<%BANNER%>

N, N`- Dialkylaminoalkylcarbonyl (DAAC) Prodrugs and Aminoalkylcarbonyl (AAC) Prodrugs of Phenol Containing Drugs

Permanent Link: http://ufdc.ufl.edu/UFE0041918/00001

Material Information

Title: N, N`- Dialkylaminoalkylcarbonyl (DAAC) Prodrugs and Aminoalkylcarbonyl (AAC) Prodrugs of Phenol Containing Drugs
Physical Description: 1 online resource (116 p.)
Language: english
Creator: Ketha, Hemamalini
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: acetaminophen, naltrexone, prodrugs, topical
Medicinal Chemistry -- Dissertations, Academic -- UF
Genre: Pharmaceutical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Oral delivery of drugs is the preferred route of drug administration because of the ease in dosing regimen and high patient compliance. However the bioavailability of an orally administered drug can be low due to first pass drug metabolism. Topical delivery of drugs circumvents the challenges associated with first pass effects. Topical delivery of phenols is an attractive field of research particularly because of applications in area of pain management (morphine), hormone replacement therapy (estradiol) and in the treatment of alcohol addiction (naltrexone). Prodrug strategy involves chemical modification of a parent drug with poor delivery or physicochemical properties to a transient form with favorable physicochemical properties, which reverts back to the parent drug chemically or enzymatically. The chemical modification that results in an increase in the aqueous and lipid solubility without a considerable increase in the molecular weight of the corresponding prodrug has been shown to give highest enhancement in flux. In the present study N, N`- dialkylaminoalkylcarbonyl (DAAC) and aminoalkylcarbonyl (AAC) prodrugs of acetaminophen were synthesized and as well as one AAC prodrug of naltrexone (NTX). The hypothesis is that the incorporation of a basic amine functionality in to an acyl prodrug will result in a favorable increase in its aqueous and lipid (biphasic) solubility resulting in an enhancement in topical delivery of the parent drug. The DAAC and AAC prodrugs hydrolyzed to the parent drug with half lives between 15 - 115 minutes in pH 6.0 buffer. Solubilities of DAAC prodrugs in buffer (acetate, pH 4.0, 50 mM), 1-octanol, propylene glycol and isopropyl myristate (IPM) have been determined. The solubilities of the AAC-HCl prodrugs were determined in 1-octanol and propylene glycol. The DAAC and AAC prodrugs were evaluated for their ability to deliver APAP or NTX through hairless mouse skin. Although one of the prodrugs showed three times higher flux than APAP, the delivery of APAP was not as high as expected. We hypothesize that this may be due to an increase in the molecular weight caused by aggregation of the prodrugs and water association with the amine group as the prodrug diffuses the membrane.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Hemamalini Ketha.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Sloan, Kenneth B.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041918:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041918/00001

Material Information

Title: N, N`- Dialkylaminoalkylcarbonyl (DAAC) Prodrugs and Aminoalkylcarbonyl (AAC) Prodrugs of Phenol Containing Drugs
Physical Description: 1 online resource (116 p.)
Language: english
Creator: Ketha, Hemamalini
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: acetaminophen, naltrexone, prodrugs, topical
Medicinal Chemistry -- Dissertations, Academic -- UF
Genre: Pharmaceutical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Oral delivery of drugs is the preferred route of drug administration because of the ease in dosing regimen and high patient compliance. However the bioavailability of an orally administered drug can be low due to first pass drug metabolism. Topical delivery of drugs circumvents the challenges associated with first pass effects. Topical delivery of phenols is an attractive field of research particularly because of applications in area of pain management (morphine), hormone replacement therapy (estradiol) and in the treatment of alcohol addiction (naltrexone). Prodrug strategy involves chemical modification of a parent drug with poor delivery or physicochemical properties to a transient form with favorable physicochemical properties, which reverts back to the parent drug chemically or enzymatically. The chemical modification that results in an increase in the aqueous and lipid solubility without a considerable increase in the molecular weight of the corresponding prodrug has been shown to give highest enhancement in flux. In the present study N, N`- dialkylaminoalkylcarbonyl (DAAC) and aminoalkylcarbonyl (AAC) prodrugs of acetaminophen were synthesized and as well as one AAC prodrug of naltrexone (NTX). The hypothesis is that the incorporation of a basic amine functionality in to an acyl prodrug will result in a favorable increase in its aqueous and lipid (biphasic) solubility resulting in an enhancement in topical delivery of the parent drug. The DAAC and AAC prodrugs hydrolyzed to the parent drug with half lives between 15 - 115 minutes in pH 6.0 buffer. Solubilities of DAAC prodrugs in buffer (acetate, pH 4.0, 50 mM), 1-octanol, propylene glycol and isopropyl myristate (IPM) have been determined. The solubilities of the AAC-HCl prodrugs were determined in 1-octanol and propylene glycol. The DAAC and AAC prodrugs were evaluated for their ability to deliver APAP or NTX through hairless mouse skin. Although one of the prodrugs showed three times higher flux than APAP, the delivery of APAP was not as high as expected. We hypothesize that this may be due to an increase in the molecular weight caused by aggregation of the prodrugs and water association with the amine group as the prodrug diffuses the membrane.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Hemamalini Ketha.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Sloan, Kenneth B.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041918:00001


This item has the following downloads:


Full Text





N, N' DIALKYLAMINOALKYLCARBONYL (DAAC) AND AMINOALKYLCARBONYL
(AAC) PRODRUGS OF PHENOLIC DRUGS




















By

HEMAMALINI DEVARAJAN KETHA


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2010

































2010 Hemamalini Devarajan Ketha


























To Amma, my mother Kalyani









ACKNOWLEDGMENTS

As I look back into time, my desire to pursue research could not have been

accomplished without the strength and optimism that my mother, Kalyani, instilled into

me. For me she is the personification of strength against all odds. I am indebted to my

father, Devarajan, for his love and care. I am thankful to my sister Indira for her

unwavering support and for the time that she spent "just listening" to my ideas. My

success as a student and an individual could not have been possible without my family.

I would like to thank Dr. Kenneth Sloan for this opportunity and for the immense

patience that he showed when I made mistakes he taught me to "stop, slow down and

learn through each mistake." Of thing I am certain as a researcher and as an

individual, I will make mistakes- but I am also confident that Dr. Sloan's words the

faster you go the behinder you get- will help me make my every mistake a learning

experience.

I am thankful to my committee members Dr. Margret O. James, Dr. Raymond

Bergeron, Dr. Kenneth Wagener and Dr. Scott Wasdo for their encouragement and

constructive ideas through the process. Scott's presence and ideas at the weekly group

meetings made it a great learning experience.

And finally, one person that needs a special mention here is Siva, my husband.

His presence and love gives me the balance and strength that I need. I want to thank

him for his patience to listen to my day to day research and for sharing my excitement

for simple achievements like a good solubility experiment or a clean NMR spectrum. I

am very fortunate that I met Siva and that he is by my side in all my endeavors big or

small. His love makes my tough times easy and my good days meaningful.









TABLE OF CONTENTS


page

ACKNOW LEDG M ENTS ............... .................................. ......................................... 4

LIST OF TABLES .......... ..... ... ..................... ............. ...... ............... 7

LIS T O F F IG U R E S .................................................................. 8

ABSTRACT .............. .............................................. 10

CHAPTER

1 BACKGROUND ................................................ ....... ......... 12

The Rationale For Topical Delivery.......................................... .................... 12
Epiderm is ............................................................................................... ......... 17
Stratum Corneum ................ ........... ........... ........ ............... 19
Strategies for Enhancing Topical Delivery ....... ................... ........... ..... 22
Optimizing Topical Delivery ............................ .... .... .. ..... ............... 23
Mechanism of Percutaneous Absorption of Drugs ............................. 24
M odel Developm ent ................. .......... ...... ................ .... 27
Transformed Potts Guy Equation- The Robert Sloan Equation .................... 29
Prodrug Strategy to Enhance Percutaneous Absorption of Drugs ...................... 32
Acyl Prodrugs ........................ ......... ........... 35
Soft Alkyl Prodrugs ...... .................. .................. 41
C o n c lu s io n s .............. ..... ............ ................. ........................................... 4 3

2 S PEC IFIC O BJECTIV ES .............................. ........................ .............. 45

First Objective ............... ......... .................. 45
S second O objective .................................................... 46
Third Objective............................................ ........ 46

3 N, N' DIALKYLAMINOALKYL CARBONYL (DAAC) PRODRUGS OF A
MODEL PHENOLIC DRUG -ACETAMINOPHEN (APAP) ............................... 48

Intro d auction ...................................................................................... ... .......... 4 8
Synthesis of DAAC Prodrugs of Acetaminophen ...... ........ ..... ................. 50
Hydrolysis of DAA C -A PA P Prodrugs ............................ ................. ........... ....... 60
Determination of Solubilities of DAAC-APAP Prodrugs ............ ............... 64
In-Vitro Flux Determination of DAAC-APAP Prodrugs ..... ..... ............... 72
In-Vitro Evaluation of Morpholinyl and Piperidinyl DAAC-APAP
P ro d ru g s ............................. .... ..... .................... ................................. 7 3
In-Vitro Evaluation of Dimethyl and Diethyl DAAC-APAP Prodrugs and
DAAC-APA P HC I Prodrugs.................................. .......................... 80









C o n c lu s io n s ......... ......... .... .................................................. 8 3

4 AMINOALKYLCARBONYL (AAC) PRODRUGS OF PHENOLIC DRUGS
ACETAMINOPHEN AND NALTREXONE......................... ..... ............. 85

Introduction ................ ...... ...... ..... ..... ......... ........ .......... 85
Synthesis of AAC Prodrugs of Acetaminophen (APAP) And Naltrexone (NTX)...... 88
H ydrolysis of A A C prodrugs................................................................................ 92
Determination of Solubilities of AAC-APAP-HCI Prodrugs And VAL-NTX-TFA
P ro d ru g ................................. ....... ..... ........................................ 9 5
In-Vitro Evaluation of VAL-APAP-HCI and VAL-NTX-TFA Prodrugs ...................... 98
C conclusions .............. ......... .. ........... ................. ........... 102

5 CONCLUSION AND FUTURE WORK............................... 104

LIS T O F R E F E R E N C E S ............... .................................. .......................... 109

BIOGRAPHICAL SKETCH ..................................... .............. .................. 116

































6









LIST OF TABLES


Table page

1-1 General representation of a prodrug, an acyl prodrug and a soft alkyl
p ro d ru g .......................................................... ..... .... ..... ...... 3 4

1-2 Acyl prodrugs formed from hydroxyl or amine containing drugs...................... 35

3-1 Characterization of DAAC-APAP-HCI and DAAC-APAP Compounds ............... 58

3-2 Half Lives (t11/2, min) and predicted pKa Values of DAAC-APAP-HCI, prodrugs
in Buffer (pH 6.0, phosphate, 20 mM, I = 0.5) at 37 0.5 C ........... .............. 61

3-3 Molar absorptivities of DAAC-APAP-HCI prodrugs in methanol (E MeOH),
acetate buffer (E 4.0) and molar absorptivities of DAAC-APAP prodrugs in
acetonitrile (E ACN) .......................... ................................................... 65

3-4 Physicochemical Properties of DAAC-APAP-HCI Prodrugs .......................... 67

3-5 Physicochemical properties of DAAC-APAP podrugs.. ... ................... 69

3-6 Results from diffusion cell experiments with morpholinyl and piperidinyl
DAA C -A PA P prodrugs .......................................... ................ .............. 76

3-7 Results from diffusion cell experiments with dimethyl and diethyl DAAC-
APAP prodrugs and DAAC-APAP-HCI prodrugs .................... ............ 81

4-1 Elemental Analysis and Melting Points for AAC.HCI Prodrugs of 4-
Hydroxyacetanilide (APAP) And AAC Prodrug of Naltrexone......................... 91

4-2 Percentage yields of the Boc-AAC-APAP compounds and the corresponding
AAC prodrugs synthesized from Method A or Method B .............................. 92

4-3 Half lives (t11/2, min) and predicted pKa values of AAC prodrugs of APAP and
naltrexone in buffer (pH 6.0, phosphate, 20 mM, I = 0.5) at 37 0.5 C............. 93

4-4 Molar absorptivities of AAC-APAP-HCI prodrugs, VAL-NTX-TFA and NTX....... 95

4-5 Physicochemical properties of AAC Prodrugs .............. ..... ............... 96

4-6 Results from diffusion cell experiments with AAC prodrugs of APAP and NTX
(VAL-APAP-HCI and VAL-NTX-TFA). ............ ......... ........... ........... 99









LIST OF FIGURES


Figure page

1-1 Chemical Structures of Skin Ceramides................................... 20

1 -2 Pathway For Diffusion of a Topically Applied Permeant Based On The
Biphasic Solubility Model ... .... .................................... ................. 26

1-3 W asdo P lot. ......................................................................................... 30

1-4 Mechanism of hydrolysis of fosphenytoin .......... ......................... ............. .. 32

1-5 Cyclosporine Prodrugs. .................. ......... ............. ............................. 33

1-6 Testosterone and Naltrexone Prodrugs ......................................................... ... 36

1-7 Chemical structures and IPM and aqueous solubilites of (A) APAP; (B) C2-
AOC-APAP; (C) C2-ACOM-APAP; (D) C2-AOCOM-APAP; (E) C2-
NANAOCAM-APAP ........ .......... .......................... ............... 38

1-8 Mechanism of of Hydrolysis of Acyl Prodrugs.. ............... ................................. 39

1-9 Chemical structures of Vitamin E and its prodrugs evaluated previously. .......... 40

1-10 Mechanism of hydrolysis of a soft alkyl prodrug. ........ ........ ... .... ............. 41

1-11 M annich B ase P rodrugs of 5 FU .................................................. .... .. ............... 42

3-1 Chemical structures of synthesized DAAC-APAP prodrugs ............................. 50

3-2 Synthesis of DAAC-APAP prodrugs ...... .................. ............. 52

3-3 Possible routes of hydrolysis of a DAAC-APAP prodrug .............................. 63

3-4 Origin of steric hindrance in 4a and 4c based on Newman Rule of Six ........... 64

3-5 Plot of IPM solubilities vs melting points for DAAC-APAP prodrugs................... 71

3-5 Plot of cumulative amount vs time for calculating steady state flux.................... 74

3-6 Plot of pKa vs J for morpholinyl and piperidinyl DAAC-APAP prodrugs. ............ 75

3-7 Plot of EXP log JM VS CALC log JM................................. ............ ................ .. 79

3-8 Cumulative amount of drug permeated vs time for MORnl-APAP, MORnl-
APAP-HCI and Me2ni-APAP and Me2ni-APAP-HCI ............................. 80









3-9 Plot of EXP log JM VS CALC log JM including dimethyl and diethyl DAAC-
APAP prodrugs ........................................ .......... 82

4-1 Chemical structures of commercially available drug Valacyclovir and valine
ester prodrug of 5-OH-DPAT by Bouwstra et. al.............................................. 85

4-2 Chemical Structures of AAC Prodrugs of APAP and NTX............................... 88

4-3 Synthesis of AAC prodrugs of APAP and NTX......................... ............... 89

4-4 Proposed mechanism of general base catalysis by the proline ring. 94

4-5 Cumulative amount of NTX permeated vs time for compounds 10 and NTX-
HCI ....... .......................................... .......... 101

5-1 Chemical structures of proposed DAAC prodrugs of phenol containing drugs. 106









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

N, N' DIALKYLAMINOALKYL CARBONYL (DAAC) AND AMINOALKYLCARBONYL
(AAC) PRODRUGS OF PHENOLIC DRUGS

By

Hemamalini Devarajan Ketha

August 2010

Chair: Kenneth B. Sloan
Major: Pharmaceutical Sciences Medicinal Chemistry

Oral delivery of drugs is the preferred route of drug administration because of the

ease in dosing regimen and high patient compliance. However the bioavailability of an

orally administered drug can be low due to first pass drug metabolism. Topical delivery

of drugs circumvents the challenges associated with first pass effects. Topical delivery

of phenols is an attractive field of research particularly because of applications in area

of pain management (morphine), hormone replacement therapy (estradiol) and in the

treatment of alcohol addiction naltrexonee). Prodrug strategy involves chemical

modification of a parent drug with poor delivery or physicochemical properties to a

transient form with favorable physicochemical properties, which reverts back to the

parent drug chemically or enzymatically. The chemical modification that results in an

increase in the aqueous and lipid solubility without a considerable increase in the

molecular weight of the corresponding prodrug has been shown to give highest

enhancement in flux.

In the present study N, N'- dialkylaminoalkylcarbonyl (DAAC) and

aminoalkylcarbonyl (AAC) prodrugs of acetaminophen were synthesized and as well as

one AAC prodrug of naltrexone (NTX). The hypothesis is that the incorporation of a









basic amine functionality in to an acyl prodrug will result in a favorable increase in its

aqueous and lipid (biphasic) solubility resulting in an enhancement in topical delivery of

the parent drug. The DAAC and AAC prodrugs hydrolyzed to the parent drug with half

lives between 15 115 minutes in pH 6.0 buffer. Solubilities of DAAC prodrugs in buffer

(acetate, pH 4.0, 50 mM), 1-octanol, propylene glycol and isopropyl myristate (IPM)

have been determined. The solubilities of the AAC-HCI prodrugs were determined in 1-

octanol and propylene glycol. The DAAC and AAC prodrugs were evaluated for their

ability to deliver APAP or NTX through hairless mouse skin. Although one of the

prodrugs showed three times higher flux than APAP, the delivery of APAP was not as

high as expected. We hypothesize that this may be due to an increase in the molecular

weight caused by aggregation of the prodrugs and water association with the amine

group as the prodrug diffuses the membrane.









CHAPTER 1
BACKGROUND

The Rationale For Topical Delivery

Oral delivery of drugs is preferred over other routes of drug administration

because of high patient compliance towards dosing regimens. However, before an

orally administered drug can reach systemic circulation, it must survive the acidic

environment of the stomach, where it might undergo deactivation or, while passing

through the upper and lower gastrointestinal (GI) tract, is subject to conjugation

reactions catalyzed by various biotransformation enzymes. After a drug is absorbed

from the gut wall, it enters the liver, which is the major site for xenobiotic

biotransformation reactions. The GI tract and liver have the highest concentrations of

drug metabolizing enzymes.1 These enzymes are responsible for converting the active

drug into more hydrophilic forms that are readily eliminated resulting in sub-optimal oral

bioavailability of the drug. Additionally, efflux transporters also contribute towards lower

bioavailability of an orally administered drug form. For example, oral bioavailability of

estradiol (administered to women undergoing Hormone Replacement Therapy (HRT)),

is only 10% and is eliminated mainly as its glucuronide in the urine.2 Additionally, certain

orally administered drugs, especially those containing an unmasked carboxylic acid

group, have been shown to cause damage to gastric mucosa3 and hepatotoxicity.4

The problems associated with an orally administered drug can be circumvented by

administering the drug via an alternate route. Topical delivery is one such alternative.

The terms topical and transdermal delivery refer to the diffusion of a drug across the

skin intended for local or systemic absorption, respectively. The undesirable effects of

the GI tract and liver can also be overcome by applying a drug topically. Skin has a









much lower concentration of drug metabolizing enzymes than the GI tract or liver. 5-8

Cytochrome P450 activity is only about 1- 5% of that found in the hepatic environment

and UGT/UDPGA (glucuronidation) and PAPS/SULT (sulfation) activity is only about

10% of that observed in the liver.9 Since topical delivery of drugs has advantages like

higher bioavailability and reduced incidence of metabolic effects over orally delivered

drug forms, topical delivery of drugs has received considerable attention in the last few

decades. Most of the drugs currently formulated for delivery through skin are effective at

low required doses (up to a few milligrams) and are delivered over a period of few

days.9 The transdermal patch of scopolamine (an alkaloid and a muscarinic antagonist)

was the first to be approved by the US FDA as a three-day patch for the treatment of

motion sickness. Over the years, patches for a number of drugs like scopolamine,

nitroglycerine, clonidine, fentanyl, lidocaine, nicotine, estradiol and testosterone have

been introduced in the market.10

Testosterone is a member of androgen group of steroid hormones. It is the primary

male sex hormone and anabolic steroid (promotes synthesis of important cellular

proteins, especially in the muscle tissue) and plays an important role in maintaining

general male sexual health. Hypogonadism is a clinical condition related to decreased

activity of gonads (organ responsible for production of sex hormones including estradiol

and testosterone) which results in decreased testosterone levels in males and females.

The symptoms include low libido, reduced bone and muscle mass and anemia.11

Several (injectable, transdermal, oral and buccal) formulations are available for

testosterone replacement therapy. Oral formulations containing methyltestosterone and

fluoxymesterone as the active pharmaceutical ingredient, even though available, are









prescribed infrequently.11 The oral delivery of testosterone has been associated with

severe hepatotoxicity and development of benign and malignant neoplasms.12 However,

the transdermal patch that delivers between 5-10 mg of testosterone per day, has been

shown to maintain uniform serum levels 13,14 of testosterone without the toxicity

associated with the oral formulation. There are other examples (fentanyl, estradiol,

nitroglycerin) where delivery of a drug through skin results in fewer side effects and

higher bioavailability.10

On the other hand topical delivery of drugs is also not without challenges. The

number of therapeutically useful compounds that can be administered through skin is

limited by the size and physicochemical properties of the permeant.5' 10 Topical patches

for low molecular weight drugs like, nicotine (162 Da) and moderately high molecular

weight drugs like oxybutin (359 Da ) are available: oxybutin being the largest molecule

being formulated commercially. The skin acts as an external physical barrier, protecting

the inner body organs against water loss, physical damage and exposure to adverse

chemical and microbial environment. The skin is composed of a highly interconnected,

complex assembly of cells and tissues that can be divided into three different layers:

epidermis (50 100 pm), dermis (1 2 mm) and hypodermis (1 2 mm). The thickness

of skin can vary depending on age, sex and anatomical location.5 It is the thickest on

areas that are more susceptible to physical abrasion, like heels and palms of hands.

The outermost layer, stratum corneum (SC), although the thinnest of all the layers,

presents the most formidable barrier towards diffusion of permeants across the skin. An

efficient delivery of a clinically relevant amount of an active therapeutic agent across

transdermall delivery) or into the skin dermall delivery) demands knowledge about the









fundamental structural components that give rise to the barrier function associated with

the skin. The structure and composition of each of the layers from innermost to the

outermost layer is discussed briefly below.

Hypodermis. Hypodermis or the subcutaneous tissue lies right below the dermis

and is composed primarily of adipocytes. The main function of this layer is to store fat

and provide insulation from external shocks and abrasions. The thickness of

hypodermis can vary depending upon the anatomic location. In addition, hypodermis

contains a vast network of blood vessels that facilitate the absorption of topically applied

compounds. Fibroblasts and macrophages are also found in this layer.

Dermis. The layer underlying epidermis, termed as dermis is a highly filamentous

and fibrous connective tissue which imparts elasticity and tensile strength to the skin.

The dermis constitutes about 70% of the dry weight of skin. Fibroblasts, the most

abundant cells found in this layer are responsible for generating connective tissue

components like elastin and collagen.5 These components are synthesized using water

soluble precursors and are released into the intercellular space by fibroblasts. Then, the

elastin and collagen are assembled into thin fibrous structures called fibrils. The

lowermost layer of dermis (reticular dermis) is comprised of more closely assembled,

highly cross-linked fibrils which become less dense as one moves towards the outer

most layers of dermis (papillary dermis) to allow for proliferation of a vast network of

nerve endings. These connective fibers act as a structural scaffold onto which other

appendages found in this layer are anchored. The region surrounding the fibrous

network is primarily comprised of highly hydrophilic macromolecules called

proteoglycans. Proteoglycans are a group of fibroblast-derived compounds that contain









numerous unbranched polysaccharide chain covalently linked to a polypepetide

backbone. The polysaccharide chains are composed of several hundred

glycosaminoglycan disaccharides that have been extensively sulphated. The anionic

charge on these chains from the sulphate and carboxylate groups results in an overall

negative charge on the side chains and considerable repulsion which is strong enough

to stabilize the chains in completely extended or a stretched state. The negative charge

also allows for water molecules to be associated with these proteins. These proteins are

capable of retaining about 1000 times their weight in water. The high degree of

hydration of the dermis makes it resemble a hydrophilic gel. Some of the important

proteoglycans include chondroitin sulfate/dermantan sulfate, heparin/heparin sulfate etc.

Appendages like sweat glands, hair follicles and sebaceous glands that extend up

into the stratum corneum (SC), originate in the dermis. Hair follicles are sheath like

structures that are embedded in the dermis. Sweat glands and sebaceous glands

extend till the outer most layer of the skin and deposit their respective contents on to the

skin surface. Sometimes, these glands are found to be linked with the hair follicle and

secrete their contents into the follicle. Like other features, the density and presence of

each appendage varies with anatomical location. Since these appendages provide a

direct access for a drug molecule on the outermost layer to reach the dermis; they may

provide a pore-mediated pathway for topically applied drugs. However, only a small

proportion (approximately 0.1%) of the skin surface is covered with pores, allowing only

a very small fraction of topically applied drug molecules to permeate via this route.

The border between dermis and epidermis, also called the dermal-epidermal

junction, is an uneven surface formed by a network of interlocked papillae from each of









the layers. The wave-like structure of the dermal-epidermal junction provides a large

surface area for the dermal papillae to interact with the epidermal cells facilitating an

effective removal of xenobiotics that permeate through skin. The extensive vascular

network that supplies nutrients and oxygen to the epidermis is also responsible for

transporting topically applied drugs into systemic circulation thereby maintaining a

continuous "sink" condition.

Epidermis

The epidermis is composed of layers of successively keratinizing cells with each

layer at a different phase of the keratinization process. The lower most to uppermost

layer from the dermal-epidermal junction to the skin surface are termed stratum basale,

stratum spinosum, stratum granulosum and stratum corneum (SC). The layers directly

below the SC (i.e stratum granulosum, stratum spinosum and stratum basale) are

together termed as the viable epidermis. The viable epidermis contains about 70%

water by weight. SC is composed of flat polyhedral shaped cells called corneocytes or

horny cells. These are about 40 pm in diameter and 0.5 pm thick. Corneocytes are

mainly composed of insoluble, aggregates of keratin. These are "dead cells" resulting

from keratinization of the cells of the viable epidermis. Although the SC is the thinnest

layer, it is the most impermeable amongst all the layers of the skin. The unique

structural organization of the SC renders it a primary barrier towards topically applied

drugs.

Stratum Basale. Keratinocytes in the basal layer are attached to the basement

membrane i.e the layer above the dermal-epidermal junction. The keratinocyte in the

stratum basale divides into a daughter stem cell and a transit amplifynig cell. The

daughter stem cell remains attached to the basement membrane whereas the transit









amplifying cell undergoes differentiation to form the spinous layer. The transit amplifying

cells however lack features characteristic of spinous and granular cells, for example

high nucleus to cytoplasm ratio. As more keratinocytes undergo differentiation, more

transit amplifying cells progress to the higher layer of the skin and to the next phase of

the keratinization process. Other components of this layer include melanocytes (cells

responsible for melanin production), Langerhans cells (mediators of immune responses)

and Merkel cells (sensory reception).

Stratum Spinosum. During their "transit" through the spinous layer, the cells

begin to undergo changes in the cellular environment like keratinization of organelles

and other proteins present inside the cell. This process leads to morphological changes

that make the keratinocytes increasingly flattened. As the cell continues its transit

towards the upper layers the diameter and volume of the cell continue to increase. This

stage also marks the beginning of formation of desmosomal plaques between the cells

that are responsible for providing cohesive strength to the corneocyte layers.

Desmosomes are surface proteins responsible for holding the flattened keratinocytes

together. The keratinocytes begin synthesizing the proteins, Keratin 1 (K1) and Keratin

10 (K10), which are precursors required in the next phase of the keratinization process.

Another important biochemical process that occurs during this phase is the formation of

lamellar granules which later become a part of the primary components that provide the

SC its unique barrier properties.

Stratum Granulosum. The cellular components, especially the nucleus, of the

keratinocytes in this layer begin to undergo enzymatic degradation. This layer is rich in

speckled or granular components called Keratohyalin Granules (KGs) and Lamellar









Bodies (LBs). KGs are responsible for producing the precursors required to form the

envelope of the corneocytes in the SC. These precursors include K1, K10, profillaggrin,

loricin and involucrin. Out of these, loricin and involucrin are involved in formation of the

outer layer of the corneocyte whereas K1 and K10 are involved in the cross linking of

the inner layer of the cornified envelope. Lamellar Bodies are ellipsoid shaped

components found in the stratum granulosum cell, the major components of which

include phospholipids, cholesterol and acylglucosylceramides. LBs are highly

compressed structures in which the lipoidal components are packed to result in a

bilayered arrangement. In addition to the above components, LBs also contain enzymes

like phospholipase A2. At the junction between the stratum granulosum and SC the LBs

are extruded in to the intracorneocyte space.

Stratum Corneum

SC can be further divided into two layers termed stratum compactum and stratum

disjunctum. The upper layer, stratum disjunctum, is constantly undergoing

desquamation. The lower of the two layers, stratum compactum, has as much as double

the amount of water per unit mass associated with it (30 % compared to 15%). These

layers differ in the overall amount of amino acids and lipids. In addition, stratum

compactum has a higher density of corneodesmosomes and, as the name suggests, is

more compact and tightly held together. The extruded LB's are fused into bilamellar

sheets which form the inter-corneocyte lipids of the SC.

Cellular remnants of the viable epidermis are packed together in a cornified

envelope to form the cellular components of the SC- the corneocyte. This cornified

envelope is stabilized by peptide cross-linkages like (y-glutamyl) lysine isopeptide

linkages, bis(y-glutamyl) polyamine linkages and disulphide bonds.











0





HN' HNOOH
OOH
Ceramide 1 Ceramide 2


,- v-0



OH
OH
Ceramide 3 Ceramide 4
OH

OH H OH OH

OH

Ceramide 5 Ceramide 6




on NOH
OH OH

Cerarnlde 7 Ceramide 8







Ceramide 9


Figure 1-1. Chemical Structures of Skin Ceramides

There is considerable overlapping observed between the corneocytes. The water


content of the SC is only about 15% by weight. The presence of bound water in the SC


plays a crucial role in determining the mechanism of permeation of drug molecules


through the SC.


The major components of the intercellular lamellae include ceramides (50% by


weight), cholesterol (30% by weight) and free fatty acids (10% by weight). Figure 1-1


shows the structures of skin ceramides isolated so far. Ceramides are important


structural components of the intercorneocyte lipids and are composed of sphingosine









(2-amino-4-octadecene-1, 3-diol) acylated at the 2-amino group with a fatty acid.

Intercellular lipids that are chemically linked via the interaction of the w

hydroxyceramides to the cornified envelope (via surface protein involucrin) occupy the

compartments formed by the partially overlapping corneocytes. The w

hydroxyceramides are derived from acylglucoceramides, i.e. ceramides 1, 4 and 9, via a

de-esterification reaction. The components of the intercellular compartments are

present as bilamellar sheets.1518 Bilamellar arrangement refers to the stacked

organization of the flat "plate like" components in the space between two corneocytes.

Swartzendruber et al. have shown that each cellular sheet is formed by the fusion of two

bilayers in LBs.

The corneocytes are further interconnected by polar components called

corneodesmosomes rendering considerable cohesive strength to the SC layer. The

overlapping of the corneocytes imparts a "tortuous" path through the lipid components

of the SC. 5 The SC has a highly dense heterogeneous microenvironment composed of

covalently linked corneocytes and intercellular lipids which are present as fused

bilamellar sheets at different phases of development. This uniquely rigid compartmental

organization of the corneocytes and the intercellular lipoidal bilamellar sheets, gives rise

to the barrier function of the SC.

The formation of the SC which is semipermeable to topically applied drugs is

therefore, a result of a series of biochemical events that lead to the conversion of viable

(living) cells into a highly rigid, heterogeneous network of impermeable cornified

envelopes and permeable inter cellular lipids.18









Strategies for Enhancing Topical Delivery

Several techniques including permeation enhancers, iontophoresis, microneedles

and prodrug strategies19 have been utilized to enhance percutaneous absorption of

drugs.10 Permeation enhancers alter the barrier property of the skin rendering it more

permeable towards active components of the formulation. Some of the properties of an

ideal penetration enhancer include predictable, reproducible and unidirectional effect

while lacking pharmacological activity and toxicity related issues. Several classes of

chemicals like oleic acid (OA), dimethyl sulphoxide (DMSO), azones, pyrrolidones and

surfactants have been shown to possess a favorable penetration enhancement effect.20

However; there are hardly any penetration enhancers that have all the above mentioned

"ideal" properties. Well-known permeation enhancers like OA, DMSO have been shown

to cause irreversible damage to the skin.21 Even water has been shown to affect the

barrier properties of skin, although the mechanism of this interaction are still under

debate.20, 22 Some proposed mechanisms by which penetration enhancers work include,

transient fluidization of the crystalline structure of stratum corneum, dissolution of

stratum corneum lipid, improved solubility of the permeant in the skin and improved

partitioning of the permeant into the skin. lontophoresis involves movement of charged

and uncharged molecules across the skin under the influence of a few milliamperes of

current applied across a small area of skin.23 The applied electric current can induce an

electric field in the skin micro-environment, which drives charged molecules across the

skin constituting as electrophoretic flow.23 Application of a small voltage of current can

also be used to generate an electro-osmotic flow, which can aid passage of charged

molecules across the skin. lontophoresis is used to deliver compounds like

priolcarpine24 (as a diagnostic test for cystic fibrosis) and lidocaine25 (local anesthetic)









into skin. The microneedle strategy for transdermal delivery involves utilization of an

array of micron-sized needles that penetrate the skin generating holes that can facilitate

entry of small molecules drugs as well as macromolecules into skin.26 This strategy has

been successfully utilized to deliver, proteins27 and DNA28 into skin.

Optimizing Topical Delivery

The observation that the SC is the primary barrier towards diffusion of permeants

across skin was made as early as 1924.29 Further experimental evidence came from a

study published by I H Blank, in which he used tape stripped human skin to

demonstrate that the water loss for the skin sample lacking stratum corneum was much

higher compared to normal skin.30 Additional work by R J Schuplein and I H Blank and

others exemplified that diffusion through skin is a passive process. 30, 31 Since a

molecule has to "dissolve" in the membrane microenvironment for the permeant to

cross, it is reasonable to state that the extent of compatibility of the physicochemical

properties of the membrane with that of the permeant dictates the permeants' diffusion

across the membrane. Since the SC has been shown to be composed of keratinized

cells embedded in intercellular lipid bilayers through which the permeant must pass,

what are the most significant physicochemical properties of the permeant that affect its

diffusion across the skin? Much effort has been made to identify these physicochemical

properties. In one such attempt, diffusion of a homologous series of straight chain

alcohols through human skin in-vitro was investigated by R J Schuplein and I H Blank.31

The choice of a homologous series was guided by the perceived usefulness of a

continuum of solubility properties and molecular weights in the series. The flux (J, in

units of pmole cm-2 h-) of alcohols when applied as neat liquids (C1-C8) as saturated

solutions in water (C6-C10) was found to be inversely proportional to the molecular









weight, with the most water soluble and the smaller member of the series exhibiting the

highest J. However the authors correlated permeability coefficient, P (defined as J/

CVEH, concentration of the permeant in the applied vehicle) in units of cm h- with the

number of carbons in the alcohol, and the partition coefficient, KLIPID: AQ. They suggested

that the most important predictor of diffusion of permeants through the skin was the

partition coefficient. Thus, P was used as an important parameter for modeling diffusion

of permeants through skin and mathematical models for predicting percutaneous

absorption, based on permeability coefficients have been developed. However, Sloan

and coworkers have shown that the "amount" permeated (J, pmole cm-2 h-) and not P

(cm h1) of the permeant is the only clinically relevant parameter for development of

mathematical models to optimize delivery of a permeant through skin. 32,33 Careful

analysis of the results from the experiments by Schuplein and Blank also demonstrated

that the solubility properties of the permeant and not the partition coefficient play the

dominant role in governing the flux and the passive diffusion of a permeant through

skin.

Mechanism of Percutaneous Absorption of Drugs

The fact that a drug molecule has to "dissolve" in the skin microenvironment for it

to surpass the SC barrier renders its permeation a "solubility-driven" process. In other

words the ability of the permeant to dissolve in the SC intercorneocyte space will decide

the maximum achievable flux, when a saturated donor phase is used and sink

conditions are maintained. Therefore, the environment of intercorneocyte lipids is crucial

in determining the mechanism of percutaneous absorption of drug molecules.

It is generally accepted that the barrier to absorption of permeants through skin is

primarily lipoidal in nature. Therefore enhanced lipid solubility of the permeant will result









in better flux. The aqueous barrier to percutaneous absorption is believed by many to

reside in the stagnant water layer existing on the SC surface for compounds delivered

from an aqueous vehicle. However, in the experiments carried out by Schuplein and

Blank it is clear, although not generally acknowledged, that the more water soluble

members of the series, applied as pure liquids (hence no stagnant water layer) gave the

best flux. Additionally, a dependence of J on water solubility of any member in a

homologous series of more lipid soluble prodrugs has been quantified and discussed in

detail in the next section.

Although the dependence of J on aqueous and lipid solubility of a permeant is

obvious, the exact location of the "aqueous layer" has been a topic of investigation by

many researchers. More recently, Sloan et. al. proposed a model for the organization of

ceramides along with bound and unbound water in the SC. The bound water has been

shown to be present in the corneocytes linked to the proteins and the unbound water is

distributed uniformly throughout the SC. The bilayers of the intercorneocyte components

are structured horizontally resulting in a parallel organization with the corneocyte

surface. The arrangement of the ceramides in the intercorneocyte space and the

suggested path followed by a topically applied permeant to reach the lower layers of the

skin has been shown in Figure 1-2. The presence of CER 6, with extended linoleic acid

side chain bridged between a broad distinct crystalline layer and a more fluid-like

narrow layer in the middle gives rise to a long periodicity phase (LPP) 13.4 nm in length.

The presence of the LPP has been shown to be essential for the barrier properties of

the intact SC. The unbound water layer (1.44 nm) in the SC is proposed to be present

close to the polar head groups of PA and Ch and CER6. The concentration of the water








in the SC has been calculated to be about 12.22 M (equivalent to 6.7 water molecules

per ceramide).

Corneocyte 1. CER 6 (AP) "'
















poa / /V. CER2M(NS) Ad
7 7 ICER I 'cEO ^i h







polar f- i ( VIl. Myristic Acid vvvv-- O
1.44 nm




Figure 1 -2. Pathway For Diffusion of a Topically Applied Permeant Based On The
Biphasic Solubility Model.

A topically applied permeant is expected to follow the path of least resistance

which is through the ceramides and other lipoidal components of the fluid lipid layer in

the LPP, explaining the dependence of J in lipid solubility of the permeant. However, for

the drug to reach systemic circulation it has to pass the aqueous layer, with high water

concentration. This explains the dependence of flux of a topically applied permeant on

its aqueous solubility.









Therefore, the mechanism of percutaneous absorption of a permeant is a

solubility-driven process where the higher lipid and aqueous solubility of the drug in the

SC favors higher J.

Model Development

Initially, for the sake of simplicity, the microenvironment of the skin has been

assumed to be homogeneous. The flux of a permeant across a homogenous membrane

can be expressed by a mathematical equation, first proposed by Aldolf Fick in 1855, for

the flow of heat, in which the flux across a homogenous membrane is directly

proportional to the concentration gradient across the membrane (CMl-CMn), where CM1

and CMn are the first few and the last layers of the SC in this case,

J = D/L (CM1-CMn) (1)

and D (diffusion coefficient) is a proportionality constant, L is the thickness of the

membrane under sink conditions, (constant uptake of the permeant by the viable layer)

and CMn is assumed to approach zero. CM1 can be estimated from equation (2)

CM1 = (CVEH) (KMI: VEH) (2)

Where KM1: VEH is the partition coefficient of the permeant between the first few layers of

the SC and the vehicle (VEH). Since D = Do exp (-3V), where Do is the diffusivity of a

hypothetical molecule with zero molecular volume, 3 is a constant characteristic of skin

and V is the van der Waals volume, equation (1) becomes equation (3).

J = (Do exp (-3V)) (CVEH) (KM1:VEH) (3)

One form of equation (3) was derived by Kasting, Smith and Cooper in 1987,34

log J = log (Do/ L) + log SM1 log (3/2.303)V (4)

Where J is maximum flux (saturated solution applied) and CM1 is SM1. Equation (4) has

been used to model data obtained from diffusion cell experiments where the permeants









were applied as a saturated propylene glycol solution. The KSC equation assumes that

the solubility of the permeant in a lipid like octanol, serves as an estimate for the

solubility of the permeant in the membrane; i.e. SM1 = SOCT. The KSC model, therefore

assumes the barrier to percutaneous absorption to be only lipoidal. Flux data for n = 36

compounds was fit to equation 4 and for a correlation of JM with SMEM and V, the r2 value

was 0.74.34

The use of permeability coefficient, P, in equation (5), allows one to normalize the

flux when different concentrations of the permeant in the vehicle are used. Theoretically,

PVEH is assumed to a constant regardless of the CVEH.

PVEH = J/CVEH (5)

Further, when the vehicle is aqueous, KMI:VEH from equation (2) becomes

(KOCT:AQ)f c and equation (3) becomes equation (6).

PAQ = Do exp (-PV) (KM1:AQ)fc (6)

log P = log (Do/ L) + f log KOCT:AQ 3o MW+ log c (7)

Equation (7), developed by Potts and Guy for permeants applied as aqueous

suspensions, was published in 1992.35 Here f is the conversion factor that accounts for

difference between the membrane-vehicle partitioning and octanol-water partitioning

behavior. Two important inferences from Equation (7) arise; first that P is positively

correlated to KOCT: AQ and inversely correlated to SAQ because SAQ is the denomination

in the octanol water partition coefficient. Equation (7) also explains the positive

correlation between P and carbon number and partition coefficient observed by

Schuplein and Blank. Equation (7) gave an r2 = 0.83 for n = 42 phenols and alcohols for









the correlation of P with KOCT: AQ and MW. However a significantly poorer r2 (0.67) was

obtained for the fit of the Flynn data base (n = 93) to the same equation.35

Transformed Potts Guy Equation- The Robert Sloan Equation

One of the drawbacks of the Potts-Guy equation is that it treats the permeability

coefficient, P and KOCT: AQ as the most important predictors of percutaneous absorption.

As mentioned above, the relevant parameter in terms of optimizing the delivery of a

permeant through skin is J not P. Partition coefficient being a ratio, reflects only the

relative affinity of the permeant towards aqueous and lipid phases of the membrane.

Therefore, equation (7) fails to represent the effect of absolute lipid and aqueous

solubilities of the permeant on its percutaneous absorption. Sloan et al observed that in

a homologous series of prodrugs, the more water soluble member of a more lipid

soluble series gave the best flux from a saturated isopropyl myristate (IPM) vehicle.19' 36,

37 Since the donor phase was a saturated IPM solution, J = JMIPM represents the

maximum possible flux available. Additionally, flux did not continuously increase as a

function of increased SIPM (or K). In fact, the flux values in a homologous series

increases initially, then, with increasing SIPM it levels off and finally decreases for the

most lipophilic members of the series. In a Wasdo plot this trend is easy to visualize

(Figurel-3), because it represents the physicochemical properties (like log SAQ, SIPM, log

J and K) as a function of carbon number. The initial shorter chain members of the

homologous series were also the more water soluble ones. The best member of each

series was the shorter chain, initial member, exemplifying the negative correlation

between J and MW and a positive dependence on SLIPID as well as SAQ. In order to

incorporate the effect of absolute aqueous and lipid solubilities, SAQ and SLIPID of the

permeant on its diffusion through skin, Equation (7) was modified as follows:











KMI: IPM = KMI: AQ / KIPM: AQ

KMI: AQ = (KIPM: AQ) C

KM1: IPM = (KIPM: AQ)f c / KIPM:AQ (8)


4 -- --log K (IPMAP )
---logJ MMIPM
logP MIPM
3 ---log J MMAQ
--log P MA
2





0-

-2

-3

-4
24 25 26 27 28 29 30 31 32 33 34 52 53 54 65 56
Prodrug


Figure 1-3. Wasdo Plot. Trends in Fluxes and Permeability Coefficients from a Lipid
Vehicle versus an Aqueous Vehicle (log JMMIPM and log PMIPM versus log
JMMAQ and log PMAQ, respectively) for Prodrugs Compared to their Partition
Coefficients (log KIPM:AQ).

Taking the logs, substituting equation (8) into equation (7) and substituting solubilities

for K gives:

log PIPM = log (Do/L) + f log SIPM f log SAQ log SIPM + log SAQ 13 MW+ log c

Collecting similar terms and adding log SIPM on both sides gave :

log J = log (Do/L) + f log SIPM (1-f) log SAQ 13 MW + log c (9)

Substitution of y for f, collecting constants into a new constant x, and substituting z for 3

gives:

log JMIPM = x + y log SIPM + (1 y) SAQ Z MW (10)









Equation (10) is the Robert Sloan (RS) equation which correlates flux of a

permeant from a saturated lipid vehicle, IPM, across a biological membrane with its SAQ,

SLIPID and MW.38 The RS equation has been used to predict flux of permeants delivered

from saturated aqueous suspensions as well. The form of the equation remains the

same regardless of the vehicle.39 A comparison of a fit of flux data for the same n = 32

compounds through hairless mouse skin in-vitro from water as a vehicle, to KSC,

equation(4), and RS equation, equation(10), gave a r2 value of 0.76 vs 0.90.40 The fit of

the Flynn data base to the RS equation also gave a much better r2 (0.93) compared to

that obtained with the KSC equation (r2 = 0.83).41 The KSC, equation (4) and the Potts-

Guy model for percutaneous absorption represent the SC as a lipoidal barrier whereas

the RS model represents the SC barrier as a biphasic barrier. Therefore, the absolute

aqueous and lipid solubilities, SAQ and SLIPID, and MW of a molecule are the statistically

significant predictors of its flux, J. The utilization of the RS equation for predicting flux

through biological membranes like hairless mouse skin and human skin has been

reviewed in detail. 33 Although a dependence of J on SAQ in a homologous series of

prodrugs was published in a review as early as in 1989 ,37 it was only in 1999 36 that a

quantitative expression reflecting this correlation was developed. The RS equation

quantifies the dependence of J on SAQ, SLIPD and MW. The extent and the nature of the

correlation (positive or negative) between the independent (SAQ, SLIPID, MW) and the

dependent variable (J) is represented by the magnitude of y and z values. The fit of the

flux database comprising n = 32 compounds to the RS equation40 gave x = -2.30, y =

0.575, z = 0.0016 and r2 = 0.90; suggesting an almost equal contribution by aqueous

and lipid solubility terms. Therefore, in terms of prodrug design, a promoiety that









facilitates a maximum improvement in SAQ as well as SLIPID without considerably

increasing the MW of the corresponding prodrug must be ideal to achieve better flux

through the skin compared to the parent drug.

Prodrug Strategy to Enhance Percutaneous Absorption of Drugs

A prodrug is an inactive, chemically functionalized derivative of the parent drug,

which upon chemical or enzymatic hydrolysis regenerates the pharmacologically active

drug form. The derivatization of drugs is not a new approach and has been widely

utilized to solve problems associated with oral delivery of drugs like poor aqueous and

lipid solubility, first-pass effects and poor absorption across biological membranes,

among many others. 42



rNds O Enzymatc \N Chemical, / H
Sr.d.s Fast N O

I -aI \ o NH
0 0

Fosphenytoin Disodium salt N-hydroxymethyl phenytoin Phenytoin

Figure 1-4. Mechanism of hydrolysis of fosphenytoin

The success of prodrug strategy is reflected by the fact that about 15 % of

marketed drug in the US are prodrugs. One such example is fosphenytoin, a disodium

phosphate ester of phenytoin that is used for short term treatment of epilepsy.

Phenytoin is a weakly acidic and a high melting (293 OC) drug. Its poor water solubility

(0.02 mg/ml) leads to an erratic bioavailability profile in humans. The prodrug was found

be about 4408 times more water soluble, and was successful in enhancing oral

bioavailability of phenetoin (Figure 1-4). 43 Fosphenytoin was shown to be enzymatically

labile with half lives 30 min and <30s in rat whole blood and rat liver homgenates









respectively.43 A proposed mechanism of hydrolysis of fosphenytoin to phenytoin has

been shown in Figure 1-4. The rate determining step is the enzymatic cleavage of the

prodrug by phosphatases generating a chemically labile species, which undergoes a

spontaneous loss of formaldehyde to regenerate the parent drug. The rate of hydrolysis

of the hydroxymethyl compound is dependent on the pKa of the leaving group.

Bungaard et al. quantified the relationship between the leaving-group pKa and the half

lives of the corresponding hydroxymethyl compounds.42


HO,, 0



NO HNO





(B) NHR7COOH
N NHRCOOH pHN CsA
0






Ph Ph 0

Figure 1-5. Cyclosporine Prodrugs. (A) Chemical structure of Cyclosporine A (CsA); (B)
Mechanism of chemical hydrolysis of the octa-arginine conjugate of CsA at
pH 7.4.

More recently, poly-arginine based prodrugs of a macrmolecular peptide

immunosuppressant, Cyclosporine A (CsA) have been shown enhance its delivery

through mouse and human skin.44 The prodrug comprised an octa-arginine (R7) chain

linked to the drug via a pH dependent, chemically reversible amino hexanoic acid linker,

which regenerated the parent drug at physiological pH (Figurel- 4). The diffusion of the

polypeptide conjugates was shown to be dependent on the membrane potential and the









number of basic arginine groups (protonated at physiological pH) in the peptide chain.

The mechanism of cellular uptake of polyarginine conjugates of CsA was shown to be a

charge dependent, active process.45 Although most of the prodrugs evaluated as

transdermal or topical delivery agents permeate the skin via a passive process, the

example of polyarginine-CsA conjugates has broadened the scope of topical delivery to

high molecular weight compounds.

A simple analysis of any prodrug form allows one to visualize that it is a "two-

component" system: the parent drug (DRUG-X-) and the promoiety (PRO). The parent

drug is derivatized at one (or more) of the functional (enabling) groups (X) with the

promoiety (PRO). It is obvious that the physicochemical properties of the prodrug form

will be dependent on the nature of PRO.

Table 1-1.General representation of a prodrug, an acyl prodrug and a soft alkyl prodrug.
DRUG-X-PRO General representation of a Prodrug

DRUG-X-C(=O)-R Acyl Prodrug

DRUG-X-CH2-Y-C(=O)-R Soft Alkyl Prodrug

Two general cases arise: Acyl prodrugs, in which the PRO is attached directly to X

and Soft alkyl prodrugs in which, X is attached through a methylene linker to a

heteroatom, Y (O, S etc). Several comprehensive reviews on application of prodrug

techniques in solving drug delivery related problems have appeared in the literature.46

In the previous section, the importance of SAQ, SLIPID and MW of a permeant as

statistically significant predictors of flux (J) across the skin has been described. A few

examples of each prodrug class and the utility of the prodrug approach in modulating

the solubility properties and consequently their flux through skin is discussed below.









Acyl Prodrugs

As the name suggests, in an acyl prodrug the enabling group is attached to the

heteroatom on the drug through a carbonyl bond and as, mentioned above, the nature

of the R group plays an important role in modulating the physicochemical properties of

the prodrug. Some of the most common functional groups targeted for prodrug

approach and the corresponding acyl prodrug formed are:

Table 1-2. Acyl prodrugs formed from hydroxyl or amine containing drugs
General
Chemical Structure Prodrug Type
DRUG-OH DRUG-O-C(=O)-R Ester
DRUG-OH DRUG-O-C(=O)-O-R Carbonate
DRUG-OH DRUG-O-C(=O)-NH-R Carbamate
DRUG-NH2 DRUG-NH-C(=O)-R Amide
DRUG-NH2 DRUG-NH-C(=O)-O-R Carbamate

There are several examples of acyl prodrugs of drugs containing a hydroxyl

group.19 Narcotic analgesics, like morphine, are used widely for management of acute

pain. Most compounds in this chemical class that are currently in clinical use are usually

administered orally or via parenteral routes. The oral and parenteral routes of

administration are associated with problems like high peak plasma concentration,

requirement of frequent dosing and sub-optimal oral bioavailability of the drug due to

extensive first pass metabolism.47 Bundgaard et al. investigated a series of 3-0

alkylcarbonyl and 6-0 alkylcarbonyl ester prodrugs for enhancing delivery of morphine

through skin. Morphine itself has a poor skin permeation profile which can be attributed

to its low water and lipid solubility. The prodrug approach can help circumvent these

challenges. Morphine is only marginally soluble in IPM (SIPM = 0.023 mg ml-1) and pH 7

phosphate buffer (SAQ = 1.8 mg ml 1). 47















) R: H
OR
Testosterone, TS
H,

SH NH C


Testosterone N N Dinmethylaminobutyric
(B) Ester Prodrug, TBSH

RO H

O Naltrexone, NTX
H N 0 0 -

0 HO H [N
n

Straight Chain Ester Proarug O
SN N,N-Dialkyl Carbamate
A, Prodrugs
n

Straight Chain Carbonate Prodrug /
0 N/

H n
Straight Chain Carbamate Prodrug

Figure 1-6. Testosterone and Naltrexone Prodrugs. (A) Chemical Structures of
Testosterone (TS) and its N, N-dimethylaminobutyric acid ester acyl prodrug;
(B) Chemical structure of Naltrexone (NTX) and its previously investigated
prodrugs.

The 3-propionyl ester that exhibited a higher aqueous (SAQ = 3.6 mg ml-1) and lipid

(SIPM = 41 mg ml-1) solubility, showed a flux of 8.7 pg cm-2 h-1 through human skin in-

vitro from a saturated IPM solution. A steady sate flux of up to 35 pg cm-2 h-1 of

morphine was achieved by its 3-hexanoyl ester prodrug (SIPM >150 mg ml-1; SAQ = 2.6

mg ml-1) applied as an IPM solution (200 mg ml-1). Only morphine was detected in the

receptor phase indicating complete conversion of the prodrug to the parent drug.









Morphine itself was undetectable (less than 0.01 pg cm-2 h-1) under the same

conditions.47 Since, saturated solutions were not used in the donor phase in all cases,

the observed flux values do not reflect the highest possible flux of morohine achievable

with these prodrugs. Similar results were obtained with alkyloxycarbonyl esters of

ketobemidone where the most lipid and water soluble member of the series exhibited

higher flux compared with ketobemidone.48

Simple acyl prodrugs (straight chain49 and branched chain50 3-O-alkylcarbonyl

esters and 3-O-alkyloxycarbonyl esters51) of NTX showed a moderate enhancement of

flux of the parent drug (-2-7 fold). 3-O-Alkylaminocarbonyl prodrugs of NTX exhibited a

3-4 fold improvement in flux of the parent drug (Figure 1-6).52 Similarly, alkyloxycarbonyl

(AOC) prodrugs of APAP have been investigated in diffusion cell experiments in our lab

previously. None were more water soluble than APAP. However the two more water

soluble members of the AOC prodrugs of APAP (Figure 1-7) exhibited a moderate

enhancement in flux (>2 times).53

Misolovich et. al.54 investigated testosteronyl-4-dimethylaminobutyrate HCI

(TSBH) as potential prodrug candidate for enhancing topical delivery of testosterone

(TS) (Figure 1-6). Solubility of TBSH was about 1000 times greater than TS in pH 7.0

phosphate buffer (>150 mg ml- for TBSH compared to 0.01 mg ml- for TS). The flux of

TSBH from a 10% solution was about 60 times higher than TS itself.55 Even the free

base form of TSBH delivered 35 times more TS through human skin in-vitro. The

protonated amine group in the promoiety has been infered to confer a more favorable

balance in lipid and aqueous solubility to the prodrug than simple alkyl promoieties by









increasing its aqueous solubility and hence enhancing skin penetration rates. But at the

pH of highest solubility, achieving optimal stability has been a challenge.

(A) ,o OH
N M SI =1.9 mM; SAO = 100mM
H

(B) 0 a'.>
() O 0 SIPM= 1.08mM;SAQ 1.31mM

H

(P) Sip = 62 mM; SAO = 24.7mM

H


(D) O Nr I O 0 SIPM = 20.7 mM; SA = 7.76mM

H

0(E) I O N OY SM = 1.15 mM; SAQ = .6mM

H

Figure 1-7. Chemical structures and IPM and aqueous solubilites of (A) APAP; (B) C2-
AOC-APAP; (C) C2-ACOM-APAP; (D) C2-AOCOM-APAP; (E) C2-
NANAOCAM-APAP

When an ionizable amine group is incorporated in the acyl portion of the ester, it

activates the acyl group towards nucleophilic attack by hydroxide ion due to two

reasons, 56, 57(a) strong electron withdrawing (-I) effect of the protonated amine group

and; (b) intramolecular catalytic effect of the protonated and unprotonated amine group

(general base catalysis). Basic hydrolysis of amino acid esters have been investigated

in great deail. 57 Steric and electronic factors and pH have been shown to affect the

rates of hydrolysis. The acyl prodrugs hydrolyze to regenerate the parent drug via the

mechanism shown in Figure 1-8.










Vitamin E, a phenol containing compound, is a powerful antioxidant and has been

shown to play a role in maintaining general physiological health of the skin.

(A)H
R 0 H20 R
H+ H20, Slow V DRUG
X O X ER 0 1O-X X "
DRUG DkUG

H H, Ht 0
H -,O -H+ O
X 0 DRUG -XH
DRUG" +' OH HO R R OH + HX-DRUG
R OH


R OH
SOH-, Slow R.- DRUG
(B) X O OX --X
I e
DRUG

0 0
Drug-X+ lOH DRUG -XH +
R OH R OH




Figure 1-8. Mechanism of of Hydrolysis of Acyl Prodrugs. (A) acid and (B) base
catalyzed mechanisms of hydrolysis of acyl prodrugs.

Vitamin E acetate, an ester prodrug, is a part of many commercial cosmetic

formulations. The poor SAQ and high MW, however, limit its permeation through skin. In

recent reports, the skin retention of eight amino acid ester prodrugs of Vitamin E (Figure

1-9) were evaluated.58 Since the prodrugs exhibited low solubility in buffer (pH 8.0,

phosphate), solubilities were evaluated in pH 8 phosphate buffer containing 5%

dimethyl 3 cyclodextrin. The enzymatic (porcine liver esterase) half life of the glycine

derivative was 49.5 min whereas the most hydrophilic member, the L-citruline prodrug,

showed a half life of 1013 min.58









All the prodrugs showed higher water solubility than Vitamin E itself and some of

the derivatives underwent hydrolysis at a rate that could ensure regeneration of the

parent drug in the skin.


OR




R: -H -0NOH
-C-O OH

-OCH3 0

-0 ~-0 O

NH,
-NNH NH,0-

NH2 O
NH2 NH2





H H
--0 HT 0 0
0


Figure 1-9. Chemical structures of Vitamin E and its prodrugs evaluated previously.

The skin retention of the prodrugs was higher than Vit-E itself. However little or no

parent drug could be detected in the skin and the receptor phase.58 The low J of Vitamin

E prodrugs was expected and can be explained by the negative correlation of the MW

with flux. The high MW, low SAQ and extremely high SLIPID (a very poor balance in

aqueous and lipid solubility) of Vitamin E itself render it a challenging molecule to

deliver through skin. To achieve an improved J of Vitamin E through the skin by its

prodrugs, the increase in aqueous solubility may have to be dramatic without increasing

the MW of the compound to an appreciable extent. However, the amino acid promoiety

does show promise to improve solubility properties of lower molecular weight drugs.54 55









Soft Alkyl Prodrugs

A soft alkyl prodrug has a methylene or a vinylygous methylene linker between the

drug heteroatom and a heteroatom in the promoiety. Soft alkyl prodrugs of heterocyclic

amines like 5 FU, Theophylline (Th), and 6-mercaptopurine (6-MP) and phenolic drug

like APAP have been evaluated for the potential to enhance topical delivery of their

parent compounds. Soft alkyl drug can undergo an esterase mediated reaction to

generate a hydroxymethyl intermediate, which undergoes spontaneous chemical

hydrolysis to regenerate the parent drug (Figure 1-10).19 One such example discussed

previously is fosphenytoin (Figure 1-3).

o
0 0 0 ROOH
|| Enzymatic 0 0
~ N/ or chemical I' +
DRUG H OR DRUG H OH
OH
o
AN OH
+ HCHO ---- DRUG H

DRUG

Figure 1-10. Mechanism of hydrolysis of a soft alkyl prodrug.

N-Mannich base (a dialkylamino) alkyl derivatives of molecules containing acidic

-NH groups have been shown to possess enhanced water as well as lipid solubility. The

masking of the polar acidic group results in an increase in lipid solubility whereas the

aqueous solubility is increased by the incorporation of a basic amine group into the

promoiety. For example, among a series of N-Mannich derivatives of 5-FU one member,

bis-(diethylamino)methyl derivative, gave a higher flux than the commercially available

formulation of 5-FU, Efudex. A 10% suspension of the bis-(diethylamino)methyl prodrug

in IPM gave a flux of 0.11 0.013 mg cm-2 h-1 whereas Efudex gave a flux of 0.017

0.046 mg cm-2 h-1 through hairless mouse skin. Because of difficulty in purification,









many members of this series were not evaluated in diffusion cell experiments.19

Nevertheless, the bis-(diethylamino)methyl prodrug of 5-FU clearly exemplifies the

effect that incorporation of a basic amine group into the promoiety has on the biphasic

solubility, and consequently on permeation of the corresponding prodrug. The

mechanism of hydrolysis of N-Mannich base prodrug involves unimolecular cleavage to

parent -NH acidic compound and a carbocation 19(Figure 1-11 ).The lower the pKa of the

parent -NH compound, the lower is the half life of its N-mannich prodrug. An equation

for predicting the half life of the N-Mannich prodrug based on the pKa of the parent -NH

compound has also been developed. 42

0 0
(A) O F (B) FN
(A) HN Et2N
O IN O -N
Hk
NEt2
bis-idilehylaminoi methyl prodrug of 5-FU

(c)
F UH2 F
OEt2N N E-t2N -c
O'N ON
NEt2 NEt2

Figure 1-11. Mannich Base Prodrugs of 5 FU. (A) Chemical structure of 5-FU; (B) N-
Mannich base prodrug of 5-FU; (C) Mechanism of hydrolysis of bis-
(diethylamino) methyl prodrug of 5-FU.

The soft alkyl approach has also been applied to phenols. Alkylcarbonyloxymethyl

(ACOM), 59 alkyloxycarbonyloxymethyl (AOCOM)60 and N-alkyl -N-

alkyloxycarbonylaminomethyl (NANAOCAM) 61prodrugs of a model phenol-

acetaminophen (APAP), were investigated in diffusion cell experiments (Figure 1-7).

The flux of the most water soluble member of the ACOM series was only 3.6 times









higher than APAP and only one member of AOCOM prodrugs series (4-AOCOM-APAP,

C2) gave higher flux (1.6 times) than APAP itself. Although the soft alkyl approach has

been utilized successfully to generate a "synthetic handle" on parent drugs, and hence

broaden the choice of functional groups that can be incorporated into the promoiety,

their performance was not significantly better than simple acyl derivatives of APAP.

Conclusions

Although delivery of drugs through the oral route is preferred over other routes of

drug administration, it is associated with problems like first pass effects, GI and liver

toxicity and poor absorption of drugs through the gut wall. Topical delivery of drugs is a

practical alternative with which the common challenges of oral delivery can be

overcome. Among several available methods for enhancing topical delivery of drugs,

the prodrug approach is the one with which optimization of maximum achievable flux of

a drug through skin is feasible. The statistically significant predictors of flux of permeant

(J) have been elucidated.62 Among a homologous series of prodrugs, the initial, low

molecular weight members exhibiting the best balance between SAQ and SLIPID have

been shown to give the best flux. 36 Therefore the relevant design directives for

developing prodrugs intended for enhanced topical delivery of its parent drug includes,

(a) higher aqueous as well as lipid solubility relative to the parent drug without a

considerable increase in the MW of the corresponding prodrug and (b) ability to

regenerate the parent drug in the skin dermall delivery) or in systemic circulation

transdermall delivery).

AOC53 (acyl), ACOM,59 AOCOM60 and NANAOCAM 61 (soft alkyl) prodrugs of a

model phenolic drug, APAP, have been evaluated in our lab previously in diffusion cell

experiments. However, enhancement in J was only minimal. In view of the previous









research on the effect of inclusion of an ionizable group in the promoiety it seemed

appropriate to investigate the development of prodrugs of poorly water soluble phenolic

drugs with improved aqueous solubility and hence delivery properties. In the present

study an efficient synthetic route to N, N'- dialkylaminoalkylcarbonyl (DAAC) and

aminoalkylcarbonyl (AAC) ester prodrugs of the phenolic drug acetaminophen will be

presented. The hypothesis is that the incorporation of the ionizable amine group will

confer to the prodrugs a higher SAQ as well as SLIPID, and hence a higher J than APAP.

If successful, these compounds can serve as models for optimization of the DAAC and

the AAC prodrug approach for improving topical delivery.









CHAPTER 2
SPECIFIC OBJECTIVES

First Objective

The first objective of this study was to synthesize a series of N, N' -

dialkylaminoalkylcarbonyl (DAAC) and aminoalkylcarbonyl (AAC) prodrugs of a model

phenolic drug. Although, acyl and soft alkyl prodrugs of a model phenol, acetaminophen

(APAP) investigated previously in our laboratory in diffusion cell experiments gave a

moderate enhancement in SIPM, none of the prodrugs were more water soluble than

APAP. For example the most water soluble member of the AOC series was C1-AOC-

APAP (21 mM) compared to 100 mM for APAP. Similarly, among the ACOM and

NANAOCAM (soft alkyl) prodrugs of APAP the most water soluble member was C2-

ACOM-APAP (25 mM) and C1-NANAOCAM-APAP (45 mM) respectively. The flux

enhancement could be attributed primarily to an increase in SIPM of the prodrugs relative

to APAP. The hypothesis for this study is that the incorporation of an ionizable amine

group (-NR1R2 or -NH2) into the acyl prodrug would result in a favorable increase in SAQ

and SLIPID Of the corresponding prodrug. The effect of incorporation of an ionizable

amine into the acyl prodrugs of APAP on its skin permeation has not been reported.

Therefore we synthesized DAAC and AAC prodrugs of APAP which can act as model

compounds for optimization of the DAAC and AAC prodrug approach.

Synthetic routes to N, N'- dialkylaminoalkylcarbonyl (DAAC) and

aminoalkylcarbonyl (AAC) acyl prodrugs of a phenolic drug acetaminophen (APAP)

have been presented. DAAC-APAP prodrugs were synthesized via a three step

procedure starting with haloalkylcarbonyl esters which were reacted with five different

amines: N, N'- dimethylamine, N, N'- diethylamine, N, N'- dipropylamine, morpholine









and piperidine. The spacing between the amino group and the carbonyl group of the

acyl group was 1 to 3 -CH2 groups. After the hydrolysis of the ester the promoiety was

subsequently coupled with the parent drug via a dicyclohexylcarbodimide (DCC)

mediated coupling to yield the DAAC-APAP-HCI prodrugs in excellent yields. The

DAAC-APAP prodrugs (free base form) were prepared by treatment of the

corresponding HCI salts with aqueous base. The AAC prodrugs were synthesized using

commercially available Boc-protected amino acids using DCC or EDCI as coupling

agents.

Second Objective

The second objective of this research was to investigate whether the DAAC and

AAC prodrugs can improve topical delivery of APAP. Hairless mouse skin was used for

in-vitro analysis of the prodrugs. Physicochemical properties, i.e., lipid and aqueous

solubility were determined for the DAAC and AAC prodrugs. Additionally, to investigate

the extent to which the prodrug might hydrolyze during the course of diffusion cell

experiment, half lives (t11/2) of a few members of the DAAC and AAC series were

measured in buffer (pH 6.0, 20 mM).

Third Objective

The third objective of this study is to develop a new Robert-Sloan equation using

an integrated solubility and flux database for compounds evaluated in diffusion cell

experiments previously in our lab. The new RS equation will be developed by fitting the

solubility and flux data for n = 73 compounds comprising n = 42 compounds from

Roberts et. al.,36 the C5-Bis 6, 9 ACOM-6MP prodrug, 53 n = 4 3 AC 5FU prodrugs, 63 n

= 6 ACOM 5FU prodrugs, 64 n = 2 bis-AC-5FU prodrugs, 65 n = 8 AOC-APAP

prodrugs,53 n = 5 ACOM-APAP59 prodrugs and n = 5 AOCOM-APAP60 prodrugs. The









new RS equation will be used to predict the flux of DAAC-APAP prodrugs through

hairless mouse skin.









CHAPTER 3
N, N' DIALKYLAMINOALKYL CARBONYL (DAAC) PRODRUGS OF A MODEL
PHENOLIC DRUG -ACETAMINOPHEN (APAP)

Introduction

First pass effects in the GI tract and the liver limit the oral bioavailability of a drug

containing a hydroxyl group phenolss or alcohols). The presence of the hydroxyl group

however facilitates the derivatization of the drug to give prodrugs. Depending on the

choice of the target site and route of administration, the prodrugs can be designed with

appropriate physicochemical properties. Presence of one or more polar hydroxyl groups

in a drug may result in low lipid solubility. Masking the polar hydroxyl group results in a

better balance in aqueous and lipid solubilties. This leads to an enhanced absorption of

orally administered drugs. For example, an enhanced oral bioavailability of terbutaline in

dogs was achieved by its carbamate esters. A single oral dose of the bis N, N-

dimethylcarbamate of terbutaline could produce sustained blood levels of the parent

drug and showed a plasma half-life of 10 h.42

An ideal promoiety for developing prodrugs of poorly water-soluble drugs

containing a hydroxyl group is one that can increase its water solubility without

compromising lipid solubility. In this regard, a promoiety containing an amine group has

been utilized successfully to improve aqueous solubility of poorly water-soluble drugs. 55

The enhanced aqueous solubility is a direct consequence of a distinct physicochemical

feature of prodrugs that contain a basic amine (primary, secondary or tertiary)

functionality ionization of the basic amine nitrogen at physiological pH.

The presence of an amine group activates the acyl group towards hydroxide ion

attack because of the (-I) effect of the amine group and promotes its intramolecular

general base catalyzed hydrolysis.46' 56,57 Kinetic behavior of esters containing an









activated acyl group (metranidazole N, N-dimethyl glycinate ester,42 hydrocortisone

lysine ester66 and acetaminophen glycine, a-aspartic acid and 3-aspartic acid esters67)

has been studied in detail. However, attaining optimal stability in aqueous solutions has

been a challenge towards developing prodrugs containing an ionizable, basic amine

functionality.

An example, where the incorporation of an ionized amine into the promoiety has

been utilized to enhance percutaneous absorption of a lipophilic, poorly water soluble

drug is testosterone (TS). The TBSH prodrug showed a 60 times higher flux through

human skin in-vitro than TS when applied as a 10% solution in pH 7.0 phosphate buffer.

It has been hypothesized that the enhancement in flux is a consequence of an increase

aqueous solubility of TBSH relative to TS.55

Only a few reports describing synthesis of acyl prodrugs of phenolic drugs in which

a basic amine group was incorporated into the promoiety have appeared in the

literature. A synthetic route for obtaining N, N'-Dialkylaminoalkylcarbonyl (DAAC)

prodrugs of APAP with different amine groups and the alkyl chain length have not been

reported. Additionally, no systematic evaluation of the effect of pKa and the distance

from the acyl group of the amine on the solubility and skin permeation properties of N, N

-dialkylaminoalkylcarbonyl (DAAC) prodrugs of phenolic drugs has been carried out. A

detailed study of the solubility, stability and in-vitro permeation behavior of DAAC

prodrugs of a model phenolic drug can aid the design of DAAC prodrugs of other poorly

water-soluble phenolic drugs. A low molecular weight phenolic drug, acetaminophen,

was chosen to optimize the DAAC prodrug approach.










Synthesis of DAAC Prodrugs of Acetaminophen

In the present study, a series of DAAC prodrugs of APAP (5a 5k) were

synthesized. The choice of the amine incorporated into the promoiety was guided by

two factors: (a) pKa of the amine and (b) steric bulk on the amine group. Five different

amines, N, N-dimethylamine, N, N-diethylamine, N, N-dipropylamine, piperidine and

morpholine with pKa values between 10.2 8.3 were chosen as candidates for the

synthesis of the amino acids (3b-3k).

(a) 0 OR1


H


RI=H

APAP


0
R1= R3R2NJl

DAAC-APAP
Prodrugs 5a-5k


-CH2-N


-CH2H2C--N


-H2C-N


,HC-H2C-N



H2C-N


,nNR2R3


5a -CH2-NE


5b -CH2*CH2'.N


5c -H2C-H2C-H2C-Nj

H2C-N 0
5d

--CHCH2-N 0

5e
-- CH2CHl O0


Figure 3-1. Chemical structures of synthesized DAAC-APAP prodrugs


Materials and Methods. Melting points were determined on a Meltemp melting

point apparatus. Thin layer chromatography (TLC) plates (Whatman AL Sil G/UV) were









purchased from Fisher. N, N-diethylamine, N, N-dipropylamine, piperidine, morpholine,

ethyl 3-(N, N-dimethylamino)propionate (1b), N, N-dimethylglycine hydrochloride (3a), 3

- (N, N-diethylamino)propionic acid hydrochloride (3d), ethyl chloroacetate, methyl 3-

bromopropionate and dicyclohexylcarbodiimide (DCC) were purchased from Sigma

Aldrich. Ethyl 4-bromobutyrate was purchased from Alfa Aesar. Spectra (1H NMR) were

recorded on a Varian Unity 400 MHz spectrometer. Pyridine, acetonitrile and

dichloromethane were dried over 3 A molecular sieves before use. pKa values were

estimated using pKa software, (ACD labs, version 12.0).

Experimental. The DAAC prodrugs of APAP (5a 5k) were synthesized in three

steps (Figure 3-1). A a, 3, or y haloalkanoic acid ester was reacted with a secondary

amine to obtain the corresponding N, N-dialkylaminoalkanioc acid ester (2b-2h). These

esters were hydrolyzed with concentrated hydrochloric acid to yield the hydrochloride

salts of the amino acids (3a-3k), which were subsequently coupled with APAP by a

dicyclohexylcarbodimide (DCC) mediated esterification to give the hydrochloride salts of

the desired prodrugs (4a 4k). The prodrugs (5a 5k) were obtained from the

corresponding hydrochloride salts by treatment with ice cold saturated aqueous

Na2HCO3 solution and immediate extraction with dichloromethane.

General Method for the synthesis of compounds 2c and 2e-2k. To a round

bottom flask containing a well-stirred solution of the a, 3, or y haloalkanoic acid ester (1

equiv.) in 25 ml acetonitrile was added, 2 equiv. of the amine drop-wise. The contents of

the flask were refluxed for 8-10 h. NMR of the crude reaction mixture confirmed

completion of the reaction. The flask was cooled in an ice bath for 1 2 h.









R2 O
0 H ACN, Reflux N H
N .RB .!oN,,J HX
X R + RNR3 8-10hR 8-lh R3 n OR R2' HR3X

1 2b-2k
X = CI, Br or I
n=1,2or3

R2 O R2 O
R .N Cone. HCI, 12 h HCI 4 DCC, Py, rt
R-(O Ro -OH _
APAP
2b-2k 3b-3k

H H
NR2 O N Sat. Aq NaHC03 RN O
R2 0 1 rR2 0

R3HI 0-4 oC.30S Rs"'
HCI n
4a-4k 5a-5k

Figure 3-2. Synthesis of DAAC-APAP prodrugs

The resulting suspension was tritrated with 100 -150 ml diethyl ether to crystallize

the HCI, HBr or HI (in case of 1c and 1e) salt of the corresponding amine. The solid

amine salt that formed was removed by filtration and the filtrate was concentrated on a

rotary evaporator to give the desired aminoalkanoic acid esters as oils. In the case of 1c

and le (sterically hindered amines) the alkylating agent, ethyl chloroacetate, had to be

converted to ethyl iodoacetate by treatment with 1 equiv of Nal in 10 ml acetone prior to

coupling with the amine otherwise the reaction did not go to completion. Ethyl

iodoacetate was not isolated or characterized. No formation of

R1R2N(CH2)nC(=O)NR1R2 was observed in this reaction. All the esters were obtained

as yellow to orange colored oils in good yields (83 98 %).

2c Ethyl 2-(diethylamino)acetate. Synthesized from 4.6 g (0.03 mole) ethyl

chloroacetate and 5.59 g (0.07 mole) diethylamine to give a brown oil. NMR (CDCI3): 5

4.18 (q, 2H), 3.31 (s, 2H), 2.66 (q, 4H), 1.27 (t, 3H), 1.06 (t, 6H). Yield: 95%.









2e Ethyl 2-(dipropylamino)acetate.Synthesized from 10.30 g (0.08 mole) ethyl

chloroacetate and 17.1 g (0.16 mole) di-propylamine to give a brown oil. 1H-NMR

(CDCI3): 5 4.05 (q, 2H), 3.20 (s, 2H), 2.42 (t, 4H), 1.35 (sextet, 3H), 2.78 (t, 3H), 0.76 (t,

6H) Yield: 86%

2f Ethyl 2-(piperidin-1'-yl)acetate. Synthesized from 4.1 g (0.03 mole) ethyl

chloroacetate and 5.7 g (0.06 mole) piperidine to give a pale pink oil. 1H-NMR (CDCI3):

5 4.18 (q, 2H), 3.18 (s, 2H), 2.50 (broad t, 4H), 1.63 (quin, 4H), 1.45 (m, 2H), 1.28 (t,

3H). Yield: 98%

2g Ethyl 3-(piperidin-1'-yl)propanoate. Synthesized from 1.53 g (0.009 mole) of

methyl 3-bromopropionate and 1.532 g (0.02 mole) piperidine to give a pale yellow oil.

1H-NMR (CDCI3): 5 3.68 (s, 3H), 2.67 (t, 2H), 2.52 (t, 2H), 2.40 (broad s, 4H), 1.58

(quin, 4H), 1.42 (m, 2H). Yield: 98%

2h Ethyl 4-(piperidin-1'-yl)butanoate. Synthesized from 4.45 g (0.02 mole) ethyl

4-bromobutyrate and 4.855 g (0.04 mole) piperidine. 1H-NMR (CDCI3): 5 4.12 (q, 2H),

2.37-2.27 (m, 8H), 1.814 (quin, 2H), 1.57 (m, 4H), 1.421 (m, 2H), 1.254 (t, 3H).Yield:

86%.

2i Ethyl 2-(morpholin-4'-yl)acetate. Synthesized from 2.3 g (0.018 mole) of ethyl

chloroacetate and 3.27g (0.037 mole) morpholine to give a yellow oil. 1H-NMR (CDCI3):

54.19 (q, 2H), 3.75 (t, 4H), 3.20(s, 2H), 2.587 (t, 4H), 1.281 (t, 3H). Yield: 83%

2j Ethyl 3-(morpholin-4'-yl)propanoate. Synthesized from 4 g (0.02 mole) of

methyl 4- bromopropionate and 4 g (0.04 mole) of morpholine to give a brown oil. 1H-

NMR (CDCI3): 5 3.68 (broad s, 7H), 2.68 (t, 2H), 2.50 (t, 2H), 2.45 (broad s, 4H).Yield:

98%









2k Ethyl 4-(morpholin-4'-yl)butanoate Synthesized from 4.6 g (0.02 mole) ethyl

4-bromobutyrate and 4.1 g (0.04 mole) morpholine. 1H-NMR (CDCI3): 5 4.11 (q, 2H),

3.68(m, 4H), 2.415 (m, 2H), 2.314-2.361 (m, 4H), 1.80 (quin, 2H), 1.254 (t, 3H).Yield:

94%

General method for the synthesis of compounds 3b, 3c, 3e 3k. The esters

2b, 2c and 2e-2k were refluxed with excess (5-10 equiv.) of conc HCL for 8-10 h after

which the water was evaporated. The residue was triturated with 50-75 ml THF

overnight. The obtained solids were dried in a vacuum oven at 40C overnight to yield

compounds 3b, 3c and 3e-3k as crystalline solids.

3b 3-(dimethylamino)propanoate hydrochloride. 1H-NMR (D20): 5 3.42 (broad

t, 2H), 2.89 (broad s, 8H). Yield: 98%

3c 2-(diethylamino)acetic acid hydrochloride. 1H-NMR (D20): 5 3.97 (s, 2H),

3.27 (q, 2H), 1.29 (t, 6H). Yield: 68%

3e 2-(dipropylamino)acetic acid hydrochloride. 1H-NMR (D20): 5 3.85 (s, 2H),

3.02 (q, 4H), 1.57 (sextet, 4H), 0.805 (t, 6H) Yield: 68%

3f 2-(piperidin-1'-yl)acetate hydrochloride. 1H-NMR (D20): 5 3.94 (s, 2H), 3.57

(broad d, 2H), 3.0 (td, 2H), 1.45-1.93 (m, 6H). Yield: 100%

3g 3-(piperidin-1'-yl)propanoate hydrochloride. 1H-NMR (D20): 5 3.53 (broad

doublet, 2H), 3.392 (t, 2H), 2.965 (td, 2H), 2.87 (t, 2H), 1.46-1.99 (m 6H). Yield: 94%

3h 4-(piperidin-1'-yl)butanoate hydrochloride. H-NMR (D20): 5 3.5 (broad

doublet 2H), 3.098 (broad triplet, 2H), 2.9 (broad triplet, 2H), 2.466 (2H, t), 1.983-1.4 (m,

8H). Yield: 80%









3i 2-(morpholin-4'-yl)acetate hydrochloride. 1H-NMR (D20): 3.8-4.2(m, 2H),

3.11 (s, 2H), 3.85 (broad t, 2H), 3.5-3.65 (m, 2H), 3.25 (broad t, 2H). Yield: 82.9%

3j 3-(morpholin-4'-yl)propanoate hydrochloride. 1H-NMR (D20): 5 4.15 (dd

2H), 3.8 (td 2H), 3.52 (broad doublet 2H), 3.44 (t, 2H), 3.2 (td, 2H), 2.95 (t, 2H). Yield:

82%.

3k 4-(morpholin-4'-yl)butanoate hydrochloride. 1H-NMR (D20): 5 4.12 (broad

doublet 2H), 3.80 (t, 2H), 3.56 (broad d, 2H), 3.54 (broad doublet, 2H), 3.15-3.24 (m,

4H), 2.5 (t 2H), 1.98-2.15 (m, 2H). Yield: 96%.

General method for the synthesis of 4a-4k. Equimolar quantities of APAP,

dialkylaminoalkanoic acid hydrochloride (3a 3k) and DCC in 35 40 ml dry pyridine

were stirred at room temperature for 24 48 h. The suspension was triturated with about

75 ml dichloromethane and 25 ml diethyl ether. The suspension was stirred vigorously

for 1 h. Then, the suspension was filtered and the solid obtained was washed

thoroughly with dichloromethane till no smell of pyridine remained. The filtered solid was

dried under vacuum for a few hours and then was refluxed with 250-350 ml chloroform

for 4 h. The resulting suspension was filtered while hot to yield 4a 4k as white solids,

while the dicyclohexyl urea remained in the filtrate.

4a 4-Acetamidophenyl 2'-(dimethylamino)acetate hydrochloride, (Me2n1-

APAP-HCI) Synthesized from 1 g (0.006 mole) APAP, 0.92 g (0.006 mole) 3a and 1.36

g (0.006 mole) DCC. 1H-NMR (D20): 5 7.47 (d, J = 8.8, 2H), 7.19 (d, J = 9.2, 2H), 4.45

(s, 2H), 3.04 (s, 6H), 2.14 (s 3H).Yield: 75%

4b 4-Acetamidophenyl 3'-(dimethylamino)propanoate hydrochloride, (Me2n2-

APAP-HCI) Synthesized from 1.48 g (9.8 mole) acetaminophen, 1.5 g (9.8 mole) 3b








and 2.02 g (9.8 moles) DCC. 1H-NMR (D20): 5 7.45 (d, 2H), 7.19 (d, 2H), 3.49 (t, 2H),

3.02 (t, 2H), 2.94 (s, 6H), 2.14 (s, 3H). Yield: 85%

4c 4-Acetamidophenyl 2'-(diethylamino)acetate hydrochloride, (Et2ni-APAP-

HCI) Synthesized from1.25 g (0.008 mole) APAP, 1.70 g (0.008 mole) DCC and 0.008

mole 3c. 1H-NMR (D20): 5 7.48 (d, J = 8.8, 2H), 7.20 (d, J = 8.8 2H), 4.47 (s, 2H), 3.39

(4H, q), 2.15 (s 3H), 1.347 (t, 6H). Yield: 70%

4d 4-Acetamidophenyl 3'-(diethylamino)propanoate hydrochloride, (Et2n2-

APAP-HCI). 1H-NMR (D20): 5 7.45 (d, 2H), 7.19 (d, 2H), 3.69 (t, 2H), 3.30 (4H, q), 3.21

(t, 2H), 2.15 (s 3H), 1.31 (t, 6H). Yield: 80%

4e 4-Acetamidophenyl 2'-(dipropylamino)acetate hydrochloride, (Pr2n1-

APAP-HCI). 1H-NMR (D20): 5 7.47 (d, J = 8.8, 2H), 7.19 (d, J = 8, 2H), 4.32 (s, 2H),

3.11 (t, 4H), 2.01 (s 3H), 1.82 (sextet, 4H), 0.824 (t, 6H). Yield: 50%

4f 4-Acetamidophenyl 2'-(piperidin-l"-yl)acetate hydrochloride, (PIPnl-

APAP-HCI). Synthesized from 2.5 g (0.014 mole) 3f, 2.87g (0.014 mole) of DCC and

2.1 g (0.014 mole) APAP. 1H-NMR (D20): 5 7.50 (d, J = 8.8, 2H), 7.22 (d, J = 8.8, 2H),

4.36 (s, 2H), 3.37 (m, 4H), 2.171 (s, 3H), 1.906-1.68 (m, 6H). Yield: 71%

4g 4-Acetamidophenyl 3'-(piperidin-l"-yl)propanoate hydrochloride, (PIPn2-

APAP-HCI). Synthesized from 1.5 g (0.08 mole) 3g, 1.7 g DCC and 1.25 g (0.08 mole)

APAP. 1H-NMR (D20): 5 7.47 (d, J = 8.4, 2H), 7.16 (d, J = 8.8, 2H), 3.50-3.57 (m, 4H),

3.19 (t, 2H), 3.01 (t, 2H), 2.16 (3H, s). Yield: 65%

4h 4-Acetamidophenyl 4'-(piperidin-l"-yl)butanoate hydrochloride, (PIPn3-

APAP-HCI) Synthesized from 1.5 g (0.008 mole) 3h, 1.6 g (0.008 mole) DCC and 1.2 g

(0.008 mole) APAP. 1H-NMR (D20): 5 7.44 (d, J = 8.8, 2H), 7.12 (d, J = 8.8, 2H), 3.57









(d, 2H), 3.16 (t, 2H), 3.19 (t, 2H), 2.92 (t, 2H), 2.76 (t, 2H), 2.14 (s, 3H), 1.40-2.15 (m,

6H). Yield: 78%

4i 4-Acetamidophenyl 2'-(morpholin-4"-yl)acetate hydrochloride, (MORnl-

APAP-HCI). Synthesized from 1.5 g (0.008 mole) 3i, 1.6 g (0.008 mole) DCC and 1.6 g

(0.008 mole) APAP. 1H-NMR (D20): 5 7.49 (d, 8.8 Hz 2H), 7.21 (d, 8 Hz 2H), 4.48

(broad s, 2H), 4.031 (broad s, 4H), 3.50 (broad s, 4H), 2.16 (3H, s).Yield: 70%

4j 4-Acetamidophenyl 3'-(morpholin-4"-yl)propanoate hydrochloride,

(MORn2-APAP-HCI) Synthesized from 1.5 g (0.007 mole) 3j, 1.5 g (0.007 mole) DCC

and 1.09 g (0.007 mole) APAP. 1H-NMR (D20): 5 7.44 (d, 8 Hz, 2H), 7.14 (d, J = 8.8 Hz,

2H), 4.11 (broad s, 2H), 3.83 (broad s, 2H), 3.60 (broad t, 4H), 3.21 (broad s, t, 4H),

2.141 (s, 3H).Yield: 50%.

4k 4-Acetamidophenyl 4'-(morpholin-4"-yl)butanoate hydrochloride,

(MORn3-APAP-HCI) Synthesized from 1.5 g (0.007 mole) 3k, 1.5 g (0.007 mole) DCC

and 1.1 g (0.007 mole) APAP. 1H-NMR (D20) 5: 7.45 (d, 8 Hz, 2H), 7.13 (d, J = 12 Hz,

2H), 4.09 (broad s, 2H), 3.8 (broad s, 2H), 3.54 (broad s, 2H), 3.21 (broad s, 2H), 3.28

(t, 2H), 2.788 ( t, 2H), 2.14 (s, 3H), 2.17-2.09 (m, 2H). Yield: 61%

General method for the synthesis of compounds 5a 5k. The hydrochloride

salts of DAAC prodrugs of APAP (4a 4k) were treated with 5 -10 ml ice cold saturated

aqueous sodium bicarbonate solution and the aqueous layer was extracted immediately

with 200 250 ml methylene chloride once. The treatment with aqueous saturated

NaHCO3 and the subsequent extraction had to be very fast (<30 sec) otherwise,

hydrolysis (20 60%) to the parent drug was observed. The organic layer was dried

over sodium sulfate for 1 h and filtered. The solvent was evaporated on a rotary










Table 3-1. Characterization of DAAC-APAP-HCI and DAAC-APAP Compounds.
Elemental Analysis (Calculated and Experimental Values of % of Carbon,


Hydrogen and Nitrogen) and Melting Points for
APAP Prodrugs


Mp

4a 200-210

4b 195-200

4c 200-205

4d 170-180

4e 175-185

4f 234

4g 202

4h 260-270

4i 220

4j 210

4k 234

5a 90-92

5b 56-60

5c 55-57

5d 60-62

5e Oil

5f 122-124

5g 155-158

5h 114-115

5i 126-128

5j 148-158

5k 98-100


C
Cal
52.85

54.45

55.90

57.23



58.50

59.90

53.42

54.79

54.62a

58.50 a

59.49 b

63.62







65.20

66.18



60.42

61.63

62.73


C
Exp
52.58

54.26

55.90

57.11



59.06

59.73

53.36

54.17

54.57

59.06 a

59.86 b

64.18







64.96

65.96



60.66

61.64

62.82


H
Cal
6.28

6.68

7.04

7.36



7.09

7.39

6.08

6.44

6.88a

7.09 a

7.22 b

7.63


DAAC-APAP-HCI and DAAC-


H
Exp
6.47

6.70

7.04

7.44



7.15

7.54

6.05

6.47

6.57

7.15a

6.91 b

7.85


7.30 7.28

7.64 7.65


6.52

6.90

7.24


6.58

6.80

7.38


N
Cal
10.27

9.77

9.31

8.90



8.57

8.22

8.90

8.52

7.96a

8.57 a

11.63 b

10.60







10.14

9.65



10.07

9.58

9.14


N
Exp
10.14

9.66

9.30

8.85



8.59

8.20

8.86

8.34

8.40

8.59 a

10.47 b

9.92







10.05

9.64



9.95

9.46

9.17


a values for hemihydrate; values for 0.25 hydrate

evaporator to yield compounds 4a-4k as colorless solids. Extraction of the free base

from the corresponding hydrochlorides was also attempted by treatment with a weaker









base, 1 equiv triethylamine, in cold methylene chloride (5 10 min) and subsequent

trituration with diethyl ether. But the desired prodrugs could not be obtained in very high

yields or purity. Further, column purification was not possible for these compounds

because of highly basic nature of the prodrugs: decomposition on silica gel to parent

drug was observed.

5a 4-Acetamidophenyl 2'-(dimethylamino)acetate, (Me2ni-APAP).1H-NMR

(CDCI3): 5 7.49 (d, J = 8.8, 2H), 7.04 (d, J = 9.2, 2H), 3.43 (s, 2H), 2.44 (s, 6H), 2.16 (s

3H). Yield: 95%.

5b 4-Acetamidophenyl 3'-(dimethylamino)propanoate, (Me2n2-APAP). NMR

(DMSO): 5 7.57 (d, J = 8.8, 2H), 7.01 (d, J = 9.2, 2H), 2.67 (t, 2H), 2.59 (t, 2H), 2.17 (s,

6H), 2.03 (s, 3H).Yield: 95%.

5c 4-Acetamidophenyl 2'-(diethylamino)acetate, (Et2ni-APAP). 1H-NMR

(CDCl3): 5 7.48 (d, 2H), 7.00 (d, 2H), 3.56 (s, 2H), 2.73 (quart, 4H), 2.09 (s, 3H), 1.10 (t,

6H).Yield: 70%.

5d 4-Acetamidophenyl 3'-(diethylamino)propanoate, (Et2n2-APAP). 1H-NMR

(CDCl3): 5 7.48 (d, J = 8.8, 2H), 7.01 (d, J = 9.2 2H), 2.95 (t, 2H), 32.74 (t, 2H), 2.63

(quart, 4H), 2.14 (s, 3H), 1.09 (t, 6H).Yield: 78%.

5e 4-Acetamidophenyl 2'-(dipropylamino)acetate, (Pr2ni-APAP). 1H-NMR

(CDCl3): 5 7.51 (d, 2H), 7.06 (d, 2H), 3.95 (broad s, 2H), 3.00 (broad s, 4H), 2.32 (s,

3H), 1.80 (broad s, 4H), 1.09 (t, 6H). Yield: 61%.

5f 4-Acetamidophenyl 2'-(piperidin-1"-yl)acetate, (PIPni-APAP). 1H-NMR

(CDCl3): 5 7.49 (d, J = 8.8, 2H), 7.04 (d, J = 8.8, 2H), 3.44 (s, 2H), 2.61 (t, 4H), 2.17 (s,

3H), 1.67-1.62 (m, 6H), 1.45 (broad d, 2H). Yield: 71%.









5g 4-Acetamidophenyl 3'-(piperidin-l"-yl)propanoate, (PIPn2-APAP). 1H-NMR

(DMSO): 5 7.57 (d, J = 9.2, 2H), 7.01 (d, J = 8.8, 2H), 2.66 (t, 2H), 2.62 (t, 2H), 2.37

(broad s, 4H), 2.035 (s, 3H), 1.49 (quin, 4H), 1.40 (broad d, 2H).Yield: 75%

5h 4-Acetamidophenyl 4'-(piperidin-1"-yl)butanoate, (PIPn3-APAP). 1H-NMR

(CDCI3): 5 7.50 (d, 2H), 7.03 (d, J = 8.0, 2H), 2.61 (broad t, 6H), 2.17 (s, 3H), 2.10

(broad s, 2H), 1.79 (broad s, 4H).Yield: 85%.

5i 4-Acetamidophenyl 2'-(morpholin-4"-yl)acetate, (MORni-APAP). H-NMR

(CDCI3): 5 7.50 (d, J=8.8Hz, 2H), 7.04 (d, J = 8.8 Hz, 2H), 3.78 (t, 4H), 3.47 (s, 2H),

2.68 (t, 4H), 2.17(s, 3H).Yield: 70%.

5j 4-Acetamidophenyl 3'-(morpholin-4"-yl)propanoate, (MORn2-APAP). 1H-

NMR (DMSO): 5 7.56 (d, 2H), 7.01 (d, J = 8.8 Hz, 2H), 3.56 (t, 4H), 2.54 (t, 2H), 2.53 (t,

2H), 2.40(broad s, 4H), 2.02 (s, 3H).Yield: 80%.

5k 4-Acetamidophenyl 4'-(morpholin-4"-yl)butanoate, (MORn3-APAP). 1H-

NMR (DMSO) 5: 7.57 (d, 8.4 Hz, 2H), 7.03 (d, J = 8.8 Hz, 2H), 3.55 (t, 4H), 2.57 (t, 2H),

2.32 (t, 6H), 2.03 (s, 3H), 1.89 (quin, 2H).Yield: 61%.

Hydrolysis of DAAC-APAP Prodrugs

The hydrolysis experiments were carried out on compounds 4a-4e (DAAC-APAP-

HCI). For determining the half lives of the members of DAAC prodrugs a solution of the

prodrug was prepared by dissolving a known amount of the prodrug in deionized water.

An aliquot of the aqueous solution was diluted with buffer (pH 6.0, phosphate, 20 mM)

to give concentrations of about 10-20 mM. The diluted solution was added into a UV

cuvette and absorbance values were recorded over 7-8 half-lives. The cuvette was

maintained at 37 0.5 C throughout the experiment. Although, the prodrugs showed

an absorbance maximum very close to that of APAP the molar absorptivities of the









prodrugs are higher than that of the parent drug, APAP, so the hydrolysis of the

prodrugs was monitored by the decrease in the absorbance values over time. The

absorbance values were recorded at 242 nm. All experiments were allowed to reach

Ao, where Ao is the absorbance at the completion of the reaction, at which time the

absorbance would reflect the concentration of APAP only. The pseudo unimolecular

rate constants were obtained from a plot of log (At A.) versus time where At is the

absorbance at time t. All the experiments were run in triplicate. The pKa values were

estimated by using the pKa prediction software (ACD Labs Inc. Version 10.0).

Table 3-2. Half Lives (t11/2, min) and predicted pKa Values of DAAC-APAP-HCI, prodrugs
in Buffer (pH 6.0, phosphate, 20 mM, I = 0.5) at 37 0.5 C
pKa t1/2
4a, Me2nlAPAP-HCI 72 24
6.94 76.92 + 0.24

4b, Me2n2APAP-HCI 8.66 113.03 2.85

4c, Et2nlAPAP-HCI 8.08 105.71 4.45

4d, Et2-n2APAP-HCI 9.59 79.32 11.92

4e, Pr2nlAPAP-HCI 9.40 105.89 3.24

4f, PIPnlAPAP-HCI 6.98 --

4g, PIPn2APAP-HCI 8.65 79.67 1.25


Table 3-2 shows the t1/2 values for the DAAC-APAP-HCI ester prodrugs. The

presence of an amine group in the acyl side chain of a DAAC prodrug predisposes the

ester group towards hydrolysis because of two reasons: (a) electron withdrawing effect

of the -NR1 R2 group and (b) general base catalysis by the -NR1R2 or -+NHR1R2 group.

In a detailed study 57 by Kirby et. al., rate enhancements up to 105 fold for the

hydrolysis of phenyl 3-dimethyl-aminopropionate vs phenyl acetate at pH > 9, where the

NMe2 group is present as a free base, was observed. This observation supports the









idea of facilitation of rate enhancement by the NMe2 group in addition to simple

electronic effects. Additionally, observation that the rate of hydrolysis of phenyl acetate

is 10 times slower than that for phenyl 3-dimethylaminopropionate at pH conditions

where the dimethylamine group is protonated indicates a rate acceleration by the+NMe2

group. However the propensity of a protonated amine group to facilitate intramolecular

catalysis is much lower compared to a free amine group.

Figure 3-3 shows pathways available to the neighboring amine group for the

facilitation of ester hydrolysis.68 These pathways include (A) nucleophilic catalysis, (B)

intramolecular general base catalysis and (C) intramolecular general acid specific base

catalysis. Mechanisms A, B and C are kinetically indistinguishable. For a prodrug

containing a basic amine group (pKa ~ 7-10) the following are true:43

* At pH << pKa the primary reaction is the hydrolysis of the protonated ester
(pathway C).

* At pH > 5 < pKa the primary reactions are the hydrolysis of free base and the
protonated ester (pathways A, B and C).

* At pH > pKa the primary reaction is the hydrolysis of the free base (pathways A
and B).

For compounds (except 4a) in the present study rate acceleration by +NRR2 group

is the primary effect observed because the compounds are expected to be protonated

(~ 99%) at pH 6.0. The t1/2 values of 4c and 4d (same amine but different n) can be

explained by a higher degree of general base catalysis by the more basic

diethylaminopropionate group compared to diethylglycinate group.

A similar trend between 4a and 4b is not observed because the calculated pKa

value of 4a is much lower than that of 4c so a higher percentage of free base is present

at pH 6.0. The trend in t1/2 values amongst compounds with same n (4a, 4c and 4e) but









different steric bulk on the amine group can be explained using Newman's Rlue of Six

which states that those atoms which are separated from the attacking atom in the

transition state by a chain of four atoms are the most effective in providing steric

hindrance.69

0 0 0

(A) C OR C-OR C-OH
) OH
S -OR- -

II


(B) 0o (C)
C-OR O


N- H
SN-

Figure 3-3. Possible routes of hydrolysis of a DAAC-APAP prodrug

For example for a DAAC acyl group, the "blocking atom" would be the atom (or

group) in the 6th position. The atom is the 6th position is more likely to be in the path of

the attacking atom than the 5th or 7th atom. (Figure 3-4)

Figure 3-4 shows that the 6th atom in 4a could be 6 H whereas in 4c the 6th or the

blocking atom could be 4 H and 2 methyl groups. Overall, the DAAC-APAP-HCI

prodrugs hydrolyze to the parent drug via a combination of nucleophilic and

intramolecular general base catalysis pathways. Although a pH rate profile can give

more detailed information about the hydrolysis behavior of prodrugs containing a basic

amine group, the data presented in this study exemplifies the transient behavior of the

DAAC-APAP-HCI prodrugs. In terms of prodrug design the DAAC prodrug approach









allows for modulation of half lives of the prodrugs by controlling the alkyl chain length

and steric bulk on the basic amine nitrogen.





0 N-



R











R




Figure 3-4. Origin of steric hindrance in 4a and 4c based on Newman Rule of Six.

Determination of Solubilities of DAAC-APAP Prodrugs

As mentioned before (Chapter 1), the permeation of a topically applied drug is a

"solubility" driven process. Determination of the solubility of a permeant in aqueous (or

aqueous-like) and lipid solvents allows for an analysis of the solubility and flux data, that

can be used to optimize topical delivery of drugs.

The molar absorptivity of DAAC-APAP-HCI prodrugs was determined in triplicate

at 244 nm in methanol and at 241 nm in buffer (pH 4.0, acetate, 50 mM). A known

amount of DAAC-APAP-HCI prodrug was dissolved in deionized water, and the solution










was diluted with methanol or buffer and analyzed by UV spectrophotometry. With the

known concentration C, E 4.0 or E MeOH was calculated with Beer's law:

A244 = E 244 1 C, where I = cell length (1)

The molar absorptivity of DAAC-APAP prodrugs, 5a-5k, (free bases) was determined in

acetonitrile (E MeOH ) by the method similar to the one described above. The molar

absorptivities have been shown in Table 3-3.

Table 3-3. Molar absorptivities of DAAC-APAP-HCI prodrugs in methanol (E MeOH),
acetate buffer (E 4.0) and molar absorptivities of DAAC-APAP prodrugs in
acetonitrile (EACN).


Compound

4a, Me2nlHCI

4b, Me2n2HCI

4c, Et2nlHCI

4d, Et2n2HCI

4f, PIPnlHCI

4g, PIPn2HCI

4h, PIPn3HCI

4i, MORnlHCI

4j, MORn2HCI

4k, MORn3HCI


APAP


E MeOHab



1.58 0.009

1.61 0.063





1.59 0.09

1.59 0.05

1.58 0.05

1.58 0.017

1.29 0.03

1.33 0.09


a,b,d
S40



1.22 0.015

1.24 0.071

1.29 0.008

1.30 0.02

1.31 0.018

1.32 0.02

1.29 0.02

1.29 0.02



1.023 0.013


Compound

5a, Me2n1HCI

5b, Me2n2HCI

5c, Et2n1HCI

5d, Et2n2HCI

5f, PIPniHCI

5g, PIPn2HCI

5h, PIPn3HCI

5i, MORniHCI

5j, MORn2HCI

5k, MORn3HCI


APAP


E ACNab

1.68 0.11



1.68 0.066

1.65 0.015

1.73 0.014

1.74 0.06

1.78 0.03

1.66 0.017

1.71 0.045

1.67 0.067

1.36e


a All experiments were run in triplicate. Units of 104 mlmole1 c UV max = 244 nm. UV max = 241 nm.
e Value from Wasdo et. al. 2004

The solubilities of DAAC-APAP-HCI prodrugs (4a 4k) were determined in PG

(aqueous-like) and 1-octanol. The solubilities of DAAC-APAP (5a 5k) prodrugs were

determined in isopropyl myristate (IPM) and buffer (pH 4.0 acetate 50 mM). The

procedure for solubility determination is described below.









For each prodrug, the solubility in 1-octanol (OCT) was determined in triplicate.

The prodrug was crushed into a fine powder and a saturated solution of the prodrug in

octanol was obtained by adding an excess of each compound to a test tube containing

1 ml octanol. The test tube was then insulated and the suspension was allowed to stir at

room temperature (23 1C) overnight (12 -14 h) on a magnetic stir plate. The

suspension was filtered through a 0.25 pm nylon syringe filter. An aliquot of the filtrate

was diluted with methanol and analyzed by UV spectrophotometry. The absorbance at

244 nm was used in each case to calculate the concentration of the prodrug in octanol,

which is the solubility in octanol (SOCT):

Csaturation = SOCT = A244 / E244 (2)

Solubilities in propylene glycol (PG), pH 4.0 acetate buffer and IPM were also

determined in triplicate by the procedure used to determine SOCT with the following

changes. The prodrugs were only stirred for 1h in pH 4.0 acetate buffer before filtration.

For determining SPG a sample of the filtrate was diluted with buffer (pH 4.0 acetate, 50

mM) and analyzed by UV spectrophotometry. For SIPM the suspensions were stirred for

2 4 h before filtration. For determining SAQ and SIPM the sample of the filtrate was

diluted with acetonitrile and analyzed by UV spectrophotometry. A 1H-NMR of the

filtered solid in the solubility experiments were recorded to ensure that the prodrugs

were intact through the course of experiment. None of the prodrugs hydrolysed to APAP

during the course of the experiment.

Partition coefficients were also determined in triplicate for the prodrugs by using

the saturated IPM solutions obtained from the solubility determinations. The saturated

IPM solution was partitioned against pH 4.0 buffer using the following volume ratios









(V4.0 / VIPM) for compounds 4a, 4b, 4f, 4g, 4j and 4k: 0.1, 0.2, 0.1, 0.05, 0.1 and 0.1

respectively. The two phases were vigorously shaken for 10 seconds, then allowed to

separate via centrifugation. An aliquot of the IPM layer was removed, diluted with

acetonitrile, and analyzed by UV spectrophotometry as described above. Using the

previously measured absorbance at 244 nm for the saturated solution, the partition

coefficient was calculated as follows:

KIPM:4.0 = [Aa/(Ab Aa)]V4.0nIPM (3)

where Ab and Aa are the respective absorbances before and after partitioning, and V4.0

and VIPM are the respective volumes of buffer and IPM in each phase.

Table 3-4. Physicochemical Properties of DAAC-APAP-HCI Prodrugs. Molecular
weights (MW), Solubility in 1-Octanol (SOCT), Solubility in Propylene Glycol
(SPG) and Estimated IPM solubilities (SIPM,est )of DAAC-APAP-HCI Prodrugs.
Compound MW SocTa SPGa SIPM,est

4a, Me2nlbHCI 272 0.352 0.01 122.05 2.90 0.004

4b, Me2n2HCI 286 0.291 0.035 150.09 2.03 0.003

4c, Et2nlHCI 300 1.297 0.067 93.80 + 0.98 0.027

4d, Et2n2HCI 314 1.162 0.055 176.10 3.42 0.018

4f, PIPnHCI 313

4g, PIPn2HCI 327 0.202 0.02 18.17 1.05 0.004

4h, PIPn3HCI 341 20.87 0.47 64.09 1.17 1.762

4i, MORnjHCI 315 0.09 0.01 14.01 0.01 0.001

4j, MORn2HCI 325 0.10 + 0.007 29.97 0.58 0.001

4k, MORn3HCI 339 1.522 0.077 39.68 0.50 0.05
APAP 151 158.94c 662.25c 1.90d
a Units of mM. bNumber of methylene groups between carbonyl carbon and amine. c Values from
Kasting, Smith and Cooper (1987, Ref. 34). dValue from Wasdo et. al. (2004, Ref. 53) was used as the
value for SIPM for APAP (2.38 mM) obtained from Kasting Simth Cooper (Ref. 34) was slightly different.









The IPM solubility of the hydrochloride prodrugs was estimated from the data of

Kasting Smith and Cooper34 by the method described below. A linear equation between

log KPG:IPM and log KPG:OCT was developed from the solubility database for n = 28

compounds. The value for SIPM,est was calculated using the corresponding SPG and the

linear equation derived above.

All the DAAC-APAP-HCI prodrugs exhibited higher melting points (Table 3-1) than

APAP which is expected because of introduction of the ionized amine group into the

prodrug. Compounds 4k and 4h exhibited decomposition (charring) close to the melting

temperature. DAAC-APAP-HCI prodrugs exhibited PG solubilities between 14 176 mM

in propylene glycol, the most PG soluble member being 4d. The trends in PG solubilities

for DAAC-APAP-HCI prodrugs were similar to the trend in aqueous solubilities (S4.0,

Table 3-5) for free bases. The member exhibiting the highest S4.0 also exhibited the

highest SPG (4d). Among the DAAC-APAP-HCI series, members containing the cyclic

amines (morpholine and piperidine, 4f 4k) in the promoiety exhibited a comparatively

lower SPG. The most PG soluble member among 4f 4k was 4h (~64 mM) compared to

4d (176 mM). The lower aqueous and PG solubility amongst morpholinyl and piperidinyl

prodrugs might be attributed to the rigid structure of the six membered ring. Among the

prodrugs containing the same amine the solubility in PG increased with an increase in

basicity of the nitrogen (Table 3-5). For example, 4a has a SPG of 122 mM which

increases to 150 mM for 4b. Similar trend is observed for 4f 4h and 4i 4k.

Similarly, amongst prodrugs containing same amine,S4.0 increased with increasing

distance from the acyl group, except for 5f-5h where the maximum aqueous solubility is

observed for the n = 2 prodrug, 5g (120 mM) and the effect of further addition of a









methylene group decreased the S4.0 for 4g to 109 mM. The trends in aqueous solubility

of the prodrugs is discussed in detail further in the section.

None of the prodrugs exhibited higher SOCT than APAP. Low lipid solubility is

expected from a hydrochloride salt. 4h exhibited the highest SOCT among the DAAC-

APAP-HCI series. Similarly, the IPM solubilities (estimated) of the prodrugs was also

generally low.

Table 3-5. Physicochemical properties of DAAC-APAP podrugs. Estimated pKa values,
solubility in pH 4.0 acetate buffer (S4.o), solubility in isopropyl myristate (SIPM),
estimated intrinsinc solubilities (Ss.5,est) and partition coefficients between IPM
and pH 4.0 acetate buffer for DAAC-APAP (free base) prodrugs.

Compound pKa SIPMa S4.0a S5.5,esta KIPM:4.0

5a, Me2nb1 6.94 15.23 + 1.43 264.50 1.40 9.26 0.055 0.007

5b, Me2n2 8.66 42.26 1.42c 897.42d -- 0.047 0.003

5c, Et2nl 8.08 91.22 0.66 116.49 0.86 0.30

5d, Et2n2 9.59 58.72 1.27 2884.03d -- 0.021 0.003

5f, PIPni 6.98 7.41 0.19 46.79 1.65 1.50 0.127 0.047

5g, PIPn2 8.65 3.75 0.07 120.89 7.47 0.85 0.13 0.005

5h, PIPn3 9.32 13.06 0.11 109.46 2.09 0.016 0.033 0.004

5i, MORni 4.65 2.52 0.05 50.51 0.55 44.25 0.048 0.037

5j, MORn2 6.42 1.25 0.02 75.93 + 1.52 8.14 0.022 0.08

5k, MORn3 7.14 12.26 0.49 380.55 1.73 8.52 0.034 0.002

APAP -- 1.90 100 -- -1.721
a Units of mM. bNumber of methylene groups between carbonyl carbon and amine. c Molar Absorptivity
of 5a was used to calculate SIPM. d Estimated from log S40 = log SIPM log K.

As shown in Table 3-5, all the members (except 5j) of the free base DAAC

prodrugs of APAP exhibited higher IPM solubility than APAP. 5c, the most lipid soluble

member of the series, was about 45 times more soluble in IPM than APAP. Although in









the DAAC prodrug series of APAP, the number of compounds having the same amine

group in the promoiety is small, (maximum three) somewhat of a trend in SIPM values

could be observed. Among members of the series having the same amine group, with

an exception of 5a and 5b, an increase in distance from the acyl group caused an initial

decrease in the SIPM with a subsequent increase in SIPM values. For example,

compound 5i showed only a moderate increase in SIPM and 5j was even less IPM

soluble than APAP. However, 5k showed about 6 times more SIPM than APAP. The

same trend in SIPM was observed among compounds 5f, 5g and 5h and between 5c and

5d. The SIPM of piperidinyl compounds (5f-5h) was higher compared to the morpholinyl

compounds (5i 5k). The higher SIPM of the DAAC prodrugs compared to APAP is

expected because of the masked polar -OH group. The trend in IPM solubilites can be

explained on the basis of melting points of the prodrugs. Figure 3-5 shows a plot of SIPM

vs melting points of DAAC-APAP prodrugs. The lower melting compounds showed a

higher SIPM.

Among the members with the same amine group the initial decrease in SIPM of the

member having two carbon atoms (n = 2) between the acyl and the amine group (5d, 5g

and 5j) compared with the n = 1 member (5c, 5f and 5i) can be attributed to a

concurrent increase in the pKa (hence basicity) of the nitrogen. Further increase in the

alkyl chain offsets the increase in the pKa and the SIPM values begin to rise again. For

example, in going from 5f (n = 1) to 5h (n = 3) the increase in the alkyl chain length

results in almost 3-4 times greater SIPM than the previous member. The trend in the pKa

values can be explained in terms of inductive effect of the carbonyl group. The increase

in the distance between the carbonyl carbon and the amine group makes the amine










more basic. For example, the pKa of ethyl N, N-dimethylaminobutyrate was very close

to dimethyl amine indicating a weak inductive effect over three carbon atoms.57

120
*

100

80 -

0 60

S40

20
*
0 S--------- -- -
0 50 100 150

Melting Point
Figure 3-5. Plot of IPM solubilities vs melting points for DAAC-APAP prodrugs

S4.0 of the DAAC series of prodrugs also showed a dependence on the pKa on the

amine in the prodrug. All members except 5f, 5i and 5j were more water soluble than

APAP. Amongst the morpholino compounds, the most water-soluble member was also

the most basic. Also a constant increase in SAQ was observed with an increasing

distance of the amine group from the carbonyl group (hence an increasing pKa).

However, the solubility among the piperidinyl compounds initially increased from 46.79

mM (5f) to 120.89 mM (5g) then decreased to 109.46 mM (5h). Even though the pKa of

the piperidinyl compounds was comparable to the 5a 5d, the SAQ was comparatively

lower. This may be explained by the trends in melting points. The more water soluble

members, 5a 5d, showed lower melting points compared to morpholinyl and piperidinyl

DAAC-APAP prodrugs.

The pH dependent solubility of a basic amine can be estimated using the method

used by the Bergstrom et. al. 70 where Stot represents the solubility of the total species









present in ionized state and Sins represents the intrinsic solubility of the amine at a

particular pH. The intrinsic solubilities of the prodrugs have been shown in Table 3-5.

For calculating Sins pH of the skin was considered to be 5.5.5 The value of the intrinsic

solubility of the least basic prodrug, 5i, was closest to its experimental solubility. Also

the higher the difference between the pKa of the permeant and the pH of the skin, the

higher is the difference between its intrinsic solubility and experimental solubility.

Overall, the solubilites of DAAC-APAP prodrugs in aqueous and lipid solvents are

governed by the basicity of and melting points of the corresponding prodrugs.

In-Vitro Flux Determination of DAAC-APAP Prodrugs

Three different mice were used to determine the flux of each prodrug. Prior to skin

removal, the mice were rendered unconscious by 002 then sacrificed via cervical

dislocation. Skins were removed by blunt dissection and placed dermal side down in

contact with pH 7.1 phosphate buffer (0.05 M, I = 0.11 M, 32 C) containing 0.11%

formaldehyde which has been shown to inhibit microbial growth and maintain the

integrity of the skins throughout the experiment.21 Prior to the application of the

prodrugs as suspensions in IPM the skins were maintained in contact with the buffer for

48 h to leach out all UV absorbing material. During this conditioning phase time, the

receptor phase was removed and replaced with buffer 3 times. The suspension of the

prodrug in IPM was prepared 4 h before application and allowed to stir at room

temperature (25 1 C) until application. The concentration of the donor phase

suspension was approximately 10 x SIPM. After the 48 hour leaching period, an aliquot

(0.5 ml) of the prodrug suspension was added to the surface of the skin (donor phase).

Samples of the receptor phase were usually taken at 8, 19, 22, 25, 28, 31, 34, and 48 h

and quickly analyzed by UV spectrophotometry. Since only parent drug was obtained in









the receptor phase in all cases the amounts of permeated APAP was quantified using

molar absorptivity of APAP in the receptor phase. At each sampling time, the entire

receptor phase was replaced with fresh buffer in order to maintain sink conditions. After

the 48 h of the first application period, the donor suspension was removed and the skins

were washed three times with methanol (3-5 ml) to remove any residual prodrug from

the surface of the skin. The remaining prodrug or APAP in the skin was leached out by

keeping the skins in contact with buffer for an additional 24 h. Then the receptor phase

was replaced with fresh buffer and an aliquot (0.5 ml) of a standard drug theophylline

was applied in PG to the skin surface. The second application fluxes were determined

by sampling of the receptor phase at 2, 4, 6 h and analysis using UV

spectrophotometry. The concentration of theophylline in the receptor phase was

determined by measuring its absorbance at 270 nm (E = 10,200 L mol-1). At each

sampling time, the entire receptor phase was removed and replaced with fresh buffer. In

each experiment, the flux was determined by plotting the cumulative amount of APAP

versus time. J in units of pmol cm-2 h could then be calculated by dividing the slope of

the steady-state portion of the graph (19 34 h, Figure 3-5) by the surface area of the

skin (4.9 cm2).

In-Vitro Evaluation of Morpholinyl and Piperidinyl DAAC-APAP Prodrugs.

The steady state flux values for DAAC-APAP prodrugs from a saturated IPM

vehicle through hairless mouse skin (EXP JM) are shown in Table 3-6. The steady state

flux values for a standard drug (Theophylline, Th) from a saturated PG donor phase (Js)

and the residual concentration of APAP in the skin has also been shown in the table.

The concentration of the APAP delivered locally to the skin was calculated from the

absorbance value of APAP in the donor phase after the 24 h leaching period.










80.000

70.000

60.000

50.000
)D
o 40.000
E
o0
2 30.000
E
t 20.000
E
E 10.000

0.000
0 10 20 30
Time (hrs)


Figure 3-5. Plot of cumulative amount vs time for calculating steady state flux

All compounds showed higher flux than APAP. Among the morpholinyl DAAC-

APAP prodrugs, the best member was MORni-APAP, 5i which gave about two times

higher flux than APAP. Although MORn2 and MORn3 prodrugs (5j and 5k) showed

higher water solubility than 5i the flux of 5j was 0.80 pmole cm-2 h-1 (1.6 times that of

APAP) and 5k was 0.71 pmole cm-2 h-1 (1.4 times that of APAP). Based on solubility

properties this was unexpected because 5i showed only a moderate enhancement in

IPM solubility (1.3 times) than APAP and was less water soluble than 5j or 5k. The flux

values for MORn2-APAP and MORn3-APAP were lower than expected. A similar trend

was observed in the piperidinyl DAAC-APAP prodrugs. The PIPni prodrug showed the

highest flux, 0.78 pmole cm-2 h-1, (1.27 time that of APAP) amongst the PIPn1-PIPn3

compounds.

Further analysis of the trend in the flux values revealed that among DAAC

prodrugs containing the same amine there is linear correlation between pKa and J

values. The plot of pKa vs J for morpholinyl and piperidinyl DAAC-APAP prodrugs has

been shown in Figure 3-6.















1-
JMORu-APAP= -0.1376pKa+ 1.687 '
R2 = 0.99
0.9


% 0.8


0.7


0.6
JPIP-APAP = -0.065 p Ka + 1.22
R2 = 0.95

0.5
0 2 4 6 8 10
pKa


Figure 3-6. Plot of pKa vs J for morpholinyl and piperidinyl DAAC-APAP prodrugs.

Among the compounds studied there is a negative correlation between J values

and pKa values. Additionally comparison of MORn1 MORn3 with PIPnl PIPn3 shows

that the relatively more basic piperidinyl prodrugs showed lower flux than the

corresponding morpholinyl compounds.

The unexpected lower flux values for more basic n = 3 morpholinyl and piperidinyl

DAAC-APAP prodrugs can be explained by a possible increase in the effective

molecular weight caused by water association with the amine group as the prodrug

diffuses through the membrane.7173 Affsprung and coworkers studied the hydration

behavior of amines in organic solvents like benzene and chloroform and proposed that

at higher concentrations the amine molecules are present as bridged hydrates involving

two base molecules and one water molecule.71 In our study, such an association will

offset the effect of enhanced water and lipid solubility because of a negative correlation









between flux and MW. Additionally, studies on permeability of model lipid bilayer

membranes to organic amine solutes suggest that interfacial charge and hydration

structure in addition to solubility dependence may influence flux of ionizable solutes.74

Table 3-6.Results from diffusion cell experiments with morpholinyl and piperidinyl
DAAC-APAP prodrugs. Maximum flux of the prodrug from a saturated
isopropyl myristate (IPM) vehicle through hairless mouse skin (JM), maximum
flux of standard drug, theophylline, from a saturated propylene glycol (PG)
vehicle (Js), log of experimental flux values (EXP log JM), residual
concentration of the drug in the skin (Cs), absolute difference between EXP
log JM and flux values calculated from Robert-Sloan equation
Compound JMa Ja Csb,d EXP CALCC A log JM
log JM log JM
5f PIPni 0.78 0.11 0.66 0.35 1.20 0.45 -0.10 0.021 0.121

5g PIPn2 0.65 0.09 0.30 0.003 2.88 0.16 -0.18 0.045 0.225

5h PIPn3 0.64 0.05 0.1 0.03 1.52 0.29 -0.19 0.262 0.452

5i MORn1 1.05 0.17 1.35 0.65 -- 0.021 -0.203 0.224

5j MORn2 0.80 0.08 0.86 0.08 -- -0.10 -0.300 0.200

5f MORn3 0.71 0.06 0.38 0.09 0.52 0.10 -0.15 0.513 0.663

APAP 0.51 0.74 2.74 0.07 -0.19 0.262 0.472
a Units of pmole cm- h 1. Concentration of APAP in the skin after a 24 h leaching period. c Calculated
from the RS equation log CALC JM = -0.599 + 0.502 log SIPM + 0.498 log S40 0.00235 MW. d Units of
pmole.

Similar results were obtained with PEG prodrugs of APAP studied in our lab

previously. The most lipid and water soluble member of that series (SAQ = 184 mM, SIPM

= 12.14 mM) gave lower flux than expected from a highly water and lipid soluble

permeant (0.69 pmole cm-2 h-1).75 The ethylene oxide moiety has been shown to be

associated with water molecules in solution. This association of water molecules with

the ethylene oxide head group causes an increase in effective molecular weight of the

permeant as it diffuses through the skin.









The residual amount of APAP in the skin after the application period (Cs) has been

shown in Table 3-6. The PIPn2 prodrug could only deliver almost the same amount of

APAP as the parent drug itself while the skin retention of all the other prodrugs was less

than that of APAP.

Additionally, the Js controll2) values for 5g, 5h and 5f were lower than usual after

a permeation experiment with IPM (1.0 pmole cm-2 h-1). This may be attributed to an

excess of HCHO present in the batch of receptor phase buffer used for these three

compounds. No microbiology experiments were carried out to confirm this. However the

control values returned to ~ 1.0 pmole cm-2 h- when a new batch of buffer containing

0.11 % formaldehyde was used for permeation experiments with other members of the

series (shown in Table 3-7).

The Robert-Sloan (RS) Equation was developed as an approach for quantifying

the dependence of flux of a permeant through skin on its aqueous and lipid solubilites

and molecular weight. The first form of RS equation was published in 1999 and the

database consisted of 5-FU and 6-MP prodrugs including the parent drugs (n = 42).

Since then the Robert-Sloan database has been updated three times and various acyl

and soft alkyl prodrugs of APAP were added to the database. The RS equation

developed by Thomas and Sloan (2009)60 is shown below.

CALC log JM = -0.562 + 0.501 log SIPM + 0.499 log SAQ 0.00248 MW (4)

However equation 4 did not include 2 bis-1,3-alkylcarbonyl 5 FU prodrugs.

Therefore a further updated database was developed comprising n = 73 prodrugs: 42

compounds from Roberts et. al. (1999),36 C5 bis-6,9-ACOM-6MP, 4 3-AC-5FU prodrugs

from Beal and Sloan ,63 6 3-ACOM 5 FU prodrugs from Roberts and Sloan ,76 8 AOC-









APAP prodrugs Wasdo and Sloan ,53 5 ACOM-APAP prodrugs from Thomas and Sloan

59 and 5 AOCOM-APAP prodrugs from Thomas et. al.60 SPSS 10.8 was used to

calculate the coefficients in the RS equation. The updated Robert-Sloan database gave

x = -0.599, y = 0.502, z = 0.00235 and an r2 = 0.92 (Equation 5). The y value in RS

equation shows that the flux is essentially equally dependent on aqueous and lipid

solubilities the permeant.

CALC log JM = -0.599 + 0.502 log SIPM + 0.498 log SAQ 0.00235 MW (5)

Equation 5 was then used to calculate the flux of the morpholinyl and piperidinyl

DAAC-APAP prodrugs (CALC log JM). The absolute difference between the EXP log JM

and CALC log JM have been shown in the Table 3-6. The average A log JM was 0.163.

The plot of EXP log JM VS CALC log JM has been shown in Figure 3-7.

The CALC log JM values for all but one of the prodrugs were higher than EXP log J

values. The one member of the DAAC-APAP series that performed better than that

predicted was MORn1 prodrug whereas the n = 3 prodrugs in both MOR and PIP series

showed much lower flux than expected and also showed the highest A log JM values.

This "over-prediction" of J values was observed due to two possible reasons.

The CALC log JM values were obtained using S4.o which are expected to be higher

than the solubility of the permeant in the skin microenvironment. When CALC log JM

was obtained using intrinsic solubilities at pH 5.5 from Table 3-5, lower calculated flux

values than experimental values were obtained (data not shown). In terms of topical

delivery of DAAC prodrugs, the pH of the skin microenviroment will govern the

percentage of ionized or unionized species in the first few layers of the skin.













1 -O *












-1 MORn-3-DAAC-APAP
4









PIPni-3-DAAC-APAP

-2

-20 -2 -15 -1 -05 0 05 1 15

CALC log JMMIPM = -0.599 + 0.502"1,y q SIpm + 0.498"1o9SAQ -0.00235 MW;
r f*. A







=73; = 0.920








Figure 3-7. Plot of EXP log JM vs CALC log JM.
Since the DAAC-APAP prodrugs are expected to be present in an equilibrium










between their ionized and unionized state,the solubility of the actual species in the skin

while the prodrug permeates the skin can be challenging to measure or predict.

The second reason for lower flux values than expected is the higher effective

molecular weight of the more basic prodrugs as they diffuse through the membrane.
-In conclusion among the morpholinyl and piperidinyl DAAC-APAP prodrugs

evaluated, MORn APAP prodrug was able to enhance delivery of APAP by two times.
-23

-25 -2 -15 -1 -05 0 05 1 15

CALC log JMMIPM =-0.599 *+L.WiuiI\ S|PM + 0.498O1ogSAQ 0.00235 MW;
n=73; r2=0.920

Figure 3-7. Plot of EXP log JM vs CALC log JM.

Since the DAAC-APAP prodrugs are expected to be present in an equilibrium









between though incorporationized and unionizable amine was able to enhance solubility properof the actual species in the skin

while the prodrugs, the inreates the skin can be challenging offset to measure flux enhancement due an
The second reason for lower flux values than expected is the higher effective

molecular weight of the more basic prodrugs as they diffuse through the membrane.

In conclusion among the morpholinyl and piperidinyl DAAC-APAP prodrugs

evaluated, MORni APAP prodrug was able to enhance delivery of APAP by two times.

Although incorporation of an ionizable amine was able to enhance solubility properties

of the prodrugs, the increased basicity, indirectly offset the flux enhancement due an

increase in effective molecular weight of the permeant due to hydrated forms in the skin

microenvironment.












In-Vitro Evaluation of Dimethyl and Diethyl DAAC-APAP Prodrugs and DAAC-
APAP HCI Prodrugs.


The results from the flux measurements of the dimethyl and diethyl DAAC-APAP


prodrugs and three DAAC-APAP-HCI prodrugs have been shown in Table 3-7. The RS


equation was used to calculate the flux of two DAAC-APAP prodrugs.


160

140 -

120

E 100 1
o I
80

Fr
E 60
5 ,..--'
40

20 4a
--A-- 5a
0
0 10 20 30 40
Time (hrs)

120


-100


Sso
E 6



E 5h
E
S40


20 9 4h



S10 20 30 40

Time (hrs)

Figure 3-8. Cumulative amount of drug permeated vs time for MORnl-APAP, MORni-
APAP-HCI and Me2n1-APAP and Me2ni-APAP-HCI


Both DAAC-APAP members that were evaluated performed better than the best


member in the morpholinyl and piperidinyl series. 5a and 5c both showed about three









times higher flux than APAP. The best member of the series was the Et2ni-APAP, which

also has the best balance in the lipid and water solubilities. Although 5c, the more lipid

and water soluble member of the series, showed the highest flux of the series, which

was expected, it did not show as high a flux as expected. 5c exhibited a SIPM of 91 mM

and a S4.0 of 116 mM where as 5a exhibited a SIPM of 15 mM and S4.0 of 264 mM. 5c is

the compound with the best balance between lipid and aqueous solubility. But 5c has a

pKa of 8.08 and 5a is less basic by about one pKa unit (6.98). As discussed in the

previous section the increase in basicity seems to offset the effect of increase in

solubility consequently leading to lower flux than expected.

Table 3-7.Results from diffusion cell experiments with dimethyl and diethyl DAAC-APAP
prodrugs and DAAC-APAP-HCI prodrugs. Maximum flux of the prodrug from
a saturated isopropyl myristate (IPM) vehicle through hairless mouse skin
(JM), maximum flux of standard drug, theophylline, from a saturated propylene
glycol (PG) vehicle (Js), log of experimental flux values (EXP log Js), residual
concentration of the drug in the skin (Cs), absolute difference between EXP
log Js and flux values calculated from Robert-Sloan equation
EXP CALCd
Compound JMa JSa CSa,c A log JM
log JM log JM
5a Me2nl 1.50 0.30 0.43 0.11 5.84 1.05 0.176 0.65 0.47

5c Et2nl 1.52 0.13 0.73 0.07 1.66 0.76 0.181 0.79 0.61

4a Me2nHHClb 0.64 0.06 1.00 0.02 0.69 0.07 -0.19 -- --

4b Me2n2HClb 0.65 0.01 0.95 0.03 0.69 0.05 -0.19

4i MORniHClb 0.54 0.02 0.96 0.25 0.71 0.18 -0.26

APAP 0.51 0.74 2.74 0.70 -0.29
a Units of pmole cm2 h 1. Suspension of the prodrug was applied in a (99:1) PG:IPM vehicle.
c Concentration of APAP in the skin after a 24 h leaching period. d Calculated from the RS equation log
CALC JM = -0.599 + 0.502 log SIPM + 0.498 log S40 0.00235 MW.

Three DAAC-APAP-HCI prodrugs were also evaluated in diffusion cell

experiments. The Me2nlAPAP-HCI prodrug gave 1.25 times higher flux than APAP

whereas MORniAPAP-HCI prodrug showed about the same flux as APAP. The DAAC-










APAP (free bases) prodrugs, 5a and 5i, gave 2.3 and 2 times higher flux than the

corresponding hydrochloride salts. The plots for cumulative amount of drug permeated

vs time for free base and corresponding hydrochlorides are shown in Figure 3-8.

Another difference between the permeation behavior of free base and hydrochloride

DAAC prodrugs is the fact that the salts showed a longer lag time to achieve steady

state flux (Figure 3-8). The skin retention of Me2ni prodrug was twice as high as APAP.

The skin retention of all the other members of the series was lower than APAP.


*
+ +






n=73prodrugs
0 MORnl-3-DAAC-APAP
SPIPnl-3-DAAC-APAP
+ Me2nlAPAP and Et2nAPAP


-25 -2 -1.5 -1 -05 0 0.5 1 1.5

CALC log JMMIPM = -0.599 +0.5i'-l:(j SIPM + 0.498"logSAQ 0.00235 MW;
n =73; 2 = 0.920


Figure 3-9. Plot of EXP log JM VS CALC log JM including dimethyl and diethyl DAAC-
APAP prodrugs

The second application flux values for all the compounds was close to the control

value of 1.0 pmole cm-2 h-1 which showed that the flux enhancement observed is not

because of the damage to the skin caused by IPM itself. The RS equation was used to









calculate the flux of 5a and 5c. As described before, the calculated flux of these

compounds was higher because the S4.0 might not precisely represent the actual

solubility of the permeant in the skin. The plot of EXP log JM VS CALC log JM has been

shown in Figure 3-9.

Conclusions

The potential of an ionizable amine in modulating solubility and permeation

properties of an acyl prodrug has not been studied or reported widely. In this study a

series of N, N'- dialkylaminoalkylcarbonyl (DAAC) acyl prodrugs of APAP as a model

phenolic drug were synthesized. The choice of amine in the promoiety was guided by

two factors: pKa of the amine and steric bulk on the amine. The prodrugs were

synthesized via a three step procedure in good yields.

The half-lives of the members evaluated show a dependence on the pKa of the

amine in the molecule and undergo hydrolysis via a combination of intramolecular

general base catalysis and nucleophilic catalysis by the basic amine group. The t1/2

values obtained in this study exemplify the transient nature of the DAAC prodrugs.

The octanol solubilities of the hydrochloride salts was generally lower than APAP

itself which is expected from an ionized compound. The IPM solubilites of all the free

base prodrugs except one was higher than APAP. With a few exceptions almost all the

prodrugs exhibited higher S4.0 than APAP. Among the compounds having the same

amine in the promoiety SAQ is dependent on the pKa of the amine in the prodrugs. The

effect of moving the amine group away from the carbonyl carbon affects the pKa values

of the prodrugs which consequently affects the solubility behavior of the prodrug. The

solubility of the prodrugs in a propylene glycol showed the same trend as water

solubility. The most water soluble member of the DAAC-APAP-HCI series exhibited the









highest SPG as well. The IPM solubilities of the DAAC-APAP prodrugs seemed to be

correlated to their melting points where the lower melting members of the series

exhibited a higher SIPM.

Although one of the prodrugs showed upto three times higher flux than APAP, the

delivery of APAP was not as high as expected. The best member of the series was also

the one that has the best balance in lipid and water solubility. We hypothesize that the

lower flux values may be due to increase in the molecular weight caused by water

association with the amine group as the prodrug diffuses the membrane.









CHAPTER 4
AMINOALKYLCARBONYL (AAC) PRODRUGS OF PHENOLIC DRUGS
ACETAMINOPHEN AND NALTREXONE

Introduction

A prodrug strategy involving modification of the parent drug, which contains

hydroxyl groups (alcoholic or phenolic), using amino acids as a promoiety has been

utilized successfully to improve potency of orally delivered drugs. Although incorporation

of a basic amine group in the prodrug facilitates its use as a water soluble salt,

achieving optimum stability of the prodrugs is challenging. One of the amino acids that

can improve the hydrophilic/lipophilic balance as well as the stability of the prodrugs is

L-valine. The -NH2 group improves water solubility and the bulky isopropyl group in the

side chain can help improve stability at physiological pH. For example, the commercially

available drug Valacyclovir Hydrochloride (Zovirax TM,GlaxoSmithKline), a valinate ester

prodrug of acyclovir (Figure 4-1), has a t1/2 = 13 h (pH 7.4, 370C), SAQ = 174 mg ml-1 and

is rapidly converted to the parent drug in vivo.46



N


0,N 0- j

NMH2


Valacyclovir Valine-5-OH-DPAT
Zovirax R, Glaxo Smith Kline Valine-5-OH-DPAT

Figure 4-1. Chemical structures of commercially available drug Valacyclovir and valine
ester prodrug of 5-OH-DPAT by Bouwstra and coworkers (Ref. 77)

There are only a few studies published about the utility of the unsubstituted amine

group in enhancing the topical delivery of a parent phenolic drug. Recently Bouwstra









and coworkers synthesized AAC prodrugs of the dopamine agonist 5-hydroxy-2-(N, N' -

propyl)-tetralin, 5-OH DPAT (Figure 4-1), which were targeted for transdermal

iontophoretic delivery. 77 The aqueous solubilities of the VAL-DPAT and 3 ALA-DPAT

were 4 times and 14 times higher than the parent drug, respectively. A higher in-vitro

iontophoretic transport was observed with 3 Ala-DPAT compared with 5-OH DPAT.

The L-proline, L-serine, L-tyrosine, L-asparagine, and L-citrulline ester prodrugs of

Vitamin E were synthesized and evaluated in skin retention experiments. Although skin

retention level of the prodrugs was found to be higher than Vitamin E, the flux of the

prodrugs was not significantly higher than Vitamin E. 58

Needham and coworkers78 investigated the transdermal delivery of the calcium

channel blocker Nicardipine Hydrochloride form different pure and blended solvent

systems. Propylene glycol (PG) was used as the primary vehicle for the development of

the transdermal product. Although several research groups have investigated the

potential of permeation enhancers20 79 and ion pairing techniques to improve the flux of

ionized compounds, a direct in-vitro transdermal evaluation of ionized prodrugs of

phenolic drugs from a saturated donor phase in a single solvent has not been reported

so far. Such an analysis would allow for optimization of the design directives that can be

used to achieve maximum improvement in flux.

One attractive target to which the AAC prodrug approach can be applied is

naltrexone (NTX). NTX is an opioid antagonist currently available as Vivitrol, Revia

and Depade for the treatment of alcohol and narcotic dependence.80-82 Revia and

Depade are administered as oral tablets whereas Vivitrol is a sustained release

formulation administered as a gluteal intramuscular injection. Orally delivered NTX has









been associated with gastrointestinal side effects in addition to plasma level

fluctuations. 83 Additionally the sustained release formulation requires frequent painful

intramuscular injections. One approach to overcome these challenges is to deliver the

drug transdermally. Treatment of alcohol and narcotic dependence require steady

delivery of drug to the body over extended period of time. Treatment of narcotic

dependence has met with a high failure rate due to compliance issues. Non-compliance

is usually followed by a relapse which is attributed to the elevated effects of heroin

following two or three days without the antagonist. The resulting "euphoria" is often an

incentive for the patient to indulge in further heroin use. 84 The non invasive nature of

transdermal drug delivery approach can facilitate sustained delivery and strict

adherence to dosing regimens. Transdermal delivery of drugs also circumvents the side

effects related to an orally delivered drug.

There is only one published report on the synthesis of amino acid esters of

APAP.67 No solubility or flux data for these AAC prodrugs of APAP has been published.

Also there are no reports of the synthesis of AAC NTX prodrugs. In order to evaluate

the ability of an unsubstituted ionizable amine group to enhance the solubility properties

and skin permeation behavior of the prodrugs, we synthesized three AAC prodrugs of

APAP and one AAC prodrug of NTX, characterized their solubility and stability

properties and evaluated them in diffusion cell experiments using hairless mouse skin.

Additionally, it is of interest to investigate the ability of the NH2- group to modulate

solubilities and flux of AAC prodrugs in comparison to the R1R2N- group in the DAAC

prodrug series.









0
(A) R1=H R4 R4 HC- Alan neT 7a
0 T 1 APAP R1= NH2 (CH3)2CH- (Valine) 7b
A^N (Proline) 7c
H AAC-APAP
Prodrugs 7a-7c N
H

(B) R50



HO H R5=

10
U^C a R5-H 25=NH, 2.TFA
10
Figure 4-2. Chemical Structures of AAC Prodrugs of APAP and NTX

Material and Methods. N-Boc-L-Amino Acids, 1 -ethyl-3-(3'-dimethylaminopropyl)

carbodiimide hydrochloride (EDCI) were purchased from Acros Organics.

Dicyclohexycarbodiimide (DCC) was purchased from Sigma Aldrich. Naltrexone

Hydrochloride (NTX-HCI) was purchased from MP Biomedicals LLC Naltrexone free

base (NTX) was obtained by treatment of NTX-HCI with aqueous NaHCO3 and

subsequent extraction with methylene chloride. Methylene chloride and pyridine were

dried over 3 A molecular sieves before use. Melting points were determined on a

Meltemp melting point apparatus. Spectra (1H NMR) were recorded on a Varian Unity

400 MHz spectrometer. UV spectra were recorded on a Shimadzu 2551 instrument.

Synthesis of AAC Prodrugs of Acetaminophen (APAP) And Naltrexone (NTX)

The AAC prodrugs of APAP and NTX were synthesized in two steps (Figure 4-3).

The N-Boc-amino acid was coupled with parent phenolic drug via a DCC or an EDCI

mediated reaction to give the Boc-AAC-APAP or Boc-AAC-NTX. The Boc protected

AAC-APAP prodrugs were treated with 2N HCI in diethyl ether to yield the AAC-APAP-

HCI prodrugs in good yields. The Boc protected AAC-NTX was treated with TFA to give










VAL-NTX-TFA. The synthetic details are discussed in the experimental section. The

results from the elemental analysis of the synthesized compounds are shown in Table

4.1.

NH2

HCI N
H
7a-7c (AAC-APAP)



BOC L N DRUG

APAP D cHN 0 H 0
NTX R4
6a Sc (Boc-AAC-APAP)
S(Boc-AAC-NTX) NH2


2 TFA 0



10 (AAC-NTX)

Figure 4-3. Synthesis of AAC prodrugs of APAP and NTX. Reaction conditions: Method
A: DCC/DMAP, r.t, 8-12 h; Method B: EDCI/CH2CI2, r.t, 6 h;

Experimental. When DCC was used as the coupling reagent to give the Boc

protected compounds purification and further characterization of these compounds was

a problem because of the presence of the DCC coupling by-product, dicyclohexyl urea

(DCU). Therefore, in order to obtain pure Boc protected AAC prodrugs, EDCI were

used as the coupling reagent. However, the yields of AAC-APAP-HCI prodrugs

synthesized by the DCC mediated coupling were relatively higher than that obtained

using EDCI mediated coupling. The relatively lower yield by the EDCI mediated

coupling may be due to the significant aqueous solubility of Boc-AAC-APAP prodrugs

leading to the loss of the prodrug during the water wash work up. For a poorly water

soluble drug, NTX, the yield of the isolated Boc-AAC prodrug was relatively high. The









relative yields of the compounds synthesized using DCC or EDCI has been shown in

Table 4-2.

General procedure for the synthesis of compounds (7a-7c). Method A

(DCC/DMAP/CH2CI2). Equimolar quantities of the APAP, Boc-amino acid, DCC and

catalytic amount of DMAP in 15-20 ml dry methylene chloride were stirred at room

temperature for 4-8 h. Completion of reaction was followed by TLC. The resulting

suspension was filtered and the solid was discarded. The filtrate was concentrated to an

oil on a rotary evaporator. The oil was resuspended in 20-25 ml methylene chloride and

the organic layer was washed twice with 10 ml 0.1 M cold aqueous HCI. The organic

layer was dried over Na2SO4 and concentrated to an oil. The oil was redissolved in 10

ml ether and 2-3 ml of 2 N HCI solution in diethyl ether was added dropwise to it with

constant stirring. The completion of deprotection step was followed by TLC. The

resulting suspension was triturated with 100-200 ml acetone:chloroform (1:10) and

filtered. Compounds 7a-7c were obtained as colorless solids.

Method B(EDCI/CH2CI2). Equimolar quantities of APAP or NTX, Boc-amino acid

and EDCI in 15-20 ml dry methylene chloride were stirred at room temperature for 12 h.

Completion of the reaction was followed by TLC. The resulting suspension was washed

with water (10 x 3 ml) and the organic layer was dried over sodium sulphate and

concentrated to an oil. The oil was purified by flash column chromatography to yield the

Boc protected compounds. The procedure for deprotection of the Boc group was the

same as described in Method A. For the NTX prodrug the deprotection was carried out

using triflouroacetic acid.









Table 4-1. Elemental Analysis and Melting Points for AAC.HCI Prodrugs of 4-
Hydroxyacetanilide (APAP) And AAC Prodrug of Naltrexone
C C H H N C
MpCal Exp Cal Exp Cal Exp
7a 200-240(d) 51.07 51.01 5.84 5.82 10.83 10.83
7b 230-250 (d) 54.45 54.51 6.68 6.65 9.77 9.74
7c 195-230 (d) 54.84 54.87 6.02 6.02 9.84 9.70
10 64-68 49.43 49.60a 5.44a 5.28 3.98a 4.01
a calculated values for the VAL-NTX-2TFA-2H20

6a N-Boc-ALA-APAP. Synthesized from 2.0 g (10.58 mmole) N-Boc-L-alanine,

2.03 g (10.58 mmole) EDCI and 1.6 g (10.58 mmole) APAP. NMR (CDCI3): 5 7.50 (d, J

= 8.8, 2H), 7.04 (d, J = 9.2, 2H), 5.05 (s, 1H), 4.50 (broad s, 1H), 2.17 (s, 3H), 1.46 (s,

9H). Yield: 60%.

6b N-Boc-VAL-APAP. Synthesized from 1.0 g (4.58 mmole) N-Boc-L-valine, 0.88

g (4.60 mmole) EDCI and 0.7 g (4.63 mmole) APAP. NMR (CDCI3): 5 7.47 (d, J = 8.8,

2H), 7.02 (d, J = 9.2, 2H), 5.02 (broad d, 1H), 4.52 (t, 1H), 2.6 (m 1H), 2.17 (s, 3H), 1.46

(s, 9H) 3H). Yield: 46% (average of two experiments).

6c N-Boc-PRO-APAP. Synthesized from 2 g (9.3 mmole) N-Boc-L-proline, 1.7 g

(9.27 mmole) EDCI and 1.4 g (9.30 mmole) APAP. NMR (CDCI3): 5 7.48 (d, J = 8.8,

2H), 7.03 (d, J = 9.2, 2H), 4.0 4.32 (dt, 1 H), 3.25-3.61 (m, 2H), 2.2-2.3 (m, 2H), 2.16 (s,

3H), 1.95-2.05 (m, 2H), 1.47 (broad s, 9H). Yield: 31%. Oil at rt.

7a ALA-APAP-HCI (Acetaminophen Alaninate Hydrochloride, ALA-APAP-

HCI). NMR (D20): 5 7.38 (d, J = 8.4, 2H), 7.04 (d, J = 8.0, 2H), 4.37 (q, 1H), 2.01 (s,

3H), 1.59 (d, 2H). Overall yield over two steps: Method A: 74%; Method B: 44%.

7b VAL-APAP-HCI (Acetaminophen Valinate Hydrochloride, VAL-APAP-HCI).

NMR (D20): 5 7.50 (d, J = 8.0, 2H), 7.18 (d, J = 8.1, 2H), 4.31 (d, 1H), 2.53 (sept, 1H),

2.14 (s, 3H), 1.14 (d, 6H). Overall yield over two steps: Method A: 69%; Method B: 57%.









7c PRO-APAP-HCI (Acetaminophen Prolinate Hydrochloride, PRO-APAP-

HCI). NMR (D20): 5 7.50 (d, J = 8.4, 2H), 7.22 (d, J = 8.2z, 2H), 4.79 (q, 1H), 3.50 (m

2H), 2.61 (m, 1H), 2.42 (m, 1H), 2.17 (t, 2H), 2.17 (s, 3H). Overall yield over two steps:

Method A: 84% Method B: 27%

Table 4-2: Percentage yields of the Boc-AAC-APAP compounds and the corresponding
AAC prodrugs synthesized from Method A or Method B
% Yield % Yield
Method A Method B
Boc-AAC-APAP
6a 60
6b -- 46a
6c 31a
AAC-APAP-HCI
7a 74 44
7b 69 57
7c 84 27
AAC-NTX
9 -- 52
10 90
a Average of two experiments

9 N-Boc-VAL-NTX. Synthesized from 0.37g (1.08 mmole) NTX, 0.21 g (1.09

mmole) EDCI and 0.23 g (1.08 mmole) L-Boc-valine-OH. NMR (CDCI3): 5 6.9 (d, J =

8.1, 1H), 6.68 (d, J = 8.2, 1H), 5.17 (d, 1H), 4.68 (s, 1H), 4.4 (d, 1H), 2.58 (sextet, 1H),

1.46 (s, 9H), 1.08 (quartet, 6H), 0.13-0.17 (quartet, 2H). Yield: 52%. Oil at room temp.

10 VAL-NTX-2TFA-2H20. NMR (D20): 5 7.09 (d, J = 8.0, 1 H), 6.98 (d, J = 8.2,

1H), 5.15 (s, 1H), 4.4 (d, 1H), 2.58 (sextet, 1H), 1.20 (quartet, 6H), 0.45-0.47 (m, 2H).


Hydrolysis of AAC prodrugs

For determining half lives of the members of AAC prodrugs a solution of the

prodrug was prepared by dissolving a small amount of the prodrug in deionized water to









give an approximate concentration of about 10-20 mM. An aliquot of the aqueous

solution was immediately diluted with buffer (pH 6.0, phosphate, 20 mM, I = 0.5). The

diluted solution was immediately added into a UV cuvette and absorbance values were

recorded till approximately >99% of hydrolysis was complete (7-8 half-lives). The

cuvette was maintained at 37 20C throughout the experiment. Since the molar

absorptivities of the prodrugs are higher than that of the parent drug, APAP, the

hydrolysis of the prodrugs was monitored by the decrease in the absorbance values

over time. Additionally, the prodrugs showed an absorbance maximum very close to

that of APAP. Therefore, the absorbance values were recorded at 244 nm. For the NTX

prodrug, the abrorbance was recorded at 278 nm. The pseudo unimolecular rate

constants were obtained from a plot of log (At Ao) versus time where At is the

absorbance at time t and Ao is the absorbance at the completion of the reaction, at

which time the absorbance would reflect the concentration of APAP only. All the

experiments were run in triplicate.

Table 4-3. Half lives (t11/2, min) and predicted pKa values of AAC prodrugs of APAP and
naltrexone in buffer (pH 6.0, phosphate, 20 mM, I = 0.5) at 37 0.5 C
pKa t11/2 (min) a
7a 8.62 32.33 0.28
7b 7.46 91.37 3.45
7c 8.65 14.90 + 0.81
10 -- 31.78 1.52
a All experiments were run in triplicate.

The AAC-APAP prodrugs showed shorter half lives than the DAAC-APAP

prodrugs. This is an expected result because of the greater ease with which the

unsubstituted amine group, -NH2, can facilitate the general base catalysis of the

hydrolysis of these compounds or a nucleophilic attack to give the diketopiperazine (see









chapter 3 for more detailed explanation). Among the AAC-APAP-HCI prodrugs, steric

effect exerted by the R4 group (Figure 4-1) seemed to play an important role in

governing the t1/2 values for these prodrugs. The valine ester prodrug 7b (VAL-APAP-

HCI) showed the longest half life amongst all the AAC compounds studied due to the

bulky isopropyl group. The shortest t1/2 amongst all the prodrugs evaluated was 14.9

min for 7c (PRO-APAP-HCI) prodrug.

(A) (B) H.. H
H2N/'cO 0O-
0 0
R HN. N I
RO" ,[ ]!:., 0


Figure 4-4. Proposed mechanism of general base catalysis by the proline ring. (A)
Presence of the proline ring "fixes" four atoms leading to rate enhancement in
hydrolysis of glycylproline ethyl ester. (B) general base catalysis by the
proline ring.

This is an unexpected result based on the steric factors alone. However, it has

been previously observed that piperazine-2, 5-diones are formed readily from dipeptide

esters particularly those containing N-methylamino acids or proline. In the case of

dipeptides containing proline, geometrical factors take control.85 For the cyclization of

dipeptide esters to piperazines to occur, the peptide bond must be in the cis

conformation so that the terminal amino group and the ester carbonyl carbon can

interact to form the six-membered ring. For example, glycylproline ethyl ester cyclizes

more readily than the prolylglycine ester because of the ease in assumption of the cis

conformation.86 The case of the 7c can be explained by similar explanation. Presence of

the proline ring system "fixes" the -NH group in an orientation that favors more effective









general base catalysis by the -NH group. The possible transition state involved in

hydrolysis of 7c has been shown (as a free base) in Figure 4-4.

Determination of Solubilities of AAC-APAP-HCI Prodrugs And VAL-NTX-TFA
Prodrug

The molar absorptivity of each AAC-APAP-HCI prodrug (7a 7c) was determined

in methanol in triplicate. Molar absorptivitiy of naltrexone (NTX) and its AAC prodrug

VAL-NTX-2TFA (10) was also determined in methanol. Molar absorptivities for parent

drugs and their corresponding AAC prodrugs have been listed in Table 4-4. A known

amount of prodrug was dissolved in deionized water, and the solution was immediately

diluted with methanol or pH 7.1 phosphate buffer and analyzed by UV

spectrophotometry. With the known concentration C, E244 was calculated with Beer's law

(shown for APAP):

A244 = E244 I C, where I = cell length (1)

Table 4-4: Molar absorptivities of AAC-APAP-HCI prodrugs, VAL-NTX-TFA and NTX
E MeOH E71
7a, ALA-APAP-HCI 1.541 0.0199a b, c
7b, VAL-APAP-HCI 1.553 0.031a, b c

7c, PRO-APAP-HCI --

10, VAL-NTX-TFA 2.175 0.059 a' d e

APAP -- 1.030 0.006a, b, c

NTX 0.94 0.005 a' d e 1.181 0.029 a d e
a All experiments were run in triplicate. b Units of 104 ml mmole-1 C UV max = 244 nm, a Units of 103 ml
mmole- e UV max = 282 nM

For each prodrug, the solubility in 1-octanol (OCT) was determined in triplicate.

The prodrug was crushed into a fine powder and a saturated solution of the prodrug in

octanol was obtained by adding an excess of each compound to a test tube containing

1 ml octanol.









Table 4-5. Physicochemical properties of AAC Prodrugs. Molecular Weight (MW),
solubility in 1-octanol (SOCT), solubility in propylene glycol (SPG) for AAC-
APAP-HCI prodrugs (7a 7c), VAL-NTX-TFA prodrug (10) and DAAC-APAP-
HCI Prodrugs (4a -4k)
MW Mp SOCTa SpGa

7a 272 200-240(d)b 1.02 0.11 335.57 4.74

7b 286 230-250 (d)b 4.65 0.38 649.79 1.13

7c 300 195-230 (d) b 1.336 0.28 260.91 13.94

10 6860 170 (64-68)d 6.514 0.358 155.57 0.39'

4ae 272 0.352 0.01 122.05 2.90

4bf 286 195-200 0.291 0.035 150.09 2.03

4Ce 300 200-205 1.297 0.067 93.80 0.98

4df 314 170-180 1.162 0.055 176.10 3.42

4fe 313 222 --

4gf 327 234 0.202 0.02 18.17 1.05

4hg 341 202 20.87 0.47 64.09 1.17

4ie 315 260-270 0.09 0.01 14.01 0.01

4jf 329 220 0.10 0.007 29.97 0.58

4kg 343 210 1.522 0.077 39.68 0.50

APAP 151 195 158.94h 662.25 h

NTX 341 175 694.77' 101.51 7.5
a Units of mM; b Decomposes with charring close to melting temperature; MW of the monohydrate; Mp
of the monohydrate and that of the bis-hydrate in paranthesis; e distance of amine nitrogen from acyl
group n = 1; n = 2; g n = 3; h Values from Kasting Smith and Cooper (1989); Values calculated from
Kaufman et. al. 197587 using log SOCT = log SAQ + log KOCTWater, Experiment run in duplicate.

The test tube was then insulated and the suspension was allowed to stir at room

temperature (23 1C) overnight on a magnetic stir plate. The suspension was filtered

through a 0.25 pm nylon syringe filter. An aliquot of the filtrate was diluted with methanol

and analyzed by UV spectrophotometry. A NMR of the flitered solid was recorded to

ensure that the prodrugs were intact during the course of solubility experiment. The

absorbance at 244 nm was used to calculate the concentration of the AAC-APAP-HCI









prodrug in octanol and the absorbance at 278 was used to calculate concentration of

the naltrexone prodrug in octanol. Solubility in octanol, SOCT:

Csaturation = SOCT = A244 / E244 (2)

Solubilities in propylene glycol (PG) were also determined in triplicate by the

procedure used to determine SOCT. The suspensions were stirred for 4 h before

filtration. A sample of the filtrate was diluted with methanol and analyzed by UV

spectrophotometry.

Melting points of AAC-APAP-HCI prodrugs were not sharp and the compounds

exhibited decomposition close to the melting temperature. All the AAC-APAP prodrugs

showed a higher melting point range than APAP. High melting point values are most

probably responsible for the low SOCT values exhibited by these compounds. Among the

AAC-APAP compounds 7b (valine ester) showed the highest SOCT. The SOCT for DAAC-

APAP-HCI prodrugs was comparable to the AAC series.

SPG of the AAC-APAP-HCI prodrugs was generally higher than that for DAAC-

APAP-HCL series. For example the most PG soluble member in the DAAC series is 4d

(176 mM) whereas the valine ester 7b is almost 3.7 times more soluble in PG (650 mM).

The trend in the PG solubility reflects the propensity of -NH2 group vs. -NR1R2 group in

modulating the aqueous-like solubility of the AAC vs. DAAC prodrug series. The reason

for this trend is that -+NH3 has more N H groups to hydrogen bond with protic solvents

water and PG.

The SOCT determination experiment with the NTX prodrug 10 showed that although

the prodrug was isolated as a bis-hydrate, the compound precipitates out of the

saturated octanol solution as a mono-hydrate. The mono-hydrate was isolated and









characterized by elemental analysis (% C Exp 50.49; Cal 50.739) and 1H-NMR. The

colorless mono-hydrate exhibited a higher melting point than that of the pale brown bis-

hydrate (shown in Table 4-5). The dependence of SOCT on melting point is again

exemplified here because the lower melting bis-hydrate was highly soluble in octanol

(>500 mg / 0.5 ml) but when the solution was stirred for about 30 minutes the less

soluble, higher melting mono-hydrate precipitated The SOCT values reported in Table 4-

5 correspond to the solubility of the mono-hydrate.

In-Vitro Evaluation of VAL-APAP-HCI and VAL-NTX-TFA Prodrugs

Two different mice were used to determine the flux of each prodrug. Prior to skin

removal, the mice were rendered unconscious by C02 then sacrificed via cervical

dislocation. Skins were removed by blunt dissection and placed dermal side down in

contact with pH 7.1 phosphate buffer (0.05 M, I = 0.11 M, 32 C) containing 0.11%

formaldehyde which has been shown to inhibit microbial growth and maintain the

integrity of the skins throughout the experiment.21 Prior to the application of the

prodrugs as suspensions in IPM:PG (99:1) the skins were maintained in contact with

buffer for 24 h to leach out all UV absorbing material. During this conditioning phase

time, the receptor phase was removed and replaced with buffer 3 times. The

suspension of the prodrug in IPM:PG was prepared 4 h before application and allowed

to stir at room temperature (25 1 C) until application. After the 24 hour leaching

period, an aliquot (0.5 ml) of the prodrug suspension was added to the surface of the

skin (donor phase). Samples of the receptor phase were usually taken at 8, 19, 22, 25,

28, 31, 34, and 48 h and quickly analyzed by UV spectrophotometry. Since only parent

drug was obtained in the receptor phase in all cases the amounts of permeated APAP

was quantified using









Table 4-6. Results from diffusion cell experiments with AAC prodrugs of APAP and NTX
(VAL-APAP-HCI and VAL-NTX-TFA). Maximum flux of the prodrug from a
saturated isopropyl myristate (IPM) vehicle through hairless mouse skin (JM),
maximum flux of standard drug, theophylline, from a saturated propylene
glycol (PG) vehicle (Js), log of experimental flux values (EXP log Js), residual
concentration of the drug in the skin (Cs), amount of parent drug (APAP or
NTX) delivered during the 48 h first application period (AAPP)
Compound JMa Js CSa,c EXP AAPPd
log JM
7c VAL-APAP-HCI 0.47 0.02 0.67 0.18 1.86 0.07 -0.32 60.73 21.9

10 VAL-NTX-TFA 0.074 0.00 0.50 0.04 2.36 0.46 -1.13 13.20 0.02

4a Me2n1APAP-HCIb 0.64 0.06 1.00 0.02 0.69 0.07 -0.19 59.62 4.06

4b Me2n2APAP-HClb 0.65 0.01 0.95 0.03 0.69 0.05 -0.19 63.30 1.51

4i MORnAPAP-HClb 0.54 0.02 0.96 0.25 0.71 0.18 -0.26 60.53 1.67

NTX-HCI 0.057 0.005 0.78 0.05 1.83 0.15 -1.24 11.42 0.43

APAP 0.51 0.74 2.74 0.70 -0.29 73.63
a Units of pmole cm2 h b Suspension of the prodrug was applied in a (99:1) PG:IPM vehicle.
c Concentration of APAP in the skin after a 24 h leaching period. d Units of pmole.

molar absorptivity of APAP or NTX in the receptor phase. At each sampling time, the

entire receptor phase was replaced with fresh buffer in order to maintain sink conditions.

After the 48 h of the first application period, the donor suspension was removed and the

skins were washed three times with methanol (3-5 ml) to remove any residual prodrug

from the surface of the skin. The remaining prodrug or APAP in the skin was leached

out by keeping the skins in contact with buffer for an additional 24 h. Then the receptor

phase was replaced with fresh buffer and an aliquot (0.5 ml) of a standard drug

theophylline was applied in PG to the skin surface. The second application fluxes were

determined by sampling of the receptor phase at 2, 4, 6 h and analysis by UV

spectrophotometry. The concentration of theophylline in the receptor phase was

determined by measuring its absorbance at 270 nm (E = 10,200 L mol-1). At each









sampling time, the entire receptor phase was removed and replaced with fresh buffer. In

each experiment, the flux was determined by plotting the cumulative amount of APAP

versus time. J in units of pmol cm-2 h could then be calculated by dividing the slope of

the steady-state portion of the graph (Figure 4-5) by the surface area of the skin (4.9

cm2).

The results for the diffusion cell experiments with VAL-APAP-HCI and VAL-NTX-

TFA are shown in Table 4-6. The flux of VAL-APAP-HCI was only as high as APAP

itself. Amongst all the DAAC-APAP-HCI and AAC-APAP-HCI prodrugs studied the

dimethyl containing salts gave best flux (1.2 times APAP). This is the first study where

permeation experiments with prodrugs of APAP as hydrochlorides has been reported.

All the hydrochloride prodrugs investigated in the study showed a longer lag time as

compared to APAP or the corresponding free base forms.

The permeation profile of NTX-HCI and its prodrug, VAL-NTX-TFA has been

shown in Figure 4-5. The VAL-NTX-TFA prodrug showed moderately higher flux than

NTX.HCI (1.3 times). The lag time for both the salts was approximately the same (~24

h). The lag times observed for NTX-HCI and its prodrug was very similar to that

observed for DAAC-APAP-HCI prodrugs and the AAC-APAP-HCI prodrug.

NTX-HCI delivered 11.42 pmole NTX whereas its prodrug, VAL-NTX-TFA,

delivered 13.2 pmole over a 48 h application period through hairless mouse skin from a

saturated IPM donor phase. The skin retention of NTX after the application period was

higher for the NTX prodrug by 1.3 times (2.4 vs 1.8 moless.


100










9

8

7 /-

6





3 ----VAL-NTX-2TFA
./---A-- NTX-HCL


1
0 5 10 15 20 25 30 35
Time (hrs)
Figure 4-5. Cumulative amount of NTX permeated vs time for compounds 10 and NTX-
HCI

The second application flux values were not higher than the control value of 1.0

pmole cm-2 h1. This showed that the skin used in the permeation experiment did not

undergo any damage due to the use of IPM as the donor phase.

The incorporation of a basic amine into the promoiety of an acyl prodrug of a

phenol containing drug resulted in an increase in lipid and aqueous solubilities of the

corresponding prodrugs in this study. However, the increased solubilities did not result

in a concomitant increase in the flux of the prodrugs when investigated as free bases

(investigated in chapter 3) or as their corresponding salt forms (investigated in chapter 3

and 4). The lower flux of the DAAC and AAC prodrugs investigated in chapter 3 and 4

can be attributed to the presence of hydrated forms of the prodrug as it diffuses through

the skin.

Based on solubility properties only, the flux of a prodrug in a salt form is expected

to be lower than the parent drug because the lipid solubility of a salt is considerably

lower than the parent drug, even though its water solubility is much higher-again a


101









"balance" in lipid and aqueous solubility is lacking. The slightly higher flux of the salt

forms observed here might be attributed to: (a) existence of an equilibrium between the

protonated and the free base forms in the skin microenvironment and (b) a favorable

interaction of the protonated amine groups with the fatty acid groups in the skin.

Conclusions

In this part of study three AAC prodrugs of APAP were synthesized via a

DCC/DMAP or a EDCI mediated coupling reaction. Based on the solubility properties of

the VAL-APAP-HCI, one AAC prodrug of NTX, VAL-NTX-TFA was also synthesized.

The Boc-protected compounds couldn't be isolated in very pure form using the DCC

method because of the presence of dicyclohexylurea by product. Thus the Boc-

protected compounds that were isolated from the EDCI coupling method were

characterized. The Boc protected NTX prodrug was isolated using the EDCI method.

Hydrolysis experiments with the AAC prodrugs were carried out in buffer (pH 6.0,

phosphate, 20 mM, 37 OC). The half lives of the prodrugs were affected by the steric

effect of the R group at the position alpha to the carbonyl group. The small t11/2 value for

the 7c (PRO-APAP-HCL) could be attributed to the possible transition state that the

proline ring in the prodrug can facilitate more readily than alanine or valine.

Solubility experiments with AAC-APAP compounds were carried out in 1-octanol

and propylene glycol. The compounds were only moderately soluble in octanol but

exhibited relatively higher PG solubility than DAAC-APAP-HCI prodrugs. The naltrexone

prodrug exists in form of mono and bis hydrates where the mono hydrate was present

as the primary species in a saturated octanol solution.

In-vitro permeation experiments were carried out with the VAL-APAP-HCI prodrug

and VAL-NTX-2TFA prodrug. The valine ester prodrug of APAP showed approximately


102









the same flux value as APAP. The naltexone prodrug showed a moderately higher flux

than NTX-HCI (1.3 times). The skin retention of the VAL-NTX-TFA prodrug was about

1.3 times compared to NTX-HCI.

In conclusion, the AAC promoiety is successful in enhancing the aqueous like

solubility of the corresponding prodrugs synthesized in this study. The solubility and

stability of the prodrugs can be modulated by changing the steric bulk on the alpha

carbon appropriately. The increase in water solubility of the AAC prodrugs did not

results in a concomitant increase in their flux. This might be due to the presence of

hydrated forms of the prodrug as it diffuses through the skin.

The introduction of the ionized group in the AAC hydrochloride prodrugs make

them an attractive targets for iontophoretic drug delivery. A combination of more than

one permeation enhancement technique, like chemical enhancers, might be the next

step in further optimization of the DAAC and AAC prodrug approach presented in this

work.


103









CHAPTER 5
CONCLUSION AND FUTURE WORK

In this study a prodrug approach in which, a basic amine functional group was

incorporated into a promoiety of an acyl prodrug of a phenolic drug, was developed. A

model phenolic drug acetaminophen (APAP) was used to demonstrate the feasibility

of the prodrug approach. The permeation experiments on hairless mouse skin with

APAP prodrugs showed that the dialkylaminoalkylcarbonyl promoiety was successful in

improving the flux of APAP by three fold. This prodrug approach can now be applied to

a broader class of phenolic parent drugs to achieve enhanced delivery through the skin.

The first objective of this study was to synthesize acyl prodrugs of a model

phenolic drug, acetaminophen (APAP), containing a basic amine group. This was

achieved by coupling the N, N'- dialkylaminoalkanoic acid hydrochlorides with the

parent phenolic drug via a DCC mediated coupling reaction. The corresponding N'-

dialkylaminoalkanoic acid hydrochlorides were synthesized using five different amines -

dimethylamine, diethylamine, dipropylamine, morpholine and piperidine. The N, N'-

dialkylaminoalkylcarbonyl (DAAC) ester prodrugs were obtained in good yields (70 -

90%). Additionally, aminoalkylcarbonyl (AAC) prodrugs of APAP and one AAC ester

prodrug of naltrexone (NTX) was synthesized. The AAC ester prodrugs were

synthesized using DCC or EDCI as the coupling reagents. The yields of the AAC-APAP

and AAC-NTX prodrugs were between 50-75%.

The second objective of this study was to evaluate the physicochemical properties

of DAAC-APAP, AAC-APAP and AAC-NTX prodrugs. Half lives of a few members was

evaluated in buffer (pH 6.0, phosphate, 50 mM, I = 0.5). The DAAC and AAC prodrugs

hydrolyzed to the parent drug with half lives between 15 115 minutes in pH 6.0 buffer.


104









The kinetic behavior of the DAAC and AAC prodrugs exemplifies the general base

catalysis by the basic amine group in the prodrugs that results in significant rate

enhancement (see page 92) compared to esters lacking a basic amine functionality.

Solubilities of DAAC prodrugs in buffer (acetate, pH 4.0, 50 mM), 1-octanol(OCT),

propylene glycol (PG) and isopropyl myristate (IPM) have been determined. The

solubilities of the AAC prodrugs were determined in 1-octanol and propylene glycol.

Amongst the DAAC-APAP-HCI prodrugs the member exhibiting the highest SPG was

also the one that exhibited the highest S4.0. SOCT of the DAAC-APAP-HCI were

generally low. All the DAAC-APAP prodrugs exhibited higher SIPM than APAP. The SIPM

of the DAAC-APAP (free bases) showed a dependence on their melting points. Among

the AAC-APAP-HCI prodrugs, the VAL-APAP-HCI exhibited the highest SOCT and SPG.

The VAL-NTX-2TFA prodrug also exhibited higher SPG than NTX. SOCT of the NTX

prodrug was lower than NTX itself which is expected from a ionic compound.

The third objective of this research was to investigate the potential of DAAC and

AAC prodrugs in enhancing the delivery of APAP through skin. The DAAC prodrugs

were evaluated for their ability to deliver APAP through hairless mouse skin. The best

member of the DAAC series of prodrugs showed three times higher flux than APAP.

Although the best member evaluated was much more lipid and water soluble than

APAP, the delivery of APAP was not as high as expected. We hypothesize that this may

be due to an increase in the molecular weight caused by water association with the

amine group as the prodrug diffuses the membrane. A new Robert-Sloan equation

using a n = 73 database was developed. The coefficients for the RS equation were x =

-0.599, y = 0.502, z = 0.00235 and r2 = 0.92. The calculated flux values for DAAC-APAP


105










prodrugs using the new RS equation were higher than experimental flux values in all

cases except one. The "over-prediction" of flux of DAAC-APAP prodrugs by the RS

equation may be due to challenges in precisely measuring or estimating the solubility of

the actual ionized species in the skin microenvironment while the prodrugs diffuse

through the skin.

One approach to improve the flux of the DAAC-HCI prodrugs is to deliver the

prodrugs through skin iontophoretically. The presence of the ionizable amine in the

DAAC or AAC prodrug allows for a utilization of iontophoresis as a permeation

enhancement technique. The pH of the skin environment has been shown to be about

5.5-6.0. 5 All the prodrugs are expected to be protonated in the skin microenvironment.

Therefore an applied electric field is expected to facilitate the passage of the ionized

molecule through skin more effectively. Recent work by Bowstra and coworkers shows

that the iontophoretic flux of the alanine ester prodrug of a parent phenolic drug was

higher than that of the parent drug.77

.R, OR,




0 N
R,= H R-= R3R2n R23 CH
APAP
n = 1-3


1 R3,

R' N
I

n

n=0-3

Figure 5-1: Chemical structures of proposed DAAC prodrugs of phenol containing
drugs.


106









The amount of the parent drug delivered by DAAC prodrugs seems to be

dependent on the nature of the donor phase solvent. Better solvation of the amine

group in an aqueous donor phase had a favorable effect on the flux of the testosterone

prodrugs.5 The primary difference between testosterone dimethylbutyrate hydrochloride

prodrug and the DAAC-APAP prodrugs is the fact that the leaving group is a phenol in

the latter series relative to a poorer alcoholic leaving group in the testosterone prodrug.

This made it feasible to deliver the prodrug from an aqueous solution whereas the

DAAC-APAP prodrugs are unstable in water over the course of a diffusion cell

experiment. One way to assess the effect of incorporation of a basic amine of the flux of

a parent drug is to synthesize aminoalkylcarbonyl derivatives of a drug containing an

alcoholic group. The corresponding prodrugs could then be delivered from an aqueous

suspension allowing an analysis of the effect of incorporation of a basic amine group on

the flux of the corresponding prodrug.

In compounds investigated so far, although incorporation of a basic amine caused

a favorable enhancement in the solubility properties of the corresponding prodrugs, the

increase in solubility is offset by increase in effective molecular weight of the more basic

prodrugs. One strategy to overcome this challenge is to synthesize DAAC prodrugs with

relatively weakly basic amines in the promoeity. The suggested compounds are shown

in Figure 5-1. This can facilitate a similar enhancement of solubilities without the

increment in effective molecular weight of the prodrug while it diffuses through the skin.

Additionally the weakly basic amine prodrugs are expected to be relatively more stable

in aqueous phase and therefore can be delivered from an aqueous donor phase.


107









In the present study, the utility of the DAAC and the AAC prodrugs approach was

investigated as a means of improving topical delivery of phenol containing drugs. When

delivered as a free base one of the DAAC-APAP prodrug was able to improve the

delivery of APAP by three times compared to its hydrochloride salt which showed

moderately higher flux than APAP (1.2 times). Similarly the AAC-NTX prodrug VAL-

NTX-TFA was able to improve the topical delivery of NTX by 1.3 times compared to

NTX-HCI.

Previous studies with topical delivery of ionized compounds have met with

success when a combination of more than one enhancement technique is used.89-91

One of the approaches to further enhance the delivery DAAC and AAC prodrugs and

their corresponding salts is by the use of chemical permeation enhancers in addition to

the iontophoretic technique mentioned above. Such a combined transdermal delivery

system can be useful in improving the delivery of compounds having an ionizable

amine.


108









LIST OF REFERENCES


1. Gibson, G.; Skett, P. An Intoduction to Drug Metabolism. Chapman and Hall:
1986.

2. Longcope, C. G. S.; Goldin, B.; Woods, M.; Dwyer, J.; Warram, J. The Metabolism
of Estradiol, Oral Compared to Intravenous Administration. J. Ster. Biochem. 1985,
23, 1065-1070.

3. Wolfe, M. M.; Lichtenstein, D. R.; Singh, G. Gastrointestinal Toxicity of Non-
Steroidal Anti-Inflammatory Drugs. N. Engl. J. Med. 1999, 340, 1888-1899.

4. Hargus, S. J.; Amouzedeh, H. R.; Pumford, N. R.; Myers, T. G.; McCoy, S. C.;
Pohl, L. R. Metabolic Activation and Immunochemical Localization of Liver Protein
Adducts of the Nonsteroidal Anti-Inflammatory Drug Diclofenac. Chem. Res.
Toxicol. 1995, 7, 575-582.

5. Schaefer, H.; Redelmeier, T. Skin Barrier. Principles of Percutaneous Absorption.
S. Karger Basel: 1996.

6. Swanson, H. Cytochrome P450 Expression in Human Keratinocytes: An Aryl
Hydrocarbon Receptor Perspective. Chem-Biol. Interact. 2004, 149, 69-79.

7. Williams, A. Transdermal and Topical Drug Delivery. Pharmaceutical Press:
London, 2003.

8. Pham, M.; Magdalou, J.; Totis, M.; Fournel-Gigleux, S.; Siest, G.; Hammock, B.
Characterization of Distinct Forms of Cytochromes P-450, Epoxide Metabolizing
Enzymes and Udp-Glucuronosyltransferases in Rat Skin. Biochem. Pharmacol.
1989, 38, 2187-2194.

9. Bashir, S.; Maibach, H. Cutaneous Metabolism of Xenobiotics. In Percutaneous
Absorption: Drugs-Cosmetics-Mechanisms-Methodology, Bronaugh, R.; Maibach,
H., Eds. Taylor and Francis: Boca Raton, 2005; pp 51-63.

10. Prausnitz, M. R.; Mitragotri, S.; Langer, R.; Current Status and Future Potential of
Transdermal Drug Delivery. Nat. Rev. Drug Discov. 2004, 3, 115-124.

11. Rhoden, E. L.; Morgentaler, A. Risks of Testosterone-Replacement Therapy and
Recommendations for Monitoring. N. Engl. J. Med. 2004, 350, 482-92.

12. Carrasco, D.; Prieto, M.; Pallardo, L. Multiple Hepatic Adenomas after Long-Term
Therapy with Testosterone Enanthate: Review of the Literature. J. Hepatol. 1985,
1, 573-578.

13. Comhaire, F. H. Andropause: Hormone Replacement Therapy in the Aging Male.
Eur. Urol. 2000, 38, 655-662.


109









14. Wang, C.; Swerdloff, R. S.; Iranmanesh, A.; Dobs, A.; Snyder, P. J.; Cunningham,
G.; Matsumoto, A. M.; Weber, T.; Berman, N. Transdermal Testosterone Gel
Improves Sexual Function, Mood, Muscle Strength, and Body Composition
Parameters in Hypogonadal Men. Journal of Clinical Endocrinology and
Metabolism 2000, 85, 2839-53.

15. Landmann, L. E. Transformation of Lamellar Granule-Disks into Intercellular
Sheets by a Membrane-Fusion Process, a Freeze-Fracture Study. J. Invest.
Dermatol. 1986, 87, 202-209.

16. Swartzendruber, D. C.; Wertz, P. W.; Kitko, D. J.; Madison, K. C.; Downing, D. J.
Molecular Models of the Intercellular Lipid Lamellae in Mammalian Stratum
Corneum. J. Invest. Dermatol. 1989, 92, 251-257.

17. Bowstra, J. A.; Gorvis, S. G.; Maria, P. Skin Lipid Organization, Composition and
Barrier Function. IFSCC Magazine 2007, 10, 297-307.

18. Bowstra, J. A.; Gorvis, S. G.; Spek, J. A.; Bras, W. Structural Investigations of
Human Stratum Corneum by Small-Angle X-Ray Scattering. Soc. Invest. Derm.
1991, 97, 1005-1012.

19. Sloan, K. B. Functional Group Considerations in the Development of Prodrug
Approaches to Solving Topical Delivery Problems. In Prodrugs: Topical and Ocular
Drug Delivery, Sloan, K. B., Ed. Marcel Dekker: New York, 1992; pp 17-116.

20. Williams, A. C.; Barry, B. W. Penetration Enhancers. Adv. Drug Del. Rev. 2004,
56, 603-618.

21. Sherertz, E. F.; Sloan, K. B.; McTiernan R. G. Effect of Skin Pretreatment with
Vehicle Alone or Drug in Vehicle on Flux of a Subsequently Applied Drug: Results
of Hairless Mouse Skin and Diffusion Cell Studies. J. Invest. Dermatol. 1987, 89,
249-52.

22. Van Hal, D. A.; Jeremiasse, E.; Junginger, H. E.; Spies, F.; Bouwstra, J. A.
Structure of Fully Hydrated Human Stratum Corneum: A Freeze-Fracture Electron
Microscopy Study. J. Invest. Dermatol. 1996, 106, 89-95.

23. Singh, J.; Maibach, H., I. Topical lontopphoresis in-Vivo: Historical Development,
Devices and Future. Dermatology 1993, 187, 235-238.

24. Warwick, W. J. et. al. Evaluation of a Cystic Fibrosis Screening System
Incorporating a Miniature Sweat Stimulator and Disposable Chloride Sensor. Clin.
Chem. 1986, 32, 850-853.

25. Miller, K. A.; Balakrishnan, G.; Eichbauer, G.; Betley, K. 1% Lidocaine Injection or
'Numby Stuff' for Topical Analgesia Associated with Peripheral Intravenous
Cannulation. Ana. J. 2001, 69, 185-187.


110









26. Matriano, J. A. et al. Macroflux Microprojection Array Patch Technology: A New
and Efficient Approach for Intracutaneous Immunization. Pharm. Res. 2002, 19,
63-70.

27. Cross, S. E.; Roberts, M. S. Physical Enhancement of Transdermal Drug
Application: Is Delivery Technology Keeping up with Pharmaceutical
Development? Curr. Drug Deliv. 2004, 1, 81-92.

28. Mikszta, J. An Improved Genetic Immunization Via Micromechanical Disruption of
Skin-Barrier Function and Targeted Epidermal Delivery. Nature Med. 2002, 8,
415-419.

29. Rein, H. Experimental Electroendosmotic Studies on Living Human Skin. Z. Biol.
1924, 81, 124.

30. Scheuplein, R. J.; Blank, I. H. Permeability of the Skin. Phys. Revs. 1971, 51,
702-747.

31. Scheuplein, R. J.; Blank, I. H. Mechanism of Percutaneous Absorption IV:
Penetration of Nonelectrolytes (Alcohols) Form Aqueous Solutions or Pure
Liquids. J. Invest. Dermatol. 1972, 5, 286-296.

32. Sloan, K. B.; Wasdo, S. C.; Rautio, J. Design for Optimized Topical Delivery:
Prodrugs and a Paradigm Change. Pharm. Res. 2006, 23, 2729-2746.

33. Majumdar, S.; Thomas, J.; Wasdo, S.; Sloan, K. B. The Effect of Water Solubility
of Solutes on Their Flux through Human Skin in Vitro. Int. J. Pharm. 2007, 329, 25-
36.

34. Kasting, G. B.; Smith, R. L.; Cooper, E. R. Effect of Lipid Solubilities and Molecular
Size on Percutaneous Absorption. In Skin Pharmacokinetics, Shroot B.; Schaefer,
H., Ed. Basel Karger: 1987; pp 138-153.

35. Potts, R. O.; Guy, R. H. Predicting Skin Permeability. Pharm. Res. 1992, 9, 663-9.

36. Roberts, W. J.; Sloan, K. B. Correlation of Aqueous and Lipid Solubilities with Flux
for Prodrugs of 5-Fluorouracil, Theophylline and 6-Mercaptopurine: A Potts-Guy
Approach. J. Pharm. Sci. 1999, 88, 515-522.

37. Sloan, K. B. Prodrugs for Dermal Dleivery. Adv. Drug Del. Rev. 1989, 3, 67-101.

38. Wasdo, S.; Sloan, K. B. Designing for Topical Delivery: Prodrugs Can Make the
Difference. Med. Res. Rev. 2003, 23, 763-793.

39. Sloan, K. B.; Wasdo, S.; Ezike-Mkparu, U.; Murray, T.; Nickels, D.; Singh, S.;
Shanks, T.; Tovar, J.; Ulmer, K.; Waranis, R. Topical Delivery of 5-Fluorouracil and
6-Mercaptopurine by Their Alkylcarbonyloxymethyl Prodrugs from Water: Vehicle
Effects on Design of Prodrugs. Pharm. Res. 2003, 20, 639-45.


111









40. Wasdo, S. C.; Juntunen, J.; Devarajan, H.; Sloan, K. B. A Comparison of the Fit of
Flux through Hairless Mouse Skin from Water Data to Three Model Equations. Int.
J. Pharm. 2009, 366, 65-73.

41. Juntunen, J.; Majumdar, S.; Sloan, K. B. The Effect of Water Solubility of Solutes
on Their Flux through Human Skin in Vitro: A Prodrug Database Integrated into the
Extended Flynn Database. Int. J. Pharm. 2008, 351, 92-103.

42. Bundgaard, H. Bioreversible Derivatives for Various Functional Groups and
Chemical Entities. In Design of Prodrugs, Bundgaard, H., Ed. Elsevier Science
Publishers B. V. (Biomedical Division) 1985; pp 1-92.

43. Varia, S. A.; Schuller, S.; Sloan, K. B.; Stella, V. J. Phenytoin Prodrugs lii Water-
Soluble Prodrugs for Oral and/or Parenteral Use. J. Pharm.Sci.1984, 73, 1068-
1073.

44. Rothbard, J. B.; Garlington, S.; Lin, Q.; Kirschberg, T.; Kreider, E.; McGrane, P.,
L.; Wender, P. A.; Khavari, P. A. Conjugation of Arginine Oligomers to Cyclosporin
a Facilitates Topical Delivery and Inhibition of Inflammation. Nat. Med. 2000, 6,
1253-1257.

45. Rothbard, J. B.; Jessop, T. C.; Wender, P. A. Adaptive Translocation: The Role of
Hydrogen Bonding and Membrane Potential in the Uptake of Guanidinium-Rich
Transporters into Cells. Adv. Drug Deliv. Rev. 2005, 57, 495-504.

46. Dhareshwar, S. S.; Stella, V. J. Prodrugs:Challenges and Review. Springer.: 2008;
Vol. 3.

47. Drustrup, J.; Fullerton, A.; Christrup, L.; Bundgaard, H. Utilization of Prodrugs to
Enhance Transdermal Delivery of Morphine. Int. J. Pharm. 1991, 71, 105-116.

48. Hansen, L. B.; Fullerton, A.; Christrup, L.; Bundgaard, H. Enhanced Transdermal
Delivery of Ketobemidone with Prodrugs. Int. J. Pharm. 1992, 84, 253-260.

49. Stinchcomb, A. L.; Swaan, P. W.; Ekabo, O.; Harris, K. K.; Browe, J.; Hammell, D.
C.; Cooperman, T. A.; Pearsall, M. Straight-Chain Naltrexone Ester Prodrugs:
Diffusion and Concurrent Esterase Biotransformation in Human Skin. J. Pharm.
Sci. 2002, 91, 2571-8.

50. Vaddi, H. K.; Hamad, M., O.; Chen, J.; Banks, S. L.; Crooks, P. A.; Stinchcomb, A.
Human Skin Permeation of Branched-Chain 3-O-Alkyl Ester and Carbonate
Prodrugs of Naltrexone. Pharm. Res. 2005, 22, 758-765.

51. Pillai, O.; Hamad, M. O.; Crooks, P. A.; Stinchcomb, A. L. Physicochemical
Evaluation, in Vitro Human Skin Diffusion, and Concurrent Biotransformation of 3-
O-Alkyl Carbonate Prodrugs of Naltrexone. Pharm. Res. 2004, 21, 1146-52.


112









52. Vaddi, H. K.; Banks, S. L.; Chen, J.; Hammell, D. C.; Crooks, P. A.; Stinchcomb,
A. L. Human Skin Permeation of 3-O-Alkyl Carbamate Prodrugs of Naltrexone. J.
Pharm. Sci. 2008.

53. Wasdo, S. C.; Sloan, K. B. Topical Delivery of a Model Phenolic Drug:
Alkyloxycarbonyl Prodrugs of Acetaminophen. Pharm. Res. 2004, 21, 940-6.

54. Milosovich, S. M.; Hussain, A. A.; Hussain, M.; Dittert, L. The Utilization of
Prodrugs to Enhance Transdermal Absorption of Testosterone,
Deoxycorticosterone & Indomethacin. Prog Clin Biol Res 1989, 292, 273-7.

55. Milosovich, S. H., A.; Dittert, L.; Aungst, B.; Hussain, M. Testosteronyl-4-
Dimethylaminobutyrate-Hcl: A Prodrug with Improved Skin Penetration Rate. J.
Pharm. Sci. 1993, 82, 227-8.

56. Hay, R. W.; Porter, L. J.; Morris, P. J. The Basic Hydrolysis of Amino Acid Esters.
Aust. J. Chem. 1966, 19, 1197-1205.

57. Kirby, A. J.; Lloyd, G. J. Intramolecular General Base Catalysis in the Hydrolysis
of 3-Dimethyl-Aminopropionates. J. Chem. Soc. Perkinll 1976, 1748-1752.

58. Ostacolo, C. M. F.; Laneri, S.; Sacchi, A.; Nicoli, S.; Padula, C.; Santi, P. Alpha-
Tocopherol Pro-Vitamins: Synthesis, Hydrolysis and Accumulation in Rabbit Ear
Skin. J. Control Release 2004, 2004, 403-13.

59. Thomas, J., D.; Sloan, K., B. In Vitro Evaluation of Alkylcarbonyloxymethyl
(ACOM) Ethers as Novel Prodrugs of Phenols for Topical Delivery: ACOM
Prodrugs of Acetaminophen. Int. J. Pharm. 2008, 346, 80-88.

60. Thomas, J. D.; Sloan, K. B. Evaluation of Alkyloxycarbonyloxymethyl (AOCOM)
Ethers as Novel Prodrugs of Phenols for Topical Delivery: AOCOM Prodrugs of
Acetaminophen. Int. J. Pharm. 2009, 371, 25-32.

61. Majumdar, S.; Sloan, K. B. Synthesis and Topical Delivery of N-Alkyl-N-
Alkyloxycarbonylaminomethyl Prodrugs of a Model Phenolic Drug:
Acetaminophen. Int. J. Pharm. 2007, 337, 48-55.

62. Sloan, K. B.; Wasdo, S. Designing for Topical Delivery: Prodrugs Can Make the
Difference. Med. Res. Rev. 2003, 23, 763-93.

63. Beall, H. D.; Sloan, K. B. Topical Delivery of 5-Fluorouracil (5-FU) by 3-
Alkylcarbonyl-5-FU Prodrugs. Int. J. Pharm. 2001, 217, 127-37.

64. Roberts, W. J.; Sloan, K. B. Topical Delivery of 5-Fluorouracil (5-FU) by 3-
Alkylcarbonyloxymethyl-5-FU Prodrugs. J. Pharm. Sci. 2003, 92, 1028-36.

65. Beall, H. D.; Sloan, K. B. Topical Delivery of 5-Fluorouracil (5-Fu) by 1,3-
Bisalkylcarbonyl-5-FU Prodrugs. Int. J. Pharm. 2002, 231, 43-9.


113









66. Johnson, K.; Amidon, G. L.; Pogany, S. Solution Kinetics of a Water-Soluble
Hydrocortisone Prodrug: Hydrocortisone-21-Lysinate. J. Pharm. Sci. 1985, 74, 87-
9.

67. Kovach, I. M.; Pitman, I. H.; Higuchi, T. Amino Acid Esters of Phenols as
Prodrugs: Synthesis and Stability of Glycine, Beta Aspartic Acid and Alpha
Aspartic Acid Esters of p-Acetamidophenol. J. Pharm. Sci. 1981, 70, 881-885.

68. Bruice, T. C.; Benckovic, S. J. Bioorganic Mechanisms. Benjamin New York: New
York, 1965; Vol. 1.

69. Gould, E. S. Mechanisms and Structures in Organic Chemistry. Rinehart and
Winston Inc.: 1959.

70. Bergstrom, C. A.; Luthman, K.; Artursson, P. Accuracy of Calculated pH-
Dependent Aqueous Drug Solubility. Eur. J. Pharm. Sci. 2004, 22, 387-98.

71. Gregory, M. D.; Christian, S. D.; Affsprung, H. E. The Hydration of Amines in
Organic Solvents. J. Phys. Chem. 1967, 71, 2283-2289.

72. Mohr, S. C.; Wilk, W. D.; Barrow, G. M. The Association of Water with Bases and
Anions in an Inert Solvent. J. Am. Chem. Soc. 1965, 87, 3048-3052.

73. Okouchi, S.; Moto, T.; Ishihara, Y.; Numajiri, H.; Uedaira, H. Hydration of Amines,
Diamines, Polyamines and Amides Studied by NMR. J. Chem. Soc. Faraday
Trans. 1996, 92, 1853-1857.

74. Bean, R. C.; Shepherd, W. C.; Chan, H. Permeability of Lipid Bilayer Membranes
to Organic Solutes. J. Gen. Physiol. 1968, 52, 495-508.

75. Thomas, J. D.; Majumdar, S.; Sloan, K. B. Soft Alkyl Ether Prodrugs of a Model
Phenolic Drug: The Effect of Incorporation of Ethyleneoxy Groups on Transdermal
Delivery. Molecules 2009, 14, 4231-45.

76. Roberts, W. J.; Sloan, K. B. Topical Delivery of 5-Fluorouracil (5-Fu) by 3-
Alkylcarbonyloxymethyl-5-FU Prodrugs. J. Pharm. Sci. 2003, 92, 1028-36.

77. Ackaert, O. W.; Graan, J. D.; Capanconi, R.; Pasqua, D. O. E.; Dijkstra, D.;
Westerink, B. H.; Danhof, M.; Bouwstra, J. A. The In- Vitro and In- Vivo Evaluation
of New Synthsized Prodrugs of 5-OH DPAT for lontophoretic Delivery. J.
Controlled Release 2010, 144, 296-305.

78. Aboofazeli, R.; Zia, H.; Needham, T. E. Transdermal Delivery of Nicardipine: An
Approach to In- Vitro Permeation Enhancement. Drug Delivery 2002, 9, 239-247.

79. Ren, C. et. al. Transdermal Delivery of Tolterodine by O-Acylmenthol: In-Vitro/in
Vivo Correlation. Int. J. Pharm. 2009, 374, 73-81.


114









80. Volpicelli, J. R.; Alterman, A. I.; Hayashida, M.; O'Brien, C. P. Naltrexone in the
Treatment of Alcohol Dependence. Arch. Gen. Psychiatry 1992, 49, 876-80.

81. Terenius, L. Rational Treatment of Addiction. Curr. Opin. Chem. Biol. 1998, 2,
541-7.

82. Krystal, J. H.; Cramer, J. A.; Krol, W. F.; Kirk, G. F.; Rosenheck, R. A. Naltrexone
in the Treatment of Alcohol Dependence. N. Engl. J. Med. 2001, 345, 1734-9.

83. Generics, P. D. R. Medical Economics. Montvale,New Jersey, 1996.

84. Sullivan, M. A.; Garawi, F.; Bisaga, A.; Comer, S. D.; Carpenter, K.; Raby, W. N.;
Anen, S. J.; Brooks, A. C.; Jiang, H.; Akerele, E.; Nunes, E. V. Management of
Relapse in Naltrexone Maintenance for Heroin Dependence. Drug Alcohol
Depend. 2007, 91, 289-92.

85. Benoiton, L.; Purdie, J. E. Piperazinedione Formation from Esters of Dipeptides
Containing Glycine, Alanine and Sacrosine: The Kinetics in Aqueous Solution.
JCS Perkin // 1973, 1845-1852.

86. Rydon, H. N.; Smith, P. W. G. Polypeptides. Part IV. The Self Condensation of
Esters of Some Peptides of Glycine and Proline. J. Chem. Soc. 1956, 3642.

87. Kaufman, J. J.; Semo, N. M.; Koski, W. S. Microelectric Titration Measurement of
the pKa's and Partition and Drug Distribution Coefficients of Narcotics and
Narcotic Antagonists and Their pH and Temperature Dependence. J. Med. Chem.
1975, 18, 647-655.

88. Smyth, H. D.; Becket, G.; Mehta, S. Effect of Permeation Enhancer Pretreatment
on the lontophoresis of Luteinizing Hormone Releasing Hormone (LHRH) Through
Human Epidermal Membrane (HEM). J. Pharm. Sci. 2002, 91, 1296-307.

89. Artusi, M.; Nicoli, S.; Colombo, P.; Bettini, R.; Sacchi, A.; Santi, P. Effect of
Chemical Enhancers and lontophoresis on Thiocolchicoside Permeation across
Rabbit and Human Skin in Vitro. J. Pharm. Sci. 2004, 93, 2431-8.

90. Nolan, L. M.; Corish, J.; Corrigan, O. I.; Fitzpatrick, D. Combined Effects of
lontophoretic and Chemical Enhancement on Drug Delivery. Transport across
Human and Murine Skin. Int. J. Pharm. 2007, 341, 114-24.


115









BIOGRAPHICAL SKETCH

Hemamalini Devarajan Ketha was born in India in 1982 to Raghavan Devarajan

and Kalyani Devarajan. Her family moved to New Delhi, India in the same year where

she was raised. She finished high school in April, 2000. The same year she enrolled in

the Bachelor of Science degree programme with a chemistry major at Hansraj College

at University of Delhi. She graduated from college in 2003 after which she pursued a

Master of Science degree in organic chemistry from 2003 to 2005. In the year 2006 she

started working on her PhD in Pharmaceutical Sciences Medicinal Chemistry at the

University of Florida where she met her husband Siva in the spring of 2007. Hema and

Siva married in 2009.


116





PAGE 1

1 N, N` DIALKYLAMINOALKYL CARBONYL (DAAC) AND AMINOALKYLCARBONYL (AAC) PRODRUGS OF PHENOLIC DRUG S By HEMAMALINI DEVARAJAN KETHA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFI LLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Hemamalini Devarajan Ketha

PAGE 3

3 To Amma my mother Kalyani

PAGE 4

4 ACKNOWLEDGMENTS As I look back into time my desire to pursue research could not have been accomplished without the strength and optimism that my mother, Kalyani, instilled into me. For me s he is the personification of strength against all odds. I am indebted to my father, Devarajan, for his love and care. I am thankful to my sister Indira for her unwavering support and for the time that she spent just listening to my ideas. My success as a student and an individual could not have been possible without my family. I would like to thank Dr. Kenneth Sloan for this opportunity and for the imm ense patience that he showed when I made mistakes he taught me to stop, slow down and learn through each mistake. Of thing I am certain as a researcher and as an individual, I will make mistakes but I am also confident that Dr. Sloans words the f aster you go the behinder you get will help me make my every mistake a learning experience. I am thankful to my committee members Dr. M argret O. James, Dr. R aymond Bergeron, Dr. K enneth Wagener and Dr. S cott Wasdo for their encouragement and constructive ideas through the process. Scotts presence and ideas at the weekly group meetings made it a great learning experience. And finally, one person that needs a special mention here is Siva, my husband. H is presence and love gives me the balance and strength t hat I need. I want to thank him for his patience to listen to my day to day research and for sharing my excitement for simple achievements like a good solubility experiment or a clean NMR spectrum. I am very fortunate that I met Siva and that he is by my s ide in all my endeavors big or small. His love makes my tough times easy and my good days meaningful.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TA BLES ............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 8 ABSTRACT ................................................................................................................... 10 CHAPTER 1 BACKGROUND ...................................................................................................... 12 The Rationale For Topical Delivery ......................................................................... 12 Epidermis ......................................................................................................... 17 Stratum Corneum ............................................................................................. 19 Strategies for Enhancing Topical Delivery .............................................................. 22 Optimizing Topical Delivery .................................................................................... 23 Mechanism of Percutaneous Absorption of Drugs ........................................... 24 Model Development ......................................................................................... 27 Transformed Potts Guy EquationThe Robert Sloan Equation ........................ 29 Prodrug Strategy to Enhance Percutaneous Absorption of Drugs .......................... 32 Acyl Prodrugs ................................................................................................... 35 Soft Alkyl Prodrugs ........................................................................................... 41 Conclusions ............................................................................................................ 43 2 SPECIFIC OBJECTIVES ........................................................................................ 45 First Objective ......................................................................................................... 45 Second Objective .................................................................................................... 46 Third Objective ........................................................................................................ 46 3 N, N` DIALKYLAMINOALKYL CARBONYL (DAAC) PRODRUGS OF A MODEL PHENOLIC DRUG ACETAMINOPHEN (APAP) ..................................... 48 Introduction ............................................................................................................. 48 Synthesis of DAAC Prodrugs of Acetaminophen .................................................... 5 0 Hydrolysis of DAACAPAP Prodrugs ...................................................................... 60 Determination of Solubi lities of DAACAPAP Prodrugs .......................................... 64 In Vitro Flux Determination of DAAC APAP Prodrugs ............................................ 72 In Vitro Evaluation of Morpholinyl and Piperidinyl DAACAPAP Prodrugs. ................................................................................................ 73 In Vitro Evaluation of Dimethyl and Diethyl DAAC APAP Prodrugs and DAAC APAP HCl Prodrugs. .................................................................... 80

PAGE 6

6 Con clusions ............................................................................................................ 83 4 AMINOALKYLCARBONYL (AAC) PRODRUGS OF PHENOLIC DRUGS ACETAMINOPHEN AND NALTREXONE ............................................................... 85 Introduction ............................................................................................................. 85 Synthesis of AAC Prodrugs of Acetaminophen (APAP) And Naltrexone (NTX) ...... 88 Hydrolysis of AAC prodrugs .................................................................................... 92 Determination of Solubilities of AACAPAPHCl Prodrugs And VALNTX TFA Prodrug ................................................................................................................ 95 In Vitro Evaluation of VALAPAPHCl and VAL NTX TFA Prodrugs ...................... 98 Conclusions .......................................................................................................... 102 5 CONCLUSION AND FUTURE WORK .................................................................. 104 LIST OF REFERENCES ............................................................................................. 109 BIOGRAPHICAL SKETCH .......................................................................................... 116

PAGE 7

7 LIST OF TABLES Table page 1 1 General representation of a prodrug, an acyl prodrug and a soft alkyl prodrug. .............................................................................................................. 34 1 2 Acyl prodrugs formed from hydroxyl or amine containing drugs ......................... 35 3 1 Characterization of DAAC APAPHCl and DAACAPAP Compounds. ............... 58 3 2 Half Lives (t1/2, min) and predicted pKa Values of DAAC APAPHCl, prodrugs in Buffer (pH 6.0, phosphate, 20 mM, I = 0.5) at 37 0.5 0C .............................. 61 3 3 Molar absorptivities of DAAC APAPMeOH), 4.0) and molar absorptivities of DAAC APAP prodrugs in ACN). .............................................................................................. 65 3 4 Physicochemical Properties of DAAC APAPHCl Prodrugs.. ............................. 67 3 5 Physicochemical properties of DAAC APAP podrugs.. ...................................... 69 3 6 Results from diffusion cell experiments with morpholinyl and piperidinyl DAAC APAP prodrugs. ....................................................................................... 76 3 7 Results from diffusion cell experiments with dimethyl an d diethyl DAACAPAP prodrugs and DAAC APAPHCl prodrugs .............................................. 81 4 1 Elemental Analysis and Melting Points for AAC.HCl Prodrugs of 4Hydroxyacetanilide (APAP) And AAC Prodrug of Naltrexone ............................. 91 4 2 Percentage yields of the Boc AAC APAP compounds and the corresponding AAC prodrugs synthesized from Method A or Method B .................................... 92 4 3 Half lives (t1/2, min) and predicted pKa values of AAC prodrugs of APAP and naltrexone in buffer (pH 6.0, phosphate, 20 mM, I = 0.5) at 37 0.5 0C ............. 93 4 4 Molar absorptivities of AAC APAPHCl prodrugs, VALNTX TFA and NTX ....... 95 4 5 Physicochemical properties of AAC Prodrugs. ................................................... 96 4 6 Results from diffusion cell experiments with AAC prodrugs of APAP and NTX (VAL APAPHCl and VAL NTX TFA). ................................................................ 99

PAGE 8

8 LIST OF FIGURES Figure page 1 1 Chemical Structures of Skin Ceramides ............................................................. 20 1 2 Pathway For Diffusion of a Topically Applied Permeant Based On The Biphasic Solubility Model. ................................................................................... 26 1 3 Wasdo Plot.. ....................................................................................................... 30 1 4 Mechanism of hydrolysis of fosphenytoin ........................................................... 32 1 5 Cyclosporine Prodrugs. ...................................................................................... 33 1 6 Testosterone and Naltrexone Prodrugs. ............................................................. 36 1 7 Chemical structures and IPM and aqueous solubilites of (A) APAP; (B) C2AOC APAP; (C) C2 ACOM APAP; (D) C2 AOCOM APAP; (E) C2 NANAOCAM APAP ............................................................................................ 38 1 8 Mechanism of of Hydrolysis of Acyl Prodrugs.. .................................................. 39 1 9 Chemical structures of Vitamin E and its prodrugs evaluated previousl y. .......... 40 1 10 Mechanism of hydrolysis of a soft alkyl prodrug. ................................................ 41 1 11 Mannich Base Prodrugs of 5 FU. ........................................................................ 42 3 1 Chemical structures of synthesized DAAC APAP prodrugs ............................... 50 3 2 Synthesis of DAAC APAP prodrugs ................................................................... 52 3 3 Possible routes of hydrolysis of a DAAC APAP prodrug .................................... 63 3 4 Origin of steric hindrance in 4a and 4c based on Newman Rule of Six. ............. 64 3 5 Plot of IPM solubilities vs melting points for DAAC APAP prodrugs ................... 71 3 5 Plot of cumulative amount vs time for calculating steady state flux .................... 74 3 6 Plot of pKa vs J for morpholinyl and piperidinyl DAAC APAP prodrugs. ............ 75 3 7 Plot of EXP log JM vs CALC log JM. .................................................................... 79 3 8 Cumulative amount of drug permeated vs time for MORn1APAP, MORn1APAPHCl and Me2n1APAP and Me2n1APAP HCl ........................................... 80

PAGE 9

9 3 9 Plot of EXP log JM vs CALC log JM including dimethyl and diethyl DAAC APAP prodrugs ................................................................................................... 82 4 1 Chemical structures of commercially available drug Valacyclovir and valine ester prodrug of 5OH DPAT by Bouwstra et. al. ................................................ 85 4 2 Chemical Structures of AAC Prodrugs of APAP and NTX .................................. 88 4 3 Synthesis of AAC prodrugs of APAP and NTX. .................................................. 89 4 4 Proposed mechanism of general base catalysis by the proline ring. 94 4 5 Cumulative amount of NTX permeated vs time for compounds 10 and NTX HCl ................................................................................................................... 101 5 1 Chemical structures of proposed DAAC prodrugs of phenol containing drugs. 106

PAGE 10

10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy N, N` DIALKYLAMINOALKYL CARBONYL (DAAC) AND AMINOALKYLCAR BONYL (AAC) PRODRUGS OF PHENOLIC DRUG S By Hemamalini Devarajan Ketha August 2010 Chair: Kenneth B. Sloan Major: Pharmaceutical Sciences Medicinal Chemistry Oral delivery of drugs is the preferred route of drug administration because of the ease in dosing regimen and high patient compliance. However the bioavailability of an orally administered drug can be low due to first pass drug metabolism. Topical delivery of drugs circumvents the challenges associated with first pass effects. Topical delivery of phenols is an attractive field of research particularly because of applicati ons in area of pain manag ement (morphine) hormone replacement therapy (estradiol) and in the treatment of alcohol addiction (naltrexone). Prodrug strategy involves chemical modification of a parent drug with poor delivery or physicochemical properties to a transient form with fav orable physicochemical properties, which reverts back to the parent drug chemically or enzymatically. T he chemical modification that results in an increase in the aqueous and lipid solubility without a considerable increase in the molecular weight of the c orresponding prodrug has been shown to give highest enhancement in flux. In the present study N, N` dialkylaminoalkyl carbonyl (DAAC) and aminoalkyl carbonyl (AAC) prodrugs of acetaminophen were synthesized and as well as o ne AAC prodrug of naltrexone (NTX ) The hypothesis is that the incorporation of a

PAGE 11

11 basic amine functionality in to an acyl prodrug will result in a favorable increase in its aqueous and lipid (biphasic) solubility resulting in an enhancement in topical delivery of the parent drug. The DAAC and AAC prodrugs hydrolyzed to the parent drug with half lives betw een 15 115 minutes in pH 6.0 buffer Solubilities of DAAC prodrugs in buffer (acetate, pH 4.0, 50 mM), 1octanol, pr opylene glycol and isopropyl myristate (IPM) have been determined. The solubilities of the AAC HCl prodrugs were determined in 1octanol and propylene glycol. The DAAC and AAC prodrugs were evaluated for their ability to deliver APAP or NTX through hairless mouse skin. Although one of the prodrugs showed three times higher fl ux than APAP, the delivery of APAP was not as high as expected. We hypothesize that this may be due to an increase in the molecular weight caused by aggregation of the prodrugs and water association with the amine group as the prodrug diffuses the membrane

PAGE 12

12 CHAPTER 1 BACKGROUND The Rationale For Topical Delivery Oral delivery of drugs is preferred over other routes of drug administration because of high patient compliance towards dosing regimens. However, before an orally administered drug can reach syst emic circulation, it must survive the acidic environment of the stomach, where it might undergo deactivation or, while passing through the upper and lower gastrointestinal (GI) tract, is subject to conjugation reactions catalyzed by various biotransformati on enzymes. After a drug is absorbed from the gut wall, it enters the liver, which is the major site for xenobiotic biotransformation reactions. The GI tract and liver have the highest concentrations of drug metabolizing enzymes.1 These enzymes are responsible for converting the active drug into more hydrophilic forms that are readily eliminated resulting in suboptimal oral bioavailability of the drug. Additionally, efflux transporters also c ontribute towards lower bioavailability of an orally administered drug form. For example, oral bioavailability of estradiol (administered to women undergoing Hormone Replacement Therapy (HRT)), is only 10% and is eliminated mainly as its glucuronide in the urine.2 Additionally, certain orally administered drugs, especially those containing an unm asked carboxylic acid group, have been shown to cause damage to gastric mucosa 3 and hepatotoxicity.4 The problems associated with an orally administered drug can be circumvented by administering the drug via an alternate route. Topical delivery is one such alternative. The terms topical and transdermal deliver y refer to the diffusion of a drug across the skin intended for local or systemic absorption, respectively. The undesirable effects of the GI tract and liver can also be overcome by applying a drug topically. Skin has a

PAGE 13

13 much lower concentration of drug met abolizing enz ymes than the GI tract or liver. 5 8 Cytochrome P450 activity is only about 15% of that found in the hepatic environment and UGT/UDPGA (glucuronidation) and PAPS/SULT (sulfation) activity is only about 10% of that observed in the liver.9 Since topical delivery of drugs has advantages like higher bioavailability and reduced incidence of metabolic effects over orally delivered drug forms, topical delivery of drugs has r eceived considerable attention in the last few decades. Most of the drugs currently formulated for delivery through skin are effective at low required doses (up to a few milligrams) and are delivered over a period of few days.9 The transdermal patch of scopolamine (an alkaloid and a muscarinic antagonist) was the first to be approved by the US FDA as a threeday patch for the treatment of motion sickness. Over the years, patches for a number of drugs like scopolamine, nitr oglycerine, clonidine, fentanyl, lidocaine, nicotine, estradiol and testosterone have been introduced in the market.10 Test osterone is a member of androgen group of steroid hormones. It is the primary male sex hormone and anabolic steroid (promotes synthesis of important cellular proteins, especially in the muscle tissue) and plays an important role in maintaining general male sexual health. Hypogonadism is a clinical condition related to decreased activity of gonads (organ responsible for production of sex hormones including estradiol and testosterone) which results in decreased testosterone levels in males and females. The sy mptoms include low libido, reduced bone and muscle mass and anemia.11 Several (injectable, transdermal, oral and buccal) formulations are available for testosterone replacement therapy. Oral formulations containing methyltestosterone and fluoxymesterone as the active pharmaceutical ingredient, even though available, are

PAGE 14

14 prescribed infrequently.11 The oral delivery of testosterone has been associated with severe hepatotoxicity and development of benign and malignant neoplasms.12 However, the transdermal patch that delivers between 5 10 mg of testosterone per day, has been shown to maintain uniform serum levels 13, 14 of testosterone without the toxicity associated with the oral formulation. There are other examples (fentanyl, estradiol, nitroglycerin) where delivery of a drug t hrough skin results in fewer side eff ects and higher bioavailability.10 On the other hand topical delivery of drugs is also not without challenges. The number of therapeutically useful compounds that can be administered through skin is limited by the size and physicochemical properties of the permeant.5, 10 Topical patches for low molecular weight drugs like, nictotine (162 Da) and moderately high molecular weight drugs like oxybutin (359 Da ) are avail able: oxybutin being the largest molecule being formulated commercially. The skin acts as an external physical barrier, protecting the inner body organs against water loss, physical damage and exposure to adverse chemical and microbial environment. The ski n is composed of a highly interconnected, complex assembly of cells and tissues that can be divided into three different layers: epidermis (50 100 m), dermis (1 2 mm) and hypodermis (1 2 mm). The thickness of skin can vary depending on age, sex and anatomical location.5 It is the thickest on areas that are more susceptible to physical abrasion, like heels and palms of hands. The outermost layer, stratum corneum (SC), al though the thinnest of all the layers, presents the most formidable barrier towards diffusion of permeants across the skin. An efficient delivery of a clinically relevant amount of an active therapeutic agent across (transdermal delivery) or into the skin (dermal delivery) demands knowledge about the

PAGE 15

15 fundamental structural components that give rise to the barrier function associated with the skin. The structure and composition of each of the layers from innermost to the outermost layer is discussed briefly below. Hypodermis Hypodermis or the subcutaneous tissue lies right below the dermis and is composed primarily of adipocytes. The main function of this layer i s to store fat and provide insulation from external shocks and abrasions. The thickness of hypodermis can vary depending upon the anatomic location. In addition, hypodermis contains a vast network of blood vessels that facilitate the absorption of topically applied compounds. Fibroblasts and macrophages are also found in this layer. Dermis The layer underlying epidermis, termed as dermis is a highly filamentous and fibrous connective tissue which imparts elasticity and tensile strength to the skin. The dermis constitutes about 70% of the dry weight of skin. Fibroblasts, the most abundant cells found in this layer are responsible for generating connective tissue compone nts like elastin and collagen.5 These components are synthesized using water soluble precursors and are released into the intercellular space by fibroblasts. Then, the elastin and collagen are assembled into thin fibrous structures called fibrils. The lowermost layer of dermis (reticular dermis) is comprised of more closely assembled, highly cross linked fib rils which become less dense as one moves towards the outer most layers of dermis (papillary dermis) to allow for proliferation of a vast network of nerve endings. These connective fibers act as a structural scaffold onto which other appendages found in this layer are anchored. The region surrounding the fibrous network is primarily comprised of highly hydrophilic macromolecules called proteoglycans. Proteoglycans are a group of fibroblast derived compounds that contain

PAGE 16

16 numerous unbranched polysaccharide chain covalently linked to a polypepetide backbone. The polysaccharide chains are composed of several hundred glycosaminoglycan disaccharides that have been ext ensively sulphated. The anionic charge on these chains from the sulphate and carboxylate groups results in an overall negative charge on the side chains and considerable repulsion which is strong enough to stabilize the chains in completely extended or a stretched state. The negative charge also allows for water molecules to be associated with these pr oteins. These proteins are capable of retaining about 1000 times their weight in water. The high degree of hydration of the dermis makes it resemble a hydrophilic gel. Some of the important proteoglycans include chondroitin sulfate/dermantan sulfate, hepar in/heparin sulfate etc. Appendages like sweat glands, hair follicles and sebaceous glands that extend up into the stratum corneum ( SC ) originate in the dermis. Hair follicles are sheath like structures that are embedded in the dermis. S weat glands and seb aceous glands extend till the outer most layer of the skin and deposit their respective contents on to the skin surface. S ometimes, these glands are found to be linked with the hair follicle and secrete their contents into the follicle. Like other features the density and presence of each appendage varies with anatomical location. Since these appendages provide a direct access for a drug molecule on the outermost layer to reach the dermis; they may provide a poremediated pathway for topically applied drug s. However, only a small proportion (approximately 0.1%) of the skin surface is covered with pores, allowing only a very small fraction of topically applied drug molecules to permeate via this route. The border between dermis and epidermis, also called the dermal epidermal junction is an uneven surface formed by a network of interlocked papillae from each of

PAGE 17

17 the layers. The wavelike structure of the dermal epidermal junction provides a large surface area for the dermal papillae to interact with the epidermal cells facilitating an effective removal of xenobiotics that permeate through skin. The extensive vascular network that supplies nutrients and oxygen to the epidermis is also responsible for transporting topically applied drugs into systemic circulatio n thereby maintaining a continuous sink condition. Epidermis The epidermis is composed of layers of successively keratinizing cells with each layer at a different phase of the keratinization process. The lower most to uppermost layer from the dermal epidermal junction to the skin surface are termed stratum basale, stratum spinosum, stratum granulosum and stratum corneum (SC). The layers directly below the SC (i.e stratum granulosum, stratum spinosum and stratum basale) are together termed as the viable epidermis. The viable epidermis contains about 70% water by weight. SC is composed of flat polyhedral shaped cells called corneocytes or horny cells. These are about 40 m in diameter and 0.5 m thick. Corneocytes are mainly composed of insoluble, aggregates of keratin. These are dead cells resulting from keratinization of the cells of the viable epidermis. Although the SC is the thinnest layer, it is the most impermeable amongst all the layers of the skin. The unique structural organization of the SC renders it a primary barrier towards topically applied drugs. Stratum Basale. Keratinocytes in the basal layer are attached to the basement membrane i.e the layer above the dermal epidermal junction. The keratinocyte in the stratum basale divides into a daughter stem cell and a transit amplifynig cell. The daughter stem cell remains attached to the basement membrane whereas the transit

PAGE 18

18 amplifying cell undergoes differentiation to form the spinous layer. The transit amplifying cells however lack features characteristic of spinous and granular cells, for example high nucleus to cytoplasm ratio. As more keratinocytes undergo differentiation, more transit amplifying cells progress to the higher layer of the skin and to the next phase of the keratinization process. Ot her components of this layer include melanocytes (cells responsible for melanin production), Langerhans cells (mediators of immune responses) and Merkel cells (sensory reception). Stratum Spinosum During their transit through the spinous layer, the cel ls begin to undergo changes in the cellular environment like keratinization of organelles and other proteins present inside the cell. This process leads to morphological changes that make the keratinocytes increasingly flattened. As the cell continues its transit towards the upper layers the diameter and volume of the cell continue to increase. This stage also marks the beginning of formation of desmosomal plaques between the cells that are responsible for providing cohesive streng th to the corneocyte layer s. D esmosomes are surface proteins responsible for holding the flattened keratinocytes together. The keratinocytes begin synthesizing the proteins, Keratin 1 (K1) and Keratin 10 (K10) which are precursors required in the next phase of the kerati ni zation process. Another important biochemical process that occurs during this phase is the formation of lamellar granules which later become a part of the primary components that provide the SC its unique barrier properties. Stratum Granulosum The cellular components, especially the nucleus, of the keratinocytes in this layer begin to undergo enzymatic degradation. This layer is rich in speckled or granular components called Keratohyalin Granules (KGs) and Lamellar

PAGE 19

19 Bodies (LBs). KGs are responsible for producing t he precursors required to form the envelope of the corneocytes in the SC. These precursors include K1, K10, profillaggrin, loricin and involucrin. Out of these, loricin and involucrin are involved in formation of the outer layer of the corneocyte whereas K 1 and K10 are involved in the cross linking of the inner layer of the cornified envelope. Lamellar Bodies are ellipsoid shaped components found in the stratum granulosum cell, the major components of which include phospholipids, cholesterol and acylglucosy lceramides. LBs are highly compressed structures in which the lipoidal components are packed to result in a bilayered arrangement. In addition to the above components, LBs also contain enzymes like phospholipase A2. At the junction between the stratum gran ulosum and SC the LBs are extruded in to the intracorneocyte space. Stratum Corneum SC can be further divided into two layers termed stratum compactum and stratum disjunctum. The upper layer, stratum disjunctum, is constantly undergoing desquamation. The l ower of the two layers, stratum compactum, has as much as double the amount of water per unit mass associated with it (30 % compared to 15%). These layers differ in the overall amount of amino acids and lipids. In addition, stratum compactum has a higher density of corneodesmosomes and, as the name suggests, is more compact and tightly held together. The extruded LBs are fused into bilamellar sheets which form the inter corneocyte lipids of the SC. Cellular remnants of the viable epidermis are packed tog ether in a cornified envelope to form the cellular components of the SC the corneocyte. This cornified envelope is stabilized by peptide cross glutamyl) lysine isopeptide glutamyl) polyamine linkages and disulphide bonds.

PAGE 20

20 Figure 11. Chemical Structures of Skin Ceramides There is considerable overlapping observed between the corneocytes. The water content of the SC is only about 15% by weight. The presence of bound water in the SC plays a crucial role in determining the mechanism of permeation of drug molecules through the SC. The major components of the intercellular lamellae include ceramides (50% by weight), cholesterol (30% by weight) and free fatty acids (10% by weight). Figure 11 shows the structures of skin cera mides isolated so far Ceramides are important structural components of the intercorneocyte lipids and are composed of sphingosine

PAGE 21

21 (2 amino4 octadecene 1, 3 diol) acylated at the 2amino group with a fatty acid. Intercellular lipids that are chemically li nk hydroxyceramides to the cornified envelope (via surface protein involucrin) occupy the hydroxyceramides are derived from acylglucoceramides i.e. ceramides 1, 4 and 9, via a deesterification reaction. The components of the intercellular compartments are pre sent as bilamellar sheets.1518 Bilamellar arrangement refers to the stacked organization of the flat plate like components in the space between two corneocytes. Swartzendruber et al. have shown that each cellular sheet is formed by the fusion of two bilayers in LBs. The corneocytes are further interconnected by polar components called corneodesmosomes rendering c onsiderable cohesive streng th to the SC layer. The overlapping of the corneocytes imparts a tortuous path through the lipid components of the SC. 5 The SC has a highly dens e heterogeneous microenvironment composed of covalently linked corneocytes and intercellular lipids which are present as fused bilamellar sheets at dif ferent phases of development. This uniquely rigid compartmental organization of the corneocytes and the i ntercellular lipoidal bilamellar sheets, gives rise to the barrier function of the SC. The formation of the SC which is semipermeable to topically applied drugs is therefore, a result of a series of biochemical events that lead to the conversion of viabl e (living) cells into a highly rigid, heterogeneous network of impermeable cornified envelopes and permea ble inter cellular lipids.18

PAGE 22

22 Strategies for Enhancing Topical Delivery Several techniques including permeation enhancers, iontophoresis, microneedles and prodrug strategies19 have been utilized to enhance per cutaneous absorption of drugs.10 Permeation enhancers alter the barrier property of the skin r endering it more permeable towards active components of the formulation. Some of the properties of an ideal penetration enhancer include predictable, reproducible and unidirectional effect while lacking pharmacological activity and toxicity related issues. Several classes of chemicals like oleic acid (OA), dimethyl sulphoxide (DMSO), azones, pyrrolidones and surfactants have been shown to possess a favorable penetration enhancement effect.20 However; there are hardly any penetration enhancers that have all the above mentioned ideal properties. Well known permeation enhancers like OA, DMSO have been shown to cause ir reversible damage to the skin.21 Even water has been shown to affect the barrier properties of skin, although the mechanism of this interaction are still under debate.20, 22 Some proposed mechanisms by which penetration enhancers work include, transient fluidization of the crystalline structure of stratum corneum, dissolution of stratum corneum lipid, improved solubility of the permeant in the skin and improved partitioning of the permeant into the skin. Iontophoresis involves movement of charged and uncharged molecules across the skin under the influence of a few milliamperes of current applied acros s a small area of skin.23 The applied electric current can induce an electric field in the skin micro environment, which drives charged molecules across the skin constit uting as electrophoretic flow.23 Application of a small voltage of current can also be used to generate an electroosmotic flow, which can aid passage of charged molecules across the skin. Iontophoresis is used to deliv er compounds like priolcarpine24 (as a diagnostic test f or cystic fibrosis) and lidocaine25 (local anesthetic)

PAGE 23

23 into skin. The microneedle strategy for transdermal delivery involves utili zation of an array of micronsized needles that penetrate the skin generating holes that can facilitate entry of small molecules drugs as well as macromolec ules into skin.26 This strategy has been successfully utilized to deliver, proteins27 and DNA28 in to skin. Optimizing Topical Delivery The observation that the SC is the primary barrier towards diffusion of permeants across skin was made as early as 1924.29 Further experi mental evidence came from a study published by I H Blank, in which he used tape stripped human skin to demonstrate that the water loss for the skin sample lacking stratum corneum was much h igher compared to normal skin.30 Additional work by R J Schuplein and I H Blank and others exemplified that diffusion through skin is a passive process. 30, 31 Since a molecule has to dissolve in the membrane microenvironment for the permeant to cross, it is reasonable to state that the extent of compatibility of the physicochemical properties of the membrane with that of the permeant dictates the permeants diffusion across the membrane. Since the SC has been shown to be composed of keratinized cells embedded in intercellular lipid bilayers t hrough which the permeant must pass, what are the most significant physicochemical properties of the permeant that affect its diffusion across the skin? Much effort has been made to identify these physicochemical properties. In one such attempt, diffusion of a homologous series of straight chain alcohols through human skin invitro was investigated by R J Schuplein and I H Blank.31 The choice of a homologous series was guided by the perceived usefulness of a continuum of solubility properties and molecular weights in the series. The flux (J, in units of mole cm2 h1) of alcohols when applied as neat liquids (C1C8) as saturated solutions in water (C6C10) was found to be inversely proportional to the molecular

PAGE 24

24 weight, with the most water soluble and the smaller member of the series exhibiting the highest J. However the authors correl ated permeability coefficient, P (defined as J / CVEH, concentration of the permeant in the applied vehicle) in units of cm h1 with the number of carbons in the alcohol, and the partition coefficient, KLIPID: AQ. T hey suggested that the most important predictor of diffusion of permeants through the skin was the partition coefficient. Thus, P was used as an important parameter for modeling diffusion of permeants through skin and mathematic al models for predicting percutaneous absorption, based on permeabilit y coefficients have been developed. However, Sloan and coworkers have shown that the amount permeated ( J, mole cm2 h1) and not P ( cm h1) of the permeant is the only clinically relevant parameter for development of mathematical models to optimize deli ve ry of a permeant through skin. 32, 33 Careful analysis of the results from the experiments by Schuplein and Blank also demonstrated that the solubility properties of the permeant and not the partition coefficient play the dominant ro le in governing the flux and the passive diffusion of a permeant through skin. Mechanism of Percutaneous Absorption of Drugs The fact that a drug molecule has to dissolve in the skin microenvironment for it to surpass the SC barrier renders its permeation a solubility driven process. In other words the ability of the permeant to dissolve in the SC intercorneocyte space will decide the maximum achievable flux, when a saturated donor phase is used and sink conditions are maintained. Therefore, the environment of intercor neocyte lipids is crucial in determining the mechanism of percutaneous absorption of drug molecules. It is generally accepted that the barrier to absorption of permeants through skin is primarily lipoidal in nature. Therefore enhanced lipid solubility of the permeant will result

PAGE 25

25 in better flux. The aqueous barrier to percutaneous absorption is believed by many to reside in the stagnant water layer existing on the SC surface for compounds delivered from an aqueous vehicle. However, in the experiments carried out by Schuplein and Blank it is clear, although not generally acknowledged, that the more water soluble members of the series, applied as pure liquids (hence no stagnant water layer) gave the best flux. Additionally, a dependence of J on water solubility of any member in a homologous series of more lipid soluble prodrugs has been quantified and discussed in detail in the next section. Although the dependence of J on aqueous and lipid solubility of a permeant is obvious, the exact locat ion of the aqueous layer has been a topic of investigation by many researchers. More recently, Sloan et. al. proposed a model for the organization of ceramides along with bound and unbound water in the SC. The bound water has been shown to be present in the corneocytes linked to the proteins and the unbound water is distributed uniformly throughout the SC. The bilayers of the intercorneocyte components are structured horizontally resulting in a parallel organization with the corneocyte surface. The arrang ement of the ceramides in the intercorneocyte space and the suggested path followed by a topically applied permeant to reach the lower layers of the skin has been shown in Figure 12. The presenc e of CER 6, with extended linolei c acid side chain bridged between a broad distinct crystalline layer and a more fluidlike narrow layer in the middle gives rise to a long periodicity phase (LPP) 13.4 nm in length. The presence of the LPP has been shown to be essential for the barrier properties of the intact SC. Th e unbound water layer (1.44 nm) in the SC is proposed to be present close to the polar head groups of PA and Ch and CER6. The concentration of the water

PAGE 26

26 in the SC has been calculated to be about 12.22 M (equivalent to 6.7 water molecules per ceramide). Figure 1 2. Pathway For Diffusion of a Topically Applied Permeant Based On The Biphasic Solubility Model. A topically applied permeant is expected to follow the path of least resistance which is through the ceramides and other lipoidal components of the fluid lipid layer in the LPP, explaining the dependence of J in lipid solubility of the permeant. However, for the drug to reach systemic circulation it has to pass the aqueous layer, with high water concentration. This explains the dependence of flux of a topically applied permeant on its aqueous solubility.

PAGE 27

27 Therefore, the mechanism of percutaneous absorption of a permeant is a solubility driven process where the higher lipid and aqueous solubility of the drug in the SC favors higher J. Model Development Initially, for the sake of simplicity, the microenvironment of the skin has been assumed to be homogeneous. The flux of a permeant across a homogenous membrane can be expressed by a mathematical equation, first proposed by Aldolf Fick in 1855, for the flow of heat, in which the flux across a homogenous membrane is directly proportional to the concentration gradient across the membrane (CM1CMn), where CM1 and CMn are the first few and the last layers of the SC in this case, J = D/L (CM1CMn) (1) and D (di ffusion coefficient) i s a proportionality constant L is the thickness of the membrane under sink conditions, (constant uptake of the permeant by the viable layer) and CMn is assumed to approach zero. CM1 can be estimated from equation (2) CM1 = (CVEH) (KM1: VEH) (2) Where KM1: VEH is the partition coefficient of the permeant between the first few layers of the SC and the vehicle ( VEH ). Since D = Do exp ( o is the diffusivity of a and V is the van der Waals volume eq uation (1) becomes equation (3). J = (Do exp ( (CVEH) (KM1:VEH) (3) One form of equation (3) was derived by Kasti ng, Smith and Cooper in 1987,34 l og J = log (Do/ L) + log SM1 (4) Where J is maximum flux (saturated solution applied) and CM1 is SM1. Equation (4) has been used to model data obtained from diffusion cell experiments where the permeants

PAGE 28

28 were applied as a saturated propylene glycol solution. The KSC eq uation assumes that the solubility of the permeant in a lipid like octanol, serves as an estimate for the solubility of the permeant in the membrane; i.e. SM1 = SOCT. The KSC model, therefore assumes the barrier to percutaneous absorption to be only lipoidal. Flux data for n = 36 compounds was fit to equation 4 and for a correlation of JM with SMEM and V, the r2 value was 0.74.34 The use of permeability coefficient, P, in equation (5), allows one to normalize the flux when different concentrations of the permeant in the vehicle are used. Theoretically, PVEH is assumed to a constant regardless of the CVEH. PVEH = J/CVEH (5) Further, when the vehicle is aqueous, KM1:VEH from equation (2) becomes (KOCT:AQ)f c and equation (3) becomes equation (6). PAQ = Do exp ( M1:AQ )f c (6) log P = log (Do/ L) + f log KOCT:AQ o MW+ log c ( 7) Equation (7), developed by Potts and Guy for permeants applied as aqueous suspen sions, was published in 1992. 35 Here f is the conversion fact or that accounts for difference between the membranevehicle partitioning and octanol water partitioning behavior. Two important inferences from Equation (7) arise; first that P is positively correlated to KOCT: AQ and inversely correlated to SAQ because SAQ is the denomination in the octanol water partition coefficient. Equation (7) also explains the positive correlation between P and carbon number and partition coefficient observed by Schuplein and Blank. Equation (7) gave an r2 = 0.83 for n = 42 phenols and alcohols for

PAGE 29

29 the correlation of P with KOCT: AQ an d MW. However a significantly poorer r2 (0.67) was obtained for the fit of the Flynn data base (n = 93) to the same equation.35 Transformed Potts Guy Equation The Robert S loan Equation One of the drawbacks of the Potts Guy equation is that it treats the permeability coefficient, P and KOCT: AQ as the most important predictors of percutaneous absorption. As mentioned above, the relevant parameter in terms of optimizing the d elivery of a permeant through skin is J not P. Partition coefficient being a ratio, reflects only the relative affinity of the permeant towards aqueous and lipid phases of the membrane. Therefore, equation (7) fails to represent the effect of absolute lipi d and aqueous solubilities of the permeant on its percutaneous absorption. Sloan et al observed that in a homologous series of prodrugs, the more water soluble member of a more lipid soluble series gave the best flux from a saturated iso pr opyl myristate (I PM) vehicle.19, 36, 37 Since the donor phase was a saturated IPM solution, J = JMIPM represents the maximum possible flux available. Additionally, flux did not continuously increase as a function of increased SIPM ( or K). In fact, the flux values in a homologous series in creases initially, then, with increasing SIPM it levels off and finally decreases for the most lipophilic members of the series. In a Wasdo plot this trend is easy to visualize (Figure1 3 ), because it represents the physicochemical properties (like log SAQ, SIPM, log J and K) as a function of carbon number. The initial shorter chain members of the homologous series were also the more water soluble ones. The best member of each series was the shorter chain, initial member, exemplifying the negative correlation between J and MW and a positive dependence on SLIPID as well as SAQ. In order to incorporate the effect of absolute aqueous and lipid solubilities, SAQ and SLIPID of the permeant on its diffusion through skin, Equation (7) was modified as follows:

PAGE 30

30 KM1: IPM = KM1: AQ / KIPM: AQ KM1: AQ = (KIPM: AQ)f c KM1: IPM = (KIPM: AQ)f c / KIPM:AQ (8) Figure 13. Wasdo Plot. Trends in Fluxes and Permeability Coefficients from a Lipid Vehicle versus an Aqueous Vehicle (log JMMIPM and log PMIPM versus log JMMAQ and log PMAQ, respectively) for Prodrugs Compared to their Partition Coefficients (log KIPM:AQ). Taking the logs, substituting equation (8) into equation (7) and substi tutin g solubilities for K gives: log PIPM = log (Do/L) + f log SIPM f log SAQ log SIPM + log SAQ o MW+ log c Collecting similar terms and adding log SIPM on both sides gave : log J = log (Do/L) + f log SIPM (1 f) log SAQ o MW + log c (9) Substitut ion of y for f, collecting constants into a new constant x, gives: log JMIPM = x + y log SIPM + (1 y) SAQ z MW (10)

PAGE 31

31 Equation ( 10 ) is the Robert Sloan (RS) equation which correlates flux of a permeant from a saturated lipid vehicle, IPM, across a biological membrane with its SAQ, SLIPID and MW.38 The RS equation has been used to predict flux of permeants delivered from saturated aqueous suspensions as well. The form of the equation remains the same regardless of the vehicle.39 A comparison of a fit of flux data for the same n = 32 compounds through hairless mouse skin invitro from water as a vehicle, to K SC, equation(4), and RS equation, equation(10), gave a r2 value of 0.76 vs 0.90.40 The fit of the Flynn data base to the RS equation also gave a much better r2 ( 0.93 ) compared to that obtained with the KSC equation ( r2 = 0.83) .41 The KSC, equation (4) and the Potts Guy model for percutaneous absorption represent the SC as a lipoidal barrier whereas the RS model represents the SC barrier as a biphasic barrier. Therefore, the absolute aqueous and lipid solubilities, SAQ and SLIPID, and MW of a molecule are the statistically significant predictors of its flux, J. The utilization of the RS equation for predicting flux through biological membranes like hairless mouse skin and human skin has been reviewed in detail. 33 Although a dependence of J on SAQ in a homologous series of prodrugs was published i n a review as early as in 1989 ,37 it was only in 1999 36 that a quantitative expression reflecting this correlation was developed. The RS equation quantifies the dependence of J on SAQ, SLIPD and MW. The extent and the nature of the correlation (positive or negative) between the independent (SAQ, SLIPID, MW) and the dependent variable (J) is represented by the mag nitude of y and z values. T he fit of the flux database comprising n = 32 compounds to the RS equation40 gave x = 2.30, y = 0.575, z = 0.0016 and r2 = 0.90; suggesting an almost equal contribution by aqueous and lipid solubility terms. Therefore, in terms of prodrug design, a promoiety that

PAGE 32

32 facilitates a maximum improvement in SAQ as well as SLIPID without considerably increasing the MW of the corresponding prodrug must be ideal to achieve better flux through the skin compared to the parent drug. Prodrug Strategy to Enhance Percutaneous Absorption of Drugs A prodr ug is an inactive, chemically functionalized derivative of the parent drug, which upon chemical or enzymatic hydrolysis regenerates the pharmacologically active drug form. The derivatization of drugs is not a new approach and has been widely utilized to solve problems associated with oral delivery of drugs like poor aqueous and lipid solubility, first pass effects and poor absorption across biological membranes, among many others. 42 Figure 14. Mechanism of hydrolysis of fosphenytoin The success of prodrug strategy is refl ected by the fact that about 15 % of marketed drug in the US are prodrugs. One such example is fosphenytoin, a disodium phosphate ester of phenytoin that is used for short term treatment of epilepsy. Phenytoin is a weakly acidic and a high melting (293 oC) drug. Its poor water solubility (0.02 mg/ml) leads to an erratic bioavailability profile in humans. The prodrug was found be about 4408 times more water soluble, and was successful in enhancing oral bioavail ability of phenetoin (Figure 14 ). 43 Fosphenytoin was shown to be enzymatically labile with half lives 30 min and <30s in rat whole blood and rat liver homgenates

PAGE 33

33 respectively.43 A proposed mechanism of hydrolysis of fosphenytoin to phenyt oin has been shown in Figure 14 The rate determining step is the enzymatic cleavage of the prodrug by phosphatases generating a chemically labile species, which undergoes a spontaneous loss of formaldehyde to regenerate the parent drug. The rate of hydrolysis of the h ydroxymethyl compound is dependent on the pKa of the leaving group. Bungaard et al. quantified the relationship between the leaving group pKa and the half lives of the corresponding hydroxymethyl compounds.42 Figure 15. Cyclosporine Prodrugs. (A) Chemical structure of Cyclosporine A (CsA); (B) Mechanism of chemical hydrolysis of the octaarginine conjugate of CsA at pH 7.4. More recently, poly arginine based prodrugs of a macrmolecular peptide immunosuppressant, Cyclosporine A (CsA) have been shown enhance its delivery through mouse and human skin.44 The prodrug comprised an octaarginine (R7) chain linked to the drug via a pH dependent, chemically reversible amino hexanoic acid linker, which regenerated the parent drug at physiological pH (Figure14). The diffusion of the polypeptide conjugates was shown to be dependent on the membrane potential and the

PAGE 34

34 number of basic arginine groups (protonated at physiologi cal pH) in the peptide chain. The mech anism of cellular uptake of polyarginine conjugates of CsA was shown to be a cha rge dependent, active process.45 Although most of the prodrugs evaluated as transdermal or topical delivery agents permeate the skin via a passive process, the example of polyarginine CsA conjugates has broadened the scope of topical delivery to high molecular weight compounds. A simple analysis of any prodrug form allows one to visualize that it is a twocomponent system: the parent drug (DRUG X ) and the promoiety ( PRO ). The parent dru g is derivatized at one (or more) of the functional (enabling) groups (X) with the promoiety (PRO). It is obvious that the physicochemical properties of the prodrug form will be dependent on the nature of PRO. Table 11. General representation of a prodrug, an acyl prodrug and a soft alkyl prodrug. DRUG X PRO General representation of a Prodrug DRUG X C(=O) R Acyl Prodrug DRUG X CH 2 Y C(=O) R Soft A lkyl Prodrug Two general cases arise: Acyl prodrugs, in which the PRO is attached directly to X and Soft alkyl prodrugs in which, X is attached through a methylene linker to a heteroatom, Y (O, S etc). Several comprehensive reviews on application of prodrug techniques in solving drug delivery related problems have appeared in the literature.46 In the previous section, the importance of SAQ, SLIPID and MW of a permeant as statistically significant predictors of flux ( J) across the skin has been described. A few examples of each prodrug class and the utility of the prodrug approach in modulating the solubility properties and consequently their flux through skin is discussed below.

PAGE 35

35 Acyl Prodrugs As the name suggests, in an acyl prodrug the enabling group is attached to the heteroatom on the drug through a carbonyl bond and as, mentioned above, the nature of the R group plays an important role in modulating the physicochemical properties of the prodrug. Some of the most co mmon functional groups targeted for prodrug approach and the corresponding acyl prodrug formed are: Table 12. Acyl prodrugs formed from hydroxyl or amine containing drugs General Chemical Structure Prodrug Type DRUG OH DRUG O C(=O) R Ester DRUG OH DRUG O C(=O) O R Carbonate DRUG OH DRUG O C(=O) NH R Carbamate DRUG NH2 DRUG NH C(=O) R Amide DRUG NH2 DRUG NH C(=O) O R Carbamate There are several examples of acyl prodrugs of drugs co ntaining a hydroxyl group.19 Narcotic analg esics, like morphine, are used widely for management of acute pain. Most compounds in this chemical class that are currently in clinical use are usually administered orally or via parenteral routes. The oral and parenteral routes of administration are associated with problems like high peak plasma concentration, requirement of frequent dosing and suboptimal oral bioavailability of the drug due to extensive fir st pass metabolism.47 Bundgaard et al. investigated a series of 3O alky l carbonyl and 6O alkyl carbonyl ester prodrugs for enhancing delivery of morphine through skin. Morphine itself has a poor skin permeation profile which can be attributed to its low water and lipid solubility. The prodrug approach can help circumvent these challenges. Morphine is only marginally soluble in IPM (SIPM = 0.0 23 mg ml1) and pH 7 phosphate buffer (SAQ = 1.8 mg ml1). 47

PAGE 36

36 Figure 16. Testosterone and Naltrexone Prodrugs. (A) Chemical Structures of Testosterone (TS) and its N, N dimethylamino butyric acid ester acyl prodrug; (B) Chemical structure of Naltrexone (NTX) and its previously investigated p rodrugs. The 3 propionyl ester that exhibited a higher aqueous (SAQ = 3.6 mg ml1) and lipid (SIPM = 41 mg ml1) solubility, showed a flux of 8.7 g cm2 h1 through human skin invitro from a saturated IPM solution. A steady sate flux of up to 35 g cm2 h1 of morphine was achieved by its 3hexanoyl ester prodrug (SIPM >150 mg ml1; SAQ = 2.6 mg ml1) applied as an IPM solution (200 mg ml1). Only morphine was detected in the receptor phase indicating complete conversion of the prodrug to the parent drug.

PAGE 37

37 Morphine itself was undetectable (less than 0.01 g cm2 h1) under the same conditions.47 Since, saturated solutions were not used in the donor phase in all cases, the observed flux values do not reflect the highest possible flux of morohine achievable with these prodrugs. Similar results were obtained with alkyl oxy carbonyl esters of ketobemidone where the most lipid and water soluble member of the series exhibited higher fl ux compared with ketobemidone.48 Simple acyl prodrugs (straight chain49 and branched chain50 3 O alkyl carbonyl e sters and 3O alkyloxy carbonyl esters51) of NTX showed a moderate enhancement of flux of the par ent drug (~27 fold). 3O Alkylaminocarbonyl prodrugs of NTX exhibited a 3 4 fold improvement in flux of t he parent drug (Figure 16).52 Similarly, alkyloxyc arbonyl (AOC) prodrugs of APAP have been investigated in diffusion cell experiments in our lab previously. None were more water soluble than APAP. However the two more water soluble members of the AOC prodrugs of APAP (Figure 17) exhibited a moderate e nhancement in flux (>2 times).53 Misolovich et. al.54 investigated testosteronyl 4 dimethylaminobutyrate HCl (TSBH) as potential prodrug candidate for enhancing topical delivery of testosterone (TS) (Figure 1 6). Solubility of TBSH was about 1000 times greater t han TS in pH 7.0 phosphate buffer (>150 mg ml1 for TBSH compared to 0.01 mg ml1 for TS). The flux of TSBH from a 10% solution was about 60 times higher than TS itself.55 Even the free base form of TSBH delivered 35 times more TS through human skin invitro. The protonated amine group in the promoiety has been infered to confer a more favorable balance in lipid and aqueous solubility to the prodrug than simple alkyl promoi e ties by

PAGE 38

38 increasing its aqueous solubility and hence enhancing skin penetration rates. But at the pH of highest solubility achieving optimal stability has been a challe nge. Figure 17. Chemical structures and IPM and aqueous solubilites of (A) APAP; (B) C2AOC APAP; (C) C2 ACOM APAP; (D) C2 AOCOM APAP; (E) C2 NANAOCAM APAP When an ionizable amine group is incorporated in the acyl portion of the ester, it activates the acyl group towards nucleophilic attack by hydroxide ion due to two reasons 56, 57( a ) strong electron withdrawing ( I) effect of the protonated amine group and; (b) i ntramolecular catalytic effect of the protonated and unprotonated amine group (general base catalysis) Basic hydrolysis of amino acid esters have been inv estiga ted in great deail. 57 Steric and electronic factors and pH have been shown to affect the rates of hydrolysis. The acyl prodrugs hydroly ze to regenerate the parent drug via the mechanism shown in F igure 18.

PAGE 39

39 Vitamin E, a phenol containing compound, is a powerful antioxidant and has been shown to play a role in maintaining general physiological health of the skin. Figure 18. Mechanism o f of Hydrolysis of Acyl Prodrugs. (A) a cid and (B) b ase catalyzed mechanisms of hydrolysis of acyl prodrugs. Vitamin E acetate, an ester prodrug, is a part of many commercial cosmetic formulations. The poor SAQ and high MW, however, limit its permeation through skin. In recent reports, the skin retention of eight amino acid ester prodrugs o f Vitamin E (Figure 1 9) were evaluated.58 Since the prodrugs exhibited low solubility in buffer (pH 8.0, phosphate), solubilities were evaluated in pH 8 phosphate buffer contai ning 5% The enzymatic (porcine liver esterase) half life of the glycine derivative was 49.5 min whereas the most hydrophilic member, the Lcitruline prodrug showed a half life of 1013 min.58

PAGE 40

40 All the prodrugs showed higher water solubility than Vitamin E itself and some of the derivatives underwent hydrolysis at a rate that could ensure regeneration of the parent drug in the skin. Figure 19 Chemical structures of V itamin E and its prodrugs evaluated previously. The skin retention of the prodrugs was higher than Vit E itself. However little or no parent drug could be detected in t he skin and the receptor phase.58 The low J of Vitamin E prodrugs was expected and can be explained by the negative correlation of the MW with flux. The high MW, low SAQ and extremely high SLIPID (a very poor balance in aqueous and lipid solubility) of Vitamin E itself render it a challenging molecule to deliver through skin. To achieve an improved J of Vitamin E through the skin by its prodrugs, the increase in aqueous solubility may have to be dramatic without increasing the MW of the compound to an appreciable extent. However, the amino acid promoiety does show promise to improve solubility properties of low er molecular weight drugs.54, 55

PAGE 41

41 Soft Alkyl Prodrugs A s oft alkyl prodrug has a methylene or a vinylygous methylene linker between the drug heteroatom and a heteroatom in the promoiety. Soft alkyl prodrugs of heterocyclic amines like 5 FU, Theophylline (Th), and 6mercaptopurine (6MP) and phenolic drug like APAP have been evaluated for the p otential to enhance topical delivery of their parent compounds. Soft alkyl drug can undergo an esterase mediated reaction to generate a hydroxymethyl intermediate, which undergoes spontaneous chemical hydrolysi s to regenerate the parent drug (Figure 1 10).19 One such example discussed previously is fosphenytoin (Figure 13). Figure 110. Mechanism of hydrolysis of a soft alkyl prodrug. N dialkylamino) alkyl derivatives of molecules containing acidic NH groups have been shown to possess enhanced water as well as lipid solubility. The masking of the polar acidic group results in an increase in lipid solubility whereas the aqueous solubility is increased by the incorporation of a basic amine group into the promoiety. For example, among a series of N Mannich derivatives of 5FU one member, bis (diethylamino)methyl derivative, gave a higher flux than the commercially available formulation of 5FU, Efudex. A 10% suspension of the bis (diethylamino)methyl prodrug in IPM gave a flux of 0.11 0.013 mg cm2 h1 whereas Efudex gave a flux of 0.017 0.046 mg cm2 h1 through hairless mouse skin. Because of difficulty in purificat ion,

PAGE 42

42 many members of this series were not evaluated in diffusion cell experiments.19 Nevertheless, the bis (diethylamino) methyl prodrug of 5FU clearly exemplifies the effect that incorporation of a basic amine group into the promoiety has on the biphasic solubility, and consequently on permeation of the corresponding prodrug. The mechanism of hydrolysis of N Mannich base prodrug involves unimolecular cleavage to parent NH acid ic compound and a carbocation 1 9(Figure 1 11).The lower the pKa of the parent NH compound, the lower is the half life of its N mannich prodrug. An equation for predicting the half life of the N Mannich prodrug based on the pKa of the parent NH comp ound has also been developed. 42 Figure 111. Mannich Base Prodrugs of 5 FU. (A) Chemical structure of 5FU; (B) N Mannich base prodrug of 5FU; (C) Mechanism of hydrolysis of bis (diethylamino) methyl prodrug of 5FU. The s oft alkyl approach has also been applied to phenols. A lkylcarbonyloxymethyl (ACOM), 59 alk yloxycarbonyloxymethyl (AOCOM)60 and N alkyl N alkyloxycarbonylaminomethyl ( NANAOCAM ) 61prodrugs of a model phenol acetaminophen (APAP), were investigated in diffusion cell experiments (Figure 17). The flux of the most water soluble member of the ACOM series was only 3.6 times

PAGE 43

43 higher than APAP and only one member of AOCOM prodrugs series (4AOCOM APAP, C2) gave higher flux (1.6 times) than APAP itself. Although the soft alkyl approach has been utilized successfully to generate a synthetic handle on parent drugs, and hence broaden the choice of functional groups that can be incorporated into the promoiety, their performance was not significantly better than simple acyl derivatives of APAP. Conclusions Although deliv ery of drugs through the oral route is preferred over other routes of drug administration, it is associated with problems like first pass effects, GI and liver toxicity and poor absorption of drugs through the gut wall. Topical delivery of drugs is a pract ical alternative with which the common challenges of oral delivery can be overcome. Among several available methods for enhancing topical delivery of drugs, the prodrug approach is the one with which optimization of maximum achievable flux of a drug throug h skin is feasible. The statistically significant predictors of flux of perm eant (J) have been elucidated.62 Among a homologous series of prodrugs, the initial, low molecular weight members exhibiting the best balance between SAQ and SLIPID have been shown to give the best flux. 36 Therefore the relevant design directives for developing prodrugs intended for enhanced topical delivery of its parent drug includes, (a) higher aqueous as well as lipid solubility relative to the parent drug without a considerable increase in the MW of the corresponding prodrug and (b) ability to regenerate the parent drug in the skin (dermal delivery) or in systemic circulation (transdermal delivery). AOC53 (acyl), ACOM,59 AOCOM60 and NANAOCAM 61 (soft alkyl) prodrugs of a model phenolic drug, APAP have been evaluated in our lab previously in diffusion cell experiments. However, enhancement in J was only minimal. In view of the pr ev ious

PAGE 44

44 research on the effect of inclusion of an ionizable group in the promoiety it seemed appropriate to investigate the development of prodrugs of poorly water soluble phenolic drugs with improved aqueous sol ubility and hence delivery properties. In the present study an efficient synthetic ro ute to N, N` dialkylaminoalkyl carbonyl (DAAC) and aminoalkyl carbonyl (AAC) ester prodrugs of the phenolic drug acetaminophen will be presented. The hypothesis is that the incorporation of the ionizable amine group w ill confer to the prodrugs a higher SAQ as well as SLIP I D, and hence a higher J than APAP. If successful these compounds can serve as models for optimization of the DAAC and the AAC prodrug approach for improving topical delivery.

PAGE 45

45 CHAPTER 2 SPECIFIC OBJECTIVES First Objective The first objective of this study was to synthesize a seri es of N, N` dialkylaminoalkyl carbonyl (DAAC) and aminoalkyl carbonyl (AAC) prodrugs of a model phenolic drug Although, acyl and soft alkyl prodrugs of a model phenol, acetam inophen (APAP) investigated previously in our laboratory in diffusion cell experiments gave a moderate enhancement in SIPM, none of the prodrugs were more water soluble than APAP. For example the most water soluble member of the AOC series was C1AOC APAP (21 mM) compared to 100 mM for APAP. Similarly, among the ACOM and NANAOCAM (soft alkyl) prodrugs of APAP the most water soluble member was C2 ACOM APAP ( 25 mM) and C1NANAOCAM APAP ( 45 mM) respectively. The flux enhancement could be attributed primarily t o an increase in SIPM o f the prodrugs relative to APAP. The hypothesis for this study is that the incorporation of an ionizable amine group ( NR1R2 or NH2) into the acyl prodrug would result in a favorable increase in SAQ and SLIPID of the corresponding prodrug. The effect of incorporation of an ionizable amine into the acyl prodrugs of APAP on its skin permeation has not been reported. Therefore we synthesized DAAC and AAC prodrugs of APAP which can act as model compounds for optimization of the DAAC and AAC prodrug approach. Synthetic rou tes to N, N` dialkylaminoalkyl carbonyl (DAAC) and aminoalkylcarbonyl (AAC) acyl prodrugs of a phenol ic drug a cetaminophen (APAP) have been presented. DAAC APAP prodrugs were synthesized via a three step pr ocedure starti ng with haloalkyl carbonyl esters which were reacted with five di fferent amines: N, N` dimethylamine, N, N` diethylamine, N, N` dipropyl amine, morpholine

PAGE 46

46 and piperidine. The spacing between the amino group and the carbonyl group of the acyl group was 1 t o 3 CH2 groups After the hydrolysis of the ester the promoiety was subsequently coupled with the parent drug via a dicyclohexyl carbodimide (DCC) mediated coupling to yield the DAAC APAPHCl prodrugs in excellent yields. The DAAC APAP prodrugs (free base form) were prepared by treatment of the corresponding HCl salts with aqueous base. The AAC prodrugs were synthesized using commerc ially available Boc protected amino acids using DCC or EDCI as coupling agents. Second Objective The second objective of thi s research was to investigate whether the DAAC and AAC prodrugs can improve topic al delivery of APAP. Hairless mouse skin was used for in vitro analysis of the prodrugs. Physicochemical properties, i.e., lipid and aqueous solubility were determined for the DAAC and AAC prodrugs. Additionally, to investigate the extent to which the prodrug might hydrolyze during the course of diffusion cell experiment, half lives (t1/2) of a few members o f the DAAC and AAC series wer e measured in buffer (pH 6.0, 20 mM). Thi rd Objective The third objective of this study is to develop a new Robert Sloan equation using an integrated solubility and flux database for compounds evaluated in diffusion cell experiments previously in our lab. The new RS equation will be developed by fitting the solubility and flux data for n = 73 compounds comprising n = 42 compounds from Roberts et. al. ,36 the C5 Bis 6, 9 ACOM 6MP prodrug 53 n = 4 3 AC 5FU prodrugs 63 n = 6 A COM 5FU prodrugs 64 n = 2 bis AC 5FU prodrugs 65 n = 8 AOC APAP prodrugs ,53 n = 5 ACOM APAP59 prodrugs and n = 5 AOCOM APAP60 prodrugs. The

PAGE 47

47 new RS equation will be used to predict the flux of DAAC APAP prodrugs through hairless mouse skin.

PAGE 48

48 CHAPTER 3 N, N` DIALKYLAMINOALKYL CARBONYL (DAAC) PRODRUGS OF A MODEL PHENOLIC DRUG ACETAMINOPHEN (APAP) Introduction First pass effects in the GI tract and the liver limit the oral bioavailability of a drug containing a hydroxyl group (phenols or alcohols). The presence of the hydroxyl group however facilitates the derivatization of the drug to give prodrugs. Depending on the choice of the target s ite and route of administration, the prodrugs can be designed with appropriate physicochemical properties. Presence of one or more polar hydroxyl groups in a drug may result in low lipid solubility. Masking the polar hydroxyl group results in a better bala nce in aqueous and lipid solubilties This leads to an enhanced absorption of orally administered drugs. For example, an enhanced oral bioavailability of terbutaline in dogs was achieved by its carbamate esters. A single oral dose of the bis N, Ndimethylcarbamate of terbutaline could produce sustained blood levels of the parent drug and show ed a plasma half life of 10 h.42 An ideal promoiety for developing prodrugs of poorly water soluble drugs co ntaining a hydroxyl group is one that can increase its water solubility without compromising lipid solubility. In this regard, a promoiety containing an amine group has been utilized successfully to improve aqueous solubility of poorly water soluble drugs. 55 The enhanced aqueous solubility is a direct consequence of a distinct physicochemical feature of prodrugs that contain a basic amine (primary, secondary or tertiary ) functionality ionization of the basic amine nitrogen at physiological pH. The presence of an amine group activates the acyl group towards hydroxide ion attack because of the ( I) effect of the amine group and promotes its intramolecular gener al base c atalyzed hydrolysis.46, 56, 57 Kinetic behavior of esters containing an

PAGE 49

49 activated acyl group (metranidazole N, Ndimethyl glycinate ester,42 hydrocortisone lysine ester66 aspartic a aspartic acid esters67) has been studied in detail. However, attaining optimal stability in aqueous solutions has been a challenge towards developing prodrugs containing an ionizable, basic amine functionalit y. An example, where the incorporation of an ionized amine into the promoiety has been utilized to enhance percutaneous absorption of a lipophilic poorly water soluble drug is testosterone (TS). The TBSH prodrug showed a 60 times higher flux through human skin in vitro than TS when applied as a 10% solution in pH 7.0 phosphate buffer. It has been hypothesized that the enhancement in flux is a consequence of an increase aqueous solubility of TBSH relative to TS.55 Only a few reports describing synthesis of acyl prodrugs of phenolic drugs in which a basic amine group was incorporated into t he promoiety have appeared in the literature. A synthetic route for obtaining N, N` Dialkylaminoalkylcarbonyl (DAAC) prodrugs of APAP with different amine groups and the alkyl chain length have not been reported. Additionally, no systematic evaluation of t he effect of pKa and the distance from the acyl group of the amine on the solubility and skin permeation proper ties of N, N dialkylaminoalkyl carbonyl (DAAC) prodrugs of phenolic drugs has been carried out. A detailed study of the solubility, stability and in vitro permeation behavior of DAAC prodrugs of a model phenolic drug can aid the design of DAAC prodrugs of other poorly water soluble phenolic drugs. A low molecular weight phenolic drug, acetaminophen, was chosen to optimize the DAAC prodrug approach.

PAGE 50

50 Synthesis of DAAC Prodrugs of Acetaminophen In the present study, a series of DAAC prodrugs of APAP ( 5a 5k) were synthesized. The choi ce of the amine incorporated into the promoiety was guided by two factors: (a) pKa of the amine and (b) steric bulk on the amine group. Five different amines, N, Ndimethylamine, N, N diethylamine, N, N di propyl amine, piperidine and morpholine with pKa values between 10.2 8.3 were chosen as candidates for the synthesis of the amino acids (3b3k). Figure 31. Chemica l structures of synthesized DAAC APAP prodrugs Materials and Methods. Melting points were determined on a Meltemp melting point apparatus. Thin layer chromatography (TLC) plates (Whatman AL Sil G/UV) were

PAGE 51

51 purchased from Fisher. N, N diethylamine, N, N di propyl amine, piperidine, morpholine, ethyl 3(N, N dimethylamino)propionate (1b), N, N dimethyl glycine hydrochloride (3a), 3 (N, N diethylamino)propionic acid hydrochloride (3d), ethyl chloroacetate, m ethyl 3bromo propionate and dicyclohexylcarbodiimide (DCC) were purchased from Sigma Aldrich. Ethyl 4 bromobutyrate was purchased from Alfa Aesar. Spectra (1H NMR) were recorded on a Varian Unity 400 MHz spectrometer Pyridine, acetonitrile and dichloromethane were dried over 3 pKa values were estimated using pKa software, (ACD labs, version 12.0). Experimental The DAAC prodrugs of APAP (5a 5k ) were synthesized in three steps ( Figure 3 1). A ester was reacted with a secondary amine to obtain the corresponding N, N dialkylaminoalkanioc acid ester (2b2h). These esters were hydrolyzed with concentrated hydrochloric acid to yield the hydrochlori de salts of th e amino acids (3 a 3k), which were subsequently coupled with APAP by a dicyclohexy l carbodimide (DCC) m ediated esterification to give the hydrochloride salts of the desired prodrugs (4a 4k). The prodrugs (5a 5k) were obtained from the corresponding hydrochloride salts by treatment with ice cold saturated aqueous Na2HCO3 solution and immediate extraction with dichloromethane. General Method for the synthesis of compounds 2c and 2e 2k To a round bottom flask containing a well alkanoic acid ester (1 equiv.) in 25 ml acetonitrile was added, 2 equiv. of the amine dropwise. The contents of the flask were refluxed for 810 h. NMR of the crude reaction mi xture confirmed completion of the reaction. T he flask was cooled in an ice bath for 1 2 h.

PAGE 52

52 Figure 32. Synthesis of DAAC APAP prodrugs The resulting suspension was trit rated with 100 150 ml diet hyl ether to crystallize the HCl, HBr or HI (in case of 1 c and 1e) salt of the corresponding amine. The solid amine salt that formed was removed by filtration and the filtrate was concentr ated on a rotary e vapor ator to give the desired aminoalkanoic acid esters as oils. In the case of 1c and 1e (sterically hindered amines) the alkylating agent, ethyl chloroacetate had to be converted to ethyl iodoacetate by tr eatment with 1 equi v of NaI in 10 ml acetone prior to coupling with the amine otherwise the reaction did not go to completion. Ethyl io doacetate was not is olated or characterized. No formation of R1R2N(CH2)nC(=O)NR1R2 was observed in this reaction. All the esters were obtained as yellow to orange c olored oils in good yields (83 98 %). 2c Ethyl 2(diethylamino) acetate. Synthesized from 4.6 g (0.03 mole) ethyl chloroacetate and 5.59 g (0.07 mole) diethylamine to give a brown oil. NMR (CDCl3 4.18 (q, 2H), 3.31 (s, 2H), 2.66 (q, 4H), 1.27 (t, 3H), 1.06 (t, 6H). Yield: 95%.

PAGE 53

53 2e Ethyl 2(dipropylamino) acetate. Synthesized from 10.30 g (0.08 mole) ethyl chloroac etate and 17.1 g (0.16 mole) di propylamine to give a brown oil. 1H NMR (CDCl3 6H) Yield: 86% 2f Ethyl 2 (piperidin1 ` yl) acetate. Synthesized from 4.1 g (0.03 mole) et hyl chloroacetate and 5.7 g (0.06 mole) piperidine to give a pale pink oil. 1H NMR (CDCl3): 3H). Yield: 98% 2g Ethyl 3 (piperidin1 ` yl) propanoate Synthesized from 1.53 g (0.009 mole) of methyl 3bromopropionate and 1.532 g (0.02 mole) piperidine to give a pale yellow oil. 1H NMR (CDCl3 (quin, 4H), 1.42 (m, 2H). Yield: 98% 2h Ethyl 4 (piperidin1 ` yl)butanoate Synthesized from 4.45 g (0.02 mole) ethyl 4 bromobutyrate and 4.855 g (0.04 mole) piperidine. 1H NMR (CDCl3 2.372.27 (m, 8H), 1.814 (quin, 2H), 1.57 (m, 4H), 1.421 (m, 2H), 1.254 (t, 3H).Yield: 86%. 2i Ethyl 2 ( morpholin 4 ` yl) acetate. Synthesized from 2.3 g (0.018 mole) of ethyl chloroacetate and 3.27g (0.037 mole) morpholine to give a yellow oil. 1H NMR (CDCl3): 2j Ethyl 3 ( morpholin 4` yl ) propanoate Synthesized from 4 g (0.02 mole) of methyl 4 bromopropionate and 4 g (0.04 mole) of morpholine to give a brown oil 1H NMR (CDCl3) 98%

PAGE 54

54 2k Ethyl 4( morpholin4` yl) bu tanoate Synthesized from 4.6 g (0.02 mole) ethyl 4 bromobutyrate and 4.1 g (0.04 mole) morpholine. 1H NMR (CDCl3) 3.68(m, 4H), 2.415 (m, 2H), 2.3142.361 (m, 4H), 1.80 (quin, 2H), 1.254 (t, 3H).Yield: 94% General method for the synthesis of compounds 3b 3c, 3e 3k The esters 2 b, 2 c and 2 e 2 k were refluxed with excess (5 10 equiv.) of conc HCL for 810 h after which the water was evaporated. The residue was triturated with 5075 ml THF overnight. The obtained solids were dried in a vacuu m oven at 40 compounds 3b, 3c and 3e3k as crystalline solids. 3b 3 (dimethylamino) propanoate hydrochloride 1H NMR (D2 t, 2H), 2.89 (broad s, 8H). Yield: 98% 3c 2(diethylamino) acetic acid hydrochloride 1H NMR (D2O): 3.27 (q, 2H), 1.29 (t, 6H). Yield: 68% 3e 2(di propylamino) acetic acid hydrochloride 1H NMR (D2 3.02 (q, 4H), 1.57 (sextet, 4H), 0.805 (t, 6H) Yield: 68% 3f 2 (piperidin1 ` yl) acetate hydrochloride 1H NMR (D2 (s, 2H), 3.57 (broad d, 2H), 3.0 (td, 2H), 1.451.93 (m, 6H). Yield: 100% 3g 3 (piperidin1 ` yl) propanoate hydrochloride 1H NMR (D2 doublet, 2H), 3.392 (t, 2H), 2.965 (td, 2H), 2.87 (t, 2H), 1.461.99 (m 6H). Yield: 94% 3h 4 (piperidin1 ` yl)butanoate hydrochloride .1H NMR (D2 doublet 2H), 3.098 (broad triplet, 2H), 2.9 (broad triplet, 2H), 2.466 (2H, t), 1.9831.4 (m, 8H). Yield: 80%

PAGE 55

55 3i 2 ( morpholin4` yl) acetate hydrochloride. 1H NMR (D2O): 3.8 4.2(m, 2H), 3.11 (s, 2H), 3 .85 (broad t, 2H), 3.53.65 (m, 2H), 3.25 (broad t, 2H). Yield: 82.9% 3j 3 ( morpholin4` yl) propanoate hydrochloride 1H NMR (D2O) 2H), 3.8 (td 2H), 3.52 (broad doublet 2H), 3.44 (t, 2H), 3.2 (td, 2H), 2.95 (t, 2H). Yield: 82%. 3k 4(morpholin4` yl) butanoate hydrochloride 1H NMR (D2O) doublet 2H), 3.80 (t, 2H), 3.56 (broad d, 2H), 3.54 (broad doublet, 2H), 3.153.24 (m, 4H), 2.5 (t 2H), 1.982.15 (m, 2H). Yield: 96%. General method for the synthesis of 4a 4k Equimolar quantities of APAP, dialkylamino alkanoic acid hydrochloride (3a 3k) and DCC in 35 40 ml dry pyridine were stirred at room temperature for 24 48 h. T he suspension was triturated with about 75 ml dichloromethane and 25 ml diethyl ether. The suspension was stirr ed vigorously for 1 h. Then, the suspension was filtered and the solid obtained was washed thoroughly with dichloromethane till no smell of pyridine remained. The filtered solid was dried under vacuum for a few hours and then was refluxed with 250350 ml c hloroform for 4 h. The resulting suspension was filtered while hot to yield 4a 4k as white solids, while the dicyclohexyl urea remained in the filtrate. 4a 4A cetamidophenyl 2 ` (dimethylamino) acetate hydrochloride, (Me2n1APAP HCl) Synthesized fro m 1 g ( 0.006 mole) APAP 0.92 g (0.006 mole) 3a and 1.36 g (0.006 mole) DCC. 1H NMR (D2O) (s, 2H), 3.04 (s, 6H), 2.14 (s 3H).Yield: 75% 4b 4 A cetamidophenyl 3 ` (dimethylamino) propanoate hydrochloride (Me2n2APAP HCl) Synthesized from 1.48 g (9.8 mole) acetaminophen, 1.5 g (9.8 mole) 3 b

PAGE 56

56 and 2.02 g (9.8 moles) DCC. 1H NMR (D2O) 3.02 (t, 2H), 2.94 (s, 6H), 2.14 (s, 3H). Yield: 85% 4c 4A cetamidophenyl 2 ` (diethylamino) acetate hydrochloride, (Et2n1APAP HCl) Synthesized from1.25 g (0.008 mole) APAP 1.70 g (0.008 mole) DCC and 0.008 mole 3c. 1H NMR (D2O) 2H), 3.39 (4H, q), 2.15 (s 3H), 1.347 (t, 6H). Yield: 70% 4d 4 Acetamidophenyl 3` (diethylamino) propanoate hydrochloride (Et2n2APAP HCl) 1H NMR (D2O) (t, 2H), 2.15 (s 3H), 1.31 (t, 6H). Yield: 80% 4e 4A cetamidophenyl 2 ` (dipropylamino) acetate hydrochloride, (Pr2n1APAP HCl). 1H NMR (D2O) 8, 2H), 4.32 (s, 2H), 3.11 (t, 4H), 2.01 (s 3H), 1.82 (sextet, 4H), 0.824 (t, 6H). Yield: 50% 4f 4 A cetamidophenyl 2 ` (piperidin1 ``yl) acetate hydrochloride (PIPn1APAP HCl). Synt hesized from 2.5 g (0.014 mole) 3 f, 2.87g (0.014 mole) of DCC and 2.1 g (0. 014 mole) APAP. 1H NMR (D2O) 4.36 (s, 2H), 3.37 (m, 4H), 2.171 (s, 3H), 1.9061.68 (m, 6H). Yield: 71% 4g 4 A cetamidophenyl 3 ` (piperidin1 ``yl)propanoate hydrochloride (PIPn2APAP HCl) Synthesized from 1.5 g (0.08 mole) 3g, 1.7 g D CC and 1.25 g (0.08 mole) APAP. 1H NMR (D2O) 3.57 (m, 4H), 3.19 (t, 2H), 3.01 (t, 2H), 2.16 (3H, s). Yield: 65% 4h 4 A cetamidophenyl 4 ` (piperidin1 ``yl)butanoate hydrochloride (PIPn3APAP HCl) Synth e sized from 1.5 g (0.008 mole) 3h, 1.6 g (0.008 mole) DCC and 1.2 g (0.008 mole) APAP. 1H NMR (D2O)

PAGE 57

57 (d, 2H), 3.16 (t, 2H), 3.19 (t, 2H), 2.92 (t, 2H), 2.76 (t, 2H), 2.14 (s, 3H), 1.402.15 (m, 6H). Yield : 78% 4i 4 A cetamidophenyl 2 ` ( morpholin4``yl) acetate hydrochloride, (MORn1APAP HCl). Synthesized from 1.5 g (0.008 mole) 3i, 1.6 g (0.008 mole) DCC and 1.6 g (0.008 mole) APAP. 1H NMR (D2O) (broad s, 2H), 4.031 (broad s, 4H), 3.50 (broad s, 4H), 2.16 (3H, s).Yield: 70% 4j 4 A cetamidophenyl 3 ` ( morpholin4``yl) propanoate hydrochloride (MORn2APAP HCl) Synthesized from 1.5 g (0.007 mole) 3j, 1.5 g (0.007 mole) DCC and 1.09 g (0.007 mole) APAP. 1H NMR (D2O) 2H), 4.11 (broad s, 2H), 3.83 (broad s, 2H), 3.60 (broad t, 4H), 3.21 (broad s, t, 4H), 2.141 (s, 3H).Yield: 50%. 4k 4A cetamidophenyl 4 ` ( morpholin4``yl) butanoate hydrochloride (MORn3APAP HCl) Synthesized f rom 1.5 g (0.007 mole) 3 k, 1.5 g (0.007 mole) DCC and 1.1 g (0.007 mole) APAP. 1H NMR (D2O) 2H), 4.09 (broad s, 2H), 3.8 (broad s, 2H), 3.54 (broad s, 2H), 3.21 (broad s, 2H), 3.28 (t, 2H), 2.788 ( t, 2H), 2.14 (s 3H), 2.17 2.09 (m, 2H). Yield: 61% General method for the synthesis of compounds 5a 5k. The hydrochloride salts of DAAC prodrugs of APAP (4a 4k) were treated with 5 10 ml ice cold saturated aqueous sodium bicarbonate solution and the aqueous layer was extracted immediately with 200 250 ml methylene chloride once. The treatment with aqueous saturated NaHCO3 and the subsequent extraction had to be very fast (<30 sec) otherwise, hydrolysis (20 60%) to the parent drug was observed. The organic layer was dried over sodium sulfate for 1 h and filtered. The solvent was evaporated on a rotary

PAGE 58

58 Table 31. Characterization of DAAC APAPHCl and DAACAPAP Compounds. Elemental Analysis (Calculated and Experimental Values of % of Carbon, Hydrogen and Nitrogen) and Melting Points for DAAC APAPHCl and DAACAPAP Prodrugs Mp C Cal C Exp H Cal H Exp N Cal N Exp 4a 200210 52.85 52.58 6.28 6.47 10.27 10.14 4b 195200 54.45 54.26 6.68 6.70 9.77 9.66 4c 200205 55.90 55.90 7.04 7.04 9.31 9.30 4d 170180 57.23 57.11 7.36 7.44 8.90 8.85 4e 175185 ------4f 234 58.50 59.06 7.09 7.15 8.57 8.59 4g 202 59.90 59.73 7.39 7.54 8.22 8.20 4h 260270 53.42 53.36 6.08 6.05 8.90 8.86 4i 220 54.79 54.17 6.44 6.47 8.52 8.34 4j 210 54.62a 54. 57 6.88a 6.57 7.96a 8.40 4k 234 58.50 a 59.06 a 7.09 a 7.15 a 8.57 a 8.59 a 5a 9092 59.49 b 59.86 b 7.22 b 6.91 b 11.63 b 10.47 b 5b 5660 63.62 64.18 7.63 7.85 10.60 9.92 5c 5557 ------5d 6062 ------5e Oil -----5f 122124 65.20 64.96 7.30 7.28 10.14 10.05 5g 155158 66.18 65.96 7.64 7.65 9.65 9.64 5h 114 115 5i 126128 60.42 60.66 6.52 6.58 10.07 9.95 5j 148158 61.63 61.64 6.90 6.80 9.58 9.46 5k 98100 62.73 62.82 7.24 7. 38 9.14 9.17 a values for hemihydrate; b values for 0.25 hydrate evaporator to yield compounds 4a4k as colorless solids. Extraction of the free base from the corresponding hydrochlorides was also attempted by t reatment with a weaker

PAGE 59

59 base, 1 equiv triet hyl amine in cold methylene chloride (5 10 min) and subsequent trituration with diethyl ether. But the desired prodrugs could not be obtained in very high yields or purity. Further, column purification was not possible for these compounds because of highl y basic nature of the prodrugs: decomposition on silica gel to parent drug was observed. 5a 4A cetamidophenyl 2 ` (di methylamino) acetate (Me2n1APAP) .1H NMR (CDCl3 3H). Y ield: 95%. 5b 4 A cetamidophenyl 3 ` (dimethylamino)propanoate (Me2n2APAP). NMR 2.17 (s, 6H), 2.03 (s, 3H).Y ield: 95%. 5c 4A cetamidophenyl 2 ` (diethylamino)acetate, (Et2n1APAP). 1H NMR (CDCl3) 6H). Yield: 70%. 5d 4 A cetamidophenyl 3 ` (diethylami no)propanoate (Et2n2APAP). 1H NMR (CDCl3) (quart, 4H), 2.14 (s, 3H), 1.09 (t, 6H).Yield: 78%. 5e 4A cetamidophenyl 2 ` (dipropylamino)acetate (Pr2n1APAP). 1H NMR (CDCl3) : 3H), 1.80 (broad s, 4H), 1.09 (t, 6H). Yield: 61%. 5f 4 A cetamidophenyl 2 ` (piperidin1 ` ` yl)acetate (PIPn1APAP). 1H NMR (CDCl3) 2H), 3.44 (s, 2H), 2.61 (t, 4H), 2.17 (s, 3H), 1.67 1.62 (m, 6H), 1.45 (broad d, 2H). Yield: 71%.

PAGE 60

60 5g 4 Acetamidophenyl 3` (piperidin1 ` ` yl)propanoate (PIPn2APAP). 1H NMR ), 2.37 (broad s, 4H), 2.035 (s, 3H), 1.49 (quin, 4H), 1.40 (broad d, 2H).Yield: 75% 5h 4 A cetamidophenyl 4 ` (piperidin1 ` ` yl)butanoate (PIPn3APAP). 1H NMR (CDCl3) (broad s, 2H), 1.79 (broad s, 4H).Yield: 85%. 5i 4 A cetamidophenyl 2 ` ( morpholin4``yl) acetate (MORn1APAP). 1H NMR (CDCl3) 2.68 (t, 4H), 2.17(s, 3H).Yield: 70%. 5j 4 A cetamidophenyl 3 ` (morpholin4``yl) propanoate (MORn2APAP). 1H 2H), 2.40(broad s, 4H), 2.02 (s, 3H).Yield: 80%. 5k 4A cetamidophenyl 4 ` ( morpholin4``yl) butanoate (MORn3APAP ). 1H Hz, 2H), 3.55 (t, 4H), 2.57 (t, 2H), 2.32 (t, 6H), 2.03 (s, 3H), 1.89 (quin, 2H).Yield: 61%. Hydrolysis of DAAC APAP Prodrugs The hydrolysis experiments were carried out on compounds 4a4e (DAAC A PAPHCl). For determining the half lives of the members of DAAC prodrugs a solution of the prodrug was prepared by dissolving a known amount of the prodrug in deionized water. An aliquot of the aqueous solution was diluted with buffer (pH 6.0, phosphate 2 0 mM) to give concentrations of about 1020 mM. The diluted solution was added into a UV cuvette and absorbance values were recorded over 78 half lives. The cuvette was maintained at 37 0.5 oC throughout the experiment. Although, the prodrugs showed an absorbance maximum very close to that of APAP the molar absorptivities of the

PAGE 61

61 prodrugs are higher than that of the parent drug, APAP, so the hydrolysis of the prodrugs was monitored by the decrease in the absorbance values over time. The absorbance values were recorded at 242 nm. All experiments were allowed to reach A absorbance would reflect the concentration of APAP only. The pseudo unimolecular rate constants were obtained from a plot of log (At A) ver sus time where At is the absorbance at time t. All the experiments were run in triplicate. The pKa values were estimated by using the pKa prediction software (ACD Labs Inc. Version 10.0). Table 32 Half L ives (t1/2, min) and predicted pKa Values of DAAC APAPHCl, prodrugs in Buffer (pH 6.0, phos phate, 20 mM, I = 0.5) at 37 0.5 0C pKa t1/2 4a Me 2 n 1 APAP HCl 6.94 76.92 0.24 4b, Me2n2APAP HCl 8.66 113.03 2.85 4c Et2n1APAP HCl 8.08 105.71 4.45 4d, Et2n2APAPHCl 9.59 79.32 11.92 4e, Pr2n1A PAPHCl 9.40 105.89 3.24 4f PIP n1APAP HCl 6.98 -4g, PIP n2APAP HCl 8.65 79.67 1.25 Table 32 shows the t1/2 values for the DAAC APAPHCl ester prodrugs. The presence of an amine group in the acyl side chain of a DAAC prodrug predisposes the ester group towards hydrolysis because of two reasons: (a) e lectron withdrawing effect of the NR1R2 group and (b) general base catalysis by the NR1R2 or -+N H R1R2 group. In a detailed study 57 by Kirby et. al. rate enhancements up to 105 fold for the hydrolysis of phenyl 3 dimethyl aminopropionat e vs phenyl acetate at pH > 9, where the NMe2 group is present as a free base, was observed. Thi s observation supports the

PAGE 62

62 idea of facilitat ion of rate enhancement by the NMe2 group in addition to simple electronic effects. Additionally, observation that the rate of hydrolysis of phenyl acetate is 10 times slower than that for phenyl 3 dimethyl aminopropionate at pH conditions where the dimethylamine group is protonated indicates a rate acceleration by the+NMe2 group. However the propensity of a protonated amine group to facilitate intramol ecular catalysis is much lower compared to a free amine group. Figure 33 shows pathways available to the neighboring am ine group for the facilitation of ester hydrolysis.68 These pathways include (A) nucleophilic catalysis, (B ) intramolecul ar general base catalysis and (C ) intramolecular general acid specific base catalysis. Mechanisms A, B and C are kinetically indistinguishable. For a prodrug containing a basic amine group (pKa ~ 710) the following are true:43 At pH << pKa the primary reaction is the hydrolysis of the protonate d ester (pathway C ). At pH > 5 < pKa the primary reactions are the hydrolysis of free base and the protonated ester (pathways A B and C ). At pH > pKa the primary reaction is the hydrolysis of the free base (pathways A and B ). For co mpounds ( except 4a) in the present study rate acceleration by +NR1R2 group is the primary effect observed because the compounds are expected to be protonated (~ 99%) at pH 6.0. The t1/2 values of 4c and 4d (same amine but different n) can be explained by a higher degree of gener al base catalysis by the more basic diethylaminoprop ionate group compared to diethylglycinate group. A similar trend between 4a and 4b is not observed because the calculated pKa value of 4a is much lower than that of 4c so a higher percentage of free base is present at pH 6.0. The trend in t1/2 values amongst compounds with same n (4a, 4c and 4e) but

PAGE 63

63 different steric bulk on the amine group can be explained using Newmans Rlue of Six which states that those atoms which are separated from the attacking atom in the transition state by a chain of four atoms are the most effective in providing steric hindrance.69 Figure 33. Possible routes of hydrolysis of a DAAC APAP prodrug For example for a DAAC acyl group, the blocking atom would be the atom (or group) in the 6th position. The atom is the 6th position is more likely to be in the path of the attacking atom than the 5th or 7th atom. (Figure 3 4) Figure 34 shows that the 6th atom in 4a could be 6 H whereas in 4c the 6th or the blockin g atom could be 4 H and 2 methyl groups. Overall the DAAC APAPHCl prodrugs hydrolyze to the parent drug via a combination of nucleophilic and intramolecular general base catalysis pathways. Although a pH rate profile can give more detailed information about the hydrolysis behavior of prodrugs containing a basic amine group, the data presented in this study exemplifies the transient behavior of the DAAC APAPHCl prodrugs. In terms of prodrug design the DAAC prodrug approach

PAGE 64

64 allows for modulation of half li ves of the prodrugs by controlling the alkyl chain length and steric bulk on the basic amine nitrogen. Figure 34. Origin of steric hindrance in 4a and 4c based on Newman Rule of Six. Determination of Solubilities of DAAC APAP Prodrugs As mentioned bef ore (Chapter 1), the permeation of a topically applied drug is a solubility driven process. Determination of the solubility of a permeant in aqueous (or aqueous like) and lipid solvents allows for an analysis of the solubility and flux data, that can be used to optimize topical delivery of drugs. The molar absorptivity of DAAC APAPHCl prodrugs was determined in triplicate a t 244 nm in methanol and at 241 nm in buffer (pH 4.0, acetate, 50 mM) A known amount of DAAC APAPHCl prodr ug was dissolved in deio nized water and the solution

PAGE 65

65 was diluted with methanol or buffer and analyzed by UV spectrophotometry. W ith the known concentration C, 4.0 or MeOH was calculated with Beers law: A244 = 244 l C, where l = cell length (1) The molar absorptivity of DAAC APAP prodrugs 5a 5k, (free bases) was determined in acetonitrile ( MeOH ) by the method similar to the one described above. The m olar absorptivities have been shown in T able 33. Table 33 Molar a bsorptivities of DAAC APAPHCl p rodrugs i n m ethanol ( MeOH) a cetate b uffer ( 4.0) and molar absorptivities of DAAC APAP prodrugs in acetonitrile ( ACN). a All experiments were run in triplicate. b Units of 104 ml mmole-1 c UV max = 244 nm d UV max = 241nm. e Value f rom Wasdo et. al. 2004 The solubilities of DAACAPAPHCl prodrugs ( 4a 4k ) were determined in PG (aqueous like) and 1octanol. The solubilities of DAAC APAP ( 5a 5k ) prodrugs were determined in isopropyl myr istate (IPM) and buffer (pH 4.0 acetate 50 mM) The p rocedure for solubi lity determination is described below. Compound MeOH a,b,c 4.0 a,b ,d Compound ACN a,b ,c 4a, Me2n1HCl --5a, Me2n1HCl 1.68 0.11 4b, Me2n2HCl 1.58 0.009 1.22 0.015 5b, Me2n2HCl -4c, Et2n1HCl 1.61 0.063 1.24 0.071 5c, Et2n1HCl 1.68 0.066 4d, Et2n2HCl -1.29 0.008 5d, Et2n2HCl 1.65 0.015 4f, PIPn1HC l -1.30 0.02 5f, PIPn1HCl 1.73 0.014 4g, PIPn2HCl 1.59 0.09 1.31 0.018 5g, PIPn2HCl 1.74 0.06 4h, PIPn3HCl 1.59 0.05 1.32 0.02 5h, PIPn3HCl 1.78 0.03 4i, MORn1HCl 1.58 0.05 1.29 0.02 5i, MORn1HCl 1.66 0.017 4j, MORn2HCl 1.58 0 .017 1.29 0.02 5j, MORn2HCl 1.71 0.045 4k, MORn3HCl 1.29 0.03 -5k, MORn3HCl 1.67 0.067 APAP 1.33 0.09 1.023 0.013 APAP 1.36e

PAGE 66

66 For each prodrug, the solubility in 1 o ctanol (OCT) was determined in triplicate. The prodrug was crushed into a fine powder and a saturated solution of the prodrug in octanol was obtained by adding an excess of each compound t o a test tube containing 1 ml octanol The test tube was then insulated and the suspension was allowed to stir at room temperature (23 1oC) overnight (12 14 h) on a magnetic stir plate. The suspension was filtered thr ough a 0.2 te was diluted with methanol and analyzed by UV spectrophotometry. The absorbance at 244 nm was used in each case to calculate the concentration of the prodrug in octanol which is the solubility in octanol (SOCT): CSaturation = SOCT = A244 244 (2) Solubilities in propylene glycol ( PG ) pH 4.0 acetate buffer and IPM were also determined in triplicate by the procedure used to determine SOCT with the following changes The prodrugs were only stirred for 1h in pH 4.0 acetate buffer before filtration. For determining SPG a sample of the filtrate was diluted with buffer (pH 4.0 acetate, 50 mM) and analyzed by UV spectrophotometry. For SIPM the suspensions were stirred for 2 4 h before filtration. For deter mining SAQ and SIPM the sample of the filtrate was diluted with acetonitrile and analyzed by UV spectrophotometry. A 1H NMR of the filtered solid in the solubility experiments were recorded to ensure that the prodrugs were intact through the course of experiment. None of the prodrugs hydrolysed to APAP during the course of the experiment. Partition coefficients were also determined in triplicate for the prodrug s by using the saturated IPM solutions obtained from the solubility determinations. The saturated IPM solution was partitioned against pH 4.0 buffer using the following volume ratios

PAGE 67

67 (V4.0 / VIPM) for compounds 4a, 4b, 4f, 4g, 4j and 4k: 0.1, 0.2, 0.1, 0.05, 0.1 and 0.1 respectively. The two phases were vigorously shaken for 10 seconds, then allowed to separate via centrifugation. An aliquot of the IPM layer was removed, diluted with acetonitrile, and analyzed by UV spectrophotometry as described above. Using the previously measured absorbance at 244 nm for the saturated solution, the partition coeffici ent was calculated as follows: KIPM:4.0 = [Aa/(Ab Aa)]V4.0/VIPM (3) where Ab and Aa are the respective absorbances before and after partitioning, and V4.0 and VIPM are the respective volumes of buffer and IPM in each phase. Ta bl e 34 Physicochem ical P roperties of DAAC APAPHCl Prodrugs. Molecular weights (MW), Solubility in 1 Octanol (SOCT), Solubility in Propylene Glycol (SPG) and Estimated IPM solubilities ( SIPM,est ) of DAAC APAPHCl Prodrugs Compound MW SOCTa SPGa SIPM,est 4a, Me2n1bHCl 272 0.352 0.01 122.05 2.90 0.004 4b, Me2n2HCl 286 0.291 0.035 150.09 2.03 0.003 4c, Et2n1HCl 300 1.297 0.067 93.80 0.98 0.027 4d, Et2n2HCl 314 1.162 0.055 176.10 3.42 0.018 4f, PIPn1HCl 313 ---4g, PIPn2HCl 327 0.202 0.02 18.17 1.05 0.0 04 4h, PIPn3HCl 341 20.87 0.47 64.09 1.17 1.762 4i MORn1HCl 315 0.09 0.01 14.01 0.01 0.001 4j MORn2HCl 325 0.10 0.007 29.97 0.58 0.001 4k MORn3HCl 339 1.522 0.077 39.68 0.50 0.05 APAP 151 158.94 c 662.25 c 1.90d a Units of mM. b Num ber of methylene groups between carbonyl carbon and amine. c Values from Kasting, Smith and Cooper (1987, Ref. 34). d Value from Wasdo et. al. (2004, Ref. 53) was used as the value for SIPM for APAP (2.38 mM) obtained from Kasting Simth Cooper (Ref. 34) wa s slightly different.

PAGE 68

68 The IPM solubility of the hydrochloride prodrugs was estimated from the data of Kasting Smith and Cooper34 by the method described below. A linear equation between log KPG:IPM and log KPG:OCT was developed from the solubility database for n = 28 compounds. The value for SIPM,est was calculated using the corresponding SPG and the linear equation derived above. All the DAAC APAPHCl prodrugs exhibited higher melting points (Table 31) than APAP which is expected because of introduction of the ionized amine g roup into the prodrug. Compounds 4k and 4h exhibited decomposition (charring) close to the melting temperature. DAAC APAPHCl prodr ugs exhibited PG solubilities between 14 176 mM in propylene glycol, the most PG soluble member being 4d. The trends in PG solubilities for DAAC APAPHCl prodrugs were similar to the trend in aqueous solubilities (S4.0, Table 35) for free bases The member exhibiting the highest S4.0 also exhibited the highest SPG (4d). Among the DAAC APAPHCl series, members containing the c yclic amines (morpholine and piperidine, 4f 4k) in the promoiety exhibited a comparatively lower SPG. The most PG soluble member among 4f 4k was 4h (~6 4 mM) compared to 4d (176 mM). The lower aqueous and PG solubility amongst morpholinyl and piperidiny l prodrugs might be attributed to the rigid structure of the six membered ring. Among the prodrugs containing the same amine the solubility in PG increased with an increase in basicity of the nitrogen (Table 35) For example, 4a has a SPG of 122 mM which increases to 150 mM for 4b. S imilar trend is observed for 4f 4h and 4i 4k. Similarly, a mongst prodrugs containing same amine,S4.0 increased with increasing distance f rom the acyl group, except for 5f 5h where the maximum aqueous solubility is obs erve d for the n = 2 prodrug, 5g (120 mM) and the effect of further addition of a

PAGE 69

69 methylene group decreased the S4.0 for 4 g to 109 mM. The trends in aqueous solubility of the prodrugs is discussed in detail further in the section. None of the prodrugs exhibited higher SOCT than APAP. Low lipid solubility is expected fr om a hydrochloride salt. 4 h exhibited the highest SOCT among the DAAC APAPHCl series. Similarly, the IPM solubilities (estimated) of the prodrugs was also generally low. Table 35 Physicochemical p roperties of DAAC APAP p odrugs Estimated pKa values, solubility in pH 4.0 a cetate b uffer (S4.0), s olubility in i sopropyl myristate (SIPM), estimated intrinsinc solubilities (S5.5, est) and partition coefficients between IPM and pH 4.0 a cetate b uffer for DAAC APAP (free base) p rodrugs. a Units of mM. b Number of methylene groups between carbonyl carbon and amine. c Molar Absorptivity of 5a was used to calculate SIPM. d Estimated from log S4.0 = log SIPM log K. As shown in T able 35 a ll the members ( except 5j ) of the free base DAAC prodrugs of APAP exhibited higher IPM solubility than APAP 5 c, the most lipid soluble member of the se ries was about 45 times more soluble in IPM than APAP. Although in Compound pKa SIPM a S4.0 a S5.5, est a KIPM:4.0 5a, Me2nb1 6.94 15.23 1.43 264.50 1.40 9.26 0.055 0.007 5b, Me2n2 8.66 42.26 1.42c 897.42d -0.047 0.003 5c Et2n1 8.08 91.22 0.66 116.49 0.86 0.30 -5d, E t2n2 9.59 58.72 1.27 2884.03d -0.021 0.003 5f PIPn1 6.98 7.41 0.19 46.79 1.65 1.50 0.127 0.047 5 g PIPn2 8.65 3.75 0.07 120.89 7.47 0.85 0.13 0.005 5 h PIPn3 9.32 13.06 0.11 109.46 2.09 0.016 0.033 0.004 5 i MORn1 4.65 2.52 0.05 50.51 0.55 44.25 0.048 0 .037 5 j MORn2 6.42 1.25 0.02 75.93 1.52 8.14 0.022 0.08 5 k MORn3 7.14 12.26 0.49 380.55 1.73 8.52 0.034 0.002 APAP -1.90 100 -1.721

PAGE 70

70 the DAAC prodrug series of APAP, the number of compounds having the same amine group in the promoiety is small, (maximum three) somewhat of a trend in SIPM values could be observed. Among members of the series having the same amine group, with an exception of 5 a and 5 b, an increase in distance from the acyl group caused an initial decrease in the SIPM with a subsequent increase in SIPM values. For example, compound 5i showed only a moderate increase in SIPM and 5j was even less IPM soluble than APAP. However, 5k showed about 6 times more SIPM than APAP. The same trend in SIPM was observed among compounds 5f, 5g and 5h and between 5 c and 5 d. The SIPM of piperidinyl compounds ( 5f 5h) was higher compare d to the morpholinyl compounds (5i 5k ). The higher SIPM of the DAAC prodrugs compared to APAP is expected because of the masked polar OH group. The trend in IPM solubilites can be explained on the basis of melting points of the prodrugs. Figure 35 shows a plot of SIPM vs melting points of DAAC APAP prodrugs. The lower melting compounds showed a higher SIPM. Among the members with the same amine group the initial decrease in SIPM of the member having two carbon atoms (n = 2) between the acyl and the amine group ( 5 d, 5 g and 5 j) compared with the n = 1 member ( 5c 5 f and 5 i) can be attributed to a concurrent increase in the pKa (hence basicity) of the nitrogen. Further increase in the alkyl chain offsets the inc rease in the pKa and the SIPM values begin to rise again. For example, in going from 5f (n = 1) to 5h (n = 3) the increase in the alkyl chain length results in almost 34 times greater SIPM than the previous member. The trend in the pKa values can be explained in terms of inductive effect of the carbonyl group. The increase in the distance between the carbonyl carbon and the amine group makes the amine

PAGE 71

71 more basic. For example, t he pKa of ethyl N, N dimethylaminobutyrate was very close to dimethyl amine indi cating a weak inductive ef fect over three carbon atoms.57 Figure 35. Plot of IPM solubilities vs melting points for DAAC APAP prodrugs S4.0 of the DAAC series of prodrugs also show ed a dependence on the pKa on the amine in the prodrug. All members except 5 f 5 i and 5 j were more water soluble than APAP. Amongst the morpholino compounds, the most water soluble m ember was also the most basic. Also a constant increase in SAQ was observed with an increasing distance of the amine group from the carbonyl group (hence an increasing pKa). However, the solubility among the piperidinyl compounds initially increased from 4 6.79 mM (5f ) to 120.89 mM ( 5g ) then decreased to 109.46 mM ( 5 h ). Even though the pKa of t he piperidinyl compounds was comparable to the 5 a 5 d, the SAQ was comparatively lower. This may be explained by the trends in melting points. The more water soluble members, 5 a 5 d, showed lower melting points compared to morpholinyl and piperidinyl DAAC APAP prodrugs. The pH dependent solubility of a basic amine can be estimated using the method used by the Bergstrom et. al. 70 w here Stot represents the solubility of the total species

PAGE 72

72 present in ionized state and Sins represents the intrinsic solubility of the amine at a particular pH. The intrinsic solubilities of the prodrugs have been shown in Table 35 For calculating Sins pH of the skin was considered to be 5.5.5 The value of the intrinsic solubility of the least basic prodrug, 5i, was closest to its experimental solubility. Also the higher the difference between the pKa of the permeant and the pH of the skin, the higher is the difference between its intrinsic solubility and experimental solubility. Overall the solubilites of DAACAPAP prodrugs in aqueous and lipid solvents are governed by the basicity of and melting points of the corresponding prodrugs. In Vitro Flux Determination of DAAC APAP Prodrugs Three different mice were used to determine the flux of each prodrug. Prior to skin removal, the mice were rendered unconscious by CO2 then sacrificed via cervical dislocation. Skins were removed by blunt dissection and placed dermal side down in contact with pH 7.1 phosphate buffer (0.05 M, I = 0.11 M, 32 oC) containing 0.11% formaldehyde which has b een shown to inhibit microbial growth and maintain the integrity of the skins throughout the experiment.21 Prior to the application of the prodrug s as suspensions in IPM the skins were maintained in contact with the buffer f or 48 h to leach out all UV absorbing material. During this conditioning phase time, the receptor phase was removed a nd replaced with buffer 3 times The suspension of the prodrug in IPM was prepared 4 h before application and allowed to stir at room temperature (25 1 oC) until application. The concentration of the donor phase suspension was approximately 10 x SIPM. After the 48 hour leaching period, an aliquot (0.5 ml) of the prodrug suspension was added to the surface of the skin (donor phase). Samples of the receptor phase were usually taken at 8, 19, 22, 25, 28, 31, 34, and 48 h and quickly analyzed by UV spectrophotometry. Since only parent drug was obtained in

PAGE 73

73 the receptor phase in all cases the amounts of permeated APAP was quantified using molar absorptivity of APAP in the receptor phase. At each sampling time, the entire receptor phase was replaced with fresh buffer in order to maintain sink conditions. After the 48 h of the first application period, the donor suspension was removed and the skins were washed three times with methanol (35 ml) to remov e any residual prodrug from the surface of the skin. The remaining prodrug or APAP in the skin was leached out by keeping the skins in contact wit h buffer for an additional 24 h. Then the receptor phase was replaced with fresh buffer and an aliquot (0.5 ml ) o f a standard drug theophylline was applied in PG to the skin surface The second application fluxe s were determined by sampling of the receptor phase at 2 4 6 h and analysis using UV spectrophotometry. The concentration of theophylline in the receptor phase was 1). At each sampling time, the entire receptor phase was removed and replaced with fresh buffer. In each experiment, the flux was determined by plotting the cumulative amount of APAP versus time J in units of mol cm2 h1 could then be calculated by dividing the slope of the steady state portion of the graph ( 19 34 h, Figure 35) by the surface area of the skin (4.9 cm2). In Vitro Evaluation of Morpholinyl and Piperidinyl DAA C APAP Prodrugs The steady state flux values for DAAC APAP prodrugs from a saturated IPM vehicle through hairless mouse skin (EXP JM) are shown in Table 36. The steady state flux values for a standard drug ( T heophylline, Th) from a saturated PG donor phase (JS) and the residual concentration of APAP in the skin has also been shown in the table. The concentration of the APAP delivered locally to the skin was calculated from the absorbance value of APAP in the donor phase after the 24 h leaching period.

PAGE 74

74 Figure 35. Plot of cumulative amount vs time for calculating steady state flux All compounds showed higher flux than APAP. Among the morphol inyl DAACAPAP prodrugs, the best member was MORn1APAP, 5i which gave about two times higher flux than APAP. Although MORn2 and MORn3 prodrugs (5j and 5k) showed higher water solubility than 5i the flux of 5j was 0.80 mole cm2 h1 (1.6 times that of APAP) and 5k was 0.71 mole cm2 h1 (1.4 times that of APAP). Based on solubility properties t his was un expected because 5i showed only a moderate enhancement in IPM solubility (1.3 times) tha n APAP and was less water soluble than 5j or 5k The flux values for MORn2APAP and MOR n3APAP were lower than expected. A similar trend was observed in the piperidinyl DAAC APAP prodrugs. The PIPn1 prodrug showed the highest flux, 0. 78 mole cm2 h1, (1.27 time that of APAP) amongst the PIP n1PIP n3 compounds. Further analysis of the trend in the flux values revealed that among DAAC prodrugs containing the s ame amine there is linear correlation between pKa and J values. The plot of pKa vs J for morpholinyl and piperidinyl DAAC APAP prodrugs has been shown in Figure 36.

PAGE 75

75 Figure 36. Plot of pKa vs J for morpholinyl and piperidinyl DAAC APAP prodrugs. Among the compounds studie d there is a negative correlation between J values and pKa values. Additionally comparison of MORn1 MORn3 with PIPn1 PIPn3 shows that th e relatively more basic piperidi nyl prodrugs showed lower flux than t he corresponding morpholinyl compounds. The unexpected lower flux values for more basic n = 3 morpholinyl and piperidinyl DAAC APAP prodrugs can be explained by a possible increase in the effective molecular weight caused by water association with the ami ne group as the prodrug diffuses through the membrane.7173 Affsprung and coworkers studied the hydration behavior of amines in organic solvents like benzene and chloroform and proposed that at higher concentrations the amine molecules are present as bridged hydrates involving two base molecules and one water molecule.71 In our study, such an association will offset the effect of enhanced water and lipid solubility because of a negative correlation

PAGE 76

76 between flux and MW. Additionally, studies on permeabi lity of model lipid bilayer membranes to organic amine solutes suggest that interfacial charge and hydration structure in addition to solubility dependence may influence flux of ionizable solutes.74 Table 36.Results from diffusion cell experiments with morpholinyl and piperidi nyl DAAC APAP prodrugs. Maximum flux of the prodrug from a saturated isopropyl myristate (IPM) vehicle through hairless mouse skin (JM), maximum flux of standard drug, t heophy lline, from a saturated propylene glycol (PG) vehicle (JS), log of experimental flux values (EXP log JM), residual concentration of the drug in the skin (CS), absolute difference between EXP log JM and flux values calculated from Robert Sloan equation Comp ound JM a JS a CS b d EXP log J M CALC c log J M M 5f PIPn1 0.78 0.11 0.66 0.35 1.20 0.45 0.10 0.021 0.121 5g PIPn2 0.65 0.09 0.30 0.003 2.88 0.16 0.18 0.045 0.225 5h PIPn3 0.64 0.05 0.1 0.03 1.52 0.29 0.19 0.262 0.452 5i MORn1 1.05 0.17 1.35 0.65 -0.021 0.203 0.224 5j MORn2 0.80 0.08 0.86 0.08 -0.10 0.300 0.200 5f MORn3 0.71 0.06 0.38 0.09 0.52 0.10 0.15 0.513 0.663 APAP 0.51 0.74 2.74 0.07 0.19 0.262 0.472 a Units of mole cm-2 h-1. b Concentration of APAP in the skin after a 24 h leaching period. c Calculated from the RS equation log CALC JM = 0.599 + 0.502 log SIPM + 0.498 log S4.0 0.00235 MW. d Units of mole. Similar results were obtained with PEG prodrugs of APAP studied in our lab previously. The most lipid and water soluble member of th at series (SAQ = 184 mM, SIPM = 12.14 mM) gave lower flux than expected from a highly water and lipid soluble permeant (0.69 mole cm2 h1).75 The ethylene oxide moiety has been shown to be associated with water molecules in solution. This association of water molecules with the ethylene oxide head group causes a n increase in effective molecular w eight of the permeant as it diffuses through the skin.

PAGE 77

77 The residual amount of APAP in the skin after the application period (CS) has been shown in T able 36. The PIPn2 prodrug could only deliv er almost the same amount of APAP as the parent drug itself while the skin retention of all the other prodrugs was less than that of APAP. Additionally, the JS ( control21) values for 5g, 5h and 5f were lower than usual after a permeation experiment with IPM (1.0 mole cm2 h1). This may be attributed to an excess of HCHO present in the batch of receptor phase buffer used for these three compounds. No microbiology experiments were carried out to confirm this. However the co ntrol values returned to ~ 1.0 mole cm2 h1 when a new batch of buffer containing 0.11 % formaldehyde was used for permeation experiments with other members of the series (shown in Table 37). The Robert Sloan (RS) Equation was developed as an approach for quantifying the dependence of flux of a permeant through skin on its aqueous and lipid solubilites and molecular weight. The first form of RS equation was published in 1999 and the database consisted of 5FU and 6MP prodrugs including the parent drugs (n = 42). Since then the Robert Sloan database has been updated three times and various acyl and soft alkyl prodrugs of APAP were added to the database. The RS equation developed by Thomas and Sloan (2009)60 is shown below CALC log JM = 0.562 + 0.50 1 log SIPM + 0.499 log SAQ 0.00248 MW ( 4 ) However equation 4 did not include 2 bis 1,3 alkyl carbonyl 5 FU prodrugs. Therefore a further updated database was developed comprising n = 73 prodrugs : 42 compounds from Roberts et. al. (1999) ,36 C5 bis 6,9 ACOM 6MP, 4 3 AC 5FU prodrugs from Beal and Sloan ,63 6 3 ACOM 5 FU prodrugs from Roberts and Sloan ,76 8 AOC -

PAGE 78

78 APAP prodrugs Wasdo and Sloan ,53 5 ACOM APAP p rodrugs from Thomas and Sloan 59 and 5 AOCOM APAP pro drugs from Thomas et. al.60 SPSS 10.8 was used to calculate the coefficients in the RS equation. The updated Robert Sloan database gave x = 0.599, y = 0.502, z = 0.00235 and an r2 = 0.92 (Equation 5). The y value in RS equation shows that the flux is essentially equally dependent on aqueous and lipid solubilities the permeant. CALC log JM = 0.599 + 0.502 log SIPM + 0.498 log SAQ 0.00235 MW (5) Equation 5 was then used to calculate the flux of the morpholinyl and piperidinyl DAAC APAP prodrugs (CALC log JM). The absolute differe nce between the EXP log JM and CALC log JM have been shown in the T able 36. The average M was 0.16 3. The plot of EXP log JM vs CALC log JM has been shown in F igure 37. The CALC log JM values for all but one of the prodrugs were higher than EXP log J values. The one member of the DAAC APAP serie s that performed better than that predicted was MORn1 prodrug whereas the n = 3 prodrugs in both MOR and PIP series showed much lower flux than expected and also showed the highest log JM values. This ov er predi ction of J values was observed due to two possible reasons. The CALC log JM values were obtained using S4.0 which are expected to be higher than the solubility of the permeant in the skin microenvironment. When CALC log JM was obtained using int rinsic solubilities at pH 5.5 from Table 35, lower calculated flux values than experimental values were obtained (data not shown). In terms of topical delivery of DAAC prodrugs, the pH of the skin microenviroment will govern the percentage of ionized or unionized species in the first few layers of the skin.

PAGE 79

79 Figure 37. Plot of EXP log JM vs CALC log JM. Since the DAAC APAP prodrugs are expected to be present in an equilibrium between their ionized and unionized state, the so lubility of the actual species in the skin while the prodrug permeates the skin can be challenging to measure or predict. The second reason for lower flux values than expected is the higher effective molecular weight of the more basic prodrugs as they dif fuse through the membrane. In conclusion among the morpholinyl and piperidinyl DAAC APAP prodrugs evaluated, MORn1 APAP prodrug was able to enhance delivery of APAP by two times. Although incorporation of an ionizable amine was able to enhance solubility properties of the prodrugs, the increased basicity, indirectly offset the flux enhancement due an increase in effective molecular weight of the permeant due to hydrated form s in the skin microenvironment.

PAGE 80

80 In Vitro Evaluation of Dimethyl and Diethyl DAAC AP AP Prodrugs and DAAC APAP HCl Prodrugs The results from the flux measurements of the dimethyl and diethyl DAAC APAP prodrugs and three DAAC APAPH Cl prodrugs have been shown in T able 37. The RS equation was used to calculate the flux of two DAAC APAP p rodrugs. Figure 38. Cumulative amount of drug permeated vs time for MORn1APAP, MORn1APAPHCl and Me2n1APAP and Me2n1APAP HCl Both DAAC APAP members that were evaluated performed better than the best member in the morpholinyl and piperidinyl series 5a and 5c both showed about three

PAGE 81

81 times higher flux than APAP. The best member of the series was the Et2n1APAP, which also has the best balance in the lipid and water solubilities. Although 5c, the mo re lipid and water soluble member of the series showed the highest flux of the series which was expected, it did not show as high a flux as expected. 5c exhibited a SIPM of 91 mM and a S4.0 of 116 mM where as 5a exhibited a SIPM of 15 mM and S4.0 of 264 mM. 5c is the compound with the best balance between lipid and aqueous solubility. But 5c has a pKa of 8.08 and 5a is less basic by about one pKa unit (6.98). As discussed in the previous section the increase in basicity seems to offset the effect of increase in solubility consequently leading to lower flux than expected. Table 37.Results from diffusion cell experiments with dimethyl and diethyl DAAC APAP prodrugs and DAAC APAPHCl prodrugs Maximum flux of the prodrug from a saturated isopropyl myristate (IPM) vehicle through hairless mouse skin (JM), max imum flux of standard drug, t heophylline, from a saturated propylene glycol (PG) vehicle (JS), log of experimental flux values (EXP log JS), residual concentration of the drug in the skin (CS), absolute difference between EXP log JS and flux values calculated from Robert Sloan equation Compound JM a JS a CS a,c EXP log JM CALC d log JM M 5a Me2n1 1.50 0.30 0.43 0.11 5.84 1.05 0.176 0.65 0. 47 5c Et2n1 1.52 0.13 0.73 0.07 1.66 0.76 0.181 0.79 0. 61 4a Me2n1HClb 0.64 0.06 1.00 0.02 0.69 0.07 0.19 --4 b Me2n2HClb 0.65 0.01 0.95 0.03 0.69 0.05 0.19 --4 i MORn1HClb 0.54 0.02 0.96 0.25 0.71 0.18 0.26 --APAP 0.51 0.74 2.74 0.70 0.29 -a Units of mole cm-2 h-1. b Suspension of the prodrug was applied in a (99:1) PG:IPM vehicle. c Concentration of APAP in the skin after a 24 h leachi ng period. d Calculated from the RS equation log CALC JM = 0.599 + 0.502 log SIPM + 0.498 log S4.0 0.00235 MW. Three DAAC APAPHCl prodrugs were also evaluated in diffusion cell experiments. The Me2n1APAPHCl prodrug gave 1.25 times higher flux than APAP whereas MORn1APAP HCl prodrug showed about the same flux as APAP. The DAAC -

PAGE 82

82 APAP (free bases) prodrugs 5 a and 5 i, gave 2.3 and 2 times higher flux than the corresponding hydrochloride salts The plots for cumulative amount of drug permeated vs time for free base and corresponding hydrochlorides are shown in F igure 38 Another difference between the permeation behavior of free base and hydrochloride DAAC prodrugs is the fact that the salts showed a longer lag time to achieve steady state flux ( F igure 3 8 ). The skin retention of Me2n1 prodrug was twice as high as APAP. The skin retention of all the other members o f the series was lower than APAP. Figure 39. Plot of EXP log JM vs CALC log JM including dimethyl and diethyl D AAC APAP prodrugs The second application flux values for all the compounds was close to the control value of 1.0 mole cm2 h1 which showed that the flux enhancement observed is not because of the damage to the skin caused by IPM itself. The RS equation w as used to

PAGE 83

83 calculate the flux of 5a and 5c. As described before, the calculated flux of these compounds was higher because the S4.0 might not precisely represent the actual solubility of the permeant in the skin. The plot of EXP log JM vs CALC log JM has been shown in F igure 39. Conclusions The potential of an ionizable amine in modulating solubility and permeation properties of an acyl prodrug has not been studied or reported widely. In this study a series of N N` dialkylaminoalkylcarbonyl (DAAC) acyl prodrugs of APAP as a model phenolic drug were synthesized. The choice of amine in the promoiety was guided by two factors: pKa of the amine and steric bulk on the amine. The prodrugs were synthesized via a three step procedure in good yields. The half lives of the members evaluated show a dependence on the pKa of the amine in the molecule and undergo hydrolysis via a combination of intramolecular general base catalysis and nucleophilic catalysis by the basic amine group. The t1/2 values obtained in this s tudy exemplify the transient nature of the DAAC prodrugs. The octanol solubilities of the hydrochloride salts was generally lower than APAP itself which is expected from an ionized compound. The IPM solubilites of all the free base prodrugs except one was higher than APAP. With a few exceptions almost all the prodrugs exhibited higher S4.0 t han APAP. A mong the compounds having the same amine in the promoiety SAQ is dependent on the pKa of the amine in the prodrugs. The effect of moving the amine group away from the carbonyl carbon affects the pKa values of the prodrugs which consequently affects the solubility behavior of the prodrug. The solubility of the prodrugs in a propylene glycol showed the same trend as water solubility. The most water soluble member of the DAAC APAPHCl series exhibited the

PAGE 84

84 highest SPG as well. The IPM solubilities of the DAAC APAP prodrugs seemed to be correlated to their melting points where the lower melting members of the series exhibited a higher SIPM. Although one of the prodr ugs showed upto three times higher flux than APAP, the delivery of APAP was not as high as expected. The best member of the series was also the one that has the best balance in lipid and water solubility. We hypothesize that the lower flux values may be due to increase in the molecular weight caused by water association with the amine group as the prodrug diffuses the membrane.

PAGE 85

85 CHAPTER 4 AMINOALKYL CARBONYL (AAC) PRODRUGS OF PHENOLIC DRUG S ACETAMINOPHEN AND NALTREXONE Introduction A p rodrug strategy involving modification of the parent drug which contains hydroxyl groups (alcoholic or phenolic) using amino acids as a promoiety has been utilized successfully to improve potency of orally delivered drugs. Although incorporation of a basic amine group in th e prodrug facilitates its use as a water soluble salt, achieving optimum stability of the prodrugs is challenging. One of the amino acids that can imp r o ve the hydrophilic/lipophilic balance as well as the stability of the prodrugs is L v aline. The NH2 gro up improves water solubility and the bulky isopropyl group in the side chain can help improve stability at physiological pH. For example, the commercially available drug Valacyclovir Hydrochloride (Zovirax TM,GlaxoSmithKline), a valinate ester prodrug of acyclovir (Figure 4 1) has a t1/2 = 13 h (pH 7.4, 37oC), SAQ = 174 mg ml1 and is rapidly converted to the parent drug in vivo.46 Figure 41 Chemical structures of commercially available drug Valacyclovir and valine ester prodrug of 5OH DPAT by Bouwstra and coworkers (Ref. 77) There are only a few studies published about the utility of the unsubstituted amine grou p in enhancing the topical delivery of a parent phenolic drug. Recently Bo u wstra

PAGE 86

86 and coworkers synthesized AAC prodrugs of the dopamine agonist 5 hydroxy 2 ( N N ` propyl ) tetralin, 5 OH DPAT (Figure 4 1) which were targeted for transdermal iontophoretic delivery 77 The aqueous s olubilities of the VAL DPAT LA DPAT were 4 times and 14 times higher than the parent drug respectively. A higher invitro iontophoretic DPAT compared with 5OH DPAT The Lproline, Lserine, Ltyrosine, Lasparagine, and Lcitrulline ester prodrugs of Vitamin E were synthesized and evaluated in skin retention experiments. Although skin retention level of the prodrugs was found to be higher than Vitamin E, the flux of the prodrugs was not significantly higher than Vitamin E 58 Needham and coworkers78 investigated the transdermal delivery of the calcium channel blocker Nicardipine Hydrochloride form different pure and blended solvent systems. Propylene glycol (PG) was used as the primary vehicle for the development of the transdermal product. Although several research groups have investigated the potential of permeation enhancers20, 79 and ion pairing techniques to improve the flux of ionized compounds, a direct invitro transdermal evaluation of ionized prodrugs of phenolic drugs from a saturated donor phase in a single solvent has not been reported so far. Such an analysis would allow for optimization of the design directives that can be used to achieve maximum improvement in flux. One attractive target to which the AAC prodrug approach can be applied is naltrexone (NTX). NTX is an opioid antagonist currently available as Vivitrol, Revia and Depade for the treatment of alcohol and narcotic dependence.80 82 Revia and Depade are administered as oral tablets whereas Vivitrol is a sustained release formulation administered as a gluteal intramuscular i njection. Orally delivered NTX has

PAGE 87

87 been associated with gastrointestinal side effects in addition to plasma level fluctuations. 83 Additionally the sustained release formul ation requires frequent painful intramuscular injections. One approach to overcome these challenges is to deliver the drug transdermally. Treatment of alcohol and narcotic dependence require steady delivery of drug to the body over extended period of time Treatment of narcotic dependence has met with a high failure rate due to compliance issues. Noncompliance is usually followed by a relapse which is attributed to the elevated effects of heroin following two or three days without the antagonist. The resulting euphoria is often an incentive for the patient to indulge in further heroin use. 84 The non invasive nature of transdermal drug delivery approach can facilitate sustained delivery and strict adherence to dos ing regimens. Transdermal delivery of drugs also circumvents the side effects related to an orally delivered drug. There is only one published report on the synthesis of amino acid esters of APAP.67 N o solubility or flux data for these AAC prodrugs of APAP has been published. Also there are no reports of the synthesis of AA C NTX prodrugs In order to evaluate the ability of an unsubs tituted ionizable amine group to enhance the solubility properties and skin permeation behavior of the prodrugs we synthesized three AAC prodrugs of APAP and one AAC prodrug of NTX, characterized their solubility and stability properties and evaluated them in diffusion cell experiments using hairless mouse skin. Additionally, it is of interest to investigate the ability of the NH2group to modulate solubilities and flux of AAC prodrugs in comparison to the R1R2N group in the DAAC prodrug series.

PAGE 88

88 Figure 42 Chemical Structures of AAC Prodrugs of APAP and NTX Material and Methods N Boc L Amino Acids, 1 ethyl 3 (3 ` dimethylamin opropyl) carbodiimide hydrochloride (EDCI) were purchased from Acros Organics. Dicyclohexycarbodiimide (DCC) was purchased from Sigma Aldrich. Naltrexone Hydrochloride (NTX HCl) was purchased from MP Biomedicals LLC Naltrexone free base (NTX) was obtained by treatment of NTX HCl with aqueous NaHCO3 and subsequent extraction with methylene chloride. Methylene chloride and pyridine were dried over 3 molecular sieves before use. Melting points were determined on a Meltemp melting point apparatus. Spectra (1H NMR) were recorded on a Varian Unity 400 MHz spectrometer. UV spectra were recorded on a Shimadzu 2551 instrument Synthesis of AAC Prodrugs of Acetaminophen (APAP) And Naltrexone (NTX) The AAC prodrugs of APAP and NTX were synt hesized in two steps (Fig ure 43 ). The N Boc amino acid was coupled with parent phenolic drug via a DCC or an EDCI mediated reaction to give the Boc AAC APAP or Boc AAC NTX. The Boc protected AAC APAP prodrugs were treated with 2N HCl in diethyl ether to yield the AAC APAPHCl pro drugs in good yields. The Boc protected AAC NTX was treated with TFA to give

PAGE 89

89 VAL NTX TFA. The synthetic details are discussed in the experimental section. The results from the elemental analysis of the synt hesized compounds are shown in T able 4.1. Fig ure 43 Synthesis of AAC prodrugs of APAP and NTX. Reaction conditions: Method A: DCC/DMAP, r.t, 8 12 h; Method B: EDCI/CH2Cl2, r.t, 6 h; Experimental When DCC was used as the coupling reagent to give the Boc protected compounds purification and further characterization of these compounds was a problem because of the presence of the DCC coupling by product, dicyclohexyl urea (DCU). Therefore in order to obtain pure Boc protected AAC prodrugs EDCI were used as the coupling reagent. However t he yields of AAC APAPHCl prodrugs synthesized by the DCC mediated coupling were relatively higher than that obtained using EDCI mediated coupling. The relatively lower yield by the EDCI mediated coupling may be due to the significant aqueous solubility of Boc AAC A PAP prodrugs leading to the loss of the prodrug during the water wash work up. For a poorly water soluble drug, NTX, the yield of the isolated Boc AAC prodrug was relatively high. The

PAGE 90

90 relative yields of the compounds synthesized using DCC or EDCI has been shown in T able 4 2 General procedure for the synthesis of compounds (7a 7c) Method A (DCC/DMAP/CH2Cl2) Equimolar quantities of the APAP, Boc amino acid, DCC and catalytic amount of DMAP in 1520 ml dry methylene chloride were stirred at room temperature for 4 8 h. Completion of reaction was followed by TLC. The resulting suspension was filtered and the solid was discarded. The filtrate was concentrated to an oil on a rotary evaporator. The oil was resuspended in 2025 ml methylene chloride and the organic layer was washed twice with 10 ml 0.1 M cold aqueous HCl. The organic layer was dried over Na2SO4 and concentrated to an oil. The oil was redissolved in 10 ml ether and 2 3 ml of 2 N HCl solution in diethyl ether was added dropwise to it with constant stirring. The completion of deprotection step was followed by TLC. The resulting suspension was triturated with 100200 ml acetone:chloroform (1:10) and filtered. Compounds 7a 7c were obtained as colorless solids. Method B(EDCI/CH2Cl2) Equimolar quantitie s of APAP or NTX Boc amino acid and EDCI in 1520 ml dry methylene chloride were stirred at room temperature for 12 h. Completion of the reaction was followed by TLC. The resulting suspension was washed with water (10 x 3 ml) and the organic layer was dr ied over sodium sulphate and concentrated to an oil. The oil was purified by flash column chromatography to yield the Boc protected compounds. The procedure for deprotection of the Boc group was the same as described in Method A. For the NTX prodrug the deprotection was carried out using triflouroacetic acid.

PAGE 91

91 Table 41 Elemental Analysis and Melting Points for AAC.HCl Prodrugs of 4Hydroxyacetanilide (APAP) And AAC Prodrug of Naltrexone Mp C Cal C Exp H Cal H Exp N Cal C Exp 7a 200240(d) 51.07 51.01 5.84 5.82 10.83 10.83 7b 230250 (d) 54.45 54.51 6.68 6.65 9.77 9.74 7c 195230 (d) 54.84 54.87 6.02 6.02 9.84 9.70 10 6468 49.43 49.60a 5.44a 5.28 3.98a 4.01 a calculated values for the VALNTX 2TFA 2H2O 6a N Boc ALAAPAP Synthesized from 2.0 g (10.58 mmole) N Boc L a lanine, 2.03 g (10. 58 mmole) EDCI and 1.6 g (10.58 mmole) APAP. NMR (CDCl3 = 8.8, 2H), 7.04 (d, J = 9.2, 2H), 5.05 (s, 1H), 4.50 (broad s, 1H), 2.17 (s, 3H), 1.46 (s, 9H). Yield: 60%. 6b N Boc VAL APAP Synthesiz ed from 1.0 g (4.58 mmole) N Boc L v aline, 0.88 g (4.60 mmole) EDCI and 0.7 g (4.63 mmole) APAP. NMR (CDCl3 2H), 7.02 (d, J = 9.2, 2H), 5.02 (broad d, 1H), 4.52 (t, 1H), 2.6 (m 1H), 2.17 (s, 3H), 1.46 (s, 9H) 3H). Yield: 46% (average of two experiments). 6c N Boc PRO APAP Synthesized from 2 g (9.3 mmole) N Boc L p roline, 1.7 g (9.27 mmole) EDCI and 1.4 g (9.30 mmole) APAP. NMR (CDCl3 2H), 7.03 (d, J = 9.2, 2H), 4.0 4.32 (dt, 1H), 3.25 3.61 (m, 2H), 2.2 2.3 (m, 2H), 2.16 (s, 3H), 1.95 2.05 (m, 2H), 1.47 (broad s, 9H). Yield: 31% Oil at rt. 7a ALA APAP HCl (Acetaminophen Alaninate Hydrochloride ALA APAP HCl ) NMR (D2 2H), 7.04 (d, J = 8.0 2H), 4.37 (q, 1H), 2.01 (s, 3H), 1.59 (d, 2H). Ove rall yield over two steps: Method A: 74%; Method B: 44%. 7b VAL APAP HCl (Acetaminophen Valinate Hydrochloride VAL APAP HCl ) NMR (D2 (d, J = 8. 0, 2H), 7.18 (d, J = 8.1, 2H), 4.31 (d, 1H), 2.53 (sept 1 H), 2.14 ( s, 3 H) 1.14 (d, 6H) Overall y i eld over two steps: Method A: 69%; Method B: 57%.

PAGE 92

92 7c PRO APAP HCl (Acetaminophen Prolinate Hydrochloride PRO APAP HCl ) NMR (D2 = 8.4, 2H), 7.22 (d, J = 8.2z 2 H), 4.79 (q, 1H), 3.50 (m 2H), 2.61 (m, 1H), 2.42 (m, 1H), 2.17 (t, 2H), 2.17 (s, 3H). Overall yield over two steps: Method A: 84% Method B: 27% Table 4 2: Percentage y ields of the Boc AAC APAP c ompounds and the c orresponding AAC p rodrugs synthesized f rom Method A or Method B % Yield Method A % Yield Method B Boc AAC APAP 6a 60 6b -46a 6c -31a AAC APAPHCl 7a 74 44 7b 69 57 7c 84 27 AAC NTX 9 -52 10 -90 a Average of two experiments 9 N Boc VAL NTX Synt hesized from 0.37g (1.08 mmole) NTX 0.21 g (1.09 mmole) EDCI and 0.23 g (1.08 mmole) LBoc v aline OH NMR (CDCl3 = 8 .1 1H), 6.68 (d, J = 8 .2 1H), 5.17 (d, 1H), 4.68 (s, 1H), 4.4 (d, 1H), 2.58 (sextet, 1H), 1.46 (s, 9H), 1.08 (quartet, 6H), 0.13 0.17 (quartet, 2H). Yield: 52%. Oil at room temp. 10 VAL NTX 2 TFA 2H2O NMR (D2J = 8 .0 1H), 6.98 (d, J = 8 .2 1H), 5.15 (s, 1H), 4.4 (d, 1H), 2.58 (sextet, 1H), 1.20 (quartet, 6H), 0.450.4 7 (m, 2H). Hydrolysis of AAC prodrugs For determining half lives of the members of AAC prodrugs a solution of the prodrug was prepared by dissol ving a small amount of the prodrug in deionized water to

PAGE 93

93 give an approximate concentration of about 102 0 mM. An aliquot of the aqueous solution was immediately diluted w ith buffer (pH 6.0, phosphate, 2 0 mM, I = 0.5). The diluted solution was immediately a dded into a UV cuvette and absorbance values were recorded till approximately >99% of hydrolysis was complete (78 half lives). The cuvette was maintained at 37 2oC throughout the experiment. Since the molar absorptivities of the prodrugs are higher than that of the parent drug, APAP, the hydrolysis of the prodrugs was monitored by the decrease in the absorbance values over time. Additionally, the prodrugs showed an absorbance maximum very close to that of APAP. Therefore, the absorbance values were recor ded at 244 nm. For the NTX prodrug, the abrorbance was recorded at 278 nm. The pseudo unimolecular rate constants were obtained from a plot of log (At A here At is the absorbance at time t and A reaction, at which time the absorbance would reflect the concentration of APAP only. All the experiments were run in triplicate. Table 43 Half lives (t1/2, min) and p redicted pKa values of AAC prodrugs of APAP a nd n altrexone in b uffer (pH 6.0, p hosphate, 20 mM, I = 0.5) at 37 0.5 0C pKa t1/2 (min) a 7a 8.62 32.33 0.28 7b 7.46 91.37 3.45 7c 8.65 14.90 0.81 10 -31.78 1.52 a All experiments were run in triplicate. The AAC APAP pro drugs showed shorter half lives than the DAAC APAP pro drugs. This is an expected result because of the greater ease with which the unsubstituted amine group, NH2, can facilitate the general base catalysis of the hydrolysis of these compounds or a nucleophilic attack to give the diketopiperazine (see

PAGE 94

94 chapter 3 for more detailed explanation). Amon g the AAC APAPHCl prodrugs, steric effect exerted by the R4 group (Figure 4 1 ) seemed to play an important role in governing the t1/2 values for these prodrugs. The v aline ester prodrug 7b (VALAPAPHCl) showed the longest half life amongst all the AAC compounds studied due to the bulky isopropyl group. The shortest t1/2 amongst all the prodrugs evaluated was 14.9 min for 7c ( PRO APAP HCl) prodrug. Figure 44. Proposed mechanism of general base catalysis by the proline ring. (A) Presence of the proline ring fixes four atoms leading to rate enhancement in hydrolysis of glycylproline ethyl ester. (B) general base catalysis by the proline ring. This is an unexpected result based on the steric factors alone. However, it has been previously observed that piperazine2, 5 diones are formed readily from dipeptide esters particularly those containing N methyl amino acids or p roline. In the case of dipeptides containing proline, geometrical factors take control .85 For the cyclization of dipeptide esters to piperazines to occur, the peptide bond must be in the cis conformation so that the terminal amino group and the ester carbonyl carbon can interact to form the six membered ring. For example, glycylproline ethyl ester cyclizes more readily than t he prolylglycine ester because of the ease in assumption of the cis conformation.86 The case of the 7c can be explained by similar explanation. Presence of the proline ring system fixes the NH group in an orientation that favors more effective

PAGE 95

95 general base catalysis by the NH group. The possible transition state involved in hydrolysis of 7c has been shown (as a free base) in Figure 44 De termination of Solubilities of AAC APAP HCl Prodrugs And VAL NTX TFA Prodrug The molar absorptivity of each AAC APAPHCl prodr ug (7a 7c) was determined in methanol in triplicate. Molar absorptivitiy of naltrexone (NTX) and its AAC prodrug VAL NTX 2TFA (10) was also determined in methanol. Molar absorptivities for parent drugs and their corresponding AAC prodrugs have been liste d in Table 44 A known amount of prodrug was dissolved in deionized water, and the solution was immediately diluted with methanol or pH 7.1 phosphate buffer and analyzed by UV 244 was calculated with Beers law (shown for APAP) : A244 244 l C, where l = cell length (1) Table 44 : Molar a bsorptivities of AAC APAP HCl p rodrugs, VAL NTX TFA and NTX MeOH 7.1 7a, ALAAPA P HCl 1. 541 0.019a, b c 7b, VAL APAP HCl 1.553 0.031a,b, c 7c PRO APAP HCl -10, VAL NTX TFA 2.175 0.059 a, d, e APAP -1.030 0.006a, b c NTX 0. 94 0.005 a, d, e 1.181 0.029 a, d, e a All experiments were run in triplicate. b Unit s of 104 ml mmole-1 c UV max = 244 nm, d Units of 103 ml mmole-1 e UV max = 282 nM For eac h prodrug, the solubility in 1o ctanol (OCT) was determined in triplicate. The prodrug was crushed into a fine powder and a saturated solution of the prodrug in octa nol was obtained by adding an excess of each compound to a test tube containing 1 ml octanol.

PAGE 96

96 Table 45. Physicochemical properties of AAC Prodrugs. Molecular Weight (MW), solubility in 1 o ctanol (SOCT), solubility in propylene g lycol (SPG) for AAC APAPHCl p rodrugs (7a 7c), VAL NTX TFA p rodrug (10) and DAAC APAPHCl Prodrugs (4a 4k) a U nits of mM; b Decomposes with charring close to melting temperature; c MW of the monohydrate; d Mp of the monohydrate and that of the bis hydrate in paranthesis; e distance of amine nitrogen from acyl group n = 1; f n = 2; g n = 3; h Values from Kasting Sm ith and Cooper (1989) ; I Values calculated from Kaufman et. al. 197587 using log SOCT = log SAQ + log KOCT:Water; j Experiment run in duplicate. The test tube was the n insulated and the suspension was allowed to stir at room temperature (23 1oC) overnight on a magnetic stir plate. The suspension was filt ered and analyzed by UV spectrophotometry. A NMR of the flitered solid was recorded to ensure that the prodrugs were intact during the course of solubility experim ent. The absorbance at 244 nm was used to calculate the concentration of the AAC APAPHCl MW Mp SOCT a SPG a 7a 272 200240(d)b 1.02 0.11 335.57 4.74 7b 286 230250 (d)b 4.65 0.38 649.79 1.13 7c 300 195230 (d) b 1.336 0.28 260.91 13.94 10 686c 170 (6468) d 6.514 0.358 155.57 0.39 j 4ae 272 0.352 0.01 122.05 2.90 4bf 286 195200 0.291 0.035 150.09 2.03 4ce 300 200205 1.297 0.067 93.80 0.98 4df 314 170180 1.162 0.055 176.10 3.42 4fe 313 222 --4gf 327 234 0.202 0.02 18.17 1.05 4hg 341 202 20.87 0.47 64.09 1.17 4ie 315 260270 0.09 0.01 14.01 0.01 4jf 329 220 0.10 0.007 29.97 0.58 4kg 343 210 1.522 0.077 39.68 0.50 APAP 151 195 158.94h 662.25 h NTX 341 175 694.77i 101.51 7.5 j

PAGE 97

97 prodrug in octanol and the absorbance at 278 was used to calculate concentration of the naltrexone prodrug in octanol. S olubility in octanol SOCT: CSaturation = SOC T = A244 244 (2) Solubilities in propylene glycol (PG) were also determined in triplicate by the procedure used to determine SOCT. The suspensions were stirred for 4 h before filtration. A sample of the filtrate was diluted with methanol and analyzed by UV sp ectrophotometry. Melting points of AAC APAPHCl prodrugs were not sharp and the compounds exhibited decomposition close to the melting temperature. All the AAC APAP prodrugs showed a higher melting point range than APAP. High melting point values are most probably responsible for the low SOCT values exhibited by these compounds. Among the AAC APAP compounds 7b ( valine ester) showed the highest SOCT. The SOCT for DAAC APAPHCl prodrugs was comparable to the AAC series. SPG of the AAC APAPHCl prodrugs was generally higher than that for DAAC APAPHCL series. For example the most PG soluble member in the D AAC series is 4d (176 mM) whereas the valine ester 7b is almost 3.7 times more soluble in PG (650 mM). The trend in the PG solubility reflects the propensit y of NH2 group vs. NR1R2 group in modulating t he aqueous like solubility of the AAC vs. DAAC prodrug series. The reason for this trend is that -+NH3 has more N H groups to hydrogen bond with protic solvents water and PG. The SOCT determination experime nt with the NTX prodrug 10 showed that although the prodrug was isolated as a bis hydrate, the compound precipitates out of the saturated octanol solution as a monohydrate. The monohydrate was isolated and

PAGE 98

98 characterized by elemental analysis ( % C Exp 50. 49; Cal 50.739) and 1H NMR. The colorless m onohydrate exhibited a higher melting point than that of the pale brown bis hydrate ( shown in Table 45 ) The dependence of SOCT on melting point is again exemplified here because the lower melting bis hydrate wa s highly soluble in octanol (>500 mg / 0.5 ml) but when the solution was stirred for about 30 minutes the less soluble, higher melti ng mono hydrate precipitated The SOCT values reported in Table 45 correspond to the solubility of the monohydrate. In V itro Evaluation of VAL APAP HCl and VALNTX TFA Prodrugs Two different mice were used to determine the flux of each prodrug. Prior to skin removal, the mice were rendered unconscious by CO2 then sacrificed via cervical dislocation. Skins were removed by bl unt dissection and placed dermal side down in contact with pH 7.1 phosphate buffer (0.05 M, I = 0.11 M, 32 oC) containing 0.11% formaldehyde which has been shown to inhibit microbial growth and maintain the integrity of the skins throughout the experiment.21 Prior to the application of the prodrugs as suspensions in IPM :PG (99:1) the skins were maintained in contact with buffer f or 24 h to leach out all UV absorbing material. During this conditioning phase time, the r eceptor phase was removed and replaced with buffer 3 times. The suspension of the prodrug in IPM :PG was prepared 4 h before application and allowed to stir at room temperature (25 1 oC) until application. After the 24 hour leaching period, an aliquot (0.5 ml) of the prodrug suspension was added to the surface of the skin (donor phase). Samples of the receptor phase were usually taken at 8, 19, 22, 25, 28, 31, 34, and 48 h and quickly analyzed by UV spectrophotometry. Since only parent drug was obtained in the receptor phase in all cases the amounts of permeated APAP was quantified using

PAGE 99

99 Table 46.Results from diffusion cell experiments with AAC prodrugs of APAP and NTX (VAL APAPHCl and VAL NTX TFA). Maximum flux of the prodrug from a saturated isopropyl myristate (IPM) vehicle through hairl ess mouse skin (JM), maximum flux of standard drug, t heophylline, from a saturated propylene glycol (PG) vehicle (JS), log of experimental flux values (EXP log JS), residual concentration of the drug in the skin (CS), amount of parent drug (APAP or NTX) de livered during the 48 h first application period (AAPP) Compound JM a JS a CS a,c EXP log JM AAPP d 7c VAL APAP HCl 0.47 0.02 0.67 0.18 1.86 0.07 0.32 60.73 21.9 10 VAL NTX TFA 0.074 0.00 0.50 0.04 2.36 0.46 1.13 13.20 0.02 4a Me2n1APAP H C lb 0.64 0.06 1.00 0.02 0.69 0.07 0.19 59.62 4.06 4 b Me2n2APAP HClb 0.65 0.01 0.95 0.03 0.69 0.05 0.19 63.30 1.51 4 i MORn1APAPHC lb 0.54 0.02 0.96 0.25 0.71 0.18 0.26 60.53 1.67 NTX HCl 0.057 0.005 0.78 0.05 1.83 0.15 1.24 11.42 0.43 APAP 0.51 0.74 2.74 0.70 0.29 73.63 a Units of mole cm-2 h-1. b Suspension of the prodrug was applied in a (99:1) PG:IPM vehicle. c Concentration of APAP in the skin after a 24 h leaching period. d Units of mole molar absorpti vity of APAP or NTX in the receptor phase. At each sampling time, the entire receptor phase was replaced with fresh buffer in order to maintain sink conditions. After the 48 h of the first application period, the donor suspension was removed and the skins were washed three times with methanol (35 ml) to remove any residual prodrug from the surface of the skin. The remaining prodrug or APAP in the skin was leached out by keeping the skins in contact with buffer for an additional 24 h. Then the receptor phas e was replaced with fresh buffer and an aliquot (0.5 ml) of a standard drug theophylline was applied in PG to the skin surface. The second application fluxes were determined by sampling of the receptor phase at 2, 4, 6 h and analysis by UV spectrophotometr y. The concentration of theophylline in the receptor phase was 1). At each

PAGE 100

100 sampling time, the entire receptor phase was removed and replaced with fresh buffer. In each experiment, the flux was determined by plotting the cumulative amount of APAP versus time. J in units of mol cm2 h1 could then be calculated by dividing the slope of the steady stat e portion of the graph (Figure 45) by the surface area of the skin (4.9 cm2). The results f or the diffusion cell experiments with VALAPAPHCl and VAL NTX TFA are shown in Table 46. The flux of VALAPAPHCl was only as high as APAP itself. Amongst all the DAAC APAPHCl and AACAPAPHCl prodrugs studied the dimethyl containing salts gave best fl ux (1.2 times APAP). This is the first study where permeation experiments with prodrugs of APAP as hydrochlorides has been reported. All the hydrochloride prodrugs investigated in the study showed a longer lag time as compared to APAP or the corresponding free base forms. The permeation profile of NTX HCl and its prodrug, VALNTX TFA has been shown in Figure 45. The VALNTX TFA prodrug showed moderately higher flux than NTX.HCl (1.3 times). The lag time for both the salts was approximately the same (~24 h ). The lag times observed for NTX HCl and its prodrug was very similar to that observed for DAAC APAPHCl prodrugs and the AAC APAPHCl prodrug. NTX HCl deliver ed 11.42 mole NTX whereas its prodrug, VALNTX TFA deliver ed 13.2 mole over a 48 h application period through hairless mouse skin from a saturated IPM donor phase. The skin retention of NTX after the application period was higher for the NT X prodrug by 1.3 times (2.4 vs 1.8 moles).

PAGE 101

101 Figure 45. Cumulative amount of NTX permeated vs time for compounds 10 and NTX HCl The second application flux values were not higher than the control value of 1 .0 mole cm2 h1. This showed that the skin used in the permeation experiment did not undergo any damage due to the use of IPM as the donor phase. The inc orporation of a basic amine into the promoiety of an acyl prodrug of a phenol containing drug resulted in an increase in lipid and aqueous solubilities of the corresponding prodrugs in this study. However, the increased solubilities did not result in a concomitant increase in the fl ux of the prodrugs when investigated as free bases (investigated in chapter 3) or as their corresponding salt forms (investigated in chapter 3 and 4). The lower flux of the DAAC and AAC prodrugs investigated in chapter 3 and 4 ca n be attr ibuted to the presence of hydrated forms of the prodrug as it diffuses through the skin. Based on solubility properties only, the flux of a prodrug in a salt form is expected to be lower than the parent drug because the lipid solubility of a salt is considerably lower than the parent drug, even though its water solubility is much higher again a

PAGE 102

102 balance in lipid and aqueous solubility is lacking. The slightly higher flux of the salt forms observed here might be attributed to: (a) existence of an equilibrium between the protonated and the free base forms in the skin microenvironment and (b) a favorable interaction of the protonated amine groups with the fatty acid groups in the skin. Conclusions In this part of study three AAC prodrugs of APAP wer e synthesized via a DCC/DMAP or a EDCI mediated coupling reaction Based on the solubility properties of the VALAPAPHCl, o ne AAC prodrug of NTX VAL NTX TFA was also synthesized. The Boc protected compounds couldnt be isolated in very pure form using the DCC method because of the presence of dicy clohexylurea by product. Thus t he Boc protected compounds that were isolated from the EDCI coupling method were characterized. The Boc protected NTX prodrug was isolated using the EDCI method. Hydrolysis experi ments with the AAC prodrugs were carried out in buffer (pH 6.0, phosphate, 20 mM, 37 0C). The half lives of the prodrugs were affected by the steric ef fect of the R group at the position alpha to the carbonyl group. The small t1/2 value for the 7c (PRO APA P HCL) could be attributed to the possible transition state that the proline ring in the prodrug can facilitate more readily than alanine or valine. Solubility experiments with AAC APAP compounds were carried out in 1octanol and propylene glycol. The com pounds were only moderately soluble in o ctanol but exhibited relatively higher PG solubility than DAAC APAPHCl prodrugs. The naltrexone prodrug exists in form of mono and bis hydrates where the mono hydrate was present as the primary species in a saturated octanol solution. In vitro permeation experiments were carried out with the VALAPAPHCl prodrug and VAL NTX 2TFA prodrug. The valine ester prodrug of APAP showed approximately

PAGE 103

103 the same flux value as APAP. The naltexone prodrug showed a moderately higher flux than NTX HCl (1.3 times). The skin retention of the VALNTX TFA prodrug was about 1.3 times comapared to NTX HCl. In conclusion, the AAC promoiety is successful in enhancing the aqueous like solubility of the corresponding prodrugs synthesized in this study. The solubility and stability of the prodrugs can be modulated by changing the steric bulk on the alpha carbon appropriately. The increase in water solubility of the AAC prodrugs did not res u l ts in a concomitant increase in their flux. This might be due to the presence of hydrated forms of the prodrug as it diffuses through the skin. The introduction of the ionized group in the AAC hydrochloride prodrugs make them an attractive targets for iontophoretic drug delivery. A combination of more than one permeation enhancement technique, like chemical enhancers, might be the next step in further optimization of the DAAC and AAC prodrug approach presented in this work.

PAGE 104

104 CHAPTER 5 CONCLUSION AND FUTURE WORK In this study a prodrug approach in which, a bas ic amine functional group was incorporated into a promoiety of an acyl prodrug of a phenolic drug, was developed. A model phenolic drug acetaminophen (APAP) was used to demonstrate the feasibility of the prodrug approach. The permeation experiments on hairless mouse skin with APAP prodrugs showed that the dialkylaminoalkylcarbonyl promoiety was successful in improving the flux of APAP by three fold. This prodrug approach can now be applied to a broader class of phenolic parent drugs to achieve enhanced delivery through the skin. The first objective of this s tudy was to synthesize acyl prodrugs of a model phenolic drug, acetaminophen (APAP) containing a basic amine group. This was achieved by coupling the N, N` d ialkylaminoalkanoic acid hydrochlorides wit h the parent phenolic drug via a DCC mediated coupling reaction. The corresponding N` d ialkylaminoalkanoic acid hydrochlorides were synthesized using five different amines dimethylamine, diethylamine, dipropylamine, morpholine and piperidine. The N, N` d ialkylaminoalkylcarbonyl (DAAC) ester prodrugs were obtained in good yields (70 90%). Additionally, aminoalkylcarbonyl (AAC) prodrugs of APAP and one AAC ester prodrug of n altrexone (NTX) was synthesized. The AAC ester prodrugs were synthesized using DCC or EDCI as the coupling reagents. The yields of the AAC APAP and AAC NTX prodrugs were between 5075%. The second objective of this study was to evaluate the physicochemical p roperties of DAAC APAP, AAC APAP and AAC NTX prodrugs. Half lives of a few members was evaluated in buffer (pH 6.0, phosphate, 50 mM, I = 0.5). The DAAC and AAC prodrugs hydrolyzed to the parent drug with half lives between 15 115 minutes in pH 6.0 buffer.

PAGE 105

105 The kinetic behavior of the DAAC and AAC prodrugs exemplifies the general bas e catalysis by the basic amine group in the prodrugs that results in significant rate enhancement (see page 92) compared to esters lacking a basic amine functionality. Solubilities of DAAC prodrugs in buffer (acetate, pH 4.0, 50 mM), 1octanol (OCT), propy lene glycol (PG) and isopropyl myristate (IPM) have been determined. The solubilities of the AAC prodrugs were determined in 1octanol and propylene glycol. Amongst the DAAC APAPHCl prodrugs the member exhibiting the highest SPG was also the one that exhi bited the highest S4.0. SOCT of the DAAC APAPHCl were generally low. All the DAAC APAP prodrugs exhibited higher SIPM than APAP. The SIPM of the DAAC APAP ( free bases ) showed a dependence on their melting points. Among the AAC APAPHCl prodrugs, the VALAPAP HCl exhibited the highest SOCT and SPG. The VAL NTX 2TFA prodrug also exhibited higher SPG than NTX. SOCT of the NTX prodrug was lower than NTX itself which is expected from a ionic compound. The third objective of this research was to investigate the potential of DAAC and AAC prodrugs in enhancing the delivery of APAP through skin. The DAAC prodrugs were evaluated for their ability to deliver APAP through hairless mouse skin. The best member of the DAAC series of prodrugs showed three times higher flux than APAP. Although the best member evaluated was much more lipid and water soluble than APAP, the delivery of APAP was not as high as expected. We hypothesize that this may be due to an increase in the molecular weight caused by water association with t he amine group as the prodrug diffuses the membrane. A new Robert Sloan equation using a n = 73 database was developed. The coefficients for the R S equation were x = 0.599, y = 0.502, z = 0.00235 and r2 = 0.92. The calculated flux values for DAAC APAP

PAGE 106

106 pr odrugs using the new RS equation were higher than experimental flux values in all cases except one. The over prediction of flux of DAAC APAP prodrugs by the RS equation may be due to challenges in precisely measuring or estimat ing the solubility of the actual ionized species in the skin microenvironment while the prodrugs diffuse through the skin. One appro ach to improve the flux of the DAAC HCl prodrugs is to deliver the prodrugs through skin iontophoretically. The presence of the ionizable amine in the DAAC or AAC prodrug allows for a utilization of iontophoresis as a permeation enhancement technique. The pH of the skin environment has been shown to be about 5.5 6.0 5 All the prodrugs are expected to be protonated in the skin microenvironment. Therefore an applied electric field is expected to facilitate the passage of the ionized molecule through skin more effectively. Recent work by Bowstra and coworkers shows that the i ontophoretic flux of the alanine ester prodrug of a p a rent phenolic drug was higher than that of the parent drug.77 Figure 51: Chemical structures of proposed DAAC prodrugs of phenol containing drugs.

PAGE 107

107 The amount of the parent drug delivered by DAAC prodrugs seems to be dependent on the nature of the donor phase solvent. B etter solvation of the amine group in an aqueous donor phase had a favorable effect on the flux of the testosterone prodrugs.55 The primary difference between testosterone dimethy l butyrate hydrochloride prodrug and the DAAC APAP prodrugs is the fact that the leaving group is a phenol in the latter series relative to a poorer alcoholic leaving group in the testosterone prodrug. This made it feasible to deliver the prodrug from an aqueous solution whereas the DAAC APAP prodrugs are unstable in water over the course of a diffusion cell experiment One way to assess the effect of incorporation of a basic amine of the flux of a parent drug is to synthesize aminoal kyl carbonyl derivatives of a drug containing an alcoholic group. The corresponding prodrugs could then be delivered from an aqueous suspension allow ing an analysis of the effect of incorporation of a basic amine group on the flux of the corresponding prodrug. In compounds investigated so far, although incorporation of a basic amine caused a favorable enhancement in the solubility properties of the corresponding prodrugs, the increase in solubility is offset by increase in effective molecular weight of the more basic prodrugs. One strategy to overcome this challenge is to synthesize DAAC prodrugs with relatively weak ly basic amines in the promoeity. The suggested compounds are shown in F igure 5 1. This can facilitate a similar enhancement of solubilities without the increment in effective molecular weight of the prodrug while it diffuses through the skin. Additionally the weakly basic amine prodrugs are expected to be relatively more stable in aqueous phase and therefore can be delivered from an aqueous donor phase.

PAGE 108

108 In the present study, the utility of the DAAC and the AAC prodrugs approach was investigated as a means of improving topical delivery of phenol containing drugs. When delivered as a free base one of the DAAC APAP prodrug was able to improve the delivery of APAP by three times compared to its hydrochloride salt which showed moderately higher flux than APAP (1.2 times). Similarly the AAC NTX prodrug VALNTX TFA was able to improve the topical delivery of NTX by 1.3 times compared to NTX HCl. Previous studies with topical delivery of ionized compounds have met with success when a combination of more than one en hancement technique is used.8991 One of the approaches to further enhance the delivery DAAC and AAC prodrugs and their corresponding salts is by the use of chemical permeation en hancers in addition to the iontophor etic technique mentioned above. Such a combined transdermal delivery system can be useful in improving the d elivery of compounds having an ionizable amine.

PAGE 109

109 LIST OF REFERENCES 1. Gibson, G.; Skett, P. An Intoduction to Drug Metabolism Chapman and Hall: 1986. 2. Longc ope, C. G. S.; Goldin, B.; Woods, M.; Dwyer, J.; Warram, J. The Metabolism of Estradiol, Oral Compared to Intravenous Administration. J. Ster. Biochem. 1985, 23, 10651070. 3. Wolfe, M. M.; Lichtenstein, D. R.; Singh, G. Gastrointestinal Toxicity of NonSteroidal Anti Inflammatory Drugs. N Engl J Med 1999, 340, 1888 1899. 4. Hargus, S. J.; Amouzedeh, H. R.; Pumford, N. R.; Myers, T. G.; McCoy, S. C.; Pohl, L. R. Metabolic Activation and Immunochemical Localization of Liver Protein Adducts of the Nonsteroidal Anti Inflammatory Drug Diclofenac. Chem. Res. Toxicol. 1995, 7, 575582. 5. Schaefer, H.; Redelmeier, T. Skin Barrier. Principles of Percutaneous Absorption. S. Karger Basel: 1996. 6. Swanson, H. Cytochrome P450 Expression in Human Keratinocytes: An Aryl Hydrocarbon Receptor Perspective. Chem Biol Interact. 2004, 149, 6979. 7. Williams, A. Transdermal and Topical Drug Delivery Pharmaceutical Press: London, 2003. 8. Pham, M.; Magdalou, J.; Totis, M.; Fournel Gi gleux, S.; Siest, G.; Hammock, B. Characterization of Distinct Forms of Cytochromes P 450, Epoxide Metabolizing Enzymes and UdpGlucuronosyltransferases in Rat Skin. Biochem. Pharmacol. 1989, 38, 21872194. 9. Bashir, S.; Maibach, H. Cutaneous Metabolism o f Xenobiotics. In Percutaneous Absorption: Drugs Cosmetics Mechanisms Methodology Bronaugh, R.; Maibach, H., Eds. Taylor and Francis: Boca Raton, 2005; pp 5163. 10. Prausnitz, M. R.; Mitragotri, S.; Langer, R.; Current Status and Future Potential of Tr ansdermal Drug Delivery. Nat. Rev. Drug Discov. 2004, 3, 115124. 11. Rhoden, E. L.; Morgentaler, A. Risks of TestosteroneReplacement Therapy and Recommendations for Monitoring. N Engl J. Med 2004, 350, 48292. 12. Carrasco, D.; Prieto, M.; Pallardo, L Multiple Hepatic Adenomas after Long Term Therapy with Testosterone Enanthate: Review of the Literature. J. Hepatol. 1985, 1, 573578. 13. Comhaire, F. H. Andropause: Hormone Replacement Therapy in the Aging Male. Eur. Urol. 2000, 38, 655 662.

PAGE 110

110 14. Wang, C.; Swerdloff, R. S.; Iranmanesh, A.; Dobs, A.; Snyder, P. J.; Cunningham, G.; Matsumoto, A. M.; Weber, T.; Berman, N. Transdermal Testosterone Gel Improves Sexual Function, Mood, Muscle Strength, and Body Composition Parameters in Hypogonadal Men. Journal of Clinical Endocrinology and Metabolism 2000, 85, 283953. 15. Landmann, L. E. Transformation of Lamellar GranuleDisks into Intercellular Sheets by a MembraneFusion Process, a FreezeFracture Study. J. Invest. Dermatol. 1986, 87, 202 209. 16. Swartzendruber, D. C.; Wertz, P. W.; Kitko, D. J.; Madison, K. C.; Downing, D. J. Molecular Models of the Intercellular Lipid Lamellae in Mammalian Stratum Corneum. J. Invest. Dermatol. 1989, 92, 251 257. 17. Bowstra, J. A.; Gorvis, S. G.; Maria, P. Skin Lipid Or ganization, Composition and Barrier Function. IFSCC Magazine 2007, 10, 297307. 18. Bowstra, J. A.; Gorvis, S. G.; Spek, J. A.; Bras, W. Structural Investigations of Human Stratum Corneum by Small Angle X Ray Scattering. Soc. Invest. Derm. 1991, 97, 10051012. 19. Sloan, K. B. Functional Group Considerations in the Development of Prodrug Approaches to Solving Topical Delivery Problems. In Prodrugs: Topical and Ocular Drug Delivery S loan, K. B., Ed. Marcel Dekker: New York, 1992; pp 17116. 20. Williams, A C.; Barry, B. W. Penetration Enhancers. Adv. Drug Del. Rev. 2004, 56, 603618. 21. Sherertz, E. F.; Sloan, K. B.; McTiernan R. G. Effect of Skin Pretreatment with Vehicle Alone or Drug in Vehicle on Flux of a Subsequently Applied Drug: Results of Hairles s Mouse Skin and Diffusion Cell Studies. J. Invest. Dermatol. 1987, 89, 24952. 22. Van Hal, D. A. ; Jeremiasse, E.; Junginger, H. E.; Spies, F.; Bouwstra, J. A. Structure of Fully Hydrated Human Stratum Corneum: A FreezeFracture Electron Microscopy Study. J. Invest. Dermatol. 1996, 106, 89 95. 23. Singh, J.; Maibach, H., I. Topical Iontopphoresis inVivo: Historical Development, Devices and Future. Dermatology 1993, 187, 235238. 24. Warwick, W. J. et. al. Evaluation of a Cystic Fibrosis Screening System I ncorporating a Miniature Sweat Stimulator and Disposable Chloride Sensor. Clin. Chem. 1986, 32, 850 853. 25. Miller, K. A.; Balakrishnan, G.; Eichbauer, G.; Betley, K. 1% Lidocaine Injection or Numby Stuff for Topical Analgesia Associated with Peripheral Intravenous Cannulation. Ana. J. 2001, 69, 185 187.

PAGE 111

111 26. Matriano, J. A. et al. Macroflux Microprojection Array Patch Technology: A New and Efficient Approach for Intracutaneous Immunization Pharm. Res. 2002, 19, 63 70. 27. Cross, S. E.; Roberts, M. S. Ph ysical Enhancement of Transdermal Drug Application: Is Delivery Technology Keeping up wi th Pharmaceutical Development? Curr. Drug Deliv. 2004, 1, 8192. 28. Mikszta, J. An Improved Genetic Immunization Via Micromechanical Disruption of Skin Barrier Functio n and Targeted Epidermal Delivery. Nature Med. 2002, 8, 415 419. 29. Rein, H. Experimental Electroendosmotic Studies on Living Human Skin. Z. Biol. 1924, 81, 124. 30. Scheuplein, R. J.; Blank, I. H. Permeability of the Skin. Phys. Revs. 1971, 51, 702747. 31. Sche uplein, R. J.; Blank, I. H. Mechani sm of Percutaneous Absorption IV : Penetration of Nonelectrolytes (Alcohols) Form Aqueous Solutions or Pure Liquids. J. Invest. Dermatol. 1972, 5, 286296. 32. Sloan, K. B.; Wasdo, S. C.; Rautio, J. Design for Op timized Topical Delivery: Prodrugs and a Paradigm Change. Pharm. Res. 2006, 23, 27292746. 33. Majumdar, S.; Thomas, J.; Wasdo, S.; Sloan, K. B. The Effect of Water Solubility of Solutes on Their Flux through Human Skin in Vitro. Int. J. Pharm. 2007, 329, 2536. 34. Kasting, G. B.; Smith, R. L.; Cooper, E. R. Effect of Lipid Solubilities and Molecular Size on Percutaneous Absorption. In Skin Pharmacokinetics Shroot B.; Schaefer, H., Ed. Basel Karger: 1987; pp 138153. 35. Potts, R. O.; Guy, R. H. Predictin g Skin Permeability. Pharm Res 1992, 9, 6639. 36. Roberts, W. J.; Sloan, K. B. Correlation of Aqueous and Lipid Solubilities with Flux for Prodrugs of 5Fluorouracil, Theophylline and 6Mercaptopurine: A Potts Guy Approach. J. Pharm. Sci. 1999, 88, 515522. 37. Sloan, K. B. Prodrugs for Dermal Dleivery. Adv. Drug Del. Rev. 1989, 3, 67101. 38. Wasdo, S.; Sloan, K. B. Designing for Topical Delivery: Prodrugs Can Make the Difference. Med. Res. Rev. 2003, 23, 763 793. 39. Sloan, K. B.; Wasdo, S.; Ezike Mkpa ru, U.; Murray, T.; Nickels, D.; Singh, S.; Shanks, T.; Tovar, J.; Ulmer, K.; Waranis, R. Topical Delivery of 5Fluorouracil and 6 Mercaptopurine by Their Alkylcarbonyloxymethyl Prodrugs from Water: Vehicle Effects on Design of Prodrugs. Pharm Res 2003, 20, 639 45.

PAGE 112

112 40. Wasdo, S. C.; Juntunen, J.; Devarajan, H.; Sloan, K. B. A Comparison of the Fit of Flux through Hairless Mouse Skin from Water Data to Three Model Equations. Int. J. Pharm. 2009, 366, 65 73. 41. Juntunen, J. ; M ajumdar S.; Sloan, K. B. The Effect of Water Solubility of Solutes on Their Flux through Human Skin in Vitro: A Prodrug Database Integrated into the Extended Flynn Database. Int. J. Pharm. 2008, 351, 92103. 42. Bundgaard, H. Bioreversible Derivatives for Various Functional Groups and Chemical Entities. In Design of Prodrugs Bundgaard, H., Ed. Elsevier Science Publishers B. V. (Biomedical Division) 1985; pp 192. 43. Varia, S. A.; Schuller, S.; Sloan, K. B.; Stella, V. J. Phenytoin Prodrugs Iii Water Soluble Prodrugs for Oral and/or P arenteral Use. J. Pharm Sci 1984, 73, 10681073. 44. Rothbard, J. B.; Garlington, S.; Lin, Q.; Kirschberg, T.; Kreider, E.; McGrane, P., L.; Wender, P. A.; Khavari, P. A. Conjugation of Arginine Oligomers to Cyclosporin a Facilitates Topical Delivery and I nhibition of Inflammation. Nat. Med. 2000, 6, 12531257. 4 5. Rothbard, J. B.; Jessop, T. C.; Wender, P. A. Adaptive Translocation: The Role of Hydrogen Bonding and Membrane Potential in the Uptake of Guanidinium Rich Transporters into Cells. Adv. Drug Deli v. Rev. 2005, 57, 495504. 46. Dhareshwar, S. S.; Stella, V. J. Prodrugs:Challenges and Review. Springer.: 2008; Vol. 3. 47. Drustrup, J.; Fullerton, A.; Christrup, L.; Bundgaard, H. Utilization of Prodrugs to Enhance Transdermal Delivery of Morphine. Int J. Pharm 1991, 71, 105 116. 48. Hansen, L. B.; Fullerton, A.; Christrup, L.; Bundgaard, H. Enhanced Transdermal Delivery of Ketobemidone with Prodrugs. Int. J. Pharm. 1992, 84, 253260. 49. Stinchcomb, A. L.; Swaan, P. W.; Ekabo, O.; Harris, K. K.; Brow e, J.; Hammell, D. C.; Cooperman, T. A.; Pearsall, M. Straight Chain Naltrexone Ester Prodrugs: Diffusion and Concurrent Esterase Biotransformation in Human Skin. J. Pharm Sci 2002, 91, 25718. 50. Vaddi, H. K.; Hamad, M., O.; C hen, J.; Banks, S. L.; Cro oks, P. A.; Stinchcomb, A. Human Skin Permeation of BranchedChain 3O Alkyl Ester and Carbonate Prodrugs of Naltrexone. Pharm. Res. 2005, 22, 758765. 51. Pillai, O.; Hamad, M. O.; Crooks, P. A.; Stinchcomb, A. L. Physicochemical Evaluation, in Vitro Human Skin Diffusion, and Concurrent Biotransformation of 3O Alkyl Carbonate Prodrugs of Naltrexone. Pharm Res 2004, 21, 114652.

PAGE 113

113 52. Vaddi, H. K.; Banks, S. L.; Chen, J.; Hammell, D. C.; Crooks, P. A.; Stinchcomb, A. L. Human Skin Permeation of 3O Alkyl C arbamate Prodrugs of Naltrexone. J. Pharm Sci 2008. 53. Wasdo, S. C.; Sloan, K. B. Topical Delivery of a Model Phenolic Drug: Alkyloxycarbonyl Prodrugs of Acetaminophen. Pharm Res 2004, 21, 940 6. 54. Milosovich, S. M.; Hussain, A. A.; Hussain, M.; Di ttert, L. The Utilization of Prodrugs to Enhance Transdermal Absorption of Testosterone, Deoxycorticosterone & Indomethacin. Prog Clin Biol Res 1989, 292, 2737. 55. Milosovich, S. H., A.; Dittert, L.; Aungst, B.; Hussain, M. Testosteronyl 4 Dimethylaminob utyrate Hcl: A Prodrug with Improved Skin Penetration Rate. J. Pharm. Sci. 1993, 82, 2278. 56. Hay, R. W.; Porter, L. J.; Morris, P. J. The Basic Hydrolysis of Amino Acid Esters. Aust. J. Chem. 1966, 19, 11971205. 57. Kirby, A. J.; Lloyd, G. J. Intramol ecular General Bas e Catalysis in the Hydrolysis of 3 DimethylAminopropionates. J. Chem. Soc. PerkinII 1976 1748 1752. 58. Ostacolo, C. M. F.; Laneri, S.; Sacchi, A.; Nicoli, S.; Padula, C.; Santi, P. Alpha Tocopherol ProVitamins: Synthesis, Hydrolysis and Accumulation in Rabbit Ear Skin. J. Control Release 2004, 2004, 40313. 59. Thomas, J., D.; Sloan, K., B. In Vitro Evaluation of Alkylcarbonyloxymethyl (ACOM) Ethers as Novel Prodrugs of Ph enols for Topical Delivery: ACOM Prodrugs of Acetaminophen. Int. J. Pharm. 2008, 346, 80 88. 60. Thomas, J. D.; Sloan, K. B. Evaluation of Alkyloxycarbonyloxymethyl (A OCOM ) Ethers as Novel Prodrugs of Phenols for Topical Delivery: AOCOM Prodrugs of Acetaminophen. Int. J. Pharm 2009, 371, 2532. 61. Majumdar, S.; Sloan, K. B. Synthesis and Topical Delivery of N Alkyl N Alkyloxycarbonylaminomethyl Prodrugs of a Model Phenolic Drug: Acetaminophen. Int. J. Pharm 2007, 337, 4855. 62. Sloan, K. B.; Wasdo, S. Designing for Topical Delivery: Prodrugs Can Make the Difference. Med Res Rev 2003, 23, 763 93. 63. Beall, H. D.; Sloan, K. B. Topical Delivery of 5 Fluorouraci l (5 FU) by 3 Alkylcarbonyl 5 FU Prodrugs. Int J. Pharm 2001, 217, 12737. 64. Roberts, W. J.; Sloan, K. B. Topical Delivery of 5Fluorouracil (5F U ) by 3 A lkylcarbonyloxymethyl 5 FU Prodrugs. J. Pha rm. Sci 2003, 92, 1028 36. 65. Beall, H. D.; Sloan, K. B. Topical Delivery of 5 Fluorouracil (5Fu) by 1,3Bisalkylcarbonyl 5 F U Prodrugs. Int J. Pharm 2002, 231, 439.

PAGE 114

114 66. Johnson, K.; Amidon, G. L.; Pogany, S. Solution Kinetics of a Water Soluble Hydrocortisone Prodrug: Hydrocortisone21Lysinate. J. Pharm Sci 1985, 74, 879. 67. Kovach, I. M.; Pitman, I. H.; Higuchi, T. Amino Acid Esters of Phenols as Prodrugs: Synthesis and Stability of Glycine, Beta Asp artic Acid and Alpha Aspartic Acid Esters of p Acetamidophenol. J. Pharm Sci 1981, 70, 881885. 68. Bruice, T. C.; Benckovic, S. J. Bioorganic Mechanisms Benjamin New York: New York, 1965; Vol. 1. 69. Gould, E. S. Mechanisms and Structures in Organic Chemistry. Rinehart and Winston Inc.: 1959. 70. Bergstrom, C. A.; Luthman, K.; Arturss on, P. Accuracy of Calculated pH Dependent Aqueous Drug Solubility. Eur J. Pharm Sci 2004, 22, 387 98. 71. Gregory, M. D.; Christian, S. D.; Affsprung, H. E. The Hydrati on of Amines in Organic Solvents. J. Phys Chem 1967, 71, 22832289. 72. Mohr, S. C.; Wilk, W. D.; Barrow, G. M. The Association of Water with Bases and Anions in an Inert Solvent. J Am Chem Soc 1965, 87, 30483052. 73. Okouchi, S.; Moto, T.; Ishihara, Y.; Numajiri, H.; Uedaira, H. Hydration of Amines, Diamines, Poly amines and Amides Studied by NMR J. Chem Soc. Faraday Trans. 1996, 92, 18531857. 74. Bean, R. C.; Shepherd, W. C.; Chan, H. Permeability of Lipid Bilayer Membranes to Organic Solutes. J. Gen Physiol 1968, 52, 495 508. 75. Thomas, J. D.; Majumdar, S.; Sloan, K. B. Soft Alkyl Ether Prodrugs of a Model Phenolic Drug: The Effect of Incorporation of Ethyleneoxy Groups on Transdermal Delivery. Molecules 2009, 14, 423145. 76. Roberts, W. J.; Sloan, K. B. Topical Delivery of 5Fluorouracil (5Fu) by 3Alkylca rbonyloxymethyl 5 FU Prodrugs. J. Pharm Sci 2003, 92, 1028 36. 77. Ackaert, O. W.; Graan, J. D.; Capanconi, R.; Pasqua, D. O. E.; Dijkstra, D.; Westerink, B. H.; Danhof, M.; Bouwstra, J. A. The In Vitro and InVivo Evaluation of New Synthsized Prodrugs of 5OH DPAT for Iontophoretic Delivery. J. Controlled Release 2010, 144, 296305 78. Aboofazeli, R.; Zia, H.; Needham, T. E. Transdermal Delivery of Nicardipine: An Approach to InVitr o Permeation Enhancement. Drug Delivery 2002, 9, 239247. 79. Ren, C. e t a l Transdermal Delivery of Tolterodine by O Acylmenthol: In Vitro/in Vivo Correlation. Int. J. Pharm 2009, 374, 7381.

PAGE 115

115 80. Volpicelli, J. R.; Alterman, A. I.; Hayashida, M.; O'Brie n, C. P. Naltrexone in the Treatment of Alcohol Dependence. Arch Gen Psychiatry 1992, 49, 87680. 81. Terenius, L. Rational Treatment of Addiction. Curr. Opin Chem Biol. 1998, 2, 5417. 82. Krystal, J. H.; Cramer, J. A.; Krol, W. F.; Kirk, G. F.; Rosen heck, R. A. Naltrexone in the Treatment of Alcohol Dependence. N Engl J. Med 2001, 345, 17349. 83. Generics, P. D. R. Medical Economics Montvale,New Jersey, 1996. 84. Sullivan, M. A.; Garawi, F.; Bisaga, A.; Comer, S. D.; Carpenter, K.; Raby, W. N.; A nen, S. J.; Brooks, A. C.; Jiang, H.; Akerele, E.; Nunes, E. V. Management of Relapse in Naltrexone Maintenance for Heroin Dependence. Drug Alcohol Depend. 2007, 91, 289 92. 85. Benoiton, L.; Purdie, J. E. Piperazinedione Formation from Esters of Dipeptides Containing Glycine, Alanine and Sacrosine: The Kinetics in Aqueous Solution. JCS Perkin II 1973, 18451852. 86. Rydon, H. N.; Smith, P. W. G. Polypeptides. Part IV The Self Condensation of Esters of Some Peptides of Glycine and Proline. J. Chem. Soc. 19 56 3642. 87. Kaufman, J. J.; Semo, N. M.; Koski, W. S. Microelectric Titration Measurement of the pK a's and Partition and Drug Distribution Coefficients of Narcotics and N arcotic Antagonists and Their pH and Temperature Dependence. J Med Chem 1975, 18, 647655. 88. Smyth, H. D.; Becket, G.; Mehta, S. Effect of Permeation Enhancer Pretreatment on the Iontophoresis of Luteinizing Hormone Releasing Hormone (LHRH) T hrough Human Epidermal Membrane (HEM). J. Pharm Sci 2002, 91, 1296307. 89. Artusi, M.; Nic oli, S.; Colombo, P.; Bettini, R.; Sacchi, A.; Santi, P. Effect of Chemical Enhancers and Iontophoresis on Thiocolchicoside Permeation across Rabbit and Human Skin in Vitro. J. Pharm Sci 2004, 93, 24318. 90. Nolan, L. M.; Corish, J.; Corrigan, O. I.; Fi tzpatrick, D. Combined Effects of Iontophoretic and Chemical Enhancement on Drug Delivery. Transport across Human and Murine Skin. Int J. Pharm 2007, 341, 11424.

PAGE 116

116 BIOGRAPHICAL SKETCH Hemamalini Devarajan Ketha was born in India in 1982 to R aghavan De varajan and Kalyani Devarajan Her family moved to New Delhi, India in the same year where she was raised. She finished high school in April, 2000. The same year she enrolled in the Bachelor of Science degree programme with a chemistry major at Hansraj Col lege at University of Delhi. She graduated from college in 2003 after which she pursued a Master of Science degree in organic chemistry from 2003 to 2005. In the year 2006 she started working on her PhD in Pharmaceutical Sciences Medicinal Chemistry at t he University of Florid a where she met her husband Siva in the spring of 2007. Hema and Siva married in 2009.