Stabilization of quinapril by incorporating hydrogen bonding interactions

Corresponding Author:

B. N. Roy
Lupin Ltd. (Research Park), 46A, 47A - Nande Village, Mulshi Taluka, Pune-411 042, India
E-mail: bnroy@lupinpharma.com

Abstract

In the present study stability of various known solvates of quinapril hydrochloride has been compared with nitromethane solvate. Nitromethane solvate was found to be more stable compared to other known solvates. Single crystal X-ray diffraction analysis of quinapril nitromethane solvate shows intermolecular hydrogen bonding between quinapril molecule and nitromethane. Stabilization of quinapril by forming strong hydrogen bonding network as in case of co-crystals was further studied by forming co-crystal with tris(hydroxymethyl)amino methane. Quinapril free base forms a stable salt with tris(hydroxymethyl)amino methane not reported earlier. Quinapril tris(hydroxymethyl)amino methane salt found to be stable even at 80° for 72 h i.e. hardly any formation of diketopiperazine and diacid impurity. As expected single crystal X-ray diffraction analysis reveals tris(hydroxymethyl)amino methane salt of quinapril shows complex hydrogen bonding network between the two entities along with ionic bond. The properties of this stable salt - stable in solid as well as solution phase, might lead to an alternate highly stable formulation.

Keywords

Quinapril; solid state; crystal structure; stabilization; stability; solvates; co-crystal

Cyclization of dipeptides to diketopiperazine is well documented in the literature [1-4]. Angiotensinconverting enzyme (ACE) inhibitors, a class of drug molecules effective in treatment of cardiovascular dysfunctions, are also dipeptides, and hence are prone to cyclization easily to diketopiperazine. Formation of diketopiperazine is a major stability issue of concern to the potent ACE inhibitors such as enalapril, moexipril, lisinopil, perindopril, ramipril and quinapril.

Quinapril hydrochloride [5-7] (‘I’ in fig. 1) easily forms the corresponding diketopiperazine on heating and, in solution, rapidly cyclize to diketopiperazine impurity (DKP impurity, ‘II’ in fig. 1), compared to the other ACE inhibitors during storage and processing. The diketopiperazine impurity is formed either during the manufacture of quinapril hydrochloride, or during drying/further formulation of quinapril hydrochloride. The said impurity once formed is difficult to remove by conventional separation techniques including fractional crystallization.

In case of various ACE inhibitors like enalapril, moexipril and perindopril the tendency of cyclization to the corresponding diketopiperazine impurity has been minimized by forming the acid or base addition salts. The techniques employed to stabilize other ACE inhibitors were found to be unsuccessful in case of quinapril. On storage specifically at elevated temperatures quinapril or its salts have been found to cyclize to DKP and also to diacid (‘III’ in fig. 1) through hydrolysis.

The product obtained by employing the process described in US Patent No. 4,344,949 for quinapril hydrochloride results in an amorphous solid. Guo et al. [8] describe stability and chemical degradation of amorphous quinapril hydrochloride. Due to the amorphous nature it gets readily converted to the diketopiperazine impurity at 80°. Studies on crystalline quinapril hydrochloride acetonitrile solvate shows that during heating of samples, initial evaporation of solvent at 60° occurs which is followed by cyclization to the diketopiperazine impurity by loss of water and hydrogen chloride.

Desolvation of stoicheometric solvates most of the times lead to a different crystal structure or results in a disordered or amorphous state [9], with some exceptions [10-13]. In case of quinapril hydrochloride solvates, the product gets totally converted to an amorphous solid on desolvation. On heating, solid quinapril hydrochloride gets converted to the diketopiperazine impurity, in the solid stage escape of hydrogen chloride is the rate limiting step, and in case of solution phase, formation of quinapril zwitterions is the rate limiting step, which can be accelerated by increasing the pH of the solution [14]. The schematic representation of various steps involved in formation of diketopiperazine impurity as described by Li et al [14] is provided in fig. 2.

Some approaches to stabilize quinapril and/or quinapril hydrochloride are provided by Harris et al [15], Sherman [16], Klokkers et al [17,18], Li et al [14,19] and Goel et al [20]. Although attempts to improve the stability of solid quinapril have been partially successful, no work has been reported for improvement of solution phase stability.

Intermolecular interactions between quinapril molecule and solvent molecule(s) would minimize the intramolecular interactions, which are responsible for formation of the diketopiperazine impurity. The crystalline structure formed in case of solvates would minimize the mobility of molecules which is more in case of amorphous solids [8]. This postulation lead to our earlier work [21] related to quinapril hydrochloride nitromethane solvate. Nitromethane solvate is comparatively more stable than other known solvates e.g. acetonitrile solvate, due to hydrogen bonding interaction between quinapril hydrochloride molecule and nitromethane. Apparently, acetonitrile solvate of quinapril hydrochloride shows inferior stability to that of nitromethane solvate because no hydrogen bonding has been observed in single crystal X-ray diffraction pattern in the former case.

In view of the above, attempts have been made to design a system wherein addition of a molecule could improve hydrogen bonding which in turn could impart stability to quinapril hydrochloride, preferably the free base. One such approach is to form a ‘co-crystal’ with a suitable molecule, which is generally recognized as safe to human beings (a GRAS molecule).

The term ‘co-crystal’ [22-29] represents a long known class of compounds, a prototypal is quinhydrone and is broadly defined as “a mixed crystal of crystals that contains two different molecules”. Supramolecular chemistry and crystal engineering concept defines it as a consequence of a molecular recognition event between different molecular species. Basically cocrystals are made from reactants that are solids at ambient temperature; this fact distinguishes solvates/ hydrates from co-crystals. In solvates, at least one component should be a liquid at ambient temperature; in hydrates, it is water [22].

Materials and Methods

All solvents used for experiments were of HPLC grade. Tris(hydroxymethyl)amino methane and tertbutylamine were purchased from M/S S. D. Fine Chemicals Ltd., Mumbai, India. 1H NMR spectra were recorded on a Bruker 400MHz spectrometer, powder X-ray diffraction was recorded on PANalytical B.V., Netherlands model PN3040/60X’Part Pro., FT IR spectra were recorded on a Perkin-Elmer model spectrum 100, thermal analyses were done on a Mettler Toledo DSC 821e.

The solvates of quinapril hydrochloride with acetonitrile [20], acetone [30], toluene [31] and methyl formate [30] were prepared by the process reported in literature. The method as published by Singh et al [21] was utilized to prepare the nitromethane solvate. Quinapril magnesium salt and quinapril hydrochloride were prepared according to the process reported by Sharman [16] and Hoefle [5] respectively.

Preparation of nitromethane solvate of (S,S,S)2- {2-[(1-ethoxycarbonyl-3-phenylpropyl) amino]- 1-oxopropyl]-1,2,3,4-tetrahydroisoquinoline-3- carboxylic acid hydrochloride

Preparation of (S,S,S)2-{2-[(1-ethoxycarbonyl- 3-phenylpropyl)amino]-1-oxopropyl]-1,2,3,4- tetrahydroisoquinoline-3-carboxylic acid hydrochloride. (quinapril hydrochloride) was achieved by following process: (S,S,S)-2-[2-{(1-ethoxycarbonyl)-3- phenylpropyl)amino]-1-oxopropyl]-1,2,3,4-tetrahydro- 3-isoquinoline carboxylic acid benzyl ester maleate salt (quinapril benzyl ester maleate salt) 25 g (0.0388 mol) was dissolved in a mixture of 125 ml water and 125 ml dichloromethane. The pH of the solution was adjusted between 7.5 and 8.5 by addition of aqueous ammonia. Reaction mixture was stirred for 30 min, the organic phase separated and washed with 50 ml water. The organic phase was evaporated under reduced pressure below 40°, the free base of quinapril benzyl ester was obtained.

The above obtained residue of quinapril benzyl ester was dissolved in 140 ml ethanol to which 1.0 g of 10% Pd/C and 6.0 g of 35% hydrochloric acid were added. The reaction mass was subjected to catalytic hydrogenation at 40-60 psi pressure and 20-30°. The reaction mass was filtered and the filtrate evaporated to give crude quinapril hydrochloride.

The residue obtained was dissolved in nitromethane (125 ml) and the solvent recovered below 35° under reduced pressure. This operation was repeated till the water content of the residue was less than 0.5%. Nitromethane (125 ml) was added to the above residue of quinapril hydrochloride and the mixture was stirred at 20-25° for 10-15 min to get a clear solution. The mixture was stirred at the same temperature for 30 min and seeded with pure quinapril hydrochloride. The mass was cooled to 5-10° and stirred at the same temperature for 2 h. The reaction mass was then filtered and the wet cake was washed with nitromethane (50 ml), to give crystalline quinapril hydrochloride nitromethane solvate.

FTIR (KBr, ν, cm^-1): 3030, 2933, 2856, 2796, 1743, 1701, 1645, 1550, 1490, 1450, 1263, 1197, 1091, 756, 729, 707. Thermo gravimetric analysis showed weight loss of 10.44% due to loss of solvent between the temperature ranges of 40° to 125°, a further loss of 10.24% was observed between the range of 125 to 200° due to loss of hydrogen chloride and water leading to degradation. DSC (50.0-300.0°; 05.00°/ min): integral (1): 233.25 mJ; integral (2): 245.18 mJ, onset(1): 96.99°; onset(2): 151.25°, peak(1): 107.39°; peak(2): 163.53°, endset(1): 111. 55°; endset(2): 172.43°. The specific optical rotation for quinapril hydrochloride nitromethane solvate: 9.53 degrees (C= 1 in water at 20° for sodium line). X-ray powder diffraction pattern showed peaks at about 7.28, 9.06, 11.03, 11.68, 12.23, 13.67, 13.77, 14.55, 15.99, 16.62, 16.72, 17.00, 18.14, 18.55, 18.81, 19.86, 20.03, 21.55, 21.68, 21.88, 22.13, 23.43, 23.63, 24.61, 25.28, 25.95, 26.42, 27.15, 27.35, 27.69, 28.19, 28.81, 29.31, 29.96, 30.24, 30.49, 30.64, 31.69, 32.06, 33.46, 34.06, 34.35, 34.89, 36.29, 37.42, 38.13, 39.08, 39.68±0.2 °2θ.

Preparation of (S,S,S)2-{2-[(1-ethoxycarbonyl- 3-phenylpropyl)amino]-1-oxopropyl]-1,2,3,4- tetrahydroisoquinoline-3-carboxylic acid tris(hydroxymethyl)amino methane salt

Quinapril benzyl ester maleate salt 25 g (0.0388 mol) was converted to quinapril benzyl ester as per the procedure provided above. The residue of quinapril benzyl ester was dissolved in 250 ml ethanol. Palladium on charcoal (10%), 2.5 g, tris(hydroxymethyl)amino methane 4.7 g (0.0388 mol) and 60 ml water were added to the ethanol solution. The reaction mass was subjected to catalytic hydrogenation at 40-60 psi pressure at 20-30° for 2 h. After completion of reaction, the reaction mass was filtered and the filtrate was evaporated to dryness. To the residue, 100 ml acetonitrile was added and stirred for 30 min. The solid was filtered and dried at 40-45° for 10 h to give 22 g quinapril tris(hydroxymethyl) amino methane salt.

FTIR Spectra (KBr, cm^-1): Please refer to Table 1. ¹H NMR (DMSO-d₆ + CDCl₃): δ 7.23 (m, 9H), 5.12 (m, 1H), 4.86-4.60 (m, 3H), 4.07 (m, 2H), 3.65 (m, 1H), 3.34 (s, 6H), 3.15 (m, 2H), 2.48-2.87 (m, 2H), 1.83 (m, 2H), 1.02-1.2 (m, 6H). Thermo gravimetric analysis shows no weight loss. DSC (50.0-300.0°; 05.00 °/min): integral: -800.04 mJ/g; onset: 155.13°; peak: 156.44°, endset: 158.53°. The specific optical rotation for quinapril tris is: -24.34° (C = 1 in water at 20° for sodium D-line). X-ray powder diffraction pattern shows peaks at about 3.80, 7.57, 8.60, 9.33, 9.46, 11.35, 14.11, 14.38, 14.78, 15.16, 16.17, 17.16, 17.59, 17.74, 18.23, 18.96, 20.14, 20.34, 20.68, 20.86, 22.49, 22.74, 23.24, 23.81, 24.56, 25.24, 26.17, 26.67, 26.93, 28.53±0.2 °2θ.

Preparation of (S,S,S)2-{2-[(1-ethoxycarbonyl- 3-phenylpropyl)amino]-1-oxopropyl]-1,2,3,4- tetrahydroisoquinoline-3-carboxylic acid tertiary butyl amine salt [quinapril erbumine salt]

Quinapril benzyl ester maleate salt 25 g (0.0388 mol) was converted to quinapril benzyl ester as per the procedure provided above. The residue of quinapril benzyl ester was dissolved in 250 ml ethanol. To this solution, 2.5 g of 10% Pd/C and 4.25 g (0.0582 mol) tertiary butyl amine were added. The reaction mass was subjected to catalytic hydrogenation at 40-60 psi pressure at 20-30° for 2 h. After completion of reaction, the reaction mass was filtered at 50° and the filtrate evaporated to dryness. To the residue, 100 ml acetonitrile was added and stirred for 30 min. The solid was filtered and dried at 40-45° for 10 h. Dry weight of quinapril tertiary butyl amine salt: 20 g

FTIR Spectra (KBr, cm^-1) were presented in Table 1. 1H NMR (CDCl3): δ 7.23 (m, 9H), 5.12 (m, 1H), 4.86-4.60 (m, 3H), 4.11 (m, 2H), 3.77 (m, 1H), 3.3 (m, 2H), 2.65 (m, 2H), 1.1.98 (m, 2H), 1.15-1.26 (m, 6H), 0.96 (s, 9H). DSC (50.0-300.0°; 05.00 °/min): integral: 1342.57 mJ; onset: 155.30°; peak: 169.25°, end set: 180.63°. TGA reveals no weight loss. X-ray powder diffraction pattern shows peaks at about 3.86, 7.62, 8.79, 9.66, 11.39, 11.44, 14.13, 14.19, 14.58, 15.18, 16.00, 16.59, 17.06, 17.55, 17.77, 18.09, 18.40, 19.00, 19.99, 20.69, 21.18, 22.37, 22.80, 23.72, 24.74, 25.08, 25.56, 26.67, 28.36, 28.55±0.2 °2θ.

Table 1: Comparison of Ftir Spectra of Quinapril Tris, Erbumine Salt and Hydrochloride Salt.

HPLC conditions set for analysis of samples

HPLC analysis was performed using Shimazdu instrument, mobile phase used was Ammonium phosphate buffer+methane sulfonic acid [pH=4] and column used was Zorbax SB-CN, 4.6×250 mm. The detector wavelength was set to 210 nm. Throughout the analysis the flow rate was set to 1.5 ml per minute. Each sample (concentration 2000 ppm) was run for 60 min and the column oven temperature was set to 40° during analysis.

Other analytical methods and description of instruments

NMR spectra were obtained on a 400 MHz Bruker instrument, with CDCl3/DMSO-d6 as solvent. Chemical shifts (δ) are given in ppm relative to tetramethylsilane (δ = 0 ppm) or to residual protons in the solvent as internal standard. IR spectra were recorded with a PerkinElmer Spectrometer (Spectrum 100), and absorption bands are given in cm^-1. TGA were recorded on PerkinElmer model Pyris 1 at the heating rate of 10 °/min and weight loss is given in percentage. DSC was recorded on Perkin Elmer model Diamond DSC at the heating rate of 10 °/min and endothermic peak is reported in ° and ΔH is reported in J/g. Melting point was recorded on Mettler Toledo apparatus with 1 °/min heating rate. Moisture content was determined by Karl Fischer titration method with a Metrohm instrument. Measurement of specific optical rotation was done on a Jasco, Model No. P1030 at the wavelength of 589 nm.

Thermal stability studies

Quinapril hydrochloride solvates (10 g each) of nitromethane, acetone, toluene, acetonitrile, methyl formate, dry quinapril hydrochloride, quinapril tris, quinapril hydrochloride, quinapril erbumine, quinapril magnesium salt(s)were kept at 60° and 80° in different drying ovens for 48 h. Samples of each were withdrawn after every 4 h for analysis of DKP impurity. Rates of degradation of quinapril to DKP were calculated by regression analysis, wherein coefficient of correlation was more than 98%. The ratio of rate of degradation of individual salt with that of dry quinapril hydrochloride was calculated (Table 2).

Table 2: Relative Thermal Degradation Rates of Various Quinapril Forms.

Solution Phase stability study

For solution phase stability study quinapril hydrochloride/quinapril tris salt(s) (100 mg each) were dissolved in 1 ml each of water, 0.1N hydrochloride solution, 4.5 phosphate buffer solution, 6.8 phosphate buffer solution, 7.5 phosphate buffer solution in separate sample vials and kept for 30 min at 25°. Then each sample was analyzed by HPLC to detect impurity level. The results are provided in Table 3.

Table 3: Comparative Solution Phase Stability of Quinapril Hydrochloride and Tris at Different ph.

Single crystal X Ray Diffraction Analysis

Single crystal of nitromethane solvate of quinapril hydrochloride was obtained by dissolving quinapril hydrochloride in nitromethane and allowing it to crystallize. Suitable crystal of quinapril tris was selected from the bulk for X-ray diffraction analysis. The crystallographic data and details of data collection of the compounds, viz. quinapril hydrochloride nitromethane solvate and quinapril tris, are given in Tables 4 and 5, respectively. Crystal of suitable size was selected and immersed in partone oil, then mounted on the tip of a glass fiber and cemented using epoxy resin. Intensity data for the compound was collected using MoK_α (λ = 0.71073Å) radiation on a Bruker Smart Apex diffractometer equipped with CCD area detector at 100K. The data integration and reduction was processed with SAINT [32] software. An empirical absorption correction was applied to the collected reflections using SADABS [33] program. The structure was solved by direct methods using SHELXTL [34] and was refined on F² by the full-matrix least-squares technique using the SHELXL-97 [35] program package. All non-hydrogen atoms except ethyl carbon atoms of the ethyl acetate moiety (which are disordered) were refined anisotropically till convergence was reached. The occupancy factor for the disordered ethyl group was calculated by the FVAR command of the SHELXTL program and the disordered atoms were refined only isotropically. The hydrogen atoms of these disordered atoms were not included in the final refinement cycles. Hydrogen atoms attached to the ligand moieties were stereochemically fixed. The diagrams of the crystal structures were generated using the programs ORTEP5 [36], Mercury 1.4.1 [37] and PALTON [38].

Table 4: Summary of Crystallographic Data for Nitromethane Solvate of Quinapril Hydrochloride.

Table 5: Crystal Data and Structure Refinement Parameters of Quinapril Tris

Results and Discussion

Fig. 3 provides plot of percentage DKP impurity versus time when quinapril hydrochloride solvates and quinapril tris salt were kept at 60° and 80°, respectively. Rate of cyclization of quinapril to its DKP impurity is lowest in case of nitromethane solvate and highest in case of toluene solvate. Varying degradation rates observed in the solvates of quinapril hydrochloride reveal that the solvents embedded in the voids of quinapril hydrochloride crystal control the rate of escape of hydrogen chloride, the rate determining step in cyclization (fig. 2).

In spite of similarity in dipole moment between nitromethane (3.46 D) and acetonitrile (3.47 D), the nitromethane solvate is approximately twice as stable as that of acetonitrile solvate at 60°. Similarly, even though boiling points of nitromethane (101.2°) and toluene (110°) are closer, nitromethane solvate was found to be three times more stable than that of toluene solvate at 60°. To understand this phenomenon the powder and single crystal X-ray diffraction of the solvates of quinapril hydrochloride were studied. Powder X-ray diffraction pattern of all the solvates were found to be almost similar to each other, hence it is likely that the crystal systems of all these solvates would be similar.

ORTEP diagram of the quinapril nitromethane solvate is depicted in fig. 4c. Details of crystallographic data are summarized in Table 4. In an attempt to understand the interactions of the chloride and lattice nitromethane molecule with the quinapril monoanion, we have analyzed the packing and hydrogen bonding interaction of the compound in detail. Packing diagram of the compound with various hydrogen bonding interactions viewed down the a-axis for the compound is shown in fig 4a. Pairs of the protonated quinapril molecules are oriented with the flexible tethered phenyl terminal in opposite direction along c-axis to make effective intermolecular N-H…O interaction from either end between the protonated amine and the carboxylic acid oxygen O1 [N(2)…O(1)= 2.820(7) Å, < N(2)-H(2A)...O(1)= 164°] generating a cleft down a-axis. These dimeric protonated quinapril molecules are set along b-axis generating a layered network with 21 screw related cavities arranged alternately. It is interesting to note that the chloride anion and the lattice nitromethane molecules via various hydrogen bonding interactions occupy these cavities. Close-up view depicting the encapsulation of the chloride ion and nitromethane molecule is shown in fig. 4c. Chloride anion is anchored inside the cavity by involvement of three hydrogen bonding interactions O-H…Cl between the carboxylic hydrogen H2, N-H…Cl with the protonated amine and C-H…Cl of the methylene hydrogen H2B of the six membered rings. The methyl hydrogen of the nitromethane present inside the cavity are involved in strong intermolecular C-H…O interaction (C(26)….O(3)= 3.514(8) Å < C(26)-H(26B)....O(3)= 171°) with the ketonic oxygen O3. In addition to the above interactions, weak intramolecular C-H…O contact exist between (i) ketonic oxygen O3 and the H12 attached to the asymmetric carbon and (ii) the acetate oxygen O4 with the H11 of the heterocyclic ring. Details of all these pertinent hydrogen-bonding interactions, along with symmetry code, are given in Table 6.

Table 6: Hydrogen Bonding Interactions in Quinapril Nitromethane Solvate.

Hence, it is evident from the crystal structure analysis the hydrogen bonding interactions between nitromethane and quinapril hydrochloride molecule provide comparatively more stability to the product. However, such interactions are weak, if present, in case of acetonitrile solvate [39,40], and probably may not be present in the other known quinapril hydrochloride solvates as evident from their stability data.

Thermo gravimetric analysis of quinapril tris salt shows no weight loss. Differential scanning calorimetric analysis shows single endothermic peak at 156°, unlike quinapril hydrochloride showing first peak at 166° of degradation and a second peak corresponding to the melting point of the diacid impurity (III) of quinapril [8].

In FTIR spectrum (Table 1), the –OH stretching vibration (3709 cm^-1) of carboxylic function in quinapril is seen in quinapril erbumine salt but absent in quinapril tris salt. Further, it can also be noted that the carbonyl stretching vibrations seen at the wavelengths of 1647 and 1648 cm^-1 in quinapril erbumine salt and quinapril hydrochloride salt respectively are shifted to 1625 cm^-1 in case of quinapril tris salt. Shifting of carbonyl stretch and absence of –OH stretch of carbonyl function of quinapril in quinapril tris salt is attributed to involvement of these groups in intermolecular hydrogen bonding.

An ORTEP [37] view of asymmetric unit of quinapril tris salt with atom numbering scheme is shown in fig. 5. The conformation of the quinapril molecule is established by the torsion angles N1-C12-C13-N2 150.3(5)˚, N2-C16-C17-C18 -62.5(6)˚ and C16-C17- C18-C19 -167.8(5)˚. The hetero ring (C1/N1/C3/C4/ C9/C10) assumes a screw-boat conformation with ring puckering parameters [41] Q=0.551(5) Å, θ=69.7(6)˚ and φ=-27.9(6)˚, respectively. The molecular structure of the title compound is stabilized by several intermolecular hydrogen bonds (Table 7) between quinapril and tris molecules. Intermolecular N(3)- H(31A)O(2), O(8)-H(8A)O(3) and O(7)-H(7A)O(6) hydrogen bonds connect the molecules into R33(18) rings which are corner fused to form one-dimensional polymeric chains propagating along the [010] direction (fig. 6). Adjacent one-dimensional chains are joined by O(6)-H(6A)O(2) and N(3)-H(31B) O(1) hydrogen bonds forming molecular channels which are further linked via C-H hydrogen bonds and interactions, so generating a three-dimensional framework (fig. 7).

Table 7: Selected Intermolecular Contacts in Quinapril Tris.

It is evident from the single crystal X-ray diffraction and FTIR data, the interactions between quinapril and tris(hydroxymethyl)amino methane occurs by two phenomena viz. (a) formation of amine salt (b) complex intermolecular hydrogen bonding network. Hence, it imparts stability not only by the ionic bond but also through hydrogen bonds similar to co-crystal formation, which was not possible in quinapril erbumine and quinapril hydrochloride salts as evident from comparison of thermal stabilities of the quinapril tris salt and quinapril erbumine salt (Table 2)

Impurity levels in quinapril tris surprisingly remains unchanged at 80° even up to 72 h, however impurity levels in quinapril hydrochloride, quinapril erbumine and quinapril magnesium increased significantly. Rate of formation of DKP impurity in quinapril tris salt is the lowest (Table 2).

Solution phase stability study of these salts reveals significant increase in concentration of DKP and diacid impurity at different pH in case of quinapril hydrochloride after 30 min at 25°, however there is no significant change in the impurity level in case of quinapril tris salt (Table 3). Solubility study of quinapril tris salt and quinapril hydrochloride in water, ethanol and methanol at 20° reveals that quinapril tris salt has comparable solubility as that of quinapril hydrochloride (Table 8).

Table 8: Comparison of Solubility of Quinapril Tris Salt with Hydrochloride Salt.

The single crystal X-ray diffraction analysis of acetonitrile solvate of quinapril hydrochloride [39,40] does not show hydrogen-bonding interaction between the solvent molecule and the host. The solvent molecule just fit into the cavities of the crystal lattice to impart stability. On the other hand, nitromethane solvate of quinapril hydrochloride is comparatively more stable among all the other solvates due to strong hydrogen bonding interaction between nitromethane molecule and quinapril molecule, thereby imparting superior stability.

In quinapril tris salt, quinapril is not only bound with tris(hydroxymethyl)amino methane by the way of ionic bond, but also bound with a complex hydrogen bonding network involving large number of hydrogen bond donors and acceptors. This formation of complex hydrogen bonding network imparts solid-state stability as evident from comparison of thermal stabilities quinapril tris salt and quinapril erbumine salt, as tert-butylamine does not contain large number of hydrogen bond donors like that of tris (hydroxymethyl)amino methane. The interactions between quinapril and tris(hydroxymethyl)amino methane occurs by two phenomena viz. (a) formation of amine salt (b) complex intermolecular hydrogen bonding network. Hence, it imparts stability not only by the ionic bond but also through hydrogen bonds similar to co-crystal formation, which was not possible in quinapril erbumine and quinapril hydrochloride salts.

Quinapril tris salt would be a potential alternative to be formulated as a drug product to the existing solvates and salt forms of quinapril in view of its stability and comparable solubility to that of quinapril hydrochloride.

Acknowledgements

The authors are grateful to Prof. Alok Kumar Mukherjee and Dr. Santu Chakraborty, of Department of Physics, Jadavpur University, Kolkata-700 032, India, for their help in carrying out single crystal analysis of quinapril tris salt. The authors are also grateful for the help in solving the crystal structure of Quinapril hydrochloride nitromethane by Dr P. K. Ghosh, Dr Parimal Paul and Dr E. Suresh of Central Salt & Marine Chemicals Research Institute, Bhavnagar–364 002, India. The authors are also thankful to Dr Pritesh R Upadhyay of Analytical Research Department, Lupin Ltd. (Research Park), Pune for assisting in analysis of various samples of this work and Dr Vishvas D Patil of Intellectual Property Cell, Lupin Ltd. (Research Park), Pune for assisting in drafting the manuscript and literature search.

References

1. Steinberg SM, Bada JL. Peptide decomposition in the neutral Ph region via the formation of diketopiperazines. J Org Chem 1983;48:2295-8.
2. Chimanlal G, Ronals TB. Kinetics of diketopiperazine formation using model peptides. J Pharm Sci 1998;87:283-8.
3. Purdie JE, Benoiton NL. Piperazinedione formation from esters of dipeptides containing glycine, alanine and sarcosine: The kinetics in aqueous solution. ChemSoc Perkin Trans 2. 1973;13:1845-52.
4. Steinberg SM, Bada JL. Diketopiperazine formation during investigations of aminoacid racemization in dipeptides. Science 1981;213:544-5.
5. Hoefle ML, Klutchko S. Substituted acyl derivatives of 1,2,3,4-tetrahydroisoquinoline-3-carboxylic acids. US Patent No. 4,344, 949, August 17, 1982.
6. Jiyon TS, Jiefurii EB, Jiyon AR. N-(substituted aminoalkanoyl) heterocyclic compounds. GB Patent No. 2,095,252, September 29, 1982
7. Patchett AA, Wu MT. Isoquinoline carboxylic acid derivates, process for preparing and pharmaceutical composition containing the same. EP Patent No. 0,065,301, November 24, 1982.
8. Guo Y, Byrn SR, Zografi G. Physical characteristics and chemical degradation of amorphous quinapril hydrochloride. J Pharm Sci 2000;89:128-43.
9. Ulrich JG. The importance of solvates. In: Hilfiker R, editor. Polymorphism in pharmaceutical Industry. Weinheim: Wiley-VCH; 2006. p. 211-33.
10. Brittain HG. Polymorphism and solvatomorphism 2006. J Pharm Sci 2007;97:3611-36.
11. Baure J, Morley J, Spanton S, Leusen FJ, Henry R, Hollis S, et al. Identification, preparation, and characterization of several polymorphs and solvates of terazosin hydrochloride. J Pharm Sci 2006;95:917-28.
12. Rollinger JM, Burger A. Physicochemical characterization of hydrated and anhydrous crystal forms of amlodipine besylate. J Thermal Anal Calorimetry 2002;68:361-72.
13. Petit S, Coquerel G. Mechanism of several solid−solid transformations betweendihydrated and anhydrous copper(II) 8-hydroxyquinolinates. proposition for a unified model for the dehydration of molecular crystals. Chem Materials 1996;8:2247-58.
14. Li J, Guo Y, Zugrufi G. Effects of a citrate buffer system on the solid-state chemical stability of lyophilized quinapril preparations. Pharm Res 2002;19:20-6.
15. Harris M, Hokanson G, Murthy K, Reisch R, Waldman F. Stabilized compositions. US Patent No. 4,743,450, May 10, 1988.
16. Sherman BC. US Patent No. Pharmaceutical compositions comprising quinapril magnesium. 6,531,486, March 11, 2003.
17. Klokkers K, Kramer KT, Fischer W, Sendl-Lang A. Matrix controlled transdermal system for stabile derivatives of ace inhibitors. US Patent Application No. 20040052835, March 18, 2004.
18. Klokkers K, Helfrich M, Nink J. Matrix controlled transdermal system for stabile derivatives of ace inhibitors. WO 2007065638, June 14, 2007.
19. Li J, Guo Y, Zografi G. The solid-state stability of amorphous quinapril in the presence of beta cyclodextrins. J Pharm Sci 2002;91:229-43
20. Goel OP, Krolls U. Crystalline quinapril and a process for producing the same. US Patent No. 4,761,479, August 2, 1988.
21. Singh GP, Rawat GS, Dhake VN, Nehate SP. Crystalline form of quinapril hydrochloride and process for preparing the same. EP1572661, September 14, 2005.
22. Vishweshwar P, McMahon JA, Bis JA, Zaworotko MJ. Pharmaceutical Co-crystals. J Pharm Sci 2006;95:499-516.
23. Etter MC, Adsmond DA. The use of co-crystallization as a method of studying hydrogen bond preferences of 2-aminopyridine. J ChemSoc ChemCommun 1990;589-91.
24. Etter MC. Graph-set analysis of hydrogen-bond patterns in organic crystals. ActaCrystallogr B 1990;46:256-62.
25. Etter MC, Urbanczyk-Lipkowska Z, Zia-Ebrahima M, Panunto TW. Hydrogen bond directed co crystallization and molecular recognition properties of diarylureas. J Am ChemSoc 1990;112:8415-26.
26. Carl HG, Hans PH On the inclusion of solvent molecules in the crystal structures of organic compounds. ActaCrystallogr B 2000;56:525-34.
27. SenthilKumar VS, Nangia A. Molecular complexes of some mono- and dicarboxylic acids with trans-l,4,-dithiane-l,4-dioxide. Crystal Growth Design 2002;2:325-8.
28. Desiraju GR. Crystal and co-crystal. Crystal EnggCommun 2003;5:466-67.
29. Almarsson O, Bourghol HM, Peterson M, Zaworotko MJ, Moulton B, Rodriguez-Hornedo N. Pharmaceutical co-crystal compositions of drugs such as carbamazepine, celecoxib, olanzapine, itraconazole, topiramate, modafinil, 5-fluorouracil, hydrochlorothiazide, acetaminophen, aspirin, flurbiprofen, phenytoin and ibuprofen. US Patent Appl. 20070059356, March 15, 2007.
30. Jennings SM. Preparation of quinapril hydrochloride. US patent No. 6,858,735, February 22, 2005.
31. Monsalvatje LM, Bartra SM, Tomas NJ, Puig TS. Process for obtainingquinapril hydrochloride and solvates useful for isolating and purifying quinapril hydrochloride. US patent No. 6,617,457, September 9, 2003.
32. Sheldrick GM. SAINT 5.1 ed., Siemens Industrial Automation Inc., Madison, WI. 1995
33. SADABS, empirical absorption Correction Program; Göttingen, Germany, University of Göttingen: 1997.
34. Sheldrick GM. SHELXTL reference manual: Version 5.1, Madison, Bruker AXS, WI. 1997.
35. Sheldrick GM. SHELXL-97: Program for Crystal Structure Refinement; Göttingen, Germany, University of Göttingen: 1997.
36. Spek AL. PLATON for MS-Windows. Journal of  AppliedCrystallography 2003;36:7-13.
37. Farrugia LJ. Ortep-3 for Windows. J Applied Crystallogr 1997;30:565- 65.
38. Macrae CF, Edgington PR, McCabe P, Pidcock E, Shields GP, Taylor R, et al.Mercury: visualization and analysis of crystal structures. J Applied Crystallogr 2006;39:453-57.
39. Hausin RJ, Codding PW. Molecular and crystal structure of MDL27,467A Hydrochloride and Quinapril Hydrochloride, Two Ester Derivatives of Potent Angiotensin Converting Enzyme Inhibitors. J Med Chem 1991;34:511-7.
40. The Cambridge Structural Database (CSD)- The world repository of small molecule crystal structures The Cambridge Crystallographic Data Centre, Copyright © 2004-2008.
41. Boeyens JCA. Calculation of ring puckering parameters. J crystallogrMol Structure 1978;8:317.