Enantioseparation of Nadifloxacin by High performance liquid Chromatography

A rapid isocratic chiral HPLC method has been developed for the separation of R -Nadifloxacin from S-Nadifloxacin. Good resolution viz. Rs> 4.0 between Rand Sforms of Nadifloxacin was achieved by RP-HPLC using endcappedC18 stationary phase and chiral mobile phase. Chirality to the mobile phase was imparted with addition o f -CD in phosphate buffer with EDTA.Column temperature was 45 ° C and flow rate was kept 1.5 mL min -1 .The elution was monitored by UV-vis detector at -290 nm. The calib ration curve showed excellent linearity over concentration range 0.040-20 μg mL -1 .Th is method was further used to determine the amount of R-Nadifloxacin in pure and active pharmaceutical ingredient of S-Nadifloxacin and is capable to quantitate and detect R-Nadifloxacin to the levels of 0.040μg mL -1 and 0.020μg mL -1 respectively. The average recovery of R-Nadifloxacin was 99.09 %. The method is better than the already reported one for the enantioseparation of the Nadifloxacin.


Introduction
In recent years, there has been considerable interest in the synthesis and separation of enantiomers of organic compounds especially because of their importance in biochemistry and pharmaceutical industry [1][2][3][4][5]. The method frequently used for separations, monitoring the progress of an asymmetric synthesis or optical purity of the product is chromatography with either liquids, gases or supercritical flu ids as the mobile phase. Separation of the enantio mers comprising the racemate i.e. the resolution of the racemate is a common problem in stereochemical research as well as in the preparation of bio logically active co mpounds, in particular drugs. One approach to separate enan tiomers sometimes referred to as indirect enantio meric resolution which involves the coupling of the enantiomers with an auxiliary chiral reagent to convert them in to diastereomers [6][7][8][9][10][11]. The diastereomers can be separated by any achiral separation technique [1]. The chirality of the enantiomeric molecu les is caused by the presence of one or more chiral atom in their structure. The chirality sense and optical activity of the enantiomers are determined by their absolute configuration.
Reversed phase high performance liquid chro matography (RP-HPLC) techniques were used to separate and quantify enantiomers with high resolution. (S)-(-)-Nadifilo xacin[(s)-( -)-9-fluoro-6,7-dihydro-8-(4-hydro xypiperidino)-5-Methyl-1-o xo -1H,5H-benzo[i,j]quino lizine-2-carbo xy licacid] is a synthetic quinolone derivative with potent broad spectrum antibacterial activ ity and inhibits the enzyme DNA gyrase that is involved in the synthesis of bacterial DNA and its replicat ion [12]. Nadiflo xacin is active against aerobic Gram-positive, Gram-negative and anaerob ic bacteria includ ing Propioniba cterium acnes and Staphylococcus epidermidi.The enantiomer co mpounds (fig-1a&1b) vary greatly in their biological act ivity. As a result of it, the assessment of the biological activ ity of each enantiomer of a chiral molecu le has been a standard practice in order to produce drugs and food products mainly as single enantio mers [13].
Cyclodextrin (CD) based chiral stationary phases (CSPs) or mobile phase additives are extensively applicab le to separate enantiomers of large class of compounds [14][15].These are quite successful in separating the enantiomers of chiral molecules with aro mat ic substituents [16][17][18][19][20]. Consequently, cyclodextrin based CSPs are natural choice for addressing the liquid chro matographic chiral separation of these compounds .The determination of enantiomeric purity of the co mpound was, therefore, developed using cyclodextrin as a chiral selector. Because of low cost and unique physical characteristics, β-CD is the most popularly used phase among the three phases of CD. A survey of literature reveals that an attempt has been made for the assay and quantification of R-(+) enantio mer impurity in the bulk drug wh ich is L-arginine salt tetrahydrate of Nadiflo xacin (W CK771) by HPLC using β-CD based CSPs [1]. However, enantiomeric resolution of Nad iflo xacin has not yet been carried out. A method has, therefore, been developed for the same with better resolution, lower limit of quantification (LOQ) and higher degree of reproducibility.

Instrumentation
Chro matographic analysis is performed with waters 2695-separation module coupled with photodiode array detector (Co mpounds were detected at -290 n m) and column YM C C18 (250 mm x 3.0 mm) 3 m pore size 120 Ă is used for chromatography. The oven temperature of HPLC was at 45°C and in jection volu me is 20 L. The flow rate of the mobile phase was adjusted at 1.5 mL min -1 and the total run time is 30 min. Chro matographic data were controlled and processed on a computer running with millenniu m empower pro version 5.00.00.00.

Chemicals
Nadiflo xacin is fro m Beijing Nine camp Medicine Technology Co.Ltd. Chaoyang (CHINA), Sodiu m hydro xide pellets, Triethylamine, d i-sodiu m hydrogen orthophosphate (anhydrous) and Potassium di-hydrogen orthophosphate (all of AR grade); water and acetonitrile (HPLC grade) are fro m qualigens (India). Ethylene diaminetetraacetic acid (EDTA) disodium salt and -Cyclodextrin (both of A R grade) fro m Across chemicals (India) were used during studies.

Preparati on of Stock Solution for Resolution
5.0 mg of R-Nadiflo xacin is weighed accurately and transferred into a standard 50 mL volu met ric flask. It is thoroughly dissolved with 2 mL of 0.1 N NaOHsolution and the volume was made up with diluent up to the mark and mixed thoroughly.

Preparati on of Soluti on for Resolution
50 mg of S-Nadiflo xacin is weighed accurately and transferred in to 25 mL volu metric flask. It is dissolved with 2 mL of 0.1 N NaOH and 0.5 mL of resolution stock solution is added. The solution is now made up with diluent up to the mark and thoroughly mixed. It is then filtered through 0.45µm filter or finer porosity memb rane filter.

Preparati on of Sample Sol ution
50 mg of S-Nadiflo xacin is weighed accurately and transferred in to 25 mL volu metric flask. It is dissolved with 2 mL of 0.1 N NaOH. The solution is now made up with diluent up to the mark and thoroughly mixed. It is then filtered through 0.45µm filter or finer porosity me mbrane filter.

Preparati on of Buffer
1.0 g of Potassium d i-hydrogen orthophosphate, 0.50 g of di-Sodiu m hydrogen orthophosphate (anhydrous), 50.0 mg of E.D.T.A and 10.0 g cyclodext rin hydrate is dissolved in 1000 mL of water by constant stiring. It is then filtered through a 0.45µm filter or finer porosity memb rane filter.

Preparati on of Mobile Phase
A mixed degassed solution of buffer and acetonitrile (ACN) in various ratios (v/v) is prepared and its pH is adjusted to the desired level with triethylamine. The mobile phase is also used as diluent.

Results and Discussion
Enantio meric separation by HPLC is generally made by immob ilizing the single enantiomers on to the stationary phase and resolution is the result of the format ion of transient diastereomer by the interaction of CSPs. Enantiomer which form the most stable diastereomer is retained and opposite enantiomer fo rming less stable diastereomer will elute first. Interactions between CSPs and enantiomers are very week and require carefu l optimizat ion by adjustment of the suitable mob ile phase and temperature of the colu mn to maximize the enantioselectivity. These interaction forces are ionic, π-π interaction, hydrophobic effect and hydrogen bonding [6].
For cyclodextrin produced by the action of Bacillus macerans amy lase on starch, the size of CD depends upon the type of reaction between the t wo. CDs are the cyclic oligosaccharides containing six to twelve or eight D (+) glucopyranose units and bonded through alpha 1-4 lin kages and form truncated conical cavity, the d iameter of which depends upon the number of glucopyranose units.
Co mmercially availab le CDs are pictured as hollow truncated cones, the torrodial structure having a hydrophilic surface. -phase has seven units of glycopyranose the value ranging fro m (6.0-8.0 Ă). β -CD has the widest application due to its pronounced kin k shape whereas α-CD and -CD are more p laner. Its molecule has secondary -2 and -3 hydroxyl g roups of the glycopyranose units lining at the mouth of the cavity and primary 6-hydro xyl groups at the rear of the cavity (fig-2).

Figure 2. Structure of β-Cyclodextrin
This means that the cavity itself is a relatively hydrophobic region of the molecule and permits the inclusion of the hydrophobic portion.
The cavity is composed of the glucoside oxygen and methylene hydrogens giving it an apolar character. Fo r the solute molecules, polar region of the mo lecules interact with the hydroxyl groups on the surface of cavity and provides the three point interaction [15,20,21]required for the chiral recognition.
The basic property of cyclodextrins that allo w them to effect a large number of chemical separations is their ab ility to form select ive inclusion complexes with a variety of guest mo lecules. The format ion of this inclusion complex [22] may be caused by (a) A hydrophilic effect (b) Hydrogen bonding (c) Release of high energy water or modifier during comp lex formation (d) Co mbined effect of all these factors.
In general, binding to the cyclodextrin is governed by the mo lecule's ability to closely fit the cavity of the CD alongwith the polarity of the molecule itself. This fit depends both on size and shape of analyte concerning the CD cavity. For too small and too bulky mo lecules, there will be little or no binding at all. Larger molecu les can be bound to CD, if certain groups or side chains of the molecule can penetrate the cavity effectively. The hydrophobic character of the cavity in a particular orientation is responsible for the stereoselective discrimination of chiral molecules that are fitted with in to the CD cav ity during chro matography.
The enantiomeric separation due to chiral addit ive (β-CD) in the mob ile phase may involve any of the following general mechanis m.
(i) Adsorption of the chiral selector to the solid phase in insitu format ion of a temporary chiral stationary phase (CSP).
(ii) Stereoselectivecomp lexat ion in the mobile phase.
(iii) Formation of the labile diastereo co mplexes with different distribution properties between the stationary and mobile phase.
However, the co mpetition between mob ile phase and stationary phase for β-CD plays a very important role. As studied earlier [23],a co mparison of co lu mn characteristics before and after the co lu mn is exposed to β-CD mob ile phase, shows that column characteristics (retention time, efficiency, peak shape etc) are not changed when a β-CD mod ified mobile phase containing 10 % A CN is used. It shows that there is no adsorption of β-CD on to the stationary phase in case of 10 % ACN solution used as the mobile phase. In the present case, this is most suitable mobile phase composition for enantio meric resolution of Nadiflo xacin (Table-1). The first important consideration to form a stable inclusion co mplex is proper fit of the molecu le to the CD-cavity. As a general rule, substituted phenyl, naphthyl and biphenyl rings can be included in the cavity of β-CD [24]. Thus, the retention mechanism is a two equilibria process i.e. a reversible equilibriu m of the solute in the bulk solvent mobile phase (a) with the stationary phase sites to from a complex and (b) with the β-CD in the mob ile phase to form an inclusion comp lex.
It has been observed earlier in case of L-argin ine salt tetrahydrate of Nadiflo xacin that the analyte form inclusion complex with β-CD and is stabilized by van der waal, hydrogen-bonding and hydrophobic interactions. The interactive selectiv ity is caused by the spatial d ifference between the isomers. β-CD is, therefore, used as chiral mobile phase additive (CMPA). The main advantage of this method is that the R-Nadiflo xacin elutes first and avoids "smearing" under main co mpound peak [1].
Seeman et al [25] observed that analytes which form more stable inclusion co mplexes would be retained longer and be eluted later and those forming less stable comp lexes will be eluted earlier. S-Nadiflo xacin can enter the β-CD cavity to form an inclusion co mplex in the reversed phase mode which leads to the observed chiral separation. It forms relat ively more stable inclusion comp lex and is eluted later. In addition to the solubility of β-CD in the mobile phase used, the position of the substituents on the aromatic ring on one hand and the hydrophilic interaction between the -OH groups outside the cavity of β-CD on the other are the determining factors for the stability of inclusion comp lexes formed.
It is well established that, for a CD to form an enantioselectivediastereomeric co mplex, the substituents of the stereogeniccenter of analyte must be in close proximity to thesecondary-OH groups at the mouth of the CD in order to achieve the necessary three points of interaction [15,20,24]. If the portion of the molecule having stereogeniccenter resides in the cavity of CD upon inclusion, the stereogeniccenter will be buried inside the CD torus, not in the close pro ximity to the secondary-OH groups on the larger rig of the molecule. In this case, the substituent (-CH 3 ) on or near the analyte'sstereogeniccenter will be unable to interact with portion of the chiral selector that is most responsible for the chiral recognition. It can be concluded that when Nadiflo xacin form an enantioselective inclusion co mplex with β -CD in the reversed phase mode, their stereogeniccenter is located near the mouth of the CD selector.
However, the main driving force for the inclusion complex format ion is through the release of enthalpy -rich water fro m the cavity due to the entrapping of guest mo lecules. Van der Waal forces, Hydrogen bonds and hydrophobic interactions help to keep the co mplex together. No covalent bonds are formed or bro ken during inclusion complex format ion. Therefore, the inclusion complexation process can be considered as a rep lacement of water mo lecules with Nadiflo xacin (Drug) mo lecules. This view is supported by the explanation given by yang [26] for β-cyclodextrinco mplexat ion and formu lation as an anti-HIV microbe.
In aqueous phase hydrophobic cavity of β-cyclodextrin is occupied by the water mo lecules, wh ich is thermodynamically unfavorable. Therefore the water mo lecules have less tendency to form the hydrogen bonds in the same way as in the solutions and result in a h igher enthalpy and higher energy.When hydrophobic guest mo lecules are incorporated in to the system, the energy of the system is lowered by substituting these enthalpy rich-water mo lecules with those of hydrophobic Nadiflo xacin (Drug) or guest molecules to form the co mplex of β-cyclodextrin and guest molecules [26].
Steric d iscrimination of the Nadiflo xacin enantio mers can affect the formation of inclusion co mplex. Due to this discrimination S-Nad iflo xacin will form mo re stable inclusion complex with β-cyclodext rin in co mparison to R-Nadiflo xacin. Therefore, S-Nadiflo xacin will elute later and resolution takes place in between R and S-Nad iflo xacin .Thus, due to mult ifunctional mechanism derived fro m mo lecular inclusion and chemical interactions, the method is quite successful in the enantioseparation of Nadiflo xacin. The method is superior to the already reported one [1] for the enantioseparation of Nadiflo xacin because of better resolution, lower limit of quantification (LOQ) and higher degree of reproducibility.

Optimization of the HPLC Enantioseparation
The main features of the enantioseparation of racemic mixtu re of Nadiflo xacin are given below. Tables 1-3 shows the effect on the USP (United States Pharmacopeia) resolution of the variation in the co mposition of the mixture of buffer and acetonitrile alongwith variation in final p H value and the flow rate of the mobile phase. The extracted chromatogram of S -Nad iflo xacin spiked with R-Nadiflo xacin (Sp iked at 0.1% level or 2 µg mL -1 respectively with R-Nadiflo xacin) is shown in figure-3 wh ile that forS-Nadiflo xacin is shown in figure-4. An excellent resolution (Rs=4.10) between the enantiomers and ideal peak shape with tailing factor 1.08 was obtained.

Validation of Analytical Method
After the systematic optimization of the method the final conditions were found to be as follows: mobile phase (buffer and ACN) rat io is 90:10, pH 7.4, flow rate 1.5 mL min -1 and total run time is 30 min.

System Suitability
Performance of the method was determined by in jecting the resolution solution. The qualification criteria were resolution between two enantio mers wh ich should be not less than 2.0 and tailing factor not more than 1.5, which ensures baseline separation and symmetrical peak shape for R-Nadiflo xacin.

System Precision
The System precision of the analytical method was determined by injecting the replicate injection of R-Nadiflo xacin with concentration 2 µg mL -1 .

Method Precision
Precision of the method was determined by the analysis of six spiked sample amount of R-Nadiflo xacin at concentration of 2 µg mL -1 shown in (table -4).

Ruggedness
Ruggedness of the method was determined by the analysis of six spiked sample amount of R-Nadiflo xacin at concentration of 2 µg mL -1 on different instrument and different colu mn shown in (table -4).

Accuracy
Accuracy of the method was confirmed by determin ing the % recovery of spiked amount ofR-Nad iflo xacin at concentration 80, 100 and 120 % of 2 µg mL -1 in thepre-analyzed samp le of Nadiflo xacin and is shown in (table -5).  The linearity of an analytical procedure is its ability (within a given range) to obtain test results which are directly proportional to the concentration (amount) of analyte in the sample. The Linearity of concentration vs peak area of R-Nadiflo xacin in this method over a concentration range of R² = 0.9993 0.00 18000.00 36000.00 54000.00 72000.00 90000.00 108000.00 126000.00 144000.00 162000.00 180000.00

Li mit of Detecti on and Quantificati on
The limits of detection and quantification of R-Nadiflo xacin in this method are su mmarized (table -6). As per Eu rochem gu idelines, the % RSD of six replicate injection peak area should be not more than 10 for LOQ. In case of LOD, the % RSD of six replicate in jection peak area should be not more than 33. The limit of quantification and limit of detection for R-Nadiflo xacin are 0.040 µg mL -1 and0.020 µg mL -1 respectively.

Conclusions
A simple, rap id and linear RP-HPLC method containing chiral mob ile phase additive is given for the quantitative separation of R-Nadiflo xacin fro m S-Nadiflo xacin. The method is simp le as it does not need any derivatizat ion to diastereomers and economical too. It excludes the use of chiral stationary phase and uses very inexpensive β-cyclodext rin as chiral mobile phase additive and easily available RP-HPLC C18 co lu mn for ch ro matography.