Stability of drugs in pharmaceuticals: kinetics and mechanisms of [PDF]

IN FO R M A TIO N TO USERS

This was produced from a copy of a document sent to us for microfilming. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the material submitted. The following explanation of techniques is provided to help you understand markings or notations which may appear on this reproduction. 1.T h e sign or "target” for pages apparently lacking from the document photographed is "Missing Page(s)”. If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure you of complete continuity. 2. When an image on the film is obliterated with a round black mark it is an indication that the film inspector noticed either blurred copy because of movement during exposure, or duplicate copy. Unless we meant to delete copyrighted materials that should not have been filmed, you will find a good image of the page in the adjacent frame. If copyrighted materials were deleted you will find a target note listing the pages in the adjacent frame. 3. When a map, drawing or chart, etc., is part of the material being photo­ graphed the photographer has followed a definite method in "sectioning” the material. It is customary to begin filming at the upper left hand corner of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again—beginning below the first row and continuing on until complete. 4. For any illustrations that cannot be reproduced satisfactorily by xerography, photographic prints can be purchased at additional cost and tipped into your xerographic copy. Requests can be made to our Dissertations Customer Services Department. 5. Some pages in any document may have indistinct print. In all cases we have filmed the best available copy.

University

Microfilms

International

300 N. ZEEB R D „ A NN ARBOR, M l 48106

8207176

D’Silva, Joseph Baptist

STABILITY OF DRUGS IN PHARMACEUTICALS: KINETICS AND MECHANISMS OF HYDROLYSIS IN LIPOSOMAL SUSPENSIONS AND AQUEOUS SOLUTIONS

PhD. 1981

The Ohio State University

University Microfilms International

300 N. Zeeb Road, Ann Arbor, M I 48106

STABILITY OF DRUGS IN PHARMACEUTICALS:

KINETICS AND MECHANISMS

OF HYDROLYSIS IN LIPOSOMAL SUSPENSIONS AND AQUEOUS SOLUTIONS

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the Graduate School of The Ohio State University

By

Joseph B. D'Silva, B. Pharm.

* * * * *

The Ohio State University 1981

Reading Committee: Professor Professor Professor Professor

Louis Malspeis, Ph.D. Robert E. Notari, Ph.D. Larry Robertson, Ph.D. Theodore Sokoloski, Ph.D.

Approved By

Adviser The College of Pharmacy

This dissertation is dedicated to,

MY MOM AND DAD

For all the unselfish love and encouragement you have bestowed and, the constant understanding you have shown this is a small token of my gratitude for a debt that can never be repayed.

ii

ACKNOWLEDGEMENTS

I wish to thank the following people for their helpful assistance during these past few years: My sister Anita and my brother Eugene for all their love and encouragement. My friend Dr. Alan Castellino for his friendship and support. My roommates and friends, Paul Savina and, Bob Henderson for all their help and understanding, especially during exam times. My adviser Dr. Robert E. Notari for having taught and guided me in the field of scientific research and, his help in the work involved in this dissertation. Dr. L. Malspeis for having supplied a sample of lndlcine Noxide and, his help in the research associated with the same. My friends and fellow graduate students, especially M. Hixon, L. McWhorter, B. Tozer and, Mr. and Mrs. Henderson. Mrs. Kay Cassidy for her assistance in typing this dissertation.

ill

VITA

July 17, 1955...................... Date of Birth 1976................................Bachelor of Pharmacy, Department of Chemical Technology, University of Bombay Bombay, India 1976 - present..................... Pre-doctoral student, Division of Pharmaceutics and Pharmaceutical Chemistry, The College of Pharmacy, The Ohio State University Columbus, Ohio. 1976-1977..........................

Fellow of the International Rotary Foundation

PUBLICATIONS Joseph B. D'Silva, Robert E. Notari, Stability-indicating Colorimetric Assay for Indicine N-oxide Using TLC, J. Pharm. Sci., 69 (1980) 471472. Joseph B. D*Silva, Robert E. Notari, Kinetics and Mechanisms of Aqueous Degradation of the Anticancer Agent, Indicine N-oxide, Int. J. Pharm., 8 (1981) 45-62.

Major Field of Study:

Pharmaceutics and Pharmaceutical Chemistry Studies in Chemical Stability of Drugs in Pharmaceut ical Formulat ions.

iv

Summary of Post-graduate Coursework:

Physical Pharmacy:

Protein Binding, Complexation, Solubilization, Stability, Interfacial Phenomena, Ionic Equilibria. Methods in Pharmaceutical Research: Drug Stability. Radioisotope Tracer Techniques. Isolation Techniques. Principles and Methodology in Pharmacokinetics. Advanced Organic Chemistry. Basic Physical Chemistry. Advanced Physical Chemistry: Chemical Kinetics, Thermodynamics. Pharmaceutical Analysis: GC, HPLC, GC-MS, and TLC. Physical Methods in Organic Chemistry Identification: NMR, MS, UV. Statistics

i

v

TABLE OF CONTENTS Page DEDICATION...................................................... ACKNOWLEDGEMENTS...............................................

11

Hi

VITA.............................................................. LIST OF TABLES..............

iv viii

LIST OF FIGURES............................................ •LIST OF SCHEMES................................................

x xiii

CHAPTER I

II

III

STABILITY-INDICATING COLORIMETRIC ASSAY FOR INDICINE N-OXIDE USING TLC............................

1

Summary............................................. Introduction........................... Experimental.......... Results and Discussion............................. References..........................................

2 3 5 9 11

KINETICS AND MECHANISMS OF AQUEOUS DEGRADATION OF THE ANTICANCER AGENT, INDICINE N-OXIDE....................

14

Summary...................................... Introduction........................................ Materials and Methods.............................. Results and Discussion ....................... References..........................................

15 16 18 24 41

KINETICS OF THE HYDROLYSIS OF CYCLOCYTIDINE,INDOMETHACIN, AND P-NITROPHENYLACETATE IN LIPOSOMAL SUSPENSIONS.................................

62

Summary............................................. Introduction.................. Statement of Problem................................ Experimental........................................ Results............................................. vi

63 64 71 76 83

Page Discussion.......................................... References.......................................... APPENDIX.......................................................

vii

87 99 117

LIST OF TABLES TABLE 2.1

2.2

2.3

2.4

3.1

3.2

3.3

3.4

3.5

Page Reaction Conditions and Apparent First-Order Rate Constants for Hydrolysis of Indicine N-oxide..........

44

Initial Concentrations (M/liter) of Indicine N-oxide [CD ] and Sodium Hydroxide [NaOH] and Observed Bimolecular Rate Constants.............................

46

Carbon-13 Shifts and Assignments for Retronecine N-oxide and Retronecine Hydrochloride..................

47

Half-life Values ( t n O for Hydrolysis of Pyrrolizidine Alkaloid Esters in 0 15 N NaOH at 2 5 ° .................

48

Apparent Molar Absorptivities of Reference Standards Recovered from Liposomal Suspensions...................

104

Apparent First-Order Rate Constants for Hydrolysis of Cyclocytidine in Buffers (kg) and in Liposomal Suspensions (kg, kg), at 40°, and the Mean Diameters of Liposomes Before (dQ) and After (d) Hydrolysis

105

Apparent First-Order Rate Constants for Hydrolysis of Indomethacin in Buffers (kg) and in Liposomal Sus­ pensions (kg, kg), at 40°, and the Mean Diameters of Liposomes Before (dQ ) and After (d) the Reaction......

106

Apparent First-Order Rate Constants for the Hydrolysis of p-Nitrophenylacetate in Buffers (kg) and in Liposomal Suspensions (kg, kg, a, 8 ), at 40°, and the Mean Diameters of Liposomes Before (dQ ) and After (d) the Reaction............................................

107

Reaction Concentrations as a Function of Time of Cyclocytidine (C ), Arabinosylcytosine (C ) and C the Sum of C anS C , as a Percentage of the Initial Concentration of Cyclocytidine C , in (A) Borate Buffer, (B) Positively-charged Liposomes, (C) Negatively-charged Liposomes and, (D) Neutral Liposomes...............................................

108

viii

Page

TABLE 3.6

Reaction Concentration as a Function of Time, of pNLtrophenylacetate ( C ), p-Nitrophenol (C ), and C the Sum of C and C , as a Percentage of the Initial Concentration CQ , o¥ p-Hitrophenylacetate in (A) Phosphate Buffer, (B) Positively-charged Liposomes, (C) Negatively-charged Liposomes and, (D) Neutral Liposomes..........................................

ix

109

LIST OF FIGURES Figure

Page

1.1

Chemical Structure of Indicine N-oxide..................

12

1.2

Beer's Law Plots for the analysis of indicine N-oxide: (A) without TLC treatment; and (B) with TLC treatment....

12

Concentration (M x 10^) of Indicine N-oxide as a Function of Time at 30° in (A) Aqueous 0.15 N NaOH; and (B) Aqueous 0.065 M Na^PO^ at pH » 12.18.............

13

Chemical Structure of Indicine N-oxide..................

49

1.3

2.1 2.2

2.3

2.4

2.5

2.6

Semi-ln Plots Based on Eqn. 2.3 to Determine the Bimolecular Rate Constant, k.2 “ slope/ (a-b), for Hydrolysis of -0.04 M. Indicine N-oxide in -0.08 N NaOH at (A) 50°, (B) 40° 1 and (0) 30^ ...... 1 ................................ ^

49

Semi-log plots of the First-order Rate Constants (ki) for Indicine N-oxide Hydrolysis as a Function of pH at (A) 90°, (B) 60°, (C) 40° and (D) 30°, where lines are Drawn to a Slope of 1.0 in Agreement with Eqn. 2.4. Insert Shows Expanded Scale for (D) at pH 12.6-13.1 to Illustrate Lack of Difference between Buffered ( ■ ) and Unbuffered ( # ) Rate-constants ....................................

50

Chemical Structure of Trachelanthic Acid Showing the Hydrogen Bonding Between the Carbonyl and Beta Hydroxyl Groups..................................

50

Proton Nuclear Magnetic Resonance Spectrum of Trachelanthic Acid in CDCI 3 , Measured at 90 M c/s........

51

Carbon-13 Nuclear Magnetic Resonance Spectrum of Trachelanthic Acid in CDCI 3 , Measured at 80 Mc/s........

52

2.7

Electron Impact Mass Spectrum of Trachelanthic Acid.

2.8

Chemical Ionization Mass Spectrum of Trachelanthic Acid..

54

2.9

Proton Nuclear Magnetic Resonance Spectrum of Retro­ necine N-oxide measured at 90 Mc/s in D 2O................

55

x

53

Figure

Page

2.10

Chemical Structure of Retronecine N-oxide..............

56

2.11

Carbon-13 Nuclear Magnetic Resonance Spectrum of Retronecine N-oxide in D 2 O Measured at 80 M c/s ........

56

2.12

Chemical Structure of Europine N-oxide.................

57

2.13

Chemical Structure of Heliotrine N-oxide............

57

2.14

Electron Impact Mass Spectrum of Retronecine N-oxlde...

58

2.15

Electron Impact Mass Spectrum of Indicine N-oxide

59

2.16

Chemical Ionization Mass Spectrum of Indicine N-oxide..

60

2.17

Structure of Indicine N-oxide Showing the Hydrogen Bonding Between the Ester Carbonyl and the Beta Hydroxyl Group..........................................

61

Semi-In Plots of (A) kj vs 1/T (Eqn. 2.6) and, (B) k 2 /T vs 1/T (Eqn. 2.8) where ^ is the Bimolecular Rate-constant for the Hydrolysis of Indicine N-oxide in Aqueous Sodium Hydroxide (Table 2.2)................

61

2.18

3.1

3.2

3.3

Time Course for Cyclocytidine in, [A] pH * 8.31 Borate Buffer (•), and in Neutral Liposomal Suspension (SI); and in [B] Positively Charged Liposomal Suspension ( ® ) and, in the Filtrate Sampled from the same ( A ) . Least Squares Regression Lines Shown Represent # , ® , and A ......

110

Time Course for Indomethacin in, pH = 9.37.Carbonate Buffer ( • ) ; and in Positively-Charged (®), NegativelyCharged (A ) , and Neutral ( 0 ) Liposomal Suspensions. Lines Drawn are from Least Square Regression Analysis of the Data...........................

Ill

Time Course for Indomethacin in, Negatively Charged Liposomal Suspension ( # ) and in the Filtrate Sampled from the same (ffi). Lines Drawn are from Least Square Regression Analysis of the Data. Liposomes are Prepared in pH - 9.37 Carbonate Buffer.................

112

xi

Figure 3.4

3.5

3.6

Page Time Course for Indomethacin int Positively Charged ( # ) and Neutral ( A ) Liposomal Suspension; and Values of R Determined from the Same. Lines Drawn are from Least Square Regression Analysis of the Data. Liposomes are Prepared in pH « 9.37 Carbonate Buffer...

113

Time Course for p-Nitrophenylacetate in pH ■ 7.51 Phosphate Buffer ( • ) ; and in Negatively-Charged ( A)» and Neutral (ffl) Liposomal Suspensions. Lines Drawn are from Least Square Regression Analysis of the Data......................................... ...........

114

Time Course for p-N.itrophenylacetate in PositivelyCharged Liposomal Suspensions. Line Drawn is from Non-Linear Regression Analysis of the Data. Lipo­ somes are Prepared in pH ■ 7.51 Phosphate Buffer

.

115

Possible locations of a compound, D, in a liposomal suspension; D^ is the compound dissolved in the external aqueous phase; D£ and ^2 * are the Compound associated with the outer and inner walls of the liposome respectively; Dg is the compound associated with the hydrocarbon portion of the lipid bilayers; D 4 is the compound dissolved in the aqueous portions in the interior of the liposome and; Dg and Dg' are the compound solubilized in the hydrocarbon portion of the lipid bilayer with portions protruding into the bulk aqueous environment and the aqueous layers in the interior of the liposome respectively...........

116

xii

LIST OF SCHEMES

SCHEME

I

II

III

IV

V

VI

VII

VIII

IX

X

Page

Reaction Products of the Hydrolysis of Indicine Noxide in Aqueous Sodium Hydroxide..................

22

Fragmentation Patterns for the Electron Impact and Chemical Ionization Mass Spectra of Trachelanthic Acid......................................................

29

Fragmentation Pattern for the Electron Impact Mass Spectrum of Retronecine N-oxide..........................

32

Fragmentation Pattern for the Electron Impact Mass Spectrum of Indicine N-oxide.............................

34

Fragmentation Pattern for the Chemical Ionization Mass Spectrum of Indicine N-oxide.............................

35

Mechanism for the Hydrolysis of Indicine N-oxide in Aqueous Sodium Hydroxide Solutions and, pH ■ 6.1 11.52 Aqueous Buffers............... ..................

36

Hydrolysis of Cyclocytidine in pH ■ 8.31 Borate Buffer Solutions...........................................

72

Hydrolysis of Indomethacin in Aqueous Sodium Hydroxide and, pH « 9.37 Carbonate Buffer........................

73

Hydrolysis of p-Nitrophenylacetate in Aqueous Sodium Hydroxide and, pH - 7.51 Phosphate Buffer........

75

Two Compartment Model for the loss of Indomethacin in Positively-charged, Negatively-charged and, Neutral Liposomal Suspensions, prepared with pH ■ 9.37 Carbonate Buffer; and p-Nitrophenylacetate in Negatively-charged and Neutral liposomal suspensions prepared with pfl ■ 7.51 Phosphate Buffer................

89

xiii

SCHEME

Page

XI

Three Compartment Model for the Loss of p-Nitrophenylacetate in Positively-Charged Liposomal Suspensions Prepared with pH ■ 7.51Phosphate Buffer... 95

XII

Modified Three Compartment Model for the Loss of p-Nitrophenylacetate in Positively-Charged Liposomal Suspensions Prepared with pH ■ 7.51Phosphate Buffer... 96

XIII

Pseudo-two Compartment Model for the Loss of p-Nitro­ phenylacetate in Positively-Charged Liposomal Suspensions Prepared with pH ■ 7.51 Phosphate Buffer...

xiv

97

CHAPTER I STABILITY-INDICATING COLORIMETRIC ASSAY FOR INDICINE N-OXIDE USING TLC*

*

Published in part in J. Pharm. Sci., 69 (1980) 471-72.

SUMMARY

Aliquots of aqueous solutions in which indicine N-oxide may be degraded were mixed with 0.5 M formic acid (1:3) to adjust the pH to ~2-4 to quench the reaction and to ensure adequate TLC resolution. Silica-coated aluminum sheets were used to isolate indicine N-oxide by cutting the appropriate region from the chromatogram.

By a modifica­

tion of a known procedure, the silica gel then was treated with an acetic anhydride-diglyme mixture, and the mixture was heated to convert the drug to a pyrrole, which was then coupled with 4-dimethylaminobenzaldehyde to produce a color.

The absorbance of the resulting

solution was determined at 566 nm, and the apparent molar absorptivity, e, based on the final indicine N-oxide concentration was 6.13 x 10^. The recovery was -92% and the assays were readily reproducible with a coefficient of variation of 4.4%.

2

INTRODUCTION

Indicine N-oxide

(Fig. 1.1), an unsaturated pyrrolizidine alkaloid

found in Heliotropium indicum Linn (Boraginaceae) (1), is undergoing clinical testing as an anticancer agent (2).

While stability data have

not been reported, alkaline ester hydrolysis is predicted from degrada­ tion studies of related alkaloids. Approximate half-lives for the decomposition of 12 pyrrolizidines were estimated in 0.5 N aqueous or hydroalcoholic sodium hydroxide at room temperature (3).

The relatively facile hydrolysis of esters of

trachelanthic and viridifloric acids was attributed to their potential for 3 -hydroxyl participation, presumably via hydrogen bonding (3). This potential for intramolecular catalysis is present in indicine Noxide, which also is a trachelanthic acid ester (of retronecine Noxide).

The products obtained from the hydrolysis of indicine in 2 N

NaOH at 100° for 2 hr were shown to be retronecine and a diastereoisomer of trachelanthic acid (4).

It is not known whether the

presence of the N-oxide in indicine N-oxide gives rise to additional degradation pathways.

Kugelman et al. (1) found that the properties

of indicine N-oxide extracted from H. indicum Linn (Boraginaceae) did not agree with those of synthesized indicine N-oxide.

1 NSC-132319

3

Although

these differences were ascribed to solvation problems, they also might reflect chemical instability. Two assays for indicine N-oxide in biological samples have been reported (2,5).

An electron-capture GLC assay after formation of the

pentafluoropropionic anhydride derivative of indicine was applied to the analysis of mixtures of indicine and indicine N-oxide in plasma and urine (2).

Prior to analysis, indicine was extracted selectively

with chloroform; the indicine N-oxide remaining in the raffinate then was reduced to indicine.

A GLC-mass spectrometric method using selective-

ion monitoring recently achieved nanogram sensitivity through formation of the trimethylsilyl derivative of indicine N-oxide (5).

Although the

method is selective for Indicine N-oxide, the equipment required is sophisticated and expensive. This study was undertaken to develop a simple, specific assay for indicine N-oxide in the presence of its degradation products.

A

colorimetric assay for unsaturated pyrrolizidine alkaloids using modified Ehrlich reagent (6 ) was adapted to assay indicine N-oxide under aqueous conditions, in which it was shown to be unstable.

TLC

on aluminum sheets was employed to isolate indicine N-oxide from buffers and reaction products.

After the appropriate region was cut and

scraped, the silica gel mixture was treated to convert the pyrrolizi­ dine structure to a pyrrole.

The pyrrole then was coupled with 4-

dimethylaminobenzaldehyde to produce a color which was measured spectrophotometrically.

EXPERIMENTAL

Materials and Chemicals - The TLC aluminum sheets were precbated with 2 0.2 mm of silica gel 60 F-254 .

3 4 Ether , absolute ethanol , ammonium

5

6

hydroxide solution , acetic anhydride , and acetone were analytical 7 reagent grade.

The diglyme

was kept free of peroxides (6 ).

Modified

Ehrlich reagent was prepared by dissolving 2% (w/v) 4-dimethylamino8

benzaldehyde

in an ethanolic solution containing 14% (w/v) boron

trifluoride, which was incorporated as its etherate complex (6,7).

pH Adjustment - The assay results, obtained from absorbance values, were consistent provided that the sample to be spotted had a pH of 2-4.5.

At pH

a\

37 activation energy and enthalpy of activation would be nearly equal to or slightly lower than for indicine N-oxide if intramolecular catalysis is taken from

into account (Fig. 2.17).

To obtain

the second-order experiments (Table

these values, the

data

2.2) were treatedusing the

Arrhenius equation:

-E /RT

.

k 2 ■ Ae

(Eq. 2.5)

where k 2 is the bimolecular rate-constant, A is the pre-exponential factor, E

is the energy of activation, R is the gas constant, and T is

the absolute temperature.

The log transformation of Eq. 2.5 is:

E In k, » In A - — ^ RT

(Eq* 2.6)

Fig. 2.18 presents the In k 2 vs 1/T plot for the data in Table 2.2. The activation energy for the attack of the hydroxide species on the indicine N-oxide molecule, obtained from the slope of the line, is 16.1 kcal/mole.

To obtain the enthalpy and entropy of activation,

second-order kinetic data (Table 2.2) were analyzed using the following equation:

k2 - H [ e ^ / R ] [e-A H*/RT * Nh

(Eq. 2.7)

where k2 is the bimolecular rate-constant, R is the gas constant, T is the absolute temperature, N is Avogadro1s number, h is Planck's

38 constant, AS* is the entropy of activation, and AH* is the enthalpy of activation.

Dividing by T followed by a log transformation gives:

(Eq. 2.8) ln(k,/T) z

Fig. 2.18 also shows

1„£_ Nh

R

RT

the lnCfc^/T) vs 1/T plot for the data in Table 2.2.

The value of AH* obtained from the slope

of the line is 15.5 kcal/mole-deg.,

A and that of AS

obtained from the intercept is -18.8 e.u.

Hydrolysis of ethyl acetate in aqueous sodium hydroxide has been reported by Halonen (18) and Pan et al. (19). energy was 11.45 (+0.35) kcal/mole.

The average activation

The data on Halonen (18) were

analyzed using Eqs. 2.7 and 2.8 to obtain values for AH

♦

and AS

*

which

were averaged with the values reported by Pan et al. (19), to obtain an average AH* of 10.85 (+0.35) kcal/mole and AS* of -26.55 (+1.25) e.u. The slower hydrolysis rate of indicine N-oxide is therefore associated with its higher energy of activation. to play a prominent role.

Inductive effects do not appear

By employing the substituent constants

reported by Charton (20) the calculated inductive potential of the tranchelanthic acid side-chain is negligible since the negative hydroxyl effects if offset by

the contribution of the alkyl groups, steric

hindrance appears to

be the prime reason for the reduced hydrolysis

rate and increased E

a

value.

An increase in the E„ values for the a

alkaline hydrolysis of several aliphatic esters with aryl and alkyl substitutions in the acyl part has been attributed to steric hindrance (21),

39 (22).

The alkaline stability of ethyl diisopropylacetate and ethyl

dicyclopentylacetate (23) illustrate just how effective steric factors can be.

The rule of six (24) states that atoms separated from the attack­

ing species in the transition state by a chain of 4 atoms provide the greatest steric hindrance.

In the case of ethyl

diisopropylacetate

and ethyl dicyclopentylacetate there are 12 and 8 hydrogen atoms respectively.

Indicine N-oxide has 10 hydrogen atoms in the trachel-

anthic acid portion that can cause maximum steric hindrance according to this rule. The higher value of AS* for indicine N-oxide is probably an anomoly and is found in other cases as well.

The hydrolysis of ethyl diphenyl-

acetate when compared to that of ethyl phenylacetate would be expected to result in a lower value of AH* due to the negative inductive effect of the second phenyl group and a lower AS* value due to additional steric hindrance.

The data for the alkaline hydrolysis of the two species

in aqueous ethanol reported by Levenson and Smith (22) were analyzed using Eqs. 2.7 and 2.8 to obtain values of AH* = 13.4 kcal/mole and 15.4 kcal/mole for ethyl phenylacetate and ethyl diphenylacetate, respectively. -7.91 e.u.

The corresponding values of AS* are -8.91 e.u. and

Thus, steric hindrance again is reflected by the value of

AH* rather than AS*. The hydrolysis data for 4 pyrrolizidine alkaloid esters listed in Table 2.4 reconfirms the acceleration of the rate by the 8 -hydroxyl group.

This has previously been attributed to hydrogen bonding between

the 8 -hydroxyl and the carbonyl group

(Fig. 2.17, (5)).

It also appears

40 that the presence of the hydroxyl group (Rg) Increases the hydrolysis rate by a factor of approximately 2 0 0 .

REFERENCES

1.

C. C. J. Culvenor, Tumor-inhibitory activity of pyrrolizidine alkaloids. J. Pharm. Sci., 57 (1968) 1112-1117.

2.

M. Kugelman, W. C. Liu, M. Axelrod, T. J. McBride and K. V. Rao, Indicine N-oxide: The antitumor principle of Heliotropium lndicum. Lloydia, 39 (1976) 125-128.

3.

M. S. Kupchan and M. I. Suffness, Tumor inhibitors XXII. Senecionine and senecionine N-oxide of Senecio triangularis. J. Pharm. Sci., 56 (1967) 541-543.

4.

M. M. Ames and 6 . Powis, Determination of indicine N-oxide and indicine in plasma and urine by electron-capture gas-liquid chromatography. J. Chromatogr., 1 6 6 (1978) 519-526.

5.

L. B. Bull, C. C. J. Culvenor and A. T. Dick, The Pyrrolizidine Alkaloids, North-Holland, Amsterdam, The Netherlands, 1968, pp. 32, 61, 57.

6.

J. B. D 1Silva and R. E. Notari, Stability-indicating colorimetric assay for indicine N-oxide using TLC. J. Pharm. Sci., j69 (1980) 471-472.

7.

A. R. Mattocks, R. Schoental, H. C. Crowley and C. C. J. Culvenor, Indicine: the major alkaloid of Heliotropium lndicum L. J. Chem. Soc., (1961) 5400-5403.

8.

M. McComish, I. Bodek and A. R. Branfman, Quantitation of the anti­ neoplastic agent indicine N-oxide in human plasma by differential pulse polarography. J. Pharm. Sci., 69 (1980) 727-729.

9.

N. K. Kochetkov, A. M. Likhosherstov and V. N. Kulakov, The total synthesis of some pyrrolizidine alkaloids and their absolute configuration. Tetrahedron, 25 (1969) 2313-2323.

10.

R. Adams and B. L. Van Duuren, Trachelanthic and viridifloric acids. J. Am. Chem. Soc., 74 (1952) 5349-5351.

11.

A. R. Mattocks, Spectrophotometric determination of unsaturated pyrrolizidine alkaloids. Anal. Chem., 57 (1967) 443-447.

41

42

12.

C. C. J. Culvenor, M. L. Heffernan, and W. 6 . Woods, Nuclear magnetic resonance spectra of pyrrolizidine alkaloids. Austr. J. Chem., 18 (1965) 1605-1624.

13.

J. C. N. Ma and E. W. Warnhoff, On the use of nuclear magnetic resonance for the detection, estimation, and characterization of N-methyl groups. Can. J. Chem., ^ 3 (1965) 1849-1869.

14.

L. M. Jackman and S. Sternhell, Applications of Nuclear Magnetic Resonance Spectroscopy in Organic Chemistry, Pergamon Press, England, 1969, p. 81-82.

15.

N. V. Mody, R. S. Sawhney and S. W. Pelletier, Carbon-13 nuclear magnetic resonance spectral assignments for pyrrolizidine alka­ loids. J. Nat. Prod., 42 (1979) 417-420.

16.

L. H. Zalkow, L. Gelbaum and E. Kleinan, Isolation of the pyrrolizidine alkaloid europine N-oxide from Heliotropium marls-mortui and H. rotundifolium. Phytochemistry, 17 (1978) 172.

17.

N. Neuner-Jehle, H. Nesvadba and G. Spiteller, Anwendung der massenspektrometric Zur strukturaufklarung von alkaloiden, 6 . Mit: Pyrrolizidinalkaloide aus dem Goldregen. Montash. Chem. 96 (1965) 321-338.

18.

E. A. Halonen, Activation energies in the alkaline hydrolysis of saturated aliphatic esters. Acta Chem. Scand., 10 (1956) 485-486.

19.

K. Pan, C. F. Chang and H. S. Hong, Kinetic studies of the alkaline hydrolysis of alkyl acetates by conductometric measurements. J. Chinese Chem. Soc., Ser. II, 9. (1962) 89-99.

20.

M. Charton, Definition of "inductive" substituent constants. J. Org. Chem., 29 (1964) 1222-1227.

21.

H. S. Levenson and H. A. Smith, The saponification of ethyl esters of aliphatic acids. J. Am. Chem. Soc., 62 (1940), 1556-1558.

22.

H. S. Levenson and H. A. Smith, Kinetics of the saponification on the ethyl esters of several phenyl-substituted aliphatic acids. J. Am. Chem. Soc., 62 (1940) 2324-2327.

23.

J. Von Braun und F. Fischer, Beitrage zur kenntnis der sterischen Hinderung, VII Mitteil.: Veresterung und Verseifung vom standpunkt der elektronischen Theorie der Bindung. Ber., 66 (1933), 101-104.

24.

M. S. Newman, Some observations concerning steric factors. Am. Chem. Soc., J72 (1950) 4783-4786.

/

J.

43

25.

H. S. H a m e d and B. B. Owen, The Physical Chemistry of Electrolytic Solutions, Reinhold, New York, 1958, p.. 729.

26.

H. S. H a m e d and M. A. Cook, The activity and osmotic coefficients of some hydroxide-chloride mixtures in aqueous solution. J. Am. Chem. Soc., 59 (1937) 1890-1893.

44 TABLE 2.1

Reaction Conditions and Apparent First-Order Rate Constants

No.

tva4a

n 4r T m

Fig. 2.18

Seml-ln Plots of (A) ^ vs 1/T (Eqn. 2.6) and, (B) kj/T vs 1/T (Eqn. 2.8) where ko Is the Blmolecular Rateconstant for the Hydrolysis of Indlclne N-oxlde.In Aqueous Sodium Hydroxide (Table 2.2).

CHAPTER III KINETICS OF THE HYDROLYSIS OF CYCLOCYTIDINE, INDOMETHACIN, AND P-NITROPHENYLACETATE IN LIPOSOMAL SUSPENSIONS

62

SUMMARY

The rates of hydrolysis of cyclocytidine, indomethacin, and p-nitrophenylacetate were studied in buffered suspensions of positivelycharged, negatively-charged, and neutral liposomes.

The apparent first-

order rate constants were compared to those in the corresponding buffers. Association between the compounds and the liposomes was evaluated by comparing assays for total suspension to those for the filtrates. Cyclocytidine showed no change in hydrolysis rate constants in the pre­ sence of liposomes and no association could be ascertained.

These

observations were attributed to the high degree of hydrophilicity of cyclocytidine.

In contrast indomethacin was found to be associated

with the liposomes (72% - 90%), and showed a decrease (68 % - 85%) in rate constants, which was attributed to selective partitioning of the hydrophobic portion of the ionized indomethacin into the lipid bilayers. p-N itrophenylacetate displayed a 31% and 26% decrease in rate constants in negatively-charged and neutral liposomal suspensions respectively where it was 91% and 8 6 % associated.

Appreciable water solubility

together with an overall neutral character, were suggested as explana­ tions for these hydrolysis characteristics of p-nitrophenylacetate. Hydrolysis of p-nitrophenylacetate in positively-charged liposomal suspensions was much more rapid than in aqueous buffer, and could be described by a biexponential equation.

This acceleration was attributed

to surface catalysis involving -stabilization of the transition state. 63

INTRODUCTION

Background - Liposomes were inadvertently prepared in 1958 when lecithin concentrations as low as 10 ““* gm/ml were claimed to form micellar solutions.

However molecular weight measurements and light

scattering experiments showed these structures to be smectic lamellae and not true micelles.

Bangham was one of the first to suggest that

these structures could have biological implications (1).

Consequently,

they were initially called "Bangosomes", though eventually they became known as liposomes, and as such have attracted a great deal of attention as potential drug delivery systems. Phospholipids in the presence of an aqueous environment, spontaneous­ ly tend to form composite structures consisting of several concentric bilayers of lipids separated by aqueous compartments (2).

These lipo­

somes in the presence of excess water represent preferred configuration for most phospholipids (3).

In these lipid bilayers, the hydrocarbon

parts of the phospholipids comprise the inner portion of the bilayer and the polar groups are exposed to the aqueous environment. These bilayers tend to be intact thus preventing the exposure of the hydrocarbon surfaces to the aqueous environment and forming con­ centric lipid bilayers separated by aqueous layers with the liposome interior itself being shielded from the exterior aqueous environment (4).

This tendency of lipids to isolate themselves from aqueous

64

65

environments is due to the water molecules having greater mutual attraction for each other than for the hydrocarbon chain of the lipid

(1). Liposomes can be classified as unilamellar or multilamellar, according to their physical structure where the former are composed of only one bilayer of phospholipids. upon the charge: liposomes.

Another classification is based

positively charged, negatively charged, and neutral

Liposomes are most commonly composed of the phospholipid,

lecithin, which is often combined with cholesterol to produce neutral liposomal suspensions.

The addition of a cationic lipid such as pro-

tonated stearylamine imparts a positive charge to the structure, whereas anionic dicetyl phosphate gives the liposome a negative charge. Liposomes can be prepared by several methods.

One of the more

widely used methods involves the hydration of a dry film of lipid material by suspending it in an aqueous buffer.

The resultant large

multilamellar liposomes can be sonicated using either a probe-type or cup-horn sonicator (5).

The latter has the advantage of not contaminat­

ing the sample with metal fragments from the probe.

Prolonged sonica-

tion produces liposomes in the size range of 280 & (6 ).

Large unilamellar

liposomes have been prepared by injecting an ether solution of the lipids into a warm aqueous solution (7).

Liposomes of uniform size distribution

and homogeneity have been prepared by extruding suspensions of multi­ lamellar vesicles through polycarbonate membranes (8 ).

A similar method

utilized a French pressure cell to produce liposomes with diameter ranges of 315-525 & (9).

Other procedures include dialysis which

66 produces

large unilamellar liposomes (1 0 ), and calcium-induced

fusion methods to prepare multilamellar cylindrical liposomes (1 1 ). Drugs to be encapsulated are generally Introduced into the aqueous phase during preparation.

A certain fraction of this drug is incorporated

Into the liposomes, and the non-associated portion may then be separated. The liposomes can be separated by centrifugation (12) or by ultra­ filtration using a membrane filter and nitrogen under pressure to concentrate the liposomes (13).

Molecular sieve chromatography using

a sepharose 4B column has also been employed.

Absorption of lecithin

on the gel can be a problem with this method and therefore special precautions have to be taken (6 ). Liposomes were initially used as models to study various properties of biological membranes which they resemble in structure and composition. They were utilized to determine the interaction of ligands with glycolipids incorporated in the liposomal membranes.

The interaction of

choleragen with the ligands mimicked the characteristics and specificity noted with biological membranes (14).

The thermodynamic properties of

bacteriorhodospin were determined using liposomes as a theoretical model (15).

Other studies Include the study of molecular motions within

membranes (16) and the Interaction of erythrocytes with local anaesthetics (17) and antibiotics (18).

Mechanisms involving catecholamine release

have been studied using stimulants entrapped within liposomes (19)• Increased intracellular sodium has been shown to induce catecholamine release from adrenal medulla by mobilizing intracellular calcium.

The

work of de Gier et al. (20) gives a good description of the relationships

67 between liposomes and biological membranes. The greatest potential of liposomes has been their possible use as drug delivery systems, as which they possess some unique advantages. One of these advantages is the protection of encapsulated drugs from degradation.

One example is insulin which if administered orally is

destroyed by the gastric juices and consequently has to be given by injection.

Insulin encapsulated with liposomes and administered orally

to animals has been found to produce decrease in blood sugar levels which are comparable to intraperitoneal or intravenous routes of administration (5, 21).

Polynucleotides, such as immune RNA, encapsu­

lated in liposomes are protected from degradation and can potentiate the production of interferon.

This therapy may find application in the

treatment of cancer and viral infections (2 2 ). Another advantage of encapsulating drugs in liposomes is the possibility of using smaller doses by targeting the liposomes to specific sites and tissues, thus reducing toxicity and side-effects. Cortisol palmitate in liposomes has been used to treat rheumatoid arthritis wherein only one-twenty fifth of the conventional therapeutic dose produced improvement in synovitis (23). Liposomes have been used in the treatment of cancer in a number of novel procedures.

Liposomes containing ergosterol have been used

to sensitize leukemic cells to the polyene antibiotic, amphotericin-B (24).

Liposomes containing lymphokines have been used to control

metastases by activating the macrophages in vivo (25).

The sensitivity

of liposomes to pH and temperature has been used to attempt to target

68

drugs to tumors.

By incorporating pH-sensitive drugs, such as palmitoyl

homocysteln into liposomes pH-speclflc release can be obtained due to changes In the environment.

Such liposomes might prove clinically useful

if they enable drugs to be targeted to areas of the body wherein the pH is less than physiological such as primary tumors and metastases (26). Liposomes with phase transitions a few degrees above physiological temperture were reported to deliver more than four times as much methotrexate to murine tumors heated to 42° as to unheated control tumors.

Higher

ratios could be obtained by optimizing liposome size, composition, and charge.

By combining local hyperthermia with generalized hypothermia

to increase the available temperature, this procedure could be used to treat localized tumors without exposing the rest of the body to the toxic slde-effects of anti-cancer agents (27). Liposomes have also been used in other dosage forms, such as the topicals.

Using triamcinolone as a model drug, liposomal encapsulation

favourably altered the drug disposition (28).

The concentration of the

drug decreased at the site of its adverse effects in the systemic circulation and increased in the skin where its activity is desired.

Stability of Drugs - The increased use of liposomes as potential drug delivery systems has not been paralleled by research work on the stability of drugs in these dosage forms.

Over the past few years there have been

only a few literature references on drug stability in liposomal sus­ pensions.

The rate of hydrolysis of procaine in neutral liposomal

suspensions was found to be slower than in the aqueous buffer, in the pH

69 range of 8.5 to 13 (29).

Two-diethylaminoethyl p-nltrobenzoate was

found to hydrolyze at a faster rate In the pH-range of 5.5 to 7.8 and slower In the pH-range of 7.8 to 10 In the presence of neutral liposomes as compared to plain aqueous buffer (30).

Both of these studies

determined the hydrolysis rates by measuring the amount of drug In the total liposomal suspension as a function of time.

Determinations of the

drug associated with the liposomes were not carried out. Micellar systems are somewhat analogous to liposomal suspensions. Micelles are formed by long-chain amphlphiles such as trlethylammonium, sulfate and carboxylate salts in aqueous systems above a certain concentration called the critical micellar concentration.

The hydro-

phobic part of the aggregate forms the core of the micelle which is hydrocarbon in nature while the polar head groups are located at the micelle water interface. liposomal systems.

This is similar to the lipid bilayers of the

The decrease in free energy of the system which

results from the preferential self-association of the hydrophobic hydrocarbon chains is the primary reason for the formation.

Detailed

explanation of the micelle formation theories are given by a number of authors (31, 32, 33). The degradation of a number of drugs has been studied in micellar systems.

The base catalyzed hydrolysis of benzocaine was depressed in

the presence of cationic and anionic micelles, with up to eighteen-fold increases in half-lives.

A dilute solution of the quaternary, cetyl-

trimethylammonium bromide, slightly accelerated benzocaine hydrolysis. The hydrolysis was shown to occur both inside the micelles as well as

70

In the aqueous phase (34).

The hydrolysis of aspirin was retarded

by a number of micellar systems, such as cetylethyldlmethylammonium bromide and sodium laurylsulfate (35), higher alcohol ethers of polyoxyethylene and fatty acid esters of polyoxyethylene sorbitan (36), and the non-ionic surfactant cetomacrogol (37).

This depression was

accounted for by the partition of non-ionized aspirin into the micelles. Indomethacin hydrolysis rate was slowed by the use of polyoxyethylene sorbitan fatty acid esters in aqueous micellar systems.

At pH ■ 9,

70°, this hydrolysis rate was reduced twelve fold in 10% surfactant as compared to plain aqueous buffer (38).

STATEMENT OF THE PROBLEM

The aim of this research was to compare the degradation of three compounds in the presence of liposomes with that in the aqueous buffer in order to determine whether effects similar to those exhibited by micelles occur in liposomal suspensions.

The research protocol was

designed to characterize the total degradation rates both inside the liposomes and in the bulk aqueous environment.

In addition, any

association between the drugs and liposomes was to be assessed by the separation of the aqueous bulk phase from the liposomes. Three candidates were chosen for this study. Cyclocytidlne(Scheme VII), a pro-drug of the anti-leukemic agent arabinosylcytosine is a highly water soluble cationic compound.

At the pH of the study,

8.31, it exhibits first-order hydrolysis to produce arabinosylcytosine exclusively (Scheme VII, 39).

Indomethacin (Scheme VIII) an anti­

inflammatory agent, is an amide, with a side-chain bearing a carboxylic acid group. water.

The neutral form of indomethacin is highly insoluble in

However at the reaction pH of 9.4, the carboxylic acid group

(pK ■ 4.5) is in the anionic form thus making the molecule sufficiently a soluble.

Alkaline hydrolysis of indomethacin gives 5-methoxy-2-

methylindole-3-acetic acid and p-chlorobenzoic acid (Scheme VIII, 40). In hydroxide solutions and in buffered solutions t(pH ■ 7 - 10) indomethacin loss is first-order (41, 42).

71

Cyclocytidine

Arabinosylcytosine

SCHEME VII

o

60

° 4ly * IV

3 ‘3 >

(Eq. 3.4)

(eix*2y_eiy*2x)*

4 2x “ Cly|/(e2x eiy-elxc2 y^’ 4 2y " ^lx4 ('f:2x Fly-elx£2 y^ ’

e*^ and e ^ are the apparent molar abBorptivities of the reactants, and e£x and respectively.

are those of the products, at wavelengths x and y The various values of x, y, and the molar absorptivities

are listed In Table 3.1.

80 For the analysis of filtrates, 2 ml aliquots of reactions in the presence of liposomal suspensions were gently filtered using disposable 8 cartridges

and a syringe.

In the case of Indomethacin and p-nitro­

phenylacetate, 0.5 ml of the filtrate was mixed with 2 ml of cold 0.5 M formic acid in water:methanol (2:8).

The procedure for cyclo­

cytidine was identical except that 0.2 M HC1 in water:methanol (1:9) was employed Instead of 0.5 M formic acid.

Absorbances were measured

in micro-cuvettes and the concentrations were determined by replacing e' values with e values (Table 3.1 in Eqs. 3.1, 3*2, and 3.4 with D ■ 5). Total particle counts

9

which were 4 x 10

6

to 8 x 10

6

per ml in the

liposomal suspensions were reduced to background (< 2000 per ml) in the filtrate by this procedure.

Kinetics of Hydrolysis in Buffers and in Buffered Liposomal Suspensions Rates of hydrolysis of cyclocytidine, indomethacin and p-nitrophenyl­ acetate were studied at 40° in the aqueous buffers described in Tables 3.2 - 3.4, respectively.

Aliquots were removed as a function of

time and analyzed for concentrations as described for filtrates. Liposomal suspensions (Tables 3.2 - 3.4) were prepared as described above and heated at 40

o

for 12 hours with constant agitation

10

.

An

aliquot of a stock solution of the reactant was then added to 12 ml of the liposomal suspension, which was maintained at 40° with continued agitation.

At appropriate intervals aliquots of the mixture were

8 9Millex-GS 0.22 ym Filter unit, Millipore Corp., Bedford, MA.

lQElzone Model 80 XY, Particle Data Inc., Elmhurst, IL. Submersion Rotator, Scientific Industries, Inc., Queens Village, NY.

81 removed, the reaction head-space flushed with nitrogen and the samples were analyzed for concentration In the total liposomal suspension and In the filtrate.

Before, and after the reaction, 6 yl aliquots of the

liposomal suspension were diluted with 10 ml of the same buffer In which they were prepared.

One ml of this dilution was further diluted

with 10 ml of buffer and the particle size of the liposomes In this 9

suspension was determined . Hydrolysis rates In filtrates In the absence of liposomes were shown to be the same as those In their corresponding buffered solutions In the following way.

Sixteen ml of positively charged liposomal

suspension was prepared using carbonate buffer (Table 3.3) and heated at 40° for 12 hours with constant agitation^.

The suspension was

centrifuged at 2600 rpm for 50 minutes, and the bulk of the liposomes which concentrated at the top of the mixture were removed.

Five ml

of the remaining suspension was filtered through several disposable 8

filter cartridges .

The filtrate was used to study the hydrolysis

of Indomethacin as described previously for aqueous buffers at 40°. This procedure was also employed to study the hydrolysis of p-nitrophenylacetate In filtrate, using positively charged liposomes prepared with phosphate buffer (Table 3.4).

No difference was observed between

the rate constants In buffer and In the corresponding filtrate.

These

two reactions were chosen because they showed the greatest change In rate constants In the presence of liposomes.

82

To estimate the volume occupied by the liposomes in a suspension, 1 0 ml of a negatively charged suspension prepared in carbonate buffer

(Table 3.3) was heated at 40° for 12 hours with constant agitation. The suspension was then heated for an additional 24 hours with no agitation.

The liposomes which concentrated at the top were removed,

and the volume of the remaining mixture was determined to be 9.2 ml.

RESULTS

Apparent First-Order Hydrolysis - Except for the loss of p-nitrophenylacetate In the presence of positive liposomes (discussed later), data from all other kinetic studies (Tables 3.2 - 3.4) were adequately described by the first-order equation,

In Ct - In CQ - kQbst

(Eq. 3.5)

where t Is time, CQ and Ct are the reactant concentrations Initially and at time t, and kQbs is the observed first-order rate constant. All rate constants were calculated using linear least squares regression analysis.

Cyclocytidine - Figure 3.1 shows typical first-order plots for decrease in cyclocytidine in each of the following: borate buffer, total liposomal suspension, and liposomal suspension filtrates.

In all

cases the sum of the cyclocytidine and arabinosylcytosine concentra­ tions as a function of time equalled the initial concentration of cyclocytidine (Table 3.5).

Comparison of the rate of hydrolysis of

cyclocytidine in the presence of liposomes to that in the absence of liposomes can be made from the results in Table 3.2.

If the ratio

of the rate constant in the presence of liposome (kL ) to that in buffered solution (k_) is defined as

84

(Eq. 3.6)

\ m h /kt it Is apparent that

“ 1.

The Rfc values are 0.992, 0.912 and 0.975

for liposomes of positive, negative and neutral charge respectively.) The filtrate concentration (C^) can be compared to the correspond­ ing concentration In the total liposomal suspension (C^,) using the ratio, Rg where

(Eq. 3.7)

The half-life concentrations were chosen as a representative estimate of R_ which changes slightly as a function of time since plots are c not exactly parallel (Fig. 3.IB). Table 3.2.

The values of Rg are listed In

Although the differences between Cj, and

are small,

the filtrate concentrations were consistently higher than the liposomal concentrations since Rg values are greater than one.

This may be due to

the decrease In volume by removal of liposomes which will be discussed later.

Indomethacin - The reduced rate of Indomethacin hydrolysis in various liposomal suspensions as compared to that In liposome-free buffer is shown in Fig. 3.2.

This reduction can be quantified using R^ values

(Eq. 3.6) which are 0.156, 0.318, and 0.226 In positively charged, negatively charged, and neutral liposomal suspensions respectively, as calculated from the data in Table 3.3.

85

In negatively charged liposomes,

- 0.28 (Eq. 3.7), and the

first-order rate constant obtained from the filtrate data was k

F

0.106 hr~^ (Fig. 3.3) which agrees with its corresponding k

L

value.

In filtrates from the neutral and positively charged liposomal suspens­ ions, the initial concentration of indomethacin was extremely low relative to assay sensitivity.

The decreasing concentrations as a

function of time caused sufficient variability to impair the accurate estimation of first-order rate constants. R

C

However average values for

could be estimated from the ratios of measured C„ to calculated C r

values instead of employing Eq. 3.7.

T

Since each ratio is comprised of

a small numerator (C ) and a large denominator (C_) the variability J?

in R

c

T

is low (Fig. 3.4, Table 3.3).

p-Nitrophenylacetate - The loss of p-nitrophenylacetate from negatively charged and neutral liposomal suspensions followed a first-order exponential pattern with reduction in rate constants as compared to aqueous buffer.

However loss from positively charged liposomes was

described by a biexponential function and was more rapid than in aqueous buffer.

The treatment of these two types of data are

described below. In all the reactions the sum of p-nitrophenylacetate and pnitrophenol concentrations as a function of time equalled the initial concentration of p-nitrophenylacetate, and representative data sets are shown in Table 3.6.

Typical first-order plots illustrating the

rate of loss of p-nitrophenylacetate in phosphate buffer, and the

86 decreased rate in neutral and negatively charged liposomal suspensions are shown in Fig. 3.5.

This reduction is reflected in the values of

R^ (Eq. 3.6), which are 0.740 and 0.688 in neutral and negatively charged liposomal suspensions respectively, as calculated from the data in Table 3.4.

The values of Rg (Eq. 3.7) are 0.144 and 0.086 for

neutral and negatively charged liposomal suspensions respectively, and the first-order rate constants in these filtrates (k^) are close to their respective

values (Table 3.4).

Semilogarithmic plots of the concentration of p-nitrophenylacetate in positively charged liposomal suspensions as a function of time showed a rapid initial decrease followed by a much slower second phase. After normalizing the data relative to the initial concentration the biphasic curves were analyzed by the following equation:

F = Ae“0t + Be_et

where F = Ct/C0 , and o > 8 (Fig. 3.6).

(Eq. 3.8)

Non-linear regression analyses

provided average values of A ■ 0.406, B * 0.595, a “ 0.233 min \ 8 = 0.0341 min \

and

Since it is not possible to determine the value of

R_ according to Eq. 3.7 due to the lack of a first-order hydrolysis half-life, an average value was determined in the following manner. Experimental

values and calculated

values determined during the

beta phase were used to calculate an average Rg value, of 0.508.

DISCUSSION

Cyclocytidine, indomethacin, or. p-nitrophenylacetate, in liposomal suspensions can hydrolyze in the bulk aqueous solution or in the liposomal phase.

The association of the reactant with the liposomal

phase could occur in a number of possible ways.

Those interactions which

bring the compounds into the proximity of an organic environment, having a lesser ion-solvating power than water (Fig. 3.7) can cause a change in the hydrolysis rate in accordance with expected solvent effects. Theory states that media with low ion-solvating power will inhibit the creation or concentration of charges while accelerating charge destruct­ ion (45).

The change in rate would therefore depend on the reaction

mechanism and the species involved. A priori, it was considered possible that rate constants in the bulk aqueous media of liposome suspensions could differ from those independently measured in aqueous buffers (kg).

To eliminate this

possibility the rate constants for the hydrolysis of p-nitrophenyl­ acetate and indomethacin initiated in filtrates from positively charged liposomal suspensions were shown to be Identical to the corresponding kg values.

Since these two reactions showed the greatest

change in the presence of liposomes, it can be concluded that the values for kg also apply to hydrolysis in the aqueous bulk of the suspensions. 87

88 The hydrolysis of cyclocytidine was not affected by the presence of liposomes as evidenced by the similarity between the values of the first-order hydrolysis rate constants determined in aqueous buffer (kg), in liposomal suspensions (k^), and using data representing the filtrates sampled from the reactions in suspensions (kp, Table 3.2). The lack of association between cyclocytidine and liposomes is further evidenced by the fact that the total amount of the drug in the liposomal suspensions was accounted for by assay of the filtrates.

The slightly

higher concentrations observed in the filtrates could be caused by the reduction in volume resulting from removal of the liposomes.

Cyclo­

cytidine appears to remain in the bulk aqueous phase of liposomal suspensions (Fig. 3.7) where it hydrolyzes to arabinosylcytosine. Cyclocytidine is a highly water soluble, positively-charged compound with two hydroxyl groups.

Consequently, the lack of associa­

tion between cyclocytidine and the liposomes can be attributed to high hydrophilicity.

However, reaction velocity could be effected

without evidence for association based on filtrate concentrations.

In

the case of positively-charged and negatively-charged liposomes, these charged surfaces could affect reaction intermediates and/or transition states; positively-charged liposomes would be expected to stabilize negatively-charged species and vice-versa.

In the case

of cyclocytidine, wherein hydroxyl anion attack on the cation would result in a neutral transition state, low sensitivity to the liposomal surface charge would be expected.

Consequently liposome surface

interactions would not affect the rate of hydrolysis.

89

In contrast to cyclocytidine, the hydrolysis of indomethacin in the presence of positively-charged, negatively-charged and neutral liposomes, and that of p-nitrophenylacetate in negatively-charged and neutral liposomal suspensions all display first-order kinetics with a rate constant (k^) which is slower than that in aqueous buffer (kg). Since the loss of these reactants from liposomal suspensions can be experimentally described by the first-order equation,

-d(DT) dt

(Eq. 3.9)

kj/DT)

where (DT) is the total amount of drug in the liposomal suspension at time t, and k^ is the apparent first-order rate constant for loss of drug, the simplest kinetic scheme that can represent this data is,

Fast ..

DF

»

DL

B

Scheme X

where (DF) is the amount of drug in the bulk aqueous phase, (DL) is the amount of drug associated with the liposomes and, kg and k^ are the rate constants for the loss of drug from the bulk aqueous and liposomal phases respectively.

It is obvious that

90

k (DT) - k (DF) + k. (DL) L B * *

(Eq. 3.10)

and therefore the rate constant k_ can be described as

k, ■ k_f_ + k.f_ Ti B B A L

where f

B

” (DF) / (DT) , and f

L

*■= (DL) / (DT) .

(Eq. 3.11)

A similar expression has

been applied to the hydrolysis of a number of compounds in the presence of micelles (46). Since the volume occupied by the liposomes in these suspensions was found to be less than 10%, the ratio Rc ■ [Cf]/[C^J (defined by Eq. 3.7) is approximately equal to fg = (DF)/(DT). substituting

Setting fL ■ (1-fg) and

for fQ in Eq. 11 results in

(Eq. 3.12)

An apparent equilibrium constant,

can be defined for the

association of drug with the liposomes,

(Eq. 3.13)

Substituting for (DL) in Eq. 3.10 gives

V DT> ■ (kB + V W (BF)

(Eg. 3.14)

91

Integrating Eq. 3.9 with respect to time provides

(DT) - (DT) e”^ o

(Eq. 3.15)

where (DT)q is the initial amount of (DT).

Substituting in Eq. 3.14

gives

k ^ D T ) oe"kL t “ (ka + kAKapp)(DF)

(Eq. 3.16)

Simplification followed by a ln-transformation yields,

ln(DF) ■ In i - k^t

(Eq. 3.17)

where i - kL (DT)£)/(kB + kAKapp)

Therefore rate constants obtained by semi-logarithmic plots of the drug in the filtrates (kp) versus time were found to be close to those obtained by analysis of drug in total liposomal suspension (k^). The rate constants for the hydrolysis of indomethacin in positively charged, negatively charged, and neutral liposomes (k^» Table 3.3) are 15.6%, 31.8% and 22.6% of kg.

Indomethacin is

associated with the liposomes to the extent of 90%, 72%, and 80% in positively charged, negatively charged, and neutral liposomes respectively (R values, Table 3.3). c

The order of the extent of this

association could be a function of electrostatic attraction.

The

92

negatively-charged indomethacin molecules would be attracted to the positively-charged liposomes to the greatest extent followed by the neutral liposomes, and the negative portion of indomethacin would be repelled to some degree by the negatively-charged liposomes.

Indo­

methacin in the unionized form is highly insoluble in water (47).

At

the reaction pH, indomethacin may be considered to be made up of a hydrophilic carboxylate anion and the remaining hydrophobic moeity. Thus the indomethacin associated with the liposomes could be positioned at the interface of aqueous bulk and the hydrocarbon portion of the lipid bilayer as shown in Fig. 3.7.

A similar effect has been reported

for the hydrolysis of the sodium salt of mono-p-nitrophenyldodecanedioate in micellar aggregates (48).

It was proposed that the carboxylate

group remained in the aqueous portion, and the hydrophobic part of the molecule including the reactive ester linkage was adsorbed into the hydrocarbon center of the micelles.

By analogy, the carboxylate anion

of indomethacin could be in an aqueous environment, while the hydrophobic part containing the amide linkage could be in a hydrocarbon environment of reduced hydroxide ion concentration.

In addition,

the hydrolysis of indomethacin involves the attack of a hydroxide ion on the negatively-charged molecule, thus causing the collection of similar charges in the transition state.

This process will be

inhibited by the low ion-solvating power of the bilipld layer.

These

two effects would explain the depression in the rate of hydrolysis of indomethacin.

The values of k^, the rate constant for the hydrolysis

93 of indomethacin in the liposomal phases were calculated by substituting the values of kg, 1^, and R c (Table 3.3), in Eq. 3.12. values of R

c

The approximate

in the case of positively charged and neutral liposomes

provided only estimated values of k^. 0.018 hr 1 , and

The values of k^ were

-0.02 hr"" ,

-0. 0 1 hr " 1 in positively charged, negatively charged,

and neutral liposomes respectively.

As can be seen by comparing kgfg

values to those for k^f^, the major portion of the hydrolysis occurs in the aqueous portion of the liposomal suspension. The hydrolysis of p-nitrophenylacetate in negatively-charged and neutral liposomes is also reduced in comparison with the aqueous buffer although the depression is less than that of indomethacin. The first-order rate constants in the presence of negatively-charged and neutral liposomes are 6 8 .8 % and 74.0% respectively of that in the aqueous buffer.

itrophenylacetate is associated to the extent of

91.4% and 85.6% with the negatively charged and neutral liposomes respectively (Table 3.4).

The structure of p-nitrophenylacetate,

unlike that of indomethacin, does not have a formal charge creating a distinct ionic portion and a hydrophobic organic part.

Therefore,

when the p-nitrophenylacetate associates with the liposomes, it would not selectively shield the ester linkage away from*the aqueous environment.

It has been shown with work done in micellar phases,

that compounds having a slight polar nature (such as nitrobenzene) are solubilized at the surface of a micelle rather than in the hydrocarbon interior (49,50).

In addition, p-nitrophenylacetate has a good degree

of solubility in water (51).

Consequently, if associated with the

liposomes, it would be expected to be exposed to a great extent to the aqueous environment both in the interior of the liposomes and in the bulk aqueous solution.

This could account for the fact that -90% ob­

served association of p-nitrophenylacetate with the liposomes resulted in only a -30% reduction in rate constant as compared to the aqueous buffer.

Since the transition state of hydroxide attack on p-nitro-

phenylacetate would be negative, the reasons given for the depression in rate of hydrolysis of indomethacin associated with liposomes would apply in this instance as well.

Rate constants for hydrolysis in the

liposomal phases were calculated from Eq. 3.12 using the values of k^, and

in Table 3.4.

Resultant k^ values are 0.238 hr”'1' and

0.252 hr”^ in negatively-charged and neutral liposomes respectively. Contrary to indomethacin, k^fg

and k^f^ comparisons indicate that the

major portion of the hydrolysis occurs inside the liposomes. The decrease of p-nitrophenylacetate concentration in the presence of positively-charged liposomes is described by a biexponential equation in which the exponential coefficients,

a and 8 , are 38.6 and 5.65 times

faster respectively, than the first-order hydrolysis rate constant in aqueous buffer (k^).

This observed acceleration in hydrolysis rate

might be attributed to electrostatic stabilization of the negativelycharged transition state by the positive charges on the surface of the liposomes.

This mechanism has been suggested for the enchanced rate of

hydrolysis of p-nitrophenylhexanoate in solutions containing cationic micelles (52), wherein the rate of hydrolysis of p-nitrophenylacetate was also accelerated. One explanation for accelerated hydrolysis described by a biexpon­ ential equation might be surface catalysis in accordance with a kinetic scheme of the following type.

95

DF

k - - "a

1 '»

PS

t

-» (DL*)

kd k

A

cat

'

'

Scheme XI

This differs from Scheme X In that (DS) represents ester associated with the exterior surface (see

3.7) while (DL') represents

ester association by the remaining mechanisms.

Here, ^cat is the

rate constant for hydrolysis of p-nitrophenylacetate at the positivelycharged liposomal surface and kfl and k^ are the rate constants for the association and disassociation with the liposomal surface. Since the ratio a/S/kg is 39/6/1 and R c (beta phase) is approxi­ mately 0.5 it follows that k cat

>>

kfi.

Since it is expected that kg >

k^ (as observed in Table 3.3 and 3.4) the dominant hydrolysis pathway in Scheme XI would be via k , k . , and k . —— a a’ cat

If it is further assumed

that the efficiency of surface catalysis, as evidenced by the observed acceleration in rate, results in a negligible ester fraction in (DL') a simplified expression can be derived from the total ester assay data (DT).

(The derivation of the equations describing Scheme XI are

shown in the Appendix).

Under these assumptions the amount of

p-nitrophenylacetate (DT) as a function of time can be described by the following equation,

96

(DT)

(DT)

- ^ 7

r

_

L(0 ' V kd-kcat)e

_n .“i + e

where (DT)q is the initial amount of p-nitrophenylacetate.

J

The fraction

of the p-nitrophenylacetate remaining, F - (DT)/(D.T)0 , may be described by,

-Bt

+1W ± '

which is of the same form as Eq. 3.8 » where, and,

B =

(ka+k B, B > 1 > A.

(Eq. 3.20)

+ (A )

the limits of the pre-exponential terms in Eq. 3.20 are The time course for F would show an initial lag phase

followed by more rapid decay.

It could not take on the shape observed

in the current studies, as typified in Fig. 3.6. A second, and perhaps more reasonable model which would provide biexponential loss of the type shown in Fig. 3.6 is shown in Scheme XII.

Fast DF

:d l ’

DS kd k

r

k.

cat ' Scheme XII

97 This differs from Scheme XI in that (DF) and (DS) are in rapid dynamic equilibrium.

Degradation occurs in the bulk and at the surface as a

single pool outside of the liposome.

Association and disassociation

constants (kfl and k^) now refer to all remaining mechanisms (DL') for liposome-ester association other than those involving surface catalysis. As discussed above for Eq. 3.18, kcat >> kg >

kA again would be

expected so that the simplest representation would be to set (Dg) = (DF + DS) and kg ■ kfi + kcat allowing k, ’a DL k

Scheme XIII

where k^ is considered insignificant relative to kg.

This is analogous

to the previous cases discussed under Scheme X wherein liposomeassociated reactant was stabilized relative to that in liposome-free buffer.

Biexponential equations and those for calculating the individual

rate constants for models described by Scheme XIII have been published (53).

The fraction of p-nitrophenylacetate remaining, F - (DT)/(DT)q ,

is given by,

(Eq. 3.21)

which describes data of the shape shown in Fig. 3.6. conditions, kj * Aa

+ Bg

kfl ■ a+g-kj-k^ ■ 5.0 hr-'*-.

■ 6.9 hr“^, k^ ■ ag/kg

Under these

* 4.2 hr-^ and,

The phase ratio, (DL)/(DF), may be

calculated from k /(k„-g) « 1.7 (54) which can be used to estimate R r * A

Cp/Ci - 0.4.

a

U

Although Scheme XIII is admittedly an oversimplification,

the estimated rate constants are reasonable and the calculated Rg is consistent with the measured distribution during the beta phase, (Rg = 0.508, Table 3.4).

REFERENCES

1.

A.D. Bangham, Advances In Lipid .Research, Vol. 1, Academic Press, New York, 1963, Pg. 65-104.

2.

A. D. Bangham, M. M. Standish, J. C. Watkins, Diffusion of univalent ions across the lamellae of swollen phospholipids, J. Mol. Biol., 13 (1965) 238-252.

3.

D. Papahadjopoulos, N. Miller, Phospholipid model membranes I. Structural characteristics of hydrated liquid crystals, Biochim. Biophys. Acta, 135 (1967) 624-638.

4.

D. A. Haydon, J. Taylor, The stability and properties of bimolecular lipid leaflets in aqueous solutions, JvTheor. Biol., 4 (1963) 281-296.

5.

H. M. Patel, B. E. Ryman, Oral administration of insulin by encapsulation within liposomes, FEBS Letters, 62 (1976) 60-63.

6.

C. Huang, Studies on phosphatidylcholine vesicles. Formation and physical characteristics, Biochemistry, j} (1969) 344-352.

7.

D. Deamer, A. D. Bangham, Large volume liposomes by an ether evaporation method, Biochim. Biophys. Acta, 443 (1976) 629-634.

8.

F. Olson, C. A. Hunt, F. C. Szoka, W. J. Vail, D. Papahadjopoulos, Preparation of liposomes of defined size distribution by extrusion through polycarbonate membranes, Biochim. Biophys. Acta, 557 (1979) 9-23.

9.

Y. Barenholzt, S. Amselew, D. Lictenber, New Method for preparation of phospholipid vesicles (liposomes) - French press, FEBS Letters, 99 (1979) 210-214.

10.

M. H. Milsmann, H. G. Weder, R. A. Schewendener, The preparation of large single' bilayer liposomes by a fast and controlled dialysis, Biochim. Biophys. Acta, 512 (1978) 147-155.

11.

D. Papahadjopoulus, W. J. Vail, K. Jacobson, G. Poste, Cochleate lipid cylinders. Formation by fusion of unilamellar lipid vesicles, Biochim. Biophys. Acta, 394 (1975) 483-491.

99

100 12.

M. Roseman, B. J. Litman, T. B. Thompson, Transbilayer exchange of phosphatidylethanolamine for phosphatidylcholine and Nacetimldoylphosphatidylethanolamlne in single-walled bilayer vesicles, Biochemistry, 14 (1975) 4826-4830.

13.

S. Batzri, E. D. Korn, Interaction of phospholipid vesicles with cells. Endocytosis and fusion as alternate mechanisms for the uptake of lipid-soluble and water-soluble molecules, J. Cell Biology, 66 (1975) 621-634.

14.

P. H. Fishman, J. Moss, R. L. Richards, R. 0. Brady, C. R. Alving, Liposomes as model membranes for ligand-receptor inter­ actions: Studies with choleragen and glycolipids, Biochemistry, 18 (1979) 2562-2567.

15.

K. J. Hellingwerf, J. C. Arents, B. J. Scholte, H. V. Westerhoff, Bacteriorhodopsin in Liposomes. II. Experimental evidence in support of a theoretical model, Biochim. Biophys. Acta, 547 (1979) 561-582.

16.

R. D. Kornberg, H. M. McConnell, Inside-outside transistions of phospholipids in vesicle membranes, Biochemistry, 10 (1971)

1111-1120. 17.

C. M. Colley, S. M. Metcalfe, B. Turner, A. S. V. Burgen, J. C. Metcalfe, The binding of benzyl alcohol to erythrocyte membranes, Biochim. Biophys. Acta, 233 (1971) 720-729.

18.

T. W. Tillack, S. C. Kinsky, A freeze-etch. Study of the effects of filipin on liposomes and human erythrocyte membranes, Biochim. Biophys. Acta, 323 (1973) 43-54.

19.

Y. Gutman, D. Lichtenberg, Pv Boonyaviroj, J. Cohen, Increased catecholamine release from adrenal-medulla by liposomes loaded with sodium or calcium ions, Biochem. Pharmacology, 28 (1979) 1209-1211..

20.

J. de Gier, M. C. Blok, P. W. M. Dijck, C. Mombers, A. J. Verkley, E. C. M. van der Neut-Kok, L. L. M. Van Deenen, Relations between liposomes and biomembranes, Vol. 308, Annals of the New York Academy of Sciences, 1978, Page 85-100.

21.

K. H. Tragi, H. Kinast, A. Pohl, Oral-administration of insulin by means of liposomes in animal experiments, Wiener Klinis.che Wochenschrift, 91 (1979) 448-451.

22.

W. E. Magee, Potentiation of lnterferon-productlon and stimula­ tion of lymphocytes by polyribonucleotides entrapped in liposomes, Vol. 308, Annals of the New York Academy of Sciences, 1978, Page 308-324.

101

23.

M. D. Silva, B. L. Hazleman, D. F. P. Thomas, P. Wraight Liposomes In arthritis - New approach, Lancet, 1_ (1979) 13201322.

24.

F. J. Schlffman, I. Klein, Rapid Induction of amphoteridn-B sensitivity In L1210 leukemla-cells by liposomes containing ergosterol, Nature, 269 (1977) 65-66.

25.

I. J. Fidler, Therapy of spontaneous metastases by intravenous Injection of liposomes containing lymphokines, Science, 208 (1980) 1469-1471.

26.

M. B. Yatvln, B. A. Horwitz, W. Kreutz, M. Shinitzky, pH-sensitive liposomes - Possible clinical implications, Science, 210 (1980) 1253-1255.

27.

J. N. Weinstein, D. S. Zaharko, R. L. Magin, M. B. Yatvln, Liposomes and local hyperthermia - Selective delivery of metho­ trexate to heated tumors, Science, 204 (1979) 188-191.

28.

M. Mezei, V. Gulasekharam, Liposomes - A selective drug delivery system for the topical route of administration. 1. Lotion dosage form, Life Sciences, 26. (1980) 1471-1477.

29.

T. Yotsuyanagl, T. Hamada, H. Tomlda, K. Ikeda, Hydrolysis of procaine in liposomal suspension, Acta Pharm. Suec., 16 (1979) 271-280.

30.

T. Yotsuyanagl, T. Hamada, H. Tomlda, K. Ikeda, Hydrolysis of 2diethylaminoethyl p-nitrobenzoate in liposomal suspension, Acta Pharm. Suec., 16 (1979) 325-332.

31.

D. C. Poland, H. A. Scheraga, Hydrophobic bondine and stability, J. Phys. Chem., j69 (1965) 2431-2442.

32.

E. D.Goddard, C. A. J. Howe, G. C. Benson, Heats ofmicelle formation of paraffin chain salts in water, J. Phys. Chem., 61 (1957) 593-598.

34.

S. Riegelman, The effect of surfactant on drug Stability I., J. Am. Pharm. Assoc., Sci. Ed., 49. (1960) 339-343.

35.

H. Nogaml, S. Awazu, N. Nakajlma, Studies on decomposition and stabilization of drugs in solution. IX. Stabilization of acetylsalicylic acid in aqueous solution by surface-active agents, Chem. Pharm. Bull. (Tokyo), 10 (1962) 503-511.

micelle

102 36.

N. Nakajima, Studies on the stabilization of aspirin In solution with surface active agents, Yakugaku Zasshi, 81 (1961) 1684-1688.

37.

A. 6 . Mitchell, J. F. Broadhead, Hydrolysis of solubilized aspirin, J. Pharm. Sci., 56 (1967) 1261-1266.

38.

H. Krasowska, The hydrolysis of indomethacin in aqueous solutions of polysorbates, Int. J. Pharm., 4_ (1979) 89-97.

39.

M. Tuncel, R. Notari, L. Malspeis, A rapid stability indicating HPLC assay for the arabinosylcytosine prodrug, cyclocytidine, J. Liquid Chrom., 4 (1981) 887-896.

40.

H. Krasowska, L. Krowczynski, Z. Bogdanik, Assay of indomethacin in the presence of its hydrolytic degradation products, Chemical Abstracts, 80 (1974) 115885m.

41.

H. Krasowska, Kinetics of indomethacin hydrolysis, Chemical Abstracts, 8£ (1975) 116047t.

42.

B. R. Hajratwala, J. E. Dawson, Kinetics of indomethacin degrada­ tion I: Presence of alkali, J. Pharm. Sci., 66 (1977) 27-29.

43.

K. Fukuzawa, H. Chlda, A. Tokumura, H. Tsukatani, Antioxidative effect of a-tocopherol Incorporation into lecithin liposomes on ascorbic acid - Fe*+ -induced lipid peroxidation, Arch. Biochem. Biophys., 206 (1981) 173-180.

44.

C. A. Hunt, S. Isang, a-Tocopherol retards autoxidation and prolongs the shelf-life of liposomes, Int. J. Pharm., J3 (1981)

101-110. 45.

J. Hine, Physical Organic Chemistry, McGraw Hill Book Co., N.Y., 1956, Pg. 83.

46.

E. J. Fendler, J. H. Fendler, Advances in Physical Organic Chemistry, Vol. 8 , Academic Press, N.Y., 1970, Pg. 271-406.

47.

The Merck Index, 9th Ed., Merck and Co., Rahway, N.J., 1976, Pg. 656.

48.

F. M. Menger, C. E. Portnoy, On the chemistry of reactions proceeding inside molecular aggregates, J. Am. Chem. Soc., 89 (1967) 4698-4703.

49.

J. C. Eriksson, 6 . Gillberg, NMR studies of the solubilization of aromatic compounds in cetyltrimethylammonium bromide solution II, Acta Chem. Scand., 20 (1966) 2019-2027.

103 50.

R. B. Dunlap, E. H. Cordes, Secondary valence force catalysis. VI. Catalysis of hydrolysis of methylorthobenzoate by sodium dodecyl sulfate, J. Am. Chem. Soc., 90 (1968) 4395-4404.

51.

Handbook of Chemistry and Physics, 57th Ed., CRC Press, Cleveland, Ohio, 1976, Pg. C-87.

52.

L. R. Romsted, E. H. Cordes, Secondary valence force catalysis. VII. Catalysis of hydrolysis of p-nitrophenylhexanoate by micelleforming cationic detergents, J. Am. Chem. Soc., 90 (1968) 44044409.

53.

R. E. Notari, A. M. Burkinan, W. K. Van Tyle, Structural effects in drug distribution: Whole animal pharmacokinetics, J. Pharm. Pharmac., 26 (1974) 481-492.

54.

P. R. Byron, R. E. Notari, Critical analysis of "flip-flop" phenomenon in two-compartment pharmacokinetic model, J. Pharm. Sci., 65 (1976) 1140-1144.

104

TABLE 3.1

Apparent Molar Absorptivities of Reference Standards Recovered from Liposomal Suspensions

X(nm) x

y

330 272

260

Compound

na

--

Indomethacin

317

285

1 0 3e '

1 0 3e*

1

6.40

-

p-Nitrophenylacetate

1

9.08

1.22

p-Nitrophenol

2

3.20

Cyclocytidine

1

Arablnosylcytoslne

2

nx

10.3 4.70

ny

11.6

3.13 14.0

an-l designates reactant and n -2 Its hydrolysis product. ^Average values from e' nx

and e' ny

where % Recovery * lOOe'/e.

% Recovery1

-100

93 94 -100 -100

105 TABLE 3.2

PH 6

Apparent First-Order Rate Constants for Hydrolysis of Cydocytidinea In Buffers (kg) and In Liposomal Suspensions (k-,kp), at 40°, and the Mean Diameters of Liposomes Before (dQ ) and After (d) Hydrolysis

Liposomal Charge

k

kB 8.31°

7.51d

in

u

hr \

kL

kp

*C

do

d

Positive

0.865e

0.858e

0.786

1.12

2.38

2.30

Negative

0.865

0.789e

0.756

1.18

2.21

2.26

Neutral

0.865

0.843e

0.774

1.16

2.43

2.52

Negative

0.121

0.109*

-—

— --

--

--

aC() - 9 x 10-^ M. ^Radiometer pH meter (Model PHM62) standardized at pH 6.97 and 10.88 at 40°. cBorate buffer - 0,320 M H 3 BO 3 ; 0.140 M NaH 2B03 ; 0.350 M NaCl. dPhosphate buffer - 0.0140 M NaH 2P04 ; 0.100 M Na 2HP0^; 0.086 M NaCl. eAverage value from two to three kinetic experiments having variation of 5%.

*

106 TABLE 3.3

pHb

Apparent First-Order Rate Constants for Hydrolysis of Indomethacina in Buffers (kg) and in Liposomal Suspensions (kL ,kp), at 40°, and the Mean Diameters of Liposomes Before (dQ ) and After (d) the Reaction

1r

Liposomal Charge kB

9.37°

- v ■■■ ■ ■■■ kL

Positive

0.346d

0.0540d

Negative

0.346

0 .110 d

Neutral

0.346

0.0783d

kp

___e

*c

do

d

- 0 .1 *

2.10

2.26

0.106

0.28

2.17

2.17

___e

* 0 .2 e

2.38

2.52

aCQ “ 9.0 x 10 M. Radiometer pH meter (Model PHM62) standardized at pH 6.97 and 10.88 at 40°. cCarbonate buffer:0.135 M NaHCOg;0.065 M Na 2 C 03 0.075 M NaCl. ^Average value from three kinetic experiments, having variation of 7%. eFiltrates were not concentrated enough to obtain more accurate estimates.

107 TABLE 3.4

pHb

Apparent First-Order Rate Constants for the Hydrolysis of p-Nitrophenylacetatea In Buffers (kg) and In Liposomal Suspensions (1^, kF , a, 3), at 40°, and the Mean Diameters of Liposomes Before (dQ) and After (d) the Reaction

.k

Liposomal Charge

hr ” 1

Jt

Negative

0.362

0.249d

0.29

0.086

2.10

2.34

Neutral

0.362

0.268d

0.31

0.144

2.14

2.38

Positive

a 0.233

0.508e

2.14

2.21

J

*C

kB 7.51°

In

do

d

min ^ e A 0.0341

* 9.0 x 1 0 . Radiometer pH meter (Model PHM62) standardized at pH 6.97 and 10.88 at 40°. cPhosphate buffer ■ 0.0140 M Na^ P O ^ ; 0.100 M Na 2HP 0 ^ ; 0.086 M NaCl. ^Average value from two to three kinetic experiments, having variation of ~9%. eAverage value during beta phase.

108 TABLE 3.5

Reaction Concentrations- as a Function of Time of Cyclo­ cytidine (Cx ), Arabinosylcytosine (Cy ) and C the Sum of Cx and C „ , as a Percentage of the Initial Concentration of Cyclocytidine CQ , in (A) Borate Buffer, (B) Positivelycharged Liposomes3 , (C) Negatively-charged Liposomes 3 and, (D) Neutral Liposomes3 .

HOURS

C

C

0.00683 0.118 0.251 0.416 0.568 0.770 1.17 1.81 2.34

96.8 88.5 78.8

0.73 9.93 19.5 29.9 38.6 47.2 62.0 77.3 83.5

97.5 98.4 98.3 98.0 98.2 97.0 97.3 97.7 96.3

(B)

1.04

95.1 96.6 100.3 96.0 95.1 96.6 95.8 95.8 94.2

(D)

0.0153 0.168 0.370 0.600 0.900 1.43 1.89 6.52 12.40

x

68.1

59.6 49.8 35.3 20.4 12.8

94.1 85.0 75.7 59.5 46.6 32.0 21.7 2.52 1.09

y

11.6

24.6 36.5 48.5 64.6 74.1 93.3 93.1

C

HOURS

T

0.015 0.146 0.268 0.550 0.778 1.09 1.57 5.68 12.57

0.0156 0.182 0.354 0.539 0.820 1.19 1.83 8.47

C

x

95.8 85.5 75.8 60.2 47.8 37.1 20.9 0.56 0.76

97.4 8 6 .6

74.3 63.9 47.3 34.7 21.5 1.63

aLiposomes are prepared in pH ■ 8.31 borate buffer.

C

CT

1.14 19.5 37.5 47.6 59.0 69.7 94.8 94.8

96.9 96.6 95.3 97.7 95.4 96.1 90.6 95.4 95.6

0.699 13.0 24.6 35.0 47.6 61.2 75.5 89.8

98.1 99.6 98.9 98.9 94.9 95.9 97.0 91.4

y

11.1

i

TABLE 3.6

Reaction Concentration as a Function of Time of p-Nitrophenylacetate (C ), p-Nitrophenol (Cy ) , and C the Sum of Cx and C„, as a Percentage of the Initial Concentration CQ , of p-Nitrophenylacetate in (A) Phosphate Buffer, (B) Positively-charged Liposomes8 , (C) Negatively-charged Liposomes8 and, (D) Neutral liposomes8 .

HOURS

(A)

(C)

C

X

cy

0.0172 0.315 0.763 1.57 3.02 3.88 4.95 16.28 16.67

0.518 97.6 90.2 11.5 76.6 26.6 56.9 47.3 33.3 72.6 82.1 24.5 16.6 90.4 0.521 107.7 0.509 108.1

0.017 0.871 1.57 2.48 3.62 6.05 11.13 22.90 25.47

99.6 81.4 68.9 54.8 42.0 22.7 7.80 3.93 3.93

0.00

17.9 30.1 43.0 55.9 72.2 87.4 92.9 93.9

C'

98.1 101.7 103.2 104.2 105.9 106.6 107.0 108.2 108.6

(B)

99.6 99.3 99.0 97.8 97.9 94.9 95.2 96.8 97.8

(D)

HOURS

C

0.0107 0.0575 0.119 0.193 0.285 0.473 0.833 1.67 2.23

92.2 51.5 37.6 27.9 15.7 6.43 1.73 1.08

0.0158 0.505 1.09 1.96 2.93 5.14 8.03 11.93 16.67

100.9 89.1 75.6 59.6 46.0 25.6 13.2 5.91 5.10

l i posomes prepared in pH = 7.51 phosphate buffer

X

68.2

cy 9.66 32.4 48.0 60.0 68.8

80.1 88.4 93.3 93.8

0.00

11.7 23.8 38.3 51.2 69.3 82.3 88.9 92.5

C

101.9 100.7 99.5 97.6 96.7 95.8 94.8 95.0 94.9

100.9 100.8

99.4 97.9 97.2 94.9 95.5 94.8 97.6

CONCENTRATION

110

05

ITS

tG HOURS

Fig. 3.1

Time Course for Cydocytidine in, pH “ 8.31 Borate Buffer (•), and in Neutral Liposomal Suspension (IS); and in [b ] Positively Charged Liposomal Suspension ( ® ) and, in the Filtrate Sampled from the same . Least Squares Regression Lines Shown Represent # ,

(Eq* A. 10)