term, and the results are in shown Table2. The area under the curve (AUC) and area under the moment curve values were calculated by the trapezoidal method and half-lives from the rate constants (Shargel and Yu, 1993). The apparent volume of distribution was calculated from the dose and C0 intercepts (i.e., A + B), and the rates of formation of the metabolites were calculated from the data over the 0- to 5-min period in a linear fashion (Durden and Philips, 1980) and from the exponential equation and are shown in Table3.
Results
Identification of Metabolites.
Figure 2 shows the EI mass spectra ofR-2HMP (plus 2HMP-D8) and metabolites, after reaction with PFB-Cl. These are from the basic extract 2 from tissues of a rat injected with the protio-deuterio mixture.R-2HMP does not react with PFB-Cl and was detected as the free base (Fig. 2a). Because the deuterium-labeled 2HMP-D8 contains five deuterium atoms in the 1 and 3 positions of the alkyl chain and three on the N-methyl group, the expected metabolites would be 2HA-D5, 2HMA-D8, and 2HPA-D5, which are shown in Fig. 2, b–d (in order of elution from the GC). Similar spectra were obtained from the lyophilized supernatants, which were also reacted with PFB-Cl.
It was not possible to obtain a spectrum of MPA from basic extracts or lyophilized extracts of the rat tissues as the intensity was extremely low and another compound tended to overlap using either GC column. MPA was detected when the molecular ion, m/z263.0370, was followed under high-resolution selected ion monitoring conditions and had the same retention time as that of authentic MPA.
From the acid fraction 1 we were able to obtain a spectrum for the major metabolite, 2HMA and 2HMA-D8, similar to that in Fig. 2c. The concentrations of the other amines were too low to detect. We did not detect any acidic or neutral compounds that had reacted with the TFEOH-derivatizing agent.
Quantitation of R-2HMP and Amine Metabolites after Administration of R-2HMP.
The time profiles of R-2HMP, R-2HMA,R-2HPA, and R-2HA are shown in Fig.3 for s.c. and in Fig.4 for oral administration and the pharmacokinetic data in Tables 1 and 2. The volume of distribution ofR-2HMP in plasma after s.c. administration was calculated to be between 8 and 11 liters (by NLR or MOR, respectively) and the mean residence time at 54 min (as area under the moment curve0–240/AUC0–240) (Shargel and Yu, 1993). After p.o. administration, the volume of distribution was calculated as 10 liters (NLR).
After s.c. or oral administration of R-2HMP the major compound found in the brain, liver, and plasma was the drug itself. In all tissues the major metabolite was R-2HMA, followed byR-2HPA and R-2HA. MPA was only detected at 30 min after s.c. administration of 10 mg/kg R-2HMP and the concentrations were estimated to be 8.3 ng/g in brain, 1.2 ng/g in liver, and 1.7 ng/g in plasma. In the 240-g rats, the peak amount of 2HMP measured in the whole brain corresponds to between 0.3 and 0.4% of the original dose and in the liver to about 1.4% (Table 1).
Figure 5 shows the results of the pharmacokinetic curve fitting by the two methods for the s.c. data. Figure 5a shows the curve fit to the data for rat brainR-2HMP using the MOR with triexponential (absorption, redistribution, and elimination phases) or biexponential (absorption and redistribution/elimination phases) models. The improvement of NLR analysis over the initial MOR analysis is shown in Fig. 5b for rat liver R-2HMP data. For the p.o. administration values of each compound only a biexponential analysis was attempted, due to the shorter time range (no elimination phase) and greater scatter of the data. The p.o. data for liver indicated that there may be a lag time but again it was not reasonable to calculate one due to the scatter of the data.
Distribution of R-2HMP and Metabolites in Rat Plasma and RBC.
After s.c. administration of 10 mg/kg R-2HMP, the drug was found to be 20 to 30% more concentrated in plasma than in RBC (617 ± 50 ng/ml versus 472 ± 47 ng/ml, mean ± S.D.,P = .0001), but did not appear to be affected by the addition of Triton X-100 (580 ± 30 versus 483 ± 61, plasma versus RBC, respectively). For R-2HMA, there was no significant difference in concentration between plasma and RBC (111 ± 45 versus 122 ± 14 ng/ml), but the addition of Triton X-100 did increase the levels in both plasma and RBC to 133 ± 16 and 165 ± 36 ng/ml, respectively (P = .04).R-2HA was found to be approximately the same concentration in plasma and RBC (10 ± 3 and 8 ± 5 ng/ml) and its levels were unaffected by Triton X-100. Similarly the levels ofR-2HPA in plasma were unaffected by the addition of Triton X-100 (65 ± 2 and 58 ± 4 ng/ml, no Triton X-100 versus Triton X-100, respectively). We were not able to measureR-2HPA in RBC because the signal levels were too low.
Urinary Excretion of R-2HMP and Metabolites.
Of the 2400 μg (14 μmol) of R-2HMP that was administered to each 240-g rat (Table 4), approximately 1.3 μmol, i.e., 10%, of the drug plus major metabolites was recovered in the first 24 h. The major excreted compound was unchanged R-2HMP. Hydrolyzing the urines with HCl did not increase the levels of the metabolites. When urine was heated with HCl at temperatures above 110°C, HCl reacted with the propargyl group by adding across the acetylene bond to make two new chlorinated alkenes, and the measured amount of R-2HMP andR-2HPA decreased. To examine the reaction, R-2HMP was heated with HBr and two similar bromo addition products were characterized by GC-MS.
Discussion
To be able to identify the metabolites of R-2HMP, rats were administered an equimolar mixture of R-2HMP and 2HMP-D8. Extracts of the supernatants of the homogenates were obtained under three conditions. First, because we expected the amine metabolites, a basic extract was reacted with PFB-Cl to derivatize free amine groups. Second, in attempts to detect other compounds, an acid extract was reacted with PFB-Cl and TFEOH to esterify possible acid groups as well as acylate the amino and alcohol groups. Third, a lyophilized acidic supernatant was reacted with PFB-Cl and TFEOH, so that we could search for all compounds at once. This extract, however, would be expected to contain a large number of extraneous compounds. Because both protio and deuterio isotopomers were administered, one would expect the GC chromatogram to contain new GC peaks (compared with that for a saline-administered rat) with mass spectra containing ions in pairs with a mass separation corresponding to the number of deuterium atoms remaining. Using the DB1701 column, the deuterio and protio isotopomers eluted essentially in the same GC peak with the deuterio amine slightly ahead. With the SPB-1301 column, the deuterio peak preceded the protio peak by a few seconds, giving almost two separate GC peaks. Thus, to obtain the spectra such as in Fig. 2, several spectra were averaged to include both isotopomers.
We have observed that the fragmentation in the EI mass spectra of various derivatives of 2-heptylamines is dominated by the loss of C5H11 from the alkyl chain, i.e., the molecular ion M less 71 Da. For the deuterated 2-heptylamines this would correspond to loss of C5H9D2to give the M-73 fragment ion.
The spectra of R-2HMP and 2HMP-D8(Fig. 2a) of the PFB derivatives of R-2HA and 2HA-D5 (Fig. 2b), of R-2HMA and 2HMA-D8 (Fig. 2c), and of 2HPA and 2HPA-D5 (Fig. 2d) are all characterized by intense ion signals due to M-71 or M-73 and molecular ions of low intensity. The spectra and GC retention times were confirmed by comparison with those of authentic compounds. The identity of each metabolite was then verified using negative ion chemical ionization, which confirmed the presence of (M-HF)− ions.
The presence of the metabolites, R-2HMA, R-2HPA, and R-2HA is consistent with our prediction based on the previous investigations of deprenyl metabolism, which is dealkylated by rats, mice, dogs, and humans, to form methamphetamine, desmethyldeprenyl (N-propargyl-amphetamine), and amphetamine (Reynolds et al., 1978, 1979; Philips, 1981; Karoum et al., 1982;Yoshida et al., 1986; Heinonen et al., 1989; Salonen, 1990; Reimer et al., 1993; La Croix et al., 1994; Szebeni et al., 1995). These metabolites are formed by oxidative dealkylation of deprenyl by cytochrome P-450 isozymes (Yoshida et al., 1986) and are ascribed to atypical metabolism by cytochrome P-450 2D6 (Grace et al., 1994).
There arose the possibility of finding a third dealkyl product due to loss of the 2-heptyl group, i.e., MPA. We did not expect this to be a major metabolite because previous work on deprenyl indicated demethylation and depropargylation to be predominant, and its concentration in brain at 30 min represents less than 1% of the total of the other metabolites.
The ratio of depropargylation to demethylation is similar for bothR-2HMP and R-deprenyl. After either s.c. (Fig. 3, Table 1) or oral (Fig. 4, Table 2) administration of R-2HMP to the rat, the AUC values show that the major metabolite isR-2HMA, at approximately three times the concentration ofR-2HPA, in brain, liver, and plasma. In the rat urine (Table4), we observed that R-2HMA was excreted in about six times greater quantities than was R-2HPA. Thus, in the rat, depropargylation of R-2HMP is more efficient than is demethylation. This parallels the metabolism of R-deprenyl in the rat, in which methamphetamine was the major metabolite followed by the desmethyl compound and amphetamine (Yoshida et al., 1986;Kalász et al., 1990; Lengyel et al., 1997).
Depropargylation appears to be more important than demethylation in other species, also; in human plasma, the overall decreasing order of concentration of deprenyl metabolites was methamphetamine, desmethyldeprenyl, amphetamine (La Croix et al., 1994; Mascher et al., 1997a,b), with desmethyldeprenyl being at its most concentrated within 1 to 2 h after dosing (Reimer et al., 1993; Heinonen et al., 1994). A similar pattern has been found in dog plasma (Salonen, 1990) and in rat liver microsome incubations (Yoshida et al., 1986). Again, in human urine, methamphetamine was the major metabolite followed either by desmethyldeprenyl (Tarjányi et al., 1998) or amphetamine (Shin, 1997).
That depropargylation of R-2HMP is more important than demethylation is confirmed also by the calculated rates of formation from the tertiary amine R-2HMP; the rate forR-2HMA is two to three times greater than that ofR-2HPA in all tissues (Table 3).
After s.c. administration, the concentrations of the drug,R-2HMP, and its metabolites, R-2HMA,R-2HPA, and R-2HA, followed a triexponential curve consisting of absorption (or appearance due to metabolic formation), followed by a redistribution or metabolic decrease and a final elimination decrease. The data for R-2HA in liver and plasma were analyzed using a biexponential model as the data were too scattered to confirm a triexponential curve.
The half-life of R-2HMP may be compared with that of deprenyl using the value for the β or redistribution/metabolism phase of 10 to 13 min in brain, plasma, and liver. Heinonen et al. (1989)report a half-life of R-deprenyl in (human) serum of 9 min after an i.v. administration. Corresponding data for deprenyl in whole brain has not been published in detail. However, after 10 mg/kg s.c administration to the rat, Melega et al. (1999) reported a half-life of 9 min over the period 10 to 60 min in caudate putamen. Similarly, data provided by Waldmeier has indicated that the redistribution half-life of R-deprenyl (10 mg/kg) in rat brain was about 7 min, over the period 15 to 30 min the elimination half-life was about 145 min, and the maximum concentration was 2200 ng/g at 15 min. From p.o. plasma data we were able to calculate a redistribution half-life of 27 min, elimination half-life of 135 min, and a volume of distribution of about 30 liter (A.-F. Steulet and P.C. Waldmeier, private communication). The latter values were obtained in experiments similar to those in this report and indicate that although the half-lives are similar, the volume of distribution ofR-deprenyl may be greater than that of R-2HMP. The relative concentrations of drugs to metabolites would indicate thatR-2HMP is metabolized to a lower extent than isR-deprenyl.
Because the concentrations of R-2HMP and metabolites were considerably lower in plasma than in either brain or liver, we decided to ascertain whether R-2HMP and metabolites had been concentrated in RBC that had been discarded when plasma was prepared. The data demonstrates that this was not the case, but thatR-2HMP was 20 to 30% more concentrated in plasma over RBC. The concentrations of the metabolites, R-2HMA andR-2HA, were not significantly different between plasma and RBC. We were not able to measure 2HPA in RBC as the signal levels were too low. The apparent volume of distribution in plasma of between 8 and 11 liters supports the view that R-2HMP is primarily taken up by the tissues, or is bound and is unavailable for metabolism or analysis. This value and the amount excreted indicate a bioavailability of approximately 10% of the administered dose.
Deprenyl has a high lipophilicity and is bound to serum proteins (Kalász et al., 1990), mostly irreversibly. It appears that most of the 2HMP is similarly bound. Similarly, R-2HMP or its metabolites might be expected to have a high lipophilicity due to the presence of the 2-heptyl chain. R-2HMP, R-2HPA, and R-2HA did not appear to be lipid-bound, as there was no effect of addition of Triton X-100 on their levels in plasma or RBC. On the other hand, R-2HMA concentrations did appear to increase 20 to 35% (in plasma and RBC, respectively) after Triton X-100 was added.
Virtually all of the excretion of R-2HMP and metabolites occurred in the first 24 h. It appears that they are excreted as the free amines, as neither R-2HMP nor its metabolites is conjugated with hydrolyzable components. We would not expect the tertiary amine to be conjugated, but the secondary amine metabolites with their free amino group do possess the possibility of being conjugated, for example, as glucuronides. One of the difficulties of measuring any increase of R-HPA after acid hydrolysis is that the acetylene group is labile under the very acidic, high-temperature conditions frequently used for hydrolysis because the acid may add across the triple bond to form addition products.
The ratio of excreted R-2HMP to its metabolites in urine is considerably different from results obtained withR-deprenyl. When R-deprenyl was administered s.c. to rats the major compound excreted in the urine over the first 24 h was methamphetamine (more than 50%) (Lengyel et al., 1997) followed by amphetamine and unreacted deprenyl. In the case of treatment of humans with R-deprenyl, again methamphetamine was observed to be the main metabolite in urine (Lengyel et al., 1997;Tarjányi et al., 1998), at approximately 70% of the total recovered drug plus metabolites, followed by amphetamine with unchanged deprenyl at less than 1% (Shin, 1997). The R-2HMP data (Table 4), in contrast, shows that the unchanged R-2HMP is excreted as 80% and R-2HMA is excreted as less than 20% of the total drug plus metabolites. This again would indicate thatR-2HMP is metabolized at a slower rate than isR-deprenyl.
Conclusion.
The metabolism of (R)-2-heptylmethylpropargylamine is analogous to that of (R)-deprenyl with some similarities and differences. Like deprenyl, R-2HMP is metabolized mostly by depropargylation, and to a lesser extent by demethylation, in both the brain and the liver. The overall rates of metabolism appear to be somewhat different with R-2HMP being metabolized at a slower rate than deprenyl and excreted to a greater extent. Because compounds such as R-2HMP demonstrate antiapoptotic actions similar to those of R-deprenyl and enhance neuronal survival in a manner stereospecific for theR-isomer, the presence of unmetabolized R-2HMP and its metabolite R-2HPA in rat brain bodes well for this process.
Acknowledgments
We thank Ciba Geigy Canada for a grant and Saskatchewan Health for continuing financial support, Saskatchewan Health Services Utilization and Research Commission for a grant to Y.-D. Liu, and RenéMag-Atas, Stephen Ambrose, and Dianne Young for expert technical assistance.
Footnotes
-
Send reprint requests to: D.A. Durden, Neuropsychiatry Research Unit, University of Saskatchewan, 103 Wiggins Rd., Saskatoon, SK, S7N 5E4, Canada. E-mail: david.durden{at}usask.ca
-
↵1 Present address: Agriculture and Agri-Foods Canada, 107 Science Place, Saskatoon, SK, S7N 0X2.
- Abbreviations used are::
- R-2HMP
- (R)-N-(2-heptyl)-N-methylpropargylamine
- R-2HMA
- (R)-N-2-heptylmethylamine
- R-2HPA
- (R)-N-2-heptylpropargylamine
- R-2HA
- (R)-2-heptylamine
- MPA
- methylpropargylamine
- TFEOH
- trifluoroethanol
- PFB
- pentafluorobenzoyl
- PFB-Cl
- pentafluorobenzoyl chloride
- RBC
- red blood cells
- GC
- gas chromatography
- MS
- mass spectrometry
- EI
- electron impact
- MOR
- method of residuals
- NLR
- nonlinear least-squares regression
- AUC
- area under the curve
- MAO
- monoamine oxidase
- 2HA-D5
- 1,1,1,3,3-pentadeutero-2-heptylamine
- 2HMA-D3
- N-trideuteromethyl-2-heptylamine
- 2HMA-D5
- N-methyl-1,1,1,3,3,-pentadeutero-2-heptylamine
- 2HMA-D8
- N-trideuteromethyl-1,1,1,3,3,-pentadeutero-2-heptylamine
- 2HMP-D8
- N-trideutero-methyl-[1,1,1,3,3,-pentadeutero-2-heptyl]-propargylamine
- 2HPA-D5
- N-[1,1,1,3,3-pentadeutero-2-heptyl]-propargylamine
- Received May 14, 1999.
- Accepted October 12, 1999.
- The American Society for Pharmacology and Experimental Therapeutics