Pharmacokinetic Study of Cyclic Glycolamide Ester Conjugates of Niflumic Acid

PHARMACOKINETIC STUDY OF CYCLIC GLYCOLAMIDE ESTER CONJUGATES OF NIFLUMIC ACID

Kokil Sachin Uttamrao.*1 and Gadad Anadannappa Karsiddappa.2

1Dept.of Pharmaceutical Chemistry Bharati Vidyapeeth College of Pharmacy, Kolhapur,

Maharashtra, India.,

2Pharmacy Programme, Faculty of Medical Sciences, The University of WestIndies, St.Augustine, Champs Fluers, MountHope, Trinidad, WestIndies.

*Author for correspondence:

Kokil Sachin Uttamrao

Bharati Vidyapeeth College of Pharmacy,

Near Chitranagri, Kolhapur.

Maharashtra, India.

Phone: +91(0)9422600264., 9423867464.

Email: sachinkokil@rediffmail

PHARMACOKINETIC STUDY OF CYCLIC GLYCOLAMIDE ESTER CONJUGATES OF NIFLUMIC ACID

ABSTRACT:

A simple synthetic route for the preparation of glycolamide ester conjugates of Niflumic acid is exploited and prepared 2-(???-trifluro-m toluidino) nicotinyl glycolamide esters in good yields and their structures were confirmed by various spectral data.Preliminary evaluation of physicochemical and in vitro reversion properties of various prodrugs is studied which includes solubility, lipophilicity, chemical hydrolysis, pH-hydrolysis, enzymatic hydrolysis, and stability study. Compounds 3, 10, 13, 14 showed solubility increased by 50 to 100 folds, compounds 4 to 7 showed solubility increased by 20 to 35 folds, compounds 8, 9, 11 and 12 showed least solubility than parent drugs indicating in general cyclic glycolamide esters are least soluble compounds in 50 mM phosphate buffer of pH 7.4 except 10 and 13. Similarly compound 3-7 and 14 showed lipophilicity more than 1 and cyclic glycolamide esters showed lesser lipophilicity except compound 10 and 11. These esters were reasonably stable in 50 mM Phosphate buffer, half lives being 0.29–3.23 h among all compounds 6 and 9 showed maximum stability, half lives being 3.23 and 2.33 h respectively. In general cyclic glycolamide esters having six membered ring found to be more stable. Compound 6 and 9 were subjected to hydrolytic studies over pH range 1.2 – 9.0 at 70?C and they showed maximum stability at pH 4.6 and pH 5.0 respectively. At pH 4.6 compound 6 and 9 showed shelf life of 1.37 h and 4.33 h at 20?C as determined from accelerated studies using Arrehenius equation. The glycolamide ester derivatives were quantitatively converted into Niflumic acid very rapidly in human plasma at 37°C and pH 7.4 except compound 8. Comparison with chemical stability data, the intrinsic reactivity of esters has no effect on their enzymatic reactivity. The results from these evaluations demonstrate that the derivatives examined have many of the ideal properties required for potential useful prodrug.

Key words: Niflumic acid, glycolamide ester, hydrolysis study, prodrug.

INTRODUCTION:

Non-steroidal anti-inflammatory drugs (NSAID’s) have differing modes of inhibitory activities, with the arachidonic acid cascades, at the level of cycloxygenase. Inhibition of this enzyme at the micromolar concentration by NSAID’s prevents the formation of thromboxanes, prostaglandins and prostacyclins and as a consequence potent anti-inflammatory activity occurs1. NSAID’s due to carboxylic acid function passes into cells of stomach mucosa. The intracellular pH of cells is more basic than that of stomach lumen and NSAID’s become ionized, this results in reverse flow of H+ ions into the cells causing cellular damage2. The direct contact mechanism appears to play a determinant role in the production of gastrointestinal lesions and it is probably a combination of local irritation produced by the free carboxylic acid group of NSAID’s and local inhibition of cytoprotective action of prostaglandines on gastric mucosa. The considerable gastrointestinal distress associated with chronic use of these compounds and their low half-life constitute the main disadvantages in clinical use of NSAID’s3.

Niflumic acid (1) is one of the newly developed non-steroidal anti-inflammatory agent which act as cycloxygenase I inhibitor, indicated in various inflammatory and rheumatic conditions4. In order to supress the gastrointestinal irritation produced by Niflumic acid which is believed to caused mainly by direct contact of acidic group with gastrointestinal mucosa, an alternative approach is to use the promising prodrug concept to mask the carboxylic acid function5.

An ideal prodrug retains the activity of the parent drug while unwanted side effects are eliminated or notably reduced6, to achieve such a pharmacological profile a prodrug should exhibit optimum physicochemical properties. The prodrug should show a good stability in aqueous solutions and in gastrointestinal fluid, it should have suitable water solubility and lipophilicity to ensure absorption by the oral route and it should be readily hydrolysed following gastric absorption to release the parent drug. Since simple alkyl and aryl ester prodrugs donot show the requirements mentioned above as they are not sufficiently labile in vivo to ensure a high rate of prodrug conversion3.

To attempt a new generally ester prodrug types possessing a high enzymatic rate of hydrolysis in plasma or in GIT wall due to esterase enzyme, we recognized the esters of certain 2-hydroxyacetamide (glycolamide) which are chemically highly stable7. The glycolamide ester structure combined with the presence of two substituents on the amide nitrogen atom, such as N,N disubstituted glycolamide esters are found be promising biolabile prodrug8.

The conversion of carboxyl group to glycolamide doesn’t negatively modify the anti-inflammatory activity of synthesized prodrugs of Niflumic acid in vivo, because, the cleavage of glycolamide linkages take place by amide hydrolysing enzymes present in various tissues as well as in plasma9. But it depends on release of Niflumic acid from prodrug.

In the novelty of present work is evaluation of a number of glycolamide esters of Niflumic acid, as potential prodrugs with respect to their solubility, lipophilicity, chemical hydrolysis, pH-hydrolysis, enzymatic hydrolysis and stability study.

MATERIALS AND METHODS:

Substituted 2-chloracetamides required for the synthesis of glycolamide esters of Niflumic acid were prepared according to reported methods10. Chemicals used for the synthesis of various 2-chloroacetamides are obtained from S. D. Fine – Chem Ltd., Mumbai, India. Reagents used for the preparation of the buffers were of analytical grade. Fresh triple distilled water from all glass apparatus was used in the preparation of all the solutions, mobile phase was prepared from HPLC grade methanol obtained from Ranbaxy Fine Chemicals Ltd, S.A.S Nagar, 160055, India. Human plasma was procured from the Mitra Industries Ltd., Haryana, India.

General method for preparation of glycolamide esters of Niflumic acid 3-14.

To a solution of the 2.82 g of Niflumic acid (0.01 mol) in 10ml of N,N dimethylformamide was added 1.52 ml triethylamine (0.011 mol), 0.149 g sodium iodide (0.001 mol) and appropriate 2-chloroactamide (0.011 mol). The mixture was stirred at 90°C for 2 hrs or in some cases stirred at room temperature overnight, poured into water (50 ml) and then extracted with ethyl acetate (2×50 ml). The combined extract were washed with a 2% aqueous solution of sodium thiosulphate, 2% sodium bicarbonate and water. After drying over anhydrous sodium sulphate, ethyl acetate was removed under reduced pressure to give the glycolamide esters which were purified by column chromatography.

HPLC procedure for the analysis of ester prodrugs of Niflumic acid.

The various physicochemical parameters of glycolamide ester prodrugs of Niflumic acid were determined by isocratic reversed phase HPLC procedures using a Milton Roy CDC analytical multiple solvent delivery system with UV variable wavelength detector. A 20 µl loop injection valve and 18 sphereimage OD 52, 5µM, 250 x 4.6 mm C.S chromatography. All solvents were of HPLC grade. The parent drug and the prodrug were subjected to HPLC and retention time was noted down for all the compounds using methanol as a solvent, and was eluted with methanol. The flow column effluent was monitored at 290nm. Quantification of the compounds was carried out by measuring the peak areas or peak heights in relation to those of standards chromatographed under the same conditions.

Determination of solubility

The solubility of prodrugs 3-14 was determined in 50 mM phosphate buffer of pH 7.4 at 25°C. Excess amount of the powdered prodrug was added to 2-5 ml of the buffer in the screw capped test tubes. The suspension was vortexed for 10 min and kept in a shaking incubator maintained at 25°C for 24 hrs. The suspension was transferred to a 10 ml glass syringe maintained at 25°C and filtered through a 0.45µm membrane filter in a warm test tube. After appropriate dilutions in methanol, 20 µl of the solution was injected for HPLC analysis. The concentration of the compound was calculated from standard plot obtained on the same day under similar conditions.

Determination of lipophilicity

The apparent partition coefficient (P) of the ester derivatives of Niflumic acid were determined in octanol-buffer system at 25°C. The aqueous phase was a 50 mM Phosphate buffer of pH 7.4. The buffer solution and octanol were mutually saturated before use. The traditional shake flask method11 was used, and concentrations were determined by HPLC to afford rapid evaluation and buffer reliability12-13. The compounds were dissolved in octanol (2ml) in 10 ml screw-capped test tube. After addition of buffer (5ml), two phases were mixed on shaking waterbath maintained at 25°C for 8 h. The tubes were centrifuged at 3000 rpm for 30 min. The octanol layer (1ml) was removed and properly diluted. Then injected into the HPLC column and the peak area was measured (AUCoct). The buffer solution was also removed and properly diluted, 20µl of this solution was injected and corresponding peak area was obtained (AUC buffer). The partition coefficient (P) was determined from the following expression.

P = (AUC oct/AUC buffer) x dilution factor (1)

The lipophilicity of these prodrug derivatives was also evaluated by means of reversed phase HPLC. In this method, the capacity factor (k’)14 of a solute was taken as a measure of the relative lipophilicity and was calculated as;

k’ = (tR-tO) / tO (2)

Where tR is the retention time of the solute and tO denotes elution time of the solvent. The k’ values were determined using methanol as a mobile phase. The flow rate was maintained at 1.2ml/min and column effluent was monitored at 290 nm.

Determination of chemical hydrolysis

Chemical hydrolysis of ester produrgs of Niflumic acid was studied under near physiological conditions at pH 7.4 in 50 mM phosphate buffer at 37°C. The reaction was initiated by adding 50-100 µl of stock solution (in methanol) of the ester to 20 ml of preheated buffer solution in screw capped test tubes. The final concentration of the compounds was in the range of 1.8 x 10-6 – 2.0 x 10-5 M. The solutions were kept in a water bath at constant temperature and at appropriate intervals samples were withdrawn and chromatographed. Pseudo first order rate constants for the hydrolysis of the derivatives were determined from the slopes of linear plots of the logarithms of residual derivative against time.

The pH hydrolysis of 6 and 9 was determined over the pH range 1.2 – 9.0 to determine the pH of maximum stability. The buffers used were hydrochloric acid, acetate, phosphate and carbonate buffers. A constant ionic strength of 0.5 was maintained for each buffer by adding a calculated amount of potassium chloride. Temperature accelerated studies for the same esters 6 and 9 were also performed at 80-90°C at pH of high stability to predict the shelf life at 20°C.

Determination of enzymatic hydrolysis

The enzymatic hydrolysis of the ester prodrugs of Niflumic acid having solubility of more than 5µg/ml was studied in human plasma diluted to 80% with 50 mM phosphate buffer at pH 7.4 at 37°C. The reaction was initiated by adding 20-50 µl of the stock solution of ester in methanol to 2-5 ml of preheated plasma being 4.2 x 10-5 – 1 x 10-4 M. The solution was kept in waterbath at 37°C. At appropriate time intervals samples of 100-250 µl were withdrawn and added to 1000-5000 µl of cold methanol in order to deproteinize the plasma. After immediate mixing and centrifugation for 5 min at 7000 rpm, 20 µl of the clear supernant was analyzed by HPLC for remaining ester prodrugs and the values of rate constants and half lives were calculated as described above.

RESULTS AND DISCUSSION:

Solubility and lipophilicity evaluation

It is recognized that the solubility and lipophilicity play an important role in governing the overall biological performance of drugs. For oral administration of drug it is mentioned that the drugs having octanol – water partition coefficient log P ? 2 are well absorbed provided they have minimum solubility of 10µg/ml15. To assess this potential, the solubility of the ester prodrugs 3-14 was determined in 50 mM phosphate buffer at pH 7.4 (25°C). The values are listed in Table I.

Compounds 3,10,13,14 showed solubility increased by 50 to 100 folds, compounds 4 to 7 showed solubility increased by 20 to 35 folds and compounds 8, 9, 11 and 12 showed least solubility than parent drug indicating in general cyclic glycolamide esters are least soluble compounds in 50 mM phosphate buffer of pH 7.4. The apparent partition coefficient (P) were determined between octanol and 50mM phosphate buffer of pH 7.4. The log P values obtained are listed in Table I. Compound 3 to 7 and 14 showed lipophilicy (log p) more than 1 and cyclic glycolamide esters showed lesser lipophilicity except compounds 10 and 11.

The lipophilicity of the ester derivatives was also evaluated by means of reversed phase HPLC capacity factor (k’) in methanol as mobile phase. The k’ values are listed in Table I. As has been observed with many different types of compounds a linear relationship existed between log k’ and log P16-17.

log P=5.063 log k’ + 3.7632 (3)

This relationship would very likely allow extrapolation to other esters in the series. So majority of compounds possess minimum physiochemical properties required for oral absorption.

Chemical hydrolysis Evaluation

Prodrug derivatives should be sufficiently stable so that they can be formulated in a stable dosage form. To assess stability, hydrolysis of the ester derivatives 3 to 14 was studied in 50 mM phosphate buffer of pH 7.4 at 37°C. The degradation of glycolamide ester prodrugs displayed strict pseudo first order kinetics over several half lives. Figure 1 and Figure 2 represents pseudo first order cleavage of 3 to 7 and 8 to 12 respectively. For all the compounds parent Niflumic acid was found in stoichiometric amounts.

The pseudo first order rate constants and half lives observed for hydrolysis at pH 7.4 and 37°C are listed in Table II. It can be seen from Table II that these esters were reasonably stable in 50 mM phosphate buffer, half lives being 0.29-3.23 h, among all compounds 6 and 9 showed maximum stability, half lives being 3.23 and 2.33 h respectively. In general cyclic glycolamide esters having six membered ring found to be more stable. The lower reactivity of N,N disubstituted glycolamide ester prodrugs compared to N-monosubstituted glycolamide esters may be due to increased steric hinderance to nucleophile OH- in the case of former. Among the disubstitued and cyclic derivatives compound 6 and 9 showed maximum stability. The stability of compounds 6 and 9 was further investigated over the pH range 1.2 – 9.0 at 70° to evaluate the effect of pH on degradation rate and to determine the pH of maximum stability.

The influence of pH on the rate of hydrolysis of 6 and 9 at 70°C is shown in Figure 3 and Figure 4 respectively, in which the logarithms of the observed pseudo first order rate constants (k) are plotted against pH. The U shaped pH rate profile indicates the occurance of specific acid catalysed (KH), neutral or water catalyzed (KO) and specific base catalyzed (KOH) processes according to the following rate expression.

Kobs = KH a H + KO + KOH a OH (4)

Where aH and aOH are the hydrogen ion and hydroxide ion activities respectively at reaction temperature18. The later was calculated from the measured pH at 70°C according to the equation.

log a OH = pH – 12.82 (70°C) (5)

Values of the second order rate constants KH and KOH were determined from the straight line portions of the pH rate profile at low and high pH values after adjusting the slope to –1 and +1 respectively. The values for the first order rate constant for spontaneous hydrolysis (KO) was obtained from Equation 4. In Figure 3 and Figure 4 the points shown are the log Kobs values and the solid curve is constructed from the derived rate constants using Equation 4.

The pH where the rate of hydrolysis is minimum (pH min) was found by differentiating Equation 4 and setting the derivative equal to zero. The pH min and values for rate constants at 70°C are given below

KH = 05.90 M-1h-1 (for compound 6)

KOH = 35.40 M-1h-1 (for compound 6)

pHmin= 4.66 (for compound 6)

KH = 01.28 M-1h-1 (for compound 9)

KOH = 14.90 M-1h-1 (for compound 9)

pHmin= 5.00 (for compound 9)

KO = 0.074 h-1

The effect of temperature (80-90°C) on rate of hydrolysis of 6 and 9 was also studied at pH 4.6 (stable range) to determine shelf life of the compound at 20°C. The rate constant obtained at different temperatures were plotted according to Arrhenius equation19.

log k = log A – Ea/2.303 RT (6)

Where ‘Ea’ is the energy of activation, ‘A’ denotes the frequency factor, ‘R’ is gas constant and ‘T’ represents absolute temperature.

On the basis of Equation 6, shelf life of 6 and 9 was found to be 1.37 h and 4.33 h at 20°C respectively.

Enzymatic hydrolysis evaluation

In order to be useful as prodrugs of Niflumic acid, the ester derivatives should be readily converted to parent drug in vivo. To assess the bioavailability the hydrolysis of ester prodrugs of Niflumic acid was studied in 80% human plasma pH 7.4 at 37°C. The progress of hydrolysis of all ester followed strict first order kinects over several half lives as illustrated in Figure 5 and Figure 6 for ester prodrugs 3 to 7 and 8 to 12 respectively.

The half lives and pseudo first order rate constants are given in Table II. Under the conditions all esters were quantitatively hydrolyzed to Niflumic acid as revealed by HPLC analysis. The glycolamide ester derivatives are quantitatively converted into Niflumic acid very rapidly in human plasma at 37°C and pH 7.4 except compound 8. Comparison with chemical stability data, the intrinsic reactivity of ester has no effect on their enzymatic reactivity. The rapid rate of hydrolysis observed in plasma is important in view of the slow rate of hydrolysis in the absence of plasma under similar conditions in buffer solutions of pH 7.4 at 37°C.

CONCLUSION:

Preparation of glycolamide esters of Niflumic acid was achieved in good yields by a simple synthetic route, various in-vitro experiments were carried out to evaluate glycolamide esters of Niflumic acid as a potential prodrugs. They are chemically stable to be presented in a proper dosage form, at the same time they get rapidly cleaved to parent drug Niflumic acid in 80% human plasma. Compounds 3,10,13,14 showed solubility increased by 50 to 100 folds and compounds 4 to 7 showed solubility increased by 20 to 35 folds than parent drug. Similarly compounds 3 to 7 and 14 showed lipophilicity more that 1 and cyclic glycolamide esters showed lesser lipophilicity except compounds 10 and 11. So majority of compounds showed optimum physiochemical properties required for oral absorption. These prodrugs are stable at gastric pH, so they possess the potential to avoid Niflumic acid mediated direct gastric damage while maintaining their efficacy via the systemic action of active metabolite Niflumic acid. It is concluded that esterification of Niflumic acid with substituted acetamide is useful in improving physiochemical properties as potential prodrugs, however further modification of these molecules required to achieve better solubility, shelf life, chemical stability, lipophilicity and release of more Niflumic acid from prodrugs.

1 2

Compound R1 R2

3 H CH2CH2OH

4 H CH3

5 H C2H5

6 C2H5 C2H5

7 CH3 CH3

8 -CH2CH2CH2CH2-

9 -CH2CH2CH2CH2CH2-

10 -CH2CH2CH2CH2CH2-

|

C2H5

11 -CH2CH2-CH2CH2-CH2-

| |

CH3 CH3

12 -CH2CH2-O-CH2-CH2-

13 -CH2 CH2-N-CH2 CH2-

|

CH2 CH2 OH

14 H2

Table I. Solubilities (S), Partition coefficients (P) and chromatographic capacity factors (k’) for Niflumic acid ester produgs.

Compound S(µg / ml) Log P Log k’

1 10.89a 1.32b -

3 499.95 1.09 -0.632

4 182.16 1.49 -0.115

5 262.95 1.56 -0.149

6 270.07 1.04 -0.074

7 346.42 1.33 -0.083

8 Neg 0.40 -0.560

9 41.98 0.24 -0.066

10 495.20 1.48 0.199

11 92.81 1.14 -0.007

12 22.35 0.07 -0.144

13 985.09 0.24 -0.698

14 842.44 1.21 -0.176

(S) Solubility in 50 mM phosphate buffer of pH 7.4 at 25ºC, (P) apparent partition coefficient in octanol – 50mM phosphate buffer of pH 7.4, (k’) capacity factor.

a. Solubility in 100 mM HCl solution.

b. Partition coefficient in octanol – 100 mM HCl system

Table II. Rate data for the hydrolysis of various ester prodrugs of Niflumic acid in 50 mM phosphate buffer solution and 80% human plasma at 37ºC and pH 7.4.

Compound First order rate constants Half lives

Buffer(h-1) 80% human plasma (S-1) Buffer(h) 80% human plasma (S)

3 0.10 x10-2 5.43×10-2 0.34 25

4 0.16×10-2 3.98×10-3 0.83 173

5 2.69×10-2 1.76×10-2 0.57 78

6 4.46×10-3 6.60×10-3 3.23 210

7 4.32 x 10-2 1.12 x 10-2 0.53 123

8 0.11 x 10-2 4.21 x 10-4 0.29 1643

9 2.03 x 10-2 3.74 x 10-2 2.33 74

10 5.04×10-3 2.97 x 10-2 1.15 46

11 1.01 x 10-2 5.60 x 10-3 1.15 123

12 5.22 x 10-2 1.03x 10-3 1.32 227

13 5.02 x 10-2 3.56 x 10-2 1.32 38

14 7.58 x 10-3 5.28 x 10-3 1.66 130

REFERENCES:

1. N.Dogu, A.Gul and G. Hankan, Synthesis and in vitro anti-inflammatory activities of some new diaryl amine derivatives as prodrug of Diclofenac, J Fac Pharma Gazi. 10(1) (1993) 69-81.

2. K.Vijay, M. Milind , C. Michael and K. Mansoor, Synthesis and evaluation of morpholinoalkyl ester prodrug of Indomethacin and Naproxen, Pharma Res.10(8) (1992) 1191-1199.

3. D. Paolo, P. Francesco, B.Franco, M.Lucia, D.Michele and R. Francesio, Synthesis and pharmacological evaluation of oligoethylene ester derivatives as indomethacin oral prodrugs, J Pharm Sci. 83 (11) (1994) 1578-1581.

4. Maritinadale, The extra pharmacopoeia, 29th ed., Pharmaceutical press, London, pp. 72-73.

5. W.Lalitkumar, D.Pritam. Glycolamide esters of 6-methoxy-2-naphthylacetic acid as potential prodrugs – physicochemical properties, chemical stability and enzymatic hydrolysis, Int J Pharma.118 (1995) 118 31-39.

6. H.Bundgaard, Design of prodrug, bioreversible derivatives for various functional groups and chemical entities, Elsevier, Amsterdam pp. 1-92.

7. H.Bundgaard, N. Nielsen, Esters of N,N disubstituted 2-hydroxyacetamides as a novel highly biolable prodrug type for carboxylic acid agents, J Med Chem. 30 (1987) 451-453.

8. H.Bundgaard, N.Nielsen, Glycolamide ester a novel biolabile prodrug type for non-steroidal anti-inflammatory carboxylic acid drugs,Int J Pharma. 43 (1988) 101-110.

9. H. Bundgaard, M. Johansen, N-mannich bases as novel prodrug candidates for amides, imides, urea derivatives, amines and other NH-acidic compounds, kinetics and mechanism of decomposition and structure reactivity relationships,Arch Pharm Chem Sci. 8 (1980) 29-52.

10. A. Speziale, P.Hamm, Preparation of some new 2-chloroacetamides, J Am Chem Soc.78 (1981) 2556-2559.

11. A.Leo, C. Hansch and D. Elikins, Partition coefficients and their uses, Chem Rev. 21 (1971) 525-616.

12. P.Hairsine, Direct measurement of partition coefficient by shake flask technique and gradient elution HPLC, Lab Practice. 38 (1989) 73 –75.

13. D.Stopher, S. Meclean, An improved method for the determination of distribution coefficients, J Pharm Pharmacol. 42 (1990) 144.

14. J. Chou, P.Jurs, Computation of partition coefficients from molecular structures by fragment addition method, In Yalkowsky, Sinkula SH, Valvani SC, Physical Chemical Properties of drugs, Dekker, New York ,1980, pp.199-463.

15. K.Wakita, M.Yoshimoto, S.Miyamoto and H. Watanabe, A method for calculation of the aqueous solubility of organic compounds by using new fragment solubility constants, Chem Pham Bull.34 (1986) 4663-4681.

16. T. Hafkenscheid, E. Tomlinson, Correlations between alkane – water and octanol – water distribution coefficients and isocratic reversed phase liquid chromatographic capacity factors of acids, base and neutrals, Int J Pharm. 16 (1983) 225-239.

17. J.Fort, A.Mitra, Physicochemical properties and chromatographic behaviour of a homologous series of methotrexate ?,?-dialkyl ester prodrugs, Int J Pharm.36 (1987) 7-16.

18. K.Connors, G. Aanidon , and V. Stella, Chemical stability of pharmaceutical, 2nd ed,Wiley, New York PP. 32-62.

19. A.Martin, J.Swarbrick, A.Cammaruta, Physical Pharmacy, 3rd ed, Lea and Febiger, Philadelphia,pp. 371-372.