Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Oct 19.
Published in final edited form as: J Pharmacol Exp Ther. 1989 Jul;250(1):79–85.

Adenosine Receptor Prodrugs: Towards Kidney-Selective Dialkylxanthines

SUZANNE BARONE 1, PAUL C CHURCHILL 1, KENNETH A JACOBSON 1
PMCID: PMC3476460  NIHMSID: NIHMS410013  PMID: 2746513

Abstract

XAC (xanthine amine congener, 8-{4-[(2-aminoethyl)-aminocarbonylmethyloxy]phenyl}-1,3-dipropylxanthine is a potent adenosine antagonist that reverses the reduction in urine flow, sodium excretion and heart rate produced by the adenosine agonist, N6-cyclohexyladenosine. New derivatives of XAC in which the primary amino group has been condensed to the γ-carboxyl group of glutamic acid have been synthesized as prodrugs. These amino acid-XAC conjugates, which are considerably less potent than XAC in competitive binding assays at A1-adenosine receptors, are designed for selective enzymatic activation in the kidneys. The γ-glutamyl xanthine derivatives are substrates for γ-glutamyl transferase (EC 2.3.2.2) to generate an amine-functionalized xanthine. N-acetyl-γ-L-glutamyl-XAC is not active in vivo, consistent with inability of renal acylase (EC 3.5.1.14) to hydrolyze the acetyl group, a prerequisite step for the production of XAC from this molecule. The xanthine derivatives, γ-L-glutamyl-XAC and γ-L-glutamyl-γ-L-glutamyl-XAC are metabolized to XAC and produce a diuresis in vivo.


Exogenous adenosine and A1-selective adenosine receptor agonists have been shown to decrease glomerular filtration rate, urine flow, sodium excretion and to inhibit renin secretion (Osswald, 1983; Churchill and Bidani, 1988). All of these renal effects can be inhibited by theophylline, caffeine and other adenosine antagonists. Theophylline, in the absence of exogenous adenosine increases glomerular filtration, induces diuresis and natriuresis and stimulates renin secretion. These effects are the opposite of those produced by exogenous adenosine which suggests that the renal effects of theophylline and other alkylxanthines are produced by antagonism of the renal effects of endogenously released adenosine (Spielman and Thompson, 1982; Osswald, 1983; Fredholm, 1984; Collis et al., 1986). This observation suggests that endogenously released adenosine plays a role in the regulation of renal hemodynamics, renin secretion and renal excretory function (Spielman and Thompson, 1982). Variations in the concentration of endogenously released adenosine have been postulated to play important roles in the renal physiological and pathophysiological phenomena of autoregulation, tubuloglomerular feedback, macula densa control of renin secretion, acute renal failure (Bidani and Churchill, 1983; Bowmer et al., 1986) and hypertension (Churchill and Bidani, 1988). The synthesis of kidney-selective adenosine receptor antagonists could have significant research and clinical applications.

We have synthesized a class of amine-functionalized xanthine derivatives (Jacobson et al., 1985a). One of these compounds, XAC, is a potent adenosine antagonist that has been found to have effects in the kidney in addition to other tissues and organs. XAC inhibits the effects of CHA in the perfused rat kidney (Rossi et al., 1987).

A strategy for desiging kidney-selective prodrugs has been described by Orlowski and co-workers for derivatives of L-dopa (Wilk et al., 1978) as a dopamine precursor and for a sulfathiazole antibiotic (Orlowski et al., 1980). By this strategy an amine-bearing drug is converted to a less active Nα-acyl-L-γ-glutamyl derivative prodrug. In the kidney, successive enzymatic cleavage by RA and by γ-GT, both present in relatively high concentration, regenerates the active parent amine-containing drug. We have applied this strategy to the amine-functionalized xanthine derivatives, including XAC, in which the amino group was deemed an important determinant of high potency for adenosine antagonists.

In the present study we have demonstrated that the γ-glutamyl xanthine derivatives are substrates for the enzyme γ-GT and do produce a diuresis in vivo.

Materials and Methods

Synthesis

The compound numbers listed in the following section refer to the structures listed in figure 1 and Schemes 1 and 2.

Fig. 1.

Fig. 1

Structure of potential xanthine prodrugs, designed to be cleaved by γ-GT at the bond indicated by an arrow.

Scheme 1a.

Scheme 1a

Boc-γ-L-glutamyl-XAC α-benzyl ester, compound 6

A xanthine amine congener, XAC (compound 1, 0.80 g, Research Biochemicals Inc., Natick, MA), t-butyloxycarbonyl-L-glutamic acid α-benzyl ester (Sigma Chemical Co., St. Louis, MO, 1.5 g) and 1-hydroxybenzotriazole (0.20 g) were added to dimethylformamide (30 ml) The suspension was treated with EDAC (1.0 g) and stirred for 2 hr. Water (60 ml) was added and the precipitate was collected. Yield was 1.25 g (89%). The white solid was recrystallized from ethyl acetate-hexanes, mp 200–205°C. The 300 MHz NMR spectrum was consistent with the assigned structure.

γ-L-Glutamyl-XAC α-benzyl ester trifluoroacetate, compound 8

Compound 3 (1.25 g) was treated with trifluoroacetic acid for 10 min to remove the urethane protecting group. After evaporation, the residue was treated with dry ether, decanted and dried in vacuo at 50°C to give the product as a pure solid (1.01 g), in 79% yield, mp 128–132°C.

N-Acetyl-γ-L-glutamyl-XAC α-benzyl ester

Compound 8 (0.77 g) was suspended in DMF (40 ml) and treated with acetic anhydride (1 ml) and pyridine (2 ml with agitation. Upon addition of water the product precipitated as a chromatographically pure solid and was collected by filtration and dried in vacuo. The yield was 0.40 g (57%), mp 238–243°C. Analysis (C35H43N7O8): calculated, 60.95%; C, 6.28%; H, 14.21% N; found, 60.73%; C, 6.32%; H, 14.12% N.

N-Acetyl-γ-L-glutamyl-XAC, compound 2

N-Acetyl-γ-L-glutamyl-XAC α-benzyl ester (0.208 g) was suspended in 15 ml of DMF (warm). Palladium on charcoal (10%, 0.05 g) was added and the mixture was hydrogenated for 4 hr, with 20 psi of hydrogen gas. The reaction was judged to be complete by thin-layer chromatography. The catalyst was separated by centrifugation. The supernatant was treated with dry ether and the product precipitated slowly (seeding helpful). The yield was 0.16 g (88.5%), mp 222–225°C. Analysis (C28H37N7O8·H2O): calculated, 54.45%; C, 6.36%; H, 15.87% N; found, 54.34%; C, 6.30%; H, 15.86% N.

L-(S-Methyl)cysteinyl-XAC trifluoroacetate, compound 12

S-Methylcysteine (Sigma Chemical Co.) was protected as the N-t butyloxycarbonyl derivative using di-t-butyloxycarbonyl dicarbonate. The protected amino acid was coupled to XAC using EDAC/1-hydroxybenzotriazole in DMF. Precipitation upon addition of pH 7 sodium phosphate (0.1 M) and recrystallization from DMF-ether afforded t-butyloxycarbonyl-L-(S-methyl)cysteinyl-XAC (mp 223–225°C) in 90% yield. The t-butyloxycarbonyl group was removed from 0.35 g of t-butyloxycarbonyl-L-(S-methyl)cysteinyl-XAC in neat trifluoroacetic acid to give L-(S-methylcysteinyl)-XAC trifluoroacetate, 12, (mp 193–196°C) in 89% yield (0.32 g).

γ-L-Glutamyl-γ-L-glutamyl-XAC, compound 4

t-Butyloxycarbonyl-γ-L-glutamic acid α-benzyl ester (0.17 g, 0.5 mmol, Sigma Chemical Co.) was dissolved in DMF and treated with EDAC (0.1 g) and l hydroxybenzotriazole (0.1 g). The mixture was stirred for 10 min and 0.225 mmol of the trifluoroacetate salt of γ-L-glutamyl-XAC benzyl ester (hygroscopic oil, see above) was added. After an additional 10 min, one equivalent of diisopropylethylamine was added. After 1 hr, two volumes of water were added and the mixture was extracted with ethyl acetate (2 times). The organic phase was dried (MgSO4) and reduced in volume. Addition of petroleum ether caused precipitation of t-butyloxycarbonyl-γ-L-glutamyl-γ-L-glutamyl-XAC dibenzyl ester, 9 [0.20 g, 88% yield (recrystallized), mp 165–169°C]. This product was dissolved in DMF (6 ml) containing 10% acetic acid and reduced under hydrogen gas using 0.05 g of 5% palladium on carbon (Engelhard, Edison, NJ) as a catalyst under vigorous agitation for one day. The mixture was clarified by centrifugation and 30 ml of saturated sodium chloride was added. The product, t-butyloxycarbonyl-γ-L-glutamyl-γ-L-glutamyl-XAC, 10, was collected (89 mg, 55% yield, decomposes at 195°C). A sample of t-butyloxycarbonyl-γ-L-glutamyl-γ-L-glutamyl-XAC (74 mg) was treated with neat trifluoroacetic acid at room temperature for 10 min. After evaporation the residue was treated with dry ether and the product, γ-L-glutamyl-γ-L-glutamyl-XAC, 4, was obtained as a solid (50 mg, 67% yield, decomposes at 180°C).

γ-L-Glutamyl-L-(S-methyl)cysteinyl-XAC, compound 5

t-Butyloxycarbonyl-γ-L-glutamic acid α-benzyl ester (0.3 g, 0.9 mmol, Sigma Chemical Co.) was dissolved in DMF (15 ml) and treated with EDAC (0.2 g, 1.0 mmol) and l-hydroxy-benzotriazole (0.1 g). The mixture was stirred for 10 min and 0.265 g (0.40 mmol) of trifluoroacetate salt of L-(S-methyl)cysteinyl-XAC, 12, was added. After an additional 10 min, one equivalent of diisopropylethylamine was added. After 1 hr, two volumes of water were added, causing precipitation of t-butyloxycarbonyl-γ-L-glutamyl-L-(S-methyl)cysteinyl-XAC benzyl ester, 13, which was recrystallized from DMF-ethyl acetate-petroleum ether to give 0.264 g of a solid, 76% yield, decomposes at 180°C). A sample of this product (95 mg) was dissolved in DMF (4 ml) containing 10% acetic acid and reduced under hydrogen gas using 0.05 g of 5% palladium on carbon as a catalyst under vigorous agitation for one day. The mixture was clarified by centrifugation and 15 ml of saturated sodium chloride was added. The product, t-butyloxycarbonyl-γ-L-glutamyl-L-(S-methyl)cysteinyl-XAC, 14, was collected (66 mg, 78% yield, decomposes at 200°C). A sample of t-butyloxycarbonyl-γ-L-glutamyl-L-(S-methyl)cysteinyl-XAC (52 mg, 67 μmol) was treated with neat trifluoroacetic acid at room temperature for 10 min. After evaporation the residue was treated with dry ether and the product, γ-L-glutamyl-L-(S-methyl)cysteinyl-XAC, 4, was obtained as a solid (49 mg, 93% yield, mp 184°C with decomposition).

Biochemical Assays

A1 receptor binding

Stock solutions of xanthines were prepared in dimethylsulfoxide and stored frozen. Solutions were warmed to 50°C before dilution. Inhibition of binding of 1 nM [3H]PIA to A1 adenosine receptors in rat cerebral cortex membranes was measured as described previously (Schwabe and Trost, 1980). Inhibition of binding by a range of concentrations of xanthines was assessed in triplicate in at least three separate experiments. IC50 values were converted to Ki values using a KD value for [3H]PIA of 1.0 nM and the Cheng-Prusoff equation (1973).

Enzyme kinetic studies

The methodology used in the determination of the rate of cleavage of the N-acetyl and γ-glutamyl group from the compounds was a modification of Orlowski et al. (1980). The experiments were carried out using purified porcine kidney acylase I and/or bovine kidney γ-GT (Sigma). Samples (0.3 ml total volume) containing the substrate, Tris buffer (0.1 M, pH 8.2), glycylglycine (7.5 μmol) and enzyme were incubated at 37°C. The optimal incubation time was determined for each compound (1–5 min). Kinetic data were derived from concentration curves that were performed for each compound. The protein content of the enzymes were determined using the Miller (1959) modification of the Lowry (1981) assay.

The products of enzyme cleavage were quantitated using HPLC. The instrumentation consisted of two Gilson pumps, a Waters WISP automatic injector and a Waters model 450 variable wavelength UV detector (300 nm). Separations were attained using a Beckman reverse-phase Ultrasphere column (5 μ, 4.6 mm × 25 cm) with an isocratic solvent system of 68% methanol and 32% 0.05 M phosphate buffer, pH 6.0, at a flow rate of 1.0 ml/min. Approximate retention times from the solvent front are: compound 1, 7 min; compound 2, 3.4 min; compound 3, 5 min; compound 4, 2.4 min; compound 5, 8 min; and compound 12, 12 min. Kinetic parameters (Km and Vmax) were calculated using ENZFITTER, a nonlinear regression data analysis program.

In vivo studies

Adult male Sprague-Dawley rats were housed and cared for in accordance with the NIH Guide for Care and Use of Laboratory Animals. The rats had free access to tap water and Purina Rodent Chow before the experiments. Rats were anesthetized with sodium pentobarbital (45 mg/kg b.wt.) through a tail vein. Supplements were given as required to maintain a surgical plane of anesthesia during the experiments. Cannulas made of polyethylene tubing were placed in both ureters and a femoral vein as described previously (Churchill et al., 1978). The arterial cannula was connected to a pressure transducer and the mean arterial pressure was monitored on a polygraph, except when arterial blood was sampled. All of the rats received a priming dose of inulin (2 ml/kg b.wt. of 10 g/ml of inulin in 150 mM NaCl), followed by a continuous i.v. infusion of inulin (0.055 ml/min of 3.8 g/ mol of inulin in 150 mM NaCl). After a 1-hr equilibration period the infusate was modified as described by the five groups listed below. Fifteen minutes after infusate modification a 30-min clearance period was begun. Urine was collected in preweighed micro test tubes and arterial blood (0.3 ml) was sampled at the clearance midpoint. The blood was centrifuged at 8000 × g for 5 min at 4°C. Urine and plasma were frozen until analysis.

Five groups of rats were studied: Group I (controls: no modification of the perfusate; n = 20); Group II (CHA at 2 nmol/min/kg b.wt; n = 10); Group III (CHA at 10 nmol/min/kg b.wt; n = 13); Group IV (n = 12) and Group V (n = 10) were identical to Groups II and III, respectively, except that a priming injection of XAC, compound 1 was given (0.2 μmol/kg b.wt.) and XAC was added to the infusate (2 nmol/ min/kg b.wt.).

The possibility that Ac-γ-Glu-XAC, compound 2, would antagonize CHA-induced effects was examined in a sixth group of rats (n = 9). CHA was infused at the lower rate (2 nmol/min/kg b.wt.) and Ac-γ-Glu-XAC was given both in priming dose (2 μmol/kg b.wt.) and by infusion (20 nmol/min/kg b.wt.). The priming and the infusion doses of Ac-γ-Glu-XAC were 10 times those of the parent XAC. Urine and plasma sodium and potassium were measured by flame photometry, using an internal lithium standardization. Inulin concentrations in the urine and plasma filtrates were determined by a spectrophotometric method that has been described previously (Churchill et al., 1978). Standard formulas were used to calculate the clearance of inulin and the sodium and potassium excretory rates. Although urine was collected directly from each ureter in order to minimize errors attributable to bladder dead space, the functions of the two kidneys were summed then normalized per kilogram body weight. All results are expressed as means ± S.K.M.s. Analysis of variance and Scheffe contrasts (Wallen-stein et al., 1980) were used to assess the statistical significance of the differences between the five groups.

Diuresis studies

Male Sprague-Dawley rats (130–170 g) were injected i.v. through a caudal vein with a xanthine derivative (≈5 μmol/ kg b.wt.) or vehicle. The compounds were dissolved in 0.1 N NaOH and the resulting solution was diluted with physiological saline to a final dilution of 1 to 20. The compound or vehicle was injected in a volume of 200 μl. After the injection, animals were placed in polycarbonate metabolic cages with water and no food. There were no differences observed between the treated and control animals in water consumption over the 3-hr period. The urine output and the amount of water consumed was measured after 3 hr. To control for the large between-day variation in the urine output, the control and treated animals injected on the same day were compared using an unpaired one-way t test (Statview statistics program). The urine samples were frozen until analysis. The urine was assayed for the presence of XAC and other xanthines. A measured aliquot of urine was acidified with trichloroacetic acid (7% final concentration) and centrifuged for 5 min on a Fisher microcentrifuge (model 235B). The concentration of xanthine derivatives in the urine was determined by the HPLC assay procedure outlined above.

Results

Synthesis

We have synthesized N-acetyl-γ-L-glutamyl-XAC (compound 2) and several other related prodrug candidates (fig. 1). The synthetic routes, based on standard peptide chemistry, are shown in Schemes 1. Because the xanthine moiety is stable to pH extremes, one may use a variety of urethane protecting groups (such as Boc, N-t-butyloxycarbonyl) for the transient protection at the glutamyl alpha amino position. Coupling of N-t-Boc-L-glutamic acid α-benzyl ester to XAC, compound 1, using a water-soluble carbodiimide, gave the fully protected, organic-soluble intermediate, compound 6 (Scheme 1a). Replacement of the urethane group with N-acetyl followed by hydrogenolysis of the benzyl ester gave compound 2, in high yield. Compound 3, γ-L-glutamyl-XAC, was synthesized from the protected intermediate in two steps. Two other related series of xanthine γ-L-glutamyl amides were prepared as potential prodrugs. The γ-carboxylate group was bound alternately to γ-L-glutamyl-XAC (compounds 4 and 9–11) or to L-(S-methyl)cysteinyl-XAC (compounds 5, 13–15, Scheme 1b).

Scheme 1b.

Scheme 1b

i, carbodiimide condensation with Boc-α-benzyl glutamic acid; ii, H2/Pd; iii, trifluoroacetic acid; iv, acetic anhydride/base.

In Vitro Experiments

A1 receptor binding

Affinity of binding to central A1-adenosine receptors was determined by inhibition of binding of [3H]PIA to rat brain membranes in the presence of adenosine deaminase. Representative binding isotherms are shown in figure 2. Hill coefficients for binding were generally in the range of 0.9 to 1.0. Table 1 shows the affinity of XAC and its γ-glutamyl derivatives at central A1-adenosine receptors. XAC, an amine derivative, is the most potent. The anionic prodrug derivatives, compounds 2 and 15, are considerably less potent in binding. L-(S-methyl)cysteinyl-XAC, compound 12, is close in receptor affinity to XAC, thus, the cys-amide bond perhaps need not be cleaved in order to obtain high potency.

Fig. 2.

Fig. 2

Inhibition of specific binding of [3H]PIA to A1 adenosine receptors in rat brain membranes by xanthine derivatives. Representative curves for XAC, 1, X; Ac-γ-Glu-XAC, 2, ▲; γ-Glu-XAC, △, A; S-methyl-Cys-XAC, 12, ○; Ac-γ-Glu-S-methyl-Cys-XAC, 15, ▲.

TABLE 1. Affinities of xanthine derivatives of A1 adenosine receptors from rat brain, measured through inhibition of binding of [3H]PIA.

Each value (X ± S.E.) represents the average of at least three determinations run in triplicate.

Compound Ki Value
nM
1 XAC 1.1 ± 0.05
2 Ac-γ-Glu-XAC 297.8 ± 12.1
3 γ-Glu-XAC 21.0 ± 2.5
4 γ-Glu-γ-Glu-XAC 174.6 ± 16.7
5 γ-Glu-(methyl)Cys-XAC 32.3 ± 2.5
12 (Methyl)Cys-XAC 4.6 ± 0.9
15 Ac-γ-Glu-(methyl)Cys-XAC 140.8 ± 30.4

Enzyme kinetics

γ-Glutamyl amide bonds are not universally cleaved by γ-GT (Magnan et al., 1982). To determine the feasibility of this prodrug approach, it was first necessary to evaluate compounds 3 to 5 as substrates for γ-GT using purified enzyme from rat kidneys. All three compounds were found to act as substrates for γ-GT. Table 2 lists the kinetic constants measured for these potential prodrug derivatives. Compound 4, the γ-glutamyl-γ-glutamyl derivative, had a low affinity for the enzyme and the cleavage of the first γ-glutamyl group occurred rapidly with all of the initial compound being metabolized within the first 5 min of incubation. The one γ-glutamyl group on compound 3 was removed at a slower rate. This compound had a much lower Km for the enzyme than compound 4. Compound 5 was found to be metabolized to compound 12 and then metabolized further to yield compound 1. The kinetics of these two reactions were not studied separately due to the difficulty of separating compounds 1 and 12 by HPLC.

TABLE 2.

Kinetic constants for xanthine derivatives as substrates of γ-GT

Compound Km Vmax
μM nmol mg−1 min−1
3 176.9 ± 32.9 21.2 ±2.5
4 535.0 ± 156.0 3790.0 ± 906.0

RA cleaves amide bonds at the Nα-position of numerous L-amino acids. Nevertheless, the N-acetylated amino acid-xanthine conjugates, compound 2, 11 and 15, were not substrates for RA. There was no significant product formation after 4 hr of incubation. The purchased RA enzyme was tested for activity using the spectrometric method of Mitz and Schlueter (1958). The commercial RA enzyme preparation was found to actively hydrolyze acetyl-L-methionine, which is known to be a good substrate for RA.

In vivo experiments

The results in figures 3 and 4 demonstrate that CHA produces concentration-dependent reductions in glomerular filtration rate, urine flow and sodium and potassium excretion. Although XAC did not antagonize CHA-induced changes in glomerular filtration rate or potassium excretion, it blocked completely the antidiuretic and antinatriuretic effects of a low dose of CHA. Urine flow and sodium excretion in the presence of XAC and CHA were higher, but not significantly, than the respective control values. The effects of XAC were not organ-selective as illustrated in figure 5. CHA depressed heart rate and blood pressure and the cardiovascular effects were antagonized by XAC. XAC reversed the chronotropic effects of CHA more effectively than the hypotensive, as expected for an A1-selective adenosine antagonist (Fredholm et al., 1987).

Fig. 3.

Fig. 3

Glomerular filtration rate (GFR) and urine flow (V) in five groups of rats: Group I (controls; n = 20); Group II (CHA at 2 nmol/min/kg b.wt.; n = 10); Group III (CHA at 10 nmol/min/kg b.wt.; n = 13); Group IV (n = 12) and Group V (n = 10) were identical to Groups II and III, respectively, except that a priming dose of Compound 1 (XAC) was given (0.2 μmol/ kg. b.wt.) and Compound 1 was added to the i.v. infusate (2 nmol/min/ kg b.wt.). Means ± S.E.M.s. P values below the curves are comparisons with control; P values above are comparisons with the corresponding group not given XAC. NS, P > .05.

Fig. 4.

Fig. 4

Sodium and potassium excretory rates (UNaV and UKV) in five groups of rats: Group I (controls; n = 20); Group II (CHA at 2 nmol/min/ kg b.wt.; n = 10); Group III (CHA at 10 nmol/min/kg b.wt.; n = 13); Group IV (n = 12) and Group V (n = 10) were identical to Groups II and III, respectively, except that a priming injection of XAC was given (0.2 μmol/kg. b.wt.) and XAC was added to the i.v. infusate (2 nmol/min/kg b.wt.). Means ± S.E.M.s. P values below the curves are comparisons with control; P values above are comparisons with the corresponding group not given XAC. NS, P > .05.

Fig. 5.

Fig. 5

Blood pressure (BP) and heart rate (HR) in five groups of rats: Group I (controls; n = 20); Group II (CHA at 2 nmol/min/kg b.wt; n = 10); Group III (CHA at 10 nmol/min/kg b.wt.; n = 13); Group IV (n = 12) and Group V (n = 10) were identical to Groups II and III, respectively, except that a priming dose of XAC was given (0.2 μmol/kg. b.wt.) and XAC was added to the i.v. infusate (2 nmol/min/kg b.wt.). Means ± S.E.M.s. P values below the curves are comparisons with control; P values above are comparisons with the corresponding group not given XAC. NS, P > .05.

In contrast, Ac-γ-Glu-XAC, given at doses 10 times higher than XAC, failed to antagonize the effects of even the lower concentration of CHA. Rats given CHA (2 nmol/min/kg b.wt.), in the absence vs. the presence of compound 2, respectively, had glomerular filtration rates of 6.9 ± 0.3 vs. 5.8 ± 0.4 ml/ min/kg, urine flows of 27 ± 2 vs. 22 ± 2 μl/min/kg, sodium excretions of 4.1 ± 0.8 vs. 4.6 ± 0.7 μmol/min/kg and potassium excretions of 5.2 ± 0.2 vs. 4.6 ± 0.4 μmol/min/kg. None of the differences between these two groups were statistically significant, but all of these mean values were significantly lower than values in the control rats.

Diuretic experiments

Compounds 1, 2, 3, 4 and 5 were tested for diuretic activity in conscious rats. The volumes of urine produced 3 hr after i.v. injection are listed in table 3. Treated and control animals injected on the same day were compared. There were no differences in water consumption between treated and control animals.

TABLE 3. Urine output 3 hr after i.v. injection of compound or vehicle.

Male Sprague-Dawley rats (130–170 g) were injected i.v. through a caudal vein with a xanthine derivative (~5 μmol/kg b. wt.) or vehicle. The between-day variation in urine output (Uv) were controlled by comparing treated and control animals injected on the same day. Uv, mean ± S.E.

Compound Dose Uv n

μmol/kg ml/kg
1 5 9.54 ± 1.29* 8
Control 5.06 ± 0.94 5
2 5 9.15 ± 1.21 10
Control 12.14 ± 1.59 6
3 4.5 10.04 ± 1.11* 10
Control 5.58 ± 1.61 6
4 5.8 18.94 ± 1.57* 10
Control 10.74 ± 2.46 6
5 5.2 10.48 ± 1.06 10
Control 9.12 ± 0.99 6
*

P < .05 unpaired t test.

Animals injected with compounds 1, 3 and 4 had significantly more urine output than the control animals. Compounds 3 and 4 were cleaved in vivo to give XAC which was measured in the urine (0.30 ± 0.05 /μmol/kg and 0.48 ± 0.09 μmol/kg b.wt., respectively). Due to daily variations in urine output it is difficult to speculate on the potency of compounds 3 and 4 as compared to the parent compound 1. Consistent with the results of the in vitro and in vivo experiments described above, compound 2, the N-acetyl derivative failed to produce an increase in urine output. Analysis of the urine showed the presence of the intact compound (0.12 ± 0.014 μmol/kg b.wt.) with no XAC as the active metabolite. Compound 5, the γ-glutamyl-(S-methyl)-cysteinyl derivative did not produce a significant diuresis.

Discussion

Caffeine and theophylline are classically used methylxanthines which act as weak nonselective adenosine receptor antagonists (Fredholm, 1980). There are numerous sites of action of adenosine throughout the body and therefore the use of theophylline and caffeine causes a variety of effects including, central, respiratory and cardiac stimulation in addition to diuresis.

There is a continuing search for more selective adenosine receptor antagonists. This has resulted in more potent xanthine analogs and antagonists which are selective for A1 (Bruns et al., 1986; Fredholm et al., 1987)- or A2 (Ukena et al., 1986)-adenosine receptor subtypes. The development of new adenosine receptor agonists and antagonists for clinical use has been impeded by the multiplicity of adenosine effects. Clearly, tissue-specific drug delivery systems for adenosine agonists or antagonists are required to achieve genuine in vivo selectivity.

Prodrugs are drug derivatives in which a key functional group has been blocked chemically to make the drug less potent or inactive (Bundgaard, 1985). The drug is activated by removal of the blocking group at a physiological site of action. Prodrugs are more organ selective and are associated with fewer side effects than the parent drug due to diminished general systemic drug activity.

Few prodrug schemes (Bodor et al., 1978) have been suggested for theophylline, because of the lack of functional groups available for derivatization. We have developed a functionalized congener approach to adenosine receptor agonists (Jacobson et al., 1985b) and antagonists (Jacobson et al., 1985a), by which a new functionality is introduced at a distal site on the drug molecule without diminishing receptor binding. XAC, 1, and a series of related congeners were found to be more potent than the parent alkylxanthine derivatives. This enhancement of receptor binding affinity was dependent on the presence of a free amino group (Jacobson et al., 1986). This amino group of XAC may be blocked temporarily, in a way which diminishes activity and which is reversible at a site of action. The structural flexibility offered by the functionalized congener approach permits the application of prodrug delivery schemes, otherwise inaccessible for a particular class of drugs.

We have synthesized a family of prodrugs based on the approach of Orlowski et al. (1978). Orlowski and others (Wilk et al., 1978; Orlowski et al., 1980; Magnan et al., 1982) have utilized the enzyme γ-GT, which is present in high concentration in the kidneys on the brush border (luminal side of the membranes), to activate certain amine-bearing γ-glutamyl derivative prodrugs in that organ. The placement of an N-acetyl, chloroacetyl, or butyryl group on the γ-glutamyl prodrug was found to enhance selectivity for the kidneys. This is due to the high concentration of RA in the kidneys, which hydrolyzes the acyl group as a first step in drug activation.

XAC was the ideal compound for application of this prodrug synthesis technique. In addition to having a free amino group, XAC was found to antagonize renal responses known to be mediated by A1 adenosine receptors. Adenosine-induced renal vasoconstriction and adenosine-induced inhibition of renin secretion are both mediated by the A1 subclass of adenosine receptors (Murray and Churchill, 1984, 1985; Churchill and Churchill, 1985). In previous studies, XAC antagonized both CHA-induced renal vasoconstriction (Rossi et al., 1987) and CHA-induced inhibition of renin secretion (Churchill et al., 1987) with a potency approximately three orders of magnitude greater than that of theophylline. XAC exhibited a 20-fold selectivity for A1 vs. the A2 receptor with respect to the renin secretory effects of CHA (Churchill et al., 1987). In the present studies, CHA decreased glomerular filtration rate, urine flow and sodium and potassium excretion, as has been reported previously (Churchill and Bidani, 1987). Many of these effects were antagonized by XAC in an apparently competitive manner. In addition to the kidney effects, XAC does not readily cross the blood brain barrier (Seale et al., 1988) which makes this compound more suitable for systemic targeting without the complication of central nervous system effects.

We have synthesized N-acetyl-γ-glutamyl-XAC, compound 2, and several other related prodrug candidates (fig. 1). γ-Glu-(S-methyl)-L-cys-XAC, compound 5, was synthesized because of its similarity in structure to glutathione (γ-Glu-Cys-Gly), the natural substrate of γ-GT.

The adenosine receptor binding affinities of these N-acetyl and γ-glutamyl prodrugs were measured to be considerately lower than the parent compound, XAC. This observation is consistent with previous findings with amino acid and other conjugates of xanthines (Jacobson et al., 1986), in which the presence of a carboxylate group on the chain tended to diminish activity. This suggested that a differential in vivo response was possible, assuming that XAC or Cys(Me)-XAC would be generated enzymatically in the kidney regions high in RA and γ-GT activity.

Using purified enzyme, we observed efficient cleavage of the γ-glutamyl amide bond resulting in the production of XAC in vitro. Unlike the N-acetyl-γ-glutamyl derivatives synthesized by Orlowski and by Hofbauer et al. (1980), the γ-glutamyl xanthine derivatives did not serve as substrates for RA. Thus, the two step prodrug cleavage scheme dependent on both RA and γ-GT could not be utilized. This lack of activation was also seen in the in vivo experiments in which N-acetyl-γ-glutamyl XAC, compound 2, failed to increase urine output or reverse the kidney effects of CHA. Lack of cleavage was also seen in vivo where the intact compound was found in the urine of rats. The reason for the lack of cleavage by RA is unclear. Steric hindrance at the active site on the enzyme is the most reasonable explanation in spite of structural similarities that exist between N-acetyl-γ-L-glutamyl XAC and the cleavable prodrug synthesized by Orlowski et al. (1980), N-acetyl-γ-L-glutamyl sulfamethoxazole.

As an alternative to the two step cleavage by different enzymes, a γ-glu-γ-glu conjugate of XAC, compound 4, was designed and synthesized for two subsequent cleavage steps by γ-GT. Cleavage of both γ-glutamyl groups was achieved in vitro with purified enzyme. The cleavage of the first γ-glutamyl group occurred so rapidly that it was doubtful that this compound would reach the kidney intact. Cleavage of the first γ-glutamyl group results in the formation of compound 3, the single γ-glutamyl derivative. Both of these compounds 3 and 4 produced a significant diuresis in vivo.

In addition to its action as a diuretic, a xanthine prodrug may be useful in the study of renal physiology and the treatment of renal pathophysiologies that are mediated by endogenous adenosine.

ABBREVIATIONS

XAC

xanthine amine congener

CHA

N6-cyclohexyladenosine

RA

renal acylase

γ-GT

γ-glutamyl transferase

EDAC

ethyldimethylaminopropyl carbodiimide hydrochloride

DMF

dimethylformamide

PIA

N6-phenylisopropyladenosine

HPLC

high-performance liquid chromatography

References

  1. Bidani AK, Churchill PC. Aminophylline ameliorates glycerol-induced acute renal failure in rats. Can J Physiol Pharmacol. 1983;61:567–571. doi: 10.1139/y83-087. [DOI] [PubMed] [Google Scholar]
  2. Bodor N, Sloan KB, Kuo YN, Higuchi T. Controlled delivery of Theophylline: Chemistry of 7-acyl and 7,7′-acylditheophylline derivatives. J Pharm Sci. 1978;67:1045–1050. doi: 10.1002/jps.2600670806. [DOI] [PubMed] [Google Scholar]
  3. Bowmer CJ, Collis MG, Yates MS. Effects of adenosine antagonist 8-phenyltheophylline on glycerol-induced acute renal failure in the rat. Br J Pharmacol. 1986;88:205–212. doi: 10.1111/j.1476-5381.1986.tb09488.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bruns RF, Lu GH, Pugsley TA. Characterization of the A2 adenosine receptor labeled by (3H)NECA in rat striatal membranes. Mol Pharmacol. 1986;29:331–346. [PubMed] [Google Scholar]
  5. Bundgaard H. Design of Prodrugs: Bioreversible derivatives for various functional groups and chemical entities. In: Bundgaard H, editor. Design of prodrugs. Elsevier Science Publishers; Amsterdam: 1985. pp. 1–91. [Google Scholar]
  6. Cheng YC, Prusoff WH. Relationship between the inhibition constant (Ki) and the concentration of inhibitor which causes 50 percent inhibition (IC50) of an enzymatic reaction. Biochem Pharmacol. 1973;22:3099–3108. doi: 10.1016/0006-2952(73)90196-2. [DOI] [PubMed] [Google Scholar]
  7. Churchill PC, Bidani AK. Adenosine and renal function. In: Williams M, editor. The Adenosine Receptors. Humana Press; Clifton, NJ: 1988. in press. [Google Scholar]
  8. Churchill PC, Bidani A. Renal effects of selective adenosine agonists in anesthetized rats. Am J Physiol. 1987;252:F299–F303. doi: 10.1152/ajprenal.1987.252.2.F299. [DOI] [PubMed] [Google Scholar]
  9. Churchill PC, Churchill MC. A1 and A2 adenosine receptor activation inhibits and stimulates renin secretion of rat renal cortical slices. J Pharmacol Exp Ther. 1985;232:589–594. [PubMed] [Google Scholar]
  10. Churchill PC, Churchill MC, McDonald FD. Renin secretion and distal tubule Na+ in rats. Am J Physiol. 1978;235:F611–616. doi: 10.1152/ajprenal.1978.235.6.F611. [DOI] [PubMed] [Google Scholar]
  11. Churchill PC, Jacobson KA, Churchill MC. XAC, a functionalized congener of 1,3 dialkylxanthine, antagonizes A1 adenosine receptor-mediated inhibition of renin secretion in vitro. Arch Int Pharmacodyn Ther. 1987;290:293–301. [PMC free article] [PubMed] [Google Scholar]
  12. Collis MG, Baxter GS, Keddie JR. The adenosine receptor antagonist, 8-phenyltheophylline, causes diuresis and saliuresis in the rat. J Pharm Pharmacol. 1986;38:850–852. doi: 10.1111/j.2042-7158.1986.tb04510.x. [DOI] [PubMed] [Google Scholar]
  13. Fredholm BB. Are methylxanthine effects due to antagonism of endogenous adenosine? Trends Pharmacol Sci. 1980;1:120–131. [Google Scholar]
  14. Fredholm BB. Cardiovascular and renal actions of methylxanthines. Prog Clin Biol Res. 1984;158:303–330. [PubMed] [Google Scholar]
  15. Fredholm BB, Jacobson K, Jonzon B, Kirk K, Li Y, Daly J. Evidence that a novel 8-phenyl-substituted xanthine derivative is a cardioselective adenosine receptor antagonist in vivo. J Cardiovasc Pharmacol. 1987;9:396–400. doi: 10.1097/00005344-198704000-00003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hofbauer KG, Sonnenburg C, Stalder R, Criscione L, Kraetz J, Fuhrer W, Habicht E. CGP 22979A, a renal vasodilator with natriuretic properties. J Pharmacol Exp Ther. 1985;232:838–844. [PubMed] [Google Scholar]
  17. Jacobson KA, Kirk KL, Padgett WL, Daly JW. Functionalized congeners of 1,3-dialkylxanthines: Preparation of analogues with high affinity for adenosine receptors. J Med Chem. 1985a;28:1334–1340. doi: 10.1021/jm00147a038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Jacobson KA, Kirk KL, Padgett WL, Daly JW. Functionalized congeners of adenosine: Preparation of analogues with high affinity for A1-adenosine receptors. J Med Chem. 1985b;28:1341–1346. doi: 10.1021/jm00147a039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Jacobson KA, Kirk K, Padgett W, Daly J. A functionalized congener approach to adenosine receptor antagonists: Amino acid conjugates of 1,3-dipropylxanthine. Mol Pharmacol. 1986;29:126–133. [PMC free article] [PubMed] [Google Scholar]
  20. Magnan SDJ, Shirota FN, Nagasawa HT. Drug latentiation by γ-glutamyl transpeptidase. J Med Chem. 1982;25:1018–1021. doi: 10.1021/jm00351a003. [DOI] [PubMed] [Google Scholar]
  21. Miller G. Protein determinations for large numbers of samples. Anal Chem. 1959;3:964. [Google Scholar]
  22. Mitz MA, Schlueter R. Direct spectrophotometric measurement of the peptide bond: Application to the determination of acylase I. Biochim Biophys Acta. 1958;27:168–172. doi: 10.1016/0006-3002(58)90305-6. [DOI] [PubMed] [Google Scholar]
  23. Murray RD, Churchill PC. The effects of adenosine receptor agonists in the isolated, perfused rat kidney. Am J Physiol. 1984;247:H343–348. doi: 10.1152/ajpheart.1984.247.3.H343. [DOI] [PubMed] [Google Scholar]
  24. Murray RD, Churchill PC. Concentration dependency of the renal vascular and renin secretory responses to adenosine receptor agonists. J Pharmacol Exp Ther. 1984;232:589–594. [PubMed] [Google Scholar]
  25. Orlowski M, Mizoguchi H, Wilk S. N-acyl-γ-glutamyl derivatives of sulfamethoxazole as models of kidney-selective prodrugs. J Pharmacol Exp Ther. 1980;212:167–172. [PubMed] [Google Scholar]
  26. Osswald H. Adenosine and renal function. In: Berne R, Rail T, Rubio R, editors. Regulatory Function of Adenosine. Martinus Nijhoff; Boston: 1983. pp. 399–415. [Google Scholar]
  27. Osswald H. The role of adenosine in the regulation of glomerular filtration rate and renin secretion. Trends Pharm Sci. 1984;5:94–97. [Google Scholar]
  28. Rossi NF, Churchill PC, Jacobson KA, Leahy AE. Further characterization of the renovascular effects of N6-cyclohexyladenosine in the isolated perfused rat kidney. J Pharmacol Exp Ther. 1987;240:911–915. [PMC free article] [PubMed] [Google Scholar]
  29. Schwabe U, Trost T. Characterization of adenosine receptors in rat brain by (—)[3H]N6-phenylisopropyladenosine. Naunyn-Schmiedeberg’s Arch Pharmacol. 1980;313:179–187. doi: 10.1007/BF00505731. [DOI] [PubMed] [Google Scholar]
  30. Seale TW, Abla KA, Jacobson KA, Carney JM. Xanthine amine congener of 1,3-dipropyl-8-phenylxanthine (XAC): A peripherally-selective adenosine receptor blocking agent. 1989. Accepted for publication. Pharmacol Biochem Behav [Google Scholar]
  31. Spielman WS, Thompson CI. A proposed role for adenosine in the regulation of renal hemodynamics and renin release. Am J Physiol. 1982;242:F423–F435. doi: 10.1152/ajprenal.1982.242.5.F423. [DOI] [PubMed] [Google Scholar]
  32. Ukena D, Shamim MT, Padgett W, Daly JW. Analogs of caffeine: Antagonists with selectivity for A2 adenosine receptors. Life Sci. 1986;39:743–750. doi: 10.1016/0024-3205(86)90023-8. [DOI] [PubMed] [Google Scholar]
  33. Wallenstein S, Zucker CL, Fleiss JL. Some statistical methods useful in circulation research. Circ Res. 1980;47:1–9. doi: 10.1161/01.res.47.1.1. [DOI] [PubMed] [Google Scholar]
  34. Wilk S, Mizoguchi H, Orlowski M. γ-Glutamyl-Dopa: A kidney specific dopamine precursor. J Pharmacol Exp Ther. 1978;206:227–232. [PubMed] [Google Scholar]

RESOURCES