<%BANNER%>

Improving the Topical Delivery of Phenol-Containing Drugs: An Alkylcarbonyloxymethyl and Alkyloxycarbonyloxymethyl Prodr...

xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101219_AAAAEY INGEST_TIME 2010-12-20T02:52:58Z PACKAGE UFE0015582_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 23388 DFID F20101219_AACYIC ORIGIN DEPOSITOR PATH thomas_j_Page_006.QC.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
c16feffdc619ad6a81b92833f9fd4d23
SHA-1
4f423b0cbe68f3d6e1ab4c7f93f6e1e33d52ca82
25387 F20101219_AACYHO thomas_j_Page_123.QC.jpg
512c6b49e74172b618f4be0e9c70dd46
47a8767c73c57e9726a53c5cb9d4c11c65357a29
1053954 F20101219_AACXEM thomas_j_Page_121.tif
150d5ac55cf739b6a204dc36d38dedf8
09c161f4e1c415e2c469b9c67fdabd66b66b27a4
35018 F20101219_AACXDY thomas_j_Page_064.QC.jpg
2ff3442fba7acb9c649b8be182e47d89
bc882da90528df60feb4d9b0bd446e9705e95740
33476 F20101219_AACXFA thomas_j_Page_016.QC.jpg
ac2665e10baec9647b82415d4b71f16a
95ffe4cea6655fa999ab7d4586ed3bca6e7818fd
34492 F20101219_AACYID thomas_j_Page_007.QC.jpg
2ae22762dac810b4ee110d79765f0568
60dbf4326a8ac1ead9699c5e9c11054d83d404ad
34237 F20101219_AACYHP thomas_j_Page_060.QC.jpg
23d928a7c4c931c3fa189463c89854a2
b66ab3d87b564cf76292a953af228e6534abc97e
111405 F20101219_AACXEN thomas_j_Page_064.jp2
f6fabf0292f6089928811e063396a68e
7d04aa087e3f597bc0cf74cdcff1c03ae3848ac2
104752 F20101219_AACXDZ thomas_j_Page_020.jp2
6e7256460964d8001aaabe9e1e473dde
e523e5ff664b50a85cd497a06346c3a7729c6f11
8450 F20101219_AACXFB thomas_j_Page_118thm.jpg
2b1caa7444727ea6457fcd687dbc0c30
8e02d3bdfdfef621dd886945c73653667f7b7e37
8414 F20101219_AACYIE thomas_j_Page_007thm.jpg
36e385461a2c5bfb915a1a0f5b0d1cba
420385c8ab571e4479d167a1c729b55c64ac5357
6912 F20101219_AACYHQ thomas_j_Page_071thm.jpg
30134e2c36af071dde62a167bbbf323a
0de4aee291fdcee0a03cd69f85ea3d61bf50ae69
7693 F20101219_AACXEO thomas_j_Page_031thm.jpg
1a5adaeae418cff6953972c26566a74e
3a56d7f40b2c3e39ba3d33afba7b3ab3671c0382
8330 F20101219_AACXFC thomas_j_Page_065thm.jpg
0036bda740fcc94668f25decba9d9b7b
6e486814eafa187840176f1efb4670bc3fa674c1
9510 F20101219_AACYIF thomas_j_Page_008.QC.jpg
cff21a314419e76c743ab4d91b2ce952
a36227301c023f189f3a26ba0745a9c8612957cf
8685 F20101219_AACYHR thomas_j_Page_037thm.jpg
175a93b68ac4c94f8b37e673373d513d
7d5759c15a19f48888eb317499488655bca42179
104489 F20101219_AACXEP thomas_j_Page_064.jpg
d8cea23c53dd3ed2662320ea43e41805
3079f7cc8b38b7b23926cf0ac52361931c62f917
20547 F20101219_AACWZJ thomas_j_Page_100.QC.jpg
3efb678a8406ae9781eda2d04a735406
3cf5d544e94a86c3f834267edf835de7d5b0374a
34556 F20101219_AACXFD thomas_j_Page_120.QC.jpg
0d426e520dc56dfa2e94b9ed6024822c
6eba11fa7f0137aeb1d9f9c545d961d61c2d052b
32204 F20101219_AACYIG thomas_j_Page_009.QC.jpg
6f6c6eba1e16ee122944c0dc2aa238e9
3c94d42078aac1f8af5f12f52fc5b321b877c427
7847 F20101219_AACYHS thomas_j_Page_138thm.jpg
3ddb1229c5789862b64c69e10ea5218d
fb8bb9bd6992167d52d91d290d53c9aa4f6d50ee
26395 F20101219_AACXEQ thomas_j_Page_012.QC.jpg
54464f4866b928f2fe111858b21363c2
17ffaea6015661fc043fe6f3e326b8270e7772aa
95715 F20101219_AACWZK thomas_j_Page_095.jpg
7bec477461d0ac73300eb9ec13a9c3c1
5e441bcb11de23c160f2cb237233afdc29f58771
93812 F20101219_AACXFE thomas_j_Page_035.jpg
1ebf09870c69a4e36319a2bbb602c933
ad0833f45a2d0378cfb759ba258b35d035bc8f4c
7716 F20101219_AACYIH thomas_j_Page_009thm.jpg
6e9f3e7e75df538e4025839336df7ac9
2d77b226ff4d13a0562ae797be078af766f1c325
8393 F20101219_AACYHT thomas_j_Page_053thm.jpg
bd2e987f3ba08b15f90df256e927ce6c
c110b3b9d53e9514a87b7612d74d6a60e85ddf81
98907 F20101219_AACXER thomas_j_Page_063.jpg
e3c654fb9d35243605650e4d39649491
a92069240abafa6fd98ef0ebdd00fcd06631442e
1823 F20101219_AACWZL thomas_j_Page_148.txt
cde106cb01ce6dc2040078ec10ba3df3
ff45d862da1148b3ec9841e41ae0e30b0380e930
875111 F20101219_AACXFF thomas_j_Page_059.jp2
084765aea69558bb774a8449be9464ab
c4982a7773985bced487579f00df7a0b7940e570
9801 F20101219_AACYII thomas_j_Page_010thm.jpg
c32e7f1ac426f5e6832e0f510f4c3b46
0b2b0a095eb9e379662b068bd641a9fa138d74f5
8009 F20101219_AACYHU thomas_j_Page_105thm.jpg
cd3a487c73d919f0422d8590ea35146b
70e657ef64dac583c4955272bf49e3d70fe367fc
50696 F20101219_AACXES thomas_j_Page_136.pro
8f16cd51ee239643ad02370d1c25860c
5bf471351c036cf5a721f0a2293a9dd861d7fe12
2053 F20101219_AACWZM thomas_j_Page_026.txt
342a831fd6c8960176aa0cc9e10862c4
3f4effa825286ab6bc923d7da5e017614d8d8e6c
98637 F20101219_AACXFG thomas_j_Page_046.jpg
72098418b513bbfb83d1e56816146909
64af5c905015a1e048af7f509111fd6c8f2887c8
5727 F20101219_AACYIJ thomas_j_Page_013thm.jpg
20ba05e2a96a01c207fb9c5956a54638
da426ac0073a6f3dc153c54c2247b1665d9de95d
6810 F20101219_AACYHV thomas_j_Page_048.QC.jpg
393d4319b8ea8d060bb1c00ae145a757
3cd8d9321bbf6b98ca399d39cfc22723f28d92a5
35742 F20101219_AACXET thomas_j_Page_054.QC.jpg
4ef1716330c44dce1dc1b53bb3a3577a
efc1d23c6012cf2f5fa96f40e38dd0e261fd554c
8417 F20101219_AACWZN thomas_j_Page_062thm.jpg
069459cb13b691b9b5560951e9416704
f3828b691936982c56c2cd00f0301603d179904a
117198 F20101219_AACXFH thomas_j_Page_082.jp2
b87284ad74be12b97b7200e9f39259f1
249d1761d28a37858021f54d3d3a4a7dc47a9864
33361 F20101219_AACYIK thomas_j_Page_015.QC.jpg
3e4babe088d687c1d153c0bafc89e43d
63dce1b05eb3b6463618d87b1af25dee37be24c1
222924 F20101219_AACYHW UFE0015582_00001.xml FULL
110a1bd13f9463712a4bd055d38ef878
35a4674e4a5e0b58744781ac49c0c7ab4f25410d
BROKEN_LINK
thomas_j_Page_001.tif
F20101219_AACXEU thomas_j_Page_132.tif
1b27371fee160e72c03ae70d6b040833
9f2054c5edff0c25d3b7e7ab69106e9fc70df7fc
49354 F20101219_AACWZO thomas_j_Page_021.pro
2ebde9527b3a2e675edf3a3211a248a9
01a35e064ee381e37a245c9640379f0eb2ca3361
F20101219_AACXFI thomas_j_Page_116.tif
afdd0843ce23d67072fcb13c7088031a
462380da4c285c64d028babf01a38a1b0e06350d
8359 F20101219_AACYIL thomas_j_Page_015thm.jpg
9eb495d8449eac173f0c8f5fa4744709
d9893842529fd46d68e1c7e7effaed910a165499
1714 F20101219_AACYHX thomas_j_Page_002.QC.jpg
0c94248c3e5c4dc06e07e1ae051b58d1
b21a88f66e57265d2f3bd25705cb81dbeb9eeb77
107358 F20101219_AACXEV thomas_j_Page_083.jp2
4a1a0b1243d84a67f84eb231b1c3ae37
1ddb4544960ea9cb48c2c717df43a08305427b51
8737 F20101219_AACWZP thomas_j_Page_068thm.jpg
04b31e03c565c02982d5f420408ce866
6ac4e20f6e423eeb2e61393b532d485d30328440
1894 F20101219_AACXFJ thomas_j_Page_131.txt
0a0c28e9ae2e97de845539c6d4f0113b
3fa555568ab3af19c37a2eba04976128ffc9a6aa
31622 F20101219_AACYJA thomas_j_Page_030.QC.jpg
8b1004786d68dd43f26f988768204f15
067d532acc240911f6a5f85fcf6378ba7365a5c5
33100 F20101219_AACYIM thomas_j_Page_017.QC.jpg
47280c582aff0f3f735d26b8137a62ee
8fefb27892a0b1b23dab91f1046b4900dcdd4a5b
660 F20101219_AACYHY thomas_j_Page_002thm.jpg
2ade65b923ff1e463d179b0488270d04
e7310d57fa85c35b31f0cebb5699237be16439ab
6826 F20101219_AACXEW thomas_j_Page_012thm.jpg
ef584366b73df0b8762749ccaaaaa149
46b8e4c39a2b8b7b69f890fe7d71aef0e99007ce
8013 F20101219_AACWZQ thomas_j_Page_124thm.jpg
2c1b7f519ca2a372a6e2ffcf669ead53
75775827ed4c52705446be22346ef239ac3ae573
88509 F20101219_AACXFK thomas_j_Page_005.jpg
c44274316cb29dfead7f567cfda455a9
35cc040ffee2d7a01a19f0e58eb9999c8c9909c9
7286 F20101219_AACYJB thomas_j_Page_032thm.jpg
70e22783f4a9faeab4f4ebec533e9027
caa92accf9e193b1874df6d68cc6d81d8339c95b
33635 F20101219_AACYIN thomas_j_Page_018.QC.jpg
7eed58fa0ec3e418450c51de29be5677
b5a6ddb1b1ad8a18a4dc6fe258d5d41b16c120ce
2274 F20101219_AACYHZ thomas_j_Page_003.QC.jpg
f2c4732f1a30bc3092587a0e3ff006ef
0fbcdd1bb1ace5567146b974e99bd97fd373a7c2
95628 F20101219_AACXEX thomas_j_Page_024.jp2
431496f81b4d11950e35fe3c32ca13d3
85d61d356ca9b9ad0c1fe3f15755cc0784656f34
F20101219_AACWZR thomas_j_Page_013.tif
935077d0303ded9edc4b2a5525fdfc97
65c8d194c5d5c51cdbedeba4c03d685d6436af11
107825 F20101219_AACXFL thomas_j_Page_138.jpg
3ee2a69956b0452f4f37a356f8ed7477
cb81a4ead16216e1981c12d7ee7ebc635bafe704
7175 F20101219_AACYJC thomas_j_Page_036thm.jpg
64f3b27c72887dab322627b47ef791b6
254e9bbf886771e745b76268c52980131f28c50d
33616 F20101219_AACYIO thomas_j_Page_019.QC.jpg
885f0c6fc8c64dede8692595a3744bb0
acaaa7630c63cb5eba1961b199b33dedacf7b0a5
1051986 F20101219_AACXEY thomas_j_Page_011.jp2
42b91cb9e03cc22829232ed86bc74225
0d6824c9eef679343ced3d497acada520522a1d2
112661 F20101219_AACXGA thomas_j_Page_133.jp2
9ccfb7657d6b483552c01de1de4bbdb2
0d39a81c0fe4e4cd6a9b24acc53a4ab25caa1019
28009 F20101219_AACXFM thomas_j_Page_122.QC.jpg
6e67ed3e6b45b5ee5edf8f03c10b1fe3
aeeb644bb2450cc0d7ae6e75d829d1b7006108b1
34927 F20101219_AACYJD thomas_j_Page_037.QC.jpg
fcfd743f991f0734e83da395719c9a14
5cfdec60fc713aa4e086fe60379bde34b78b6edb
8353 F20101219_AACYIP thomas_j_Page_019thm.jpg
93733b34584d788dbc92b3489e063003
7f419110a33c6232b1a3b111bc966525f930c372
F20101219_AACXEZ thomas_j_Page_109.tif
80372330baf5bdccd7fa19bc68193ca5
100f5c681364ae9310a06de51ded87485118c4b6
1944 F20101219_AACWZS thomas_j_Page_021.txt
9a539094b39998186f37f803193b2a39
a449b2605fc1c14c35e9ffd10577a13a159be755
7556 F20101219_AACXGB thomas_j_Page_087thm.jpg
9c7a7a2f339cd19b65e2565deb91ca0d
4fd778e5c37e1683984553a3b71e3e382eb09793
21318 F20101219_AACXFN thomas_j_Page_025.QC.jpg
3834f12abd27b233c76d90760cc62fc3
1a8ecfd0970c1cf2ac6ceed909c4d05cb980559b
26912 F20101219_AACYJE thomas_j_Page_038.QC.jpg
c86912a561c5568a99dd5244fea71e54
1f336b3c19989face7d7b8bda3770bcd91c93c63
8267 F20101219_AACYIQ thomas_j_Page_021thm.jpg
68cf9fff9a0b1aead5dee19b21bbb254
a10f90d471d160571eebec3816c5d46201b536df
51370 F20101219_AACWZT thomas_j_Page_056.pro
a6f0e22b5a02d0b1b90ad88a5fe53d0c
dacb84fa2971be6009936161fcc437f58a018028
32044 F20101219_AACXGC thomas_j_Page_119.QC.jpg
19803147d1b80210eacea9f9479896d6
dd5ccf1bd3c2533a7a90be71b53a36fda6a722ee
8143 F20101219_AACXFO thomas_j_Page_095thm.jpg
972abf1cf1c414f0f1a50f7de206b147
23498ae541dc089939a9eab12ff2764eb6a14583
7148 F20101219_AACYJF thomas_j_Page_038thm.jpg
6ba2a8f0cf91b52730fb6aaf9d127354
7bf91c8d24feb08750ab076755d16d07beec0032
31050 F20101219_AACYIR thomas_j_Page_022.QC.jpg
04d68acb92e55cd2ada51e65e7c10566
5edc387fa2a9c017caa6572dbe434ffe78159bb5
25271604 F20101219_AACWZU thomas_j_Page_011.tif
6df6ab3c745985319258efa48f7fc91f
7bfd2ab36d88814fbae57eb745a9810e2d32bc26
F20101219_AACXGD thomas_j_Page_137.tif
dab698c79627123cfac28fc326bd3789
0bfcf86f65f5004b41f9a6fdccaff5e93780b641
9203 F20101219_AACXFP thomas_j_Page_085thm.jpg
a471c6b238579e58fcfbfd7122aa2d82
82d97c015d34c13201f95410b534c5387c382025
6327 F20101219_AACYJG thomas_j_Page_039thm.jpg
ff1d485e9557a3c02bbca82167734907
627d33b5ae093ba60365107733a4f15c02b394b9
7548 F20101219_AACYIS thomas_j_Page_023thm.jpg
e79c9858427667fe92dbd51af82a8fa9
0fe4e83c2a7d9bb57f0ce4e3999cc235fd83428f
39212 F20101219_AACWZV thomas_j_Page_127.pro
5bca774a99c1f1ff3d82faffeb485309
7ae31d6ab0b3fdd3343589e2abea4e61b44688d5
73295 F20101219_AACXGE thomas_j_Page_076.jpg
be62c3bf58866fc218c1daf7205e0dee
1b84c9ca569cadc6066c48f5b4941e5ce408a25e
69573 F20101219_AACXFQ thomas_j_Page_092.jpg
00bf2b56b42af6615cdd57aee8334997
c15a036e5d0c4c9d17bf2d4369d4d37f84c26fc6
8088 F20101219_AACYJH thomas_j_Page_040thm.jpg
c2998e94025d1f353591e41a0d2da964
3c40b0686ccacc66e9fceca64a756548012befd2
7423 F20101219_AACYIT thomas_j_Page_024thm.jpg
a095fb6e3da81b460ba6f095604880e3
1b88ee0ff3b635f4ebeb391f5dc3baf7b35a6757
7530 F20101219_AACWZW thomas_j_Page_033thm.jpg
4c04851825b8ddc632dcfbc86359535d
f3406c538afc3e9ff79aae84c1cf05de45681e03
2541 F20101219_AACXGF thomas_j_Page_142.txt
5f734598363d5f1c1d2142e54005c237
402e1c7948c846d44e3d4a2e5933de187f641e61
44511 F20101219_AACXFR thomas_j_Page_057.pro
11fa6871f417bd8dd9029e6013d4f5ed
79e85ca6fb91df912bd9e7e0cd051d1a135ceede
26143 F20101219_AACYJI thomas_j_Page_041.QC.jpg
e3d36cdb345de48fc3596e309e6db131
8b7a48a6467ca4c3e2c41941d914203f4e7910f6
6386 F20101219_AACYIU thomas_j_Page_025thm.jpg
08e92e8d8885ae2a042af605d8b2c88b
6e3e0e136ad9a43118cb29d7cca1fb5b7529dbf3
76393 F20101219_AACWZX thomas_j_Page_061.jpg
cee91573f9cc2d6ebda1f403956788fc
31e4bbae7a4092d373be3e5e26ce1b1b1364df71
1302 F20101219_AACXGG thomas_j_Page_002.pro
f380bc5c88366d0e652eb8debb64c3ff
4879937d1e46472a69bc779260cf1e6700b4952e
86466 F20101219_AACXFS thomas_j_Page_127.jpg
0d2a33fc9a820027f0eaf25d4df34b55
32115c0f004055e0038972e8dbe946001cee4567
6688 F20101219_AACYJJ thomas_j_Page_041thm.jpg
96823d250b64478f7d162dde76a540b2
546ba4bf4e74c24bd41cd319599e7d1a71fe7a09
31598 F20101219_AACYIV thomas_j_Page_026.QC.jpg
860af0f1885a95b3cb2c4419eddcb639
386beaec42a31e430691eda46adf75bd0a177c29
1851 F20101219_AACWZY thomas_j_Page_035.txt
13a2a70c0986775d60ae5fd8532ec164
7e253cd79b06d5a13334e5f121863cc64c9dde0d
785 F20101219_AACXGH thomas_j_Page_025.txt
6b4aa0bffd8a4e13544fcb29ca800d9e
b4141cab0b0fd83559031a7ebb2a149b97c77002
F20101219_AACXFT thomas_j_Page_005.tif
65608e490773db03fbbf8badb66eef46
dc8dd0935bf466cfca36c3bc1cd43ae4364c662f
24856 F20101219_AACYJK thomas_j_Page_042.QC.jpg
aff2b5f23b8d6b5a73767b3b79175417
9ebd893f5695db53c7451096aae86cdebb414037
8511 F20101219_AACYIW thomas_j_Page_027thm.jpg
65c6498de413604e6489051d9e103cbc
a6ce74c825a786a6e6bbb951f2aa691767247bb6
1975 F20101219_AACWZZ thomas_j_Page_083.txt
8c82f5102ce663928eef72cff837b493
0d2f49ff2d21c57478a8b1c98a719de2442b250f
7968 F20101219_AACXGI thomas_j_Page_016thm.jpg
6902e316f423f1143cb471ff3b43fea3
b806dc4b786cca6501c9c6f0452f18fd905ef5a1
5621 F20101219_AACXFU thomas_j_Page_006thm.jpg
ca2191e6bc224b4842f94462324905ab
e66816ff948c6d51e8334dbaa3bb9395c48cf5ad
6574 F20101219_AACYKA thomas_j_Page_061thm.jpg
0d14c5f04ff0eff34b9b6e4824746a80
d03773bdbbff3ed9f14621cac9c8189931c2e00e
26831 F20101219_AACYJL thomas_j_Page_043.QC.jpg
932b6fa5464969cba00ef2f49d6e3937
9dcd6f8bc1e6bd82454c79988a89a58f714c7ab6
F20101219_AACYIX thomas_j_Page_028.QC.jpg
1b41cbc1125b3374d68ada2e321b5b60
dc2b32935ba4c5cb9a64bcb8983bd1519eb3618c
96472 F20101219_AACXGJ thomas_j_Page_062.jpg
cd14b8b087ab2b7c438155e12003eff8
d988ac949bb7d06f229355113ee1ddffd1ca65d7
F20101219_AACXFV thomas_j_Page_056.tif
4c7cd3a59e2bae3ed34995acfda16c04
5135cb5d753de2cbe454345e824d7975723cb556
34249 F20101219_AACYJM thomas_j_Page_045.QC.jpg
b696a421f6006833126de350c6ac3a9a
e5b4cccdff5f41b8e7607a935d56d3b3c2b3112b
8331 F20101219_AACYIY thomas_j_Page_028thm.jpg
052f6114bf560a3bed3db9d818a625e3
6c23bbdc0850583fdd87b4d7f22144fc1ecc36f6
1998 F20101219_AACXGK thomas_j_Page_110.txt
23a3e7cbe8b6642ec1616935728be727
150c6e858b42d60f797d7b5e60941ddc92820287
8584 F20101219_AACXFW thomas_j_Page_080thm.jpg
348317220eb8db681e1cd256399d9c9e
a3e26c2177957cd78190afdd428c2639367f1964
31917 F20101219_AACYKB thomas_j_Page_062.QC.jpg
71de61df740268f4c2f4c853e6903603
9ccb4d901699dc6a91958bccd50e266a352cdafd
8531 F20101219_AACYJN thomas_j_Page_045thm.jpg
a152601da478960a3cb0e67c95e87af1
3bad963e6713e8b3fb26c83047b11a97cecb5d97
8494 F20101219_AACYIZ thomas_j_Page_029thm.jpg
9842a47713112199931c432058a44f10
a32b334c2addf6c4f0ad264a9bf1d2f2f307837c
109702 F20101219_AACXGL thomas_j_Page_080.jpg
e61c473b2f4ba52044ce1127aa9eecff
edc0122908d479a71f4114ab6b3da22b33c85644
1570 F20101219_AACXFX thomas_j_Page_078.txt
480778671b19861cd4d7ab65eb6f2a23
a334926e3610fce94bcda5a2696e6c66abf7fa69
32126 F20101219_AACYKC thomas_j_Page_063.QC.jpg
80bab06babb3b0cab2345a0e437c8a92
c7d072e73da6e007fc0b0693c76526de5d687a00
8099 F20101219_AACYJO thomas_j_Page_046thm.jpg
199d2f879d0f53f397c669aa4b248ab0
223325c2f06c3b07fc99f0fad0473859ab651560
2550 F20101219_AACXHA thomas_j_Page_145.txt
834426c0e09c6020b820f3ac6354a8f4
ccd935cbc54657325b2faf2b7342af04e346bfb6
F20101219_AACXGM thomas_j_Page_068.tif
3802684172cbc40f30e53758e03e3414
1c7712ef4a80b970fef55d00ca8d688d03225c5f
F20101219_AACXFY thomas_j_Page_092.tif
4019725ff53682ccf22c30168a1ff0fa
e42e352383fa15d9cb81012f5ad136174c08d58b
8552 F20101219_AACYKD thomas_j_Page_066thm.jpg
b1730e3bc08587d957020a45161c6114
f28a644b782251456332c63540ae8fbdd7e1a9f9
32167 F20101219_AACYJP thomas_j_Page_049.QC.jpg
3288f65388e56db328282f256e811de0
a087e44c746504354f36805bfa832195251fde55
97737 F20101219_AACXHB thomas_j_Page_105.jp2
f74ac33996d18d674b50460f7ddf5ad8
a667c2875ac34dd52e316e8f9441e97aa1f9d908
37833 F20101219_AACXGN thomas_j_Page_106.QC.jpg
23e014bd8041cd82244e54841fd1c227
b487b662028470da9424b2972de5573031c994e5
96846 F20101219_AACXFZ thomas_j_Page_031.jp2
0f65e0c116b101fc82816f456c323006
c6aeb2ff8ca9637a4409ad5834cc275dfbc07631
33124 F20101219_AACYKE thomas_j_Page_068.QC.jpg
6290540b5fca02e2b0b176eb9a429668
dcdf7a4bdc25baceef41e4de99058683025470d8
7921 F20101219_AACYJQ thomas_j_Page_049thm.jpg
ab533f3d9f9d2388f119cc24c628032d
3c588c152a633d539590abc74ce98e32c896df25
76563 F20101219_AACXHC thomas_j_Page_107.jpg
87e59ba34eb61e35f1d691bc458d4ddd
ced162cd0653c36472e777c897951a879a987fa0
32467 F20101219_AACXGO thomas_j_Page_001.jpg
cab46aa36eaedd47cf7208168a79f8c4
1d26ac71b40254a38d017b21be021021f6ef1e31
34573 F20101219_AACYKF thomas_j_Page_069.QC.jpg
60ff802053951571c2d63cc665d7ee42
e911427eef35c5e65dacbeac780f5c9f64d04a12
33512 F20101219_AACYJR thomas_j_Page_051.QC.jpg
ccd36e4f0b5566d317c50004975c08e0
d56edfe7613002c7d66d64b7d8ab272c86ecfaf1
37260 F20101219_AACXHD thomas_j_Page_147.QC.jpg
9cee53e9d854bbaa6a3976221ab7709f
da522915f3931223427bf3b1b5b4469cb701ef1d
105772 F20101219_AACXGP thomas_j_Page_121.jpg
06458d498ca99a47007258769ec4d3a2
7c46ca22d9b816b7453161ea7f1af5609391d8bd
8567 F20101219_AACYKG thomas_j_Page_070thm.jpg
d61ffa649fe771130252fe4fb48e8684
f7e14feea5f02d28a6a58fe883beff0bd58673ee
8639 F20101219_AACYJS thomas_j_Page_052thm.jpg
23e58585e7f404bb567119dd30649b93
6b73d35d2627316324cfef964bf7ff730c5cbc4b
102627 F20101219_AACXHE thomas_j_Page_050.jp2
a86457c4be7bb6910d53f2197e7abbbc
b54ec84b78b96a00a7b93ebad5537ff7969c4ee5
1740 F20101219_AACXGQ thomas_j_Page_071.txt
927a388cc4d253bc26d43c53995ac79b
8c91e61acfcf9bc980dd4e64be0a48dda7d4e568
31386 F20101219_AACYKH thomas_j_Page_072.QC.jpg
c3abbdacb5043a6284a09c59210ef820
56e5020be9c49fd2655e558b2da6d2035494b61e
34218 F20101219_AACYJT thomas_j_Page_053.QC.jpg
ac6e2a255117e355531e11a750bfa3cf
c823efa7664f7f4455cdfef02e0b5e7da9f8866b
109314 F20101219_AACXHF thomas_j_Page_067.jp2
5a0d02cb6ece12e2af19c583c21b2aa4
2ae00529910cce1434dae3c4ab6310006be80f53
F20101219_AACXGR thomas_j_Page_008.tif
199e4b45e4d9c4e31531d570e6d480f8
aa634bbe341dc99bdaf01f1850c94c98659e088f
32454 F20101219_AACYKI thomas_j_Page_073.QC.jpg
166b9f340dfb0a2d4307935211d20f98
465d27bf902695fcf258c62267dc3b8e22c6144a
8449 F20101219_AACYJU thomas_j_Page_055thm.jpg
c6811b60f88d7b159906e6f93a05e23f
6a6aea0875c33978f707370b265b8cea1551038a
31222 F20101219_AACXHG thomas_j_Page_046.QC.jpg
62870902b6133ac7c9cbb7187531fb33
61c67be39f2d5f8fae754ea405996ef2119a199e
1482 F20101219_AACXGS thomas_j_Page_047.txt
acc2397943f6cdf686f9f35783da2460
abaca2010e71a19683352bca3afaf3e2026466d4
8023 F20101219_AACYKJ thomas_j_Page_074thm.jpg
75ddc615a5c4094fac77a531690fd593
9de253a21b4a4156701927fb4d2952ff8a9280d3
28919 F20101219_AACYJV thomas_j_Page_056.QC.jpg
555d47a061c8ecd99f8de916b749a793
4340a1a73e6eccc8b71870b9fcb5421fb28fa659
1886 F20101219_AACXHH thomas_j_Page_075.txt
ea80f908d8a77868489ae269036877de
3054861ad7f470b010284e8ebc3d7d8dc55272fb
47678 F20101219_AACXGT thomas_j_Page_046.pro
1f29b04a5c7eb916d95958f3b72e78fe
b561cb4461cfc629a3981a8267ba9d41307cf254
30258 F20101219_AACYKK thomas_j_Page_075.QC.jpg
02c40af60316dc305c8560c8bae3f890
9dc6cfc35613ef0e7fbd075f040117347966781e
8460 F20101219_AACYJW thomas_j_Page_056thm.jpg
502852234aaefeed76c1d2abcfa3f30b
ad4c4084ecf7745b5d31b408e9af5979a44e1cc0
F20101219_AACXHI thomas_j_Page_064.tif
21e830ff062a129def2c49937b44ee57
cc704b10605fa7153bfb4d25fbea12ec68a427d1
101017 F20101219_AACXGU thomas_j_Page_062.jp2
aa512654ac4bb3552be00dd64458d1a7
22a15b06cf9ea8520e8ac3466548149a260ae1f5
8500 F20101219_AACYLA thomas_j_Page_088thm.jpg
dfbd87f7441a61ed437e1ab232c1f1b3
6872de61420c944851b2600181950ae463f99067
7648 F20101219_AACYKL thomas_j_Page_075thm.jpg
3a93b4e81520025bdea5c9a852e58d4d
f917bc2afd46a45d10441ed3190f04cce48171fe
6698 F20101219_AACYJX thomas_j_Page_057thm.jpg
d2a7bb6d26fe6453ece46c2d47944db4
0adcb0357a9350510d4f87cef8b4ad05a9b2cf78
48801 F20101219_AACXHJ thomas_j_Page_118.pro
6c0bd78b8b2e4c8bb65dc07538714a0f
7f07cce8cff9135fe4948f97ac5678aba45368c0
33000 F20101219_AACXGV thomas_j_Page_050.QC.jpg
b34b770aa579ccaf9fce657722471331
9ffdb6237863039be704cfb21c68b1c96d0eea3a
7425 F20101219_AACYLB thomas_j_Page_089thm.jpg
373a30807681a6dbf258a710d696ade5
352ae2e8cc9331aa438c5bb206ad7b9fa0699933
7789 F20101219_AACYKM thomas_j_Page_077thm.jpg
012acbcc3fa2af5da59deeacadbb6b51
5cea2025054541196deb2d3b5d1ed7171a7e719a
33347 F20101219_AACYJY thomas_j_Page_058.QC.jpg
0972ea26d91d857b5667729a0f21c55a
ecfa0f1085fcb4fa8ca51f87c7f7bdf97e1695c8
24399 F20101219_AACXHK thomas_j_Page_094.QC.jpg
a6f64f336fe35417278a44da989bc317
ef513cb9a4498f6857fd3eb171438fa44c2d2045
104528 F20101219_AACXGW thomas_j_Page_066.jpg
d881da69e29d927eee31bbfbf9eb1f25
8fe99ae0bd49b00517072fadfcff0c047cd2f0b4
25689 F20101219_AACYKN thomas_j_Page_078.QC.jpg
4a7ff3a91d7688f3b7512abc77b99b22
0d2ae6e26c973b6873a000d473fdd8bca2a91fdc
24540 F20101219_AACYJZ thomas_j_Page_061.QC.jpg
c309b78f6223eb2994d749e45b002284
ad0fe3e981842c3fb81593a3b31509d5d2a6b442
2250 F20101219_AACXHL thomas_j_Page_138.txt
19873a4141c5c3437a66054f99451f1e
eb2145cfdef818af5c4163b185efe9c43b79c1b4
107856 F20101219_AACXGX thomas_j_Page_134.jp2
84dd56e73ab455f5c430e37cc98baf0a
9e03aab02743401dbc054a11ce012b4edf93c5b0
26959 F20101219_AACYLC thomas_j_Page_091.QC.jpg
716137da1cf782407b1ad613de13486f
4519f2de7bc33825da41d8353181189347f228b7
33605 F20101219_AACYKO thomas_j_Page_079.QC.jpg
b997791e4e49d12f5441e4a23e60b431
842287b5e59a235e661c4d305a78a50fd1f1c347
F20101219_AACXHM thomas_j_Page_007.tif
7e07876f5b6ee08a669f77953d1d134b
4e2e41c705db1ac8fa074eb8cf89073f39aa2af3
104069 F20101219_AACXGY thomas_j_Page_136.jpg
0d1fec292dee50dcb1feb6f070f38863
77adc49cbba8b5ec71b5fa189d35d84d3cd063b1
26795 F20101219_AACXIA thomas_j_Page_126.QC.jpg
d150658370689c696c1018fb2b5180f3
8a21b6cbd89c7804d52aa0245d4a7dd3be38fbe6
5868 F20101219_AACYLD thomas_j_Page_092thm.jpg
d53d4420f953e0d5c8cfd00da65e684f
b168d1378a4b32c7b67811d3ee832bccc8da6e25
34119 F20101219_AACYKP thomas_j_Page_080.QC.jpg
799ee1d7bddf2bcb24b47a088f30cac3
1de79509db3b6bf1e6dcb09ec2d217f7a6390551
100908 F20101219_AACXHN thomas_j_Page_022.jp2
0cb44cd77941530562a9567da3cc3c4e
9a90549ea42dba9c9793bf1e953493de91e14c40
50065 F20101219_AACXGZ thomas_j_Page_017.pro
d33d97c7f65883d2d849e14d6602f694
3be533ea50c6f945fdd0088cce1bebd37b7ddd91
49429 F20101219_AACXIB thomas_j_Page_101.pro
044a6eb21d9d9e89847806420680183b
89e27c6e78d4165ba63369458925eaa7ff84e53b
29949 F20101219_AACYLE thomas_j_Page_093.QC.jpg
16f83fd4485147a198e1dc2d81b33221
c585b229b6b7c5bfde94a83acd6654cfdbf2f5e6
28913 F20101219_AACYKQ thomas_j_Page_081.QC.jpg
b8b2d47d5b59f35b43a2632a1bc5e22d
1d6737f92c0e840041a3ae5e91204a651da55216
7696 F20101219_AACXHO thomas_j_Page_099thm.jpg
a8c4d5743d96e11b7b5cdeb2bb61e7f3
d87fc8c675e177cb7be199491fb0f211cff5b9ca
F20101219_AACXIC thomas_j_Page_025.tif
4c42ea9a708c34ffd62c56a09118b95d
90269de618b9c42cf57fc93adb4979deca994035
31404 F20101219_AACYLF thomas_j_Page_095.QC.jpg
a5dbb65285ed0f7cc37e829b24428b1c
ff868fa65f147a2b5b49c0b933f2a77a533f8bd1
7260 F20101219_AACYKR thomas_j_Page_081thm.jpg
dca7a81145aa202e6e0e0c9c6027c613
69fc7a566e756b803996dc6278bd254422806639
F20101219_AACXHP thomas_j_Page_063.tif
425a4213f1339f45f74bf295de24b52d
ef9fba78f54cb2efb693b698ef3ee2192b0bdec3
37237 F20101219_AACXID thomas_j_Page_091.pro
962156e2fc32fd1cc7cb136aa8ccffa7
ba8019491f86ec31931580079d7acb6b4f9f3a8a
7533 F20101219_AACYLG thomas_j_Page_096thm.jpg
c34f9f4d214b5e28620193596b346a98
2734af00d60ac6b28bf9a3b5ba4633127900b485
33526 F20101219_AACYKS thomas_j_Page_082.QC.jpg
fed7016720341ef161684729cf7cb4c5
606d55dab7c57f518bf81ac56d0539aa9c8ad2de
106952 F20101219_AACXHQ thomas_j_Page_017.jp2
d22a0bb79da311491d8fabfbc31348c2
80963e7c33a845e3fd3b0f24f93761293ee72795
30532 F20101219_AACXIE thomas_j_Page_013.pro
4830e1a67068cf2f41f8f67dcd7245e5
bf2655b91ff14086bc5ec0b2f4bdeb63f1ecb98f
34781 F20101219_AACYLH thomas_j_Page_097.QC.jpg
e7ba85f452e11920e59bfd91f7903156
521374ef84b0dce0b4fcc7c8dc069b801b7f84c0
8446 F20101219_AACYKT thomas_j_Page_083thm.jpg
d4550faa8ba88f2e0baaed951e4a7ad9
56046478d3a68a5280e75c94a137d22d8e33b44a
8812 F20101219_AACXHR thomas_j_Page_125thm.jpg
106c28373164486c45ed64f7352133ac
022d1ed126e68cc1d2ff70653bd0093cc6bdea7f
F20101219_AACXIF thomas_j_Page_046.tif
7198d64d73f22b2629f0e1480baebacb
d8d1cf4f6a173fc7af315ebd842c0d45b54bebf6
32387 F20101219_AACYLI thomas_j_Page_098.QC.jpg
61e7aefedc10e3a977228a0c24ead6ca
f9aabb60e322d575301a3f27209784f23638c569
8299 F20101219_AACYKU thomas_j_Page_084thm.jpg
282bf88d8d0235248a986c1b574d43fb
71b1ecd594108941d64ef3051b17f14de94b2224
30195 F20101219_AACXHS thomas_j_Page_024.QC.jpg
8cb3f0876903caec54ce1248c9cbcd63
e2f348760340011f798e702a0355b30e18df554e
F20101219_AACXIG thomas_j_Page_104.tif
6fd0c6709ba4fd947b2b2cc367683b98
6f937de50ff77cb55578135955b9f076d6d25044
8456 F20101219_AACYLJ thomas_j_Page_098thm.jpg
1da5ae777473674a486b33763baa3d7e
c16701d883247b578e9315098ca66d64845b7aa0
38237 F20101219_AACYKV thomas_j_Page_085.QC.jpg
e58710335f7e729f0cbc7a09196209f6
6a0f2720560c229eeed43c533806080e20f2bab0
86768 F20101219_AACXIH thomas_j_Page_014.jp2
6a3c5d743b1438be917d9cbdac1c518d
b9b707bad7aa5bea46498cbc6cd0839cfa88394f
87160 F20101219_AACXHT thomas_j_Page_032.jp2
98f1682224bbdf36eeb7e7d558ad12ba
fd125ac6a808fb7b1155a4869b3acb5c687c805c
34003 F20101219_AACYLK thomas_j_Page_101.QC.jpg
dc86f6f50e4f7397e22317765eb31205
2e4a0d7265378296d7f8d82845f5d1c28cc34592
33270 F20101219_AACYKW thomas_j_Page_086.QC.jpg
9161c7d669fc6d6a6a78c1fc11e96ec6
bb2af436fb9da1e4e9f65cd9dd99d89434414454
104794 F20101219_AACXII thomas_j_Page_117.jpg
2b4d6885279cb8bf9e68e0a8f2aadf48
708e96eb5f8961fd996a807edb8292308ab06599
F20101219_AACXHU thomas_j_Page_111.tif
fc5f5f05c40d56317676aa7ffbde1353
4d1ddad8dc2e51faa60d44e73792c52331bcc3de
30478 F20101219_AACYMA thomas_j_Page_116.QC.jpg
6d3e2feb59088c95e32b9639bb6df625
6a6868d994a1903c8167dfac9a7e1a9b3fc2ccc2
8265 F20101219_AACYLL thomas_j_Page_102thm.jpg
50671aa29714134845c7541e3723ed06
fc36ec4563d076e226960c8499ea7d5e8d0d1a75
8343 F20101219_AACYKX thomas_j_Page_086thm.jpg
7c6e250c8b2f9192bd3b02075921063e
49d63402c6300e06d3e6456f98ab690b504e9d03
46991 F20101219_AACXIJ thomas_j_Page_049.pro
770ad8bceccebef4043b2376edbe6a98
e406aaea2616555e8a10e7ce3301c22df33ea72a
F20101219_AACXHV thomas_j_Page_026thm.jpg
205aaf638e0e588e8b3d623cc320651b
244785341ffa24a31d73f0eae71d129adf649737
7702 F20101219_AACYMB thomas_j_Page_116thm.jpg
613858c60004054fc2a8ebbc8c533506
a54a88466732d332e7ca739fcc58d9f73a4994b7
34326 F20101219_AACYLM thomas_j_Page_103.QC.jpg
9987a52fd9dd7d79f4879d1150d2df83
fe35de30bd288849de48b227a8f73699e1f66906
27067 F20101219_AACYKY thomas_j_Page_087.QC.jpg
6ee2bb182a357c39d9f4a680539c1c1f
8659af9a5b327ef67e0c17d4d73fc2cc4a53b5ec
6271 F20101219_AACXIK thomas_j_Page_011thm.jpg
b4099f98e451cdecd13798b0376708fd
d0566c8c9eb1e3488f8599178542ed934d742d23
101496 F20101219_AACXHW thomas_j_Page_068.jpg
b6923309c2a4a9f744ceddc51c8a7756
d3faab93e381ebe6854327b2993756aac3cf778b
34341 F20101219_AACYMC thomas_j_Page_117.QC.jpg
4397ad14e1ac831f34bdb81c8d279441
91e2b982440afa478704200f7ac8e8a27e30c9f6
8719 F20101219_AACYLN thomas_j_Page_103thm.jpg
80d241c86fe61cc7b8a2820bbd909a47
8e1b784fc0b6197238d3dca28c564e498a56fed0
33113 F20101219_AACYKZ thomas_j_Page_088.QC.jpg
2aa503816f83d04fa43b8ca99d2a7150
2ec43b8b465d39642792c755b314d6f67df96019
91574 F20101219_AACXJA thomas_j_Page_056.jp2
1ef6260d8f00a35996c13c8db8268b18
7b9b201e04bbfba2cb88aad91563ad5964a4f6c1
24760 F20101219_AACXIL thomas_j_Page_130.QC.jpg
ddde107506f0b987c6c2848f14e990c7
62e5b02e331524639e6d390e49689ec9b7f1bb0a
6539 F20101219_AACXHX thomas_j_Page_094thm.jpg
251d40c09fbd3a7adc6e4e60dbd50ab7
5f18069e7306cd53eb1b37aec4bf4051e77f2485
34425 F20101219_AACYLO thomas_j_Page_104.QC.jpg
c4ec2c3c20e6f2a44724dc355f204888
b1ab4381b3fb4892d6394df2eedc8da0faf61bfc
37756 F20101219_AACXIM thomas_j_Page_041.pro
5800687dfadae5ac62e41e4fcb479992
e981a900489a23947bff538f8ceb46ee2d6ab9da
7504 F20101219_AACXHY thomas_j_Page_109thm.jpg
854145477edf76a7ca7c050c5eb8a44c
3455cc33b8ae0b20cf86d293f0c0d816fbdb4cf9
8759 F20101219_AACYMD thomas_j_Page_117thm.jpg
5f3b7c4069976f85444d2c762d5ddd37
313fd32425caa79d652086c0bbee3e788449109a
8873 F20101219_AACYLP thomas_j_Page_106thm.jpg
1be36853a0882fa147566e82b8883b15
2c3e9e6cbdde538e5e739707ce915c85132ccabd
849119 F20101219_AACXJB thomas_j_Page_126.jp2
6ee9165176b475ba93cdff46a9e6cec5
fde68a3bc06c3ad94aff2f5fdb64520bda3d35d6
105311 F20101219_AACXIN thomas_j_Page_070.jpg
a80eba48368e46ed2c84ab65a77ec310
b74fe2a4e59d3feeb549ff4550a95844c336326d
99906 F20101219_AACXHZ thomas_j_Page_131.jp2
1c8cc58e7f4d59e1cddd64846a7d7eba
855eed7015e9664f6296712d20cc8c641566faa2
33808 F20101219_AACYME thomas_j_Page_118.QC.jpg
e9f589355a1e0ae4a186138c18ea4c15
426a49aca5967a7dd0950a2ea94e828a30c9b93b
24597 F20101219_AACYLQ thomas_j_Page_107.QC.jpg
1278936eccbe9df1704cceed73045b93
faab94ff3702da2b734ec83e41a81f038e830102
9082 F20101219_AACXJC thomas_j_Page_141thm.jpg
4710438db90f5b36b71fcdc51531797e
ba98022a7220d6f0f4637f05d80de64d8769d715
F20101219_AACXIO thomas_j_Page_041.tif
79a151a80403dddc4e0bb7f44f33d975
df825cd580ada7ddb0eff69543edab6af90012a4
7851 F20101219_AACYMF thomas_j_Page_119thm.jpg
32f9a18a2c1d223ae559461da6893bf9
c6f335a54e05b4003ca31b5820294692acefcf86
6637 F20101219_AACYLR thomas_j_Page_107thm.jpg
3c3a49f590790b35b70f746fd99e1e81
b52e99af579ec8eac335c66f77d27b0a1984c93a
F20101219_AACXJD thomas_j_Page_057.tif
f5da6af3fb9209aa840b4417a69aec24
36cf14f7474b72156db197113106a63ac4f2089f
F20101219_AACXIP thomas_j_Page_143thm.jpg
c4f5106de6b738e153bfbf09c861bbb3
eef6cb2b940f41d9d07305ebe7f7745a912fe8d3
9268 F20101219_AACYMG thomas_j_Page_120thm.jpg
a117daebc5dc3ffd04d20c65a2118413
331f501733d6da17373f88b37b9e0a8017481b8e
31458 F20101219_AACYLS thomas_j_Page_108.QC.jpg
2fe4d8eb91032f9a0152c81ba960839b
98d3c0a4a7e598fd643aef5fd9e6cee23b4ffbbd
7805 F20101219_AACXJE thomas_j_Page_020thm.jpg
770b623dbb679fa8897d4ef4b71c7b67
d2926083b3119b4a073b4d545d5a8ad5a94c44ad
8197 F20101219_AACXIQ thomas_j_Page_017thm.jpg
1d3182d26befaec788de3bfc66b4a396
f2d374eaacd8d04b471c442775374dcefd9ee761
8722 F20101219_AACYMH thomas_j_Page_121thm.jpg
8e87341b6c5b1c3e1daad04967fe17b9
7cfcb84b395ddfe1ee39792eac604c1198c8cb23
8275 F20101219_AACYLT thomas_j_Page_108thm.jpg
fa8d0126a52b3d55d05d3a1ea8db90fc
1245228f4dc6f045be1c4886a294362ef5a7c4bc
91937 F20101219_AACXJF thomas_j_Page_044.jpg
fc498c2b94d876e0c69f21afe0f46b69
ff73b7f244439ae14b787e0346eaf8d4caf6bca9
F20101219_AACXIR thomas_j_Page_014.tif
709c177e2772a6782efaf232d0881cd6
41532fa11a2029187ae1a23b2d1553d037220e86
7277 F20101219_AACYMI thomas_j_Page_123thm.jpg
0a9fd634b59eaf74880e1efe96ce71ed
6c0412a2bb4a71cae9a3c4a939e183d0736ad5ef
31865 F20101219_AACYLU thomas_j_Page_110.QC.jpg
76e350af206666ca5435e065e7100807
98b5bad351bfb71146d6e74b44bf276f4544f665
103169 F20101219_AACXJG thomas_j_Page_019.jpg
1e08b87d6dc732a42ac4d92a20e1ec96
d295057dbbfaf713ecaaedb58dda5b34310d0e3d
F20101219_AACXIS thomas_j_Page_123.tif
de7f8077ed7badd46d05d70fb6913de0
6a9a33eda2a997f6528f20eeff96371588154334
28402 F20101219_AACYMJ thomas_j_Page_124.QC.jpg
0ef322f5471ee921f13968fff1241e05
8adc52001b218e9144f0c06fadaabb09246b51f1
31601 F20101219_AACYLV thomas_j_Page_112.QC.jpg
33159b001708bde809c11d095ec69f3e
fe45f252eb56888b44c00329d77febadac0e2e84
1628 F20101219_AACXJH thomas_j_Page_038.txt
ad84eff2463e5156c3dc1c4e67141435
d488a58868ca588b72f2a243ea89aa5c2b2497f2
F20101219_AACXIT thomas_j_Page_035.tif
aff388d6561d8c7dcc2bcee6cb397982
b12448444869f4071f62cee7d81247c0b7f33e9a
35231 F20101219_AACYMK thomas_j_Page_125.QC.jpg
bd4ee51b0a3fbc90927c4e4ef0d4bf98
3fb8e8c32b4958896e9632d25365d7fd64daea17
8090 F20101219_AACYLW thomas_j_Page_112thm.jpg
2067016dbe796b4d30758b5012e961dc
d9bbdc532a4cf141b4ea82813b496aff3e799bc9
35176 F20101219_AACXJI thomas_j_Page_067.QC.jpg
8268d691bd6658bccd92d1de3275c3e1
028a9817ddb7992dbeca1de578ab70c5d36653d6
F20101219_AACXIU thomas_j_Page_019.tif
ac7f526065f9fa828a22311d0f6f76e2
b0560f3a052e4e5bb33590b16dc5d4384de7e468
3614 F20101219_AACYNA thomas_j_Page_137thm.jpg
e2407ab94e10ffbe04fa432059aa876c
26e8894f303650e3f83f3ea51cec21625c8678a6
28461 F20101219_AACYML thomas_j_Page_127.QC.jpg
fc914a97ffc11a905be282f0b7a87b7d
b228669b6988d3c7c57705d3eae46b140e6cb9df
32938 F20101219_AACYLX thomas_j_Page_114.QC.jpg
833764ab6ad8e5f42076b9ae109cab43
58d98045def01124b0c94c5bc147db84ba567f4d
28214 F20101219_AACXJJ thomas_j_Page_023.QC.jpg
e977b25d1328c4b85630454faaf20ff3
4a4d477593c573a949426d2e0861a79d2e97814c
31727 F20101219_AACXIV thomas_j_Page_020.QC.jpg
981f1ad60a9cea390713121349b33503
914b4fecabd191556968c2ad0364f67cbda64095
31200 F20101219_AACYNB thomas_j_Page_138.QC.jpg
851ff0415c047af9bb75d2d69600c215
bdd6f2877186269b70573768550df012460d2703
7622 F20101219_AACYMM thomas_j_Page_127thm.jpg
8ca93a8716e7debb566cad5e0b063a94
ade9770f9e230cee86ca3172dcd319ad213df80d
8204 F20101219_AACYLY thomas_j_Page_114thm.jpg
409fb17e035686c217faba7f9b51e1ac
b54cb4d1eaf2ca30eee88b099b75e57f8f05bc77
56691 F20101219_AACXJK thomas_j_Page_006.pro
c0d24de39d673490fcaa4a55a96d8db3
7ca8d0604312669f1079c2188329cfcef7ae4c6f
108690 F20101219_AACXIW thomas_j_Page_103.jp2
bcfd5364d7c34907044d828406151ed0
fd861a4ff5caec2259fea5351a45df8765efcd8d
37743 F20101219_AACYNC thomas_j_Page_139.QC.jpg
324c1373cb30b7a2f28ec5b70391a054
226a674675a0efedc10dfaea07dd0eacf69918d5
34469 F20101219_AACYMN thomas_j_Page_128.QC.jpg
353b23b9be450e4b45ea8a5f65b11520
bd2fe06a7b5aebc6448d35356b4249b09a7f8227
8141 F20101219_AACYLZ thomas_j_Page_115thm.jpg
d4e4cd635cf940e78fee55cce4aee4cf
7f2e25c0d26f7edd10e471ae8ab52c15aed24fba
29057 F20101219_AACXJL thomas_j_Page_044.QC.jpg
82e2ce37f4fbd40d9b80dfd00a6d07e0
5057b8b99245090b6f626a30524a1522ebcb0859
8024 F20101219_AACXIX thomas_j_Page_072thm.jpg
40c337ea571bd58307d54fe5fd09aa25
1500d4c9cbe27adfb12939053c7391d73f26251d
1529 F20101219_AACXKA thomas_j_Page_094.txt
1df65e21e7c7c2c078336415a059e5d7
a1c65489b7d13a2a61610b297ed951b60035f19c
9038 F20101219_AACYND thomas_j_Page_139thm.jpg
0ae674a5b10793480fd2a604f97a223c
ef9e3df66f9d26efdfd9b1b81e48f322cc388078
23982 F20101219_AACYMO thomas_j_Page_129.QC.jpg
2fe2d2a6ae2092a2a19fb444182cd304
bfa0535db6d1a477308b43771325791983814a0a
8575 F20101219_AACXJM thomas_j_Page_067thm.jpg
8da85496d799a2f7353ed3de8de67965
07bfa79b420c7cc6a6fae54837d50b53f7fc2dda
8146 F20101219_AACXIY thomas_j_Page_034thm.jpg
112866fcbfcca0394c2abee05933a60d
b7b57255148d64023b9187a9b2bc08ebfb07e7a6
88309 F20101219_AACXKB thomas_j_Page_096.jpg
273d86e9f280e9628a101c0f17f5b7af
c43a1e4a6198e7aac8537b11382a4156bdf0d145
6164 F20101219_AACYMP thomas_j_Page_130thm.jpg
049e1854bb7f12354bd60e8fa928d82e
aa11a455dee6a7bbc9b4b704cd96bc583d8ccf34
42802 F20101219_AACXJN thomas_j_Page_010.QC.jpg
9142cdcb24427e4d0c821f7817f2e2f9
879ea527189f68586e82094b0937270c6bd77c34
24130 F20101219_AACXIZ thomas_j_Page_100.pro
03665b0efea08f54fdbe354e120da0e0
ae7ebcc2863a40e637c9e8dc9d41dc24fe632c39
35125 F20101219_AACYNE thomas_j_Page_140.QC.jpg
bad9b38f2ca0c54f70df9cf4ecfe02d3
ca9df5f3d59658c1b6f740d9d62c039e848d1604
30509 F20101219_AACYMQ thomas_j_Page_131.QC.jpg
c11657167a24ae4b8e4188c7a8a5871c
7ebef7e694862f04c6e089bbed3c9892cbafa279
F20101219_AACXJO thomas_j_Page_142.tif
7169d564a700e32d75269e4c32fb1da8
f229395c7576b04fe1d520cf21d8f43e6927e947
27267 F20101219_AACXKC thomas_j_Page_011.QC.jpg
3fe8295eb879f1253b0d81bcc52d310f
ac80fba3eaaa70d5e05748fb23a3aede4565468d
8644 F20101219_AACYNF thomas_j_Page_142thm.jpg
0f9ec29de8351086ec431f8886d4d921
bd3c9059ac6786f51268330b7e83bbae15597219
33190 F20101219_AACYMR thomas_j_Page_132.QC.jpg
bdff670c252e68b89bfb00556bf07dbf
35c082c7bbff114f5183c7f779724316c1fdc427
8297 F20101219_AACXJP thomas_j_Page_050thm.jpg
fc75abcdb3deb0feb5ceee3ec65f7abe
d8edfd6d4a2e02b3bb0ce16104d72e0e6e5fa9c5
1567 F20101219_AACXKD thomas_j_Page_129.txt
fd6f6ac79015a8179a99eb9215533f5c
17198fed7cbbb8d09c5adbe9f84ff9dd0973b297
35141 F20101219_AACYNG thomas_j_Page_144.QC.jpg
dab4afe732233d11b44ff35fcf65cb77
fc351074f9c70ad766277234d731d683d5bb6c17
35177 F20101219_AACYMS thomas_j_Page_133.QC.jpg
f694a657a35262974ffa83e3e689576c
07a669482179056448506ef3e2eb0cd8f40d21ea
2461 F20101219_AACXJQ thomas_j_Page_007.txt
93d2793505d0f230aaa44d8b751f649f
8536bba81d83dab49989e0278578e0e1e14d1ec0
78917 F20101219_AACXKE thomas_j_Page_047.jp2
60391131ce4a314cc74c2347c92c12e6
b3819f828c36cbae647f2485c23ea799c1a6037f
8361 F20101219_AACYNH thomas_j_Page_144thm.jpg
e7c2497fe42cad58b3b7466982d24c62
05d49d05f1c23a4295e42636893e2113fdebf604
8676 F20101219_AACYMT thomas_j_Page_133thm.jpg
eb7c0d62ce4ef2ececc88304b17744fb
5094e99bac10986afec2d45c194f83f9acba529c
F20101219_AACXJR thomas_j_Page_143.tif
9e3fef545b282853f18b0b72b67b41d1
650f52f3828ae52cd214286474a6cdb068fb5a8b
F20101219_AACXKF thomas_j_Page_065.tif
80f2c62fdbd6eac9a1631d7b8daebdeb
2062b8f8bc3e33d6211df809b138bbb618004f6f
8512 F20101219_AACYNI thomas_j_Page_145thm.jpg
0810a22b0ee9579429be9bce334fe20c
79d199baec2b36bb475cd3f52a6e8ee21b149939
32868 F20101219_AACYMU thomas_j_Page_134.QC.jpg
0421398386fcf6aa2d0e1656d729c673
7ff6b8db8f69dbf2ebe7386d3951fc259ba000ff
7370 F20101219_AACXJS thomas_j_Page_093thm.jpg
2e57b3e815ea94a905a1177d4d7cec7f
00478f16561bf209d9471abf92dc59cea079c1a0
107814 F20101219_AACXKG thomas_j_Page_133.jpg
5c8eb5034f0e6bb8327ea37bb8a4985e
04d6adb0a2ead41d848fc330ebe5b0146136d178
37587 F20101219_AACYNJ thomas_j_Page_146.QC.jpg
26e5a15bde716a8b793bb519de0a341c
cf0373ed9e1b718272433cc7cf02597a2ea5da4d
8228 F20101219_AACYMV thomas_j_Page_134thm.jpg
24b761e2754f415430d3702a677eb4f2
a34a156ee4774960d23e533a9151e95a5b43099f
34592 F20101219_AACXJT thomas_j_Page_121.QC.jpg
1755b2b404db6714aca38fac18616b12
e59c0ec3da8f038572466be2858c13d731527e0f
52690 F20101219_AACXKH thomas_j_Page_037.pro
c09506a29c984050be5cd7776954a4ec
f9df803ac4406c3539ce2c878dcd3b48d8404f0b
8791 F20101219_AACYNK thomas_j_Page_146thm.jpg
56869bbeca2e0d3eaaf1b35e7abf3967
fd26df27e836237eeee182bb0ab951da56b20f4c
25916 F20101219_AACYMW thomas_j_Page_135.QC.jpg
e22942b9339d51a4ef5b8203996ecbd1
ec8fde42c783ca47508ef97356bf45283b66fec8
47677 F20101219_AACXJU thomas_j_Page_030.pro
886068f70dec6abe37c4464b71a258ba
441703212f24bb1febf1fb1ea1cd045a295936f8
1935 F20101219_AACXKI thomas_j_Page_065.txt
74b1e7990b2b4b8b9a6f56832837b479
35500ff604722c935c23d3215e840c74b4ad6a6e
8884 F20101219_AACYNL thomas_j_Page_147thm.jpg
d92a634a250c84165aaf07a2e8bdc6dc
4acf86303fd87502860b97f3731e539338f84586
34313 F20101219_AACYMX thomas_j_Page_136.QC.jpg
eb088d169e8d2ee4af49e86bca1bace1
2bd5126e7e9e3cbe91127f9a426158f63d1fd57d
8108 F20101219_AACXJV thomas_j_Page_110thm.jpg
0f5db5604ce25b319b171a24936496af
d50f65d5073456986014780342d983b6d23ad4c9
101281 F20101219_AACXKJ thomas_j_Page_088.jpg
70fa5b5fe9d767ba76998ef34d72f490
0a8c80b5807a4a538e38b960c964516ece8f696b
25926 F20101219_AACYNM thomas_j_Page_148.QC.jpg
5885665c74f995a8e0c78f9ecfbed5a0
c8732f5e703b2b1f8f64476143d808fe53e11eae
8573 F20101219_AACYMY thomas_j_Page_136thm.jpg
12976698d834a92e1ae98830593cec45
1138ef281380d9214a33ffda99e5ed39ef308ae2
104216 F20101219_AACXJW thomas_j_Page_034.jpg
d50dd37170d7f116e47b99c9cd91adb9
11727212e6fb30ebe286252a47777b9ad59ca575
83341 F20101219_AACXKK thomas_j_Page_091.jpg
20af81e93e994fefc36c0b5ff39b714a
6f696fc71453adea385fbfa6d1d828cc532b54b1
6161 F20101219_AACYNN thomas_j_Page_148thm.jpg
f310140a549d323d1069a615bd078aea
82932c9d9f637631e9a8f03d5338f37ba0d437f8
13920 F20101219_AACYMZ thomas_j_Page_137.QC.jpg
d24b3087a4fec35508e66f6132573729
cbe1a49e7f57631e2acfb1f8dd1aa90b34957f73
103106 F20101219_AACXJX thomas_j_Page_055.jpg
6e9ee5443875fb00feacb531477a1c22
ecb4e02d65bd469bd4a42d9e708a3707228c7a83
2430 F20101219_AACXLA thomas_j_Page_008thm.jpg
e10000fab29aabd65d1ed039c79766d1
16c804d2d6f8059bcd084c319d5ced2424f31f26
46391 F20101219_AACXKL thomas_j_Page_131.pro
e42d64194ce24d44d4dd12b33a501f40
240c6d1abe3e5d6df0b5c0ec5ac0b4123cc20945
11784 F20101219_AACYNO thomas_j_Page_149.QC.jpg
58324d9423669d0edbdb06fc8020df98
20027786f45fae53979da9a046ca4527de067349
F20101219_AACXJY thomas_j_Page_115.tif
82cded566e61510594b3f78cdb7c3ecc
141a436d86893e5b26c2efe66b267d18787ab697
29934 F20101219_AACXLB thomas_j_Page_111.QC.jpg
c381fbe78390d271d514c97a15e6a861
735061180bee72726c1b1c8e1815c658b3ea9b66
64517 F20101219_AACXKM thomas_j_Page_100.jpg
d724138173f4533916779fa4bd462aef
f0e5bca06fbf0b59eddb742376b2822afa797791
3136 F20101219_AACYNP thomas_j_Page_149thm.jpg
9cb788f6041da318848e1c32c6c402af
ea26de4f3cc5963fcf987668deb5dc57ffe208f5
1051973 F20101219_AACXJZ thomas_j_Page_009.jp2
b44e25c7dd2406d0ff78c125989460d8
394a9751493091a35628b2e3328b1b28f9a63e25
2007 F20101219_AACXLC thomas_j_Page_016.txt
0fdaddcbae665938d82c30fdd572828f
c166b0d5f65631b8b581289a89c3defca37747bd
F20101219_AACXKN thomas_j_Page_080.tif
9bd99eb6a6bbbabfdf4ba0a2d3b12705
1cdeb6ca2cb6cabd1bf4fddbc1df546e599cab4c
F20101219_AACXKO thomas_j_Page_070.tif
d589311ff5997f23db5b1d8ee27cb0a4
d128e4aa637340c048bdd5a6839455a2fab5734b
6544 F20101219_AACXLD thomas_j_Page_043thm.jpg
287a965781aaa6a3ced66e9bb4f8ee35
5172bc691f8e0582548072c494f3ec33a1a764ba
47641 F20101219_AACXKP thomas_j_Page_063.pro
9a121f2c6d302304a100953f217dbf70
21be3b0f868b19b892a17651ea002e6c39074753
108 F20101219_AACXLE thomas_j_Page_002.txt
1fd907560c289fea5ae3b8f4654cdf14
5ea97d1fd072041f56d9efb7a92acd26a5999236
31645 F20101219_AACXKQ thomas_j_Page_074.QC.jpg
99ae04c999ffb97ba00ab16caef0fbf1
9a83f47970bcda1e5cb2eaf8b372c18ade766fe7
106897 F20101219_AACXLF thomas_j_Page_117.jp2
a31fcb25b8046aaa2ced6851ce1e1c42
0cb82db29c9d59b5e1e0f98274506f64bc64e941
67166 F20101219_AACXKR thomas_j_Page_013.jpg
9a9dbbb9de4228cd0105d7c6d41f4d54
f319098b4bf7e236cb28273e18f6df95f4c57ed8
1918 F20101219_AACXLG thomas_j_Page_046.txt
57496e4c5e3292cf36ea954a22ad24c4
4a5f0b85fb2ae86d81ed6fde462ae44c4c3c90d7
91762 F20101219_AACXKS thomas_j_Page_036.jp2
fb85d7390097437b1bc195f3323cb820
e50979379b8646dd3c10bd2c450e366ec68c329e
105492 F20101219_AACXLH thomas_j_Page_115.jp2
87bf3d35b14b7929282582b9830d96fe
a600d4c5e439e8dae87eb5fb725c31d37a6b28dd
8706 F20101219_AACXKT thomas_j_Page_128thm.jpg
01af8404094c6d9640ef4b0f3cd7d3c1
5552c8b6e2754de0ab7dfad5fa5acd2e1fd4c2a7
61605 F20101219_AACXLI thomas_j_Page_145.pro
f054570afb47cd2e3c6fa42d6fcdcc86
11085f3b6d3aa95c8feb70345f64b156878e2059
8819 F20101219_AACXKU thomas_j_Page_058thm.jpg
e13a3cb8fee4f5cab1baa78c40200484
b38c7e4da041aa9a0f73a57f9dab06c6252cd2c3
84156 F20101219_AACXLJ thomas_j_Page_041.jp2
03177c9af505e02663231c1cd9a5c353
d0c29fdd5c69859a410de1d7588e5bc050731e93
8641 F20101219_AACXKV thomas_j_Page_097thm.jpg
5a8459d1840fcbbf01bf5883a8a2e7fa
511e17408893c99cd45a1843504ee7b0e10e9444
F20101219_AACXLK thomas_j_Page_077.tif
58540e31108f3fb3f664983f3bfd726d
f91ce780de349c898ef3ce3bb037b47888da4f79
88996 F20101219_AACXKW thomas_j_Page_081.jpg
32fb863ee849109a38a42d777224062a
b2c827dcc0aa2376e557ae7861baaf6608b2ed80
111410 F20101219_AACXMA thomas_j_Page_060.jp2
b4a9dc5257a51c9bb72e1e19c4fcfcaa
076c1436136045bcc04d0adfa798dd0ad6a9042e
52633 F20101219_AACXLL thomas_j_Page_097.pro
bad5ccd1eab2b3945e3457f61456a2c4
622cbfff09959e84dd8a47b9f3c7862c450298ac
87829 F20101219_AACXKX thomas_j_Page_043.jp2
ffef8c2218157d55e605f5ed692ba16c
d8baf6de52370f219e5723dd03417a86b84d4f4a
F20101219_AACXMB thomas_j_Page_126.tif
e58028e355066df098de6a41d04cf775
1259da354979b820836017ef94467f2c834b0b07
F20101219_AACXLM thomas_j_Page_005.jp2
0734deebd65328ee64ee49c15d8daeb8
2cd326df323cde5afdd0724ad1eed2c144622fd5
127764 F20101219_AACXKY thomas_j_Page_145.jpg
f6541b1d71d04dc34328471b5a98a4ef
4b5ffaf5468db7c66a083fcd8d6d84f4678ef7d0
19866 F20101219_AACXMC thomas_j_Page_005.QC.jpg
0b411502ac78f79167685edd20a17c5d
1d41a9ff0f1561c10c0d20d2138702a70b8cfd78
35875 F20101219_AACXLN thomas_j_Page_145.QC.jpg
a4494e4e6ace3476658bb85ccbe76abd
420bd69acd1af821c2c883387355877e86d5388d
156166 F20101219_AACXKZ thomas_j_Page_010.jpg
634764eb256c1f2f429910f2590ad369
1d34c738740e87f19738dcec4d09a35efe23b168
F20101219_AACXMD thomas_j_Page_010.tif
cc95aea95f33581e734335620ce7dfe9
1b681908ef0230e5724f9b13ecc1bb5bb5a5f367
102105 F20101219_AACXLO thomas_j_Page_104.jpg
7e55223f160e55c1189d9ecfe82240ba
5b2118d515f02013ed10e87dc2760c1b5ff86360
1991 F20101219_AACXLP thomas_j_Page_133.txt
b4b70298aba002fe2a2bd226210f968c
994b8db994408a62ec03231a05649f7c2e8311f3
F20101219_AACXME thomas_j_Page_055.tif
89798fed7e27965e55eb2d8b380a33a3
1cb340c8a6f7de62c2f9b0e26b699b45e8962a87
99506 F20101219_AACXLQ thomas_j_Page_040.jpg
eb203a4692bfe581d3c933d252490161
344dcf8b11b2209dee4b1f9eae7939b936c517fb
172184 F20101219_AACXMF UFE0015582_00001.mets
d47a3af09076bce13b8eb7dcc4041143
b16ef74ab971c03914b32659c99b0428f0ecb7e9
thomas_j_Page_001.tif
320 F20101219_AACXLR thomas_j_Page_048.txt
c2647bedbeed42ff47b855cbd3baff67
975d8669b486c7859e46bb52555b3ec47c61cae7
F20101219_AACXLS thomas_j_Page_100.tif
e9fd821c4bcaeb2d22c838ef2fc8b229
e0cdde617cfb5b6b367a89ccda9a7c695590a5e4
32863 F20101219_AACXLT thomas_j_Page_105.QC.jpg
2433bb186f3e2fa9f0adf4154bd03ac7
9b61b9129c173aaf84253b2f490c41557c7da00a
5100 F20101219_AACXMI thomas_j_Page_002.jpg
720ef04539b918900462785c155eb8fa
16b1654d581f2085815ebc46d504b9e9e29bd5f6
120563 F20101219_AACXLU thomas_j_Page_144.jpg
a3b895a18375473912da75afcf256d61
c54d24f52620e13f403aa8a84dbedc863b7e17ef
8880 F20101219_AACXMJ thomas_j_Page_003.jpg
ff110fe122fd4ce8a4f4f7e03d04c512
de8fa6e463ce3f6de885870c44398f3b8fece79a
109899 F20101219_AACXLV thomas_j_Page_082.jpg
0083aceb21434823b4cde35e224da29a
15369481f711a6d0a4426cac8bf3d7ce684975ad
73506 F20101219_AACXMK thomas_j_Page_004.jpg
12e969535909671d8897124ded1240f7
49c5aa18f86f7ff844c1765574ff934a9104d80a
6595 F20101219_AACXLW thomas_j_Page_076thm.jpg
9b02dd87f8c91b59f2a0a0cc50958ba9
97942ef1468d3d01d2a0f594d12a085bd9e9480b
101517 F20101219_AACXML thomas_j_Page_006.jpg
70058251fc5e6d2321244003ebfeae24
a7a886e1829416f43de8a93743ad69d7a7cd401f
50617 F20101219_AACXLX thomas_j_Page_133.pro
f48053203aa377123092df4355d24596
2fcb3bb5bfe4d4fbec0e8239b14db869108a309f
91219 F20101219_AACXNA thomas_j_Page_024.jpg
983fb14075bbd5179d24a4471e9181d2
c1297487b711cc3f1cae6a218083bd9b265ef0ed
126676 F20101219_AACXMM thomas_j_Page_007.jpg
708920292c0eacb5435aa13c102f0830
35b76382a7ecb69ef0e851936abe100f60477f26
31990 F20101219_AACXLY thomas_j_Page_040.QC.jpg
8ce9d99538317ebd8a709c9da262b817
99c16180a1b8a0d0b95138d911f036111c2b796b
60992 F20101219_AACXNB thomas_j_Page_025.jpg
2d0d739488b10252abcd82e621a26784
9b1345f07b780442e5465678170f8b25172e5fb8
35957 F20101219_AACXMN thomas_j_Page_008.jpg
29dbbc805721a366f0a76ebfb01dc88d
f340c380f257900000c9fcd822dfedacd7dd0953
14768 F20101219_AACXLZ thomas_j_Page_025.pro
dde2a0c19a6646d744a64a62d2941dcf
6643bb04b06627f2009af4257d87a47dc455c4d7
99955 F20101219_AACXNC thomas_j_Page_026.jpg
990b58bb66f48e50ec10eca3e52a1ae1
bc6833d5b8e3df1035cd4a38c1bbd7734a9ae18d
123197 F20101219_AACXMO thomas_j_Page_009.jpg
dcf665aee93463588bb83f2a816512c0
076a98abf4f32be4e8239e941c0c461e3f616b70
108171 F20101219_AACXND thomas_j_Page_027.jpg
0afe3438118967851e97c3f6da4f1101
987629bf94c354f1c95fb62c1cc92bf53e3241fc
101533 F20101219_AACXMP thomas_j_Page_011.jpg
e7fee24ae4eb82916056ba34d761d701
bcf00100e36090646442d12a222749c5b22d82be
103079 F20101219_AACXNE thomas_j_Page_028.jpg
e041c354ba1029b7cb120f9a1d8374a6
c3661c2c16a4347fcbe4c7862f376bbc835614c3
86605 F20101219_AACXMQ thomas_j_Page_012.jpg
eb880a1a022856a13d160665aec7ceb0
7afff446e0b7d400564588ce2b72277d17be89c8
97796 F20101219_AACXNF thomas_j_Page_030.jpg
65fe13cd1e23b49a8d1d1d913b69ad43
16453c2826036332f250742e75b9fa3c96a0ab4f
81856 F20101219_AACXMR thomas_j_Page_014.jpg
cc896b25e1032dd91023e4f09fb22ec8
2fde9b4c4e6c38bc7ff32488335850f0a06a61b0
90752 F20101219_AACXNG thomas_j_Page_031.jpg
d4282a2ed2ec85828d308f1bfaa0b0fa
b2662eafa326e4ce67e3a47764c05e93f1857062
102055 F20101219_AACXMS thomas_j_Page_015.jpg
15f5a7892be8652c0b5f57cafd96ac88
a3bda1a321da36752c018e8f4fc77dfd9505f072
81584 F20101219_AACXNH thomas_j_Page_032.jpg
a5527b7e70c84f6895ebef789e421f64
386db8e52720f202861b3b9313fd4d9b8f3d2a18
103459 F20101219_AACXMT thomas_j_Page_016.jpg
b5c10b7bb118b24d8bf16273205682d8
827b8808936affd068ac30cbecf1bb523ed621c0
86554 F20101219_AACXNI thomas_j_Page_033.jpg
666a1707c46744254ab83ce7e1935290
ab0da0ac47ecdf0c2fe1463a1513e02b5d0af4f4
101831 F20101219_AACXMU thomas_j_Page_017.jpg
fc4e4a2c045482ac26df7669a2379645
737fb763ab8987fc8e17b7eb7664154c2b1d54e1
86729 F20101219_AACXNJ thomas_j_Page_036.jpg
8190b91d27b5dd7253d71c5947245fd0
b72733089d6da47361127626c39d94a208209130
103309 F20101219_AACXMV thomas_j_Page_018.jpg
5f48078062837cd87fc89f96bf14b110
dd2cb3f12c2f1d4e579383a078d6b0e3f7abaf0c
107942 F20101219_AACXNK thomas_j_Page_037.jpg
d55d996916cfad1b42e84876545fbdb6
81d727518725681eb29d9ba240ff8ed104fd1f04
97179 F20101219_AACXMW thomas_j_Page_020.jpg
332fa96377cf9c20bb4a7f42520be046
972891f27617b66404a6bf79094f4d0c0f8a5f61
85372 F20101219_AACXOA thomas_j_Page_059.jpg
4e7f67c21e03b807799695679958e15b
68a6c9cd12f476154d0d3c13605c33eb15064a0d
78659 F20101219_AACXNL thomas_j_Page_038.jpg
8a6a08cbd78cbda9458d00f731093cfb
308688ec4ce526cf43ae61b468962c42469f6a2c
100230 F20101219_AACXMX thomas_j_Page_021.jpg
28e32265ffd24d1508d45713e67c991f
bab6689599f0f7251067bb9bf3e054f970e4c39a
107366 F20101219_AACXOB thomas_j_Page_060.jpg
33a29254be307391164d2bffc096746a
0469a478b6075cfb176d9f51d92ad43b9502b3dc
67469 F20101219_AACXNM thomas_j_Page_039.jpg
6ac22412741162f790a0f6a2f3d77b33
5e15a50a98cf6bdc7adeb0cc87e40b0bfc289b46
94675 F20101219_AACXMY thomas_j_Page_022.jpg
1ab36bf6b64fdd0208d2eb56b0ffd31d
43c06c47f2ca3fb0646bbe57917f0c862075742e
102927 F20101219_AACXOC thomas_j_Page_065.jpg
8d82d07e13aad46b9221a10e60e8c661
839c4e3bbaf6c044e0818daa786c4fb2b486f84c
81494 F20101219_AACXNN thomas_j_Page_041.jpg
0ca5512cb22563de90b55d994c00f0ca
99c80c5547c466ffa26737772fcb8cd925fa8f0b
87952 F20101219_AACXMZ thomas_j_Page_023.jpg
de24879e1658ef2e7c27d740c86a58fc
56a59764f43ffd51bcf6a9a08975ae17780c1b2d
105361 F20101219_AACXOD thomas_j_Page_067.jpg
2a5a3e9dc6e46fbc9775fadfcdd55d7e
f896097ea62592736f2bc670f56dc3b53061e9e6
81649 F20101219_AACXNO thomas_j_Page_043.jpg
0f969b3e2d596178a6d636327950a83a
523ecc9f0018dc0beb84670d4457010aff83fc6a
101511 F20101219_AACXOE thomas_j_Page_069.jpg
028e543e378a887741e604f7cb64436e
d0befdae4603e9ac9704d6ed012032e6315135ce
105044 F20101219_AACXNP thomas_j_Page_045.jpg
291477302e6e9b14d7cf63d267c32dfe
7469e8f0e46773bb36dcfaa02408b642c98e4619
81872 F20101219_AACXOF thomas_j_Page_071.jpg
05000562d816f21ff74d3b4cefebb8d9
deeb589ea2eede3c72813b099b92ed0766d2bdff
75306 F20101219_AACXNQ thomas_j_Page_047.jpg
b4a4d08635b7705bd7b76a105488f550
63cd2e0eb45da327638bf6908438bc6adf0d6d15
20099 F20101219_AACXNR thomas_j_Page_048.jpg
2645ec475087268bd5aea57c02108ca6
73548e655c5894c9533db669b8225384064dcfa8
97841 F20101219_AACXOG thomas_j_Page_072.jpg
241c8fe18e8cef2a9be17d74f876b570
663d8d943150665e8ac653f7213d7f7dc4de8415
99830 F20101219_AACXNS thomas_j_Page_049.jpg
a95f02b24650857faecbdb1b29bca1a3
0ea638d56cf78c90fbf6abfe86dab5aacf7c982f
99680 F20101219_AACXOH thomas_j_Page_073.jpg
7eb16f701f32384eb2363ec40acbc43d
efadff91cd121f10f895897a98dfbd79ecdb5a29
101253 F20101219_AACXNT thomas_j_Page_050.jpg
53e742beeca56a1c8209e04fb4cc20ab
7827ae79f2f1854a210151bb27075490fc156ee6
97089 F20101219_AACXOI thomas_j_Page_074.jpg
e070fa787ec4fdc2cb70b011b91f495b
e09a7a9cb7ead9f4eab17325997a5f0483aed3e4
103599 F20101219_AACXNU thomas_j_Page_051.jpg
22fb489372b5304f2b996b40317e4b75
23becf4ade8797d3ef25980a44e1f7d72151a2fb
89748 F20101219_AACXOJ thomas_j_Page_075.jpg
c75697f2a486d63e475bde3fba20ad39
e3a3e1530b5668768fa707ed854be16141e07dff
117882 F20101219_AACXNV thomas_j_Page_052.jpg
d70214ffb7aa0d5d8837a6a6bb649485
33b0cc1a52f6c3925ba7b5b16253d3070ef175c6
95457 F20101219_AACXOK thomas_j_Page_077.jpg
857fd8ad34a337957b9e7910e80f8933
1e5b033db926e1b07ea6623f1066adc0966a6aac
105060 F20101219_AACXNW thomas_j_Page_053.jpg
f9a6fc3783fa797196e29b4bb33ec7d2
55e71b5effbf5eb18981f4569061f9995bf5556b
79446 F20101219_AACXOL thomas_j_Page_078.jpg
669477c16d5e5ac50d9110db26da87fa
c8fc6596159efe4c161b26ed7f2a71970aa73e35
87258 F20101219_AACXNX thomas_j_Page_056.jpg
5b81913cbd7ce3441a60a0a258e1db76
3cf74b54e51dbf94ba097f90d53a34beea9b9bec
95529 F20101219_AACXPA thomas_j_Page_108.jpg
f252b3f08e25139af9895605b518b30d
b15a613e921e8b93f4cd354c47c73d915440f23a
103245 F20101219_AACXOM thomas_j_Page_079.jpg
638097e1664e01644962436c78575635
acc4cdf352f76ff8b6f54a645acc21a3ba03c6a7
81203 F20101219_AACXNY thomas_j_Page_057.jpg
e6dfc03c3a6647b7c796532c90805f25
e0dc80753ec40df359c180eee17f8b9fecabb865
93095 F20101219_AACXPB thomas_j_Page_109.jpg
7af9de3649bfcef3f597c57496d9a415
43470752b92a9770aa1661996a12c3199be4414f
104702 F20101219_AACXON thomas_j_Page_083.jpg
bfc1bb7a07314cf3b003c4f12c516189
37fa3273dd94116e54546ad80b62f86c3299436b
102433 F20101219_AACXNZ thomas_j_Page_058.jpg
00d08aa0e591b8fdcb8d7145de06bf0f
b558c412e1ca6d09002b08ab952cdfca8bf37012
99448 F20101219_AACXPC thomas_j_Page_110.jpg
476f1c21a826b68a318b982ac62a104f
41d051da4d4a1954da194a3cdfc92536bbbed5e4
101895 F20101219_AACXOO thomas_j_Page_084.jpg
cc3bb046175aff8ed390f9156b595b17
e901d74be1455d6b5249f2aed10e7b2b06899a0d
93368 F20101219_AACXPD thomas_j_Page_111.jpg
a3323389b2119f3aebe5dbc07639ad73
62f3570ef9c58f0d1a31b0334d123e3e55160d6a
131902 F20101219_AACXOP thomas_j_Page_085.jpg
68b1c446ac20f460c8ca4a00683a0a50
05842e92810084a145727f916d1ea173a76c7490
96739 F20101219_AACXPE thomas_j_Page_112.jpg
cfc293d3bedb728b83e8aeab982d58cb
a407bfedff72e9a71e3f3d908b445e22a67994b9
108837 F20101219_AACXOQ thomas_j_Page_086.jpg
8b33877706a90fe8fb360efe0ea9085f
4cadf0162b90e004cbfe533c69bf3ab82283137b
78201 F20101219_AACXPF thomas_j_Page_113.jpg
8556eaff32c6b377f4a61cb527707b63
1f452b427e2a91bf0dd6c821fdc9dfbfcfeeb461
83002 F20101219_AACXOR thomas_j_Page_087.jpg
c28a144997260ed25acf6cb0bb7b956c
7fd6976fb09b5eaf6e3932c3ff16002274cac8ad
103885 F20101219_AACXPG thomas_j_Page_114.jpg
0eb46895ef91afe8d29543d614e8c378
0e25dd260329572a0d03c8189ef20fb693ef9eed
81466 F20101219_AACXOS thomas_j_Page_089.jpg
6cea422de77b19fb59fe45c68bc459f3
227bde2626b04a506b33534e26d2dc3722ded30d
104983 F20101219_AACXOT thomas_j_Page_090.jpg
9eb2af248678f665731628218461b33a
8dfcf11176143b5cd92ab5c16762bead7a560a65
99276 F20101219_AACXPH thomas_j_Page_115.jpg
9a5c35c01c47f2f2d5e618b5921536c8
2b36ff0064fa99f5a9b239d13a7031b9acbe4ed5
106364 F20101219_AACXOU thomas_j_Page_097.jpg
7b156befb8708c4511d01291a1cd41a3
379bbf586212fa25424a98307d843dd78b2d3d43
100707 F20101219_AACXPI thomas_j_Page_116.jpg
6c887bbb6bd094343ac83a96c381944f
d64775772fd35ff2908baccab721ed820e68a07c
99742 F20101219_AACXOV thomas_j_Page_098.jpg
a9af9227253674c7f28b7c385b5d361d
5455b418130e35f7d5ddb0250c9e6cc85ccb20cf
101220 F20101219_AACXPJ thomas_j_Page_118.jpg
9e8c2775ef2c52c492c4572fee7bbb09
ec150bb624475bd38cdaa1cfa3ef349c597ba297
102256 F20101219_AACXOW thomas_j_Page_101.jpg
d73582140610805276ee7cd00981ef4b
78a6f7d038f75c7a8286663129888350e405a5b1
97911 F20101219_AACXPK thomas_j_Page_119.jpg
8db762ca058b3f1dd650a86d028d7466
42bfc1750b783234fc680a8b414137f173fa76fc
104388 F20101219_AACXOX thomas_j_Page_103.jpg
ab2d15c42430fc480d9e925ba9b85afc
f2666e1854d8f0b327b5ea8b4de06e39c691190c
133670 F20101219_AACXQA thomas_j_Page_147.jpg
54cbee6b1e63353443b30dfb58198161
d0f92ed5e2694eae99738900194279f06e24fe7c
116741 F20101219_AACXPL thomas_j_Page_120.jpg
a3d77934123bf277826992d61c356804
d254950cc6da6e29ec3e479b21c6e74630a06e1a
98423 F20101219_AACXOY thomas_j_Page_105.jpg
7109156e923e9d33f4319b950e160564
d1b2a126f398a41395792eeeaa0003c253d2e3cc
91600 F20101219_AACXQB thomas_j_Page_148.jpg
17687c14f2f131d6c1d5fd29f5804c6a
4f25144b75bd55f872754f6747dd58c9e3641806
91804 F20101219_AACXPM thomas_j_Page_122.jpg
24bb3016746740e0caf57a376337fb10
a98ad2a4811839ba1951e58cd1d53b529d7aeb4e
106844 F20101219_AACXOZ thomas_j_Page_106.jpg
aac13586561db7a910217bba57f8c4fc
0d4b21cc428196cff145192ae8b11eae826f9c11
35810 F20101219_AACXQC thomas_j_Page_149.jpg
7388ed765476448fea562733152c79b5
4b7e5fe0b5a497baf3fa094c9b5dec556e34d836
78815 F20101219_AACXPN thomas_j_Page_123.jpg
5932d4505726dd10105f64ff2184ab57
6ddeb323f8b7a078b1b1e374f925c2dc4991c794
28856 F20101219_AACXQD thomas_j_Page_001.jp2
cbeb343748fe9d4109b7b587775136a1
087e235af3e59fbfb757a6fa38e469243c6b6598
106969 F20101219_AACXPO thomas_j_Page_125.jpg
82c1661c1b99304b0db595bd9030d622
6b952762d259af79ad652912109470bdcddfa82e
6021 F20101219_AACXQE thomas_j_Page_002.jp2
82bec5e636d887d4be79afcf39889543
b019acad658d7541e409523e54fce99b0eefd7f8
81975 F20101219_AACXPP thomas_j_Page_126.jpg
974087c0e34b756f53fc1234108f81c5
f045484e028b19c5bf3eb0cd86e8fa2ac9d6b5b8
9875 F20101219_AACXQF thomas_j_Page_003.jp2
4e2a8eb8a4d02fa84498f062fcb13820
124be745b71140c1d2b93de5a229bab7f5abef43
76144 F20101219_AACXPQ thomas_j_Page_129.jpg
8455029cf4fb5f710602aae93dc8e747
31b73b01206bb94b21f042415e1f0827a73e2257
1051970 F20101219_AACXQG thomas_j_Page_006.jp2
4043ed119d022ee9f107a37f32be0322
e4aa065a07bd2e4eb3245c5177e27e83af23b6e0
77455 F20101219_AACXPR thomas_j_Page_130.jpg
df057b9fd802c2d8ebf2206f404f7c50
f63e015b915a015136ea3bc89372676d163080cc
1051969 F20101219_AACXQH thomas_j_Page_007.jp2
cf8a185cab753d5f9b5b52d0aad06a26
2fca6544f20d9925a76ad753b356ab40354887c3
95072 F20101219_AACXPS thomas_j_Page_131.jpg
292e21c66e9171919a94004ee81c006b
7f7df62fa7b072637439cb637cfe0b430806cda4
103647 F20101219_AACXPT thomas_j_Page_132.jpg
3ff322347c0c104d647443b60d1db68e
e81cff18de62790723fff8e1f39785655d1a2a20
625290 F20101219_AACXQI thomas_j_Page_008.jp2
8f8cdd360f9cf1b441c5e6556b12f0b5
bf38baafa279f1ad40dfced69876fa0161fa4667
100930 F20101219_AACXPU thomas_j_Page_134.jpg
4466037670f40eee4ac2218090b26c53
6ba9292140b0aeda18e09ec266eb8cdc74404ec8
1051980 F20101219_AACXQJ thomas_j_Page_010.jp2
c32ee60220fe7b5098bc1bfefef31d92
a2cbc2e1d86228fa432c5e4514ba98445f96b96a
79824 F20101219_AACXPV thomas_j_Page_135.jpg
fe85c013cd9f210d023d9fd45a38df73
9c7a23caae7a403bc07c97e9047537326c8dff1a
87341 F20101219_AACXQK thomas_j_Page_012.jp2
3351cba1ed4297872fd710f9cb354e78
fbf401b8d38a6cdae13ab3153839e29b03595b37
42397 F20101219_AACXPW thomas_j_Page_137.jpg
691c51a6e37ac85b44bc472522de8adf
4e532fd8db4780f82dce5648d864e667db1b8246
97072 F20101219_AACXRA thomas_j_Page_044.jp2
a21f7bba44974af6cc147ee1cffeb22d
1d4c51fca1c1dfa1090aa454e3de1be63f5d43e3
70418 F20101219_AACXQL thomas_j_Page_013.jp2
3181849449eb4306921f7bc49709169e
0ca76307a94a30b6a8afc9a9656b6256104fa5ca
134543 F20101219_AACXPX thomas_j_Page_139.jpg
6f2879434aedb155618a1fe8eee82333
4b1548d51f2fadce683e5cfb63a6933431349544
111013 F20101219_AACXRB thomas_j_Page_045.jp2
32604309d937cff40a616c6673d6d2dd
d00105365cc9071a864041c9dd7f2fa5a6cb989b
108366 F20101219_AACXQM thomas_j_Page_018.jp2
9d05df756a7bd2dc10663ed0eb729503
828c3b995b9db3fdd6687dabce43236d9aa0c2b9
121529 F20101219_AACXPY thomas_j_Page_140.jpg
6ccf17ecdc6210f01433b3f902e1541c
8284ebf5a8a6ac28ff1ead33e0d4c5d4f934ea1d
104978 F20101219_AACXRC thomas_j_Page_046.jp2
0f581e0ef1c04db228b60ddd4733f180
87ee377be69cf9cad60c71666f3f64e707eae187
110361 F20101219_AACXQN thomas_j_Page_019.jp2
1c862a39e7f68329fa9e2cb163a3e957
8da7f9b8e1cf472c2bcf5cee70857129f03f6bbf
133856 F20101219_AACXPZ thomas_j_Page_141.jpg
f0d827e21dbbe486e5f45ccfc0aadeb1
c0c882a90ded28f707db44a4b78206e16fa45620
21212 F20101219_AACXRD thomas_j_Page_048.jp2
a457865ea4efe7ffcd0762a9dd0f79db
9cace376aad04eab3fa979d23d2c6ac896de0928
107512 F20101219_AACXQO thomas_j_Page_021.jp2
ed189bca9c86a7e7d4c8d75c88d5e031
66bc05c0b923daffa7a6c5e122c37d73a2c74e17
109940 F20101219_AACXRE thomas_j_Page_051.jp2
be1e39c58b975d614af9920f2a36db59
f1bc7a0d3d4cce02232e737eace250d0ad9503cb
91471 F20101219_AACXQP thomas_j_Page_023.jp2
5bd26457444fc20a9bf2e3d9d14cfad9
b864da9355f052ac53967607e608369e63268801
116017 F20101219_AACXRF thomas_j_Page_052.jp2
3333cd2c3519d53197bf41bb834c3109
84d1658fc6923ba0c1ead1e3768ce5f6f83a1bc5
1018669 F20101219_AACXQQ thomas_j_Page_026.jp2
78b6e1a5455909962e0d244dbfd6c47f
e76c110d4993ae70faffe72b8804100fe368752a
111980 F20101219_AACXRG thomas_j_Page_054.jp2
9a299b1b86c640032b9b579eba762762
02d2c3ae3aff109fff9431a3b6ad93dc3bfa705c
113401 F20101219_AACXQR thomas_j_Page_027.jp2
24a9c63b8a0bd0c65643a6a907fd4239
0ed188e7dca312db2a754d45dd06b4fc30898ca6
107198 F20101219_AACXRH thomas_j_Page_055.jp2
a6ebf5c020e06a54c8048e13551763ac
c653ef14c1524b67b3248ef0ec4121bba6a9c17e
110341 F20101219_AACXQS thomas_j_Page_028.jp2
abe94f95d1683f4cee5471f911f89039
382a06b0f5af370cc139fc895299a7b566690a13
102989 F20101219_AACXRI thomas_j_Page_058.jp2
8f29446bd65b65d183bdbbe308c53678
a886c25d3f34a497704b7afce173b49004b54247
104554 F20101219_AACXQT thomas_j_Page_030.jp2
2620305bd15f595f1f6b3a6a376ba91a
5df75310f57a9992cf066f9289bf11a8e6dfeef0
92029 F20101219_AACXQU thomas_j_Page_033.jp2
894baf72db34212921bf3373ee2b14d9
f39313dae52f93ba74ca5aa677a1f869df8b5d00
781780 F20101219_AACXRJ thomas_j_Page_061.jp2
89cf8e01d563da4879badeecb63c4e7d
24c3f5f466bb934c8bc6c2618678caa7a4f79033
110227 F20101219_AACXQV thomas_j_Page_034.jp2
1d3bd40248122f6fa7ed7ac318b0332f
027d9fa7b613f5591d2f65ef20edfb9b757f58dc
104654 F20101219_AACXRK thomas_j_Page_063.jp2
5b2e5c332fdc4925b0c4788b739f4a0b
76cd8ceff4af74cdc7677eff72c176f446214cf4
99969 F20101219_AACXQW thomas_j_Page_035.jp2
7cd9f1105fdb90f882155bacda66a73f
a20bc5c6c58540a369eaf5a7b02afceda2adc54a
110392 F20101219_AACXRL thomas_j_Page_065.jp2
d1d54eaf1902ec1b48a73bde170f0462
ff8af5315693f2b79c41b14a669f8cdf3aeb4249
113224 F20101219_AACXQX thomas_j_Page_037.jp2
4b6131a9f38b4524191629ca73c39273
21256a1aa2dc10c9ed008cf2074c8d91c36d935b
1051953 F20101219_AACXSA thomas_j_Page_085.jp2
b7a906bd55d76769cdcaa16a456e7bef
4bd40da28c5d3eaf8bfae2ddb45e9d695595ad8c
109276 F20101219_AACXRM thomas_j_Page_066.jp2
0081585b2169a29cb3358a7c2c51bace
17a11858ab187c53f4368bae606b8ed77add747b
81105 F20101219_AACXQY thomas_j_Page_038.jp2
bdc8d5f0988d90d7300c458373f1d919
48f998ac7fd10bc08de26dc1414422bab900fde9
113216 F20101219_AACXSB thomas_j_Page_086.jp2
13c0e7ab7eb21db74a504683d4d24ddd
189128463984c787976c215dbdd6d0b8a3cc0824
75685 F20101219_AACXQZ thomas_j_Page_042.jp2
16647c579e5330d7b7c1e8ba8eb1e50c
4a79bfdf614fce485927302018b19e0ebba6c9bc
850667 F20101219_AACXSC thomas_j_Page_087.jp2
25be1faf309902496c885140dbcb9994
bee2287ee07be08e3f3216e90239d2cfdb7c70b6
1051983 F20101219_AACXRN thomas_j_Page_068.jp2
d9ec0b02ab75990022edb8c586d8103b
3d84117f8ad421c2fa48548321a13ee729798189
107952 F20101219_AACXSD thomas_j_Page_088.jp2
06fbe441fa81293f2ff1d75eafbb36bd
46bdfac8652826c76ea81a74ef7d94f1fedaa7b5
107612 F20101219_AACXRO thomas_j_Page_070.jp2
97cb28d50bbf81873f829b4e73681285
79552da66f1f1efb6b81e67e6671348cab756a60
892862 F20101219_AACXSE thomas_j_Page_089.jp2
30de15945b699767d6f0dd11307191be
faddca23a1915120378ad301b7f3963a06a6bcd4
83282 F20101219_AACXRP thomas_j_Page_071.jp2
22c7905bf62dedfc13b691b7b758e0b9
abf5b3cc7c446da903c6dadc96241b8b351ee0f2
109525 F20101219_AACXSF thomas_j_Page_090.jp2
bc6f91521284c97364819428b9c6e37b
353c8f86b04ac2882a8861d9f4f01e15445602b8
1051978 F20101219_AACXRQ thomas_j_Page_072.jp2
370b0b2dc47fc95c6dba648ae0593fbb
3bd331650d4c4161426a287f0612040c34518606
867122 F20101219_AACXSG thomas_j_Page_091.jp2
d6803a125cd34ec7ea3e3cf4b7ec73f4
8d8a6de67c2177437dc72a8c5e5261435093d541
105266 F20101219_AACXRR thomas_j_Page_073.jp2
45bdb3c925add77a5d03d1eeed5f42f3
7969e707bcfab7121168bb7750a493c4f5984303
73550 F20101219_AACXSH thomas_j_Page_092.jp2
c409214883bd442dc0ecbb8237aa73c6
9aa6b16bfdb0e4abe3b20527d6db5f33e7c32ac0
103142 F20101219_AACXRS thomas_j_Page_074.jp2
aa8ff5c319b56915b408060abc8822e4
3fc6015041ecb4fba1bbebcc271d4b0a0e703ede
73874 F20101219_AACXSI thomas_j_Page_094.jp2
82fa0c9aedb7d688bf94b0ace85d1a56
05af677032115a41a96d8525bfe4fc6ed6eefcb4
96734 F20101219_AACXRT thomas_j_Page_075.jp2
63ca3f1d783ac7c91cd2b4c3275c5f3c
b1551eb68d9a2011113fc52817606802a44aa1e0
100804 F20101219_AACXSJ thomas_j_Page_095.jp2
a43e0deb282c8ccbd19256d335b2cee5
8f8fb57296e45172848837d9b93423e8458e2b89
887307 F20101219_AACXRU thomas_j_Page_076.jp2
cc664a2468ab16af6f08f1abbabc8707
1860de17ec6ec6a9e6d1758b8087b8ae41976ee1
100149 F20101219_AACXRV thomas_j_Page_077.jp2
f5dcec4b38e85f8f94ba9f797961c814
dd3a4be6e18071946aee46f0384712184236c100
112449 F20101219_AACXSK thomas_j_Page_097.jp2
66d41415a4492b4bbabab8ef77710a45
341cdc6d29aeff553eeaf210a09bfff4eaa50ff7
824726 F20101219_AACXRW thomas_j_Page_078.jp2
bf180f60d1ad755d2c7a9bb1d85baf1c
c83876eb20c064aa0ee859b95e4c6e0fa7876960
102755 F20101219_AACXTA thomas_j_Page_119.jp2
6ece1e4cafb9ad558a31db23d2d4681c
1fbc1d0a9423d16545263ee906beb1d222776dd6
104087 F20101219_AACXSL thomas_j_Page_098.jp2
f40e61d025a872a9fa38b3b9c8dfa8ed
c4ebc48080536f9bfd7091466c3411489031bb0e
112945 F20101219_AACXRX thomas_j_Page_080.jp2
77ad933363365cf0c8bc4f27dc0b69a5
c1a64758a73b60131f1ede773704a58f99edd77f
F20101219_AACXTB thomas_j_Page_120.jp2
b1c7b69d38ddf14664377c0c9cfb1d40
a400cb0e10314d7a3432d4e986bb51245de7ac91
102737 F20101219_AACXSM thomas_j_Page_099.jp2
1baa781ad649b89fddd673de5e143f8c
a856d3b925b3172e39b03248efb9e251383ecf81
91234 F20101219_AACXRY thomas_j_Page_081.jp2
c104eeb9520e8fbe78616c8a6d35eab4
f16e47ec5e3a9c7fc73b71a93e0699c7066a2734
110574 F20101219_AACXTC thomas_j_Page_121.jp2
bb30c2df8c858b93f18cbbf8c717845d
c5149b0d0ede8675e61c8e44bc840b36c0eb6bc3
628555 F20101219_AACXSN thomas_j_Page_100.jp2
73847b3ee68aee06bf8ff04da53780ef
7198667f769143085d6db14957564bf8a2bef40c
108200 F20101219_AACXRZ thomas_j_Page_084.jp2
3589a802885306fb565122b6bf1bc8f8
4e24527a019974b4998535a543daf09533ab416c
93603 F20101219_AACXTD thomas_j_Page_122.jp2
783c561bcffaaa50a42f62584584a552
b8f930f9fe0e6daebea0c7c847c9e48ce3415946
110312 F20101219_AACXSO thomas_j_Page_101.jp2
21d0302a3da63c42e70023a95de90319
f90688abbf667413daae42a4e39cdfc9b348dcc7
802207 F20101219_AACXTE thomas_j_Page_123.jp2
484eaa617a329e2c415be7180ac39c50
e35701b306928093b7e22ea846a2f996fc40ace3
103899 F20101219_AACXSP thomas_j_Page_102.jp2
ca61546932b0ded5d0d89d9426ff66ac
16c270a7cc0b97190f2a1d7669ce0e08f219e931
111227 F20101219_AACXTF thomas_j_Page_125.jp2
fd885916fe29a63f663332578b73f701
f0c3aebdae6d1dd80297fae157de4e6c1b6e8c2f
F20101219_AACXSQ thomas_j_Page_104.jp2
6daf7b8dc90f25b5179af855727ce08f
9498bfca2599041d0600b118c52fd473e412c9be
896167 F20101219_AACXTG thomas_j_Page_127.jp2
796e386cab1bbe6d0d2cb78cd154bc15
764f7d88d285f14219970eb7ff25a14efe7f9ef1
107008 F20101219_AACXSR thomas_j_Page_106.jp2
9b14b5111f9032869d900bb7ba68a696
95a58af3354fd57c7b95027a510536c56a21c76d
109067 F20101219_AACXTH thomas_j_Page_128.jp2
4c2f7e831fcc458873d3906977194d6b
fbb8112bd6571b02c41096b8f9934299882898bd
78349 F20101219_AACXSS thomas_j_Page_107.jp2
a1b5efafa8ef2050851a84e0fba241d8
e1a287b814958cd2dee3182083bcf035f0a90df7
781956 F20101219_AACXTI thomas_j_Page_129.jp2
346d6bb4a458467747004f2611f0c44d
f4d16b380b40c7d9c3438d65f87b65a4fc475f55
1050078 F20101219_AACXST thomas_j_Page_108.jp2
9f9a397db212164697d68b0874e9e09a
6b07d4e9439afbb985d95e60c58f0f89b9dbbd4e
81620 F20101219_AACXTJ thomas_j_Page_130.jp2
8afbbc1e4a26e36a77fbec303d1af111
9fb8872220c91e741ec38d02612cca268dcda6cf
105041 F20101219_AACXSU thomas_j_Page_110.jp2
a0d809917f0214be2492c9b9f6a69564
742c59b4f2de0b428fb37f0a7d61a022f2692f1d
109063 F20101219_AACXTK thomas_j_Page_132.jp2
f3f86ef517f77741770f6615a3a2b564
bc85681811da4ebb039c2190285a007235990159
98970 F20101219_AACXSV thomas_j_Page_111.jp2
7c76977a058af673569dca75ce188a3f
be9d21c8e0663d44ac9fd082bb60f0bae1e729b3
102558 F20101219_AACXSW thomas_j_Page_112.jp2
a9a1a5a629879f415c44fcaf8decc694
d6dbececdfa71eca9989d4733dc8c85633686121
108665 F20101219_AACXTL thomas_j_Page_136.jp2
5ae3440169e45e64f043ef987efaa360
58c110e14cf74cf8f634dcb449c7e46edd6562a1
807296 F20101219_AACXSX thomas_j_Page_113.jp2
abc4c584deab646c19d439ce06de118b
0823c8d4fd72ed5ec7804d8ae6f18bd0c59510d7
F20101219_AACXUA thomas_j_Page_006.tif
ecdd56c7441622afd40a8dd9c9a595e4
988cc59ebc0081a480e95065b32d7fbed8b03694
44547 F20101219_AACXTM thomas_j_Page_137.jp2
152f9e763f70337e71344ff729490ae9
91fa245b259d749a24c34dcca76005e06fe078f6
103846 F20101219_AACXSY thomas_j_Page_116.jp2
14bd818a924b7163bbc60d147665effc
4fedfc0d67e27d86df992a9af720d0ed25e47790
F20101219_AACXUB thomas_j_Page_009.tif
3357bc6614ce2f723b364bd5c1fa9a09
b5563a855102eec86dce99292bb1f074cfbb66b5
115030 F20101219_AACXTN thomas_j_Page_138.jp2
03903d7f847682753a2699775e0adf75
034c92dbe9f682741d53402d3076d9ed18278402
107422 F20101219_AACXSZ thomas_j_Page_118.jp2
3c3cada6358d9084829051c6ed6efb20
3ef26cd6cee21042944a66011293ceeb421f7b7c
F20101219_AACXUC thomas_j_Page_012.tif
08919bbec14183e1d31837827ee2dffb
d869f4e7225d8fead92d223e39b543a37c094699
135958 F20101219_AACXTO thomas_j_Page_139.jp2
db5c6ca962b312a45b2fcf96ebe0a19f
1e6b1c606d329e03809e7df0064f6f7a9b3ba3de
F20101219_AACXUD thomas_j_Page_015.tif
303370db71a9c2fe41f9912bb66ac60b
48718e0219880633828e2534ee91258e846840d4
127759 F20101219_AACXTP thomas_j_Page_140.jp2
dcaff0107bdb5ab68f60db4de63a0a84
d9cbc085f0ba591b8514ffa91ee23bdb4c93a43b
F20101219_AACXUE thomas_j_Page_016.tif
8fed37bd29a0bed22dc09718230cca04
ea2c38a8a0e2479dfc3dc62b8d33fe9eeed4d31a
122754 F20101219_AACXTQ thomas_j_Page_143.jp2
c777d74b9fcb5b5e67529cea9ca63bce
1e39cf04aa0512d039ea66733238c471bd02dd8a
F20101219_AACXUF thomas_j_Page_017.tif
243c67c802f04379a602b9c1b9749776
58f9bdded14d0cf82ea68282e8052f28959a0d20
121540 F20101219_AACXTR thomas_j_Page_144.jp2
401414586bc1639d62b7e2067d596845
ea7d7504eb2503fda5fbd49a77c88f343a8ef388
51301 F20101219_AACYAA thomas_j_Page_088.pro
b120cf6eb14dcdcb91cbc582d50c7e56
fdddf25e4314ff00c52ace9319eb036f645057bc
F20101219_AACXUG thomas_j_Page_020.tif
284281b7bb26d3e6f54a97012d5bec14
5edb392daf8aaa87799c6eee3d810d26d62b471e
129563 F20101219_AACXTS thomas_j_Page_145.jp2
f4763313e3aafa20e2c1d5e99129774a
285b75d87df0125b629f5c4d49ed3f1d7c3abb40
30958 F20101219_AACYAB thomas_j_Page_089.pro
35c5bb12185dbbd5cbbadf254993db15
5e8cf51a73129e024f5716e9fbce0d487a6dad6e
F20101219_AACXUH thomas_j_Page_021.tif
3d26b2f681d8999ee3df9a5ee2d8350a
3c4536baf745d46b074758d172c8bec3cec76a03
132421 F20101219_AACXTT thomas_j_Page_146.jp2
37f451f80ff2d96b792889dbdc0a52eb
0079f53d609bdf27928c5f3a46b0c60d771789a2
52493 F20101219_AACYAC thomas_j_Page_090.pro
54aecfe9035c903d23718c10068571d4
50625b18f618f392824f4f4d65b6c185d2adae64
F20101219_AACXUI thomas_j_Page_022.tif
86d80d5489b84651e44449ccd5729d6f
6ecf93250bc99a87301a70d30eec28d448dd6598
136308 F20101219_AACXTU thomas_j_Page_147.jp2
7451bbc6074dc3f2847f43d046ca6ba5
0785aa7e8615b4ac18ac5f2afaea44fc48bae893
32324 F20101219_AACYAD thomas_j_Page_092.pro
3d57ae3f39e058f7fafa5f5e57ab10f4
c1c78e855f361e99e4e14c60c5f484e64e592862
F20101219_AACXUJ thomas_j_Page_023.tif
3ee2d446bcf7562d59b5b0a5ca1514eb
8a7d4618276014b72c067b68b7d6e956f8a87b90
91423 F20101219_AACXTV thomas_j_Page_148.jp2
c28ed1404811ceb92220661f16e8f926
30b4951e36b03d92d6850c4645ff14783324000b
42467 F20101219_AACYAE thomas_j_Page_093.pro
710b4d50f977e57ffadc359d411048d2
5a082bffc5c55ec07c66842bbbb7a22d0130af4b
F20101219_AACXUK thomas_j_Page_024.tif
a5381058bb201b70dd6e31abd9d8bed0
8048b6c78c2e7b08d24a86ac7b701dc09f7561c5
36261 F20101219_AACXTW thomas_j_Page_149.jp2
964c7aa7ba2e4811e8c2abda31d6eb2e
b8f98877f24cf186cae64ed75a8f7d810e7ffe07
32792 F20101219_AACYAF thomas_j_Page_094.pro
25e54edbc2f70e9324ac0ed23bd65a34
a5fe012344076009ee133994fb267edf5407c41c
F20101219_AACXUL thomas_j_Page_026.tif
4fcb1019198b64128200850082c9dcbe
bfec247ec5c723165db6853d2b4c8731eeeb13db
F20101219_AACXTX thomas_j_Page_002.tif
eff8d20e5b9859f4d066ba7ae6e4844b
4e6f5a2b9221c2f11da7f334ff668d7850d0d957
44762 F20101219_AACYAG thomas_j_Page_095.pro
fff1ff1d723f4472129834cddf097cee
9c08f0fb27afff8f85c13e3b7669ce01ead32b55
F20101219_AACXVA thomas_j_Page_047.tif
d8ce2cb9279a04ef52588fc802531f7e
ab36f2d4c619d99f5fbca1414ba57849da305a2f
F20101219_AACXTY thomas_j_Page_003.tif
1b7e5ad3b88d752b790d55e4d5430f9f
1a673772988dee09b5e5366e1fdb6ff85a59cdf1
49766 F20101219_AACYAH thomas_j_Page_096.pro
820e0d31042be01994896e5898db6032
180b8e71564f6ae5295cd8c6dfcbde90f96b7f15
F20101219_AACXVB thomas_j_Page_048.tif
7e17e962dfaa2b79f330cc2b620ce6cc
7efc82f14fe05ba84b5a115f0cae84b2b98f767c
F20101219_AACXUM thomas_j_Page_027.tif
4edbc002c8605c7a1bb7c2d6ccf8c5a7
627f9994fce85b303aaa9bd8f1cacc0983cc40d1
F20101219_AACXTZ thomas_j_Page_004.tif
c7970ff813544a07f173a038ca0bdfb1
f582cd836eb56c824a7eeb5d727ccc0921f9cc54
47043 F20101219_AACYAI thomas_j_Page_099.pro
aff1e611864f969b4a77272ade492b60
94c36bb6016a6e8fde8e74810bc1055c2716f578
F20101219_AACXVC thomas_j_Page_049.tif
1b614c2c4d04fe66432ed805d0d4ab0a
d5910233fd2c7d3cde483f3af49e44069d34b97b
F20101219_AACXUN thomas_j_Page_028.tif
148b1974dda534ad0726e15c0eba8b34
d7c36ec5890a939e75d31515a8c313930cfef397
49219 F20101219_AACYAJ thomas_j_Page_103.pro
2519a3785c388952a84e3af5bbed88b6
1c7d8ba4563607867a93e67a382f6f2f5e9e46db
F20101219_AACXVD thomas_j_Page_050.tif
65e4a9ebc4b38e938ae95a554b5e4e13
69f85fb746a6daa97ca839d766ebe6b925251e49
F20101219_AACXUO thomas_j_Page_029.tif
dfdb7a663b50f5c172cec77106b1c680
6a92bd64fdf68d039f181e4bd156d8bdf66cb4be
48001 F20101219_AACYAK thomas_j_Page_104.pro
c281f3c2fc7443b3e2889e4b46d17ad8
e769349c611b8229f11b0d8dea80c78d33507e2e
F20101219_AACXVE thomas_j_Page_051.tif
621481786d8a59e6994a35acf5fd0751
f99f058e068ad39a983d64d45072b6090213f3f5
F20101219_AACXUP thomas_j_Page_030.tif
3910e18ffbacf2720969f9b80206ee72
84ef3e3e2347c22646e4fca86eef98426e1f127a
45542 F20101219_AACYAL thomas_j_Page_105.pro
2aaf39d0c5f250b2d6df37acf44f726d
fde9b1e422e9de3bea23fae7b4e79e4fb09bd2ad
F20101219_AACXVF thomas_j_Page_052.tif
ced91e825cce8d5b655dc35fa9dde3dc
b1d59d73857c12c32b38647063df763610c13423
F20101219_AACXUQ thomas_j_Page_031.tif
363bc3c26cfbde2744edd75be627746d
4946f9a19c4712104ebb9c753c372b902cf89fad
49674 F20101219_AACYAM thomas_j_Page_106.pro
9dd78c1cca27c0782a4150f10c693d1e
2b5333dbf5321b0398e32edd5086950b40dc0877
F20101219_AACXVG thomas_j_Page_053.tif
5e04720644094510c1d83bd3d6b02932
3fb519db438f2bedf6552873d762fa759a8070d1
F20101219_AACXUR thomas_j_Page_032.tif
5ec51dbd963ce848a855ff6b43b6a64a
e4b3477f405d2ff44c839671505167241c92b92f
45711 F20101219_AACYBA thomas_j_Page_122.pro
b067f93a612a7e0c63c1354ded318645
61120271edcccf00d85017ff9d39d2c5b39adde8
33435 F20101219_AACYAN thomas_j_Page_107.pro
b44bf5db90edc1961b89910f02ef7e3e
6ed94a4cec327293140f4301c60cea324358022d
F20101219_AACXVH thomas_j_Page_058.tif
bcd0f86c5dc1bcca2b4677d36b23fba4
c432c04ec0e8ddc98e4fa618fb2b21f97a7cb284
F20101219_AACXUS thomas_j_Page_033.tif
f9e9ab51892589353229f22f41596dd4
30fe8eab8fb817349740f37e8f4bf7e4e981c60b
36005 F20101219_AACYBB thomas_j_Page_123.pro
cf47583d18076208ad66e15757cc395e
5514690f6a9af143ad31c0ad15e5b17d88cc81ac
47769 F20101219_AACYAO thomas_j_Page_108.pro
f61cad0241eb6afb4b8375d00cb41457
c31b9383e61644ae8504ceac5cf6e6d77ade9ee0
F20101219_AACXVI thomas_j_Page_059.tif
209f99401eeaf8c335034357a414640e
c0cdf41b2ae9243d2772a273f38e943f67abedb3
F20101219_AACXUT thomas_j_Page_034.tif
efffef1cb51dc6492293d6734b0e4630
67f30b2b1b37c6f6ca503619787d68f2836d009b
35083 F20101219_AACYBC thomas_j_Page_124.pro
b78c33c71e861439f6b1e3515f667549
7d84ad7f3c32b1ab146ab72c67c0c89cc113ad4f
48738 F20101219_AACYAP thomas_j_Page_110.pro
46dbd31da2338ae82cd1036d3f854322
271e4acceb80b3bc18860d8b302e27477ca14cfb
F20101219_AACXVJ thomas_j_Page_060.tif
7439a99dc6e08ac2fcdb6d9dc03662db
db9ece5ee2cd03aa474d8d2d1cf072750863e8b9
F20101219_AACXUU thomas_j_Page_037.tif
d44536c0a774b0a3798446c13ab39f3a
4883cfcd99ec55a7efddcc447d0204553b06551e
36277 F20101219_AACYBD thomas_j_Page_126.pro
a26593d835224243b98ebf88ba5aa0e0
24944f6cfd366c8d5009b797162db203eed9c3e7
45414 F20101219_AACYAQ thomas_j_Page_111.pro
d712b56d61e15484b1dbf308619e02d6
54fe62779349d03d67c607f0238fa168c74952fb
F20101219_AACXVK thomas_j_Page_061.tif
95c8035060156f1caea80288bb73e404
09e041beaf9195750aa60895444bb9ab21283483
F20101219_AACXUV thomas_j_Page_038.tif
1fc30e1d5e1d20189057750fcf8b215b
b84972d875d5ffd651e5b2aa7ef36ea1275f8d35
50767 F20101219_AACYBE thomas_j_Page_128.pro
c65ec7d61464362a0cca7c736fb888d2
a564422253726d0834d46089ac472b470c38caa1
46995 F20101219_AACYAR thomas_j_Page_112.pro
bfe84bfca053c1f13e3ba087a612fd8a
8966c5863fbaa9c4749d5073c79aa7d23932eb22
F20101219_AACXVL thomas_j_Page_062.tif
037344a8b773468e51b076a59a05cce9
046fc91770ef2b207b3ed5c0ed47db5a9cd0fd4b
F20101219_AACXUW thomas_j_Page_039.tif
fd8f26c06539ae36c3015fd39cc06c38
76bc8ea3cbea8a7ee73fc86a7f6a4f4679996140
34134 F20101219_AACYBF thomas_j_Page_129.pro
4f0cd7e0d4f2483cd54079a969af5aca
5c074b91ad479f7916233872d8cc2fe316685675
F20101219_AACXWA thomas_j_Page_087.tif
cfb75facdf8476764fbb0b9579741062
88ddacb1de2cd56064cdac4cbbbf5899d5b53476
34275 F20101219_AACYAS thomas_j_Page_113.pro
d9490040a2bdb34935ce2dbc6740e6a9
923729fb3b4e90820f06003a57406515888a8c82
F20101219_AACXVM thomas_j_Page_066.tif
dc65738905972f771d7308978a3e8d60
738152f5da69aa5ec2b4a2e385e01627ee7dad6b
F20101219_AACXUX thomas_j_Page_042.tif
c5300d292f09ea484bbf7147dc410dfd
53721f604bda142e4af20ebfad400dc8e4a745a7
36394 F20101219_AACYBG thomas_j_Page_130.pro
9ee87e96fb3bf66eee1b7c93669fafb3
928de57eb9bf8442e8be7179cb30374d06ee981f
F20101219_AACXWB thomas_j_Page_088.tif
9d28beb16e84e874f6842da138747371
02d2e892afce5c35acaa00bb830d61592eb6f825
52353 F20101219_AACYAT thomas_j_Page_114.pro
8e812a9018295bd6675d23d0d8abd145
e750e7a7b1e9d442d74b6de37b4953389b6da27b
F20101219_AACXUY thomas_j_Page_043.tif
63f8a26b399a33ddfb5b07b5a28da9d0
f0dc2c9b017e1b4d3a683b40d69d2d307ff87037
50048 F20101219_AACYBH thomas_j_Page_132.pro
92d95e83f5d76b4e44b9ef25d00a8337
b65f6c57e090926208d02f97ec0ca2d42d0cdca8
F20101219_AACXWC thomas_j_Page_089.tif
df1abe0bcad52ba99bb0ad3a9db6ed4b
a6d740fbeb070929cb60c12617cca4a3706430b3
49634 F20101219_AACYAU thomas_j_Page_115.pro
1de4bbeccf678db7885509549493a64f
10accaa46733ced4095541259f23122bc61f29d9
F20101219_AACXVN thomas_j_Page_067.tif
3a9e1223b6cbbb307de5aad45acda003
08ad2da9fdb8be9f944593a16b04cadad476617c
F20101219_AACXUZ thomas_j_Page_045.tif
4581eb2dbd90ae8d5d5691fb0d75ef95
9868238e4c583585fd198b07551045c3d4aea0f3
49852 F20101219_AACYBI thomas_j_Page_134.pro
364e1a7f68dba4cd41780671e4d4caaa
3054e8d35e6e860224f1bfca4778b69d19235d6d
F20101219_AACXWD thomas_j_Page_090.tif
8cfcae3a906a27346c71ed86b4f20309
eef52bf4f749626765581882016f393b5a090750
49241 F20101219_AACYAV thomas_j_Page_116.pro
3352695b4be9dca7dc197ec7264bebeb
a104def3d0ef3012eada1675f8c760dc1c1c9267
F20101219_AACXVO thomas_j_Page_069.tif
292e1d7d4ae438f782737090ac8be282
0dac998c3a064a9bfff3870df55ee0cb595b3827
35081 F20101219_AACYBJ thomas_j_Page_135.pro
5055ad3f74437b4b699e28ba40106f40
51ec7a4c086084d2a08adc9ebb79ebfb12959f19
F20101219_AACXWE thomas_j_Page_091.tif
de63e9dc13a20c6107e355e512bf3ff1
079ff020c447878f1e2997c320b075205b82dc23
46252 F20101219_AACYAW thomas_j_Page_117.pro
8084840cdf3a3fc1e88150731206845d
7acfce7f1baee82640d77d119c1d20084b9bb11d
F20101219_AACXVP thomas_j_Page_071.tif
c0bfedb0690c2b5c42e2441c1a28de02
c17578a930a2cd5dcee79f479a49a3b105a809c9
54937 F20101219_AACYBK thomas_j_Page_138.pro
d9f4c97d15d3b9b435417bc678415e39
1c462e49d542193064e8adac1e700afe24b28628
F20101219_AACXWF thomas_j_Page_093.tif
ceda833d662e7ecffd3b96d14d604645
8bc672cb3537a2664207abd1a3dfa3786750b55e
F20101219_AACXVQ thomas_j_Page_072.tif
3b44e00dc6e264ea99a0056793ef8880
4416ceedc4e8e2dd054f06204b5317440dbc25de
65971 F20101219_AACYBL thomas_j_Page_139.pro
d5c4c87f2d6df7b210ba5455aad876f5
1ad46289b2fd5ee2190796f4cc47ef01049c8a00
F20101219_AACXWG thomas_j_Page_094.tif
040caa9789b1c8abae15da03a3b2930a
2919e6c06fde02ce795b908349d0dfd5914fc2b9
47945 F20101219_AACYAX thomas_j_Page_119.pro
c9b7023864f4492678361a636badb74f
922699e03f4428952fa5dd6b7c80fc13ced8445e
F20101219_AACXVR thomas_j_Page_073.tif
a8aa86b74f132d1b98aadd60eafa6fcb
af348ac4782ac80c4e3b524bf026370ec81b3124
1701 F20101219_AACYCA thomas_j_Page_012.txt
d9c0567f558ba9f863bd60b222406eb0
3febf17850aeb44bc66ae545a748e758d4c84a52
61547 F20101219_AACYBM thomas_j_Page_140.pro
eec825cf360b1c8ccd2c135a6afff879
feda4abcf088dff612535e3aad79540d83a5ca50
F20101219_AACXWH thomas_j_Page_095.tif
0159855d4eeb52d35c4896d54743c038
7b5017c4db965434dc8920a2cfffc803323ebb0e
70670 F20101219_AACYAY thomas_j_Page_120.pro
2f9a41f0d73114dec173b67724a13546
9e4a9eab9c3e90135464e503d5bcd1c71894321d
F20101219_AACXVS thomas_j_Page_074.tif
327cdc0904f1b298b8fa1765e93f17d1
c37a5b9e653b78558cb6d50a1a146832b444735d
1723 F20101219_AACYCB thomas_j_Page_014.txt
b712b4fd6408ef830a896f406bff460c
994e4898d81fe1e717b040c6ff8b937c28362a74
65069 F20101219_AACYBN thomas_j_Page_141.pro
c150f34349172985083385e3208d0325
4fa646beb5c50bdf438f3cdb431da6cd7c4944c5
F20101219_AACXWI thomas_j_Page_096.tif
6c7d210969a81151f37ee81d0b955ec2
6613458830c8d4d75d70306c044768c06893a40d
52847 F20101219_AACYAZ thomas_j_Page_121.pro
9e5e0344b60187aa9d6228a47b2b0ce7
c0ce8bbaac7677a1ac73221036fc0e32df9c40f5
F20101219_AACXVT thomas_j_Page_076.tif
f109f5e797dafa970e12818403ea05de
2f6c5600292d6b579c8abf5efd16c45793dc5207
2000 F20101219_AACYCC thomas_j_Page_015.txt
05dd43f2b425230fc4ef8dead95117a5
35a0562d5b5189953dbe92f39c27c390c2fbcec4
61488 F20101219_AACYBO thomas_j_Page_142.pro
7d8b91e661c933dc97a13d1ccb1801f6
f2f1adf5a51810ff350ec5aea620bc35ed4b12cc
F20101219_AACXWJ thomas_j_Page_097.tif
d136476968072143bffa431e402fa19b
18309e7d1f2401225026ab72f8d4512d9ec1fc3b
F20101219_AACXVU thomas_j_Page_078.tif
4e0b23888f689a89b24e6a2dd0175ff6
1812f03bc394e8732e04e8a968b5df254ada0902
1977 F20101219_AACYCD thomas_j_Page_017.txt
ab8893c11c3c8d8acc209e54a5dcec92
7b4958638733786a103b20fad19e87f42fdb6c09
59100 F20101219_AACYBP thomas_j_Page_143.pro
572cc453e08ed914246a79e860d968d0
28f4570fa5f30b65902d3a7cbfdc2275754c996f
F20101219_AACXWK thomas_j_Page_098.tif
64545efb9ab15439980f3151deb35820
95ce56695a95a256d13b0e86f2db063785b44a20
F20101219_AACXVV thomas_j_Page_082.tif
6c4e9c58b903f423ee8b4fa68a9ecc0a
9546361e4a1430313b18e9c846bc75af99601889
2018 F20101219_AACYCE thomas_j_Page_018.txt
1fdfae559742935fcc84682f9d8ee66c
e75e0feb48eef146014d36b06c9415b65f61e0b2
58054 F20101219_AACYBQ thomas_j_Page_144.pro
78a5745f2695dbef5355425efb9e03fa
24deecd91d5b8158d9b6dddf01af45ae28fb7bb0
F20101219_AACXVW thomas_j_Page_083.tif
87c7f781a3849480e3fce12a96d67128
8e4b1fda45cd6ae8283e9f68f4aabc99843cd02b
2034 F20101219_AACYCF thomas_j_Page_019.txt
6664b95ff0c54d83ab300f8bba6b307f
e17f5c6349d835bf24cdc2d566b8f48ed4300444
65318 F20101219_AACYBR thomas_j_Page_146.pro
8ae0ef491de212d8be66419f2c59f657
9e27d77514d92b55dcad8a59082fe569892eb14a
F20101219_AACXWL thomas_j_Page_099.tif
201612ed1e766d68b9a2a927bb3b0850
efd89cc8a8acabc4f12d43b7ac780598a5df9a40
F20101219_AACXVX thomas_j_Page_084.tif
442660ea0307f147abc08a9c130f9690
392d72f186872e010df8939d849c2557236443f7
1908 F20101219_AACYCG thomas_j_Page_020.txt
4e7dd5ac55df94428fcafedb77445d1c
42eaaa4be24dfb8a3ef2c81a3994a1c8104ad071
F20101219_AACXXA thomas_j_Page_128.tif
8d6b3a831b0fb54c6f03abdbc1c4bb91
409b429175c67c87e4dc8eb495b15d22e2f67d58
66234 F20101219_AACYBS thomas_j_Page_147.pro
1778b0648d109105e9775e1cdd4fef7b
0a5e235a2af32e506861c90f1e40e753d5b4c2a3
F20101219_AACXWM thomas_j_Page_102.tif
70f45adf04add4b42556611fa82f4611
71347072abdba3c74652bd111a7de5a1c0831787
F20101219_AACXVY thomas_j_Page_085.tif
66c3d1a5ea6d60eacb7d6b5e4a71245a
147a42cf78a3ef70c0325a1a2a160c62cb5c95fe
1833 F20101219_AACYCH thomas_j_Page_022.txt
d3a77d52f7bd9f947f042cf167d34822
8a292eb65110c43ef70311af9b5e8a55bed68200
F20101219_AACXXB thomas_j_Page_130.tif
5a99082343181cd200a97f127a557003
997de0eaec6026e2b1273ad7e4a9d7b69da11b68
14975 F20101219_AACYBT thomas_j_Page_149.pro
5e39b7d7a27152f8ed2f1ecea9f3e805
d678d9dd702f3f9d505312971e2bff28409a0e75
F20101219_AACXWN thomas_j_Page_105.tif
e8da86c530e2d502bffdd4bdcf7fb398
7f13f414db1487076a89415fffe1830a5db8bb60
F20101219_AACXVZ thomas_j_Page_086.tif
0a9ed072be38af4f2da9cbfb115f9c0b
1501ed212de7633333731c62e2d7ec9bc65fd4ae
1726 F20101219_AACYCI thomas_j_Page_023.txt
bff8d7bdedbbbc77bcb0cf2e3baef1d2
f7d432393a3a6f7bcb6dadafcc1e38fb95b96b8e
F20101219_AACXXC thomas_j_Page_131.tif
05d22694689d6b4453d5761cb688b2c0
a7c4d96db348f1a1423bdd9e1c68e926305837dd
519 F20101219_AACYBU thomas_j_Page_001.txt
72c2d96d045339fdfe8396b7740ccc37
96c04f4833ca64cd98c60fe0250192d56fe5c91a
1769 F20101219_AACYCJ thomas_j_Page_024.txt
7cedb39a97287cc3484e5b1233ee7884
a378b12768a4db5435714aa72774b4966e547564
F20101219_AACXXD thomas_j_Page_133.tif
a1117ddd9674dc793e93f13f43f84090
72bf22b56d45527b0a2aa37c17b94b1e7b8dd27f
1390 F20101219_AACYBV thomas_j_Page_004.txt
37a4f758eb60ac256daf81cb24346e44
3421fb60ddb871b5a9da5096256bc45990a6d9dc
F20101219_AACXWO thomas_j_Page_106.tif
1765bdbf58b92b9ffaebec7e8a9c5f3c
08869dbee1a658dc1af95bb1246219c51e73fb67
2122 F20101219_AACYCK thomas_j_Page_027.txt
6658a6cdecfbb0c19160e803fab52609
728c92774028c1593b45648352c4bca46e8aa0bb
F20101219_AACXXE thomas_j_Page_134.tif
71256d464d6d243ef0a1dea77b84084f
5ad525c070835e49bf11d4f6fdfdc848606be8c1
2509 F20101219_AACYBW thomas_j_Page_005.txt
0032547b01c724e274f8b2507df0bcb1
859d390aabd5bc88f36c7a35e66923a4ffea9e90
F20101219_AACXWP thomas_j_Page_108.tif
711363e70c6bb6d032d9dfb31cfeedfb
7618710039f8dfa11381f4a5e6b538dd2ce9a8d3
2051 F20101219_AACYCL thomas_j_Page_028.txt
d5f1660d1c7480d9bcefde33bdd78658
0f8eddf41e86d11406215ee0b9b143981ed2e4df
F20101219_AACXXF thomas_j_Page_136.tif
06f6666531c65f7eb80dde49f7b88dd5
1995b547bd8bd177024e88e416a20b5e59494c2b
2458 F20101219_AACYBX thomas_j_Page_006.txt
16070708e555b02940379386a5ebf669
45f8cb50074a3855dfbe391f6b6e17d3cf653dc9
F20101219_AACXWQ thomas_j_Page_110.tif
e8a0d8bf671e211a3e20ca0af7cec281
e94d28fde7574815e790b2e206921f4f223257b6
2838 F20101219_AACYDA thomas_j_Page_052.txt
e66c17fa0462164fcd84f0e7f9dca940
b0a86af0e29a551f34ba29a6e1cbae98961bf5d6
2035 F20101219_AACYCM thomas_j_Page_029.txt
82d942184041b460762d2bca4cf74497
47841aaedcf6fe8e4d61bc6f42bec5e51ee0a111
F20101219_AACXXG thomas_j_Page_139.tif
929d8f266f117bbe019f6f49ec797321
b8c685ee14086b12940136d4adafcc7f70f114a9
F20101219_AACXWR thomas_j_Page_112.tif
a7793515b21f9359eaf246655fe2c60c
0ba130f7ac393cab972620dc8d562cab977a900a
2033 F20101219_AACYDB thomas_j_Page_053.txt
399b4be32d6ce828884b5d5473c2fa74
04f931c9e42aae17fd81e32e661a4952086dd43a
1938 F20101219_AACYCN thomas_j_Page_030.txt
99e6850ef2b89df88e0c4a5e923e989c
633dbe20358232d0b864bd0871aee0decd908f38
F20101219_AACXXH thomas_j_Page_140.tif
738c4e9ccc92d6a5b11f415035171170
d6328f2942bfb245c53c51c651a3ae509ddde237
521 F20101219_AACYBY thomas_j_Page_008.txt
7f473fd43db31d15534d9f0af95ad3ff
5c7a9ec204c3527f739e501c141caf859c46b6a1
F20101219_AACXWS thomas_j_Page_114.tif
dcbd6cb0af4509eccead91e3ac31b3d5
06942dd41fd51b1ebe85f8f3ae7476679f848451
2067 F20101219_AACYDC thomas_j_Page_054.txt
664e411c68673e548b29a8cc0bff40cc
efeffcd22353bc6e4429b2d87c498ec48a50c3aa
1857 F20101219_AACYCO thomas_j_Page_031.txt
9ab15d8067629e11a172ff565cae4b6f
a525f88317d7b5ac8ddec4f64540a45f0c87c37b
F20101219_AACXXI thomas_j_Page_141.tif
74edfd272d1a668a789b857cfa475467
fb5c24e0789023f0e94ea91cd208dab71ba80a0f
2772 F20101219_AACYBZ thomas_j_Page_010.txt
d2063a387cb099ff283f93b9d6609827
bc8acd122c7914a410caef214ab479fc221ba535
F20101219_AACXWT thomas_j_Page_117.tif
7dda1e454c5fa55f37fb386b6804e2fa
0bb62fcf78ede0c2f5b1e2da4a8b7b4ebcb742cf
2026 F20101219_AACXAA thomas_j_Page_136.txt
9856e472048263995ed5a83dd4bb1fa7
25daf63c72669c5ce625e160515e536a58209716
1980 F20101219_AACYDD thomas_j_Page_055.txt
203cb490e833a9ee96b408a78d6d1e1c
c67c12a34764ce4c86ac08bee5b645429f6bdfe8
1685 F20101219_AACYCP thomas_j_Page_032.txt
55f0725efe98527143d43dab6a997232
23c5f09e8c2d75db010e5c23fa1089162ebd3fe5
F20101219_AACXXJ thomas_j_Page_144.tif
6c7daba0c3c43da723d7f631a8d3a199
8c8a0e80032b7414692c5061d8fc84d3d8ae4f5b
F20101219_AACXWU thomas_j_Page_118.tif
82432823ff89892ec217c006b0ef936c
19a09b595b8dac9b379ddb66983964aec3a79718
27645 F20101219_AACXAB thomas_j_Page_089.QC.jpg
d8b93f4462757107ef342833e385a6aa
a0771773f0a932aeeaf7802580616a360ca159bc
2439 F20101219_AACYDE thomas_j_Page_056.txt
3e50fad7bfbd786ed5414df2e392f41d
529f99723438e15ac5b5868897ad21a74d88ef2d
1819 F20101219_AACYCQ thomas_j_Page_033.txt
9e7ef72bf2c8945862ad8cdb4b47584e
4db3ae00ab2d5163c448b79aba9d69a03a10af94
F20101219_AACXXK thomas_j_Page_145.tif
a463ad75fe360c87b1cc6c1fa9001b70
f148451dd852112cc86735246e803ef6004cabbd
F20101219_AACXWV thomas_j_Page_119.tif
1b4b232fa531ddbe1d207ad748d0cd29
53aa6e5b5e6419b79b2ba0299ec2c0aef544ad84
46999 F20101219_AACXAC thomas_j_Page_102.pro
520d745dfbdb1625e6be0b6d4b32a057
b1f6875a11c15dee3c54b5b9a5c6e4c298fbeab2
2028 F20101219_AACYDF thomas_j_Page_057.txt
fc805f7e4faea4e25fa386492da78f51
135f74288c1233ec44c69880f5bdf0f9f7169632
2070 F20101219_AACYCR thomas_j_Page_037.txt
7cee4927a4725bd0ff062d16f23fdb6c
37606612c17c2ae1e5e81f824aec712b29c53293
F20101219_AACXXL thomas_j_Page_146.tif
c54351d33f50cef31727a73c973c365b
d3f8981c88fa97d3a4cfa6368d4279f0b541e10a
F20101219_AACXWW thomas_j_Page_120.tif
9ee5eec6a896ebca610297d9a99d4777
b3faacab1baa10743e18e694a3d7edd27c4ff855
34148 F20101219_AACXAD thomas_j_Page_038.pro
c7a6bac1b94780e72692c55a4c6e9cde
2c38cb9b6ec20b1b749427268d7226e880119c39
2325 F20101219_AACYDG thomas_j_Page_058.txt
eb3f77d7dbc381c0c9acdedbc13b5faf
4fdd03520228219f121594dda54a5ff0c1b4d21d
50969 F20101219_AACXYA thomas_j_Page_016.pro
5215cfddc9a9453d4c3401cbb75b06d4
8b75e32f2d7e6be206ed27b03dc776ab6f38a1ac
1614 F20101219_AACYCS thomas_j_Page_039.txt
6911a517da86b6d4967e91c8dd7f14c0
1ded60de2353ce09f8925f1b0a73afa162d7bd82
F20101219_AACXXM thomas_j_Page_147.tif
128985cbac7cf38e8bec94d091a9d9a5
54ae123b061b6368e0cc573e57bb9b8cafe8ae5f
F20101219_AACXWX thomas_j_Page_124.tif
f4c76cac66009aa7db002ffe9388e340
412857294f27c2d03a1d9c0e835decc291666b4b
110142 F20101219_AACXAE thomas_j_Page_029.jp2
8bfa0ab32f519eb521735fa2ec486688
4f10a55c4850653cac1767c0e77e4206224c0d9c
1736 F20101219_AACYDH thomas_j_Page_059.txt
2f2de51d085a85fce581f3200c2d7ba0
408235d81fbeb58dd50a54f91ec4e1d53655a0b4
51061 F20101219_AACXYB thomas_j_Page_018.pro
6636f4593878e14e59ba838b2472c2b4
cc08f0bc431c3e91b42c4f191f539fd0f41bacbf
1401 F20101219_AACYCT thomas_j_Page_042.txt
3931fc4f465b4fe7398fac16485dfe8e
8cf003819375509bb861a6ed6ed89a10a268d31d
F20101219_AACXXN thomas_j_Page_148.tif
0da92282e3ad61661293cd2976be54bc
dc88cf2d93b1f155b42969669d9f18f55db4fedd
F20101219_AACXWY thomas_j_Page_125.tif
163db14bbcf5a052354aa06f316f56dd
9792e9862c3c9437d38a9a5eeb7d250ba81c13ab
F20101219_AACXAF thomas_j_Page_113.tif
48dad1f42cb58157d26ad26296a89ce0
a11293f267a7199c32acc3c21e7de431d03ca5a3
48464 F20101219_AACXYC thomas_j_Page_020.pro
093f6c716ccbcc2988ad159012bff367
730fc8a09a8bfaa8350fb3d309b675e5de7dd16f
1569 F20101219_AACYCU thomas_j_Page_043.txt
30203171b3e6d5859f182b3e1a02a094
7f3e498ae272cd8c0cbf90aaa0976bd606b02b1f
F20101219_AACXXO thomas_j_Page_149.tif
efc8d4fd4c1cb675f480f255c562e04f
f051404f0103264b233744b5088d9f55d77c2d35
F20101219_AACXWZ thomas_j_Page_127.tif
5c0e29a88d780bf640bcf9b9370f4c7b
24e01ae761b6a9e11e0cb6b1c511f489fe20f4f1
189 F20101219_AACXAG thomas_j_Page_003.txt
d1dcefc9db6f4cf9cdc770b2baf74b80
13467069c780a604ce156ba61d4821d3a5d68898
2080 F20101219_AACYDI thomas_j_Page_060.txt
3cb5a8290e59c43669b1efe450b60a5a
6cceb4ca68c0abf9035a072f559c4a8806620f60
46315 F20101219_AACXYD thomas_j_Page_022.pro
ccdfd8eef3191dfdbecd1e459b5e044f
686c6b58580b839b42b430e958bee9eab8adec08
F20101219_AACYCV thomas_j_Page_044.txt
f1d536cc66bcd1e46752029cee8a3fd5
d830b52437bb01d0e11e999034402c6293a8b4a5
105955 F20101219_AACXAH thomas_j_Page_054.jpg
f41f6af84bbc22e2472958bbb92f41ee
bb163185a8d1fce383e7aae1ccfe7dab8cb029fd
1607 F20101219_AACYDJ thomas_j_Page_061.txt
7cd7964b03fb4292e4ccd76dd12cfe17
2df9270273148fe99b56622fed007aae83042b1c
40210 F20101219_AACXYE thomas_j_Page_023.pro
7973bc249bac9cba01d640c2d44dee95
f3c2f939535f6f2eaee406b3e229b3853378ba29
2006 F20101219_AACYCW thomas_j_Page_045.txt
a338e22c5cab1b544b3a45ea27ec2fd5
c9a6595002ebaf7eb1082598d9e876ab6fc27fde
9702 F20101219_AACXXP thomas_j_Page_001.pro
18ecc31566d310e5d62644225b692ecc
fad63c2819bc6d3811101c67a790bd050382d472
1699 F20101219_AACXAI thomas_j_Page_041.txt
1b2298fa4603962adfeadae0078e6f53
05b447e9ea73e2c70bf8b3bdf4257e5475c5f9a6
1801 F20101219_AACYDK thomas_j_Page_062.txt
911433a0ec5ccc56dc9932b2ae234536
5d5508381088a3439038f89c7bfe3fd20e2f4b55
44555 F20101219_AACXYF thomas_j_Page_024.pro
6d5d367f5f3e6edba3c445b9c6cb1c04
5d4269816a97c86d50569347ca66ccd3e2d38369
1900 F20101219_AACYCX thomas_j_Page_049.txt
905a08a3d937316a1150e18dc1164441
032cb565a7e2da6f088e07146418b54268cfa454
2977 F20101219_AACXXQ thomas_j_Page_003.pro
d7710d0d139fe75ecb74c609f366be14
ccfecee1365044f2322798a69fa479a41789e9df
5633 F20101219_AACXAJ thomas_j_Page_100thm.jpg
61b81a7fb761efb3a76d46da9d43c6ba
f18cd7cc6411e3fbe791869fb974a089719f8437
1888 F20101219_AACYDL thomas_j_Page_063.txt
1a2d6a443dceda42c5531a0c54abd61b
a76bc16d847ad1529c9ba01e62ebea4f81dbf45d
1965 F20101219_AACYEA thomas_j_Page_084.txt
1f4b58341a5062ff8d65d82d88f60245
9b394b2fff73ea07357326a955fdc87e0801d7e7
46825 F20101219_AACXYG thomas_j_Page_026.pro
ddee0560ce10e855a089de6b1a2fc68b
bafafa72ea0548ac16ae87afbd38cb1b784047c7
2258 F20101219_AACYCY thomas_j_Page_050.txt
d46ac0068300b935bfd9c376e19a8aa5
fcc7da67456eff5eb9a80588d97689123fd5c91f
33696 F20101219_AACXXR thomas_j_Page_004.pro
4dc47649f6d42ad96a2b6ce3a91afe7a
702ce7086eb0500f16d078c222e540e28e15e30b
31522 F20101219_AACXAK thomas_j_Page_047.pro
f621039867301da36d10276cb46490ef
a5826cac03b16e23edda406b2f10ae4534f71506
F20101219_AACYDM thomas_j_Page_064.txt
9f783d716f5dfc235e2eeab89732e318
8c26376e2d250c25d39f850dadfbcee3c7bdb543
2955 F20101219_AACYEB thomas_j_Page_085.txt
6566c93101ca56051f063f608118d806
05fb3fe5988e673c58c18196099cab6e8892d0f6
54016 F20101219_AACXYH thomas_j_Page_027.pro
0d70d885c66750d518324573e3df2bdf
69cfa110647b4c21378cfe547b87d2357074c022
59167 F20101219_AACXXS thomas_j_Page_007.pro
335fe9b810462d1f3fa91aa0cac82c17
09fbd575520beb880ba69135e23d943418d69492
90821 F20101219_AACXAL thomas_j_Page_093.jpg
4ed3105b234b88e580810b89e19a6289
9957f712fbdc1d31a4721e2f3c68fba8f28c7c4a
2012 F20101219_AACYDN thomas_j_Page_066.txt
fceb5a205f8b61263ec4857ef3dbfef2
27ae973b099f35e7e802b20755110aba61a86e16
2665 F20101219_AACYEC thomas_j_Page_086.txt
453b0619ae9ec338ad66c73b87900793
43c5dd698a8d623f033fc13a2716ad42484a6db2
52217 F20101219_AACXYI thomas_j_Page_028.pro
6f3c4fd6995cdd92c489dc0c054c1700
db921348fda11ee043085d068debddf6e385bbac
2003 F20101219_AACYCZ thomas_j_Page_051.txt
b2093278fdd4952d9377743106af34c6
f4ccd9d1b49c5921e4a8347e25d24a5a55ff6221
12705 F20101219_AACXXT thomas_j_Page_008.pro
9b051f5f7a629a9eaca9b57df168d0b9
8c8cd15187686a4274939e0bb2849da5c12bdad5
125823 F20101219_AACXBA thomas_j_Page_142.jp2
3bcf35badbe87caacf522a645db95a28
6f83c3d332c97ce1ff3bae3b4af7e2377af5bf18
134629 F20101219_AACXAM thomas_j_Page_146.jpg
98deb05b7bc27abae1a23405bd9d8c43
75882b22ecd284d0129bcb400fe63b7ad1826985
1984 F20101219_AACYDO thomas_j_Page_067.txt
24a634235c6154bc20aab2e1c0d3d02f
73603f81127730979f691dd5ed3d83948613f8d1
1679 F20101219_AACYED thomas_j_Page_087.txt
e66b5a0101f11d69f3af0635bcf45193
2e74618f06d9f497f478c502f53ec68329850e62
1917 F20101219_AACYDP thomas_j_Page_068.txt
30eb2f45511898bf3133804e6b9aa599
f7123cb67a28ce04a71f400ee611c34be9ca0f36
51752 F20101219_AACXYJ thomas_j_Page_029.pro
a77639cd937b1b9f509e04e777c85f32
1da44bf637fcc19f50d76ee83f864eaea4e0ae0e
54450 F20101219_AACXXU thomas_j_Page_009.pro
4ad8861b9b98bd397b5cce3fa5261a55
8315f37083f3eab770f2700a5106f51453aafc27
F20101219_AACXBB thomas_j_Page_079.tif
ffa467aeec418cfd3ae3ca400848cf2c
b8c4fe2becd72f79bfcd5114a89f8977662dae47
96220 F20101219_AACXAN thomas_j_Page_109.jp2
51fcd2690ac919dc9b31130b7fcdb87a
7c1b67371ef4760e7e5fda53bafa5776b66309dd
2055 F20101219_AACYEE thomas_j_Page_088.txt
31ca0aad63d05b3715da57161a38734d
879c343cd5d6ae78cff818f2846def9dcb5ff6b3
1889 F20101219_AACYDQ thomas_j_Page_069.txt
452718ac3b5cbd2195754066d6087f70
505d23d9f69f648f74cb7ef40cfb1e213a41d2fd
44933 F20101219_AACXYK thomas_j_Page_031.pro
8728475ae9069ce9736f81baa1439676
3f26b16c852a450bfbe286160dd1d03dd6e5e777
68420 F20101219_AACXXV thomas_j_Page_010.pro
1d823960c2ebbf1cb7e871a6c13b79a7
88e6a5bea062cf46165026eda94b986bea07a2b9
1923 F20101219_AACXBC thomas_j_Page_118.txt
e9b493699e11a42dbbbd2bb1c2d6d0c4
56a4fd218588822d72f952c322dc11963bd5d225
F20101219_AACXAO thomas_j_Page_054.tif
d7c678c8dcb85fdeacbf1d0e054bb9d7
2a2571fb2859822c5664633e705bf404e0de3a09
1325 F20101219_AACYEF thomas_j_Page_089.txt
7b7b504aa3c85ee39db609e940099763
7926ff9a4670704f455784f8896db1ed1ef80db6
F20101219_AACYDR thomas_j_Page_070.txt
755a40149d944751f0000f1c17edbbd6
5cd2ae7bf8cf60bdc41a53b22745ea6665cd4bf6
39112 F20101219_AACXYL thomas_j_Page_032.pro
4ee7ec0489d4fe4b8cfb4f2c83151e75
1854818c2c8935e18c38a14d05db803f6ad71d0b
43119 F20101219_AACXXW thomas_j_Page_011.pro
3217b8ed9e3e5c5aaec94adbf3f09430
5d44747bd5735fcb3414160d37b674b7178ab8fb
7242 F20101219_AACXBD thomas_j_Page_044thm.jpg
49ed1c00a18aa58af1faa2376a3ccb25
a473cc5ecf9c8a046e8a949bf37358bd81e3da51
99777 F20101219_AACXAP thomas_j_Page_069.jp2
30eb4002a8473a3cd43063c4cd7bc971
44936ced5b99bbb65f2adab6d9c8c3d156f7d0d9
2069 F20101219_AACYEG thomas_j_Page_090.txt
31c435a7ab4158ddc8335fbcf4fe70f6
a88d2ef743cfe8d7a967232f5871cbf47ca804b6
1877 F20101219_AACYDS thomas_j_Page_072.txt
955b43804ba99107f2d5594f5663332e
13d522dca9acb9dbb35856042ef6a22097e6d4eb
42683 F20101219_AACXYM thomas_j_Page_033.pro
a9f53c7059edad0ed3a62b5c8d39e4dc
0719c7e3bb4feb183fcd774ea0807cd22da1d39d
38402 F20101219_AACXXX thomas_j_Page_012.pro
7efa4d51d14dcaa1ad90eb681ecf102e
009748b100086c1dd86eb12920c34d4d7f5fe113
F20101219_AACXBE thomas_j_Page_075.tif
5eda921856994476578e01d08438fd39
0248543609ddbd5c21e1ac48542be5c7871d5d35
8608 F20101219_AACXAQ thomas_j_Page_140thm.jpg
661faa0ca2ba705f56a394eef73f6aa9
9c9ad63149d37778132c28f03689d7ee961cd816
51607 F20101219_AACXZA thomas_j_Page_058.pro
3bc8b72d59f2f2155457dc7d58ac2e19
43e49445425d1edd095ef6343f0268172ffcb8b4
1639 F20101219_AACYEH thomas_j_Page_091.txt
f01721aef5066a0712e1c8087688846f
828b81d494e7855d1cc7b44757bc2174e4473ac7
2178 F20101219_AACYDT thomas_j_Page_073.txt
3b1b493261747810beb53aef303e4a27
543c9cf9095b82d31c511b67f749073840f5da3f
46750 F20101219_AACXYN thomas_j_Page_035.pro
e5a0ea28a6a50045950f956d69cf992b
33dfd81c283f39da438c329787448d1535294522
40746 F20101219_AACXXY thomas_j_Page_014.pro
3e6d39edbdea20b2b441f390cb70e256
78d3bd9c4071767353950261321b1071f2d22729
8537 F20101219_AACXBF thomas_j_Page_079thm.jpg
4fbfa45e755f4c3df6db75edc7304ca7
218efaaa217a1e7d50c9e1ab437e59732177cc19
35155 F20101219_AACXAR thomas_j_Page_142.QC.jpg
03f228370459aa5ace7a084421193d0a
e8cd2c69d4403e048f89c839df92d7dd98c8a025
39363 F20101219_AACXZB thomas_j_Page_059.pro
b9fd681ab5aec95c963cc53d93a547f2
9439fce6421e6744e16b6b6a835474374065b7ab
1291 F20101219_AACYEI thomas_j_Page_092.txt
88dde71c9db7baa6fbcb615dbe669d7c
6febe9bb93881a2034a948e07b1fa5a04f1d636a
1963 F20101219_AACYDU thomas_j_Page_074.txt
c6eea386eaf220c1836ec4220577b2fd
200f44232e2af30e826d6d5beaa4a1459827596c
31179 F20101219_AACXYO thomas_j_Page_039.pro
9dea1f061744a7a09844bd2b23683190
1486002bf16e9db5c6c1143bd44f1a00dc540c85
50804 F20101219_AACXXZ thomas_j_Page_015.pro
d27d4adb003ad6d597f22e8265006a60
20c24a7ff8449e8c34d8f37562a58e4555840e9e
F20101219_AACXBG thomas_j_Page_138.tif
3f76bb4f3e398bb84878987c3588ea2b
5ebdddc46fd1eb0d705f949b9632112fefea341d
36435 F20101219_AACXAS thomas_j_Page_070.QC.jpg
bb8934187734df005502e9fcdd862c9d
41b4bcb335503df280a52bd27062e7b783347ce4
52895 F20101219_AACXZC thomas_j_Page_060.pro
276b07f585c5342363ea61f1b63ba4ae
783ec39d5b0e0ee93dc108311349876f1ac3267f
1759 F20101219_AACYEJ thomas_j_Page_093.txt
6b6527b7f090c89b18c1c860a0398af8
ab40e098bfa817c685e485cccb1d8ef22126ad69
1786 F20101219_AACYDV thomas_j_Page_076.txt
b93eb46dbc5f2c7b4a2a3fca8b1cda68
84c99ba59f9461a9a39015c0114648eecf21a76a
47608 F20101219_AACXYP thomas_j_Page_040.pro
61f3a820593e11191b92fb986ddbd319
519a9888ac55a0b1faf23e1c786473afe6d78fa7
94091 F20101219_AACXBH thomas_j_Page_093.jp2
215ef259717729caa64ea1392fefc634
88c09c2ccd0451d4409077567286807d3594eb7d
85370 F20101219_AACXAT thomas_j_Page_135.jp2
309a4184224ba542d95f82668c347b4f
1adfc6f0f94b52ab8e1dc0e27af3111465566389
35335 F20101219_AACXZD thomas_j_Page_061.pro
dbd854399b2e79d87d7b15ae1c6994c1
91e879cc083b55cd2f45c57464808c7b5089b1cb
1808 F20101219_AACYEK thomas_j_Page_095.txt
6afe63c15126998c379d281045fe9fea
fb23126e2ce1881f1fbc21d69c8431a83c442a48
F20101219_AACYDW thomas_j_Page_077.txt
018fe2a679e811c55aadcf80f2fb8df3
43df48d2e2611e04d0378427a7734953f05d04ec
92336 F20101219_AACXBI thomas_j_Page_096.jp2
0c9d744f196c6ebbecec03ff00bd1749
7e7b98244f1db8a21597a8ab283b6894a194533f
83877 F20101219_AACXAU thomas_j_Page_057.jp2
f521d2315e05144d6f23953d8fc8c842
a427023977e18c58b9f329b9feb48461a77d2f1c
44569 F20101219_AACXZE thomas_j_Page_062.pro
885efcce47ef0afb80db90ffd7ad7079
6f6560ef8f8218d7678ad1acb08ab1d68db92046
2224 F20101219_AACYEL thomas_j_Page_096.txt
6e66a2e701c00dc19395a53002fb9396
9574e13060872170ac2bb0306ff010d9fe8f2cf0
1966 F20101219_AACYDX thomas_j_Page_079.txt
48ab5280c22d243e12fed528ed6e053b
0ba7cfcfa17455819a54b1b97be70711fa9daaaa
29941 F20101219_AACXYQ thomas_j_Page_042.pro
88fa1d9f1449764a1c74a3984bb5c844
bd83465deb98a5772c2db31bb27dab89b36e1f38
28152 F20101219_AACXBJ thomas_j_Page_096.QC.jpg
00ced4e57ab20d2ef3d189cf9b24a1fa
c8263911bad2839841e7f233f72d366c02aebaf8
1451 F20101219_AACXAV thomas_j_Page_130.txt
1da2090a4fb67aba41004ecc45a5c50b
14d103098bf0d6e4178512bfad2450a9f018bfea
51869 F20101219_AACXZF thomas_j_Page_064.pro
30fe8fc75ee492346be72a5555e66f56
6a9c37cd4543619edd1aa4348ad7d43f75224dd8
1864 F20101219_AACYFA thomas_j_Page_112.txt
424e0647e5c0dfbc515e9b7fa4a37056
e52b65f8f95845e79d85d0118c548b93275037e4
2066 F20101219_AACYEM thomas_j_Page_097.txt
0a26a40fbfc7f50f205e1f5fc8406b44
b2ece86d47766ec2d13e4fbcb7c2d36f1815844b
2286 F20101219_AACYDY thomas_j_Page_080.txt
91fe2196fb34d917de6e82c450f58661
7c2173bd4300dc51ce5198d3b8ad3b3ee898da77
39340 F20101219_AACXYR thomas_j_Page_043.pro
688c94640c33d82740ed857d5e2d3a23
daf3ef47118078a49e18fec615ce387323f2021d
35813 F20101219_AACXBK thomas_j_Page_078.pro
fd26ae3633b3f078a74f660b9bc46a74
cd8f854f52bee30fd72ca4863e88bd90c6628b0d
8625 F20101219_AACXAW thomas_j_Page_054thm.jpg
f83a57a96bb31f57da2096b43d784d9d
8258a283b43a2ab00182e666c57e12ad13343e45
49158 F20101219_AACXZG thomas_j_Page_065.pro
3c406a6db6ec195fa9bd1d9562b598c0
f98bf652ab05c1052e74a2a3bf016750717577fe
1472 F20101219_AACYFB thomas_j_Page_113.txt
b66fd58a77f88d71605c5db48e12aad8
fa1eaa6f8eedb1fd14d9b6e52cf6b553159f38db
2032 F20101219_AACYEN thomas_j_Page_098.txt
0a7daafa7180826ad3f81721c3156360
47e14ba4520097c2f415f39162281fe36dbfa407
2287 F20101219_AACYDZ thomas_j_Page_082.txt
d85c73afc60857c04fbdf22a161b9a66
c590ef908fe34ca1706442e00bffe935b8bd92b3
44336 F20101219_AACXYS thomas_j_Page_044.pro
011172e0aea32c28df0c1a02117284c3
867459ad77a471219ffe59f5970297ae2f4e5875
F20101219_AACXBL thomas_j_Page_081.tif
c4f72092e212c3fb73a6d99ab1c2d2e0
8e952a2360de39d3599c360d9064a00c367279ff
F20101219_AACXAX thomas_j_Page_036.tif
7c416965cf9c31fea3c2d1240a2ddbfc
3949b198a1f469cdd97b1fcc9b34cf24a17011ad
50399 F20101219_AACXZH thomas_j_Page_067.pro
da6a3e1b1d60960287b331547931cb3c
35f58ddb98c4e72c6ca038b9e5d5b80d2325d189
2110 F20101219_AACYFC thomas_j_Page_114.txt
607f895d5fe52b4e86bc26d448b82a3c
b98fd2e25f004b490affa32c9b9da28b1c6a2952
1854 F20101219_AACYEO thomas_j_Page_099.txt
3002e2662989cc1831355111447ed342
f2e322d8e6181c049f2d825cc3173e49e2d2301a
51022 F20101219_AACXYT thomas_j_Page_045.pro
e1030b09305bd6e50009e7dbe2ff65b7
a341c2baea55cf19921588a4c7bfa58c85d6f98c
7020 F20101219_AACXCA thomas_j_Page_135thm.jpg
10cd0a5a533f26b62c93af46fb2051ca
b3f2f21fc979cebbc4a5194b1aea1b951bb77310
108258 F20101219_AACXBM thomas_j_Page_079.jp2
5118631f6c844fccf3624fd3df393d74
6d5bf21a26e7141db0c5899cfec12a19ad2a5b07
48562 F20101219_AACXZI thomas_j_Page_068.pro
81113a219846be892f299712f5fab933
bd55489248142ab10784b2b6e9a2c569d917eb5a
1959 F20101219_AACYFD thomas_j_Page_115.txt
7362bc7e11eb0976707c1100703c5a53
5cb1b68740faee294261df54c0f79bfc4d0137e8
1106 F20101219_AACYEP thomas_j_Page_100.txt
455f7a02caf3cdfef306dc71a560b0da
3f17ef41ee7379083b4cd47d94d85726a2d12b38
7967 F20101219_AACXYU thomas_j_Page_048.pro
4dbd189ea28311c3c0a0338790253b35
586287260d41cef14f8dfd360ee30a8fcf7e7427
995643 F20101219_AACXCB thomas_j_Page_124.jp2
9e1d5e4f581882a100710d009ebcfbce
60728be22d33e2c7cd3745596c456e5691ea0f32
F20101219_AACXBN thomas_j_Page_055.QC.jpg
45b287168015d9b307c5ad9caea35dc0
a8cb856d1a2338338831f7802e834b2259ebc9b3
F20101219_AACXAY thomas_j_Page_135.tif
9c1d336dad3a74ab110f15aba5579393
17868568f2de8aa65df14f8c92519152ee0a4247
47732 F20101219_AACXZJ thomas_j_Page_069.pro
be47a75e3007ee69dbb00b93f46a554a
149c1468f652704b8441a3ae4a8dde3988b4c32f
F20101219_AACYFE thomas_j_Page_116.txt
9aea73608dbf91e87f166f78e43ef267
6efff01da8411ba712773bba136fd7cc1065ae1c
1946 F20101219_AACYEQ thomas_j_Page_101.txt
9495109d43d92a2878f087c6b404dee5
f7fb93b49ff782782aeaf32eebbb055142161adb
53103 F20101219_AACXYV thomas_j_Page_050.pro
42060c9603aafbcb692096cc72ab8bbb
bbdbfbd4e74e7bd715b41776d8d29c92dc40dce8
68023 F20101219_AACXCC thomas_j_Page_039.jp2
b52e7631eff4d45cc5a2c3202d939281
f78220d2d82a528223d50d5ac19e29011df72ea9
35373 F20101219_AACXBO thomas_j_Page_143.QC.jpg
80c2a9ef6e8ba1c9dd1b0bf55bc263e5
ac540ba62d2126ac157a0f3478edb0675c968fd8
1779 F20101219_AACXAZ thomas_j_Page_011.txt
73d06a4161f5a1897d36cd581b4a4a42
127dbc98f36c7fdea1feb9b393bc6f3951ea5d72
50993 F20101219_AACXZK thomas_j_Page_070.pro
51684249b13624437249a9a7a454ff6c
e841b3ea97631e87ceeb5638858c420256f6645d
1913 F20101219_AACYFF thomas_j_Page_117.txt
b9954a2ae606dc36f666bfc43f714aec
f8706bffc819b93dc25455b9b9e18df797d321cc
1852 F20101219_AACYER thomas_j_Page_102.txt
6eba3b9c77076425cd092b8ed6ac7dde
179d777675f872c40c0fe9d6a703eddfa5d1aa03
F20101219_AACXYW thomas_j_Page_051.pro
d178a8e5c238377878cf60c6c56179e1
ddc63e9405e6fb04d3a53ba289a43af22ad5dfb6
51089 F20101219_AACXCD thomas_j_Page_066.pro
c9172b1340c723d3ea5d8f735092fdc2
4bb3dc044ed2765cc872d9b1bb073b549d9730c9
F20101219_AACXBP thomas_j_Page_104thm.jpg
b2a6987d56bd5f931e18a39bfda11f58
11d6bf2df09c1af63279bef0413ca52f16a5609d
46916 F20101219_AACXZL thomas_j_Page_072.pro
9f7f44d47dfd3c977edb052dd425ea04
ccc4211c937c3e4afdadfc14f4ac2896f7c07e7b
1896 F20101219_AACYFG thomas_j_Page_119.txt
ea7fe5e61bba99eb22e71d8194a24d4f
4402b92d76069cfeb48c34096a047071f8658121
1943 F20101219_AACYES thomas_j_Page_103.txt
56517a4765de5fe42189640de7394d91
51e21803b0c614ad23bc17913d0372dd271df8d5
51924 F20101219_AACXYX thomas_j_Page_053.pro
c775af124b5fa2259369192511331a08
772de9001f2f507e5193405fa5b96b0c1e4cf3eb
F20101219_AACXCE thomas_j_Page_122.tif
37c49075424451cac0d52f0996d9e4ad
437fec12fc5df008f9cef33efcd7bbf015247f18
72414 F20101219_AACXBQ thomas_j_Page_042.jpg
04fe38b616e3a5102696aeb0dc9f7de2
5e4a7c845f0c4851b9283f4a1b1f0715d8e6e05d
51982 F20101219_AACXZM thomas_j_Page_073.pro
f1bb3130e76092b0d571535c1303d95e
e2789c49409b78af0f1d6db0627fe7ce9c5f1077
3012 F20101219_AACYFH thomas_j_Page_120.txt
bffd2803089ceef3ae2986fce035d14f
70133737d70ba99274c1e003a19cf4f597eea94a
1901 F20101219_AACYET thomas_j_Page_104.txt
d7d05d2398cba177e47f891666ba3086
773cf6f1255dd4846224f8ac05ebeece8c4ba955
52338 F20101219_AACXYY thomas_j_Page_054.pro
3e5b7e85344292663f4d1beab861767c
2fa9cb63853a8739245372720eb5f1a9a3c543fb
F20101219_AACXCF thomas_j_Page_036.txt
f47fe4536a1a09045b01265e5d5d109d
3fdcf65ec74b1473ac52d7c39df366af69f33565
53624 F20101219_AACXBR thomas_j_Page_125.pro
e27ad2039d5a739f9e7c8a7b40041991
a3c0863917cad88b4590062165c3fa41da5e450c
48107 F20101219_AACXZN thomas_j_Page_074.pro
e91a2c417c4272abf75c10177b738e59
0f0d2938fe84d51d7d5bf78c86d6a510cca89e96
F20101219_AACYFI thomas_j_Page_121.txt
a64f87b2cf17b795e92fde7b6c30078a
8035b4562e6b88391863d1b54b09ea26b975da79
1800 F20101219_AACYEU thomas_j_Page_105.txt
bf90d60508ed23dccc15eb7980e96176
f18e0567c0531f79437273dc6b7eabf7d089082e
48818 F20101219_AACXYZ thomas_j_Page_055.pro
348fdc0b4090b2951685af9d6b616cfc
8795003237ff42a892ee089746286429d2018ec6
22371 F20101219_AACXCG thomas_j_Page_013.QC.jpg
a6aa1b63b617895b0f08d15882cbecfa
5c581be8ef1e81526af3bf92aadc52c0a387445e
8406 F20101219_AACXBS thomas_j_Page_018thm.jpg
a045fe4e7d378afc8b6c738219ddbd1d
f136f45873418cc6712ef9d688df3f44400ccbb5
44001 F20101219_AACXZO thomas_j_Page_075.pro
e0769ebe77c652b1cc47bc3ef341c2aa
c92cc30a897906e371ed8e900c3d83666da60078
F20101219_AACYFJ thomas_j_Page_122.txt
ef46960a81105f4a8e4018762d3d7104
d2b31cb0ed1bb68c4e9dd27b8e32f4e2903de16c
1955 F20101219_AACYEV thomas_j_Page_106.txt
949d8932f13ec3dbc540ad61205063b9
f358ec2f4f5f4b9016fac3ace13f5d1c3af0c154
18992 F20101219_AACXCH thomas_j_Page_137.pro
ebb5472c0bb08f23f5c3d471fc985ec3
6e62542c74c543e4bcca16db4b385049434f4f3c
107730 F20101219_AACXBT thomas_j_Page_114.jp2
dfdbe869e58037632ec9584cf49c7994
5c46d4271a17660ee142409afff5587d6da0c550
29760 F20101219_AACXZP thomas_j_Page_076.pro
cff53a324948c7aeb29067896f0f7e55
e5e6eaf08f3cf75c8cade21d199ee64f23c3e595
1512 F20101219_AACYFK thomas_j_Page_123.txt
9a58db410221ad100136f0a7971f0b96
e0c66ed269ad1b8872611539fc2e16797e1d260b
1592 F20101219_AACYEW thomas_j_Page_107.txt
67db37b8ff30e05de399af142c15a3eb
ceb987676249340341039565f4eeedeb848e6860
31970 F20101219_AACXCI thomas_j_Page_021.QC.jpg
2a0e5bd5e874e27332e2f2c09a311545
0bc0b60eab0f670f603af0fbfe8acc0211c6c74d
7392 F20101219_AACXBU thomas_j_Page_035thm.jpg
7c261882227a80586d13193e8ed871bc
c164f52f951822e82b0c3439f4dac30e49ad681f
45837 F20101219_AACXZQ thomas_j_Page_077.pro
18f61e7c3e268b6db1b441b8ab5df13a
c7fbbdc788ea46d32b2b6ee8207cc5a0ac567fbf
1509 F20101219_AACYFL thomas_j_Page_124.txt
8533e953e1a1e1a7784c653814688706
545cf29e235271e2fcc1c6eb60a85dd260f001aa
F20101219_AACYEX thomas_j_Page_108.txt
c24ca4df2a7c7d1ebfa5dde6909a8923
206a4e7a0def66cb4c68ea8d70b0cd530dfe2daa
51605 F20101219_AACXCJ thomas_j_Page_034.pro
62d360b93fc877098c1df10381e628e9
d41360e8699373a94b945854369053aadfefa22c
111064 F20101219_AACXBV thomas_j_Page_053.jp2
0c96aef736c75d2f8bfbccbf8d860e03
8ebde81442508fd2dc26aab20a78f3d47cbd7f40
2737 F20101219_AACYGA thomas_j_Page_147.txt
4a80c2d67d05bc6d168b2e9b9f53b8c2
409a6d33a26a41a8b3f288d801becf4c3a82dc7d
2114 F20101219_AACYFM thomas_j_Page_125.txt
34ea1a8493071da0dbd4939946f59627
4328edd5f847eda779e4df0c0d80da8b5a4661c3
1996 F20101219_AACYEY thomas_j_Page_109.txt
d53d659d09c9c0e2acd8884d85a373e4
5096bce25be1b868239b25db11b561b668ef055d
7099 F20101219_AACXCK thomas_j_Page_059thm.jpg
d7d57002c60bb81d4c4206ec3dab063d
5f3610c7fd9bf7c0bd66f31de3a62b8b1a594105
F20101219_AACXBW thomas_j_Page_101.tif
e95383a21187fdfe6af569d0811fa25a
36c0ac2b741839f2b5a9669171a0bd9054b6cbcb
50047 F20101219_AACXZR thomas_j_Page_079.pro
54ab9ef12897d933b117792335ca55e9
88c98df784946b771f321c0dc68d86c810aab38a
645 F20101219_AACYGB thomas_j_Page_149.txt
e8197d7a6d953a011b8127e3e0e70f6c
f0aefb56e16b3422c6de0c31a99ae3b96007c9d2
1583 F20101219_AACYFN thomas_j_Page_126.txt
a4a70292cfa9564940c68cbc345b322f
54bb7c767d045619eed2583c44a5743bdbf22922
1898 F20101219_AACYEZ thomas_j_Page_111.txt
a6fd8d13409291cbb1c86db47a041eec
f0736a944ce1734cbb0f0d6f7077a0aa978ce2f1
104864 F20101219_AACXCL thomas_j_Page_128.jpg
1760f0ea3a6d24929146ab034f32b210
e157c1b3d112c779dbfaf90d0d1dfd719f388b0b
7984 F20101219_AACXBX thomas_j_Page_063thm.jpg
509096c49bf92812c4fd320d68bf1983
6d8f598b51d30a5b710e7ee1ad6ef1b071d54618
54008 F20101219_AACXZS thomas_j_Page_080.pro
94dbf750aeb5e7fb33f891e77063af18
5e577986536150af93ee6ab857888438296a6789
2628 F20101219_AACYGC thomas_j_Page_001thm.jpg
f9dfeba446d60530df0345e306c5a7d3
a8ae9891444d0c12674b8537359dc5c0bc3af1d4
1705 F20101219_AACYFO thomas_j_Page_127.txt
6edf237b02160ad54e5fe53c51ed9036
ae22cebfe33428c9f4c6f04eaf80d30307802104
101843 F20101219_AACXCM thomas_j_Page_102.jpg
3184fba8984d778c48690e868c679ed1
eaef95098039d69c3b7e16a3874a28d2fe446b0e
F20101219_AACXBY thomas_j_Page_018.tif
1f3cfe861a43eb5c6ef0026180b07996
4fc79a1267d4e69e8600a52f448fe2087b37f32b
42208 F20101219_AACXZT thomas_j_Page_081.pro
ce82953199ba56c6bf96427466564734
adfc4f66183a3e3718f4a054b44a84400cafb57b
6590 F20101219_AACXDA thomas_j_Page_042thm.jpg
0b6df10cff9ea8fc5c7a9bf1172d9a5c
c44691aa181707467d7b4e4a50c74f0fec07064e
693045 F20101219_AACYGD thomas_j.pdf
d80a54ae9660e024c8fd7c71b4f7ea60
45375994e2bd7f09072036a3ad69f467cfa65edd
1995 F20101219_AACYFP thomas_j_Page_128.txt
0f31b44730e986b8ab350ab0a3b7d0ae
3a42df334e7679d491cb64bb4987c4c920242061
9594 F20101219_AACXCN thomas_j_Page_001.QC.jpg
af6ba5fcc4ca3343af7d5b24d4b64287
71c7f4c974b8a8a3924c3299c36755dd33214198
57297 F20101219_AACXZU thomas_j_Page_082.pro
5f8a67b157c2d428292fcca370d10953
15bef4965763eea623aa2b2569a6afaa3b055b43
48283 F20101219_AACXDB thomas_j_Page_098.pro
1c84752884601cb7ecea00e94fca45b7
8f50576f5afced8fcf5bc297f5567a5a5eb93d52
29996 F20101219_AACYGE thomas_j_Page_109.QC.jpg
8707aba4123f20ecd4c3488a9b2168ef
7ec2043b2dee6bb037ea52a818b8e658d8fab19c
1973 F20101219_AACYFQ thomas_j_Page_132.txt
7b187caa4d8fd4b89c75eabc7a71eee6
82c428b059396e0854e28a8503840d088b073385
109980 F20101219_AACXCO thomas_j_Page_015.jp2
475140f8695bff12e9250553981be3e2
3e4c5c8412bca42e52d997733f95c1abba0d02fd
F20101219_AACXBZ thomas_j_Page_044.tif
8175319541dea716d56ca4b7c6e25276
e1d1b21399ebafa106052a0b642d1214217e2d9d
50287 F20101219_AACXZV thomas_j_Page_083.pro
d80c895bcf7704c9305d98cfa69fa163
85a85020c74ab6701281b8d670e17ef112acf20e
29288 F20101219_AACXDC thomas_j_Page_031.QC.jpg
9c6bec5c45c3fe5ee7c16293c53547d7
c08f0528cb6af4e10c8654d0194c9a51c5de144e
7189 F20101219_AACYGF thomas_j_Page_122thm.jpg
cba631124a310d0bc128bf02ba52857e
e80b1d22e964100b1bdeeccd343460a227f719b6
F20101219_AACYFR thomas_j_Page_134.txt
c27a70b6e06ba46cfb3b18c4d72a6af9
f22669714ad6697a3a76569721c61357db5a6454
F20101219_AACXCP thomas_j_Page_029.QC.jpg
8372e18b274ca42a8f94731398e9d6b2
c970d96ececbd2c77e10de1d93563a569fd38ff2
49728 F20101219_AACXZW thomas_j_Page_084.pro
9c2814f9de3d205ead4dbd7171e52960
7bf6bbf9b40f5031099b985c931e223ccea2202d
47058 F20101219_AACXDD thomas_j_Page_109.pro
0f29cdbb8c9c27a32762f1ff6ff024b0
d274aa28671bd07bc4d44cc683597ae6598e08dd
34706 F20101219_AACYGG thomas_j_Page_034.QC.jpg
0add3b15a2a8fe0a1b100b2b3e650d2f
d7e6fe17c7199f83278493b417306ccdb129e046
1532 F20101219_AACYFS thomas_j_Page_135.txt
64f135c8e00f67acfe0f963db145d60d
a87df0b8f31f6f7f81ac273ec9fff2f3b08c2ff2
37452 F20101219_AACXCQ thomas_j_Page_071.pro
7016a42636e2d5c587c6720204761331
23c3a7d1157aa342394d6f93628217da16f374b5
70486 F20101219_AACXZX thomas_j_Page_085.pro
2dd6e2e949ef93afb7fe8d0422967f4c
531db6fd3db3371ce019a165ca1b033a532998b7
40608 F20101219_AACXDE thomas_j_Page_036.pro
6e5702adbb1efe8c3a7db03881d6cee3
d2c94808fb7717cd496aead4d9fd2d37dcc98a4c
28124 F20101219_AACYGH thomas_j_Page_071.QC.jpg
933c2b6d472d63afd26dab6720aa98d8
164e97f6739227477c5844f5bfdd5de65d172e03
756 F20101219_AACYFT thomas_j_Page_137.txt
14c61b3769351fb7677814ae0190ce1f
e9b856339ffde3c83de5203993d5252bf689afb5
121244 F20101219_AACXCR thomas_j_Page_143.jpg
e8cd453bc6d88b8280f36bc266277fcf
3310d521035322b2a3f118afa2a45d3b4e8d5725
60650 F20101219_AACXZY thomas_j_Page_086.pro
1d67f778313113eaf62d665a27a31030
0edd03fc9a98c7d51a0fcd5b554e018a30a8a0a7
102254 F20101219_AACXDF thomas_j_Page_040.jp2
9272915ca70583d7d48c6ca3b420932c
2b522e112efa02b5c3f66f11ff02e8dbe73101b8
24750 F20101219_AACYGI thomas_j_Page_076.QC.jpg
49208f3f97c5635da9ea03b584ac4297
8cc73e6461ef629106af2368ad8241342f431b8f
2709 F20101219_AACYFU thomas_j_Page_139.txt
9b90b2e201c2f34e6451f361bf626289
592cc5eeea8dff91b867935a21d234153cc3d96c
55791 F20101219_AACXCS thomas_j_Page_005.pro
fc9a549ee3314bdbafd5c2315af7ddc0
a5da3f596c84cfc22e5456517b3c1930e35697c3
39273 F20101219_AACXZZ thomas_j_Page_087.pro
a645f3aac007e6801bb574417e52becf
0bff9919c582572d8298186a5397870d2be6f2d1
7193 F20101219_AACXDG thomas_j_Page_047thm.jpg
766d83319d242fdefbba63b12f0bd207
cb19bf4ee72c9b3d0e019eaafd6f63c8e692385f
1921 F20101219_AACYGJ thomas_j_Page_048thm.jpg
432ffba9e60364ca5e5f7d0fa76d5e7a
f611cdaa9d442bde4ad8377aa8b9af78a40dcffa
2542 F20101219_AACYFV thomas_j_Page_140.txt
df1de8ece99286d67a0bf695136c7d9a
ed6400b8bf5e6d879a239297f99e7977d78f4f13
71324 F20101219_AACXCT thomas_j_Page_094.jpg
ac14558ae8f0d7b9b346f88b1c8db7a4
eb0e75d2388f38bab278e013a69f03771b4273fa
F20101219_AACXDH thomas_j_Page_064thm.jpg
b4df79d52e8b86cd70544356afd4bcbc
b66e409ef0c8fe8491a7d3daec40419b23bf37d3
5060 F20101219_AACYGK thomas_j_Page_005thm.jpg
68d7dfd72864aa76473b910d8597ea45
1126477bbf121b997bc541a3a7cced1d73447a02
2669 F20101219_AACYFW thomas_j_Page_141.txt
4ec2da87744c278b8530e1ea679693be
f4f33fcb080d933ef2fd4f6d373ba90de97c953c
7627 F20101219_AACXCU thomas_j_Page_022thm.jpg
063a2dd59066a4471203aeb35d17c755
0057a9ec3f3f35ab96cc88c55ae261aba7f73556
F20101219_AACXDI thomas_j_Page_049.jp2
b6e561bfd512591e00fde23e693aa58f
3617be5a0ac41d3e92ccb6448d548bfbb732ee11
7107 F20101219_AACYGL thomas_j_Page_091thm.jpg
e361ae0c4d42b8280d423827151e61c0
bc741f6ca6e485f743a63636cb1d952ab610b193
2444 F20101219_AACYFX thomas_j_Page_143.txt
2d35194f88945d29c60f4378ede8b80d
953724400670e963892c0e089fcc39ee39ae0786
135293 F20101219_AACXCV thomas_j_Page_141.jp2
e5ea4139473fa1b0c88a1a7f63f29ca5
72683430bc4185f27712769391dc136e653d954a
6951 F20101219_AACXDJ thomas_j_Page_078thm.jpg
49af3e6c4ea40f84e64f8d0cec51ce0f
c1f581a25c50bb49d33a23c85a603d4cb4b15b01
6525 F20101219_AACYHA thomas_j_Page_014thm.jpg
16785d90f3ed5e50989d60cd62d5f7d8
a210e55978dc30b6a99197f00d38a1711789936a
7402 F20101219_AACYGM thomas_j_Page_126thm.jpg
35874ea3aca5b33c22588dd397ab0fbe
97f29c532b83b2fbf0356a7874a8820bc06d32e9
2393 F20101219_AACYFY thomas_j_Page_144.txt
e0c870bcf49feefcc14678aeb9ddacc1
4151f128a6650df47d928586ccf03284f728ffc5
75179 F20101219_AACXDK thomas_j_Page_004.jp2
8786a0cdeb9a44b837605a630ba70984
ae482635e0d3f2759c2195d52283bf55e5aeb2d1
27091 F20101219_AACXCW thomas_j_Page_032.QC.jpg
a007e98b6b65263c1bdb4d2b11e7ae61
631549b8d9e59715109e63229e4e154304031982
34367 F20101219_AACYHB thomas_j_Page_065.QC.jpg
b104bff4dff847fd35f157c566a282dc
9ce6d986c46582c6fffaae1f82f0434ac302aecc
7784 F20101219_AACYGN thomas_j_Page_030thm.jpg
20bfaf44307a44640e66cfa10caa5171
a3fb095a7cf7efe1d5a9addaf6401331be8fef2b
2678 F20101219_AACYFZ thomas_j_Page_146.txt
b5c56c9c0257382e3f62a5cf0af0a282
992228bdbe1c1cb795ae4b388c59b3dfc30a77d5
124162 F20101219_AACXDL thomas_j_Page_142.jpg
3dbea53cbb7208153a948edabf806ffe
b4ac5d6f3eda463e87925b0f74d1a2d82e5d7f1d
51611 F20101219_AACXCX thomas_j_Page_019.pro
4b8e363291b875e93f7fe8d2673f47f6
e39fb7d78b7c32cca1fb12b3e56a87434c4d3707
27580 F20101219_AACYHC thomas_j_Page_059.QC.jpg
802782bcee1061265b399379628adb70
6783a19593598e87e6686c6773863cc9b2b3018c
30852 F20101219_AACYGO thomas_j_Page_035.QC.jpg
51ef440d8c8a0ce4bc78ec959bf12dbd
a34c02fca320f57e210a35f5a72c9f3525cb0e42
1220 F20101219_AACXEA thomas_j_Page_013.txt
8490cb3714cb47ec5c50771c2e887f0d
5bc3594b097934fc34acbcdbb23846eef0035191
F20101219_AACXDM thomas_j_Page_081.txt
255c2453a4d0a10fa291770e633820a8
d57cf87113bc3cc2abcca41c3ada02b3ff510813
24235 F20101219_AACXCY thomas_j_Page_047.QC.jpg
44fa53d1adec53dcf78025a7e0d2e689
68019844aceef709085f40f29f48d185c4fc4c78
37878 F20101219_AACYHD thomas_j_Page_141.QC.jpg
5ed0fee028e35b14f672a7704549a879
424f7cd70e789496add11fc53630c89004da3565
819 F20101219_AACYGP thomas_j_Page_003thm.jpg
995979c9c5f1c304c6a7ad7085593d90
260cfa0dbc9886c6554abee38678dd83842885b5
64759 F20101219_AACXEB thomas_j_Page_025.jp2
c352d4b6c0c1d5d92aaa19655876bc0e
3c998d3c6ba5bbce213685a3c612871dc1442072
F20101219_AACXDN thomas_j_Page_103.tif
480ce75c1dc47aed6060ea89492e6a45
76a209966d70b78ea7251a520fb21aede40cc929
64348 F20101219_AACXCZ thomas_j_Page_052.pro
45a83e86a69fa2b3a6627e8e0d0842d8
f26a20738277b6bdab49f7d17f5b9fa6c99cacde
23383 F20101219_AACYHE thomas_j_Page_092.QC.jpg
5512b670ea0a5b2bec32750af62ae920
e20e76599974f1c8d32a7190ec9710b394d4c907
34140 F20101219_AACYGQ thomas_j_Page_066.QC.jpg
999c948f966c92cb3c8f02eb181324c6
3a94d4b7e2df532e54818eb362a70e04c94de1ac
32791 F20101219_AACXEC thomas_j_Page_115.QC.jpg
a0a5083f76ad18e347f1e2b14291b01f
e31df76920b0aa4f08a2806c80fd8e654a8c0d02
35400 F20101219_AACXDO thomas_j_Page_027.QC.jpg
9b3a397098b88bf1d0999ff7b4797960
89d31958a9588d8f1167abcea02c3159309df9f2
26784 F20101219_AACYHF thomas_j_Page_014.QC.jpg
16b9c11123867bc5aa028df120f0bce2
cfe5bb71fdfd7e039d226bd0d95b5d53801ba807
8634 F20101219_AACYGR thomas_j_Page_060thm.jpg
80abe2c279c8b030a8d8dd9d371b67b5
db89a5b63d48b5da92ac523a8ebc121ed46d8988
27499 F20101219_AACXED thomas_j_Page_033.QC.jpg
880821a16e6869223b0247d8b8c9fd3b
77e02e5f11738e014edb3240243b409cbb14b95c
88121 F20101219_AACXDP thomas_j_Page_124.jpg
7accc6d1936540caa744159dcc396e2f
38ae2c371c7d49060de64f6a9e7c5a1d3d64ae25
8193 F20101219_AACYHG thomas_j_Page_101thm.jpg
0384db22056cf1acbe65708becf4aca7
567e018cca32c644f1ef7f42a92d07e7d8a68595
8272 F20101219_AACYGS thomas_j_Page_132thm.jpg
cec5096a81dc22418be43a80f200d07e
7e919d483a92750d54d8bb879f6f6254c552870f
43649 F20101219_AACXEE thomas_j_Page_148.pro
031a4a102e323a6b1e5c483de802aa94
b2a59d070fd0d72e8df04c1d243beaad46bc1f78
103400 F20101219_AACXDQ thomas_j_Page_029.jpg
fa3627c184caef7351ab99c3e9e006f3
06a91634f96a78378e9f9c135fe3bd647a49632e
33018 F20101219_AACYHH thomas_j_Page_102.QC.jpg
cd7fb2f6b5ec7527203361bd67bc0d92
5f9f8e86919a2c6520ff43c38d5f4f99f6bbaeb0
8433 F20101219_AACYGT thomas_j_Page_082thm.jpg
302def9fd60ec922e4da42d107995115
0d2831570de950fcb5b4603359d9c2e2718dc4b4
7053 F20101219_AACXEF thomas_j_Page_113thm.jpg
f26ae018733d4ab75213fe7665ec6804
e3f07033e4571b5e3e290cc5c653a9d90d8fc419
8300 F20101219_AACXDR thomas_j_Page_051thm.jpg
bf2b3800bfd7b41fe716f099b2fada15
48a61013441258f8edb910c85ed6868b7a793376
34132 F20101219_AACYHI thomas_j_Page_083.QC.jpg
9bbcbfac9d4f7a8355fa5b575adcd444
7d0e0cc72f110bb8556f49f840c5bff0ef38193e
8151 F20101219_AACYGU thomas_j_Page_073thm.jpg
95855a62265e4300ea8e6ed625a651b6
27543c699172f876d9fb2c8b900acedce26fc230
1876 F20101219_AACXEG thomas_j_Page_040.txt
46ac0a06d9604956949576a696f13da5
f754daf16a949cd056c52712b298409a6c7b0c70
97106 F20101219_AACXDS thomas_j_Page_099.jpg
e11a687b5e6313a53f06dd9ecfb182b6
91682f39ff684a4be0eeabc4c6d90603f5ca8881
27585 F20101219_AACYHJ thomas_j_Page_036.QC.jpg
769a56b359c33aa755f01b1edeefd50f
e433379d4e014e1bc08f6df79f84589aae2ab527
6709 F20101219_AACYGV thomas_j_Page_129thm.jpg
30df2e2fbbf549482251a30cfe150cb2
acb7e9ef708079c31157f3120edbcf85dfaa48dd
F20101219_AACXEH thomas_j_Page_129.tif
c7ff5568b7e96c0ec7d617b8ddc9c447
a5c6ccb0132d724bf81cd1e938ab029ef7f299b4
F20101219_AACXDT thomas_j_Page_040.tif
187fd21aaa568cfac9b4484a28849eee
3ceec2a14a3b9b55daaeb40c0b4522b3b89450e8
26683 F20101219_AACYHK thomas_j_Page_113.QC.jpg
6a3657222b47f7a6f0d30acc9ded9a40
d830619a2e4f2c751b205f741fcdeada85fe0c32
8783 F20101219_AACYGW thomas_j_Page_090thm.jpg
052d7fb1171fa62020314eca567be003
e24ab6994a3237a48981e7ceab1645a46d7a65b0
7413 F20101219_AACXEI thomas_j_Page_131thm.jpg
bef15d5178f57efd292b4d808074fa0a
aec54172d536af9ab31f7db628a2226a5ed75059
109293 F20101219_AACXDU thomas_j_Page_016.jp2
b29644a99452b9653ac209cf32abeaa1
dcd6f7718b9ed9e72096c43b979d4930facf85ce
7705 F20101219_AACYHL thomas_j_Page_111thm.jpg
670a0ca9bd12ebd99f4f4dc8b46cc779
9f9f527cfd948697eca181d5bacfd0384c68b454
34317 F20101219_AACYGX thomas_j_Page_090.QC.jpg
fff8e6ec3ca3dad7abb0977217dd3a63
8cb667ae73279b0c5f4063e951d0bb2ffb16749b
F20101219_AACXEJ thomas_j_Page_107.tif
48f33871f5095e50b856d169f4fc9ca8
cc53fed11f0cf4c6d841dfe4fc4dacaf3cd0724d
23724 F20101219_AACXDV thomas_j_Page_039.QC.jpg
9fad8ed5052d6120cc163a78b662d156
1d89d2a4b2e1c82d98d2ff931a8e6ee7b9bc7943
23298 F20101219_AACYIA thomas_j_Page_004.QC.jpg
8b7c528ea44974c597a01b70bb38b2db
2a0f583eaff6e8d60a8e91af363f9cd488649a14
32295 F20101219_AACYHM thomas_j_Page_099.QC.jpg
0dc52cbbe06dba92a78ae842f62f9d63
96eb1f93a5b285f370aeccc3bcc3ea074402c7d2
25224 F20101219_AACYGY thomas_j_Page_057.QC.jpg
37614f5029ba6361de819201df87e19b
788b5f45399c3204eb2946c3df104c7b48c9761b
2272 F20101219_AACXEK thomas_j_Page_009.txt
6f1ec1ccb3b01945437af3657f17ca13
85f2dbf3e04a1a4ed77916e1ecdcd0c34dc987fc
8776 F20101219_AACXDW thomas_j_Page_069thm.jpg
9e1f4fd549f6bf647256bfdabb93c2b6
9d41d49e4bfb9b599f499acf069550e8c1030416
6268 F20101219_AACYIB thomas_j_Page_004thm.jpg
54f0c9bdf5afec7bdef3367f73dbf5bd
9a42ed5d10f9ef4e9846049a495ded3750c790b4
34276 F20101219_AACYHN thomas_j_Page_052.QC.jpg
325e5b7aa01bba20b416bcaa5df8e9bf
871147c4e3ca562213a899743df343928b9a3ed4
31821 F20101219_AACYGZ thomas_j_Page_077.QC.jpg
a9e120d9aa28d1acf5b96da6c105e6ff
d67c42ac9ff272c6e5f77ac5590259293ef52d6b
2030 F20101219_AACXEL thomas_j_Page_034.txt
1d91c6ec73057e7dac32a13fd970351d
e1da9fac8d5ad4c012c8618a2778155f0695e3b2
32755 F20101219_AACXDX thomas_j_Page_084.QC.jpg
2ac69342f6583f8f9045e6cfab9767f2
4c88b52dc141e3c9a0353c86d2844290b0ce32e6



PAGE 1

IMPROVING THE TOPICAL DELIVERY OF PHENOL-CONTAIN ING DRUGS: AN ALKYLCARBONYLOXYMETHYL AND ALKYLOXYCARBONYLOXYMETHYL PRODRUG APPROACH By JOSHUA DENVER THOMAS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Joshua Denver Thomas

PAGE 3

This document is dedicated to my wife Ambe r, my daughter Miriam, and to my parents Richard and Delores.

PAGE 4

iv ACKNOWLEDGMENTS It is clear to me that every accomplishmen t in my life has been fueled by the love of my family and friends and by the wisdom and knowledge of my advisors. It is with this realization that I would like to tha nk my wife Amber for her unwavering love, support, and encouragement (especially during my first and last semesters of graduate school); and my daughter Miriam, whose smile is sometimes all I need I would also like to thank my parents, Richard and Delores, who have taught me that there is no greater purpose in life than to know my Creator. I would be remiss if I did not also thank Christopher E. Dahm, James M. Gibson, and James W. Hall for the advice and early research opportunities they provided; my committee members Margaret O. James and William R. Dolbier for their help at critical junctures in my graduate career; and Raymond Booth for graciously accepting a pos ition on my committee. Finally, I will always be indebted to Kenneth B. Sloan for his direction and immense patience and for giving me the opportunity to conduct graduate res earch. I am grateful to know him as my mentor. Above all, I would like to thank my Savi or Jesus Christ for his unconditional love and for the peace that comes from knowi ng that my life is in his hands.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii LIST OF FIGURES...........................................................................................................ix ABSTRACT......................................................................................................................xii CHAPTER 1 BACKGROUND..........................................................................................................1 Topical Delivery...........................................................................................................1 Rationale................................................................................................................1 Anatomy and Physiology of Skin..........................................................................3 Hypodermis....................................................................................................4 Dermis............................................................................................................5 Epidermis.......................................................................................................7 Barrier Properties of the Skin..............................................................................12 Physicochemical barrier...............................................................................12 Biochemical barrier......................................................................................14 Overcoming the Skin Barrier...............................................................................15 Strategies......................................................................................................15 Predictive models for optimizing topical delivery.......................................16 Prodrugs......................................................................................................................21 Acyl Prodrugs......................................................................................................23 Soft Alkyl Prodrugs.............................................................................................26 Conclusions.................................................................................................................29 2 SPECIFIC OBJECTIVES...........................................................................................31 First Objective............................................................................................................31 Second Objective........................................................................................................33 Third Objective...........................................................................................................34 3 ALKYLCARBONYLOXYMETHYL PROD RUGS OF ACETAMINOPHEN (APAP).......................................................................................................................36

PAGE 6

vi Synthesis of Alkylcarbonyloxym ethyl (ACOM) Iodides...........................................36 Coupling Reaction of ACOM Iodides with 4-Hydroxyacetanilide............................41 Conclusions.........................................................................................................51 Experimental........................................................................................................52 In Vitro Determination of Flux of ACOM Prodrugs of APAP..................................58 Materials and Methods........................................................................................59 Physicochemical properties and analysis.....................................................59 Diffusion cell experiments...........................................................................62 Results and Discussion........................................................................................65 Physicochemical properties..........................................................................65 Diffusion cell experiments...........................................................................69 Conclusions.........................................................................................................79 4 ALKYLOXYCARBONYLOXYMETHYL (AOCOM) PRODRUGS OF ACETAMINOPHEN (APAP)....................................................................................80 Synthesis of AOCOM Prodrugs of 4-Hydroxyacetanlide (APAP)............................80 Conclusions.........................................................................................................87 Experimental........................................................................................................88 In Vitro Determination of Flux of AOCOM APAP Prodrugs....................................94 Methods and Materials........................................................................................94 Physicochemical properties and analysis.....................................................95 Diffusion cell experiments...........................................................................98 Results and Discussion......................................................................................100 Physicochemical properties........................................................................100 Diffusion cell experiments.........................................................................105 Conclusions.......................................................................................................117 5 CONCLUSIONS AND FUTURE WORK...............................................................118 LIST OF REFERENCES.................................................................................................125 BIOGRAPHICAL SKETCH...........................................................................................136

PAGE 7

vii LIST OF TABLES Table page 3-1 Variation in Reacti on Conditions, Crude Yielda of 3 4 and 5 and Percentage of 1 Remaining at the End of the End of the Experimentb...........................................39 3-2 Product Distribution of the Reactiona of ACOM Halides 3 with Phenols 6 : Data Taken from the Literature........................................................................................43 3-3 Product Distribution of the Reactiona of ACOM Halides 3 with Phenols 6 : Data from the Present Work.............................................................................................45 3-4 Molar Absorptivities ( ) of APAP 6a and Prodrugs 7a-e ........................................60 3-5 Physicochemical Properties of 4-Hydroxyacetanilide 6a 4-ACOM-APAP Prodrugs 7a-e and 4-AOC-APAPa Prodrugs 8i-m ...................................................67 3-6 Log Solubility Ratios (log SRIPM:AQ), Differences Between Log SRIPM:AQ ( SR), Log Partition Coefficients (log KIPM:4.0), Differences Between Log KIPM:4.0 ( K), and Solubility Parameters ( i) for Prodrugs 7a-e .....................................................69 3-7 Flux of Total APAP Species through Ha irless Mouse Skin from Suspensions of 4-ACOM-APAP and 4-AOC-APAPa Prodrugs in IPM (JM), Second Application Flux of Theophylline through Hairless Mouse Skin from a.....................................72 3-8 Percent Intact Prodrug Detected in Receptor Phase during Steady-State (% Intact), Log Permeability Coefficients (log PM), Concentrations of APAP Species in Skin (CS), and Dermal/Transdermal Delivery Ratios for APAP 6a ,......73 4-1 Product Distribution of the Reaction of RCO2CH2X 3 with Phenols 6 Under Various Reaction Conditions...................................................................................85 4-2 Molar Absorptivities ( ) of APAP 6a and Prodrugs 7i-m ........................................96 4-3 Physicochemical Properties of 4-Hydroxyacetanilide 6a 4-ACOM-APAP Prodrugs 7a-e ,a 4-AOC-APAP Prodrugs 8i-m ,b and 4-AOCOM APAP Prodrugs 7i-m ........................................................................................................................101 4-4 Log Solubility Ratios (log SRIPM:AQ), Differences between Log SRIPM:AQ ( SR), Log Partition Coefficients (log KIPM:4.0), Differences between Log KIPM:4.0 ( K), and Solubility Parameters ( i) for Prodrugs 7i-m ..................................................103

PAGE 8

viii 4-5 Flux of Total APAP Species through Hair less Mouse Skin from Suspensions of 4-ACOM-APAP,a 4-AOC-APAP,b and 4-AOCOM-APAP Prodrugs in IPM (log JM), Second Application Flux of Theophylline through.........................................107 4-6 Percent Intact Prodrug Detected in Receptor Phase during Steady-State (% Intact), Log Permeability Coefficients (log PM), Concentrations of APAP Species in Skin (CS), and Dermal/Transdermal Delivery Ratios for.....................109

PAGE 9

ix LIST OF FIGURES Figure page 1-1 Structure of Acylglucosylceramide and General Orientation in Lamellar Bodies...10 1-2 Structure of Ceramides f ound in Human Stratum Corneum....................................12 1-3 Tortuous Path of Permeant Through the Stratum Corneum and Expanded View of Alternating Nonpolar (White Bands, Electron Lucent) and Polar (Dark Bands, Electron Dense) Phases Found With in the Intercellular Matrix..............................13 1-4 Bioconversion of Minoxidil to Minoxidil Sulfate by S calp Sulfotransferase in the Presence of 3 -Phosphoadenosine-5 -phosphosulfate (PAPS)...........................23 1-5 Structures of Acyl Prodrugs for the Topical Delivery of Captopril Testosterone, and Acetaminophen..................................................................................................25 1-6 Most Common Mechanisms by whic h Acyl Prodrugs are Hydrolyzed Chemically...............................................................................................................26 1-7 Mechanism of Hydrolysis of Soft Alkyl Prodrugs (Alkylcarbonyloxymethyl and Hydroxymethyl Derivatives are shown) and Comparison to Metabolism of Hard Alkyl Derivatives (General Mechanism......................................................28 1-8 Examples of Alkylcarbonyloxymethyl (ACOM) and Alkyloxycarbonyloxymethyl (AOCOM) Prodrugs..................................................29 2-1 Phenol-Containing Therapeutic Agents th at may benefit from Topical Delivery via Alkylcarbonyloxymethyl (ACOM) or Alkyloxycarbonyloxymethyl (AOCOM) Derivatization........................................................................................34 3-1 Reaction of Trioxane 1a and Parald ehyde 1b with Acid Chlorides in the Presence of NaI........................................................................................................37 3-2 General Reaction of Alkyl carbonyloxymethyl (ACOM) Halide 3 with Phenol 6 to Give Aryl Acylal 7 and Aryl Ester 8 ....................................................................42 3-3 Structures of ACOM Deriva tive of a Protected Amino Acid 9 (R = Protecting Group) and its Corresponding Aliphatic Derivative 10 and Structure of Byproduct 11 ..........................................................................................................s43

PAGE 10

x 3-4 Reaction of ACOM Iodides 3a-f with Phenols 6a-c ................................................44 3-5 Plot of the Percentage of 4 (RCO2CH2Cl) in Crude 3 Versus the Ratio of 8 / 7 (Acylated/Alkylated phenol) Resulting from the Reactions of 3a-3e with 6a and 6b (Taken from Entry 4, Table 3-2 and Entries 1-4, and 8, Table 3-3 .................46 3-6 Plot of Chartons steric parameter for R Versus the Ratio of 8/7 (Acylated/Alkylated Product) Re sulting from the Reactions of 3a-3e with 6a and 6b (Taken from Table 3-2: Entry 4, Table 3-3: Entries 1-4, and 8 and Entry 5...48 3-7 Speculative Mechanism for Reactions of Protected Amino Acid Derivatives 9 with Phenols 6 ..........................................................................................................49 3-8. Structure of 4-Hydroxyacetanilid e and Corresponding 4-ACOM Prodrugs..............58 3-9. Diagram of Franz Diffusion Cell (Metal Clamp Not Shown)...................................63 3-10 Flux of Compound 7a through Hairless Mouse Skin...............................................65 3-11 Structure of 4-alkyloxycarbonyl (AOC) derivatives of APAP................................68 3-12 Plot of Solubility Parameter vers us Log P for 4-ACOM-APAP Prodrugs 7a-e ....74 3-13 Log SIPM ( ), Log S4.0 ( ), Log KIPM:4.0 ( ), and Log JM ( ) Values for APAP 6a 4-ACOM-APAP Prodrugs 7a-e and 4-AOC-APAP Prodrugs 8i-m .................76 3-14 Plot of Experimental Versus Calcul ated Flux for 5-FU, 6-MP, and Th Prodrugs ( n = 53), APAP ( ), 4-AOC-APAP Prodrugs ( n = 5, plus two additional compounds mentioned in Reference 1 to give n = 7)...............................................78 4-1 Synthetic Routes to Alkyloxy carbonyloxymethyl (AOCOM, R = Oalkyl) Prodrugs of 4-hydroxyacetanilide (APAP)..............................................................81 4-2 Generalized Reaction of AOCOM halides (R = Oalkyl) and ACOM halides (R = alkyl) 3 with phenols 6 .............................................................................................82 4-3 Reaction of AOCOM iodides with ph enols under phase-transfer conditions..........83 4-4 Plot of Charton s Steric Parameter for R Versus the Ratio of Acylated/Alkylated Product ( 8/7 ) Resulting from the Reactions of 6 with AOCOM Iodides (Entries 3-5 in Table 4-1, ) and ACOM Iodides (Entries 14.....87 4-5 Plot of Charton s Steric Parameter for R Versus the Ratio of Acylated/Alkylated Product ( 8/7 ) Resulting from the Reactions of 6 with AOCOM Iodides (Entries 6-11 in Table 4-1, ) Under Phase-Transfer..................87 4-6 Structure of 4-Hydroxyacetanilid e (APAP) and Corresponding 4-AOCOMAPAP Prodrugs........................................................................................................94

PAGE 11

xi 4-7 Flux of Compound 7j through Hairless Mouse Skin.............................................100 4-8 Structures of Alkylcarbonyloxymethyl (ACOM) and Alkyloxycarbonyl (AOC) Derivatives of APAP and Comparisons between Homologs of Approximately Equal Size...............................................................................................................104 4-9 Plot of Solubility Parameters versus Log PM for 4-AOCOM-APAP Prodrugs 7im .............................................................................................................................110 4-10 Log SIPM ( ), Log S4.0 ( ), Log KIPM:4.0 ( ), and Log JM ( ) Values for APAP 6a 4-ACOM-APAP Prodrugs 7a-e 4-AOC-APAP Prodrugs 8i-m, and 4AOCOM-APAP Prodrugs 7i-m .............................................................................111 4-11 Plot of Experimental Versus Calcul ated Flux for 5-FU, 6-MP, and Th Prodrugs ( n = 53), APAP ( ), 4-AOC-APAP Prodrugs ( n = 5, plus two additional compounds mentioned in Reference 1 to give n = 7), 4-ACOM-..........................113 4-12 Plot of Experimental Versus Calcul ated Flux for 5-FU, 6-MP, and Th Prodrugs ( n = 53), APAP ( ), 4-AOC-APAP Prodrugs ( n = 5, plus two additional compounds mentioned in Reference 1 to give n = 7), 4-ACOM-..........................114 4-13 Plot of Experimental Versus Calcul ated Flux for 5-FU, 6-MP, and Th Prodrugs ( n = 53), APAP ( ), 4-AOC-APAP Prodrugs ( n = 5, plus two additional compounds mentioned in Reference 1 to give n = 7), 4-ACOM-..........................116 5-1 Structures of Naproxen, Naproxen Prodrugs,133, 135 Proposed Methylpiperazinyl ACOM and AOCOM Prodrugs of APAP and Potential Mechanism for Hydrolysis of Methylpiperz inylmethyloxycarbonyloxymethyl.............................122

PAGE 12

xii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy IMPROVING THE TOPICAL DELIVERY OF PHENOL-CONTAIN ING DRUGS: AN ALKYLCARBONYLOXYMETHYL AND ALKYLOXYCARBONYLOXYMETHYL PRODRUG APPROACH By Joshua D. Thomas August 2006 Chair: Kenneth B. Sloan Major Department: Medicinal Chemistry Although most drugs are administered orall y, this route is not suitable for many compounds due to their extensive metabolism in the GI tract and liver Topical delivery is an alternative route of admi nistration for such drugs that a voids this first-pass effect and permits the drug to enter the systemic circulation following penetration of the skin a much less metabolically active tissue than th e liver. One of the most effective methods for improving topical delivery while minimi zing side effects involves the use of prodrugs. Most previous attempts to improve the topical delivery of phenols via a prodrug have involved some type of aryl ester, carbonate or carbamate. In the present work, alkylcarbonyloxymethyl (ACOM) and al kyloxycarbonyloxymethyl (AOCOM) prodrugs of 4-hydroxyacetanilide (acetaminophen) have been evaluated in vitro as novel permeation-enhancing derivatives of phenol -containing drugs. Alkylcarbonyloxymethyl iodides were synthesized by way of a new one -step route and were subsequently reacted

PAGE 13

xiii with various phenols to obtain the target ACOM derivatives. The coupling reaction between ACOM iodides and phenols was shown to favor the alkylated product regardless of the steric hindrance in th e alkylating agent or the pheno l. On the other hand, the coupling reaction of AOCOM iodides with phenols seemed to be more sensitive to steric effects, with the acylated product being favor ed when steric effects were minimal. However, under phase-transfer conditions, th e influence of steric hindrance was minimized and yields of AOCOM phenol were increased. More importantly, the ACOM and AOCOM prodrugs were able to improve the topical delivery of APAP up to 3.6 and 1.3-fold, respecti vely. The ACOM and AOCOM prodrugs were also added to the Roberts-Sloan database (n = 61) to obtain a new database of 71 compounds. A fit of this new database (n = 71 r2 = 0.92) to the Roberts-Sloan ( RS ) equation resulted in a more robust model for predicting flux (JM) through hairless mouse skin: log JM = -0.562 + 0.501 log SIPM + 0.499 log S4.0 0.00248 MW where SIPM and S4.0 are the solubilities in is opropyl myristate and pH 4.0 bu ffer, and MW is molecular weight.

PAGE 14

1 CHAPTER 1 BACKGROUND Topical Delivery Rationale Although there are many available routes of drug administration, the oral route is by far the most popular. This is primarily due to a high incidence of patient compliance. While it is true that patients often find an oral drug regimen more palatable than the parenteral alternative (e.g., intravenous or intramuscular injection), oral drug absorption is a much more complicated problem1, 2 for the drug discovery scient ist to solve. If given orally, a drug molecule must surmount num erous chemical and enzymatic hurdles in order to reach the systemic circulation. For example, if the drug survives the acidic environment of the stomach, it still f aces efflux transporters and various biotransformation enzymes in the gut wall. Following absorption in the gut, the drug enters the liver, where a host of biotransfo rmation enzymes await. At each stage of absorption, there is the potential for the dr ug to be inactivated and excreted, thereby reducing the amount of the original dose that reaches the intended site of action in the body. Given the extent to which a drug can be in activated as it is absorbed into the systemic circulation, alternative methods that avoid first-pass metabolism yet retain the simplicity needed to achieve high patient comp liance are desirable. Topical delivery is one such approach. In general, the levels of drug-metabolizing enzymes in the skin are much lower than those in the liver and intestine.3-6 For example, transferase activity in

PAGE 15

2 the skin (e.g., glucuronidati on and sulfation) may appro ach 10% of the liver while cytochrome P-450 activity in the skin is t ypically 1-5% of the corresponding hepatic activity.7 In fact, skin permeability rather than drug metabolism appears to be the major barrier to topical bioavailability.8, 9 Although it is an important consider ation in drug delivery, minimal drug inactivation is not the only advantage to be gained from avoidance of first-pass metabolism. Potential side effects must also be taken into account Topically applied drugs frequently exhibit fewer side effects than the corresponding oral dosage forms. One of the most studied medications in that respect is estrogen. Several recent studies have indicated that the detrimental eff ects of hormone replacement therapy in postmenopausal women may be due to the route of dr ug administration.10-13 In a comparison between oral and transdermal estrogen therapies, both treatments were equally effective at increasing bone mineral density and decreasing luteinizing hormone levels.10 However, patients treated with oral estrogen for six months experienced an increase in triglyceride levels and fat ma ss with an accompanying decrease in lean body mass. Triglyceride levels and body compositi on of patients treated with transdermal estrogen did not significantly change over the course of the six month treatment.10 Other studies indicate that or al estrogen may play a role in the elevated levels of Creactive protein (CRP)11, 13 and serum amyloid A (SAA)12 detected in women undergoing hormone-replacement therapy. These studies found no such side effects in patients undergoing transdermal estrogen therapy. In both cases, the evidence suggests that the differences in side effects between the routes of administration are di rectly related to the action of oral estrogen in the liver.11-13 Since both CRP and SAA have been identified as

PAGE 16

3 important indicators of systemic inflammation and are predictive of future cardiovascular disease,14 transdermal estrogen replacement ther apy appears to offer a better safety profile than the more common oral route. In fact, in the case of SAA, transdermal estrogen may exert a protective effect compared to the oral route. Abbas and coworkers12 found that the levels of SAA and the SAAHDL complex (HDL-SAA) in postmenopausal women receiving transdermal estrogen were substantially lower than those in women receiving oral estrogen. While the examples given above for es trogen support the case for transdermal delivery (to the systemic circulation), it is pe rhaps more obvious that topical delivery is an important route for treating skin diseases (dermal delivery). The main advantage of topical over oral administration fo r the treatment of skin diseases is that high levels of the drug can be delivered to the skin with mi nimal exposure to the rest of the body. One example of the benefits of topical delivery for the treatment of a skin condition is the topical application of dapsone (4,4-sulfonyldianiline).15, 16 Although dapsone is normally given orally for the treatment of leprosy,17 oral dapsone has also proven effective in treating moderate cases of acne.15 However, the effectiveness of orally administered dapsone is limited due to its hemotoxic side effects. In a recent study, topically applied dapsone was successfully used to treat moderate acne with side effects no different than those of th e vehicle (a gel) itself.16 Anatomy and Physiology of Skin Although topical delivery presents fewer comp lications than the oral route, this does not mean that overcoming the barrier proper ties of the skin is a small task. Unlike the gastrointestinal tract, th e primary purpose of skin is to restrict the passage of endogenous and exogenous substances into a nd out of the body. As a consequence,

PAGE 17

4 topical delivery is a viable option for a re latively small percenta ge of drugs. For example, all the drugs currently approved for use by the FDA as transdermals have molecular weights less than 400 Da, exhibit relatively high lipid solubility, and are therapeutically e ffective at low doses (0.04-10 ng/ml).3, 9 Furthermore, since most transdermal drug candidates were origin ally designed for oral administration,18 they typically do not possess the pa rticular physiochemical properties required for adequate diffusion through skin.19 Although the relationship between flux and the physicochemical properties of the pe rmeant is still a matter of debate,20 a knowledge of skin anatomy and physiology is helpful in understanding why some compounds permeate the skin better than others. The skin is composed of three main laye rs of varying thickness: the hypodermis (12 mm), dermis (1-5 mm), and epidermis (60-120 m).3, 5 The actual composition of each layer varies with age, dis ease state, and anatomical lo cation. Though one might expect the thickest of these layers to be the primary barrier to pe rcutaneous absorption, this is not the case. In fact, the most impervious la yer of the skin is actually the thinnestthe outermost layer of the epidermis which is refe rred to as the stratum corneum (10-20 m). Although diffusion through the stratum corneum is generally recognized as the ratelimiting step to percutaneous absorption, disrup tions in the integrity of the other layers can also affect skin permeability. Thus, the structure and function of each layer will be reviewed in the following sections. Hypodermis The deepest layer of the sk in, the hypodermis, is primarily composed of adipose tissue. As such, it functions as an energy depot, a layer of insulation, and as a shock absorber. As with the other layers of th e skin, the thickness of the hypodermis varies

PAGE 18

5 from one part of the body to another. For instance, the eyelids are altogether missing a hypodermal layer. Variations in diet can aff ect the thickness of th is layer as well. The hypodermis serves as the entry poin t for the major blood vessels and nerves that service the skin. Although adipose tis sue may sometimes function as a depot for highly lipophilic xenobiotics, this is gene rally not the case with the hypodermis. Compounds that reach this layer by diffusi on are usually taken up by the network of blood vessels that run throughout the subcutan eous fat. Because the loose connective tissue of the hypodermis is inte rwoven with that of the de rmis, there is no distinct boundary between these two layers In addition, although most ha ir follicles originate in the dermis, course hair can often extend deep (3 mm) within the hypodermis.3, 5, 21 Dermis Directly above the hypodermis is the dermis the thickest layer of the skin. In sharp contrast to the underly ing layer of adipose tissue, the dermis is a much more aqueous-like environment. For instance, the gelatinous substance in which the various structures of the dermis are imbedded consists of proteoglycans and glycosaminoglycanscompounds that are cap able of binding up to 1000 times their weight in water. Running throughout this gel-like ground substance is a dense, irregular network of collagen fibers. Thes e fibers make up the bulk of the dermal connective tissue and act as a supporting fram ework for blood vessels, hair follicles and various other structures. Micr ofibrils composed of elastin, fibrillin, and vitronectin make up the elastic connective tissue (the second mo st abundant tissue in the dermis), and provide a certain amount of elasticity to the skin.3, 5 Most of the appendages of the skin origin ate in the dermis. These include the hair follicles, sebaceous glands, and sweat glands. As with other features of the skin, the

PAGE 19

6 density and presence of these structures vary with anatomical locat ion. For example, of these three appendages, only the sweat glands are found in the palms and soles. Hair follicles are sheath-like structures that enclose each hair and extend from the surface of the skin into the dermis. Although the follicle consists of living epidermal cells, the hair shaft inside the follicle is mainly composed of dead, keratinized cells. Attached to the follicle is a band of smooth muscle fibers th at are collectively known as an arrector pili muscle. Under conditions of emotional st ress or cold temperatures, these muscles contract, causing the hair to stand erect and the skin to take on the familiar goose bump appearance. In most regions of the skin, se baceous glands merge with hair follicles and secrete their contents (sebum) directly into the follicle. However, in various sites throughout the body the sebaceous glands extend to the outermost layers of the skin and deposit their contents directly at the surface. In a similar fashion, sweat glands either connect to the hair follicle (as in apocrine glands) or open up at the skin surface (as in eccrine glands). Sebum (a mixture of fatty acids, triglycerides, and wax secreted by the sebaceous glands) and sweat (a mixture of salts and various wast e products (e.g., urea and uric acid)) help keep the surface of the sk in slightly acidic (pH 5). With regard to topical delivery, skin appendages may offe r an alternative pathway to permeating compounds that avoids the stratum corneum. However, since the appendages make up a such a small percentage of the total surface area of the skin (approximately 0.1%), these shunt routes are not expected to signif icantly affect the observed flux of most permeants.3, 5, 21 The dermal-epidermal border resembles a transverse wave runni ng parallel to the skin surface. As a result of these undulations (referred to as dermal papillae), sections of

PAGE 20

7 the dermis come within 200 m of the skin surface. Capillaries also extend into the dermal papillae and help maintain sink conditions within the skin by efficiently transporting permeated compounds to the system ic circulation. In addition, the vascular network of the dermis is responsible for s upplying nutrients and oxygen to the skin and also plays a role in regulating body temper ature. A system of lymphatic vessels comprises an additional dermal circulatory system. These vessels are involved in removing cellular waste and help regulate th e volume of the interstitial fluid in the dermis. During times of wound healing and inflammation, the lymphatic system also delivers macrophages, lymphocytes, and leucocytes to the affected areas of the dermis. These cells facilitate the healing process by destroying i nvading bacteria via phagocytosis or via the secretion of certai n cytotoxic agents. In general, the lymphatic circulatory system plays only a minor role in the cl earance of permeated compounds from the dermis.3, 5 Epidermis Directly above the dermis lies the epidermi s. The epidermis is composed of four distinct regions, each represen ting a different phase of kert inocyte differentiation. From the dermal-epidermal border to the skin su rface they are the stra tum basale, stratum spinosum, stratum granulosum, and stratum co rneum. Since there are no blood vessels in this layer of the skin, nutrients reach epid ermal cells by way of passive diffusion across the basement membrane at the dermal-epiderm al border. The passage of nutrients and other materials across the basement membrane is facilitated by the relatively high surface area provided by the dermal papillae. The final stage of keratinocyte differentiation is represented by the stratum corneumthe out ermost layer of skin. Although it is

PAGE 21

8 essentially dead tissue, the stratum corneum is the rate-limiting barri er to percutaneous absorption.3, 5, 21 Stratum basale. Keratinocytes of the stratum basale are unique in that they are the only epidermal cells that undergo mitosis. Following mitotic division, one cell remains in the stratum basale while the other daughter cell detaches from the basement membrane and migrates outward through the remaining ep idermal layers. Basal keratinocytes are attached to the basement membrane by structures known as hemidesmosomes. Similar desmosome plaques are found throughout the epidermis and function as proteinaceous rivets linking adjacent cells. Other cell types found in the stratum basale include melaninocytes, Langerhans cells, and Merkel cells. Melaninocytes are responsible for producing the pigment melanin. Though melani n is produced by the melaninocytes, it is also transferred to neighbor ing cells through dendritic conne ctions. Langerhans cells play an important role in the immune re sponse by binding to foreign antigens in the epidermis and presenting them to T-lymphocyt es in the lymph nodes. Merkel cells are involved in sensory reception and are found at sites along th e basement membrane where dermal nerve endings extend into the papillae.5 Stratum spinosum. Upon migration from the stra tum basale to the stratum spinosum, keratinocytes undergo several mor phological changes including the formation of desmosomal plaques between adjacent ce lls. These intercellular linkages make substantial contributions to the overall cohesiveness and organization of the epidermis. Besides forming desmosomes, the keratinocytes of this layer also lose their columnar shape and begin to take on a more flattened appearance. Both the volume and diameter of the keratinocyte continue to increase as the cell makes its way through the remaining

PAGE 22

9 strata. In addition to changes in structure, keratinocytes also begin to synthesize keratins 1 and 10 and develop special organelles called lamellar granules that play an important role in maintaining the barrier properties of the stratum corneum.5, 22 Stratum granulosum. At this stage of keratinocyt e differentiation, the cell begins to die and the nucleus and orga nelles are enzymatically degraded. As the name suggests, the cells of the stratum granulosum (SG) ar e filled with keratohyalin granules (KHGs) and lamellar bodies (LB, also known as lamella r granules). Kerat ohyalin granules are enriched in the precursors of intracellular corneocyte proteins and of the cornified envelope. Included among these precursors are profillaggrin, loricrin, and keratins 1 and 10. Lamellar bodies are ovoid organelles c ontaining stacks of lipid membranes composed of phospholipids, cholesterol, and glucosylceramides. In addition, LB contain high levels of various catabo lic enzymes including acid hydrolases, sphingomelinase, and phospholipase A2.3 The accordion-like appearance of these lipoidal structures likely results from the compression and subsequent stacking of Golgi-deri ved lipid vesiclesa process thought to be mediated by acylglucosylceramide (Figure 1-1).23 The incorporation of the glucose a nd linoleic acid moieties into a -hydroxyceramide backbone allows acylglucosylceramide to be an chored in the polar pha se of one vesicle, span the lipid interior, and in sert itself into the polar su rface of an adjacent vesicle thereby functioning as a molecular rivet.8 At the stratum corneum-stratum granulosum interface, lamellar bodies are excreted from the cell and their contents made ready for incorporation into the stratum corneum (SC).

PAGE 23

10 O N H O O O O H OH OH O H O OH Vesicle Membrane Lipid Phase Polar Phase Lipid Phase Vesicle 2 Lipid Phase Vesicle 1 Figure 1-1: Structure of Ac ylglucosylceramide and Genera l Orientation in Lamellar Bodies Stratum corneum. Although the stratum corneum (SC) is the last major layer of the epidermis, it can be further divided into inner (stratum compactum) and outer (stratum disjunctum) layers. As the name im plies, the cells of the stratum compactum are packed together more tightly than those of the stratum disjunctum. This difference in packing and cell cohesion between the two layers is primarily due to the loss of linkages (corneodesmosomes) between cells in the outer laye r in a process known as desquamation. At the SG-stratum compactum interface, LB fuse with one another24 to form the intercellular lipid lamellae of the stratum corneum. The cells of the SC are known as corneocytes. They are nonliving and are generally considered to be impermeable to most compounds. Compared to the other layers of the skin, the overal l water content of the SC is quite low (approximately 15% by weight versus 70% for viable epidermis)the majority of which is associated with the proteinaceous material (mainly keratins 1 and 10 and various degradation products of filaggrin) that comprises the inne r compartment of the corneocytes.3 An impermeable membrane (the cornified envelope) composed of highly cross-linked protein encloses the core protein of the corneocytes. This me mbrane not only functions as a barrier to

PAGE 24

11 permeation, but it also plays an important role in the organization of the intercellular lipid lamellae via the interaction of -hydroxyceramides that are covalently bound to the exterior surface of the cornified evelope.3, 8 These particular lipids are derived from ceramides 1, 4 and 9 (Figure 1-2) by a deeste rification reaction that removes the linoleic acid group. The primary constituen t of the exterior lipids is a -hydroxyceramide derived from ceramide 1 which itself is deri ved from another important LB lipid (i.e. acylglucosylceramide, Figure 1-1). These very long chain lipids are li kely attached by an ester linkage at the -hydroxyl end to a surface protei n (possibly involucrin) on the envelope.8 The intercelluar lamellae of the SC consist of the following three lipids in their approximate order of abundance: ceramide s (50% by weight), cholesterol (30% by weight) and free fatty acids (10% by weight).22 To date, nine different ceramides (Figure 1-2) have been isolated from human SC. Th ey have traditionally been labeled in a way that reflects their relative polar ities on thin layer chromatography (TLC). In that regard, it should be noted that the recently discovere d ceramide 9 exhibits a retardation factor (Rf) on TLC that is between ceramides 2 and 3. Interestingly, ceramide 1 may also serve the same molecular rivet role in the lipid lamellae as its precursor, acylglucosylceramide, does in LB.8 Although it is evident from the brie f overview presented here that the composition of the SC is much different from the plasma membranes found in most tissues of the body, this point is further emphasized by the absence of phospholipids in the SC.5

PAGE 25

12 N H O OH OH O O Ceramide 1 N H O OH OH Ceramide 2 N H O OH OH O O OH Ceramide 9 N H O OH OH O O OH Ceramide 4 N H O OH OH OH Ceramide 5 N H O OH OH OH OH Ceramide 7 N H O OH OH OH Ceramide 3 N H O OH OH OH Ceramide 6 N H O OH OH OH OH Ceramide 8 Figure 1-2: Structure of Ceramide s found in Human Stratum Corneum Barrier Properties of the Skin Physicochemical barrier The primary barrier to percutaneous absorption is presented by the SC.3, 5 Given that the enzymatic activity of the SC is much lower than that of the viable epidermis and dermis,3 the barrier properties of the SC are mainly physicochemical rather than

PAGE 26

13 biochemical in nature. One of the key features of this barr ier is the organization of the corneocytes within the intercellular matr ix. As the corneocytes are practically impermeable to most compounds, they act as road blocks in the path of diffusion. In fact, Potts and Francoeur25 have shown that the diffusion of water throug h the SC is 1000-times lower than its diffusion through a comparable homogeneous lipid phase. They also found that the diffusion pathlength was 50-times greater th an the thickness of the membrane. From these results, it was concluded that the diffusion of permeants through the SC occurs by way of a meande ring path around the corneocytes and through the intercellular lamellae (Figure 1-3).25 Figure 1-3: Tortuous Path of Permeant Through the Stratum Corneum and Expanded View of Alternating Nonpolar (White Bands, Electron Lucent) and Polar (Dark Bands, Electron Dense) Phases F ound Within the Inte rcellular Matrix (Phases Presented as they Generally A ppear in Ruthenium Tetroxide Fixation of Normal Skin)24 As the intercellular lamellae represent th e only continuous pathway in the SC, the composition and organization of the lipids in this matrix are of primary importance to percutaneous penetration. Due to the edge -to-edge fusion of the LB at the SG-SC interface, the intercellular domain is composed of continuous lipid sheets consisting of repeating units of polar a nd nonpolar phases (Figure 1-3).24, 26 Despite the high degree of

PAGE 27

14 order within the lipid lamellae, the lipi d phases are often interrupted by hydrophilic bridges that link two neighboring polar phases. As a conseque nce of the structure of the intercellular matrix, a permeant must pass th rough alternating lipi d-poor and lipid-rich layers. The implication for drug design is that in order to maximize flux, the solubilities of the drug in both lipid and aque ous solvents must be increased.19, 20 Biochemical barrier Although the skin is primarily a physical barr ier, the enzymatic activity of the skin is significant and should not be ignored. Of the three main skin layers, the epidermis exhibits the highest enzymatic activity per unit tissue mass and is considered the major region of drug metabolism in the skin.27 Many of the major types of phase I and phase II reactions are known to occur in the skin including oxidation, reduc tion, ester hydrolysis, epoxide hydrolysis (microsomal and cytosolic) methylation, glucur onidation, sulphation, glycine conjugation, and glutathione conjugation.6, 27, 28 It is particul arly important to note that many of the cytochrome P450 enzy mes responsible for metabolizing a wide variety of pharmaceutical compounds in the liv er and gut are also found in the skin.4 One of the major obstacles to the oral abso rption of drugs is the presence of efflux transporters such as multidrug resistance-associated proteins (MRP) and P-glycoprotein (P-gp) in the gut wall.1 Early attempts to determine the tissue distribution of P-gp found evidence of this protein in the liver, pancre as, intestine, and kidney but were unable to detect P-gp in the skin.29 However, recent work in this area has shown that the skin contains several constitutively expressed MR Ps (1 and 3-6). P-gp was also found but only after induction wi th dexamethasone.30 Current knowledge about the function of MRP in the skin is limited.31 In contrast to its infamous role as a contributor to multidrug resistance, Randolph and coworkers have demons trated that P-gp plays an important role

PAGE 28

15 in the migration of Langerhans cells out of the skin by way of the lymphatic vessels.32 Thus P-gp helps maintain a healthy immune res ponse in the skin. Li and coworkers have also found evidence to suggest that MRP-1 acts as an effl ux transporter in the skin.33 Specifically, they found that th e tissue-to-plasma c oncentration ratio of grepafloxacin in the skin of MRP-1 knockout mice was higher th an the corresponding ratio in the skin of wild type mice following an i.v. injecti on of grepafloxacin. Other experiments demonstrated that the uptake of another MRP1 substrate (fluo 3) into the keratinocytes was significantly increased in th e presence of an MRP-1 inhibitor.33 Though these results provide evidence of active tran sport of xenobiotics out of ke rainocytes via MRP-1, it is unclear whether such transport would ultimatel y result in the expulsion of the xenobiotics to the skin surface (though this does not seem lik ely given the nature of the SC barrier). Yet if an active xenobiotic efflux system ex ists in the skin, it would probably have a greater effect on delivery into the skin (dermal delivery) rather than through it (transdermal delivery). In short, the presence of efflux transporters in the skin raises the possibility of an additional biochemical barrier (efflux transport out of the skin) to skin permeability, but the current evidence fo r such a barrier is not definitive. Overcoming the Skin Barrier Strategies Much research has gone into developi ng effective methods for overcoming the barrier properties of the skin.5, 9, 34 Typical examples of such strategies incl ude the use of electricity35 to either create temporary holes in the skin (electroporation) or to electrostaticly push charged drug molecules into the skin (iontophoresis); penetration enhancers,36 chemicals designed to temporarily decrea se the barrier proper ties of the skin; microneedles37 which physically create micron-si zed holes in the skin through which

PAGE 29

16 drug molecules bypass the stratum corneum altogether; and prodrugs19, 20 which are transient derivatives of active drugs that tem porarily improve the solubility of the drugs in the skin (thereby increasing their flux through the skin) and then rapidly convert to the parent drugs in the skin or in the systemic circulation. Of the methods listed above penetration enhancers have received the most attention in industry. Howeve r, despite this predilection for chemical enhancers, the improvement in drug flux is often only modest at best.9 Moreover, the enhancing effects are often directly proportional to the concentratio n of the enhancera situation which often results in toxic side effects.36 In order to reduce or av oid the adverse side effects associated with penetration enhancers, it has been suggested19, 38, 39 that a prodrug/formulation combination might be a better way to approach the problem. In many cases, a drug molecule exhibits poor solubil ity in the skin due to one or more polar functional groups in the molecu le that are either highly ch arged at physiological pH or that promote hydrogen bonding and high crysta l lattice energies. A prodrug approach attempts to overcome this problem by te mporarily masking the offending functional group. Since the prodrug is alr eady more soluble in the skin than the parent drug, a much lower concentration of the chemical enhan cer would be needed to experience great improvement in drug permeability. Predictive models for optimizing topical delivery As mentioned in previous sect ions of this chapter, the intercellular lipid matrix of the SC is the rate-limiting barrier to the passi ve diffusion of drugs through skin. Due to the particular arrangement of the intercel lular lipid lamellae (Figure 1-3), permeating compounds must pass through alte rnating polar and nonpolar laye rs within the SC. On this knowledge alone one might expect percutaneous absorption to be positively

PAGE 30

17 dependent on lipid and aqueous solubilities. Though such dependency is most clearly seen in homologous series of prodrugs in which the homolog exhibiting the highest flux also exhibited the best balance of hi gh lipid and high a queous solubilities.20, 38 Although such qualitative relationships can serve as a general guide for optimizing topical delivery, a mathematical model for accurately predic ting permeation through skin based on easilydetermined physicochemical properties would be of even greater value as a tool for quickly identifying lead compounds (i.e. those compounds expected to exhibit the highest flux). Mathematical modeling of diffusion thr ough a complex heterogeneous membrane like the skin can be a formidable challenge. However, the problem can be simplified by assuming that the skin behaves like a homoge neous membrane. Once this assumption is made, most quantitative treatments of skin pe rmeability data begin by considering Ficks first and second laws of diffusion expre ssed by equations 1 a nd 2, respectively: J = -D( C/ x) (1) C/ t = D( 2C/ x2) (2) Ficks first law (equation 1) states that the amount of material passing through a given area of a homogeneous membrane over time (f lux, J) is directly proportional to the concentration gradient across the membrane wh ere D (the diffusion coefficient) functions as the proportionality coefficient. Ficks s econd law (equation 2) stat es that the rate at which the concentration changes ( C/ t) at any point within the membrane is proportional (again, D is the proportionality coeffi cient) to the rate of fluctuation in the concentration gradient at that point ( 2C/ x2).40 If the concentration of the permeant in the first layer of skin does not change with time, equations 1 and 2 simplify to equation 3:

PAGE 31

18 J = (D/L)(CMEM C0) (3) where L is the distance traveled by the permean t on passage through the skin (note: this is not the same as the thickness of the skin; se e Physicochemical Barrier section above) and CMEM and C0 are the concentrations of the permeant in the first and last layers of the skin. For all practical purposes, the body func tions as a limitless rese rvoir on one side of the skin where the concentration of the permean t is essentially zero (i.e. sink conditions). In this case, CMEM >> C0 and equation 3 reduces to J = (D/L)(CMEM) = (D/L)(KMEM:V)CV (4) where KMEM:V is the partition coefficient between th e membrane and the vehicle (solvent) in which the permeant has been applied, and CV is the concentrati on of the permeant in the vehicle.41 In the development of the Ka sting-Smith-Cooper (KSC) model,41 the authors noted that in order to make reliable comparis ons of flux the experi mental conditions under which flux was measured should ensure th at each permeant exhibited the same thermodynamic activity. To meet this require ment, Kastings and coworkers decided to only consider those cases in which the perm eant is applied as a saturated solution (CV = SV, where SV is the solubility in the vehicle). Th is approach ensures that each permeant experiences the same thermodynamic drivi ng force since each permeant is at its respective maximum concentration (i.e. saturati on) in the first layer of the skin. Under these conditions, equation 4 becomes JM = (D/L)(SMEM) = (D/L)(KMEM:V) SV (5) where JM is the maximum flux, and SMEM is the solubility in the sk in. In order to arrive at the diffusion coefficient D, Kasting et al assumed41 that diffusion through the

PAGE 32

19 intercellular lipids of the SC can be approximated from similar models that describe diffusion through polymer membranes. By this approach, D becomes D = Do exp (MV) (6) where Do is the diffusivity of a hypothetical molecule having zero molecular volume,42 is a constant that is specific to the skin,43 and MV is molecular volume. The value for SMEM in equation 5 was either calculated from ideal solution theory or was assumed to be approximately equal to the solubility in a model lipid (SLIPID) such as octanol (SOCT).41 The general form of the KSC model is shown below in logarithmic form: log JM = log (Do/L) + log SMEM ( /2.303) MV (7) As noted by Potts and Guy,42 one of the weaknesses of the KSC model is the assumption that SOCT can approximate the solubilizing capa city of the intercellular lipids of the SC (SMEM). To account for the differences between SMEM and SOCT, Potts and Guy proposed that when the vehicle is water (SV = SAQ), KMEM:AQ and KOCT:AQ are related by equation 8 KMEM:AQ = (KOCT:AQ)y (8) in which the coefficient y is a measure of the similarities between the two partitioning domains. Since SMEM = (KMEM:AQ)(SAQ) (9) substitution of equation 9 in to equation 7 gives the following equation for flux log JM = log (Do/L) + y log KOCT:AQ + log SAQ MW (10) where molecular weight (MW) has been substituted for mol ecular volume and = /2.303 but also includes a conversion f actor for using MW in place of MV.42 Whereas

PAGE 33

20 the Potts-Guy model (PG)42 is an expression of the permeability coefficient (log P = log J log SV) equation 10 is a modified vers ion of PG that describes flux. Though equation 10 is an improvement over KCS, it suffers from the fact that it only applies to aqueous vehicles. Furthermor e, it offers little insight into the relative impact of aqueous and lipid solubilites on flux since the SOCT term is hidden within KOCT:AQ. In order to address thes e issues, Roberts and Sloan43 were able to extend the applicability of equation 10 to vehicles other than water in a model which clearly shows the dependency of flux on aqueous and lipid solubilities. Using isopropyl myristate (IPM) as example of when a lipophilic vehicle is applied, the following identity may be used:43 KMEM:IPM = KMEM:AQ/KIPM:AQ (11) Modification of equation 8 to include IPM gave equation 12 KMEM:AQ = (KIPM:AQ)y (12) Substitution of equation 12 into equation 11 gave equation 13 KMEM:IPM = (KIPM:AQ)y/KIPM:AQ (13) The general form of the Roberts-Sloan (RS) equation43 (equation 14) followed from the assumption that solubility ratios could be s ubstituted for partition coefficients and that equation 13 could be substituted into equati on 10 to give (after collecting terms): log JM = x + y log SIPM+ (1-y) log SAQ z MW (14) where x = log (Do/L) and z = It is important to note that all three models predict a negative dependence of flux on the size of the permeant (expressed as either molecular volume MV or molecular weight MW). However, in contrast to KSC (equation 7 where SMEM = SOCT)41 and PG (equation

PAGE 34

21 10 where SMEM = (KOCT:AQ)y(SAQ),42 RS (equation 14 where SMEM = (SIPM)y(SAQ)1-y) indicates that the intercellular matrix of the SC is a biphasic material consisting of aqueous and lipid phasesa description which is consistent with electron micrographs of normal24, 26 and hydrated44 human skin. A fit of the flux, molecular weight, and solubility data from 61 prodrugs ( in vitro mouse) to RS suggested that water solubility was nearly as important as lipid solubility (0.52 SIPM, 0.48 SAQ, r2 = 0.91).45 When a similar analysis was performed on a smaller dataset (n = 10) from the delivery of nonsteroidal antiinflammatory drugs from minera l oil (MO) through human skin in vivo flux was again positively dependent on solubi lities in water (0.28 SAQ) and in a lipid (0.28 SAQ, 0.72 SMO, r2 = 0.93). A recent analysis of a much larger database (n = 103) of in vitro human skin data gave similar values for octanol and water solubilities (0.56 SOCT, 0.44 SAQ, r2 = 0.90).46 Prodrugs By definition, an inactive derivative of an active drug that does not revert to the parent compound in vivo can not be considered a prodr ug, and more importantly, is not therapeutically useful. For example, Billich and coworkers recently reported that certain trimethylammonio-alkyl carbonyl derivatives of cyclosporin A (C sA) exhibited fluxes that were 180-times greater than CsA.47 However, the authors were unable to detect any CsA in the skin and only trace amounts (< 5% total CsA species as CsA) were found in the receptor phases of th e diffusion cells. In this case, since the derivative was inactive,47 the improvement in flux was therapeutically useless except as a demonstration of the potential permeation-enhancing effect of a trimethylammonio-alkyl carbonyl group. Most prodrugs are designed to be enzymati cally labile in order to avoid chemical stability problems that might arise during formulation. On e major benefit of enzymatic

PAGE 35

22 activation is the potentially greater tissu e-specific delivery of the active drug.48 An example for purely enzymatic activation is the conve rsion of minoxidil (6-(1piperidinyl)-2,4-pyrimidinediamine-3-oxide) to minoxidil sulfate following topical application of minoxidil to the scalp (Figure 1-4).49 At least four different sulfotransferase enzymes are believed to be responsible for the bioactivation of minoxidil.49, 50 Although it was originally given oral ly as an antihypertensive agent, it was later found to stimulate hair growth a nd is now used as a treatment for alopecia.50 While the benefits of enzymatic activation are clear, it is important to recognize that enzyme-mediated reactions are subject to inte rspecies and inter-i ndividual variation, whereas chemical activation is largely under the control of the re searchera situation that results in more predictabl e rates of delivery of active drug In the case of minoxidil, there is evidence to suggest that the inefficacy of topical minoxidil in some individuals is due to relatively low sulfotransfe rase activity in those patients.51 The rationale for using prodrugs to ov ercome the skin barrier was briefly mentioned in Section A-4. Although the mo st well-known and prof itable prodrugs have been developed for oral administration,2, 48, 52 many of the same types of prodrugs have been evaluated as topical delivery agents as well.19, 20 A comprehensive review of all the major classes of prodrugs evaluated to date in topical delivery inve stigations is beyond the scope of this thesis. However, the inte rested reader may find such information in several detailed reviews of the subject.19, 20, 38 In this section, only two of the major classes of prodrugs, acyl and soft alkyl, will be discussed.

PAGE 36

23 N N O NH2 H2N N Sulfotransferase PAPSN N OSO3 NH2 H2N N Minoxidil Minoxidil Sulfate Figure 1-4: Bioconversion of Mi noxidil to Minoxidil Sulfate by Scalp Sulfotransferase in the Presence of 3 -Phosphoadenosine-5 -phosphosulfate (PAPS) Acyl Prodrugs The most common type of prodrug found on the market today is one in which a heteroatom on the drug has b een acylated to give the co rresponding ester, carbonate, amide, or carbamate.2 Most of these promoieties cont ain simple aliphatic groups in the acyl chain such as the esters of capto pril ((2S)-1-(3-mercapto-2-methylpropionyl)-Lproline) recently evaluated by Moss et al (Figure 1-5).53 Six esters of captopril were synthesized in which only the le ngth of the alkyl chain was va ried from the methyl to the hexyl ester. As expected, all of th e prodrugs were less soluble in water (SAQ) than captopril (range of SAQ = 0.03-0.58 times the SAQ value for captopril). However, all were much more soluble in octanol (SOCT) than the parent. Although solubilities in octanol (SOCT) were not measured, they may be estimated from the calculated partition coefficients (KOCT:AQ) reported by the authors. By this approach, all of th e ester prodrugs were approximately 4to 89-times more solubl e in octanol. As a result of their higher lipophilicity, five of the deri vatives permeated porcine sk in more effectively than captopril. Within this series of more lipoph ilic homologs, the member that exhibited the greatest increase in flux (40-fo ld) was also the second-most water soluble member of the

PAGE 37

24 series. Thus these results agree with literature precedent19, 20 and the RS model (equation 14),43 and they demonstrate the dependen ce of flux on biphasic solubility. While most acyl-type prodrugs contain simp le aliphatic groups in the acyl chain, there are many reports19, 20 of the benefits of incorporat ing other functional groups into the acyl chain. Milosovich and coworkers54 have shown that in lie u of the aliphatic ester approach that is typically used to deliver steroids,19 introduction of a te rtiary amine into the promoiety can lead to dramatic improve ments in flux. To prove the usefulness of such an approach, the authors reported that a 10% solution of th e hydrochloride salt of testosteronyl-4-dimethylaminobut yrate (TSBH) exhibited a 60-fold greater flux through human skin in vitro than a 10% suspension of testoster one (TS) (Figure 1-5). The free base of TSBH also exhibited a flux that was 35-times greater than TS.54 As noted by Milosovich et al.,54 the relatively high fluxes of the prodrugs are likely the result of increasing aqueous solubility without compro mising lipophilicity. For instance, TSBH is at least 340-times more sol uble in pH 7 phosphate buffer than TS, yet the decrease in partition coefficient (KOCT:AQ) on going from TS (log KOCT:AQ = 3.3) to TSBH (log KOCT:AQ = 2.7) is minimal. Similar results were reported by Wasdo and Sloan45 in a study of alkylcarbonyloxy (AOC) derivatives of acetaminophen (4-hydroxyacetanilde, APAP) (Figure 1-5). In this case, the goal was to im prove the biphasic solubility of the parent by replacing a methylene group in the acyl chain with oxygen to give an ether. Thus, the difference between this and the previous exampl e is the absence of an ionizable group in the acyl chain of the AOC promoiety. The effect of heteroatom substitution on the physicochemical properties of the prodrugs is most apparent in a comparison of 4butyloxycarbonyl-APAP (4-BuOC-APAP) with 4-(2 -methoxyethyloxycarbonyl)-APAP

PAGE 38

25 (4-MOC2-APAP). Although 4-MOC2-APAP wa s 0.74-times less soluble in isopropyl myristate (IPM) than 4-BuOC-APAP, it was 81-times more soluble in water than 4BuOC-APAP and consequently exhibited 8times the flux of 4-BuOC-APAP. Both prodrugs were more soluble in IPM (5to 7-fold) than APAP, but neither was more soluble in water than the parent. Howe ver, since 4-MOC2-APAP exhibited better biphasic solubility than APAP, its flux wa s 1.5-times higher than the flux of APAP. N OH O S H O N O O S H O Captopril Captopril Ethyl Ester O OH O O O N+ H Testosterone (TS)Cl TSBH O NHCOCH3 O O O O NHCOCH3 O O O H NHCOCH3 Acetaminophen (APAP) 4-MOC2-APAP 4-BuOC-APAP Figure 1-5: Struct ures of Acyl Prodrugs for th e Topical Delivery of Captopril Testosterone, and Acetaminophen A variety of mechanisms have been iden tified for the conversion of acyl prodrugs to their respective parent compounds.19, 20 However, simple ali phatic acyl prodrugs are typically hydrolyzed by one of th e mechanisms shown in Figure 1-6.2, 55 Although both reactions are theoretically reversible, the base-catalyzed hydrolysis is usually driven to completion by the formation of the carboxylate anion55 and is shown in Figure 1-6 as an irreversible process.

PAGE 39

26 Acid-Catalyzed Hydrolysis of Esters (AAC2) Base-Catalyzed Hydrolysis of Esters (BAC2) slow DrugX C R O CR O O H DrugX C R O OH H + H2O slow DrugX C R O H OH2 DrugX C R O H + DrugX C R O + DrugX C R O H OH H + DrugXH slow + C R O O H DrugX + CR O O DrugXH +-CR O H OH DrugXH + H + +-HOFigure 1-6: Most Common Mechanisms by which Acyl Prodrugs are Hydrolyzed Chemically Soft Alkyl Prodrugs The term soft alkyl was first given56 to the alkylcarbonyloxymethyl (ACOM) derivative of the amide-type compound show n in Figure 1-7 because it is an ester derivative of the corresponding hydroxymethyl compound which is an al kyl derivative of the parent drug. Whereas the hydroxymethyl prodrug requires chemical activation to give the parent, the corresponding ACOM de rivative generally undergoes a two-step process involving an initial enzymatic (o r chemical) hydrolysis followed by chemical activation to give the parent.19 This is in contrast to the hard alkyl prodrug shown in Figure 1-7 for which bioconversion is restricted to enzymatic oxidation.56 Although soft alkyl derivatives cover a wi de range of promoieties,19 only ACOM and

PAGE 40

27 alkyloxycarbonyloxymethyl (AOCOM) derivatives w ill be considered since they are the focus of this thesis. Much of the work on soft alkyl appr oaches to improve topical delivery19, 20, 38 has focused on polar heterocycles such as the ophylline (Th) and 6-mercaptopurine (6-MP). Three of these examples are shown in Figure 1-8. In their report on the synthesis and in vitro evaluation of a homologous series of 7-ACOM-Th derivatives, Kerr and coworkers57 noted that all of the homologs (R = CH3 to C5H11 and (CH3)3C) were substantially more so luble in IPM (8to 229-times) than Th. However, the maximum flux exhibited by any of the prodrugs was only 2.2-times higher (for R = C3H7) than the flux of Th. Such a modest increase in flux is probably due to the loss of water solubility (SAQ = 0.04 to 0.27-times the SAQ of Th) on going from the pa rent to the prodrug. This situation is much different for th e ACOM prodrugs of 6-MP (R = CH3 to C5H11 and C7H15). The first three members of the 6-AC OM-6-MP series were 2 to 6-times more soluble in water than the parent. As w ith the Th series, all of the 6-ACOM-6-MP prodrugs were much more soluble (50 to 200-time s) in IPM than the parent. In contrast to the Th series, the 6-MP prodrugs permeated the skin much more effectively than 6-MP (53 to 69-fold improvement in flux for the first four members of the series). The relative ineffectiveness of the ACOM approach in the case of Th may be rationalized by considering the fact that Th itself is 41-tim es more soluble in water and 15-times more soluble in IPM than 6-MP. Consequently, Th is much more effective (126-times higher flux) at penetrating the skin than 6-MP. Thes e results demonstrate that it is easier to improve the flux of a poorly soluble compound such as 6-MP with a prodrug approach.

PAGE 41

28 Drug N H O R O O Drug NH2 O ++CH2ON H CH3 Drug O Drug N H O O H Hard Alkylated: Soft Alkylated: Soft Alkylated: Drug N H OH O R'CO2Enzyme-H2O 1. HO 2. H2O H2O HO-HOFigure 1-7: Mechanism of Hydrolysis of Soft Alkyl Prodrugs (Alkylcarbonyloxymethyl and Hydroxymethyl Deriva tives are shown) and Co mparison to Metabolism of Hard Alkyl Derivatives (Gener al Mechanism for an Enzymatic NDemethylation Reaction is given as an Example) In spite of their proven eff ectiveness in oral drug delivery,2 AOCOM prodrugs have received little attention in topical delivery. In fact, the 7-AOCOM derivative of Th shown in Figure 1-8 appears19, 20 to be the only example of the use AOCOM prodrugs to improve percutaneous absorption.58 However, the authors of the study for which it was synthesized were more interested in the hydr olytically more labile 7-ACOM-Th prodrugs and chose not to evaluate this part icular derivative in diffusion cells.58 The example of bacampicillin, an orally administered prodrug of ampicillin, has been included in Figure 1-8 as a reminder of the potential usef ulness of the AOCOM promoiety. In a comparative study of the pharmacokinetics of orally administered pivampicillin (an ACOM prodrug of ampicillin), bacampicillin and ampicillin, bacampicillin exhibited the highest rate of absorption and shortest abso rption lag time. Both prodrugs were equally effective at improving the oral bioavailability of ampicillin.59

PAGE 42

29 N N N H N O O N N N N O O O R O Theophylline (Th) 7-ACOM-Th N H N N H N S N N N H N S O R O 6-Mercaptopurine (6-MP) 6-ACOM-6-MP N N N N O O O OR O 7-AOCOM-Th O O O O O S N O N H O NH2 BacampicillinO H O S N O N H O NH2 Ampicillin Figure 1-8: Examples of Al kylcarbonyloxymethyl (ACOM) and Alkyloxycarbonyloxymethyl (AOCOM) Prodrugs Conclusions Although oral drug delivery will likely remain the method of choice for drug administration, it is not a suitable route fo r many different medications due to the substantial biochemical barrier presented by the GI tract and liver. One of the main advantages of transdermal delivery is the avoidance of first-pass metabolism that stems from the relatively low enzymatic activity of the skin compared to the liver. As illustrated in the case of transdermal versus or al estrogen, topical delivery is often a safer alternative to the oral route. In addition, topical delivery provides a means for treating local conditions without expos ing the systemic circulation to high levels of the therapeutic agent. In contrast to the GI tract and liver, the skin functions mainly as a physical barrier to drug absorption with the outermost layer, the stratum corneum, providing most of the

PAGE 43

30 resistance to permeation. Electron micrographs of the stratum corneum have shown that intercellular matrix through which a permean t must pass is compos ed of alternating layers of polar and nonpolar ma terial. Such evidence supports in vivo and in vitro skin penetration experiments in which flux th rough skin was positively dependent on the aqueous as well as lipid solubility. These qualitative observations were subsequently used to develop a mathematical model (i.e. the Roberts-Sloan model, RS) for accurately predicting flux through skin based on the solu bility properties and molecular weight of the permeant. Among the many methods used to overcome the skin barrier, a prodrug/formulation approach is one of the most attractive as it would likely increase permeation while minimizing side effects. Two of the most su ccessful promoieties used in topical delivery are the acyl and soft alkyl-type. Of these two types, the acyl promoiety is the most common perhaps by virtue of its relatively faci le synthesis and genera lly low toxicity of its hydrolysis byproducts. Though they are not as common, soft alkyl prodrugs have a long history of improving oral bioavailability as well as topical deli very. AOCOM derivatives are a sub-type of soft alkyl prodrugs that are underre presented in topical delivery and should be further investigated using skin permeation experiments. Regardless of the promoiety, flux was shown to depend directly on the lipid and aqueous solubilities of the prodrug.

PAGE 44

31 CHAPTER 2 SPECIFIC OBJECTIVES First Objective The first objective of the present investigation was to synthesize a homologous series of alkylcarbonyloxymethyl (ACO M) and alkyloxycarbonyloxymethyl (AOCOM) derivatives of a model phenol. There are currently no examples of the topical delivery of ACOM and AOCOM derivatives of phenols. Th is is in spite of their well-documented effectiveness at improving the oral bioavailability2 of phosphates and carboxylic acids, and the topical delivery19 of amides, imides, thioamide, a nd carboxylic acids. Most of the previous work on the topical delivery of phe nols via a prodrug approach has focused on the corresponding acyl derivatives.19, 45, 60-65 One of the most studied classes of drug in that respect is the narcotic analgesics (see Fi gure 2-1 for examples). Narcotic analgesics are usually given intravenously, sublinguall y, or intramuscularly in order to avoid extensive first-pass metabolism on oral admini stration, but the parenter al routes are also associated with high peak plasma levels and require frequent dosing. In addition to its avoidance of first-pass metabolism, transder mal administration is typically associated with constant rates of delivery into the systemic circulation and has a relatively high degree of patient compliance.5 Thus, topical delivery is an attractive alternative to the current methods by which these compounds are administered. Most reports on the use of ester (alkylcarbonyl AC)60, 62-64 and carbonate (alkyloxycarbonyl AOC)45, 61, 65 prodrugs to increase the percutaneous absorption of narcotic analgesics indicate that the improve ment in flux is only modest (2-7 fold).

PAGE 45

32 However, Sung et al.63 found that the decanoate ester of nalbuphine was 40-times more permeable than the parent when delivered from pH 4 buffer, and Drustrup et al.60 found that the 3-hexanoate ester of morphine wa s approximate 3500-times more permeable than morphine when delivered from IPM. Alt hough it is impossible to know whether ACOM and AOCOM prodrugs of phenols will work better than the corresponding acyl derivatives,19 there does not appear to be great differences in permeation enhancement when an acyl promoiety is used in place of an ACOM in the same parent drug (compare 1-AC66 to 1-ACOM-5-fluorouracil67 and 3-AC68 to 3-ACOM-5-fluorouracil69). On the other hand, since the carbonyl moiety of the prodrug is separated from the parent compound by a methylene spacer, the physicochemical properties of soft alkyl derivatives are governed less by the parent drug and more by the promoiety. The result is that soft alkyl prodrugs such as ACOM and AOCOM are more easily customized to meet the particular objectives (drug solubilit y, stability, etc.) of the investigator.38 One example where a soft alkyl prodrug may be more effective than the corresponding AC derivative is -tocopherol (Vitamin E). Vitamin E is one of several key compounds responsible for maintaining an effective barrier against free-radical damage in cellular membranes.70 In fact, it is the primary antioxidant for membranes and lipids. Since the body does not synthesize vita min E, it must be taken in through diet or given as a supplement. However, there is currently no efficient way to administer supplemental Vitamin E.70 Oral administration of V itamin E suffers from slow absorption rates71 and generally provides inferior photoprotection compared to topically applied Vitamin E.71, 72 Intravenous formulations of Vitamin E have also been administered, but in some cases,73 life-threatening side effects have ensued. Part of the

PAGE 46

33 difficulty in delivering Vitamin E is that it is practically insoluble in water74 and readily oxidizes in air. The problem of instability has traditionally been solved by converting Vitamin E to its acetate or succinate esters. However, this appro ach introduces a new problem: the acetate and succinate esters do not readily revert to the active compound in vivo.75 A similar problem has been addressed before in the case of -lactam antibiotics.76 Alkyl derivatives of the carboxylic acid group of these drugs exhibi t poor bioavailability in vivo, but often see dramatic improvemen ts in prodrug-to-drug conversion when a ACOM or AOCOM approach is used.2 In the case of Vitamin E, nucleophilic attack at the carbonyl carbon is limited due to the fla nking methyl groups on the aromatic ring. One potential solution to this problem is to move the site of hydrolysis away from the sterically hindered chromanol head of Vitamin E by way of a soft alkyl (ACOM or AOCOM) derivative. Before applying the soft alkyl approach to the narcotic analgesics and Vitamin E, it seemed prudent to first validate the strate gy using a simple phenol. Acetaminophen (4hydroxyacetanilide, APAP) was selected as a model because its ACOM and AOCOM derivatives were expected to be solids (APAP mp = 167-170) and hence more easily characterized. Since a series of AOC derivati ves of APAP had been previously evaluated in diffusion cell experiments,45 it would also be possible to compare the effects of using an acyl versus a soft alkyl promoiety. Second Objective The second objective of this project wa s to determine whether the ACOM and AOCOM prodrugs could improve the topical de livery of APAP. Hairless mouse skin in vitro was selected as a model for human skin due to its relatively low cost and in order to

PAGE 47

34 be consistent with all previous work by our lab. Mouse skin also has the advantage of exhibiting less variation than human skin. O O OH N O H Naltrexone (NTX) O N O H MeO C CH3 O H C(CH3)3 O O H OH N O H Nalbuphine (NA) Buprenorphine O H N CH3 O H O H Morphine (MOR) OH N CH3 O Ketobemidone O H O H H alpha-Tocopherol (Vitamin E) Figure 2-1: Phenol-Containing Therapeutic Agents that may benefit from Topical Delivery via Alkylcarbonyl oxymethyl (ACOM) or Alkyloxycarbonyloxymethyl (AOCOM) Derivatization Third Objective The third objective of the present investigation was to improve the accuracy of the Roberts-Sloan equation (RS)43 for predicting flux through hairless mouse skin. At present, the database (n = 61) upon which the RS equation is based is heavily dependant on data from heterocyclic compounds: 59% 5-fluorouracil rela ted entries, 18% 6mercaptopurine related entries, and 10% The ophilline related entries in the database. Only 8 of the 61 entries (13%) are of a phe nolic compound (i.e. APAP). An earlier study found that in general, the e rror in predicting flux using RS was greater for a phenolic

PAGE 48

35 prodrug (4-AOC-APAP) than for a heterocyclic prodrug.45 In order to extend the applicability of RS to a wider range of drugs, the structural di versity of the database must be expanded. Incorporation of the ACOM and AOCOM prodrugs into the database would likely result in a more robust RS model.

PAGE 49

36 CHAPTER 3 ALKYLCARBONYLOXYMETHYL PRODRUGS OF ACETAMINOPHEN (APAP) Synthesis of Alkylcarbony loxymethyl (ACOM) Iodides A key feature of the RobertsSloan database (Chapter 1)20 is that it is almost entirely comprised of homologous series. Such homogeneity was intentional as it is easier to determine the impact of physicoche mical properties on flux when structural differences are minimal. In keeping w ith that theme, synthetic routes to 3 that allowed R to be simple aliphatic groups was desired. Currently, there are three reported methods for synthesizing such alkylating agents. In two of these procedures, ACOM chloride 4 functions as the intermediate from whic h the corresponding iodide is subsequently generated via a Finkelstein-type halide excha nge. Chloromethyl chlorosulfate has proven to be a useful reagent for obtaining ACOM chloride from carboxylic acids under phasetransfer conditions.77, 78 However, since this method fa ils for carboxylic acids with fewer than 6 carbon atoms,77 it was not suitable for th e present study. Compound 4 may also be generated via the condensation of acid chlo rides with aldehydes in the presence of a Lewis acid.79, 80 However, this route to ACOM iodide frequently provides low yields of the desired compound.81 A different approach was take n by Fleischmann and coworkers: they synthesized pivaloyloxyethyl iodide directly from acetaldehyde and pivaloyl chloride in the presence of NaI.82 In an effort to extend the applicability of this reaction to 3 where R = H, it was found that trioxane 1a reacts with acid chlorides in the presence of NaI to give predominately compounds 3a-f in one step (Figure 3-1). Paraldehyde 1b exhibited a similar reactivity with acid chlorides under the same conditions to give 3

PAGE 50

37 where R = CH3. The structure of 3 was arrived at by comparison of its 1H NMR spectra with 1H NMR spectra reported for 3 in the literature.81 O O O R R R +R'COCl NaIR'CO2CH(R)IR'CO2CH(R)Cl(R'CO2)2CH(R)+ +a: R' = b: R' = c: R' = d: R' = e: R' = f: R' = CH3C2H5C3H7C5H11C7H15a: R' = R = H b: R' = R = H c: R' = R = H d: R' = R = H e: R' = R = H f : R' = R = g: R' = R = H CH3C2H5C3H7C5H11C7H15CH3CH31 a: R = H b: R = CH32 3 4 5 a: R' = R = H b: R' = R = H c: R' = R = H d: R' = R = H e: R' = R = H g: R' = R = H CH3C2H5C3H7C5H11C7H15a: R' = R = H b: R' = R = H c: R' = R = H d: R' = R = H e: R' = R = H g: R' = R = H CH3C2H5C3H7C5H11C7H15(CH3)3C (CH3)3C (CH3)3C(CH3)3C Figure 3-1: Reaction of Trioxane 1a and Paraldehyde 1b with Acid Chlorides in the Presence of NaI Unfortunately, various amounts of 4 and 5 formed along with 3 as well. Byproduct 4 was identified as the chloride analogue of 3 based on the 1H NMR spectra reported for this compound in the literature,81 and by comparison to an authentic sample of 4 prepared via a previously reported method.83 Compound 5 was assigned the structure shown in Figure 3-1 by comparison of its 1H NMR spectra with the pr oduct of the reaction of acetic acid with 3a by a modification of the method of Folkmann and Lund.84 1H NMR analysis of the product mixture revealed an upfield shift in the diagnostic methylene singlet from 5.99 ppm in RCO2CH2I to 5.73 ppm in the product. Furthermore, the product gave a spectrum that was consistent with bis(acetyloxy)methane. It should be noted that others have observed the formati on of bis(acetyloxy)meth ane in the reaction of trioxane with acetyl mesylate.85 Thus, it is not surprising that 5 is also formed in the present case. However, in reac tions involving paraldehyde (R = CH3), 4 and 5 could not be detected in the 1H NMR spectrum of the reaction mixture. In an effort to optimize the reaction, va rious reaction conditions were employed; the results from some of these experiments are listed in Table 3-1. Given that ACOM

PAGE 51

38 iodides are relatively unstab le above room temperature,81 25 oC was set as the upper temperature limit for all ACOM iodide synthese s. No reaction occurs in the absence of NaI, and a slight excess of NaI is necessa ry to achieve good conversion of the starting materials, regardless of solvent. Similarl y, pyridine was unable to catalyze the reaction (entry 5) and only carboxylic acid, acid anhydr ide, and starting materi al was detected in the product mixture. This result is interesting since French and Adams86 had previously found that mixtures of pyridine and aromatic acid halides react with aromatic aldehydes to yield the corresponding ACOM halides. T hus, in the present case depolymerization of 1 may be rate-determining. Yields of the desired compound 3 appear to be unaffected by variations in temperature below 25 oC. For example, the yield of 3 does not change substantially if the re actants are allowed to stir for 1 hour at 0 oC after initial mixing, versus allowing the mixture to stir at room temperature immediatel y after all reactants have been added (data not show n). Likewise, the yield of 3 is substantially unaffected by the length of time over which 2 is added (entry 7 versus en try 9) and by the degree to which 2 is converted to the acyl iodide before 1 is added (entry 8). On the other hand, the formation of 3 appears to be more sensitive to the form of the aldehyde undergoing conversion. This relationship is most appa rent in entries 6 and 10. As shown in the Table 3-1, trioxane reacts with octanoyl chlo ride to give octanoyloxymethyl iodide in 86 % yield (entry 6). In contrast, paraformalde hyde reacts under the same conditions to give only 45% yield of the desired ACOM iodide (entry 10). Though the reaction was run only once, byproduct 4 seems to be more favored when paraformaldehyde is used instead of trioxane (entry 10 versus entry 6). As shown in entries 3 and 11, the reaction is also

PAGE 52

39 able to accommodate a certain am ount of steric hindrance in 1 and 2 ; however, the reaction of 1b with 2f was not attempted. Table 3-1: Variation in Reaction Conditions, Crude Yielda of 3 4 and 5 and Percentage of 1 Remaining at the End of the End of the Experimentb % Yield Entry R R Molecular Ratioc 1 : 2 : NaI Solvent 3 4 5 % of 1 Remaining 1d, e H C2H5 1 : 1 : 1 CD3CN 46 11 11 18 2d, e H C2H5 1 : 1 : 1.2 CD3CN 54 7 11 7 3d, e CH3 CH3 1 : 1 : 1.2 CDCl3 83 f f 17g 4d, e H C2H5 1 : 1 : 0 CDCl3 -0 0 100 5h H C7H15 1 : 1 : 0 CH2Cl2 -0 0 100 6e H C7H15 1 : 1 : 1.2 CH2Cl2 86 10 4 0 7e H C7H15 1 : 1 : 1 CH2Cl2 74 11 6 6 8d, i H C3H7 1 : 1 : 1 CH3CN 33 f f f 9j H C7H15 1 : 1 : 1 CH2Cl2 70 16 8 8 10e, k -C7H15 1 : 1 : 1.2 CH2Cl2 45 40 15 0 11e H (CH3)3C 1 : 1 : 1.2 CH2Cl2 70 24 6 0 12e, l H C7H15 1 : 1 : 1.2 CH2Cl2 87 2 6 1 3 0.6 6 3 13e H C5H11 1 : 1 : 1.2 CH2Cl2 89 7 4 2 14e, l H C3H7 1 : 1 : 1.2 CH2Cl2 82 4 14 44 1 0 15e, l H C2H5 1 : 1 : 1.2 CH2Cl2 80 6 16 45 2 0 16e, l H CH3 1 : 1 : 1.2 CH2Cl2 72 2 19 311 2 0 a Unless otherwise noted, entries represent a single experiment (n = 1). b Reaction time was usually 20-24 hours. c Molecular ratio shown is based on equi valents of formaldehyde or in the case of paraldehyde, equivalents of acetaldehyde. d Crude yield determined using benzene as an internal standard. e 2 is added to a mixture of 1 and NaI within 1-20 min. f Could not determine from 1H NMR spectrum. g Present as the monomer, acetaldehyde. h 0.3 mol % pyridine added as a catalyst. i 2 is allowed to react with NaI for 1 h at 25 oC. After 1 h, a solution of 1 is added over 40-60 min at 0 oC. j 2 is added over 2 h to a mixture of 1 and NaI. k Paraformaldehyde used instead of trioxane. l Average SD, n = 3. As this study progressed, it became a pparent that the product distribution was dependent on the type of NaI being used. Pr actically all of the AC OM iodides used in this study (including those represented in Tabl e 3-1) were prepared using NaI from three different lots and purity grades purchas ed from Aldrich during the 1980s (see Experimental). These particular batches of NaI were eventually consumed and additional NaI of the same purity and catalog number was ordered from Aldrich. However, when this new (purchased 2005) NaI was used as shown in Figure3-1, the reaction failed to

PAGE 53

40 reach completion even after 48 hours. Moreover, the mixtures resulting from such reactions were always contam inated with a large amount of unwanted byproducts. In other experiments, this new NaI was sti ll able to convert al kyl chlorides to the corresponding iodides as expected for a Finke lstien reaction. These divergent results were rationalized by assuming that the older batches of NaI were contaminated with traces of an unidentified catal yst. Subsequent experiments in which various transition metals and Lewis acids were added to the react ion mixture indicated that this was indeed the case. For example, zinc dust87 (23 mol %) was found to cat alyze the reaction by fully converting 1 to products, but unfortunately compound 5 was the major product. Other transition metal catalysts such as iron also failed to improve the yield of 3 Aluminum metal, as well as AlCl3 ( 23 mol %), suppressed the formation of 4 and 5 but failed to fully convert 1 to products. However, if a combination of AlCl3 ( 10 mol %) and I2 ( 5 mol %) was used, total 4 and 5 were minimized (< 14% and < 15% of product mixture, respectively) and 1 was completely consumed. It was further noted that aluminum metal is completely consumed during the reaction and that AlCl3 gives the same results as aluminum metal under identic al reaction conditions. Thes e results suggest a reaction mechanism that involves Lewis acid (formed by traces of HCl in the acid chloride reacting with traces of metal in the NaI) cata lysis. Interestingly, in the one case where AlI3 and AlCl3 were allowed to react separately under identical conditions, the resulting product mixtures differed considerably. Conve rsion rates in those experiments differed by 50% (though in neither case was 1 completely consumed), and in the AlI3 reaction, several unidentifiable byproducts were formed as well. These results suggest that AlCl3, and not AlI3 is the principle catalyst in this reaction.

PAGE 54

41 Though a Lewis acid such as AlCl3 is (apparently) important for successful formation of 3 other experiments indicate that iodide ion and I2 are needed for the depolymerization of 1 If AlCl3 is replaced with an equivalent amount of I2, 1 is completely converted to products: molar ratio 3 : 4 : 5 = 1:1:1.5 plus an unidentifiable byproduct. Also, in the absence of NaI, 1 reacts slowly with 2 and AlCl3 to give a mixture of unidentifiable bypr oducts and only minor amounts of 4 (approximately 50% conversion of 1 to these products after 24 hours). Thus iodide ion likely aids in opening compound 1 perhaps through an SN2 process similar to that proposed by Balme and Gore88 for the cleavage of acetals by TiCl4/LiI. Since I2 also increases the rate at which compound 1 is converted to products, it ma y facilitate the cleavage of 1 by coordinating with the oxygen atoms in the ring thereby polarizing the CH2O--CH2 bond in the formal moiety. Unfortunately, a catalyst system that consistently matched the reactivity of the older batches of NaI could not be identifie d. However, it should be noted that when crude reaction mixtures of 3 generated via the modified procedure (5 mol % AlCl3 and 2 mol % I2 included in reaction mixture) were a llowed to react with 4-hydroxyacetanilide, the product mixtures were no diffe rent than those obtained using 3 generated from the older batches of NaI. Coupling Reaction of ACOM Iodides with 4-Hydroxyacetanilide It has long been known that ACOM halides 3 display ambident reactivity sometimes nucleophiles react at the carbonyl to give acylated produc ts while at other times the alkyl halide carbon is attacked to give alkylated products (Figure 3-2). Such reactivity has been observed in reactions of ACOM halides with a variety of nucleophiles including amines,89 phenols,90 and alcohols.90 In the initial repor t on the reactions of ACOM halides with phenols,90 it was noted that th e nucleophilicity of the phenol and the

PAGE 55

42 nucleofugicity of the halide are key determ inants of the product distribution. More nucleophilic phenols tend to give acylated pr oducts, while better le aving groups and less nucleophilic phenols shift the product distribution in favor of the alkylated phenol. It was also suggested90 that 7 is favored by functional groups at the methylene spacer that are capable of stabilizing a positive charge. R' O X O R + O H O O R' O R O R' O 3 6 7 8+X = Cl, Br, I R'' = alkyl, aromatic, etc. R'' R'' R'' Figure 3-2: General Reac tion of Alkylcarbonyloxymethyl (ACOM) Halide 3 with Phenol 6 to Give Aryl Acylal 7 and Aryl Ester 8 Recently, Ouyang and coworkers have s uggested that the percentage of 7 in the product mixture is also directly proportiona l to the degree of steric hindrance in 3 and in 6 .91 According to Ouyang, compound 8 is the major product if 3 and 6 are relatively free from steric hindrance, but as th e degree of steric hindrance in 3 and 6 increase so does the percentage of 7 Ouyangs conclusions were base d on reactions between various phenols and compounds 9 in which the size of the amino-protecting group was varied from the relatively small allyloxycarbonyl to the bulky 9-fluorenylmethoxy carbo nyl (Figure 3-3). As shown in Table 3-2, the product distribut ion was shifted almost entirely toward acylated phenol 8 when the protecting group was small (e ntry 13). As the steric bulk of the protecting group increased, so did th e percentage of alkylated product 7 reaching as high as 15% of the product mixture (ent ry 15). Although higher yields of 7 were realized if both 3 and 6 were sterically hindered (entry 16), 8 remained the major product in all cases. Compound 7 became the major product (58%) onl y when the base was changed from K2CO3 to Cs2CO3, and both 3 and 6 were sterically hindered (R = Boc-D-Leu, R =

PAGE 56

43 CO2CH3 H, Y = H, Z = CCCO2Pac). Based on these results, it was concluded91 that both 3 and 6 must be sterically hindered in order to shift the product distribution in favor of 7 N H O I O R'''O O 9 NHCOCH3 O O O R' O R R 11R'''CH2 CH2 O I O 10 Figure 3-3: Structures of ACOM Derivative of a Protected Amino Acid 9 (R = Protecting Group) and its Corre sponding Aliphatic Derivative 10 and Structure of Byproduct 11 Table 3-2: Product Distribution of the Reactiona of ACOM Halides 3 with Phenols 6 : Data Taken from the Literature Distribution (%)b Entry R R X Y Z 7 8 c Ref 1 (CH3)3C H Cl H H 0 100 1.24d 90 2 (CH3)3C H Cl OCH3 H 0 100 90 3 (CH3)3C H Cl NO2 H 50 50 90 4 (CH3)3C H I H H 100 0 90,91e 5 (CH3)3C H I OCH3 H 100 0 90 6 (CH3)3C H I NO2 H 100 0 90 7 CH3 H I H CONH2 (25)f g 0.52d 93 8 C3H7 H I H CONH2 (47)f g 0.68d 93 9 (CH3)3C H I H CONH2 (29)f g 93 10 CH3 H Br (25)f g 92 11 C2H5 H Br (24)f g 92 12 (CH3)2CH H I (30)f g 0.76d 92 13 Alloc-DLeu H I H H 5 95 1.75h (0.69)i 91 14 F-moc-DLeu H I H H 10 90 1.75 (1.41)j 91 15 Boc-D-Leu H I H H 15 85 1.75 (1.24)k 91 16 Boc-D-Leu H I H 38 62 91 a For entries 1-9 and 13-16, base = K2CO3, solvent = acetone or acetonitrile. For entries 10-12, base = NaH, solvent = THF. b Determined from 1H NMR spectrum of the cr ude reaction mixture. c Chartons steric parameter for R d Reference 94. e In this case, Cs2CO3 was used as a base in lieu of K2CO3. f Isolated yield. g Reference makes no mention of any products other than 7 h Calculated as described in the text. i Steric parameter of the allyl group (reference 95). j Steric parameter of the 9-Methyl-9-fluorenyl group (reference 95). k Steric parameter of the t -butyl group (reference 94). O O O H O O O H O O O H

PAGE 57

44 On the other hand, Bensel and coworkers92 have demonstrated that good yields of 7 may be obtained when neither 3 nor 6 is sterically hindered (e ntries 10-12). Bundgaard and coworkers93 have also shown that if 6 is sterically hindered but 3 is not, 7 may still be obtained in good yield (entries 7-9). Yet, it was unclear whether sterically unhindered ACOM derivatives of 4-hydroxyacetanilide (APA P) could be synthesized given the prior assertions of Ouyang on the im portance of steric hindrance.91 The results from the reactions of 3a 3f with APAP 6a phenol 6b and 2,2,5,7,8-pentamethyl-chroman-6-ol 6c (Figure 3-4) are show n in Table 3-3. R'CO2CH(R)IK2CO3, CH3CN 60-90% R'CO2CH(R)O Y z R'COO Y z+a: R' = Y = Z = H b: R' = Y = Z = H c: R' = Y = Z = H d: R' = Y = Z = H e: R' = Y = Z = H g: R' = Y = H, Z = H h: R' = (phenol = 2c) CH3C2H5C3H7C5H11C7H15NHCOCH3NHCOCH3NHCOCH3NHCOCH3NHCOCH3CH3CH3 O H Y za: R' = R = H b: R' = R = H c: R' = R = H d: R' = R = H e: R' = R = H f : R' = R = CH3C2H5C3H7C5H11C7H15CH3CH3a: Y = Z = H b: Y = H Z = H c: 2,2,5,7,8-pentamethylchroman-6-ol NHCOCH33 6 7 8 a: R' = R = H, Y = Z = H b: R' = R = H, Y = Z = H c: R' = R = H, Y = Z = H d: R' = R = H, Y = Z = H e: R' = R = H, Y = Z = H f : R' = R = Y = Z = H g: R' = R = H, Y = H, Z = H h: R' = R = H, (phenol = 2c) CH3C2H5C3H7C5H11C7H15CH3CH3NHCOCH3NHCOCH3NHCOCH3NHCOCH3NHCOCH3NHCOCH3CH3CH3+ Figure 3-4: Reaction of ACOM Iodides 3a-f with Phenols 6a-c As shown in Table 3-3, 7 was the major product in every case regardless of the steric hindrance pr esented by the phenol 6 or the ACOM iodide 3 (entries 1-6, 8, 9) despite the predictions of Ouyang.91 There did seem to be a vague relationship between

PAGE 58

45 product distribution and alkyl ch ain length, however. For alkyl chain lengths longer than propyl, the percentage of 7 remained close to 70% (entri es 1-3). As the alkyl chain length decreased from propyl to methyl, there wa s an incremental decrease in the ratio of 7 / 8 (entries 3-5). The only instance where 8 formed in preference to 7 was when chloride was used as the leaving group X (entry 7). In addition, the reacti on of the sterically hindered phenol 6c with the relatively sterically unhindered 3a gives credence to the idea91 that sterically hindered phe nols give higher ratios of 7 / 8 than sterically unhindered phenols (entry 9 versus entries 5 and 8). Further increases in the percentage of 7 were realized by introducing a methyl group in pl ace of hydrogen in the methylene linker R of 3 (entry 5 versus entry 6). Table 3-3: Product Distribution of the Reactiona of ACOM Halides 3 with Phenols 6 : Data from the Present Work Distribution (%)b Entry R R X Y Z 7 8 11 c 1d C7H15 H I NHCOCH3 H 71 (1.7) 27 (1.7) 2 (1.2) 0.73e 2f C5H11 H I NHCOCH3 H 66 27 7 0.68e 3f C3H7 H I NHCOCH3 H 73 24 3 0.68e 4g C2H5 H I NHCOCH3 H 59 (7) 31 (3) 11 (9) 0.56e 5d CH3 H I NHCOCH3 H 49 (2.9) 37 (4) 15 (7.5) 0.52e 6f CH3 CH3 I NHCOCH3 H 60 40 0 7f CH3 H Cl NHCOCH3 H 0 100 0 8f CH3 H I H H 63 37 h 9f CH3 H I 68 32 h a Base = K2CO3, solvent = acetone or acetonitrile. b Determined from 1H NMR spectrum of the crude reaction mixture. c Chartons steric parameter for R d Average (SEM, 3 experiments). e Reference 94. f n =1. g Average, 2 experiments; value in parenthesis is the range. h Could not determine by 1H NMR. It is important to recognize that compounds 3a 3f were not purified before they were used in the c oupling reactions with 6a c As such, they (with the exception of 3f ) contained various amounts of 4 (Table 3-1) which may or may not have influenced the O O H

PAGE 59

46 product distributions. In their initia l report on the coupling reactions of 3 with 6 Sloan and Koch observed that acylated products readily formed when X = Cl even though 4 was sterically hindered (e ntries 1-3, Table 3-2).90 Although 3 is contaminated with 4 in the present study, 3 is much more reactive, and is in excess of 4 by at least 3-4 fold. Thus 6 is more likely to react with 3 than 4 If the reaction of byproduct 4 with 6 is significant under the present conditions, then there shoul d be a correlation be tween the ratio of 8 / 7 and the percentage of 4 in the crude product 3 (Table 3-1). Using en try 4 (Table 3-2) as a reference point for when 3 is pure, a plot of the ratio of 8 / 7 versus the percentage of 4 in crude 3 is shown in Figure 3-5. As shown in Figure 3-5, there does not appear to be a strong relationship between the purity of 3 and the ratio of 8/7 It is therefore reasonable to assume that the product distributions observe d in the present investigation result solely from the reaction of 3 with 6 y = 0.0243x + 0.1204 R2 = 0.7207 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 05101520 % RCO2CH2ClAcy/Alk Figure 3-5: Plot of the Percentage of 4 (RCO2CH2Cl) in Crude 3 Versus the Ratio of 8 / 7 (Acylated/Alkylated phenol) Resulting from the Reactions of 3a-3e with 6a and 6b (Taken from Entry 4, Table 3-2 and Entries 1-4, and 8, Table 3-3 ; and Entry 5 ; Note: Entry 5 not Included in Linear Regression Analysis as it appears to be an Outlier)

PAGE 60

47 When ascertaining the affect of steric hindrance on a given reaction, it is often helpful to use a quantitative measure of steric hindrance. In the present work, this was done by relating Chartons steric parameter 94 to the ratio of alkylat ed / acylated phenol (Tables 3-2 and 3-3). Since most of the derivatives of 3 shown in Tables 3-2 and 3-3 contain simple aliphatic groups in the acyl portion R, the values could be taken directly from the literature.94-96 To our knowledge, values for the R groups in entries 13-16 (Table 3-2) have not been reported. Since it was desirable to make all comparisons of steric effects using the same scale, the steric parameter for these groups were estimated by assuming that the van der W aals radius of the carbamate moiety in 9 is approximately equal to the corresponding arrangement of methylene groups in 10 .97 Using 10 as a surrogate for 9 values were then calculated96 from = 0.497n + 0.409n + 0.0608n 0.309, where n, n, and n, are the number of carbon atoms attached to the alpha, beta, and gamma carbon atoms, respectively, in 10 Alternatively, the steric effect of R in entries 13-16 (Table 3-2) may be ev aluated by assuming that for this series, the ratio 7 / 8 is determined primarily by the steric bu lk of the amino protecting group R. In this case, values may be taken from the literature since the steric para meters of R are known (values in parentheses, entries 13-16, Table 3-2).94, 95 Since neither Bundgaard93 nor Bensel92 mentioned product distributions in there reports, entries 7-12 (Table 3-2) offer only i ndirect evidence of the effect of steric hindrance on the formation of 7 and 8 What is clear from their findings is that good yields of 7 may be obtained under essentially th e same conditions used by Ouyang91 but from a sterically unhindered ACOM halide (X = Br or I). For entry 4 (Table 3-2) and entries 1-5, and 8 (Table 3-3), the variation in 7 / 8 appears to be directly related to the

PAGE 61

48 variation in A plot of versus the ratio of 8/7 for these entries (Figure 3-6) gave a good correlation (r2 = 0.95). If these results are repres entative of all reactions of acyclic 3 (where R is aliphatic) with 6 then the effect of R on the product distribu tion is related to its ability to discourage nucleophilic attack at the carbonyl. Such a finding should not be surprising since nucleophilic substitution at a carbonyl carbon is known to be sensitive to steric hindrance.55 y = -0.7693x + 0.9358 R2 = 0.951 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.400.600.801.001.201.40steric parameterAcy/Alk Figure 3-6: Plot of Char tons steric parameter for R Versus the Ratio of 8/7 (Acylated/Alkylated Product) Re sulting from the Reactions of 3a-3e with 6a and 6b (Taken from Table 3-2: Entry 4, Table 3-3: Entries 1-4, and 8 and Entry 5 Note: entry 5 not included in linear regression analysis as it appears to be an outlier) On the other hand, analysis of the steric effect in entries 13-16 (Ouyangs data, Table 3-2)) is more complicated. If one assumes that 9 (R = protected amino acid) and 3 (R = simple aliphatic chain) react with 6 by the same mechanism, and that the acyl group of 10 can approximate the steric effect of the acyl group in 9 then variations in the amino

PAGE 62

49 protecting group should have no effect on product distribution, contrary to the conclusions of Ouyang.91 This follows from the work of Charton96 that showed that for aliphatic acyl groups, substitution at the delta carbon contributes nothing to the effective van der Waals radius of the acyl group. Indeed, the fact that the ratio of 7 / 8 increases on going to bigger protecting groups (see values in parentheses, Table 3-2) implies that 9 reacts with 6 by a different mechanism than that prescribed90 for simple derivatives of 1 (where R is aliphatic). On e potential mechanism for ratio nalizing the results of Ouyang is shown in Figure 3-7. It may be possible for compounds such as 9 to cyclize to give 5oxazolidinone 12 5-Oxazolidinones are known to unde rgo nucleophilic addition at the carbonyl carbon to give 13 followed by loss of formaldehyde to give 8 .98 In this scenario, bulky protecting groups lik ely retard the conversion of 9 to 12 and thus permit 9 to exhibit a reactivity with phenols similar to that displayed by more conventional derivatives of 3 (i.e. where R = aliphatic). 12 9R'''O N O H O O I .. R'''O N O O O HI R'''O N O O O O H z R'''O N H O O O z CH2O138 6 Figure 3-7: Speculative Mechanism for Reactio ns of Protected Amino Acid Derivatives 9 with Phenols 6 In addition to the expected products 7 and 8 there was also the unanticipated formation of byproduct 11 (Figure 3-3) in react ions involving APAP 6a (entries 1-5 Table 3-2). Compound 11 was assigned the structure shown by comparison of its 1H NMR to the corresponding derivatives 7 (compound 11 was also analyzed by IR, but no useful structural information could be gleaned from the spectrum). At present, it is not

PAGE 63

50 clear why 11 is generated in reactions involving APAP, but fails to form when other phenols react with 3 (entries 1-16 Table 3-2 and entries 8 and 9 Table 3-3). An analysis (1H NMR) of the crude reaction mixture resulting from the synthesis of 3 showed no evidence of alkylating agents such as RCO2CH2OCH2I or bis(acetoxymethyl) ether99 which might react with 6 to give 11 Presumably, the formaldehyde generated during the acylation of 6 by 3 goes on to react with another molecule of 3 to form RCO2CH2OCH2I. Several reaction conditions were employed in an effort to maximize the yield of 7 Methods such as solid-liquid phase-transfe r catalysis or the use of a non-nucleophilic organic base failed to improve the yield of 3 Interestingly, the use of 1,8diazabicyclo[5.4.0]undec-7-ene as a base resulted in an increase in the percentage of 7 by approximately 20% for the least sterica lly hindered member of the series ( 3a ). However, since the conversion of 6 to 7 was lower is this case, this technique was not synthetically useful. Replacing K2CO3 with Cs2CO3 as recommended by Ouyang91 resulted in an increase in the conversion of 6 to 7 (50% versus 40% when K2CO3 was used as a base) when 3a was used but such effect was not observed with the longer alkyl chain derivatives. Likewise, the use Cs2CO3 resulted in a slight increase in the ratio of 7 / 8 when 3a was used (59/32 versus 53/44), but ha d no effect on product distribution for a longer alkyl chain derivative such as 3e As it turns out, the original ACOM/phenol coupling method of Sloan and Koch90 proved to be the most effective in the present case as well.100 As mentioned above and shown in Tables 3-2 and 3-3, the mixtures resulting from the coupling of 3 and 6 are frequently contaminated with a large pe rcentage of 8

PAGE 64

51 especially when R offers little steric hi ndrance. Unfortunately, isolated yields of 7 suffer as a consequence (see Experimental). Compounds 7 and 8 could not be separated by simple crystallization, and could only be isol ated in poor to low yield (1-30%) by way of a time-consuming chromatographic procedure involving multiple passes through a column of silica gel. Reverse-phase chro matography failed to improve the separation. However, other have reported that a phenolic ester can be selectively cleaved in the presence of an aliphatic ester.101-105 Yet when these techniques were applied to mixtures of 7 and 8 a large portion of 7 was destroyed along with 8 Aminolysis with hydrazine106 and t -butylamine107 proved ineffective as well. Selective cleavage of 8 was finally achieved by subjecting the crude reaction mixt ure to a solution of imidazole in 30% aqueous acetonitrile.108 In general, the selectivity for 8 varied with the steric hindrance in R and R. This trend is reflected in the differences in isolated yield of 7 discussed in the Experimental Section. Even though a portion of 7 is cleaved via this procedure, it was quite practical in that it simplified the purification of 7 For example, compound 7 is easily separated from the product of the cleavage of 8 (parent phenol 6 ) via a single elution from a column of sili ca gel. Interestingly, byproduct 11 appeared to be unaffected by this procedure. Conclusions A new method has been developed for synthesizing ACOM iodides 3 in one step and in good yield starting from trioxane or paraldehyde. This reaction was found to be dependent on an unidentified catalyst that wa s present in older batches of NaI, but is absent in newer, purer batches. Although an optimized procedure for synthesizing the 3 using the newer brands of NaI was not develo ped, potential catalysts were identified. The coupling reaction of 3 with phenols 6 appears to be somewhat dependent on steric

PAGE 65

52 hindrance as measured by Chartons steric parameters. In fact, the percentage of alkylated phenol 7 in the product mixture increases w ith increasing steric hindrance in 3 and in 6 However, based on literature precedent92, 93 and new data from our lab (Table 3-3), alkylated phenol is favor ed over acylated phenol regard less of the steric hindrance in 3 or 6 contrary to the findings of Ouyang.91 As Ouyangs is the only report where R is a protected amino acid, this particular ac yl group may impart a unique reactivity to 3 not found in more common derivatives (i.e. where R = hydrocarbon). Experimental Batches of sodium iodide designated as old in the text were purchased from Aldrich (99+%, catalogue number 2 1763-8, lot numbers 1327 DK and 04229 CV; 99.5%, catalogue number 38311-2, lot number 11717 MG). Batches of sodium iodide designated as new in the text were purchased fr om Aldrich (99+%, catalogue number 217638, lot number 05412 BC; 99.5% catalogue number 383112, lot number 07908 CC) and from Fisher (Certified, catalogue number S 324-500, lot number 037120). Thin layer chromatography (TLC) plates (Polygram Sil G/UV 254) were purchased from Brinkman. Spectra (1H NMR) were recorded on a Varian Unity 400 MHz spectrometer or on a Varian EM-390 90 MHz spectrometer. Melting points were determined on a Meltemp melting point apparatus. Sodium sulfate and all solvents were purchased from Fisher. Trioxane and paraldehyde were purchased from Eastman Chemical Company. Iodine (crystalline) was purchased from Mallinckrodt. All other reagents were from Aldrich. Containers of NaI and Cs2CO3 were wrapped in parafilm and stored in a vacuum desiccator. Solvents listed as dry below were obtained as such following storage over 4-angstrom molecular sieves. Microanalyses were performed by Atlantic Microlab, Inc., Norcross, GA.

PAGE 66

53 General procedure for the synthesis of 3a-e and 3gsynthesis of 3a : Sodium iodide (12 mmol, from any of the old ba tches listed above) was added to a stirred solution of 1a (3.3 mmol) in 12 mL dichloromethane, and the suspension that resulted was cooled to 0 oC. A solution of 2a (10 mmol) in 12 mL dichloromethane was then added, and the resulting mixture was allowed to reach room temperature. The reaction vessel was protected from light with aluminum foil while the contents were allowed to continue stirring at room te mperature for 20-24 hours. The reaction mixture was filtered by vaccum followed by concentration of the f iltrate at room temperature on a rotary evaporator to give an orange-colored oil. A sample of this oil was dissolved in CDCl3 and analyzed by 1H NMR. The yield of 3a was then calculated on the basis of the molar ratio of the products. No further effort was made to purify 3a and it was used as such in subsequent reactions with phenol s. Representative spectrum (1H NMR, CDCl3) from the reaction of 1a with 2a to give 3a (R = CH3): 5.90 (s, 2 H), 2.10 (s, 3 H). Reaction of 1 with 2 by modified procedure using AlCl3/I2synthesis of 3f : Sodium iodide (15.2 mmol, 2.28 g, from Fish er) was added to a stirred solution of 1b (R = CH3) (4.2 mmol, 0.55 g) in 25 mL dichloromethan e, and the suspension that resulted was cooled to 0 oC. A solution of 2a (12.7 mmol, 1.00 g) in 10 mL dichloromethane was then added. Subsequent addition of alumin um chloride (0.42 mmol, 0.056 g) and iodine (0.084 mmol, 0.021 g) gave a mixture that was then allowed to warm to room temperature. The reaction vessel was protec ted from light with aluminum foil while the contents were allowed to c ontinue stirring at room temp erature for 20-24 hours. After such time, the reaction mixture was filtered by vacuum, diluted with 25 mL dichloromethane, then washed with 10 mL 10% aqueous Na2S2O3 followed by 10 mL

PAGE 67

54 brine. The organic phase was then dried over Na2SO4, filtered, and concentrated at room temperature on a rotary ev aporator to give 10.2 mmol 3f in Cl2CH2 (80% yield). Reaction of 6a with 3the reaction of 6a with 3e: To a stirred suspension of 6a (19.9 mmol, 3.01 g) and K2CO3 (39.8 mmol, 5.50 g) in 50 mL dry acetonitrile was added a solution of 3e (as indicated above, this soluti on is actually a mixture of 87% 3e 7% C7H15CO2CH2Cl, 4% (C7H15CO2)2CH2), 1% trioxane) in 15 mL dry acetonitrile. The mixture that resulted was allowed to stir ov ernight at room temperature. The reaction mixture was then filtered and concentrated in vacuo to give 10.82 g oily residue. 1H NMR (DMSO-d6) analysis of the solid retained in the filter cake revealed only a trace amount of unreacted APAP. 1H NMR (DMSO-d6) analysis of the oily residue showed 89 % conversion to products and the product di stribution shown in Table 3-2. Column chromatography (3 consecutive expe riments) on silica gel (gradient = hexane dichloromethane acetone) gave 2.37 g of 4-octanoyloxymethyloxyacetanilide 7e as an oil (39%). This oil was then triturated with pentane to give 1.89 g of 7e as colorless crystals (31%); mp = 53-54 oC; one spot on TLC (CHCl3 : acetone, 97 : 3) Rf 0.13; 1H NMR (CDCl3) 7.42 (d, J = 8 Hz, 2 H), 7.09 (brs, 1H), 6.99 (d, J = 8 Hz, 2 H), 5.73 (s, 2 H), 2.35 (t, J = 7 Hz, 2 H), 2.16 (s, 3 H), 1.62 (m, 2 H), 1.26 (quint, J = 7 Hz, 8 H), 0.87 (t, J = 7, 3 H); Anal. Calcd for C17H25NO4: C, 66.43; H, 8.20; N, 4.56. Found: C, 66.51; H, 8.19; N, 4.55. In addition to 7e the chromatography procedure described above also gave 2.24 g solid material composed of a mixture of 7e and 4-octanoyloxyacetanilide 8e in a ratio of 1.3 : 1.0. By way of simple crystal lization (EtOAc : hexane), 0.48 g of 8e was isolated from this mixture as colorl ess crystals (9%); mp = 106-108 oC (lit37 103-105 oC); one

PAGE 68

55 spot on TLC (CHCl3 : acetone, 97 : 3) Rf 0.10; Anal. Calcd for C16H23NO4: C, 69.29; H, 8.36; 5.05. Found: C, 69.06; H, 8.34; N, 5.04. The reaction of 6a with 3d was carried out and processed as described above for 3e, except that in this case, the sc ale was larger (52.1 mmol) and a different solvent gradient (hexane dichloromethane EtOAc) was used for column chromatography (3 consecutive experiments). In this way, 2.96 g of 4hexanoyloxymethyloxyacetanilide 7d was isolated as colorle ss crystals (20%); mp = 5052 oC; one spot on TLC (Cl2CH2 : EtOAc, 85 : 15) Rf 0.20; 1H NMR (CDCl3) 7.42 (d, J = 8 Hz, 2 H), 7.10 (brs, 1H), 6.99 (d, J = 8 Hz, 2 H); 5.73 (s, 2 H), 2.35 (t, J = 7 Hz, 2 H), 2.16 (s, 3 H), 1.63 (quint, J = 7 Hz, 2 H), 1.29 (m, 4 H), 0.87 (t, J = 7 Hz, 3 H), Anal. Calcd for C15H21NO4: C, 64.50; H, 7.58; N, 5.01. Found: C, 64.54; H, 7.56; N, 4.97. In addition to 7d 4-hexanoyloxyacetanilide 8d was isolated in a fashion similar to that described above for 8e : 0.30 g of pale blue crystals (2%), mp = 105-109 oC (lit37 107109 oC); one spot on TLC (Cl2CH2 : EtOAc, 85 : 15) Rf 0.17; Anal. Calcd. for C14H19NO4: C, 67.45; H, 7.68; N, 5.62. Found: C, 67.17; H, 7.64; N, 5.59. The reaction of 6a with 3c was carried out as described above for 3e, except that in this case, the scal e was much larger (112 mmol). The corresponding compound 8c was selectively destroyed as describe d below to give 51.24 g oil containing 7c 11c and 6a in the ratio of 50 : 1 : 3. The oil was then subjected to column chromatography (silica gel, EtOAc : hexane, 1 : 1) to give 12.51 g of 4-butryloxymethyloxyacetanile 7c as an oil (44%). Crystallization from diet hyl ether : 2-methylbutane gave 7.03 g of 7c as colorless crystals (25%); mp = 56-58 oC; one spot on TLC (EtOAc: hexane, 1 : 1) Rf

PAGE 69

56 0.16; 1H NMR (CDCl3) 7.42 (d, J = 8 Hz, 2 H), 7.13 (brs, 1H), 6.99 (d, J = 8 Hz, 2 H), 5.74 (s, 2 H), 2.34 (t, J = 7 Hz, 2 H), 2.17 (s, 3 H), 1.65 (m, 2 H), 0.94 (t, J = 7 Hz, 3 H); Anal. Calcd for C13H17NO4: C, 62.14; H, 6.82; N, 5.57. Found: C, 61.92; H, 6.85; N, 5.52. The reaction of 6a with 3b was carried out and processed as described above for 3c, except in this case, two consecuti ve column chromatography experiments (acetone : hexane 3 : 7) were requi red to separate 7b from 11b. Following crystallization from ether : pentane, 3.64 g of 4-propionyloxym ethyloxyacetanilide 7b was obtained as colorless crystals (15%); mp = 56-59 oC; one spot on TLC (acetone : hexane, 35 : 65) Rf 0.26; 1H NMR (CDCl3) 7.42 (d, J = 8 Hz, 2 H), 7.10 (brs, 1 H) 6.99 (d, J = 8 Hz, 2 H), 5.74 (s, 2 H), 2.39 (quart, J = 8 Hz, 2 H), 2.16 (s, 3 H), 1.15 (t, J = 8 Hz, 3 H); Anal. Calcd for C12H15NO4: C, 60.75; H, 6.37; N, 5.90. Found: C, 60.85; H, 6.35; N, 5.84. In addition to 7b column chromatography gave 3.14 g oil composed of a mixture of 4-propionyloxymethoxymethoxyacetanilide 11b solvent, and an unidentified compound. Crystallization from Cl2CH2 : hexane gave 1.05 g of 11b as colorless crystals (4%); mp = 71-73 oC; one spot on TLC (acetone : hexane, 3 : 7) Rf 0.18; 1H NMR (CDCl3) 7.08 (d, J = 8 Hz, 2 H), 6.88 (d, J = 8 Hz, 2 H), 5.40 (s, 2 H), 5.19 (s, 2 H), 2.35 (quart, J = 8 Hz, 2 H), 1.91 (s, 3 H), 1.31 (t, J = 8 Hz, 3 H); Anal. Calcd for C13H17NO5: C, 58.42; H, 6.41; N, 5.24. Found: C, 58.45; H, 6.43; N, 5.24. The reaction of 6a with 3a was carried out and processed as described above for 3c, except that in this case, an aqueous workup was no t performed on the aminolysis reaction (reaction mixture was too complex to det ermine ratio 7a, 8a, 11a

PAGE 70

57 and 6a). Instead, the crude mixture was subj ected to three consecutive column chromatography experiments (first two experiments used hexane Cl2CH2 acetone; final experiment used EtOAc : hexane, 1 : 1). In this way, 4acetyloxymethyloxyacetanilde 7a was obtained as 1.81 g pale green crystals (6.5%, crystallized from ether : 2-methylbutane); mp = 92-95 oC; one spot on TLC (Cl2CH2 : acetone, 95 : 5) Rf 0.21; 1H NMR (CDCl3) 7.43 (d, J = 8 Hz, 2 H), 7.14 (brs, 1 H), 7.00 (d, J = 8 Hz, 2 H), 5.73 (s, 2 H), 2.18 (s, 3 H), 2.12 (s, 3 H); Anal. Calcd for C11H13NO4: C, 59.19; H, 5.87; N, 6.27. Found: C, 58.96; H, 5.84; N, 6.22. In addition to 7a column chromatography also gave 1.61 g of 4acetyloxymethoxymethoxyacetanilide 11a as an oil. Crystallizat ion from diethyl ether : 2-methylbutane gave 0.40 g 11a as colorless crystals; mp = 91-93 oC; one spot on TLC (acetone : hexane, 3 : 7) Rf 0.15; 1H NMR (CDCl3) 7.10 (d, J = 8 Hz, 2 H), 6.87 (d, J = 8 Hz, 2 H), 5.38 (s, 2 H), 5.18 (s, 2 H), 2.07 (s, 3 H), 1.90 (s, 3 H); Anal. Calcd for C12H15NO5: C, 56.91; H, 5.97; N, 5.53. Found: C, 56.72; H, 5.96; N, 5.47. The reaction 6a with 3f was carried out and processed as described above for 3c except that in this case, the scale was much smaller (8.5 mmol). Using this procedure, 1.54 g of oil containing 7f : 6a in the ratio of: 16 : 1 was obtained. The oil was subjected to column chromatography (silica gel, acetone : hexane, 3 : 7) to give 0.79 g 4-acetyloxyethyloxyacetanilide 7f as a colorless solid. This solid was recrystallized from ether : 2-methylbutane to give 0.56 g 7f as colorless crystals (28%). Upon heating, 7f displayed an initial melting point of 82-92 oC. Once this material had cooled to room temperature and solidified, it was heated again. This time, 7f displayed a sharp melting point: 81-83 oC; one spot on TLC (acetone : hexane, 3 : 7) Rf 0.20; 1H NMR (CDCl3)

PAGE 71

58 7.41 (d, J = 9 Hz, 2 H), 7.08 (brs, 1 H), 6.92 (d, J = 9 Hz, 2 H), 6.51 (quart, J = 5 Hz, 1 H), 2.16 (s, 3 H), 2.10 (s, 3 H), 1.60 (d, J = 5 Hz, 3 H); Anal. Calcd for C12H15NO4: C, 60.75; H, 6.37; N, 5.90. Found: C, 60.69; H, 6.40; N, 5.91. General procedure for the selective aminol ysis of 7 in the presence of 8. The procedure described above for the reaction of 6a with 3 gave various mixtures of 7 8 11 unreacted 3 and 6a (determined by 1H NMR as described above). The mixture was then triturated in dichloromethane, filtered, and conc entrated in vacuo to give an oil. The oil was dissolved in 30% aqueous CH3CN (approx. 17 mL / 1 mmol 8 ), and imidazole was added (10 equiv. based on mmol 8 present in the oil, as determined by 1H NMR). The resulting mixture was allowed to reflux overnight. After such time, the solvent was removed in vacuo. The residue was dissolved in dichloromethane, washed with 1 M HCl (1/6 vol. of organic phase), a nd water (1/6 vol. of organic phase). The organic phase was dried over Na2SO4, filtered, and concentrated in vacuo to give an oil containing various ratios of 7 : 8 : 11 (determined by 1H NMR: specific ratios listed above). In Vitro Determination of Flux of ACOM Prodrugs of APAP O N H O O R O O H N H O 4-ACOM-APAP4-Hydroxyacetanilide (APAP) 6a 7a, R = CH37b, R = C2H57c, R = C3H77d, R = C5H117e, R = C7H15 Figure 3-8. Structure of 4-Hydroxyacetan ilide and Corresponding 4-ACOM Prodrugs

PAGE 72

59 Materials and Methods Melting points were determined on a Melt emp capillary melting point apparatus and are uncorrected. Ultraviolet (UV) sp ectra were obtained on a Shimadzu UV-265 or UV-2501 PC spectrophotometer. The vertical Franz diffusion cells (surface area 4.9 cm2, 20 ml receptor phase volume, 15 ml donor pha se volume) were purchased from Crown Glass (Somerville, NJ, USA). A Fisher (Pi ttsburgh, PA, USA) circul ating water bath was used to maintain a constant temperature of 32 oC in the receptor phase. Isopropyl myristate (IPM) was purchased from Givauda n (Clifton, NJ, USA). Theophylline (Th) was purchased from Sigma Chemical Co. (S t. Louis, MO, USA); all other chemicals were purchased from Fisher. The female hairless mice (SKH-hr-1) were obtained from Charles River (Boston, MA, USA). All proced ures involving the care and experimental treatment of animals were performed by Prof essor K. B. Sloan of the department of Medicinal Chemistry in agreement with the NIH Principles of Laboratory Animal Care. Physicochemical properties and analysis The molar absorptivity of each prodrug at 240 nm ( 240) in acetonitrile was determined in triplicate by dissolving a known amount of prodrug in acetonitrile, and analyzing the dilute solution by UV spectrophotometry. Since the concentration C was known, 240 could be calculated by way of Beers law: A240 = 240 l C, where l = cell length (1) For each prodrug, the solubility in isopropyl my ristate (IPM) was determined in triplicate by crushing a sample of the prodrug into a fi ne powder. Excess powder was added to a test tube containing 3 ml IPM. The test t ube was then insulated and the suspension was allowed to stir at room temperature (23 1 oC) for 24 hours on a magnetic stir plate. The

PAGE 73

60 suspension was filtered through a 0.25 m nylon sy ringe filter. A sample of the filtrate was diluted with acetonitrile and analyzed by UV spectrophotometry. In order to be consistent with a previous investigation of acetaminophen prodrugs,45 the absorbance at 240 nm (A240) was used to calculate the prodrug concentration C in the IPM solution using the Beers law relationship. In this case, since C is the concentration of a saturated solution, C is the solubility in IPM (SIPM): CSaturation = SIPM = A240 / 240 (2) Solubilities in water were also determined in triplicate using an identical protocol to the one described above, except that the suspen sions were only stirred for one hour before filtering. This was done in order to make direct comparisons between the present investigation and previous studies.45, 68 In each case, a sample of the filtrate was diluted with acetonitrile and analyzed by UV spectrophotometry using 240 in acetonitrile (Table 3-4). Table 3-4: Molar Absorptivities () of APAP 6a and Prodrugs 7a-e Compound 240 in ACNa, b 240 in Buffera, c 280 in Buffera, d 6a, APAP 1.36e 1.01 0.053f 0.174 0.020f 7a 1.48 0.011 7b 1.64 0.067 7c 1.56 0.057 1.20 0.025g 0.119 0.0025g 7d 1.58 0.050 7e 1.46 0.044 a Units of 1 x 104 L mol-1. b Molar absorptivities at 240 nm acetonitrile ( SD, n = 3). c Molar absorptivities at 240 nm in pH 7.1 phosphate buffer with 0.11% formaldehyde. d Molar absorptivities at 280 nm in pH 7.1 phosphate buffer with 0.11% formaldehyde. e Taken from Reference 45. f n = 5 ( SD). g n =6 ( SD). Partition coefficients were also determ ined in triplicate for each prodrug by using the saturated IPM solutions obtained from the so lubility determinations. Since solubility in pH 4.0 buffer (S4.0) is a parameter in the Roberts-Sloan database,20 acetate buffer (0.01 M, pH 4.0) was used as the aqueous phase in the partition coefficien t experiments. In

PAGE 74

61 this way, S4.0 could be estimated as described previously109 and the values included in the database. Thus, an aliquot of the satura ted IPM solution was partitioned against pH 4.0 buffer using the following volume ratios (V4.0 / VIPM) for compounds 7a, 7b, 7c, and 7d : 0.5, 2, 10, and 20, respectively. The two pha ses were vigorously shaken for 10 seconds,109 then allowed to separate via centrifuga tion. An aliquot of the IPM layer was removed, diluted with acetonitrile, and anal yzed by UV spectrophotometry as described above. The partition coefficien t was calculated as follows: KIPM:4.0 = [Aa/(Ab Aa)]V4.0/VIPM (3) where Ab and Aa are the respective absorbances be fore and after partitioning, and V4.0 and VIPM are the respective volumes of buffer and IPM in each phase. Due to the high solubility ratio exhibited by compound 7e it was not possible to accurately determine its partition coefficient using this proce dure. Therefore, in this case KIPM:4.0 was estimated from the average methylene K obtained for compounds 7a-d according to the following relationship log Kn + m = ( K)(m) + log Kn (4) where n is the number of methylene units in the promoiety of one prodrug and m is the number of additional methylene units in the promoiety with which it is compared. UV spectrophotometry was also us ed to determine the amount of 6a and prodrug present in the receptor phase of the diffusion cell. Since all the prodrugs in this study were part of a homologous seri es, it was assumed that satisf actory results would attain for the entire series from the use of the mola r absorptivity of one homolog. Thus, the molar absorptivities of compounds 7c and 6a were determined in pH 7.1 phosphate buffer (0.05 M, I = 0.11 M) containing 0.11% formaldehyde by first dissolving a known amount of

PAGE 75

62 either compound in acetonitrile. An aliquo t (0.500 mL) of the acetonitrile solution was removed, diluted with buffer, and analyzed by UV spectrophotometry to obtain the molar absorptivities shown in Table 3-4. Because there is considerable overlap between the UV spectra of APAP an d its ACOM prodrugs 7a-e the relative concentrations of each were determined using the following approac h. The differences in absorption were found to be greatest at 240 nm and at 280 nm. Th erefore, considering th e additive nature of absorption, the absorbance at each wavele ngth (assuming constant cell length) is A240 = P240CP + A240CA (5) A280 = P280CP + A280CA (6) where A is the absorbance at the respective wavelengths, is the molar absorptivity of either the prodrug (P) or APAP (A) at th e respective wavelengths, and C is the concentration of the respectiv e compounds in the mixture. Solving the two simultaneous equations gives the following solution for the prodrug concentration CP CP = (A280A240 A240A280) / (A280P240 A240P280) (7) Once CP is known, it may be inserted into equatio n 5 to give the following solution for the concentration of APAP CA: CA = (A240 P240CP) / A240 (8) Solubility parameters. Solubility parameters were calculated by the method of Fedors110 as demonstrated by Martin and coworkers111 and Sloan and coworkers.112 Diffusion cell experiments The flux of each prodrug was measured using skin samples from three different mice. Prior to skin removal, the mice were rendered unconscious by CO2 then sacrificed via cervical dislocation. Sk ins were removed by blunt diss ection and placed dermal side

PAGE 76

63 down in contact with pH 7.1 phosphate buffer (0.05 M, I = 0.11 M, 32 oC) containing 0.11% formaldehyde (2.7 ml of 36% aqueous formaldehyde/liter) to inhibit microbial growth and maintain the integrity of the skins113 throughout the experi ment. A rubber Oring was placed on top of the skin to ensu re a tight seal, and the donor and receiver compartments were fastened together with a metal clamp (see Figure 3-9). Water In Water Out Water Jacket Buffer Magnetic Stir Bar Sampling Arm Skin Rubber O-ring Suspension of Drug or Prodrug Donor Compartment Receiver Compartment Open to Atmosphere Figure 3-9. Diagram of Franz Diffusi on Cell (Metal Clamp Not Shown) Prior to the application of the prodrug, the skins were kept in contact with buffer for 48 h to allow any UV absorbing material to leach out. During this time, the receptor phase was removed and replaced with buffer 3 times in order to facilitate the leaching process. Twenty four hours before applica tion of the prodrug, a suspension (0.09 M to 0.80 M, i.e. roughly 10 SIPM) of the prodrug in IPM was pr epared and allowed to mix

PAGE 77

64 until it was needed in the diffusion cell experi ments. After the 48 hour leaching period, an aliquot (0.5 ml) of the prodrug suspension wa s 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 spectrophotome try (Table 3-4, equations 7 and 8) to determine the amounts of permeated APAP and prodrug. At each sampling time, the entire receptor phase was replaced with fres h buffer in order to maintain sink conditions. After the 48 h of the first application period, the donor susp ension 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 skin s were kept in contac t with buffer for an additional 24 h to allow all APAP species (i.e APAP and prodrug) to leach from the skin. Following this second leaching period, the rece ptor phase was replaced with fresh buffer and an aliquot (0.5 ml) of a standard drug/ vehicle (theophylline/propylene glycol) was applied to the skin surface: the second application period. Samples of the receptor phase were taken at 1, 2, 3, and 4 h and analyzed by UV spectrophotometry. The concentration of theophylline in the receptor phase was determined by measuring its absorbance at 270 nm ( = 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 species (APAP plus prodrug) against time as shown by the example in Figure 310. Flux could then be calculated by dividing the slope of the stead y-state portion of the graph by the surface area of the skin (4.9 cm2).

PAGE 78

65 y = 3.5788x 51.368 R2 = 0.9943 0 10 20 30 40 50 60 70 80 010203040 Time (h)Cumulative amount of APAP species (mol) Figure 3-10: Flux of Compound 7a through Hairless Mouse Skin Results and Discussion Physicochemical properties The solubilities of compounds 7a-e in IPM (SIPM) and in water (SAQ) are shown in Table 3-4. The relative sta ndard deviations were all 5% except for the SAQ value measured for compound 7e ( 9%). As expected, all th e prodrugs were more soluble in IPM than APAP (Table 3-5). Although there was a thirteen fold range in SIPM between the first and last member of the se ries, there was little variation in SIPM between the second and last member of the se ries. The biggest increase in SIPM (7 fold) occurred on going from the first (C1) to the second me mber (C2) of the series. Beyond C2, SIPM gradually increased until the fourth member of the series (C5), but began to decrease thereafter. It is reasonable to anticipate a p oint of diminishing re turns where no further increases in lipid solubility are realized by extending the length of the alkyl chain. Typically, the increase in lipid solubility ex hibited by the first memb er of a series of prodrugs or analogues results from mask ing a hydrogen bond donor in the parent compound. Elimination of the offending f unctional group results in a compound with

PAGE 79

66 lower crystal lattice energy than the parent, and is thus more easily solvated. For a homologous series in which the only differe nce between members is the length of an aliphatic chain, lipid solubility will increase as the chain is extended due to the incorporation of lipophilic groups. Howeve r, at a certain po int van der Waals interactions between the aliphatic chains become dominant, causing an increase in melting point and a decrea se in lipid solubility In general, the trends in SIPM for 7a-e appear to follow the trends in melting point, though there was less variation in melting point among 7b to 7e than there was in SIPM. It is important to note that the trends in SIPM shown here were observed previously in other prodrug series including 1-ACOM-5fluorouracil (1-ACOM-5U),67 3-ACOM-5-FU,69 1-AOC-5-FU,114 and bis-6,9-ACOM-6mercaptopurine (6,9-ACOM-6-MP).39 In addition to the 4-ACOM-APAP series physicochemical data from a recently described series of alkyloxycarbonyl (AOC) de rivatives of APAP (Figure 3-11) is also listed in Table 3-5. If homologs of the same alkyl chain length are compared ( 7a to 7c versus 8i to 8k ), the ACOM derivatives all exhib it lower melting points and, with the exception of 7a (C1), are more soluble in IPM and water than the corresponding members of the AOC series. However, comparis ons such as this do not take into account the structural differences between the prom oieties in question. In order to make comparisons between homologs of approxi mately equal size, it is perhaps more appropriate to consider th e fact that members of th e ACOM series contain a CH2O spacer between the phenoxy group of APAP and the carbonyl of the promoiety which extends the alkyl chain further from the phenyl ring of the parent. Disregarding the differences in size between a methylene unit and oxygen, the C1 member of the ACOM series should

PAGE 80

67 be compared to the C2 member of the AOC series. If similar comparisons are made for the remainder of the two series, the ACOM pr odrugs are 4 to 17-times more soluble in water and, with the exception of 7a are 3 to 5-times more soluble in IPM than the corresponding members of the AOC series. Table 3-5: Physicochemical Propertie s of 4-Hydroxyacetanilide 6a 4-ACOM-APAP Prodrugs 7a-e and 4-AOC-APAPa Prodrugs 8i-m Compound MWb mp oCc SIPM d, e, f SAQ d, f, g S4.0 d, h KIPM:4.0 i 6a, APAP 151 167-170 1.9a 100a 7a, C1 223 92-95 8.41 0.44 15.2 0.34 16.2 0.52 0.016 7b, C2 237 56-59 62.0 1.91 24.7 0.33 26.6 2.33 0.039 7c, C3 251 56-58 73.5 1.45 7.12 0.0073 8.26 8.90 1.00 7d, C5 279 50-52 109 1.48 0.597 0.018 0.90 121 19.1 7e, C7 307 53-54 98.7 3.77 0.0637 0.0060 0.048 2077 j 8i, C1 209 112-115 12.0 20.4 17.0 0.692 8j, C2 223 120-122 9.33 3.80 4.47 2.09 8k, C3 237 104-106 23.4 2.70 3.02 7.94 8l, C4 251 118-120 13.8 0.427 0.447 31.6 8m, C6 279 108-110 16.7 0.0479 0.0324 513 a Data from Reference 45. b Molecular weight. c Melting point (uncorrected). d Units of mM. e Solubility in isopropyl myristate (IPM). f Measured at 23 1 oC. g Solubility in water. h Solubility in pH 4.0 buffer estimated from SIPM/KIPM:4.0. i Partition coefficient between IPM and pH 4.0 acetate buffer. j Extrapolated from previous KIPM:4.0 in the series as described in the text. Although 7a-e were 4 to 60 times more lipid so luble than APAP, they were all much less soluble in water than APAP. In fact, the most water soluble member of the series, C2, exhibited only one-fourth the aque ous solubility of APAP (Table 3-5). SAQ increased on going from C1 ( 7a ) to C2 ( 7b ), but dropped off quickly as the alkyl chain length increased. Interestingly, the C2 member was also the most water soluble member of the 1-ACOM-5-FU67 and 3-ACOM-5-FU69 prodrug series. Contra ry to its effect on SIPM, masking a hydrogen bond donor in the pare nt compound can often lead to lower SAQ relative to the parent. Such was the case in the present study and in previous prodrug

PAGE 81

68 series including 7-ACOM -theophylline (7-ACOM-Th),57 1-alkylaminocarbonyl-5-FU (1AAC-5-FU),115 and 4-AOC-APAP.45 O N H O R O 4-AOC-APAP8 i, R = OCH38 j, R = OC2H58 k, R = OC3H78 l, R = OC4H98 m, R = OC6H13 Figure 3-11: Structure of 4-alkyloxyc arbonyl (AOC) derivatives of APAP In order to incorporate the physicochemical property data for 7a-e into the RobertsSloan database,20 pH 4.0 buffer was used as the a queous phase in determinations of partition coefficients. Partiti on coefficients obtained in this manner were then used to estimate the solubilities of 7a-e in pH 4.0 buffer (S4.0, Table 3-5). Part ition coefficients between IPM and pH 4.0 buffer (KIPM:4.0) were experimentally determined for all compounds except for 7e (Table 3-5). The relative standard deviations in KIPM:AQ were all less than 10% except for 7c ( 11%) and 7d ( 16%). Although the average methylene K for the 4-ACOM-APAP series (0.60 0.05) is somewhat higher than the 4AOC-APAP series (0.55 0.06), it is c onsistent with average methylene K values seen in other ACOM prodrug series: 1-ACOM-5-FU,67 K = 0.60 0.14; 3-ACOM-5-FU,69 K = 0.59 0.01; 7-ACOM-Th,57 K = 0.58 0.05). Since the pa rtition coefficients and K values for 7a-d (Table 3-6) were reasona bly well-behaved, the average K value was used to estimate the partition coefficient for 7e (Table 3-5). Use of the solubility ratios SRIPM:AQ as a surrogate for KIPM:4.0, resulted in an average methylene SR value that was slightly higher than K, but exhibited a smaller standa rd deviation (0.62 0.03). The

PAGE 82

69 estimated solubilities in pH 4.0 buffer S4.0 were somewhat higher than SAQ for 7a-c (10 5%), while the calculated S4.0 for 7e was only 0.75 times the experimentally measured SAQ for 7e Due to the relatively large difference between SIPM and SAQ of 7d (SIPM/SAQ = 182), it was difficult to experimentally determine KIPM:4.0 with reasonable precision. As a consequence, S4.0 for 7d was 1.5 times higher than its SAQ, which is somewhat greater than the largest variation observed pr eviously in the 4-AOC-APAP series (S4.0 was 0.59 times the experimentally measured SAQ in the case of 4-(2 methoxyethyloxycarbonyloxy)acetanilide).45 Table 3-6 : Log Solubility Ratios (log SRIPM:AQ), Differences Between Log SRIPM:AQ ( SR), Log Partition Coefficients (log KIPM:4.0), Differences Between Log KIPM:4.0 ( K), and Solubility Parameters ( i) for Prodrugs 7a-e Prodrug log SRIPM:AQ a SR b log KIPM:4.0 c K d i e 7a -0.257 -0.285 12.04 7b 0.400 0.66 0.368 0.65 11.77 7c 1.01 0.61 0.949 0.58 11.54 7d 2.26 0.62 2.09 0.57 11.18 7e 3.19 0.57 3.32f 10.89 a Log of the ratio of the solubilities in IPM (SIPM) and water (SAQ). b SR = (log SRn + m log SRn)/m; n is the number of methylene units in th e promoiety of one prodrug and m is the number of additional methylene units in the promoiety with which it is compared. c Log of the partition coefficient between IPM and pH 4.0 buffer. d Same definition as in b with the exception that log KIPM:4.0 is used in place of log SRIPM:AQ. e Calculated as described in Reference 112 (units = (cal cm-3)1/2. f Extrapolated from previous KIPM:4.0 in the series as described in the text. Diffusion cell experiments To date, there has been only one re port of the topi cal delivery of 4hydroxyacetanilide (APAP) by a homologous series of prodrugs.45 In order to facilitate comparisons between the results of the present investigation to those of the prior study of 4-alkyloxycarbonyloxyacetanilide de rivatives (4-AOC-APAP), data from both prodrug series are listed in Table 3-7. As shown in Table 3-7, the fluxes ( SD) of the ACOM prodrugs with the exception of 7e ( 32%) were within the typical45 30% variation of in

PAGE 83

70 vitro experiments with hairless mice. Three of the five members of the ACOM series were more effective at delivering APAP through the skin than APAP itself. This is in contrast to the AOC series in which only one member (C1) pe rmeated the skin better than APAP. If comparisons are made between me mbers of the same alkyl chain length ( 7a to 7c versus 8i to 8k ), the ACOM derivatives are, with the exception of 7a 2 to 11-times more permeable than the corresponding memb ers of the AOC series. The flux of the most permeable derivative 7b was 3.6 times greater than that of APAP. An improvement of this magnitude is modest when compared to the results of other prodrug series. For instance, 6-ACOM derivatives of 6-mercaptopurine (6-MP)116 and 1-ACOM derivatives of 5-fluorouracil67 improve the flux of the parent by as much as 69 and 16 times, respectively. The apparent ineffectiveness of the ACOM promoiety in the present case may be explained by considering the differen ces in the physicochemical properties of the parent compounds. Compared to APAP, 5-FU and 6-MP are much less soluble in IPM and water. Thus it is not surp rising to find that the flux of APAP is two fold higher than the flux of 5-FU and 134 times greater than that of 6-MP. As a consequence of its relatively high SIPM and SAQ values, it is more difficult to improve the flux of APAP than it is to improve the flux of polar heterocycles such as 5-FU and 6-MP. It is also worth mentioning that the 7-ACOM derivatives of theophylline (Th),57 a polar heterocycle, exhibited only modest (2 fold) improvements in flux. Though Th is less soluble in lipid and aqueous solvents than APAP, it is 7 times more soluble in IPM than 5-FU while still exhibiting 54% of the water solubility of 5-FU Again, the better the biphasic solubility of the parent compound, the more difficult it is to improve the flux via a prodrug approach.

PAGE 84

71 When the receptor phases from the application of 7a-e were analyzed during steady-state flux conditions, only APAP was found. The exception was compound 7b in which the intact prodrug accounted for 9% of the total APAP species in the receptor phase (Table 3-8). Since this particular de rivative was also the most permeable member of the series, the system of cutaneous esterases in this case may have been overwhelmed and unable to completely hydrolyze the prodr ug on its way through the skin. A similar phenomenon was observed in the 4-AOC-APAP series45 in which the derivative that exhibited the highest flux also delivered the highest percentage of intact prodrug through the skin (Table 3-8). Although no effort was made to determine the half-lives of 7a-e in the receptor phase buffer, the aqueous stabilit y may be estimated based on similar studies by others.93, 117 For example, Bundgaard and coworkers93 found that the 2acetyloxymethyl and 2-butyrloxymethyl derivative s of salicylamide exhi bit half-lives of 46 and 98 h, respectively at 37 oC in pH 7.4 buffer. Others have found that 4hexanoyloxyacetanilide displays an a pproximate half-life of 19 hours at 37 oC in pH 7.8 buffer.117 Given the generally higher pKa of an aryl hemiacetal compared to its corresponding phenol, the ACOM derivatives 7a-e should exhibit half-lives greater than 19 hours under the present experimental conditio ns. Thus, it is reasonable to assume that the absence of intact prodrug in the recep tor phase is due to extensive enzymatic hydrolysis in the skin and is not the result of substantial chemi cal hydrolysis in the receptor phase. Apparently, the fluxes of 7a-e are not artificially high due to damage sustained by the skin over the course of the experime nt. This assessment is based on control experiments in which a suspension of the ophylline in propylene glycol (Th/PG) was

PAGE 85

72 applied to the skin followi ng the removal of the prodrug donor phase. This second application of Th/PG resulted in Th flux values that were not significantly different from those through skins treated with IPM alone (T able 3-7). However, it is important to recognize that IPM is a well-know n penetration enhancer whic h can increase flux 50-fold compared to experiments where water was the vehicle.118 Although the apparent flux values of 7a-e are likely inflated due to IPM, this is not expected to change the rank order of flux within or between series.118 Table 3-7: Flux of Total APAP Species through Hair less Mouse Skin from Suspensions of 4-ACOM-APAP and 4-AOC-APAPa Prodrugs in IPM (JM), Second Application Flux of Theophylline through Hairless Mouse Skin from a Suspension in Propylene Glycol (JJ), Error in Predicting Log JM using the Roberts-Sloan Equation ( log Jpredicted), Error in Calculating Log JM using the Roberts-Sloan Equation ( log Jcalculated), and Ratio of the Flux of the Prodrug to the Flux of APAP (Jprodrug / JAPAP). Compound JM b JJ b log JM b log Jpredicted c log Jcalculated d Jprodrug / JAPAP 6a, APAP 0.51a 0.74a -0.29a -0.496e -0.484 7a, C1 0.730 0.23 0.934 0.136 -0.136 -0.104 -0.0911 1.4 7b, C2 1.86 0.24 0.935 0.0764 0.270 -0.213 -0.197 3.6 7c, C3 0.777 0.20 0.780 0.224 -0.109 -0.350 -0.331 1.5 7d, C5 0.344 0.062 0.857 0.148 -0.464 -0.254 -0.231 0.67 7e, C7 0.110 0.028 0.687 0.147 -0.957 -0.0366 -0.00703 0.22 8i, C1 1.00 1.12 0.00 -0.0953e -0.0794 2.0 8j, C2 0.174 0.64 -0.76 -0.482e -0.464 0.51 8k, C3 0.355 1.14 -0.45 -0.260e -0.240 0.69 8l, C4 0.0977 0.85 -1.01 -0.264e -0.241 0.20 8m, C6 0.0324 0.76 -1.49 -0.162e -0.133 0.063 Controlf 1.02 0.13g a From Reference 45. b Units of mol cm-2 h-1. c Predicted from log JM = -0.497 + 0.519 log SIPM + 0.481 S4.0 0.00268 MW (coefficients from n = 61 da tabase, Reference 45, were recalculated using SAS 8.1). Error in prediction = log JM predicted log JM. d Calculated from log JM = -0.545 + 0.511 log SIPM + 0.489 S4.0 0.00253 MW (n = 61 + current data gives a new database of n = 66 compounds). Error in calculation was from log JM calculated log JM. e Already included in the n = 61 database, so the value listed here is actually the difference between log JM and a calculated value for flux, log Jcalculated. f Skins were sequentially subjected to 48 h conditioning, 48 h contact with IPM, methanol wash, 24 h leaching. g From Reference 112. If the fluxes of 7a-e are normalized by their respec tive solubilities in IPM, the corresponding permeability coefficients PM are obtained (Table 3-8). PM has units of

PAGE 86

73 distance per time (usually cm h-1) and is thus a measure of how quickly a compound diffuses through the skin. Because PM gives no indication of the amount, or dose, of the permeant that is entering the body, it is not cl inically useful apart from the appropriate solubility data. Nevertheless, PM is frequently used in the literature to quantify the permeation efficiency of a compound through skin.5, 18 One of the most popular expressions of PM, the Potts-Guy equation (9),42 shows that PM is positively dependent on the octanol-water partition coefficient (KOCT:AQ) and negatively dependent on molecular weight (MW): log PM = -6.3 + 0.71 log KOCT:AQ 0.0061 MW (9) Table 3-8: Percent Intact Prodrug Detected in Receptor Phase during Steady-State (% Intact), Log Permeability Coefficients (log PM), Concentrations of APAP Species in Skin (CS), and Dermal/Transdermal Delivery Ratios for APAP 6a 4-ACOM-APAP 7a-e and 4-AOC-APAP Prodrugsa 8i-m Compound % Intactb log PM c CS d D/Te 6a, APAP -0.571 2.74 0.70f 0.046 7a, C1 0 -1.06 2.67 0.572 0.031 7b, C2 9 -1.52 13.1 2.10 0.060 7c, C3 0 -1.98 5.56 0.535 0.061 7d, C5 0 -2.50 3.55 1.05 0.088 7e, C7 0 -2.95 2.72 1.55 0.21 8i, C1 64 -1.08 5.45 1.57f 0.046 8j, C2 14 -1.73 1.08 0.13f 0.053 8k, C3 25 -1.82 2.84 1.44f 0.068 8l, C4 0 -2.15 1.91 0.08f 0.17 8m, C6 0 -2.71 1.79 0.43f 0.47 a From Reference 45. b Percent intact prodrug detected in the 31 h receptor phase sample. c Calculated from log JM log SIPM, units of cm h-1. d Amount of total APAP species (in units of mol) in receptor phase after 24 hours follo wing donor phase removal to allow APAP and prodrug to leach out of skin. e Calculated from D/T = [(CS/4.9 cm2 24 h)]/JM. f From Reference 119. Such a relationship suggests that percut aneous absorption is positively dependant on lipid solubility and negatively dependant on the water solubility of a permeant. However, a plot of the log PM values for 7a-e versus their respective log KIPM:4.0 values gave a negative slope (-0.519, r2 = 0.975, plot not shown). Similarly, a plot of log PM

PAGE 87

74 versus the calculated so lubility parameters of 7a-e gave a positive slope (Figure 3-12), demonstrating an inverse relationship between log PM and alkyl chain length (i.e. higher SIPM, lower i). These results are consiste nt with the findings of others45, 69, 118 and support the idea20 that lipophilicity alone is not a good predictor of flux. y = 1.6415x 20.853 R2 = 0.9974 -4.0 -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 10.81111.211.411.611.81212.2 Solubility ParameterLog PM Figure 3-12: Plot of Solubility Parameter versus Log P for 4-ACOM-APAP Prodrugs 7a-e To further illustrate the relatively weak dependence of flux on lipid solubility, consider the SIPM and SAQ values for APAP prodrugs 7a-e and 8i-m (Table 3-5). Although compound 7c is 6.1 times more soluble in IPM than 8i compound 8i is 2.9 times more soluble in water than 7c This increase in water solubility on going from 7c to 8i though modest, resulted in 1.3 fold greater flux for 8i compared to 7c The impact of SAQ on flux is more distinct when comparis ons are made between individual members of a series. For instance, 7e is 1.6 times more soluble in IPM than 7b but 7b is 388 times more soluble in water. As a result, the flux of 7b is 17 times greater than the flux of 7e In order to ascertain the relative impact of so lubility in a lipid, sol ubility in water, and partition coefficient on flux, the trends in SIPM, S4.0, KIPM:4.0, and JM for APAP 6a and its

PAGE 88

75 prodrugs 7a-e and 8i-m are graphically represented in Figure 3-13 (a Wasdo plot).119 What is clear from such a representation is that KIPM:4.0 is of little positive predictive value in determining the rank order of flux. For each increase in alkyl chain length, there is a corresponding increase in KIPM:4.0 regardless of the trends in JM. It is interesting to note that while similar observati ons have been made by others,118 the idea that partition coefficient is predictive of flux120 remains an erroneous yet persistent5 concept. A similar conclusion may be reached by examining the trends in SIPM. Within the AOC series and to a lesser extent in the AC OM series, the trends in SIPM are relatively flat across the series despite the fact that JM grows progressively smaller. In contrast, the trends in S4.0 generally mirror the trends in JM across a series. Although such trends imply that water solubility is a better predictor of flux than lipid solubility, th e reality is that flux is best predicted when both properties are considered.43 This is demonstrated in the present case by the fact that the most permeable members of both series ( 7b and 8i ) exhibit the best mixture of high SIPM and high S4.0. Such behavior is no doubt related to the biphasic nature of the absorption barrier presente d by the stratum corneum (see Chapter 1). Although it is obvious that flux is pos itively dependent on lipid and aqueous solubility, there is currently only one mathematical model available for quantifying such a relationship (see Chapter 1): log JM = x + y log SIPM + (1 y) log S4.0 z MW (10) log JM = -0.491 + 0.520 log SIPM + 0.480 log S4.0 0.00271 MW (11) Equation 10, or the Roberts-Sloan ( RS ) model,43 was originally based on a database (n = 42) of 7 different series of prodrugs of polar heterocycles. This database was recently updated45 to include two new series of hetero cyclic prodrugs and one new series of

PAGE 89

76 phenolic prodrugs (4-AOC-APAP) resulting in a more structurally diverse database of 61 compounds. A fit of that data to equation 10 gave the form of RS expressed by equation 11.45 In its present state, the model is h eavily dependent on data from heterocyclic compounds: 59% 5-FU related entries, 18% 6MP related entries, and 10% Th related entries in the database. On ly 8 of the 61 entries (13%) are of a phenolic compound (i.e. APAP). Therefore, it was of interest to determine whether equation 11 could accurately predict the flux of the 4-ACOM prodrugs 7a-e of APAP. Applica tion of equation 11 to prodrugs 7a-e resulted in predicted flux values (Jpredicted, data not shown) that were consistently higher than the expe rimentally determined fluxes (JM). The differences between log JM and log Jpredicted ( log Jpredicted) for 7a-e are listed in Table 3-7. On average, the error in predicting log JM ( log Jpredicted) for 7a-e was 0.192 0.124 log units. -2 -1 0 1 2 3 4CompoundLog Parameter Value Figure 3-13: Log SIPM ( ), Log S4.0 ( ), Log KIPM:4.0 ( ), and Log JM ( ) Values for APAP 6a 4-ACOM-APAP Prodrugs 7a-e and 4-AOC-APAP Prodrugs 8i-m 6a 7a 7c 8j 7d 8i 8l 8k 8m 7e 7b

PAGE 90

77 In order to increase the dive rsity of the database and improve the predictive power of RS prodrugs 7a-e were incorporated into th e database. A fit of the SIPM, S4.0, and MW for the resulting n = 66 entries to equation 10 gave the following estimates for x, y, and z: x = -0.545, y = 0.511, z = 0.00253, r2 = 0.915. These parameter estimates were then used to calculate JM for all 66 compounds (data not shown). A plot of JM versus the calculated flux values is shown in Figure 3-14. The differences between the experimental and calculated fluxes ( log Jcalculated) for APAP 6a and its prodrugs 7a-e and 8i-m are listed in Table 3-7. As shown in Table 3-7, the log Jcalculated for 6a 7a-e and 8i-m decreased with the incorporation of the 4-ACOM-APAP da ta into the database. On average, the log Jcalculated for 7a-e (0.171 0.126 log units) was some what higher than the average log Jcalculated for the entire n = 66 database (0.155 0.118 log units), but was much lower than the average log Jcalculated for 8i-m (0.231 0.148 log units). Interestingly, APAP and its prodrugs all exhibit lo wer than expected fluxes based on the present form of RS (Figure 3-14). In addition, the average log Jcalculated for APAP and its prodrugs ( 6a plus 7a-e plus 8i-m ; 0.227 0.133 log units) is quite a bit higher than the average log Jcalculated for the database as a whole. In order to determine whether 4-ACOM-A PAP prodrugs function better as dermal (delivery into the skin itself) or transder mal (delivery through the skin and into the systemic circulation) delivery agents, the sk ins were kept in contact with buffer for 24 hours after removing the donor phase to allo w APAP and prodrugs to leach out. The amount of total APAP species leached from the skin (CS) is shown in Table 3-8. As shown in Table 3-8, the rank order of CS generally follows the rank order of flux. In other words, the most permeable members of th e series were also the most effective at

PAGE 91

78 increasing the concentration of APAP in the skin. Three out of the five ACOM derivatives were able to deliver more APAP into the skin than suspensions of topically applied APAP alone, with derivative 7b delivering up to 5-times more APAP. Using the CS values as an estimate for the amount of total APAP species delivered into the skin, dermal/transdermal delivery ratios (D/T, Table 3-8) were calculated from equation 12: D/T = [(CS/4.9 cm2 24 h)]/JM (12) -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 -2.5-2-1.5-1-0.500.511.5log JMIPM = 0.545 + 0.511 log SIPM + 0.489 log S4.00.00253 Experimental log JMIPM Figure 3-14: Plot of Experimental Versus Calculated Flux for 5-FU, 6-MP, and Th Prodrugs ( n = 53), APAP ( ), 4-AOC-APAP Prodrugs ( n = 5, plus two additional compounds mentioned in Refere nce 1 to give n = 7), and 4-ACOMAPAP Prodrugs ( n = 5) Most of the prodrugs exhibited D/T ra tios that were higher than APAP. Thus, compared with topically applied APAP al one, all but one of the ACOM prodrugs ( 7a ) were more effective at delivering APAP to the skin itself rather than through it. Among 7a-e the prodrugs that preferentially delivered more APAP into the skin itself were also the most lipophilic and least permeable member s of the series. Thus, compounds such as 7d and 7e are best suited for a therapeutic regi men involving sustained delivery of low

PAGE 92

79 levels of a drug, while the shorter chain de rivatives would allow for maximum exposure of the drug to the sy stemic circulation. Conclusions Despite the success of ACOM prodrugs in improving the transdermal delivery of heterocyclic drugs, there are currently no exam ples of this approach being applied to a phenol. The results presented here demonstrat e for the first time that ACOM derivatives are capable of improving the topical deliver y of a phenol. In general, the ACOM derivatives of acetaminophen (APAP) exhibi ted better biphasic solubility and lower melting points than the previously studied45 AOC derivatives. As a result, the 4-ACOMAPAP prodrugs were capable of improving th e delivery of acetaminophen by 4-fold. The trends in flux were found to depend on a bala nce between lipid and aqueous solubility. Addition of the 4-ACOM-APAP prodrugs to th e Roberts-Sloan database increased the structural diversity of the current da tabase and resulted in a more robust RS model. Given that all of the 4-ACOM-APAP derivatives contained simple aliphatic groups in the acyl chain, it is likely that even greater improvements in flux will be realized by incorporating more hydr ophilic functional groups into the acyl chain.20

PAGE 93

80 CHAPTER 4 ALKYLOXYCARBONYLOXYMETHYL (AOCOM) PRODRUGS OF ACETAMINOPHEN (APAP) Synthesis of AOCOM Prodrugs of 4-Hydroxyacetanlide (APAP) To date, there has been only one report in the literature of the synthesis of an AOCOM derivative of a phenol.121 In that study, Seki and co workers arrived at the target AOCOM compound (4-ethyloxycarbonyloxymethyloxya cetanilide) by way of a four-step synthetic route starting from methyl chloroformate (Figure 4-1, R = C2H5). At the time of Sekis investigation, one of the ke y reagents, chloromethylchloroformate 16 was commonly synthesized via the chlori nation of methyl chloroformate.84, 122 This method requires fractional distillation of the product mixture to obtain pure 16 and often provides low yields of the desired product. Currently, chloroformate 16 may be purchased from several suppliers and it is no longer s ynthesized in the lab on a regular basis.123 Since the AOCOM and ACOM promoieties are structurally similar, it was of interest to determine whether the same strategy that was used to synthesize ACOM iodide s (Chapter 3) could be used to eliminate the use of 16 (and 4 R = Oalkyl) altogether (alternative routes shown in Figure 4-1 starting from 1a ). In keeping with this strategy, chloroformates were allowed to mix with trioxane and NaI at r oom temperature. Unfortunately, no reaction was observed at room temperature, and at higher temperatures the chloroformate apparently underwent decarboxyl ation as indicated by the generation of gas. Various Lewis acid / NaI mixtures also failed to result in product. If a cataly tic amount of

PAGE 94

81 pyridine was added, approximately 70% of th e chloroformate was converted to alkyl iodide124 even at room temperature. CH3O Cl O +SO2Cl2ClCH2O Cl O ClCH2O R O ICH2O R O OO O 1415 16 4 Bz2O2RCOCl / NaI RCOCl APAP O CH2O R O OH APAP 3 1a RH NaI Figure 4-1: Synthetic Rout es to Alkyloxycarbonyloxymet hyl (AOCOM, R = Oalkyl) Prodrugs of 4-hydroxyacetanilide (APAP) An alternative two-step route to AOCOM iodides involving an intermediate AOCOM chloride 4 (R = Oalkyl) was also attempted in order to avoid purchasing the relatively expensive 16 (Figure 4-1). There are a few repor ts in the literature that suggest such an approach is feasible.81, 125 For example, ethyloxycarbonyloxyethyl chloride had been synthesized81 previously in good yield (48%) by reacting acetaldehyde with ethyl chloroformate in the presence of a catalytic amount of ZnCl2. Yet this method failed to work in the present case where the aldehyde is the formaldehyde trimer trioxane 1a Furthermore, although certain AOCOM alkyl halides can be synthesized from a monomeric aldehyde and chloroformate in the presence of a pyridine catalyst,125 this method also failed in the present investigation.

PAGE 95

82 As it was not possible to neither shor ten the synthesis of AOCOM iodide 3 nor make the corresponding chloride 4 (R = Oalkyl) parsimoniously, 3 (R = Oalkyl) was obtained through a two-step process starting from 16 (Figure 4-1). With 3 (R = Oalkyl) in hand, the target prodrugs could be obtained by coupling 3 (R = Oalkyl) with APAP. The coupling reaction used by Seki121 was an adaptation of the method of Sloan and Koch for the synthesis of ACOM ethers of phenols.90 Under those conditions (K2CO3 as base, acetone as solvent), Seki and coworkers noted that 4ethyloxycarbonyloxymethyloxyacetanilide was obtain ed in 20 % yield from the coupling of APAP with ethyloxycarbonyloxymethyl i odide following a reaction time of seven days. In an effort to improve the yield and ascertain the reaction parameters by which this reaction is governed, a series of AOC OM derivatives of phenols (with emphasis on APAP) was synthesized by the method of Seki121 and by a more efficient method involving phase-transfer catalys is. In addition, the results were compared to those obtained from the coupling reaction of AC OM halides with phenolsan analogous system whose reaction parameters are known (see Chapter 3).90 R O X O + 3 7 8+6 X = Cl, Br, I R = Oalkyl, alkyl R' = Alkyl, aromatic, etc. O H R' O O R O R' O R O R' Figure 4-2: Generalized Reac tion of AOCOM halides (R = Oalkyl) and ACOM halides (R = alkyl) 3 with phenols 6 In the present investigation, 4-hydroxyacet anilide (APAP) was chosen as a model phenol in order to make a direct comparison between this work and the work of Seki.121 In addition, if the reactivity of AOCOM halides was found to parallel that of ACOM

PAGE 96

83 halides,90 then the reaction mixtures were expected to contain various percentages of acylated phenol 8 as a byproduct (Figure 4-2). In th e present case, since the carbonate derivatives of APAP had been characterized previously,45, 126 adoption of this particular phenol as a model facilitated byproduct iden tification. As shown in Figure 4-3, AOCOM iodides may be obtained from the corres ponding chlorides via halogen exchange in acetone, preferably in the presence of sodi um bicarbonate to neut ralize traces of HI formed during the reaction.84 Subsequent reaction with phenols under the standard conditions (acetonitrile or acetone as solvent, K2CO3 as base)90, 121 or in a biphasic system in the presence of tetrabutylammonium hydroge n sulfate (Figure 4-3) gave mixtures of 7 and 8 RH+ pyridine Cl2CH2R O Cl O acetone NaI / NaHCO3R O I O O H Y K2CO3Bu4NHSO4/ Cl2CH2/ water 70-100% O Y O R O 6a: Y = 6b: Y = H 6c: 2,2,5,7,8-pentamethylchroman-6-ol NHCOCH3 O Y R O +17 16 4Cl O Cl O 3 7 8 a: R = OCH3b: R = OC2H5c: R = OC3H7d: R = OC8H17 e: R = OC10H21f: R = O i -Pr g: R = O t -Bu h: R = OCH3 (89%) i: R = OC2H5 (92%) j: R = OC3H7 (93%) k: R = OC8H17 (98%) l: R = OC10H21(90%) m: R = i -Pr (85%) n: R = t -Bu (59%) h: R = OCH3 (86%) i: R = OC2H5 (90%) j: R = OC3H7 (72%) k: R = OC8H17 (96%) l: R = OC10H21(93%) m: R = i -Pr (92%) n: R = t -Bu (87%) i: R = OCH3, Y = NHCOCH3 j: R = OC2H5, Y = NHCOCH3 k: R = OC3H7 Y = NHCOCH3 l: R = OC8H17 Y = NHCOCH3 m: R = OC10H21 Y = NHCOCH3 n: R = i -Pr, Y = H o: R = t -Bu, Y = H p: R = OCH3 (phenol = 2c) i: R = OCH3, Y = NHCOCH3 j: R = OC2H5, Y = NHCOCH3 k: R = OC3H7 Y = NHCOCH3 n: R = OC8H17 Y = NHCOCH3 o: R = OC10H21 Y = NHCOCH3 p: R = i -Pr, Y = H q: R = t -Bu, Y = H r: R = OCH3 (phenol = 2c) Figure 4-3: Reaction of AOCOM iodides with ph enols under phase-transfer conditions

PAGE 97

84 Previously, Sloan and Koch90 had shown that the coupling of ACOM halides with phenols is sensitive to the nucle ofugicity of X, with better leaving groups giving more alkylated product 7 Recently, others91 have suggested that the ratio 7 / 8 is also dependent on the steric hindrance of the acyl group (R group in 3 ). Although the data presented in Table 4-1 is not exhaustive, it suggests that the trends obs erved in reactions of ACOM halides with phenols are operative in the analogous reactions of AOCOM halides. For example, if X is a poor leaving group, 8 is favored, but as th e nucleofugicity of X increases, the product dist ribution shifts toward 7 (compare entries 1 and 2 with 4). For X = I, alkylated phenol 7 becomes the major product when the alkoxy chain length extends beyond OCH3. Interestingly, the ratio 7 / 8 when R = OCH3 increases by more than 3 fold when the reaction is carried out under phase-transfer conditions instead of the standard protocol (entry 3 versus entry 6). Under thes e conditions, there is an incremental increase in the percentage 7 with increasing steric hindrance (as measured by Chartons steric parameters127) in R (entries 6-8), but beyond propyloxy, the percentage of 7 remains fairly constant for the straight chain derivatives studied. However, the product distribution shif ts entirely toward 7 on going to more bulky R groups (entries 1112). On the other hand, the percentage of 7 may be increased even for sterically unhindered R if the phenol is sufficiently hinde red (entry 6 versus entry 13). This particular result (entry 13) is not without precedent since others91 have observed a similar trend in reactions of ACOM halides with phenols. Aside from its effect on product distribution, the advantages of the phase-tra nsfer reaction include shorter reaction times (one day) and higher ove rall yield compared to the method of Seki.121 Although no mention was made of product distribution, it is also worth noting that Wolff and

PAGE 98

85 Hoffmann128 have used a similar reaction system to successfully alkylate phenols with cyclic ACOM halides. Table 4-1: Product Distribut ion of the Reaction of RCO2CH2X 3 with Phenols 6 Under Various Reaction Conditions Distribution (%)a Entry R X Phenol Solvent Base 7 8 vb 1 OC2H5 [MeNC4H8]+ 6a acetonitrile MeNC4H8 0 100 (28)c 0.48d 2 OC2H5 Cl 6a acetonitrile K2CO3 3 58 3 OCH3 I 6a acetonitrile K2CO3 36 64 0.36d 4 OC2H5 I 6a acetone K2CO3 57 (17)c 43 (13)c 5 OC4H9 I 6a acetone K2CO3 58 42 0.58d 6e OCH3 I 6a Cl2CH2/H2O K2CO3 66 (18)c 34 (6)c 7e OC2H5 I 6a Cl2CH2/H2O K2CO3 74 (50)c 26 (13)c 8e OC3H7 I 6a Cl2CH2/H2O K2CO3 84 (43)c 16 (6)c 0.56 9e OC8H17 I 6a Cl2CH2/H2O K2CO3 82 (45)c 18 (3)c 0.61 10e OC10H21 I 6a Cl2CH2/H2O K2CO3 78 (41)c 22 (6)c 0.56f 11e OiPr I 6b Cl2CH2/H2O K2CO3 100 0 0.75d 12e Ot -Bu I 6b Cl2CH2/H2O K2CO3 100 0 1.22d 13e OCH3 I 6c Cl2CH2/H2O K2CO3 90 (33)c 10 (0)c 14g CH3 I 6b acetonitrile K2CO3 63 37 0.52h 15g C2H5 I 6a acetonitrile K2CO3 59 31 0.56h 16g C3H7 I 6a acetonitrile K2CO3 73 24 0.68h 17g C5H11 I 6a acetonitrile K2CO3 66 27 0.68h 18g C7H15 I 6a acetonitrile K2CO3 71 27 0.73h a Determined from 1H NMR spectrum of the crude reaction mixture. b Chartons steric parameter for R. c Isolated yield. d Reference 127. e Reaction mixture includes 1 equivalent tetrabutylammonium hydrogen sulfate. f Estimated from the relationship v = 0.406n + 0.108n + 0.059n 0.00839 in Charton, M. J. Org. Chem. 1978 43 3995-4001. g Data taken from Chapter 3. h Reference 94. As discussed previously in Chapter 3, ACOM halides react with phenols under the standard conditions to give mainly 7 as long as X is a good leaving group ( Br). Thus, the relatively low ratio 7 / 8 in the AOCOM series compared to the ACOM series (compare entries 3-5 with entries 14-18) was unanticipated. Moreover, since the

PAGE 99

86 carbonyl of a carbonate is usually less reactiv e than the carbonyl of the corresponding ester,55 one might expect less acylation when R is alkyloxy (as in AOCOM) than when it is alkyl (as in ACOM). Since the AOCOM iodides 3 (R = Oalkyl) were not purified other than to filter off NaCl and unreacted NaI (see Experimental below), it is worth considering whether any remain ing AOCOM chloride in crude 3 affected the product distribution. If 4 (R = Oalkyl) was reacting with 6 to any significant extent then the percentage of acylated product 8 would have increased as the percentage of 4 increased. In the case of entries 3, 4, and 5, the percentages of unreacted AOCOM chloride 4 in crude 3 (R = Oalkyl) were 2%, 9%, and 9% respec tively. Thus it does not appear that the product distribution was affected by the presence of AOCOM chloride in crude 3 (R = Oalkyl). On the other hand, an alysis of the steric paramete rs for both series (ACOM and AOCOM) suggests that differences in 7 / 8 between the series are directly related to differences in the steric hindrance of R ba sed on Chartons steric parameters (compare entries 3-5 to entries 14-18).94, 127 A plot of versus the ratio of 8 / 7 for the entries 3-5 and entries 14-18 is shown in Figure 4-4. Although the plot of the AOCOM series consists of only three data point s, the trends in the data s uggest that the coupling reaction of AOCOM iodides with phenols is much more sensitive to steric effects than the analogous reactions of ACOM iodides (slope = -4.9 versus slope = -0.77). A plot of versus 8/7 for entries 6-11 (Figure 4-5) demons trates a much weaker dependence of product distribution on steric effects when phase -transfer conditions are used in lieu of the standard conditions (slope = -1.3 versus slope = -4.9).

PAGE 100

87 y = -0.7714x + 0.9388 R2 = 0.9505 y = -4.9125x + 3.4107 R2 = 0.814 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 00.20.40.60.811.21.4 Steric ParameterAcy/Alk Figure 4-4: Plot of Char tons Steric Parameter for R Versus the Ratio of Acylated/Alkylated Product ( 8/7 ) Resulting from the Reactions of 6 with AOCOM Iodides (Entries 3-5 in Table 4-1, ) and ACOM Iodides (Entries 14-18 in Table 4-1, ) Under the Standard Reaction Conditions. y = -1.2955x + 0.9766 R2 = 0.9608 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.30.40.50.60.70.8 Steric ParameterAcy/Alk Figure 4-5: Plot of Char tons Steric Parameter for R Versus the Ratio of Acylated/Alkylated Product ( 8/7 ) Resulting from the Reactions of 6 with AOCOM Iodides (Entries 6-11 in Table 4-1, ) Under Phase-Transfer Conditions. Conclusions In conclusion, the data presented here sugge sts that steric hindrance plays a greater role in the coupling reactions of AOCOM ha lides with phenols than in the analogous

PAGE 101

88 reactions of ACOM halides. However, this problem may be circumvented through the use of phase-transfer catalysis. Under these conditions, the influence of steric hindrance (as characterized by Chartons steric parame ters) is minimized, reaction time is reduced, and overall yields are increased. Experimental Melting points were determined on a Meltem p melting point apparatus. Thin layer chromatography (TLC) plates (Polygram Sil G/UV 254) were purchased from Brinkman. Spectra (1H NMR) were recorded on a Varian Unity 400 MHz spectrometer or on a Varian EM-390 90 MHz spectrometer; chemical shifts listed below are in reference to Me4Si. Sodium iodide was from Fisher or Aldrich. Note: there was no difference in reactivity between old and new batches of NaI (see Chapter 3) when used in the Finkelstein reactio ns described here Sodium sulfate and all solvents were purchased from Fisher. Trioxane was purchased from Eastman Chemical Company. Chloromethylchloroformate was purchased from TCI America and Lancaster Synthesis. All other reagents were from Aldrich. All bulk solvents and silica gel for chromatography were from Fisher. Containe rs of NaI were wrapped in parafilm and stored in a vacuum desiccator. Solvents li sted as dry below were obtained as such following storage over 4-angstrom molecular sieves. Methanol, ethanol (absolute), propanol, and butanol were dried over 3-angs trom molecular sieves before they were used as reagents. Pyridine was dried over 4-angstrom molecular sieves before it was used. Microanalyses were performed by Atlantic Microlab, Inc., Norcross, GA. Note: In general, the compounds described below were selected for large scale synthesis on the basis of whether they were solids (oils are usually more difficult to characterize and isolate than solids). With that in mind, it should be noted that 4-

PAGE 102

89 butyloxycarbonyloxymethyloxyacetanilide and 4hexyloxycarbonyloxymethyloxyacetanilide were also synthesized but since they were oils they were never isolated on a large enough scale to eval uate in diffusion cells. General procedure for the synthesis of alkyloxycarbonyloxymethyl iodides (3, R = Oalkyl)methyloxycarbonyloxymethyl iodi de 3h (Note: it is not necessary to add NaHCO3 as suggested below. However, since NaHCO3 minimized the formation of 4-hydroxy-4-methyl-2-pentano ne during the Finkelstein reaction, it was almost always used in this study to synthesize the AOCOM iodides): To an icecold solution of chloromethyl chloroformate 16 (82.8 mmol) and methanol 17a (69 mmol) in methylene chloride (130 ml) wa s added pyridine (82.8 mmol) in methylene chloride drop-wise over 10 minutes. The mixture was allowed to warm to room temperature and continue stirri ng overnight. The reaction mi xture was then washed with 1 M HCl (35 ml) and water (35 ml), dried over Na2SO4, filtered, and concentrated on a rotary evaporator to give 4h as a pale yellow oil ( 61.1 mmol, 89% yield; by 1H NMR, this oil also contained 5.8 mmol Cl2CH2 and 2.4 mmol H2O; no 16 or 17a remained); 1H NMR (400 MHz, CDCl3): 5.74 (s, 2 H) and 3.89 (s, 3 H). Compound 4h was subsequently dissolved in 70 ml dry acetone and NaI (91.7 mmol) was added. This was immediately followed by the addition of NaHCO3 (6.1 mmol) and the resulting mixture was allowed to react at 40 oC for 4 hours. After such time, the mixture was concentrated on a rotary evaporator and triturated in met hylene chloride for approximately 30 minutes. The resulting mixture was filtered and concentrated as before to give 3h as a dark oil (52.5 mmol, 86% yield; by 1H NMR, this oil also contained 3.9 mmol CH3OCO2CH2Cl,

PAGE 103

90 29 mmol (CH3)2C(OH)CH2COCH3, 24 mmol Cl2CH2, 2.6 mmol acetone, and 3.7 mmol H2O); 1H NMR (400 MHz, CDCl3): 5.96 (s, 2 H) and 3.87 (s, 3 H). General procedure for the phase-transfer reactions : A mixture of phenol 6 (28.4 mmol) and K2CO3 (85.2 mmol) in 140 ml water was a llowed to stir several minutes before adding tetrabutyla mmonium hydrogen sulfate (28.4 mmol) and 70 ml methylene chloride. After several minut es of stirring, a solution of 3 (R = Oalkyl, 36.8 mmol) in 70 ml methylene chloride was added in portions to the reaction mixture. The resulting biphasic system was allowed to mix overnight at the maximum stirring rate of a standard magnetic stir plate. After such time, the phases were separated and the water layer was extracted with methylene chloride. The orga nic phases were combined and concentrated under vacuum to give an oily residue. A sample of this residue was analyzed by 1H NMR in order to determine the product distribut ions shown in Table 4-1 (Estimated % conversion of phenol to its corresponding AOCOM derivative for all reactions = 70-100 % by 1H NMR. In all cases, with the excep tion of entry 6, TLC of the water phase showed no evidence of unreacted 6 ). The residue was then triturated in ether and tetrabutylammonium iodide was removed by vacuum filtration. Compounds 7 and 8 were separated by column chromatography on si lica gel and recrystall ized from various solvents to obtain pure samples as described below. Isolation of compounds 7 and 8 (No effort was made to purify compounds 7n and 7o and the product distributions were determined by 1H NMR of the crude reaction mixtures.): 4-methyloxycarbonyloxymethyloxyacetanilide (7i) and 4methyloxycarbonyloxyacetanilde (8i) (scale = 40.4 mmol 6a ) were separated by column chromatography on silica gel (gradient = 100% hexane 70:30 hexane : acetone) to give

PAGE 104

91 2.16g 7i (22% crude yield) as a solid. Recr ystallization of this solid from Cl2CH2 : hexane gave 1.77 g 7i (7.41 mmol, 18% yield) as colorless crystals: mp = 104-106 oC; 1H NMR (400 MHz, CDCl3) 7.43 (d, J = 7 Hz, 2 H), 7.13 (brs, 1 H), 7.01 (d, J = 7 Hz, 2 H), 5.73 (s, 2 H), 3.83 (s, 3 H), 2.16 (s, 3 H); Anal. Calcd for C11H13NO5: C, 55.23; H, 5.48; N, 5.85. Found: C, 55.23; H, 5.52; N, 5.89. Compound 8i was obtained from the column as a solid (0.98 g, 4.7 mm ol, 12 % crude yield). This solid was recrystallized from EtOAc : hexane to give 0.53 g (2.53 mmol, 6% yield) 8i as colorless crystals: mp = 115-117 oC (lit. = 115.5-116.5 oC)126 4-Ethyloxycarbonyloxymethyloxyacetanilide (7j) and 4ethyloxycarbonyloxyacetanilide (8j) (scale = 3.2 mmol 6a) were separated by column chromatography on silica gel (gradient = 100% hexane 70:30 hexane : acetone) to give 0.58 g (2.3 mmol, 72% crude yield) 7j as a solid. This solid was recrystallized from ether : pentane to give 0.40 g (1.6 mmol, 50% yield) 7j as colorless crystals: mp = 83-85 oC (lit = 74-77 oC);121 1H NMR (400 MHz, CDCl3) 7.42 (d, J = 9 Hz, 2 H), 7.18 (brs, 1 H), 7.02 (d, J = 9 Hz, 2 H), 5.73 (s, 2 H), 4.24 (quart, J = 7 Hz, 2 H), 2.16 (s, 3 H), 1.32 (t, J = 7 Hz, 3 H); Anal. Calcd for C12H15NO5: C, 56.91; H, 5.97; N, 5.53. Found: C, 56.91; H, 6.05; N, 5.54. Compound 8j was also obtained from the column as a solid (0.20 g, 0.90 mmol, 28% crude yield). Recrystallization of this solid from EtOAc : hexane gave 0.08 g (0.4 mmol, 13% yield) 8j as colorless crystals; mp = 119-120 oC (lit = 121-122 oC).126 4-Propyloxycarbonyloxymethyloxyacetanilide (7k) and 4propyloxycarbonyloxyacetanilide (8k) (scale = 24.8 mmol 6a) were separated by column chromatography on silica gel (gradient = 100% hexane 75:25 hexane :

PAGE 105

92 acetone) to give 4.38 g 7k (16.4 mmol, 66% crude yield) as an oil. This oil was crystallized from ether : pentane to give 2.83 g (10.6 mmol, 43% yield) 7k as colorless crystals: mp = 68-69 oC; 1H NMR (400 MHz, CDCl3) 7.42 (d, J = 9 Hz, 2 H), 7.16 (brs, 1 H), 7.02 (d, J = 9 Hz, 2 H), 5.73 (s, 2 H), 4.14 (t, J = 7 Hz, 2 H), 2.16 (s, 3 H), 1.71 (m, 2 H), 0.96 (t, J = 7 Hz, 3 H); Anal. Calcd for C13H17NO5: C, 58.42; H, 6.41; N, 5.24. Found: C, 58.46; H, 6.42; N, 5.25. Compound 8k was isolated from the column as a solid (0.80 g, 3.4 mmol, 14% crude yield). Recrystallization of this solid from EtOAc : hexane produced 0.39 g (1.6 mmol, 6% yield) 8k as colorless crystals: mp = 107-110 oC (lit = 105-108).129 4-Octyloxycarbonyloxymethyl oxyacetanilide (7l) and 4octyloxycarbonyloxyacetanilide (8n) (scale = 28.4 mmol 6a) were separated by column chromatography on silica gel (gradient = 100% hexane 80:20 hexane : acetone) to give 4.10 g 7l (12.2 4 3% crude yield) as a white solid. A second fraction from the column contained a mixture of 7l and 8n Both fractions were recrystallized from ether : pentane and combined to give 4.36 g (12.9 mmol, 45% yield) 7l as colorless crystals: mp = 64-65 oC; 1H NMR (400 MHz, CDCl3) 7.42 (d, J = 9 Hz, 2 H), 7.23 (brs, 1 H), 7.02 (d, J = 9 Hz, 2 H), 5.73 (s, 2 H), 4.17 (t, J = 7 Hz, 2 H), 2.16 (s, 3 H), 1.67 (m, 2 H), 1.40-1.20 (m, 10 H), 0.87 (t, J = 7 Hz, 3 H); Anal. Calcd for C18H27NO5: C, 64.07; H, 8.07; N, 4.15. Found: C, 64.04; H, 8.12; N, 4.10. Compound 8n was also isolated from the column as a solid. This solid wa s recrystallized from EtOAc : hexane to produce 0.29 g (0.94 mmol, 3% yield) 8n as colorless crystals: mp = 80-82 oC (lit = 82.5-83 oC).126

PAGE 106

93 4-Decyloxycarbonyloxymethyl oxyacetanilide (7m) and 4decyloxycarbonyloxyacetanilide (8o) (scale = 23.2 mmol 6a) were separated by column chromatography on silica gel (gradient = 90:10 hexane : acetone 80:20 hexane : acetone) to give 4.45 g 7m as a white solid. Recrystallization of this solid produced 3.55 g (9.73 mmol, 42% yield) 7m as a white powder: mp = 54-56 oC; 1H NMR (400 MHz, CDCl3) 7.42 (d, J = 9 Hz, 2 H), 7.08 (brs, 1 H), 7.03 (d, J = 9 Hz, 2 H), 5.73 (s, 2 H), 4.17 (t, 7 Hz, 2 H), 2.17 (s, 3 H), 1.67 (m, 2 H), 1.40-1.20 (m, 14 H), 0.88 (t, J = 7 Hz, 3 H); Anal. Calcd for C20H31NO5: C, 65.73; H, 8.55; N, 3.83. Found: C, 65.90; H, 8.62; N, 3.82. Compound 8o was also isolated from the column as a solid (1.19 g, 3.36 mmol, 14% crude yield). Th is solid was recrystallized from EtOAc : hexane to give 0.42 g (1.3 mmol, 6% yield) 8o as colorless crystals: mp = 85-88 oC. Although 8o had not been previously synthe sized, the chemical shifts for CH2C H2O2C ( 4.24, t, 2 H) and the AB quartet ( 7.50, d, 2 H; 7.13, d, 2 H) in 8o were consistent with those exhibited by other members in the series. 6-Methyloxycarbonyloxymethyloxy-2,2,5,7,8pentamethylchroman (7p) and 6methyloxycarbonyloxy-2,2,5,7,8-pentamethylchroman (8r) (scale = 8.6 mmol 6c) were separated by column chromatography on silica gel (gradient = 100% hexane 97:3 hexane : acetone) to obtain 1.56 g 7p (5.06 mmol, 59% crude yield) as a yellow oil. The oil was crystallized from ethe r : pentane to get 0.86 g (2.8 mmol, 33% yield) 7p as pale yellow cr ystals: mp = 94-95 oC; 1H NMR (400 MHz, CDCl3) 5.51 (s, 2 H), 3.82 (s, 3 H), 2.59 (t, J = 7 Hz, 2 H), 2.15 (s, 3 H), 2.11 (s, 3 H), 2.08 (s, 3 H), 1.79 (t, J = 7 Hz, 2 H), 1.29 (s, 6 H); Anal. Calcd for C17H24O5: C, 66.21; H, 7.84. Found: C, 66.16; H, 7.86. Compound 8r eluted from the column as a mixture of 7p and

PAGE 107

94 8r in a ratio of 85:15. Compound 8r could not be separated from this mixture by crystallization, and no further effort was made to isolate 8r In Vitro Determination of Fl ux of AOCOM APAP Prodrugs Figure 4-6: Structure of 4-Hydroxyacet anilide (APAP) and Corresponding 4-AOCOMAPAP Prodrugs Methods and Materials Melting points were determined on a Melt emp capillary melting point apparatus and are uncorrected. Ultraviolet (UV) sp ectra were obtained on a Shimadzu UV2501 PC spectrophotometer. The vertical Fr anz diffusion cells (surface area 4.9 cm2, 20 ml receptor phase volume, 15 ml donor phase vol ume) were purchased from Crown Glass (Somerville, NJ, USA). A Fisher (Pittsburgh, PA, USA) circulating water bath was used to maintain a constant temperature of 32 oC in the receptor phase. Isopropyl myristate (IPM) was purchased from Givaudan (Clif ton, NJ, USA). Theophylline (Th) was purchased from Sigma Chemical Co. (St. L ouis, MO, USA); all other chemicals were purchased from Fisher. The female hairless mice (SKH-hr-1) were obtained from Charles River (Boston, MA, USA). All proced ures involving the care and experimental O N H O O R O O H N H O 4-AOCOM-APAP7i, R = OCH37j, R = OC2H57k, R = OC3H77l, R = OC8H177m, R = OC10H214-Hydroxyacetanilide (APAP) 6a

PAGE 108

95 treatment of animals were performed by Prof essor K. B. Sloan of the department of Medicinal Chemistry in agreement with the NIH Principles of Laboratory Animal Care. Physicochemical properties and analysis The molar absorptivity of each prodrug at 240 nm (240) in acetonitrile (Table 4-2) was determined in triplicate by dissolving a known amount of prodrug in acetonitrile, and analyzing the dilute solution by UV spectrophotometry. Since the concentration C was known, 240 could be calculated by way of Beers law: A240 = 240 l C, where l = cell length (1) For each prodrug, the solubility in isopropyl my ristate (IPM) was determined in triplicate by crushing a sample of the prodrug into a fi ne powder. Excess powder was added to a test tube containing 3 ml IPM. The test t ube was then insulated and the suspension was allowed to stir at room temperature (23 1 oC) for 24 hours on a magnetic stir plate. The suspension was filtered through a 0.25 m nylon sy ringe filter. A sample of the filtrate was diluted with acetonitrile and analyzed by UV spectrophotometry. In order to be consistent with a previous investigation of acetaminophen prodrugs,45 the absorbance at 240 nm (A240) was used to calculate the prodrug concentration C in the IPM solution using the Beers law relationship. In this case, since C is the concentration of a saturated solution, C is the solubility in IPM (SIPM): CSaturation = SIPM = A240 / 240 (2) Solubilities in water were also determined in triplicate using an identical protocol to the one described above, except that the suspen sions were only stirred for one hour before filtering. This was done in order to make direct comparisons between the present investigation and previous studies.45, 68 In each case, a sample of the filtrate was diluted

PAGE 109

96 with acetonitrile and analyzed by UV spectrophotometry using 240 in acetonitrile (Table 4-2). Table 4-2: Molar Absorptivities () of APAP 6a and Prodrugs 7i-m Compound 240 in ACNa, b 240 in Buffera, c 280 in Buffera, d 6a, APAP 1.36e 1.01 0.053 0.174 0.020 7i 1.44 0.023 7j 1.53 0.041 1.11 0.036 0.101 0.014 7k 1.46 0.056 7l 1.52 0.048 7m 1.54 0.0027 a Units of 1 x 104 L mol-1. b Molar absorptivities at 240 nm acetonitrile ( SD, n = 3). c Molar absorptivities at 240 nm in pH 7.1 phosphate buffer with 0.11% formaldehyde ( SD, n = 5). d Molar absorptivities at 280 nm in pH 7.1 phosphate buffer with 0.11% formaldehyde ( SD, n = 5). e Taken from Reference 45. Partition coefficients were also determin ed in triplicate for each prodrug by using the saturated IPM solutions obtained from the so lubility determinations. Since solubility in pH 4.0 buffer (S4.0) is a parameter in the Roberts-Sloan database,20 acetate buffer (0.01 M, pH 4.0) was used as the aqueous phase in the partition coefficien t experiments. In this way, S4.0 could be estimated as described previously109 and the values included in the database. Thus, an aliquot of the saturated IP M solution was partitioned against pH 4.0 buffer using the following volume ratios (V4.0 / VIPM) for compounds 7i, 7j, and 7k : 0.7, 2.5, and 10, respectively. The two phases were vigorously shaken for 10 seconds,109 then allowed to separate via centri fugation. An aliquot of the IPM layer was removed, diluted with acetonitrile, and analyzed by UV spectr ophotometry as described above. Using the previously measured absorbance at 240 nm for the saturated solution, the partition coefficient was calculated as follows: KIPM:4.0 = [Aa/(Ab Aa)]V4.0/VIPM (3)

PAGE 110

97 where Ab and Aa are the respective absorbances be fore and after partitioning, and V4.0 and VIPM are the respective volumes of buffer and IPM in each phase. It was not possible to experimentally determin e partition coefficients for compounds 7l and 7m since their respective solubility ratios (SIPM/SAQ) were much too high. Therefore, in these cases KIPM:4.0 was estimated from the average methylene K obtained for compounds 7i-k according to the following relationship log Kn + m = ( K)(m) + log Kn (4) where n is the number of methylene units in the promoiety of one prodrug and m is the number of additional methylene units in the promoiety with which it is compared. UV spectrophotometry was also us ed to determine the amount of 6a and prodrug present in the receptor phase of the diffusion cell. Since all the prodrugs in this study were part of a homologous seri es, it was assumed that satisf actory results would attain for the entire series from the use of the mola r absorptivity of one homolog. Thus, the molar absorptivities of compounds 7j and 6a were determined in pH 7.1 phosphate buffer (0.05 M, I = 0.11 M) containing 0.11% formaldehyde by first dissolving a known amount of either compound in acetonitrile (n = 5). An aliquot (1 ml) of the acetonitrile solution was removed, diluted with buffer, and analyzed by UV spectrophotometry to obtain the molar absorptivities shown in Table 4-2. Because there is considerable overlap between the UV spectra of APAP an d its AOCOM prodrugs 7i-m the relative concen trations of each were determined using the following approac h. The differences in absorption were found to be greatest at 240 nm and at 280 nm. Th erefore, considering th e additive nature of absorption, the absorbance at each wavele ngth (assuming constant cell length) is A240 = P240CP + A240CA (5)

PAGE 111

98 A280 = P280CP + A280CA (6) where A is the absorbance at the respective wavelengths, is the molar absorptivity of either the prodrug (P) or APAP (A) at th e respective wavelengths, and C is the concentration of the respectiv e compounds in the mixture. Solving the two simultaneous equations gives the following solution for the prodrug concentration CP CP = (A280A240 A240A280) / (A280P240 A240P280) (7) Once CP is known, it may be inserted into equatio n 5 to give the following solution for the concentration of APAP CA: CA = (A240 P240CP) / A240 (8) Solubility parameters. Solubility parameters were calculated by the method of Fedors110 as demonstrated by Martin and coworkers111 and Sloan and coworkers.112 Diffusion cell experiments The flux of each prodrug was measured using skin samples from three different mice. Prior to skin removal, the mice were rendered unconscious by CO2 then sacrificed via cervical dislocation. Sk ins were removed by blunt diss ection 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 (2.7 ml of 36% aqueous formaldehyde/liter) to inhibit microbial growth and maintain the integrity of the skins113 throughout the experi ment. A rubber Oring was placed on top of the skin to ensu re a tight seal, and the donor and receiver compartments were fastened together with a metal clamp (see Chapter 3, Figure 3-9). Prior to the application of the prodrug, the skins were kept in contact with buffer for 48 to allow any UV absorbing material to leach out. During this time, the receptor phase was removed and replaced with buffer 3 times in order to facilitate the leaching

PAGE 112

99 process. Twenty four hours before applica tion of the prodrug, a suspension (0.095 M to 0.664 M, i.e. generally 10 SIPM) of the pr odrug in IPM was prepared and allowed to mix until it was needed in the diffusion cell experiments. After the 48 hour leaching period, an aliquot (0.5 ml) of the prodrug susp ension 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 spect rophotometry (Table 4-2; equations 7 and 8) to determine the amounts of permeated APAP and prodrug. 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 susp ension 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 skin s were kept in contac t with buffer for an additional 24 h to allow all APAP species (i.e APAP and prodrug) to leach from the skin. Following this second leaching period, the rece ptor phase was replaced with fresh buffer and an aliquot (0.5 ml) of a standard drug/ vehicle (theophylline/propylene glycol) was applied to the skin surface: the second application period. Samples of the receptor phase were taken at 1, 2, 3, and 4 h and analyzed by UV spectrophotometry. The concentration of theophylline in the receptor phase was de termined by measuring its absorbance at 270 nm ( = 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 species (APAP plus prodrug) against tim e as shown by the example in Figure 4-7.

PAGE 113

100 Flux could then be calculated by dividing the slope of the steady-state portion of the graph by the surface area of the skin (4.9 cm2). y = 3.234x 38.853 R2 = 0.9996 0 10 20 30 40 50 60 70 80 010203040 Time (h)Cumulative Amount of APAP species (mol) Figure 4-7: Flux of Compound 7j through Hairless Mouse Skin Results and Discussion Physicochemical properties The solubilities in IPM (SIPM) and in water (SAQ) for prodrugs 7i-m are listed in Table 4-3. The relative sta ndard deviations of the SIPM and SAQ values were all 5% except for the SAQ value for 7l which was 11%. As expected, all of the AOCOM prodrugs exhibited lower melting points than AP AP and were more soluble in IPM than APAP. There was a steady increase in SIPM on going from the first to the last member of the series, with the last member of the seri es (C10) exhibiting the greatest increase (68fold) in SIPM over APAP. As seen in the alkyloxycarbonyloxy (AOC)45 and alkylcarbonyloxymethyl (ACOM, Chapter 3) prodrugs of APAP, all of the AOCOM derivatives were much less soluble in water than APAP. In fact, the most water soluble member of this series, 7j exhibited only 0.08-times the SAQ of APAP. In general, the SAQ values decreased along the series except for a slight increase in SAQ on going from

PAGE 114

101 C1 to C2. Interes tingly, the present SAQ value for 7j (C2) is twice as high as the value previously reported by Seki (Table 4-3).121 Although the reason for this discrepancy is unclear, it must be noted that the SAQ value of 4-ethyloxycarbonyloxyacetanilide 8j measured by Seki121 (SAQ = 2.15 mM, 25 oC, 0.01 M phosphate buffer, pH 7.0) is also about one-half the SAQ value measured by others45 under similar conditions. Table 4-3 : Physicochemical Propertie s of 4-Hydroxyacetanilide 6a 4-ACOM-APAP Prodrugs 7a-e ,a 4-AOC-APAP Prodrugs 8i-m ,b and 4-AOCOM APAP Prodrugs 7i-m Compoundc MWd mp oCe SIPM f, g, h SAQ f, h, i S4.0 f, j KIPM:4.0 k 6a, APAP 151 167-170 1.9b 100b 7a, C1 223 95-95 8.41 15.2 16.2 0.519 7b, C2 237 56-59 62.0 24.7 26.6 2.33 7c, C3 251 56-58 73.5 7.12 8.26 8.90 7d, C5 279 50-52 109 0.597 0.90 121 7e, C7 307 53-54 98.7 0.0637 0.048 2077l 8i, C1 209 112-115 12.0 20.4 17.0 0.692 8j, C2 223 120-122 9.33 3.80 4.47 2.09 8k, C3 237 104-106 23.4 2.70 3.02 7.94 8l, C4 251 118-120 13.8 0.427 0.447 31.6 8m, C6 279 108-110 16.7 0.0479 0.0324 513 7i, C1 239 104-106 7.93 0.14 7.20 0.14 8.39 0.946 0.022 7j, C2 253 83-85 (74-77)m 20.7 1.0 7.76 0.41 (3.72)n 7.51 2.76 0.22 7k, C3 267 68-69 45.8 1.5 2.00 0.091 4.97 9.21 0.51 7l, C8 337 64-65 66.4 1.9 0.00440 0.00047 0.029 2720l 7m, C10 365 54-56 130 2.4 o 0.0062 26500l a Data from Chapter 3. b Data from reference 45. c C1, C2... refer to the length of the alkyl chain. d Molecular weight. e Melting point (uncorrected). f Units of mM. g Solubility in isopropyl myristate (IPM). h Measured at 23 1 oC. i Solubility in water. j Solubility in pH 4.0 buffer estimated from SIPM/KIPM:4.0. k Partition coefficient between IPM and pH 4.0 acetate buffer. l Extrapolated from previous KIPM:4.0 in the series as described in the text. m Previously reported value from reference 121. n Value measured at 25 oC in 0.01 M phosphate buffer, pH 7.0 from reference 121. o Could not be determined

PAGE 115

102 In order to incorporate the physicochemical property data for 12 to 16 into the Roberts-Sloan database,20 pH 4.0 buffer was used as the aqueous phase in partition coefficient determinations (KIPM:4.0). Partition coefficients obtained in this manner were then used to estimate the solubilities of 7i-m in pH 4.0 buffer (S4.0, Table 4-3). Partition coefficients between IPM and buffer could be determined for all but the last two members of the series. These last two ho mologs, C8 and C10, exhibited such low solubilities in water that the present met hod for measuring partition coefficient was not useful. Relative standard deviations for the KIPM:4.0 values were all 8%. The average methylene K for this series (0.49 0.04) wa s much lower than the average K for the 4ACOM-APAP series ( K = 0.60 0.05), but was within the standard deviation of the average K for the 4-AOC-APAP series ( K = 0.55 0.06).45 While an average K of 0.49 is certainly lower than the values typica lly seen in prodrug se ries, an even lower value ( K = 0.44) has been reported45 for a series of me thoxyethyleneoxycarbonyl derivatives of APAP. Thus it seems that the experimental KIPM:4.0 values of the present series are reasonably well-behaved. Since the KIPM:4.0 values obtained for the first three homologs were reasonable, the average K value was used to calculate KIPM:4.0 for the last two members of the series (C8 and C10) Use of the solubility ratios SIPM:AQ as a surrogate for partition coefficient resulted in a slightly higher value for the average SR (0.54 0.14). The estimated solubility in pH 4.0 buffer was somewhat higher than the experimentally determined SAQ in the case of 7i and somewhat lower in the case of 7j For 7k and 7l the values for S4.0 were all much higher (2.5 and 5.5-times higher, respectively) than the co rresponding values for SAQ.

PAGE 116

103 Table 4-4 : Log Solubility Ratios (log SRIPM:AQ), Differences between Log SRIPM:AQ ( SR), Log Partition Coefficients (log KIPM:4.0), Differences between Log KIPM:4.0 ( K), and Solubility Parameters ( i) for Prodrugs 7i-m Prodrug log SRIPM:AQ a SR b log KIPM:4.0 c K d i e 7i C1 0.0424 -0.0242 11.87 7j C2 0.427 0.38 0.441 0.47 11.62 7k C3 1.36 0.66 0.964 0.52 11.41 7l C8 4.18 0.56 3.43f 10.68 7m C10 4.42f 10.48 a Log of the ratio of the solubilities in IPM (SIPM) and water (SAQ). b SR = (log SRn + m log SRn)/m; n is the number of methylene units in the promoiety of one prodrug and m is the number of additional methylene units in the promoiety with which it is compared. c Log of the partition coefficient between IPM and pH 4.0 buffer. d Same definition as in b with the exception that log KIPM:4.0 is used in place of log SRIPM:AQ. e Calculated as described in Reference 112 (un its = (cal cm-3)1/2. f Extrapolated from previous KIPM:4.0 in the series as described in the text. In order to facilitate comparisons be tween the AOCOM APAP series and other APAP derivatives (Figure 4-8), the relevant physicochemical property data of the 4AOC-APAP45 and 4-ACOM APAP series (Chapter 3) has been included in Table 4-3. If comparisons are made between members of the same alkyl chain length (C1 to C3), the AOCOM series is generally more soluble in IPM and less soluble in water than the AOC series. For instance, C2 and C3 AOCO M are 2.2 and 2.0-times, respectively, more soluble in IPM than the co rresponding members of the AOC series, while C1 and C3 AOC are 2.8 and 1.4-times, respectively, more soluble in water than the corresponding members of the AOCOM series. If simila r comparisons are made between the AOCOM and ACOM series, the C1 to C3 ACOM deriva tives exhibit higher so lubilities in both water and IPM than the correspondin g members of the AOCOM series.

PAGE 117

104 O N H O O R O 4-AOC-APAP 4-ACOM-APAP O N H O R O O N H O O R O 4-AOCOM-APAP 7a, R = CH37b, R = C2H57c, R = C3H77d, R = C5H117e, R = C7H15 7i, R = OCH37j, R = OC2H57k, R = OC3H77l, R = OC8H177m, R = OC10H21 8i, R = OCH38j, R = OC2H58k, R = OC3H78l, R = OC8H178m, R = OC10H21 Figure 4-8: Structures of Alkylcarbonyloxymethyl (A COM) and Alkyloxycarbonyl (AOC) Derivatives of APAP and Co mparisons between Homologs of Approximately Equal Size. If the structural differences between the pr omoieties are taken into account, slightly different conclusions are reached. Si nce the AOCOM derivatives contain a CH2O linker between the phenoxy group of APAP and the ca rbonyl of the prodrug, the alkyl chain in this series is extended two atoms farther fr om APAP than members of the same alkyl chain length in the AOC series. Therefore, rather than simply counting the number of methylene units in the alkyl chain, it may be more appropriate to include this two-atom unit in the total chain length when making comparisons between homologs of approximately equal size. Using this rati onale, C1 and C2 AOCOM are 2.7 and 18-times, respectively, more soluble in water than the corresponding members of the AOC series (C3 and C4). The differences in SIPM are not as one-sided. In one case the AOC member (C3) is more soluble in IPM (compared to C1 AOCOM), while in the other case the AOCOM member (C2) is more soluble (compare d to C4 AOC). This approach may also be used to compare the AOCOM derivatives with the corresponding ACOM derivatives. In this case, C2 and C3 ACOM are 7.8 a nd 3.6-times more soluble in IPM than the corresponding members of the AOCOM series (C1 and C2). In addition, C2 ACOM is 3.4-times more soluble in water than C1 AOCO M. Based on these results, it appears that

PAGE 118

105 substitution of oxygen for a methylene unit for in the carbonyl group of the prodrug (ACOM AOCOM) results in a decrease in li pid solubility w ith little to no improvement in water solubility. Diffusion cell experiments Results from the diffusion cell expe riments for the 4-AOCOM-APAP prodrugs are listed in Table 4-5. For most of the pr odrugs, samples of the receptor phase were taken every 3 h once steady-state flux wa s established. The exception was compound 7m in which samples only were taken every 12 h. Unfortunately, only two of the four samples were concentrated enough to be detected using the UV spectrophotometric method described above. As a consequence, the flux value for 7m listed in Table 4-5 is an estimate of JM based on the samples taken at 31 and 43 h. Also included in this table are the diffusion cell results from the 4-AOC-APAP45 and 4-ACOM-APAP (Chapter 3) series. With the exception of 7j ( 47%) and 7k ( 32%), the fluxes of 7i-m were all within the 30% varia tion typically observed45 in diffusion cell experiments with hairless mice. As a whole, the AOCOM deriva tives were not very effective at increasing the transdermal delivery of APAP. In the one case ( 7j ) where the flux of the prodrug was greater than that of APAP, the improvement was only marginal (1.3-fo ld). If the fluxes of members of the same alkyl chain length are compared, the first three homologs of the AOCOM series (C1 to C3) performed worse on average than the corresponding members of the ACOM series (average ratio of fluxes JAOCOM / JACOM = 0.44) but performed better on average than the correspondi ng members of the AOC series (average ratio of fluxes JAOCOM / JAOC = 1.7). If structural differences be tween the promoieties are taken into account (as in Figure 4-8), the AOCOM series is even more effective at delivering APAP

PAGE 119

106 than the AOC series (ave rage ratio of fluxes JAOCOM / JAOC = 4.0), but still less effective than the ACOM series (a verage ratio of fluxes JAOCOM / JACOM = 0.55). When the receptor phases from the application of 7i-m were analyzed during steady-state flux conditions, vari ous percentages of intact pr odrug and APAP were found (Table 4-6). The entries in Table 4-6 are from samples taken at 31 h and are representative of percentage s of intact prodrug observed at other times during steadystate. Although no effort was made to determine the half-times of 7i-m in the receptor phase buffer, aqueous stability may be estimated based on the work of others. For example, Seki and coworkers121 found that 7j exhibited a half-life of 200 h in pH 7.0 phosphate buffer (0.01 M) at 25 oC. Thus, under the present experimental conditions it is reasonable to assume that presence of APAP in the receptor phase is due to enzymatic hydrolysis of the prodrugs in the skin and is not the result of chemi cal hydrolysis in the receptor phase. In that regard, it is importa nt to recognize that the skins were kept in contact with buffer for 48 h prior to application of the prodrugs. During this preapplication period, the enzyma tic activity of the skin decr eases as hydrolytic enzymes are leached from the skin.116 Therefore, the extent to which 7i-m are hydrolyzed in the skin should be greater in vivo In general, the percent of intact prodrug d ecreased as the alkyl chain length increased. A sim ilar trend was previously observed45 in the 4-AOCAPAP series (Table 4-6) and in fact s hould not be surprising based on literature precedent.130 Given that 7j is the most permeable member of the series, the relatively high percentage of intact prodrug in this case is likely due to saturation of the esterase system in the skin.

PAGE 120

107 Table 4-5 : Flux of Total APAP Species through Hair less Mouse Skin from Suspensions of 4-ACOM-APAP,a 4-AOC-APAP,b and 4-AOCOM-APAP Prodrugs in IPM (log JM), Second Application Flux of Th eophylline through Hairless Mouse Skin from a Suspension in Propylene Glycol (JJ), Error in Predicting Log JM using the Roberts-Sloan Equation ( log Jpredicted), Error in Calculating Log JM using the Roberts-Sloan Equation ( log Jcalculated) and Ratio of the Flux of the Prodrug to the Flux of APAP (Jprodrug / JAPAP) Compound JM c JJ c log JM c log Jpredicted d log Jpredicted e log Jcalculated f Jprodrug / JAPAP 6a, APAPg 0.51 0.74 -0.29 -0.496h -0.484h -0.492 7a, C1 0.730 0.23 0.934 0.14 -0.136 -0.104 -0.091h -0.088 1.4 7b, C2 1.86 0.24 0.935 0.076 0.270 -0.213 -0.197h -0.188 3.6 7c, C3 0.777 0.20 0.780 0.22 -0.109 -0.350 -0.331h -0.317 1.5 7d, C5 0.344 0.062 0.857 0.15 -0.464 -0.254 -0.231h -0.207 0.67 7e, C7 0.110 0.028 0.687 0.15 -0.957 -0.037 -0.0070h 0.028 0.22 8i, C1 1.00 1.12 0.00 -0.095h -0.079h -0.074 2.0 8j, C2 0.174 0.64 -0.76 -0.482h -0.464h -0.455 0.51 8k, C3 0.355 1.14 -0.45 -0.260h -0.240h -0.226 0.69 8l, C4 0.0977 0.85 -1.01 -0.264h -0.241h -0.221 0.20 8m, C6 0.0324 0.76 -1.49 -0.162h -0.133h -0.103 0.063 7i, C1i 0.443 0.051 0.884 0.087 -0.353 -0.096 -0.083 -0.077 0.87 7j, C2i 0.660 0.31 1.12 0.43 -0.181 -0.117 -0.103 -0.094 1.3 7k, C3i 0.283 0.091 1.12 0.26 -0.549 -0.342 -0.323 -0.305 0.55 7l, C8i 0.0211 0.0018 1.03 0.0056 -1.67 -0.088 -0.056 -0.014 0.041 7m, C10i 0.00739 0.00018 0.713 0.059 -2.13 -0.313 -0.277 -0.227 0.014 Controlj 1.02 0.13k a From Chapter 3. b From Reference 45. c Units of mol cm-2 h-1. d Predicted from equation 10 (coefficients from n = 61 database, Reference 45, were recalculated using SAS 8.1). Error in prediction = log JM log Jpredicted. e Predicted from equation 11 (n = 61 + 4-ACOM-APAP (Chapter 3, n = 5) to give a database of n = 66 compounds). Error in prediction = log JM log Jpredicted. f Calculated from equation 12 (n = 61 + 4ACOM-APAP (n = 5) + present data (n = 5) to give a new database of n = 71 compounds). g From Reference 45. h Already included in the database, so the valu e listed here is actually the difference between log JM and a calculated value for flux. i Directly measured SAQ values were used in all equations to calculate flux. In the case of 7m SAQ was calculated from the average SR for the series. j Skins were sequentially subjected to 48 h conditioning, 48 h co ntact with IPM, methanol wash, 24 h leaching. k From Reference 112.

PAGE 121

108 Apparently, the fluxes of 7i-m are not artificially high due to damage sustained by the skin over the course of the first application or the leaching periods. This assessment is based on control experiments in whic h a suspension of theophylline in propylene glycol (Th/PG) was applied to the skin fo llowing the removal of the prodrug donor phase. This second application of Th/PG resulted in Th flux values that were not significantly different from those through skins treated with IPM alone (Table 4-5). However, it is important to recognize that IPM is a well-know n penetration enhancer which can increase flux 50-fold compared to experiments where water was the vehicle.118 Although the apparent flux values of 7i-m are likely inflated due to IPM, this is not expected to change the rank order of flux within or between series.118 If skin damage did not influence the rank order of flux, then the rank order of the observed flux is directly related to the rank order of the solubility of the prodrug in the skin (SMEM)a property which must be determined indirectly.41 Since the stratum corneum is a highly lipophilic membrane,3 it is commonly believe d that percutaneous absorption is directly dependant on lipid solubility (octanol, SOCT, is a typical model)131 or its surrogate, partition coefficient KOCT:AQ.5, 18, 120 Given the emphasis in the literature on the importance of lipid solubility in govern ing flux, it was of interest to determine the effect of lipid solubility on the fluxes of 7i-m If the fluxes of 7i-m are normalized by their respective solubilities in IPM, the corresponding permeability coefficients PM are obtained (Table 4-6). Permeability coefficients PM will be used in this section instead of JM since PM is frequently used in the literature to quantify the permeation efficiency of compounds through skin.5, 18 For the sake of comparison, PM of the ACOM and AOC prodrugs of APAP have also been include d in Table 4-6. A plot of the log PM values for

PAGE 122

109 7i-m versus their respective log KIPM:4.0 values gave a negative slope (-0.654, r2 = 0.987, plot not shown). These results are c onsistent with the findings of others45, 69, 118 and support the idea20 that lipophilicity alone as defined by KIPM:4.0 is not a good predictor of flux. Similarly, a plot of log PM versus the calculated solubility parameters of 7i-m gave a positive slope (Figure 4-9), demonstrati ng an inverse relationship between log PM and alkyl chain length (i.e. higher SIPM, lower i). Table 4-6: Percent Intact Prodrug Detected in Receptor Phase during Steady-State (% Intact), Log Permeability Coefficients (log PM), Concentrations of APAP Species in Skin (CS), and Dermal/Transdermal Delivery Ratios for 4-ACOMAPAP,a 4-AOC-APAP,b and 4-AOCOM APAP Prodrugs Compound % Intactc log PM d CS e D/Tf 6a, APAP -0.57 2.74 0.70g 0.046 7a, C1 0 -1.06 2.67 0.57 0.031 7b, C2 9 -1.52 13.1 2.1 0.060 7c, C3 0 -1.98 5.56 0.54 0.061 7d, C5 0 -2.50 3.55 1.05 0.088 7e, C7 0 -2.95 2.72 1.55 0.21 8i, C1 64 -1.08 5.45 1.57g 0.046 8j, C2 14 -1.73 1.08 0.13g 0.053 8k, C3 25 -1.82 2.84 1.44g 0.068 8l, C4 0 -2.15 1.91 0.08g 0.17 8m, C6 0 -2.71 1.79 0.43g 0.47 7i, C1 32 -1.25 2.83 0.62 0.054 7j, C2 46 -1.50 3.03 2.17 0.039 7k, C3 25 -2.21 4.53 1.38 0.14 7l, C8 0 -3.50 1.57 0.37 0.63 7m, C10 0 -4.25 0.825 0.118 0.95 a From Chapter 3. b From Reference 45. c Percent intact prodrug detected in the 31 h receptor phase sample. d Calculated from log JM log SIPM, units of cm h-1. e Amount of total APAP species (in units of mol) in receptor phase after 24 hours following donor phase removal to allow APAP and prodrug to leach out of skin. f Calculated from D/T = [(CS/4.9 cm2 24 h)]/JM. g From Reference 119.

PAGE 123

110 y = 2.1244x 26.359 R2 = 0.9845 -5 -4 -3 -2 -1 10.410.610.81111.211.411.611.812 Solubility ParameterLog PM Figure 4-9: Plot of Solubili ty Parameters versus Log PM for 4-AOCOM-APAP Prodrugs 7i-m If lipid solubility as defined by KIPM:4.0 is poorly correlated with skin permeability, then on which physicochemical properties is fl ux dependant? In order to ascertain the relative impact of solubility in a lipid, solubi lity in water, and partition coefficient on flux, the trends in SIPM, S4.0, KIPM:4.0, and JM for APAP 6a and its prodrugs ( 7a-e 8i-m and 7i-m ) are graphically represented in Figure 4-10 (a Wasdo plot).119 The most consistent trend between the seri es is the steady increase in KIPM:4.0 with increasing alkyl chain length. This is spite of the fact the JM generally decreases along a series. Thus it is clear from the present results that KIPM:4.0 is of little positive predictive value in determining the rank order of flux. Similarl y, there is no obvious relationship between SIPM and flux as SIPM grows larger along the AOCOM series ( 7i-m ), but remains relatively constant along the ACOM ( 7a-e ) and AOC ( 8i-m ) series. On the other hand, the trends in S4.0 generally mirror the trends in flu x. Such a relationship should not be surprising as the literature is replete with similar examples.57, 67, 68, 114-116 Although the

PAGE 124

111 dependence of flux on water solubility is most apparent in homologous series of compounds, such dependence has recently been demonstrated for a large number of unrelated compounds through human skin in vitro46 and for a small set of nonsteroidal anti-inflammatory drugs through human skin in vivo .132 Though water solubility is clearly important, flux is not governed by this property alone. In fact, most quantitative treatments of skin permeation data indicate that lipid solubility is either more important than42, 46, 132 or is equal in importance43 to water solubility. This is demonstrated in the present case by the fact that the most permeable compounds in each series ( 7b 8i and 7j ) are those that exhibit th e best mixture of high SIPM and high S4.0. -3 -2 -1 0 1 2 3 4 5Log Parameter Value Figure 4-10: Log SIPM ( ), Log S4.0 ( ), Log KIPM:4.0 ( ), and Log JM ( ) Values for APAP 6a 4-ACOM-APAP Prodrugs 7a-e 4-AOC-APAP Prodrugs 8i-m, and 4-AOCOM-APAP Prodrugs 7i-m Although it is obvious that flux is pos itively dependent on lipid and aqueous solubility, the Roberts-Sloan equation (RS, e quation 9) is currently the only mathematical model available for quantifying such a relationship (see Chapter 1): log JM = x + y log SIPM + (1 y) log S4.0 z MW (9) 6a 7c 7d 8i 8l 8k 8m 7e 7b 7a 8j 7i 7j 7k 7l 7m Compound

PAGE 125

112 log JM = -0.491 + 0.520 log SIPM + 0.480 log S4.0 0.00271 MW (10) Since it was first introduced in 1999,43 the database upon which RS (originally referred to as the Transformed Potts-Guy model) is based has been modified only once by the addition of 19 new entries to give an extended database of 61 compounds.45 A fit of that data to the RS model gave the form of RS expressed by equation 10.45 Use of equation 10 to predict the fluxes of 7i-m resulted in flux values (Jpredicted, not shown) that were higher than the experimentally determ ined values. In particular, the Jpredicted values for 7k to 7m were unusually high. A plot of log JM versus log Jpredicted using equation 10 is shown in Figure 4-11. The error in predicting log JM ( log Jpredicted) for 7i 7j 7k 7l and 7m using equation 10 was 0.128, 0.110, 0.532, 0.481, and 0.692 respectively. The average log Jpredicted for 7i to 7m (0.388 0.258 log units) was much higher than the average log Jpredicted for the entire database (n = 61, 0.154 0.117 log units). In addition, the average log Jpredicted for 7i to 7m was also substantially higher than the average log Jpredicted obtained for the 4-ACOM-APAP prodrugs 7a to 7e (0.192 0.124 log units) using equation 10 (Chapter 3). However, when the measured SAQ values for 7i to 7m [SAQ for 7m (0.00101 mM) was calculated fr om the average methylene SR (0.54) for the series] were used in equation 10 instead of their respective estimated S4.0 values (Table 4-3), the Jpredicted values were much closer to th e experimental flux values (i.e. lower log Jpredicted, Table 4-5). This improvement in accuracy is apparent in a comparison of Figure 4-11 with a new plot of log JM versus log Jpredicted using equation 10 (Figure 4-12). Use of the measured SAQ values for 7i to 7m also resulted in an average log Jpredicted for 7i to 7m (0.191 0.125 log units) that was much closer to the average log Jpredicted for the database as a whole (n = 61). Since RS predicts the flux of the

PAGE 126

113 AOCOM compounds ( 7i to 7m ) with greater accuracy when their respective SAQ values are used instead of their S4.0 values, the SAQ values for 7i to 7m will be used in all subsequent flux equations presented in this thesis. -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 -2.5-2-1.5-1-0.500.511.5log JMIPM = 0.497 + 0.519 log SIPM + 0.481 log S4.00.00268 MW Experimental log JMIPM Figure 4-11: Plot of Experimental Versus Calculated Flux for 5-FU, 6-MP, and Th Prodrugs ( n = 53), APAP ( ), 4-AOC-APAP Prodrugs ( n = 5, plus two additional compounds mentioned in Refe rence 1 to give n = 7), 4-ACOMAPAP Prodrugs ( n = 5), and 4-AOCOM-APAP Prodrugs ( n = 5) As discussed in Chapter 3, equation 10 is heavily dependent on data from heterocyclic prodrugs, and theref ore lacks a certain structural diversity. It was for this reason that the 4-ACOM-APAP 7a-e prodrugs were added to the database in Chapter 3. A fit of this new database (now n = 66) to the model gave the form of RS expressed by equation 11. log JM = -0.545 + 0.511 log SIPM + 0.489 log S4.0 0.00253 MW (11) With the incorporation of the 4-ACOM-APA P prodrugs into the database, equation 11 should be able to predict the fluxes of non-heterocyclic compounds with somewhat greater accuracy. This hypothesis was tested by using equation 11 to pr edict the fluxes of

PAGE 127

114 7i-m The individual log Jpredicted values for 6a 7a-e 8i-m and 7i-m using equation 11 are shown in Table 4-4. Alt hough the experimental fluxes of 7i-m were all lower than predicted based on equation 11, the average log Jpredicted for 7i-m (0.168 0.122 log units) decreased compared to when equation 10 was used. -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 -2.5-2-1.5-1-0.500.511.5log JMIPM = 0.497 + 0.519 log SIPM + 0.481 log S4.00.00268 Experimental log JMIPM Figure 4-12: Plot of Experimental Versus Calculated Flux for 5-FU, 6-MP, and Th Prodrugs ( n = 53), APAP ( ), 4-AOC-APAP Prodrugs ( n = 5, plus two additional compounds mentioned in Refe rence 1 to give n = 7), 4-ACOMAPAP Prodrugs ( n = 5), and 4-AOCOM-APAP Prodrugs ( n = 5). Note: In the 4-AOCOM-APAP series, SAQ has been substituted for S4.0 In order to further diversify the databa se and improve the predictive power of RS the 4-AOCOM-APAP prodrugs 7i-m were incorporated into th e database. A fit of the SIPM, S4.0 (again, SAQ is used for 7i-m instead of S4.0), MW, and JM for the resulting n = 71 entries to equation 9 gave the followi ng estimates for x, y, and z: x = -0.562, y = 0.501, z = 0.00248, r2 = 0.923: log JM = -0.562 + 0.501 log SIPM + 0.499 log S4.0 0.00248 MW (12) Equation 12 was then used to calculate JM for all 71 compounds (data not shown). A plot of JM versus the calculated flux values is s hown in Figure 4-13. The differences between

PAGE 128

115 the experimental and calculated fluxes ( log Jcalculated) for APAP 6a and its prodrugs ( 7ae 8i-m and 7i-m ) are listed in Table 4-5. As shown in Table 4-5, the log Jcalculated for 7a-d 8i-m and 7i-m decreased with the inclusio n of the 4-AOCOM-APAP data. However, the average log Jcalculated for APAP and its prodrugs ( 6a 7a-e 8i-m and 7im ; 0.195 0.143 log units) is sti ll higher than the average log Jcalculated for the database as a whole (0.156 0.117 log units). Although e quation 12 was able to predict the fluxes of APAP and its prodrugs with greater accuracy than equations 10 and 11, there was no advantage in using equation 12 to predict the rank order to fl ux since all three equations predicted the same rank order within each seri es. Interestingly, although the rank order of flux within the 4-AOC and 4-ACOM-APA P series was predicted with complete accuracy, the rank order of only three of the five 4-AOCOM-APAP compounds was accurately predicted. In order to determine whether AOCOM prodrugs of phenols would be more effective at delivering the pa rent compound to the skin (der mal delivery) or through the skin and into the systemic circulation (trans dermal delivery), the skins were left in contact with buffer for 24 hours after re moving the donor phase to allow APAP and prodrug to leach out. The amount of total APAP species leached from the skin (CS) is shown in Table 4-6. If the homologs of e qual alkyl chain length are compared (C1 to C3), the AOCOM prodrugs are generally more effective than the AOC prodrugs (average CS AOCOM / CS AOC = 4.9), but less effective than the ACOM prodrugs (average CS AOCOM / CS ACOM = 0.71) at increasing the co ncentration of APAP in the skin. In addition, all but the most lipophilic AOCOM derivatives 7l and 7m delivered more APAP to the skin than APAP itself. Using the CS values as an estimate of th e amount of total APAP species

PAGE 129

116 delivered to the skin, dermal/transdermal delivery ratios (Table 4-6) were calculated from equation 13: D/T = [(CS/4.9 cm2 24 h)]/JM (13) -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 -2.5-2-1.5-1-0.500.511.5log JMIPM = 0.562 + 0.501 log SIPM + 0.499 log S4.00.00248 Experimental log JMIPM Figure 4-13: Plot of Experimental Versus Calculated Flux for 5-FU, 6-MP, and Th Prodrugs ( n = 53), APAP ( ), 4-AOC-APAP Prodrugs ( n = 5, plus two additional compounds mentioned in Refe rence 1 to give n = 7), 4-ACOMAPAP Prodrugs ( n = 5), and 4-AOCOM-APAP prodrugs ( n = 5). Note: In the 4-AOCOM-APAP series, SAQ has been substituted for S4.0 Within each series, the derivatives that prefer entially delivered more APAP into the skin than through the skin are also the least permeable members of the series. Based on the D/T ratios, the AOCOM C1 to C3 derivativ es are on average more effective dermal delivery agents than the corresponding members of the AOC (average [D/T AOCOM]/[D/T AOC] = 1.3 ) and ACOM (average [D/T AOCOM]/[D/T ACOM] = 1.6) series. Regardless of the differences between the series, all but two of the derivatives ( 7j and 7a) delivered more APAP to the skin than topically applied APAP itself.

PAGE 130

117 Conclusions Although there are numerous repo rts of the use of prodrugs to improve the topical delivery of phenols,45, 60-63 all have made exclusive use of the acyl promoiety in which the prodrug is directly attached to the parent phenol through an ester-type bond. Such derivatives frequently exhibit higher melting points and poorer biphasi c solubility than the corresponding soft alkyl derivatives. Moreover, the only re ported example of an AOCOM derivative of a phenol (compound 7j ) was not evaluated in topical delivery experiments. Thus, the results presented here are significant in that they demonstrate for the first time that AOCOM derivatives of a phenol are capable of improving the topical delivery of the parent compound. While th e improvement in flux was marginal (1.3fold), three out of the five prodrugs test ed were more effective at increasing the concentration of APAP in the skin than topically applied APAP itself. Furthermore, all but one of the compounds tested were more ef fective than APAP at selectively delivering APAP to the skin rather than through it. The AOCOM derivatives of APAP are generally more effective dermal delivery agents than the previously described ACOM derivatives (Chapter 3). Based on thes e results, AOCOM prodrugs of phenols appear to be best suited for targeted delivery to the skin itself as opposed to the systemic circulation.

PAGE 131

118 CHAPTER 5 CONCLUSIONS AND FUTURE WORK The main advantages of topical drug deliv ery over other routes of administration are avoidance of first-pass metabolism, mini mal side effects, high incidence of patient compliance, and targeted delivery to the skin for treating local conditions. In order for topical delivery to be effective, the barrier properties of the skin must be overcome in such a way that the skin does not become i rreversibly damaged or that local irritation does not limit patient compliance. The rate-l imiting barrier to percutaneous absorption is the stratum corneum, and more specifically, it is the intercellular matrix of the stratum corneum that is responsible for limiting diffu sion. Electron microscopic analysis of the stratum corneum indicates that the intercellular matrix consists of alternating polar and nonpolar regions. These findings are indirectly supported by nu merous skin permeability experiments which show that flux through sk in is positively dependant on the aqueous and lipid solubilities of the permeant. In that respect, the Roberts-Sloan model ( RS ) for flux is particularly useful si nce it allows flux to be predic ted based on molecular weight and solubilities in aqueous and lipid solvents. Prodrug modification has been identified as a useful approach to overcome the skin barrier by transiently improving th e biphasic solubility of the active drug. In this thesis, alkylcarbonyloxymethyl (ACOM) a nd alkyloxycarbonyloxymethyl (AOCOM) promoieties were selected as novel deriva tives for improving the topical delivery of phenol-containing drugs based on their successful use in the oral delivery of a wide range of drugs and (in the case of ACOM) the t opical delivery of heterocyclic drugs.

PAGE 132

119 Acetaminophen (4-hydroxyacetanilide, APAP) was chosen as a model phenol in order to justify further work on more pharmaceutically interesting phenols. The first objective of this work was to synthesize a homologous series of ACOM and AOCOM prodrugs of APAP. In the ACOM series, ACOM iodides were synthesized in good yield via a new one-step route. Subs equent reactions between the ACOM iodides and various phenols gave mainly alkylated phen ol regardless of the steric hindrance in the ACOM iodidea finding that contradicted previous asser tions that the ACOM iodide must be sterically hindered in order to shif t the product distribution in favor of alkylated phenol.91 A slightly different situation was f ound in the AOCOM series. Compared to the ACOM series, steric hindrance (as measur ed by Chartons steric parameters) in the AOCOM iodide was more influential in dete rmining the product distributionespecially when the length of the alkoxy chain was shor t. However, under phase-transfer conditions the influence of steric hindrance was minimize d, reaction time was reduced, and yields of the alkylated product were improved. Although a potentially usef ul reaction for synthesizing ACOM iodides was identified, it is currently of little va lue since its success was dependant upon an unidentified catalyst that was present in older batches of NaI but is absent from newer, purer batches of NaI. A ne w catalyst system involving AlCl3 and I2 was identified but was not optimized due to time constraints. In order to make this reaction available for future work, the new catalyst system must be optimized (i.e. determine optimum molar ratios of AlCl3 and I2). Whether this catalyst system is optimized or not, this does not preclude future work with ACOM derivativ es of phenols since the method of Adams79 is still available for synthesizing the requisite ACOM iodides.

PAGE 133

120 The second objective of this work was to determine whether the ACOM and AOCOM derivatives were capable of impr oving the topical delivery of APAP. The diffusion cell experiments demonstrate that both types of pr odrug are capable of improving the flux of APAP. If comparisons are made between the two series, the ACOM prodrugs are more soluble in wate r and isopropyl myristate (IPM) than the AOCOM prodrugs. As a consequence, the great est improvement in flux (4-fold) was by a member of the ACOM series (4-propionyl oxymethyloxyacetanilide). Three out of the five members of the ACOM se ries exhibited higher fluxes than APAP as compared to only one member from the AOCOM series. Although both types of prodrug delivered mainly APAP through the skin, the ACOM series delivered a somewhat greater percentage of APAP (90-100% of total APAP species in receptor phase as APAP) than the AOCOM series (50-100% of total APAP sp ecies in receptor pha se as APAP). In general, both series delivered more APAP to the skin than topically applied APAP itself. Although the ACOM and AOCOM prodrugs we re capable of improving the topical delivery of APAP, the maximum increase in flux was only 4-fold. Such a modest increase in flux is due to the substantial loss in water solubility that occurs on conversion of APAP to its prodrugs. In order to expe rience further increases in flux, water solubility must be increased without significantly decreasing lipid solubility.20 Simple ACOM and AOCOM derivatives are able to improve the lipid (and often aqueous) solubility of a parent compound by eliminating a hydrogen bon d donor in the parent, thereby lowering the crystal lattice energy. Lipid solubility may be further increased by extending the alkyl chain, but this only decrea ses water solubility. Therefore, in order to increase water solubility by an ACOM or AOCOM approac h, hydrophilic groups must be incorporated

PAGE 134

121 into the acyl chain. Though there are many wa ys to proceed with such a strategy, some of the most successful methods for simultane ously improving water and lipid solubility involve incorporating a basi c amine into the promoiety.19, 20 An interesting series of articles with pa rticular relevance to the present situation was recently published by Rautio and cowork ers on alkylcarbonyloxyalk yl derivatives of naproxen (2-(6-methoxy-2-naphthyl) propionic acid, Figure 5-1).133-135 In the first paper,133 an acetyloxy group was attached by way of an alkyl linkage to the carboxylic acid portion of naproxen. Though none of the al kylcarbonyloxyalkyl derivatives were as soluble in octanol (SOCT) and pH 7.4 buffer (SAQ) as naproxen, the prodrug that exhibited the best biphasic solubility (the acetyloxyethyl ester, Fi gure 5-1) also exhibited the highest flux. However, since there was no im provement in aqueous and lipid solubilities when naproxen was converted to its prodr ugs, the increase in flux was only 1.9-times higher than the flux of naproxen. In the next two articles,134, 135 Rautio et. al. incorporated various amino groups into acyl po rtion of the promoiety in an attempt to improve biphasic solubility. The best resu lts were finally obtained by incorporating methylpiperazine into the acyl chain as s hown in Figure 5-1. With this promoiety, SOCT of the derivative (methylpi perazinylacetyloxyethyl este r) was 120-times higher than naproxen, but SAQ was still only 0.49-times the SAQ of naproxen. Even though the SAQ of the methylpiperazinylacetyl oxyethyl ester was less than the parent, the SAQ of this derivative was still 830-times higher than the SAQ of the best performing member of the previous series of acetyloxyethyl prodrugs (F igure 5-1). As a consequence, the flux of the methylpiperazinylacetyl oxyethyl prodrug was 50-times higher than the flux of naproxen.

PAGE 135

122 Based on the success of Rautio et al. w ith alkylcarbonyloxyalkyl derivatives of naproxen,133-135 similar derivatives may be propos ed for ACOM and AOCOM prodrugs of phenols as shown in Figure 5-1 (APAP is us ed as a model). Such derivatives would likely exhibit higher solubilitie s in water and lipids than th e corresponding simple ACOM and AOCOM prodrugs investigated in the present work. In the AOCOM case, an additional methylene unit would likely be required in the alkyl spacer between the methylpiperazine and carbonyl moieties in order to prevent unin tentional chemical hydrolysis (see Figure 5-1).19 CH3O OH O Naproxen CH3O O O O O N N Naproxen Methylpiperazinylacetyloxyethyl Ester SOCT = 200 mM SAQ = 100 mM JMAQ = 0.23 nmol cm-2 h-1 SOCT = 2.5 x 104 mM SAQ = 50 mM JMAQ = 13 nmol cm-2 h-1 CH3CONH O O N O N APAP Methylpiperzinylacetyloxymethyl Ether CH3CONH O O O O N N APAP Methylpiperzinylethyloxycarbonyloxymethyl Ether. CH3CONH O O O O N N CH3O O O O O Naproxen Acetyloxyethyl Ester SOCT = 190 mM SAQ = 0.06 mM JMAQ = 0.44 nmol cm-2 h-1 APAP Methylpiperzinylmethyloxycarbonyloxymethyl Ether Figure 5-1: Structures of Naproxen, Naproxen Prodrugs,133, 135 Proposed Methylpiperazinyl ACOM and AOCOM Prodrugs of APAP, and Potential Mechanism for Hydrolysis of Meth ylpiperzinylmethyloxycarbonyloxymethyl Ether of APAP

PAGE 136

123 The results from the coupling reactions of ACOM and AOCOM iodides with 2,3,5,7,8-pentamethyl-chroman-6-ol indicate that the correspondi ng reactions with Vitamin E should favor alkylated phenol. Vita min E is so lipophilic that it is unlikely that masking its phenolic OH with a simple ACOM or AOCOM promoiety will increase its water solubility and ultimately its fl ux. On the other hand, an ACOM or AOCOM derivative of Vitamin E should be much more la bile than the acetate and succinate esters of Vitamin E that are currently on the market. Therefore, even if the flux of Vitamin E is not improved by a ACOM/AOCOM derivative, the application of such a soft alkyl approach is justified if the in vivo conversion of the soft alkyl derivative is higher than the currently available derivatives of Vitamin E. Future work should focus on determining the half-lives of ACOM and AOCOM de rivatives of Vitami n E in the skin. The third objective of this work was to improve the accuracy of the Roberts-Sloan ( RS ) equation for predicting flux through hairless mouse skin. This objective was met by incorporating the physicochemical data a nd flux values for the ACOM and AOCOM prodrugs into the prodrug database (n = 61) to obtain a new database of 71 compounds. A fit of the solubility, molecular weight, and flux (JM) values to RS gave the following estimates for x, y, and z: x = -0.562, y = 0.501, z = 0.00248, r2 = 0.923: log JM = -0.562 + 0.501 log SIPM + 0.499 log S4.0 0.00248 MW (1) The previously published RS equation45 based on the n = 61 database is shown below: log JM = -0.491 + 0.520 log SIPM + 0.480 log S4.0 0.00271 MW (2) The average error in ca lculating the fluxes ( log JM) of all 71 com pounds using equation 1 (0.15 0.12 log units) was somewhat less than the log JM associated with using equation 2 to calculate the fluxes of all 71 compounds (0.16 0.12 log units). In other

PAGE 137

124 words, the model is only slightly more accu rate for the database as a whole when equation 1 is used instead of equation 2. On the other hand, w ith the incorporation of the ACOM and AOCOM data into the prodrug data base, the new model (equation 1) should be able to predict the flux of a wi der range of compounds (e.g., nonheterocyclic compounds) with greater accuracy. This is demonstrated by the lower log JM obtained for APAP and its prodrugs (AOC, ACOM, and AOCOM) when equation 1 is used instead of equation 2 (0.19 0.14 log units versus 0.23 0.14 log units). In order to further extend the applicability of RS to a wider range of drugs, more nonheterocyclic compounds need to be added to the database.

PAGE 138

125 LIST OF REFERENCES 1. Waterbeemd, H.; Smith, D.; Beaumont, K.; Walker, D., Property-Based Design: Optimization of Drug Absorp tion and Pharmacokinetics. J. Med. Chem. 2001, 44, (9), 1313-1333. 2. Beaumont, K.; Webster, R.; Gardner, I.; Dack, K., Design of Ester Prodrugs to Enhance Oral Absorption of Poorly Pe rmeable Compounds: Challenges to the Drug Discovery Scientist. Curr. Drug Metab. 2003, 4, 461-485. 3. Schaefer, H.; Redelmeier, T., Skin Barrier: Principles of Percutaneous Absorption S. Karger: Basel, 1996. 4. Swanson, H., Cytochrome P450 Expressi on in Human Keratinocytes: an Aryl Hydrocarbon Receptor Perspective. Chem-Biol. Interact. 2004, 149, 69-79. 5. Williams, A., Transdermal and Topical Drug Delivery Pharmaceutical Press: London, 2003. 6. Pham, M.; Magdalou, J.; Totis, M.; Fourne l-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, (13), 2187-2194. 7. Bashir, S.; Maibach, H., Cutaneous Metabolism of Xenobiotics. In Percutaneous Absorption: Drugs--Cosmetics--Mechanisms--Methodology 4th ed.; Bronaugh, R.; Maibach, H., Eds. Taylor and Francis: Boca Raton, 2005; pp 51-63. 8. Madison, K., Barrier Function of the Skin: "La Raison d'Etre" of the Epidermis. J. Invest. Dermatol. 2003, 121, (2), 231-241. 9. Prausnitz, M.; Mitragotri, S.; Langer, R., Current Status and Future Potential of Transdermal Drug Delivery. Nat. Rev. Drug Discovery 2004, 3, 115-124. 10. O'Sullivan, A.; Crampton, L.; Freund, J.; Ho, K., The Route of Estrogen Replacement Therapy Confers Divergent Effects on Substrate Oxidation and Body Composition in Postmenopausal Women. J. Clin. Invest. 1998, 102, 1035-1040. 11. Vongpatanasin, W.; Tuncel, M.; Wang, Z.; Arbique, D.; Mehrad, B.; Jialal, I., Differential Effects of Oral versus Tr ansdermal Estrogen Replacement Therapy on C-Reactive Protein in Postmenopausal Women. J. Am. Coll. Cardiol. 2003, 41, (8), 1358-1363.

PAGE 139

126 13. Eilertsen, A.; Hoibraaten, E.; Os, I.; Anderson, T.; Sandvik, L.; Sandset, P., The Effects of Oral and Transdermal Ho rmone Replacement Therapy on C-reactive Protein Levels and other Inflammatory Markers in Women with High Risk of Thrombosis. Maturitas 2005, 52, 111-118. 14. Ridker, P.; Hennekens, C.; Buring, J.; Ri fai, N., C-Reactive Protein and other Markers of Inflammation in the Predicti on of Cardiovascular Disease in Women. N. Engl. J. Med. 2000, 342, 836-843. 15. Helton, D.; Osborne, D.; Pierson, S.; Buona rati, M.; Bethem, R., Pharmacokinetic Profiles in Rats After Intravenous, Oral or Dermal Administration of Dapsone. Drug Metab. Dispos. 2000, 28, (8), 925-929. 16. Carter, E.; Elewski, B.; Sharata, H.; Rafa l, E., Dapsone Gel 5 % is an Effective, Safe, and Novel Topical Treatment of Ac ne Vulgaris: First of 2 Controlled, Parallel-Design Studies (DAP0203). J. Am. Acad. Dermatol. 2006, 54, (3), AB26AB26 Suppl. S. 17. Pieters, F.; J., Z., The Absolute Oral Bioavailability of Dapsone in Dogs and Humans. Int. J. Clin. Pharmacol. Ther. Toxicol. 1987, 25, 396-400. 18. Hadgraft, J.; Guy, R., Feasibility Assesmen t in Topical and Transdermal Delivery: Mathematical Models a nd In Vitro Studies. In Transdermal Drug Delivery 2 ed.; Hadgraft, J.; Guy, R., Eds. Marc el Dekker: New York, 2003; pp 1-23. 19. Sloan, K. B., Functional Group Considera tions 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 17116. 20. Sloan, K. B.; Wasdo, S., Designing for Topi cal Delivery: Prodrugs Can Make the Difference. Med. Res. Rev. 2003, 23, 763-793. 21. Skin and the Integumentary System. In Hole's Human Anatomy and Physiology Shier, D.; Butler, J.; Lewis, R., Eds. McGraw-Hill: Boston, 1999; pp 160-183. 22. Menon, G. K., New Insights Into Skin Structure: Scratching the Surface. Adv. Drug Delivery Rev. 2002, 54, (Suppl. 1), S3-S17. 23. Wertz, P.; Downing, D., Glycolipids in Mammalian Epidermis: Structure and Function in the Water Barrier. Science 1982, 217, 1261-1262. 24. Madison, K. C.; Swartzendruber, D. C.; Wert z, P. W.; Downing, D. T., Presence of Intact Intercellular Lipid Lamellae in the Upper Layers of the Stratum Corneum. J. Invest. Dermatol. 1987, 88, (6), 714-718. 25. Potts, R. O.; Francoeur, M., The Influence of Stratum Corneum Morphology on Water Permeability. J. Invest. Dermatol. 1991, 96, (4), 495-499.

PAGE 140

127 26. Hou, S. Y. E.; Mitra, A.; White, S.; Me non, G. K.; Ghadially, R.; Elias, P. M., Membrane Structures in Normal and Essential Fatty Acid-Deficient Stratum Corneum: Characterization by Ruthen ium Tetroxide Staining and X-Ray Diffraction. J. Invest. Dermatol. 1991, 96, (2), 215-223. 27. Martin, R.; Denyer, S.; Hadgraft, J., Skin Metabolism of Topically Applied Compounds. Int. J. Pharm. 1987, 39, 23-32. 28. Kao, J.; Carver, M., Cutaneous Metabolism of Xenobiotics. Drug Metab. Rev. 1990, 22, (4), 363-410. 29. Thiebaut, F.; Tsuruo, T.; Hamada, H.; Go ttesman, M.; Pastan, I.; Willingham, M., Cellular Localization of the Multidrug-Resistance Gene Product P-Glycoprotein in Normal Human Tissues. Proc. Natl. Acad. Sci. U S A 1987, 84, 7735-7738. 30. Baron, J. M.; Holler, D.; Schiffer, R.; Frankenberg, S.; Neis, M.; Merk, H. F.; Jugert, F. K., Expression of multiple cytochrome p450 enzymes and multidrug resistance-associated transport prot eins in human skin keratinocytes. J Invest Dermatol 2001, 116, (4), 541-8. 31. Merk, H. F.; Abel, J.; Baron, J. M.; Krutmann, J., Molecular Pathways in Dermatotoxicology. Toxicol. Appl. Pharmacol. 2004, 195, 267-277. 32. Randolph, G. J.; Beaulieu, S.; Pope, M.; S ugawara, I.; Hoffman, L.; Steinman, R. M.; Muller, W. A., A physiologic functi on for p-glycoprotein (MDR-1) during the migration of dendritic cells from sk in via afferent lymphatic vessels. Proc Natl Acad Sci U S A 1998, 95, (12), 6924-9. 33. Li, Q.; Kato, Y.; Sai, Y.; Imai, T.; Tsuji, A., Multidrug Resistance-Associated Protein 1 Functions as an Efflux Pump of Xenobiotics in the Skin. Pharm. Res. 2005, 22, (6), 842-846. 34. Bronaugh, R.; Maibach, H., Percutaneous Absorption: Drugs--Cosmetics-Mechanisms--Methodology 4th ed.; Taylor and Fr ancis: Boca Raton, 2005. 35. Banga, A.; Bose, S.; Ghosh, T., Iontophores is and Electroporation: Comparisons and Contrasts. Int. J. Pharm. 1999, 179, 1-19. 36. Williams, A.; Barry, B., Penetration Enhancers. Adv. Drug Delivery Rev. 2004, 56, 603-618. 37. Prausnitz, M., Microneedles for Transdermal Drug Delivery. Adv. Drug Delivery Rev. 2004, 56, 581-587. 38. Sloan, K. B.; Wasdo, S., The Role of Pr odrugs in Penetration Enhancement. In Percutaneous Penetration Enhancers 2nd ed.; Smith, E.; Maibach, H., Eds. Taylor and Francis: Boca Raton, 2006; pp 51-64.

PAGE 141

128 39. Waranis, R. P.; Sloan, K. B., Effects of Vehicles and Prodrug Properties on the Delivery of 6-Mercaptopur ine Through Skin: Bisacyl oxymethyl-6-mercaptopurine Prodrugs. J. Pharm. Sci. 1987, 76, 587-595. 40. Flynn, G. L., Physicochemical Determinants of Skin Absorption. In Principles of Route-to-Route Extrapolation for Risk Assessment Gerrity, T.; Henry, C., Eds. Elsevier: New York, 1990; pp 93-127. 41. Kasting, G. B.; Smith, R. L.; Cooper, E. R., Effect of Lipid Solubility and Molecular Size on Percutaneous Absorption. In Skin Pharmokinetics Shroot, B.; Schaefer, H., Eds. Karger: Basel, 1987; Vol. 1, pp 138-153. 42. Potts, R.; Guy, R., Predicting Skin Permeability. Pharm. Res. 1992, 9, (5), 663-669. 43. Roberts, W.; Sloan, K. B., Correlation of Aqueous and Lipid Sol ubilities with Flux for Prodrugs of 5-Fluorouracil, Theophylline and 6-Mercaptopurine: A Potts-Guy Approach. J. Pharm. Sci. 1999, 88, 515-532. 44. van Hal, D. A.; Jeremiasse, E.; Junginger, H. E.; Spies, F.; Bouwstra, J., Structure of Fully Hydrated Human Stratum Corneum: A Freeze-Fracture Electron Microscopy Study. J. Invest. Dermatol. 1996, 106, (1), 89-95. 45. Wasdo, S. C.; Sloan, K. B., Topical Delivery of a Model Phenolic Drug: Alkyloxycarbonyl Prodrugs of Acetaminophen. Pharm. Res. 2004, 21, (6). 46. Majumdar, S.; Thomas, J.; Wasdo, S.; Sloan, K. B., The Effect of Water Solubility of Solutes on their Flux through Huma n Skin: the Extended and Edited Flynn Database. AAPS J. 2004, 6, (S1), W4250. 47. Billich, A.; Vyplel, H.; Gra ssberger, M.; Schmook, F.; Steck, A.; Stuetz, A., Novel Cyclosporin Derivatives Featuring Enhan ced Skin Penetration Despite Increased Molecular Weight. Bioorg. Med. Chem. 2005, 13, 3157-3167. 48. Ettmayer, P.; Amidon, G.; Clement, B.; Test a, B., Lessons Learned from Marketed and Investigational Prodrugs. J. Med. Chem. 2004, 47, (10), 2393-2404. 49. Anderson, R.; Kudlacek, P.; Clemens, D ., Sulfation of Minoxidil by Multiple Human Cytosolic Sulfotransferases. Chem. Biol. Interact. 1998, 109, 53-67. 50. Messenger, A.; Rundegren, J., Minoxidil: Mechanisms of Action on Hair Growth. Br. J. Dermatol. 2004, 150, 186-194. 51. Buhl, A.; Baker, C.; Dietz, A., Minoxidil Su lfotransferase Activity Influences the Efficacy of Rogaine Topical Solution (T S): Enzyme Studies using Scalp and Platelets. J. Invest. Dermatol. 1994, 102, 534. 52. Testa, B., Prodrug Research: Futile or Fertile. Biochem. Pharmacol. 2004, 68, 2097-2106.

PAGE 142

129 53. Moss, G.; Gullick, D.; Cox, P.; Alexander, C.; Ingram, M.; Smart, J.; Pugh, W., Design, Synthesis, and Char acterization of Captopril Prodrugs for Enhanced Percutaneous Absorption. J. Pharm. Phamacol. 2006, 58, 167-177. 54. Milosovich, S.; Hussain, A.; Dittert, L.; Aungst, B.; Hussain, M., Testosteronyl-4dimethylaminobutyrate HCl: A Prodrug with Improved Skin Penetration Rate. J. Pharm. Sci. 1993, 82, (2), 227-228. 55. Smith, M.; March, J., Aliphatic Nucleophilic Substitution. In March's Advanced Organic Chemistry: Reactions Mechanisms, and Structure 5th ed.; Smith, M.; March, J., Eds. John Wiley and Sons: New York, 2001; pp 389-675. 56. Sloan, K. B.; Hashida, M.; Alexander, J.; Bodor, N.; Higuchi, T., Prodrugs of 6Thiopurines: Enhanced Delivery Through the Skin. J. Pharm. Sci. 1983, 72, (4), 372-377. 57. Kerr, D.; Roberts, W.; Tebbett, I.; Sloan, K. B., 7-Alkylcarbonyloxymethyl Prodrugs of Theophylline: Topical Delivery of Theophylline. Int. J. Pharm. 1998, 167, 37-48. 58. Sloan, K. B.; Bodor, N., Hydroxymet hyl and Acyloxymethyl Prodrugs of Theophylline: Enhanced Delivery of Polar Drugs Through Skin. Int. J. Pharm. 1982, 12, 299-313. 59. Ehrnebo, M.; Nilsson, S.; Boreus, L., Ph armacokinetics of Am picillin and its Prodrugs Bacampicillin and Pivampicillin in Man. J. Pharmacokinet. Biopharm. 1979, 7, (5), 429-451. 60. Drustrup, J.; Fullerton, A.; Christrup, L.; B undgaard, H., Utilization of Prodrugs to Enhance the Transdermal Absorption of Morphine. Int. J. Pharm. 1991, 71, 105116. 61. Hansen, L. B.; Fullerton, A.; Christrup, L.; Bundgaard, H., Enhanced Transdermal Delivery of Ketobemido ne with Prodrugs. Int. J. Pharm. 1992, 84, 253-260. 62. Stinchcomb, A. L.; Paliwal, A.; Dua, R. ; Imoto, H.; Woodard, R. W.; Flynn, G. L., Permeation of Buprenorphine and Its 3-Al kyl-Ester Prodrugs through Human Skin. Pharm. Res. 1996, 13, (10), 1519-1523. 63. Sung, K. C.; Fang, J.; Hu, O. Y., Delivery of Nalbuphine and its Prodrugs Across Skin by Passive Diffusion and Iontophoresis. J. Control. Rel. 2000, 67, 1-8. 64. Stinchcomb, A. L.; Swaan, P.; Ekabo, O.; Harris, K.; Browe, J.; Hammell, D.; Cooperman, T.; Pearsall, M., Straight-C hain Naltrexone Ester Prodrugs: Diffusion and Concurrent Esterase Biot ransformation in Human Skin. J. Pharm. Sci. 2002, 91, (12), 2571-2578.

PAGE 143

130 65. Pillai, O.; Hamad, M.; Crooks, P.; Stinchcomb, A. L., Physicochemical Evaluation, in Vitro Human Skin Diffusion, and Concurrent Biotransformati on of 3-O-Alkyl Carbonate Prodrugs of Naltrexone. Pharm Res 2004, 21, (7), 1146-1152. 66. Beall, H. D.; Sloan, K. B., Transderma l Delivery of 5-fluor ouracil (5-FU) through Hairless Mouse Skin by 1Alkylcarbonyl-5-FU Prodrugs. Int. J. Pharm. 1996, 129, 203-210. 67. Taylor, H. E.; Sloan, K. B., 1-Alkylca rbonyloxymethyl Prodrugs of 5-Fluorouracil (5-FU): Synthesis, Physicochemical Properties, and Topical Delivery. J. Pharm. Sci. 1998, 87, (1), 15-20. 68. Beall, H. D.; Sloan, K. B., Topical Delivery of 5-Fluorouracil (5-FU) by 3Alkylcarbonyl-5-FU Prodrugs. Int. J. Pharm. 2001, 217, 127-137. 69. Roberts, W.; Sloan, K. B., Topical De livery of 5-Fluorouracil (5-FU) by 3Alkylcarbonyloxymethyl-5-FU Prodrugs. J. Pharm. Sci. 2003, 92, 1028-1036. 70. Pinnell, S., Cutaneous Photodamage, Oxid ative Stress, and Topical Antioxidant Protection. J. Am. Acad. Dermatol. 2003, 48, (1), 1-19. 71. Fuchs, J.; Kern, H., Modulation of UVLight-Induced Skin Inflammation by alphaTocopherol and L-Ascorbic Acid: A C linical Study Using Solar Simulated Radiation. Free Radical Bio. Med. 1998, 25, (9), 1006-1012. 72. Mireles-Rocha, H.; Galindo, I.; Huerta, M. ; Trujillo-Hernadez, B.; Elizalde, A.; Cortes-Franco, R., UVB Photoprotection w ith Antioxidants: Effects of Oral Therapy with d-alpha-Tocopherol and Asco rbic Acid on the Minimal Erythema Dose. Acta Derm-Venereol. 2002, 82, (1), 21-24. 73. Alade, S.; Brown, R.; Paquet, A., Po lysorbate-80 and E-Ferol Toxicity. Pediatrics 1986, 77, (4), 593-597. 74. Windholz, M., The Merck Index: An Encyclopedia of Chemicals, Drugs, and Biologicals 10th ed.; Merck & CO.: Rahway, 1983. 75. Pedraz, J.; Calvo, B.; Bortolotti, A.; Cela rdo, A.; Bonati, M., Bioavailability of Intramuscular Vitamin-E Acetate in Rabbits. J. Pharm. Phamacol. 1989, 41, (6), 415-417. 76. Jansen, A. B. A.; Russell, T. J., Some Novel Penicillin Derivatives. J. Chem. Soc. 1965 2127-2131. 77. Binderup, E.; Hansen, E. T., Chlorosulfat es as Reagents in the Synthesis of Carboxylic Acid Esters Under Phase-Transfer Conditions. Synth. Commun. 1984, 14, (9), 857-864.

PAGE 144

131 78. Harada, N.; Hongu, M.; Tanaka, T.; Kawaguchi T.; Hashiyama, T.; Tsujihara, K., A Simple Preparation of Chloromethyl Esters of the Blocked Amino Acids. Synth. Commun. 1994, 24, (6), 767-772. 79. Adams, R.; Vollweiler, E. H., The React ion Between Acid Halides and Aldehydes. I. J. Am. Chem. Soc. 1918, 40, 1732-1746. 80. Ulich, L. H.; Adams, R., The Reaction Betw een Acid Halides and Aldehydes. III. J. Am. Chem. Soc. 1921, 43, 660-667. 81. Iyer, R.; Yu, D.; Ho, N.; Agrawal, S., Synt hesis of Iodoalkylacylates and their use in the Preparation of S -Alkyl Phosphorothiolates. Synth. Commun. 1995, 25, (18), 2739-2749. 82. Fleischmann, K.; Adam, F.; Durckheimer, W. ; Hertzsch, W.; Horlein, R.; Jendralla, H.; Lefebvre, C.; Mackiewicz, P.; Roul, J.; Wollmann, T., Synthesis of HR 916 B: The First Technically Feasible Route to the 1-(pivaloyloxy)e thyl Esters of Cephalosporins. Liebigs Ann. 1996, 11, 1735-1741. 83. Taylor, H. E. Bioreversible Derivative s of 5-Fluorouracil: The Synthesis and Evaluation of a Series of 1-Alkylcarb onyloxymethyl-5-fluorouracil Derivatives. University of Florida, Gainesville, 1997. 84. Folkmann, M.; Lund, F., Acyloxymethyl Carbo nochloridates. New Intermediates in Prodrug Synthesis. Synthesis 1990, 12, 1159-1166. 85. Burness, D. M.; Wright, C. J.; Perkins, W. C., Bis(methylsulfonoxymethyl) Ether. J. Org. Chem. 1977, 42, (17), 2910-2913. 86. French, H. E.; Adams, R., The Reaction Be wteen Acid Halides and Aldehydes. II. J. Am. Chem. Soc. 1921, 43, 651-659. 87. Bhar, S.; Ranu, B. C., Zinc-Promoted Select ive Cleavage of Ethers in Presence of Acyl Chloride. J. Org. Chem. 1995, 60, 745-747. 88. Balme, G.; Gore, J., Conversion of A cetals and Ketals to Carbonyl Compounds Promoted by Titanium Tetrachloride. J. Org. Chem. 1983, 48, 3336-3338. 89. Sloan, K. B.; Koch, S., Effect of Nucle ophilicity and Leaving Group Ability on the SN2 Reactions of Amines with (Acy loxy)alkyl alpha-Halides: A Product Distribution Study J. Org. Chem. 1983, 48, 635-640. 90. Sloan, K. B.; Koch, S., Reaction of (Acy loxy)alkyl alpha-Halides with Phenols: Effects of Nucleofugicity and Nucl eophilicity on Product Distribution J. Org. Chem. 1983, 48, (21), 3777-3783.

PAGE 145

132 91. Ouyang, H.; Borchardt, R.; Siahaan, T., St eric Hindrance is a Key Factor in the Coupling Reaction of (Acyloxy)Alkyl-alpha-H alides with Phenols to Make a New Promoiety for Prodrugs. Tet. Lett. 2002, 43, 577-579. 92. Bensel, N.; Reymond, M.; Reymond, J., Pi valase Catalytic Antibodies: Towards Abzymatic Activation of Prodrugs. Chem. Eur. J. 2001, 7, (21), 4604-4612. 93. Bundgaard, H.; Klixbull, U.; Falch, E., Prodrugs as Drug Delivery Systems. 44: OAcyloxymethyl, O-acyl, and N-acyl Salicylam ide Derivatives as Possible Prodrugs for Salicylamide. Int. J. Pharm. 1986, 30, 111-121. 94. Charton, M., Steric Effects. I. Esterifi cation and Acid-Catalyzed Hydrolysis of Esters. J. Am. Chem. Soc. 1975, 97, (6), 1552-1556. 95. Charton, M., Steric Effects.7. A dditional Steric Constants. J. Org. Chem. 1976, 41, (12), 2217-2220. 96. Charton, M., Steric Effects.13. Composition of the Steric Parameter as a Function of Alkyl Branching. J. Org. Chem. 1978, 43, (21), 3995-4001. 97. Obviously, this is a rough estimation. Si nce nitrogen and oxygen are smaller than a methylene unit, the v value obtained by th is method likely overestimates the true steric parameter for 9. 98. Bundgaard, H.; Rasmussen, G., Prodrugs of Peptides. 9. Bioreversible N -alphaHydroxalkylation of the Peptide Bo nd to Effect Protection Against Carboxypeptidase or Other Proteolytic Enzymes. Pharm. Res. 1991, 8, (3), 313322. 99. Wadsworth, D.; Vinal, R., Reactions of Bis(acetoxymethyl) Ether and Several of Its Aryloxy Analogues. J. Org. Chem. 1982, 47, 1623-1626. 100. Although Sloan and Koch recommend usi ng acetone as a solvent, under those conditions the product mixtures were freque ntly contaiminated with a substantial amount of 3-hydroxy,3-methyl,2-butanone form ed from the aldol condensation of acetone in the presence of base. This problem could be circumvented by using acetonitrile instead. The ratio of 7:8 di d not change on going from acetone to acetonitrile. 101. Ramesh, C.; Mahender, G.; Ravindranath, N.; Das, B., A Mild, Highly Selective and Remarkably Easy Procedure for Depr otection of Aromatic Acetates Using Ammonium Acetate as a Neutral Catalyst in Aqueous Medium. Tetrahedron 2003, 59, 1049-1054. 102. Blay, G.; Cardona, M.; Garcia, M.; Pedro, J., A Selective Hydrolysis of Aryl Acetates. Synthesis 1989 438-439.

PAGE 146

133 103. Kunesch, N.; Miet, C.; Poisson, J., Mil d, Rapid, and Selective Deprotection of Acetylated Carbohydrates and Phenols with Guanidine. Tet. Lett. 1987, 28, (31), 3569-3572. 104. Bell, K., Facile Selective Aminolysis of Phenolic Benzoates with 1-Butamine in Benzene. Tet. Lett. 1986, 27, (20), 2263-2264. 105. Chakraborti, A.; Sharma, L.; Sharma, U., A Mild and Chemoselective Method for Deprotection of Acryl Acetates and Be nzoates Under Non-hydrolytic Condition. Tetrahedron 2001, 57, 9343-9346. 106. Datta, A.; Hepperle, M.; Georg, G., Selec tive Deesterification Studies on Taxanes: Simple and Efficient Hydrazinolysis of C-10 and C-13 Ester Functionalities. J. Org. Chem. 1995, 60, 761-763. 107. Roberts, W.; Sloan, K. B., Synthesis of 3-Alkylcarbonyloxymethyl Derivatives of 5-Fluorouracil. J. Heterocyclic Chem. 2002, 39, 905-910. 108. Nagase and Cowokers have used si milar procedure for deprotecting 1Benzyloxycarbonyloxymethyl-5-fluorouracil as described in Chem. Lett., 1988, 1381-1384. 109. Beall, H. D.; Getz, J. J.; Sloan, K. B., Th e Estimation of Relative Water Solubility for Prodrugs that are Unstable in Water. Int. J. Pharm. 1993, 93, 37-47. 110. Fedors, R. F., A Method for Estimating Both the Solubility Parameters and Molar Volumes of Liquids. Polym. Eng. Sci. 1974, 14, (2), 147-154. 111. Martin, A.; Wu, P. L.; Velasquez, T., Extended Hildebrand Solubility Approach: Sulfonamides in Binary and Ternary Solvents. J. Pharm. Sci. 1985, 74, 277-282. 112. Sloan, K. B.; Koch, S.; Siver, K.; Flowers, F., The Use of Solubility Parameters of Drug and Vehicle to Predict Flux. J. Invest. Dermatol. 1986, 87, 244-252. 113. Sloan, K. B.; Beall, H. D.; Weimar, W. R.; Villaneuva, R., The Effect of Receptor Phase Composition on the Permeability of Hairless Mouse Skin in Diffusion Cell Experiments. Int. J. Pharm. 1991, 73, 97-104. 114. Beall, H. D.; Prankerd, R. J.; Sloan, K. B., Transdermal Delivery of 5-Fluorouracil (5-FU) Through Hairless Mouse Skin by 1-Alkyloxycarbonyl-5-FU Prodrugs: Physicochemical Characterization of Prodr ugs and Correlations with Transdermal Delivery. Int. J. Pharm. 1994, 111, 223-233. 115. Sloan, K. B.; Getz, J. J.; Beall, H. D.; Prankerd, R. J., Transdermal Delivery of 5Fluorouracil (5-FU) Through Hairless M ouse Skin by 1-Alkylaminocarbonyl-5-FU Prodrugs: Physicochemical Characterizati on of Prodrugs and Correlations with Transdermal Delivery. Int. J. Pharm. 1993, 93, 27-36.

PAGE 147

134 116. Waranis, R. P.; Sloan, K. B., Effects of Vehicles and Prodrug Properties on the Delivery of 6-Mercaptopurine through Skin: S6-Acyloxymethyl-6-mercaptopurine Prodrugs. J. Pharm. Sci. 1988, 77, 210-215. 117. Bauguess, C. T.; Sadik, F.; Fincher, J. H. ; Hartman, C. W., Hydrolysis of Fatty Acid Esters of Acetaminophen in Buffe red Pancreatic Lipase Systems I. J. Pharm. Sci. 1975, 64, (1), 117-120. 118. Sloan, K. B.; Wasdo, S.; Ezike-Mkparu, U.; Murray, T.; Nickels, D.; Singh, S.; Shanks, T.; Tovar, J.; Ulmer, K.; Waranis, R. P., Topical Delivery of 5Fluorouracil and 6-Mercaptopurine by Their Alkylcarbonyloxymethyl Prodrugs from Water: Vehicle Effects on Design of Prodrugs. Pharm. Res. 2003, 20, (4), 639-645. 119. Wasdo, S. Topical Delivery of a M odel Phenolic Compound: Alkyloxycarbonyl Prodrugs of Acetaminophen. Ph.D. Dissertation, University of Florida, Gainesville, 2006. 120. Flynn, G. L.; Yalkowsky, S. H., Correlation and prediction of ma ss transport across membranes. I. Influence of alkyl chai n length on flux-determining properties of barrier and diffusant. J Pharm Sci 1972, 61, (6), 838-52. 121. Seki, H.; Kawaguchi, T.; Higuchi, T., Speci ficity of Esterases and Structure of Prodrug Esters: Reactivity of Various Acylated Acetami nophen Compounds and Acetylaminobenzoated Compounds. J. Pharm. Sci. 1988, 77, (10), 855-860. 122. Alexander, J.; Fromtling, R.; Bland, J.; Pelak, B.; Gilfillan, E., (Acyloxy)alkyl Carbamate Prodrugs of Norfloxacin. J. Med. Chem. 1991, 34, 78-81. 123. Ichikawa, T.; Kitazaki, T.; Matsushita, Y.; Yamada, M.; Hayashi, R.; Yamaguchi, M.; Kiyota, Y.; Okonogi, K.; Itoh, K., Op tically Active Antifungal Azoles. XII. Synthesis and Antifungal Activity of th e Water-Soluble Prodrugs of 1-[(1 R ,2 R )-2(2,4-Difluorophenyl)-2-hydroxy-1-methyl-3-(1 H -1,2,4-triazol-1-yl)propyl]-3-[4(1 H -1-tetrazoyl)phenyl]-2-imidazolidinone. Chem. Pharm. Bull. 2001, 49, (9), 1102-1109. 124. Hoffmann, H. M. R.; Iranshahi, L., Synt hesis and CuCN-Promoted Cyanation of Iodoformic Esters. J. Org. Chem. 1984, 49, 1174-1176. 125. Senet, J.; Sennyey, G.; Wooden, G., A Convenient New Route to 1-Haloalkyl Carbonates. Synthesis 1988 (5), 407-410. 126. Dittert, L.; Caldwell, H.; Adams, H. ; Irwin, G.; Swintosky, Acetaminophen Prodrugs I: Synthesis, Physicochemical Properties, and Analgesic Activity. J. Pharm. Sci. 1968, 57, (5), 774-780. 127. Charton, M., Steric Effects.9. Substitu ents at Oxygen in Carbonyl Compounds. J. Org. Chem. 1977, 42, (22), 3531-3535.

PAGE 148

135 128. Wolff, S.; Hoffmann, H., Aflatoxins Revi sited:Convergent Synthesis of the ABCMoiety. Synthesis 1988, 10, 760-763. 129. Merck, E., Zur Kenntnis der Einwirkung von Phosgen bxw Chlorkohlensaure Ester auf p-Aacetylaminophenole und p-Oxyphenylurethane. Chem. Zentralbl. 1897, I, 468-469. 130. Milstien, J. B.; Fife, T. H., Steric Effect s in the Acylation of alpha-Chymotrypsin. Biochemistry 1969, 8, (2), 623-627. 131. Kasting, G. B.; Smith, R. L.; Anderson, B. D., Prodrugs for Dermal Delivery: Solubility, Molecular Size, and Functional Group Effects. In Prodrugs: Topical and Ocular Drug Delivery Sloan, K. B., Ed. Marcel Dekker: New York, 1992; pp 142-158. 132. Roberts, W. J.; Sloan, K. B., Application of the Tran sformed Potts-Guy Equation to In vivo Human Skin Data. J. Pharm. Sci. 2001, 90, (9), 1318-1323. 133. Rautio, J.; Taipale, H.; Gynther, J.; Vepsal ainen, J.; Nevalainen, T.; Jarvinen, T., In Vitro Evaluation of Acyloxyalkyl Esters as Dermal Prodrugs of Ketoprofen and Naproxen. J. Pharm. Sci. 1998, 87, (12), 1622-1628. 134. Rautio, J.; Nevalainen, T.; Taipale, H.; Vepsalainen, J.; Gynther, J.; Pedersen, T.; Jarvinen, T., Synthesis and In Vitro Eval uation of Aminoacyloxyalkyl Esters of 2(6-methoxy-2-naphthyl)propionic Acid as Novel Naproxen Prodrugs for Dermal Drug Delivery. Pharm Res 1999, 16, (8), 1172-1178. 135. Rautio, J.; Nevalainen, T.; Taipale, H.; Vepsalainen, J.; Gynther, J.; Laine, K.; Jarvinen, T., Synthesis and In Vitro Evaluation of Novel Morpholinyland Methylpiperazinylacyloxyalkyl Prodrugs of 2-(6-Methoxy-2-naphthyl)propionic Acid (Naproxen) for Topical Drug Delivery J. Med. Chem. 2000, 43, 1489-1494.

PAGE 149

136 BIOGRAPHICAL SKETCH Joshua D. Thomas was born in Peachland, North Carolina, on January 5, 1978, where he lived until graduating from Anso n County High School in June, 1996. In the fall of that year, he enrolled in Wingate Univ ersity where he met his wife Amber. After graduating from Wingate University in Ma y 2001, he and Amber married. Later that year they moved to Gainesville, where Joshua began his studies in the graduate program in medicinal chemistry at the University of Florida. He and Amber are the parents of Miriam Faith Thomas, born July 19, 2005.


Permanent Link: http://ufdc.ufl.edu/UFE0015582/00001

Material Information

Title: Improving the Topical Delivery of Phenol-Containing Drugs: An Alkylcarbonyloxymethyl and Alkyloxycarbonyloxymethyl Prodrug Approach
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0015582:00001

Permanent Link: http://ufdc.ufl.edu/UFE0015582/00001

Material Information

Title: Improving the Topical Delivery of Phenol-Containing Drugs: An Alkylcarbonyloxymethyl and Alkyloxycarbonyloxymethyl Prodrug Approach
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0015582:00001


This item has the following downloads:


Full Text











IMPROVING THE TOPICAL DELIVERY OF PHENOL-CONTAINING DRUGS: AN
ALKYLCARB ONYLOXYMETHYL AND ALKYLOXYCARB ONYLOXYMETHYL
PRODRUG APPROACH














By

JOSHUA DENVER THOMAS


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


2006



























Copyright 2006

by

Joshua Denver Thomas





























This document is dedicated to my wife Amber, my daughter Miriam, and to my parents
Richard and Delores.
















ACKNOWLEDGMENTS

It is clear to me that every accomplishment in my life has been fueled by the love

of my family and friends and by the wisdom and knowledge of my advisors. It is with

this realization that I would like to thank my wife Amber for her unwavering love,

support, and encouragement (especially during my first and last semesters of graduate

school); and my daughter Miriam, whose smile is sometimes all I need. I would also like

to thank my parents, Richard and Delores, who have taught me that there is no greater

purpose in life than to know my Creator. I would be remiss if I did not also thank

Christopher E. Dahm, James M. Gibson, and James W. Hall for the advice and early

research opportunities they provided; my committee members Margaret O. James and

William R. Dolbier for their help at critical junctures in my graduate career; and

Raymond Booth for graciously accepting a position on my committee. Finally, I will

always be indebted to Kenneth B. Sloan for his direction and immense patience and for

giving me the opportunity to conduct graduate research. I am grateful to know him as my

mentor.

Above all, I would like to thank my Savior Jesus Christ for his unconditional love

and for the peace that comes from knowing that my life is in his hands.





















TABLE OF CONTENTS


page

ACKNOWLEDGMENT S .............. .................... iv


LIST OF TABLES .........._.... ..............vii...__........


LIST OF FIGURES .............. .................... ix


AB STRAC T ................ .............. xii


CHAPTER


1 BACKGROUND ................. ...............1.......... ......


Topical Delivery ................. ...............1.......... ......
Rationale ................ .. .. .......... ...............1.......
Anatomy and Physiology of Skin ................. ...............3............ ...
Hypodermis .............. ...............4.....
D erm is .............. ...............5.....

Epidermis .............. ..... ...............7..
Barrier Properties of the Skin .....__.......____...... ...............12..
Physicochemical barrier ................. ...............12.................
Biochemical barrier ................. ......... ...............14.......
Overcoming the Skin Barrier ................. ...............15................
Strategies ............... .. ......... .. ... ....... .............1
Predictive models for optimizing topical delivery .............. ....................16
Prodrugs ................. ...............21.................
Acyl Prodrugs ................. ...............23.................
Soft Alkyl Prodrugs ................. ...............26........... ....
Conclusions............... ..............2


2 SPECIFIC OBJECTIVES ........._.._ ........... ...............31....


First Obj ective ........._... ......___ ...............31....
Second Obj ective ........._.. ..... ._ ...............3 3....
Third Obj ective ........._... ...... ._ ._ ...............34...


3 ALKYLCARBONYLOXYMETHYL PRODRUGS OF ACETAMINOPHEN

(APAP) .............. ...............36....












Synthesi s of Alkylcarbonyloxymethyl (ACOM) lodides ................. ............... .....3 6
Coupling Reaction of ACOM Iodides with 4-Hydroxyacetanilide ........._................41
Conclusions .............. ...............51....

Experim ental .............. .. ......_._... .... .. ... .. ......... .......5
In Vitro Determination of Flux of ACOM Prodrugs of APAP .............. ..............58
Materials and Methods .............. ...... ...............59

Physicochemical properties and analysis ................... ...............5
Diffusion cell experiments .............. ...............62....
Results and Discussion ................. ...............65......._.. ....
Phy sicochemical properties............... ...............6
Diffusion cell experiments .............. ...............69....
Conclusions .............. ...............79....


4 ALKYLOXYCARB ONYLOXYMETHYL (AOCOM) PRODRUGS OF
ACET AMINOPHEN (APAP) ................. ......... ...............80.......


Synthesis of AOCOM Prodrugs of 4-Hydroxyacetanlide (APAP) ...........................80
Conclusions .............. ...............87....

Experim ental ............... .. .......... ..............._ .. .. ..........8
In Vitro Determination of Flux of AOCOM APAP Prodrugs .............. ................94
Methods and Materials .............. ...... ...............94
Physicochemical properties and analysis ................... ...............9
Diffusion cell experiments .............. ...............98....
Results and Discussion ................. ................. 100.....__.....

Phy sicochemical properties............... ..............10
Diffusion cell experiments .............. ...............105....
Conclusions ................. ...............117......... ......


5 CONCLUSIONS AND FUTURE WORK .......__................. ........._.._. ......11


LIST OF REFERENCES .......__................. ........_.. .........12


BIOGRAPHICAL SKETCH ..........._.__........... .........._..........13

















LIST OF TABLES


Table pg

3-1 Variation in Reaction Conditions, Crude Yielda of 3, 4, and 5, and Percentage of
1 Remaining at the End of the End of the Experimentb ..........._.. .........__......39

3-2 Product Distribution of the Reactiona of ACOM Halides 3 with Phenols 6: Data
Taken from the Literature .............. ...............43....

3-3 Product Distribution of the Reactiona of ACOM Halides 3 with Phenols 6: Data
from the Present W ork .............. ...............45....

3-4 Molar Absorptivities (E) of APAP 6a and Prodrugs 7a-e ..........._... ..............60

3-5 Physicochemical Properties of 4-Hydroxyacetanilide 6a, 4-ACOM-APAP
Prodrugs 7a-e and 4-AOC-APAPa Prodrugs 8i-m ......___ ........__ ..............67

3-6 Log Solubility Ratios (log SRIPM:AQ), Differences Between Log SRIPM:AQ (ESR),
Log Partition Coefficients (log KIPM:4.0), Differences Between Log KIPM:4.0 RnK),
and Solubility Parameters (6i) for Prodrugs 7a-e .............. ....................6

3-7 Flux of Total APAP Species through Hairless Mouse Skin from Suspensions of
4-ACOM-APAP and 4-AOC-APAPa Prodrugs in IPM (Jhl), Second Application
Flux of Theophylline through Hairless Mouse Skin from a............... ..................72

3-8 Percent Intact Prodrug Detected in Receptor Phase during Steady-State (%
Intact), Log Permeability Coefficients (log Phl), Concentrations of APAP
Species in Skin (Cs), and Dermal/Transdermal Delivery Ratios for APAP 6a, ......73

4-1 Product Distribution of the Reaction of RCO2CH2X 3 with Phenols 6 Under
Various Reaction Conditions .............. ...............85....

4-2 Molar Absorptivities (E) of APAP 6a and Prodrugs 7i-m............. ............_ ...96

4-3 Physicochemical Properties of 4-Hydroxyacetanilide 6a, 4-ACOM-APAP
Prodrugs 7a-e,a 4-AOC-APAP Prodrugs 8i-m,b and 4-AOCOM APAP Prodrugs
7i-m .............. ...............101....

4-4 Log Solubility Ratios (log SRIPM:AQ), Differences between Log SRIPM:AQ (ESR),
Log Partition Coefficients (log KIPM:4.0), Differences between Log KIPM:4.0 (RK),
and Solubility Parameters (6i) for Prodrugs 7i-m .............. .....................0










4-5 Flux of Total APAP Species through Hairless Mouse Skin from Suspensions of
4-ACOM-APAP,a 4-AOC-APAP,b and 4-AOCOM-APAP Prodrugs in IPM (log
Jhl), Second Application Flux of Theophylline through .............. ............... ...107

4-6 Percent Intact Prodrug Detected in Receptor Phase during Steady-State (%
Intact), Log Permeability Coefficients (log Phl), Concentrations of APAP
Species in Skin (Cs), and Dermal/Transdermal Delivery Ratios for .....................109

















LIST OF FIGURES


Figure pg

1-1 Structure of Acylglucosylceramide and General Orientation in Lamellar Bodies...10

1-2 Structure of Ceramides found in Human Stratum Corneum ................ ................12

1-3 Tortuous Path of Permeant Through the Stratum Corneum and Expanded View
of Alternating Nonpolar (White Bands, Electron Lucent) and Polar (Dark Bands,
Electron Dense) Phases Found Within the Intercellular Matrix .............. ................13

1-4 Bioconversion of Minoxidil to Minoxidil Sulfate by Scalp Sulfotransferase in
the Presence of 3 '-Phosphoadenosine-5 '-phosphosulfate (PAPS) ................... ........23

1-5 Structures of Acyl Prodrugs for the Topical Delivery of Captopril Testosterone,
and Acetaminophen ........... ..... ._ ...............25....

1-6 Most Common Mechanisms by which Acyl Prodrugs are Hydrolyzed
Chem ically .............. ...............26....

1-7 Mechanism of Hydrolysis of Soft Alkyl Prodrugs (Alkylcarbonyloxymethyl and
Hydroxymethyl Derivatives are shown) and Comparison to Metabolism of
"Hard Alkyl" Derivatives (General Mechanism .............. ...............28....

1-8 Examples of Alkylcarbonyloxymethyl (ACOM) and
Alkyloxycarbonyloxymethyl (AOCOM) Prodrugs ....._____ ......... .............29

2-1 Phenol-Containing Therapeutic Agents that may benefit from Topical Delivery
vi a Alkyl carb onyloxymethyl (ACOM) or Alkyl oxycarb onyl oxymethyl
(AOCOM) Derivatization .............. ...............34....

3-1 Reaction of Trioxane la and Paraldehyde lb with Acid Chlorides in the
Presence of Nal .............. ...............37....

3-2 General Reaction of Alkylcarbonyloxymethyl (ACOM) Halide 3 with Phenol 6
to Give Aryl Acylal 7 and Aryl Ester 8............... ...............42...

3-3 Structures of ACOM Derivative of a Protected Amino Acid 9 (R"' = Protecting
Group) and its Corresponding Aliphatic Derivative 10, and Structure of
By product 11 ........._..._.._ ................ s43.._..._....










3-4 Reaction of ACOM Iodides 3a-f with Phenols 6a-c .......____ ..... ...._............44

3-5 Plot of the Percentage of 4 (RCO2CH2C1) in Crude 3 Versus the Ratio of 8/7
(Acylated/Alkylated phenol) Resulting from the Reactions of 3a-3e with 6a and
6b (Taken from Entry 4, Table 3-2 and Entries 1-4, and 8, Table 3-3 m ...............46

3-6 Plot of Charton' s steric parameter v for R' Versus the Ratio of 8/7
(Acylated/Alkylated Product) Resulting from the Reactions of 3a-3e with 6a and
6b (Taken from Table 3-2: Entry 4, Table 3-3: Entries 1-4, and 8 and Entry 5 ...48

3-7 Speculative Mechanism for Reactions of Protected Amino Acid Derivatives 9
with Phenols 6 .............. ...............49....

3-8. Structure of 4-Hydroxyacetanilide and Corresponding 4-ACOM Prodrugs ..............58

3-9. Diagram of Franz Diffusion Cell (Metal Clamp Not Shown) ........._._... ..............63

3-10 Flux of Compound 7a through Hairless Mouse Skin ................. ............ .........65

3-11 Structure of 4-alkyloxycarbonyl (AOC) derivatives of APAP ................ ...............68

3-12 Plot of Solubility Parameter versus Log P for 4-ACOM-APAP Prodrugs 7a-e....74

3-13 Log SIPhi (O), Log S4.0 (a), Log KIPM:4.0 (0), and Log Jhl (*) Values for APAP
6a, 4-ACOM-APAP Prodrugs 7a-e, and 4-AOC-APAP Prodrugs 8i-m. ...............76

3-14 Plot of Experimental Versus Calculated Flux for 5-FU, 6-MP, and Th Prodrugs
(0, n = 53), APAP (m), 4-AOC-APAP Prodrugs (*, n = 5, plus two additional
compounds mentioned in Reference 1 to give n = 7)............... .... ............... 7

4-1 Synthetic Routes to Alkyloxycarbonyloxymethyl (AOCOM, R = Oalkyl)
Prodrugs of 4-hydroxyacetanilide (APAP) ................ ............. .................81

4-2 Generalized Reaction of AOCOM halides (R = Oalkyl) and ACOM halides (R =
alkyl) 3 with phenols 6 .............. ...............82....

4-3 Reaction of AOCOM iodides with phenols under phase-transfer conditions ..........83

4-4 Plot of Charton' s Steric Parameter v for R Versus the Ratio of
Acylated/Alkylated Product (8/7) Resulting from the Reactions of 6 with
AOCOM Iodides (Entries 3-5 in Table 4-1, o) and ACOM Iodides (Entries 14.....87

4-5 Plot of Charton' s Steric Parameter v for R Versus the Ratio of
Acylated/Alkylated Product (8/7) Resulting from the Reactions of 6 with
AOCOM Iodides (Entries 6-11 in Table 4-1, o) Under Phase-Transfer. ...............87

4-6 Structure of 4-Hydroxyacetanilide (APAP) and Corresponding 4-AOCOM-
APAP Prodrugs .............. ...............94....










4-7 Flux of Compound 7j through Hairless Mouse Skin .............. .....................100

4-8 Structures of Alkylcarbonyloxymethyl (ACOM) and Alkyloxycarbonyl (AOC)
Derivatives of APAP and Comparisons between Homologs of Approximately
Equal Size ................. ...............104................

4-9 Plot of Solubility Parameters versus Log Phl for 4-AOCOM-APAP Prodrugs 7i-
m ............... ...............110...

4-10 Log SIPM (O), Log S4.0 (a), Log KIPM:4.0 (0), and Log Jhl (*) Values for APAP
6a, 4-ACOM-APAP Prodrugs 7a-e, 4-AOC-APAP Prodrugs 8i-m, and 4-
AOCOM-APAP Prodrugs 7i-m. ........... ..... ._ .....__.............1

4-11 Plot of Experimental Versus Calculated Flux for 5-FU, 6-MP, and Th Prodrugs
(0, n = 53), APAP (m), 4-AOC-APAP Prodrugs (*, n = 5, plus two additional
compounds mentioned in Reference 1 to give n = 7), 4-ACOM- ................... .......113

4-12 Plot of Experimental Versus Calculated Flux for 5-FU, 6-MP, and Th Prodrugs
(0, n = 53), APAP (m), 4-AOC-APAP Prodrugs (*, n = 5, plus two additional
compounds mentioned in Reference 1 to give n = 7), 4-ACOM- ................... .......114

4-13 Plot of Experimental Versus Calculated Flux for 5-FU, 6-MP, and Th Prodrugs
(0, n = 53), APAP (m), 4-AOC-APAP Prodrugs (*, n = 5, plus two additional
compounds mentioned in Reference 1 to give n = 7), 4-ACOM- ................... .......116

5-1 Structures of Naproxen, Naproxen Prodrugs,133, 135 Proposed Methylpiperazinyl
ACOM and AOCOM Prodrugs of APAP, and Potential Mechanism for
Hydroly si s of Methylpiperzinylmethyl oxycarb onyloxymethyl ................... ..........1 22
















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

IMPROVINTG THE TOPICAL DELIVERY OF PHENOL-CONTAINING DRUGS: AN
ALKYLCARB ONYLOXYMETHYL AND ALKYLOXYCARB ONYLOXYMETHYL
PRODRUG APPROACH

By

Joshua D. Thomas

August 2006

Chair: Kenneth B. Sloan
Major Department: Medicinal Chemistry

Although most drugs are administered orally, this route is not suitable for many

compounds due to their extensive metabolism in the GI tract and liver. Topical delivery

is an alternative route of administration for such drugs that avoids this "first-pass effect"

and permits the drug to enter the systemic circulation following penetration of the skin--

a much less metabolically active tissue than the liver. One of the most effective methods

for improving topical delivery while minimizing side effects involves the use of

prodrugs.

Most previous attempts to improve the topical delivery of phenols via a prodrug

have involved some type of aryl ester, carbonate or carbamate. In the present work,

alkylcarbonyloxymethyl (ACOM) and alkyloxycarbonyloxymethyl (AOCOM) prodrugs

of 4-hydroxyacetanilide (acetaminophen) have been evaluated in vitro as novel

permeation-enhancing derivatives of phenol-containing drugs. Alkyl carb onyl oxymethyl

iodides were synthesized by way of a new one-step route and were subsequently reacted









with various phenols to obtain the target ACOM derivatives. The coupling reaction

between ACOM iodides and phenols was shown to favor the alkylated product regardless

of the steric hindrance in the alkylating agent or the phenol. On the other hand, the

coupling reaction of AOCOM iodides with phenols seemed to be more sensitive to steric

effects, with the acylated product being favored when steric effects were minimal.

However, under phase-transfer conditions, the influence of steric hindrance was

minimized and yields of AOCOM phenol were increased.

More importantly, the ACOM and AOCOM prodrugs were able to improve the

topical delivery of APAP up to 3.6 and 1.3-fold, respectively. The ACOM and AOCOM

prodrugs were also added to the Roberts-Sloan database (n = 61) to obtain a new database

of 71 compounds. A fit of this new database (n = 71 r2 = 0.92) to the Roberts-Sloan (RS)

equation resulted in a more robust model for predicting flux (Jhl) through hairless mouse

skin: log Jhl = -0.562 + 0.501 log SIPM + 0.499 log S4.0 0.00248 MW where SIPM and

S4.0 are the solubilities in isopropyl myristate and pH 4.0 buffer, and MW is molecular

weight.















CHAPTER 1
BACKGROUND

Topical Delivery

Rationale

Although there are many available routes of drug administration, the oral route is

by far the most popular. This is primarily due to a high incidence of patient compliance.

While it is true that patients often find an oral drug regimen more palatable than the

parenteral alternative (e.g., intravenous or intramuscular injection), oral drug absorption

is a much more complicated problem, 2 for the drug discovery scientist to solve. If given

orally, a drug molecule must surmount numerous chemical and enzymatic hurdles in

order to reach the systemic circulation. For example, if the drug survives the acidic

environment of the stomach, it still faces efflux transporters and various

biotransformation enzymes in the gut wall. Following absorption in the gut, the drug

enters the liver, where a host of biotransformation enzymes await. At each stage of

absorption, there is the potential for the drug to be inactivated and excreted, thereby

reducing the amount of the original dose that reaches the intended site of action in the

body.

Given the extent to which a drug can be inactivated as it is absorbed into the

systemic circulation, alternative methods that avoid first-pass metabolism yet retain the

simplicity needed to achieve high patient compliance are desirable. Topical delivery is

one such approach. In general, the levels of drug-metabolizing enzymes in the skin are

much lower than those in the liver and intestine.3-6 For example, transferase activity in









the skin (e.g., glucuronidation and sulfation) may approach 10% of the liver while

cytochrome P-450 activity in the skin is typically 1-5% of the corresponding hepatic

activity.' In fact, skin permeability rather than drug metabolism appears to be the maj or

barrier to topical bioavailability.8, 9

Although it is an important consideration in drug delivery, minimal drug

inactivation is not the only advantage to be gained from avoidance of first-pass

metabolism. Potential side effects must also be taken into account. Topically applied

drugs frequently exhibit fewer side effects than the corresponding oral dosage forms.

One of the most studied medications in that respect is estrogen. Several recent studies

have indicated that the detrimental effects of hormone replacement therapy in

postmenopausal women may be due to the route of drug administration. 10-13 In a

comparison between oral and transdermal estrogen therapies, both treatments were

equally effective at increasing bone mineral density and decreasing luteinizing hormone

levels.10 However, patients treated with oral estrogen for six months experienced an

increase in triglyceride levels and fat mass with an accompanying decrease in lean body

mass. Triglyceride levels and body composition of patients treated with transdermal

estrogen did not significantly change over the course of the six month treatment.10

Other studies indicate that oral estrogen may play a role in the elevated levels of C-

reactive protein (CRP)11, 13 and serum amyloid A (SAA)12 detected in women undergoing

hormone-replacement therapy. These studies found no such side effects in patients

undergoing transdermal estrogen therapy. In both cases, the evidence suggests that the

differences in side effects between the routes of administration are directly related to the

action of oral estrogen in the liver.11-13 Since both CRP and SAA have been identified as









important indicators of systemic inflammation and are predictive of future cardiovascular

disease,14 transdermal estrogen replacement therapy appears to offer a better safety

profile than the more common oral route. In fact, in the case of SAA, transdermal

estrogen may exert a protective effect compared to the oral route. Abbas and coworkersl2

found that the levels of SAA and the SAA-HDL complex (HDL-SAA) in postmenopausal

women receiving transdermal estrogen were substantially lower than those in women

receiving oral estrogen.

While the examples given above for estrogen support the case for transdermal

delivery (to the systemic circulation), it is perhaps more obvious that topical delivery is

an important route for treating skin diseases dermall delivery). The main advantage of

topical over oral administration for the treatment of skin diseases is that high levels of the

drug can be delivered to the skin with minimal exposure to the rest of the body. One

example of the benefits of topical delivery for the treatment of a skin condition is the

topical application of dapsone (4,4' -sulfonyldianiline) .15, 16 Although dapsone is

normally given orally for the treatment of leprosy," oral dapsone has also proven

effective in treating moderate cases of acne.l5 However, the effectiveness of orally

administered dapsone is limited due to its hemotoxic side effects. In a recent study,

topically applied dapsone was successfully used to treat moderate acne with side effects

no different than those of the vehicle (a gel) itself.16

Anatomy and Physiology of Skin

Although topical delivery presents fewer complications than the oral route, this

does not mean that overcoming the barrier properties of the skin is a small task. Unlike

the gastrointestinal tract, the primary purpose of skin is to restrict the passage of

endogenous and exogenous substances into and out of the body. As a consequence,









topical delivery is a viable option for a relatively small percentage of drugs. For

example, all the drugs currently approved for use by the FDA as transdermals have

molecular weights less than 400 Da, exhibit relatively high lipid solubility, and are

therapeutically effective at low doses (0.04-10 ng/ml).3, 9 Furthermore, since most

transdermal drug candidates were originally designed for oral administration,'" they

typically do not possess the particular physiochemical properties required for adequate

diffusion through skin.19 Although the relationship between flux and the

physicochemical properties of the permeant is still a matter of debate,20 a knowledge of

skin anatomy and physiology is helpful in understanding why some compounds permeate

the skin better than others.

The skin is composed of three main layers of varying thickness: the hypodermis (1-

2 mm), dermis (1-5 mm), and epidermis (60-120 Clm).3, 5 The actual composition of each

layer varies with age, disease state, and anatomical location. Though one might expect

the thickest of these layers to be the primary barrier to percutaneous absorption, this is

not the case. In fact, the most impervious layer of the skin is actually the thinnest--the

outermost layer of the epidermis which is referred to as the stratum corneum (10-20 Clm).

Although diffusion through the stratum corneum is generally recognized as the rate-

limiting step to percutaneous absorption, disruptions in the integrity of the other layers

can also affect skin permeability. Thus, the structure and function of each layer will be

reviewed in the following sections.

Hypodermis

The deepest layer of the skin, the hypodermis, is primarily composed of adipose

tissue. As such, it functions as an energy depot, a layer of insulation, and as a shock

absorber. As with the other layers of the skin, the thickness of the hypodermis varies









from one part of the body to another. For instance, the eyelids are altogether missing a

hypodermal layer. Variations in diet can affect the thickness of this layer as well.

The hypodermis serves as the entry point for the maj or blood vessels and nerves

that service the skin. Although adipose tissue may sometimes function as a depot for

highly lipophilic xenobiotics, this is generally not the case with the hypodermis.

Compounds that reach this layer by diffusion are usually taken up by the network of

blood vessels that run throughout the subcutaneous fat. Because the loose connective

tissue of the hypodermis is interwoven with that of the dermis, there is no distinct

boundary between these two layers. In addition, although most hair follicles originate in

the dermis, course hair can often extend deep (3 mm) within the hypodermis.3, 5, 21

Dermis

Directly above the hypodermis is the dermis--the thickest layer of the skin. In

sharp contrast to the underlying layer of adipose tissue, the dermis is a much more

aqueous-like environment. For instance, the gelatinous substance in which the various

structures of the dermis are imbedded consists of proteoglycans and

glycosaminoglycans-compounds that are capable of binding up to 1000 times their

weight in water. Running throughout this gel-like "ground substance" is a dense,

irregular network of collagen fibers. These fibers make up the bulk of the dermal

connective tissue and act as a supporting framework for blood vessels, hair follicles and

various other structures. Microfibrils composed of elastin, fibrillin, and vitronectin make

up the elastic connective tissue (the second most abundant tissue in the dermis), and

provide a certain amount of elasticity to the skin.3, 5

Most of the appendages of the skin originate in the dermis. These include the hair

follicles, sebaceous glands, and sweat glands. As with other features of the skin, the










density and presence of these structures vary with anatomical location. For example, of

these three appendages, only the sweat glands are found in the palms and soles. Hair

follicles are sheath-like structures that enclose each hair and extend from the surface of

the skin into the dermis. Although the follicle consists of living epidermal cells, the hair

shaft inside the follicle is mainly composed of dead, keratinized cells. Attached to the

follicle is a band of smooth muscle fibers that are collectively known as an arrector pili

muscle. Under conditions of emotional stress or cold temperatures, these muscles

contract, causing the hair to stand erect and the skin to take on the familiar "goose bump"

appearance. In most regions of the skin, sebaceous glands merge with hair follicles and

secrete their contents sebumm) directly into the follicle. However, in various sites

throughout the body the sebaceous glands extend to the outermost layers of the skin and

deposit their contents directly at the surface. In a similar fashion, sweat glands either

connect to the hair follicle (as in apocrine glands) or open up at the skin surface (as in

eccrine glands). Sebum (a mixture of fatty acids, triglycerides, and wax secreted by the

sebaceous glands) and sweat (a mixture of salts and various waste products (e.g., urea

and uric acid)) help keep the surface of the skin slightly acidic (pH 5). With regard to

topical delivery, skin appendages may offer an alternative pathway to permeating

compounds that avoids the stratum corneum. However, since the appendages make up a

such a small percentage of the total surface area of the skin (approximately 0. 1%), these

"shunt routes" are not expected to significantly affect the observed flux of most

permeants .35,2

The dermal-epidermal border resembles a transverse wave running parallel to the

skin surface. As a result of these undulations (referred to as dermal papillae), sections of










the dermis come within 200 Clm of the skin surface. Capillaries also extend into the

dermal papillae and help maintain "sink" conditions within the skin by efficiently

transporting permeated compounds to the systemic circulation. In addition, the vascular

network of the dermis is responsible for supplying nutrients and oxygen to the skin and

also plays a role in regulating body temperature. A system of lymphatic vessels

comprises an additional dermal circulatory system. These vessels are involved in

removing cellular waste and help regulate the volume of the interstitial fluid in the

dermis. During times of wound healing and inflammation, the lymphatic system also

delivers macrophages, lymphocytes, and leucocytes to the affected areas of the dermis.

These cells facilitate the healing process by destroying invading bacteria via phagocytosis

or via the secretion of certain cytotoxic agents. In general, the lymphatic circulatory

system plays only a minor role in the clearance of permeated compounds from the

dermis.3, 5

Epidermis

Directly above the dermis lies the epidermis. The epidermis is composed of four

distinct regions, each representing a different phase of kertinocyte differentiation. From

the dermal-epidermal border to the skin surface they are the stratum basale, stratum

spinosum, stratum granulosum, and stratum corneum. Since there are no blood vessels in

this layer of the skin, nutrients reach epidermal cells by way of passive diffusion across

the basement membrane at the dermal-epidermal border. The passage of nutrients and

other materials across the basement membrane is facilitated by the relatively high surface

area provided by the dermal papillae. The final stage of keratinocyte differentiation is

represented by the stratum corneum-the outermost layer of skin. Although it is









essentially dead tissue, the stratum corneum is the rate-limiting barrier to percutaneous

absorption.3, 5, 21

Stratum basale. Keratinocytes of the stratum basale are unique in that they are the

only epidermal cells that undergo mitosis. Following mitotic division, one cell remains

in the stratum basale while the other daughter cell detaches from the basement membrane

and migrates outward through the remaining epidermal layers. Basal keratinocytes are

attached to the basement membrane by structures known as hemidesmosomes. Similar

desmosome plaques are found throughout the epidermis and function as proteinaceous

rivets linking adjacent cells. Other cell types found in the stratum basale include

melaninocytes, Langerhans cells, and Merkel cells. Melaninocytes are responsible for

producing the pigment melanin. Though melanin is produced by the melaninocytes, it is

also transferred to neighboring cells through dendritic connections. Langerhans cells

play an important role in the immune response by binding to foreign antigens in the

epidermis and presenting them to T-lymphocytes in the lymph nodes. Merkel cells are

involved in sensory reception and are found at sites along the basement membrane where

dermal nerve endings extend into the papillae.5

Stratum spinosum. Upon migration from the stratum basale to the stratum

spinosum, keratinocytes undergo several morphological changes including the formation

of desmosomal plaques between adj acent cells. These intercellular linkages make

substantial contributions to the overall cohesiveness and organization of the epidermis.

Besides forming desmosomes, the keratinocytes of this layer also lose their columnar

shape and begin to take on a more flattened appearance. Both the volume and diameter

of the keratinocyte continue to increase as the cell makes its way through the remaining









strata. In addition to changes in structure, keratinocytes also begin to synthesize keratins

1 and 10 and develop special organelles called lamellar granules that play an important

role in maintaining the barrier properties of the stratum corneum.5, 22

Stratum granulosum. At this stage of keratinocyte differentiation, the cell begins

to die and the nucleus and organelles are enzymatically degraded. As the name suggests,

the cells of the stratum granulosum (SG) are filled with keratohyalin granules (KHGs)

and lamellar bodies (LB, also known as lamellar granules). Keratohyalin granules are

enriched in the precursors of intracellular corneocyte proteins and of the cornified

envelope. Included among these precursors are profillaggrin, loricrin, and keratins 1 and

10. Lamellar bodies are ovoid organelles containing stacks of lipid membranes

composed of phospholipids, cholesterol, and glucosylceramides. In addition, LB contain

high levels of various catabolic enzymes including acid hydrolases, sphingomelinase, and

phospholipase A2.3 The accordion-like appearance of these lipoidal structures likely

results from the compression and subsequent stacking of Golgi-derived lipid vesicles--a

process thought to be mediated by acylglucosylceramide (Figure 1-1).23 The

incorporation of the glucose and linoleic acid moieties into a co-hydroxyceramide

backbone allows acylglucosylceramide to be anchored in the polar phase of one vesicle,

span the lipid interior, and insert itself into the polar surface of an adj acent vesicle

thereby functioning as a "molecular rivet."s At the stratum corneum-stratum granulosum

interface, lamellar bodies are excreted from the cell and their contents made ready for

incorporation into the stratum corneum (SC).











VesiclePolar
Lipid Phase MrbaeLipid Phase Phase Lipid Phase







OH OH

SVesicle 1~ I Vesicle 2
Figure 1-1: Structure of Acylglucosylceramide and General Orientation in Lamellar
Bodies

Stratum corneum. Although the stratum corneum (SC) is the last major layer of

the epidermis, it can be further divided into inner (stratum compactum) and outer

(stratum disjunctum) layers. As the name implies, the cells of the stratum compactum are

packed together more tightly than those of the stratum disjunctum. This difference in

packing and cell cohesion between the two layers is primarily due to the loss of linkages

(corneodesmosomes) between cells in the outer layer in a process known as

desquamation. At the SG-stratum compactum interface, LB fuse with one another24 to

form the intercellular lipid lamellae of the stratum corneum.

The cells of the SC are known as corneocytes. They are nonliving and are

generally considered to be impermeable to most compounds. Compared to the other

layers of the skin, the overall water content of the SC is quite low (approximately 15% by

weight versus 70% for viable epidermis)--the maj ority of which is associated with the

proteinaceous material (mainly keratins 1 and 10 and various degradation products of

filaggrin) that comprises the inner compartment of the corneocytes.3 An impermeable

membrane (the cornified envelope) composed of highly cross-linked protein encloses the

core protein of the corneocytes. This membrane not only functions as a barrier to









permeation, but it also plays an important role in the organization of the intercellular lipid

lamellae via the interaction of co-hydroxyceramides that are covalently bound to the

exterior surface of the cornified evelope.3, 8 These particular lipids are derived from

ceramides 1, 4 and 9 (Figure 1-2) by a deesterification reaction that removes the linoleic

acid group. The primary constituent of the exterior lipids is a co-hydroxyceramide

derived from ceramide I which itself is derived from another important LB lipid (i.e.

acylglucosylceramide, Figure 1-1). These very long chain lipids are likely attached by an

ester linkage at the co-hydroxyl end to a surface protein (possibly involucrin) on the

enve ope.8

The intercelluar lamellae of the SC consist of the following three lipids in their

approximate order of abundance: ceramides (50% by weight), cholesterol (30% by

weight) and free fatty acids (10% by weight).22 To date, nine different ceramides (Figure

1-2) have been isolated from human SC. They have traditionally been labeled in a way

that reflects their relative polarities on thin layer chromatography (TLC). In that regard,

it should be noted that the recently discovered ceramide 9 exhibits a retardation factor

(Rf) on TLC that is between ceramides 2 and 3. Interestingly, ceramide 1 may also serve

the same mol ecular rivet role i n the li pi d lamellae as its pre curs or, acyl gluco syl cerami de,

does in LB.8 Although it is evident from the brief overview presented here that the

composition of the SC is much different from the plasma membranes found in most

tissues of the body, this point is further emphasized by the absence of phospholipids in

the SC.5







12



O O/-~ ar


nide 1



~O

Ceramide 3


Ceramide 2


Ceramide 4 O


h~H~OH

Ceramide 5






Ceramide 7


Ceramide 6






Ceramide 8


Ceramide 9 O



Figure 1-2: Structure of Ceramides found in Human Stratum Corneum


Barrier Properties of the Skin

Physicochemical barrier

The primary barrier to percutaneous absorption is presented by the SC.3, 5 Given

that the enzymatic activity of the SC is much lower than that of the viable epidermis and

dermis,3 the barrier properties of the SC are mainly physicochemical rather than










biochemical in nature. One of the key features of this barrier is the organization of the

corneocytes within the intercellular matrix. As the corneocytes are practically

impermeable to most compounds, they act as "road blocks" in the path of diffusion. In

fact, Potts and Francoeur25 have shown that the diffusion of water through the SC is

1000-times lower than its diffusion through a comparable homogeneous lipid phase.

They also found that the diffusion pathlength was 50-times greater than the thickness of

the membrane. From these results, it was concluded that the diffusion of permeants

through the SC occurs by way of a meandering path around the corneocytes and through

the intercellular lamellae (Figure 1-3).25

I I I I I I I I l
[ I i IIII I Ir 1 r
I I I [I I I I I I I
L 1 I I I Ir I I I IL
I I I I I I 1
0 1 II1 I Ir\ II 1
I II II 11 Rr\ 1 II I







Figure 1-3: Tortuous Path of Permeant Through the Stratum Corneum and Expanded
View of Alternating Nonpolar (White Bands, Electron Lucent) and Polar
(Dark Bands, Electron Dense) Phases Found Within the Intercellular Matrix
(Phases Presented as they Generally Appear in Ruthenium Tetroxide Fixation
of Normal Skin)24

As the intercellular lamellae represent the only continuous pathway in the SC, the

composition and organization of the lipids in this matrix are of primary importance to

percutaneous penetration. Due to the edge-to-edge fusion of the LB at the SG-SC

interface, the intercellular domain is composed of continuous lipid sheets consisting of

repeating units of polar and nonpolar phases (Figure 1-3).24, 26 Despite the high degree of









order within the lipid lamellae, the lipid phases are often interrupted by hydrophilic

bridges that link two neighboring polar phases. As a consequence of the structure of the

intercellular matrix, a permeant must pass through alternating lipid-poor and lipid-rich

layers. The implication for drug design is that in order to maximize flux, the solubilities

of the drug in both lipid and aqueous solvents must be increased.19, 20

Biochemical barrier

Although the skin is primarily a physical barrier, the enzymatic activity of the skin

is significant and should not be ignored. Of the three main skin layers, the epidermis

exhibits the highest enzymatic activity per unit tissue mass and is considered the major

region of drug metabolism in the skin.27 Many of the maj or types of phase I and phase II

reactions are known to occur in the skin including oxidation, reduction, ester hydrolysis,

epoxide hydrolysis (microsomal and cytosolic), methylation, glucuronidation, sulphation,

glycine conjugation, and glutathione conjugation.6 27, 28 It is particularly important to

note that many of the cytochrome P450 enzymes responsible for metabolizing a wide

variety of pharmaceutical compounds in the liver and gut are also found in the skin.4

One of the maj or obstacles to the oral absorption of drugs is the presence of efflux

transporters such as multidrug resistance-associated proteins (MRP) and P-glycoprotein

(P-gp) in the gut wall.' Early attempts to determine the tissue distribution of P-gp found

evidence of this protein in the liver, pancreas, intestine, and kidney but were unable to

detect P-gp in the skin.29 However, recent work in this area has shown that the skin

contains several constitutively expressed MRPs (1 and 3-6). P-gp was also found but

only after induction with dexamethasone.30 Current knowledge about the function of

MRP in the skin is limited.31 In COntrast to its infamous role as a contributor to multidrug

resistance, Randolph and coworkers have demonstrated that P-gp plays an important role









in the migration of Langerhans cells out of the skin by way of the lymphatic vessels.32

Thus P-gp helps maintain a healthy immune response in the skin. Li and coworkers have

also found evidence to suggest that MRP-1 acts as an efflux transporter in the skin.33

Specifically, they found that the tissue-to-plasma concentration ratio of grepafloxacin in

the skin of MRP-1 knockout mice was higher than the corresponding ratio in the skin of

wild type mice following an i.v. inj section of grepafloxacin. Other experiments

demonstrated that the uptake of another MRP-1 substrate ("fluo 3") into the keratinocytes

was significantly increased in the presence of an MRP-1 inhibitor.33 Though these results

provide evidence of active transport of xenobiotics out of kerainocytes via MRP-1, it is

unclear whether such transport would ultimately result in the expulsion of the xenobiotics

to the skin surface (though this does not seem likely given the nature of the SC barrier).

Yet if an active xenobiotic efflux system exists in the skin, it would probably have a

greater effect on delivery into the skin dermall delivery) rather than through it

transdermall delivery). In short, the presence of efflux transporters in the skin raises the

possibility of an additional biochemical barrier effluxx transport out of the skin) to skin

permeability, but the current evidence for such a barrier is not definitive.

Overcoming the Skin Barrier

Strategies

Much research has gone into developing effective methods for overcoming the

barrier properties of the skin.5 9 34 Typical examples of such strategies include the use of

electricity to either create temporary holes in the skin electroporationn) or to

electrostaticly push charged drug molecules into the skin (iontophoresis); penetration

enhancers,36 chemicals designed to temporarily decrease the barrier properties of the skin;

microneedleS37 which physically create micron-sized holes in the skin through which









drug molecules bypass the stratum corneum altogether; and prodrugsl9, 20 which are

transient derivatives of active drugs that temporarily improve the solubility of the drugs

in the skin (thereby increasing their flux through the skin) and then rapidly convert to the

parent drugs in the skin or in the systemic circulation.

Of the methods listed above, penetration enhancers have received the most

attention in industry. However, despite this predilection for chemical enhancers, the

improvement in drug flux is often only modest at best.9 Moreover, the enhancing effects

are often directly proportional to the concentration of the enhancer--a situation which

often results in toxic side effects.36 In Order to reduce or avoid the adverse side effects

associated with penetration enhancers, it has been suggestedl9, 38, 39 that a

prodrug/formulation combination might be a better way to approach the problem. In

many cases, a drug molecule exhibits poor solubility in the skin due to one or more polar

functional groups in the molecule that are either highly charged at physiological pH or

that promote hydrogen bonding and high crystal lattice energies. A prodrug approach

attempts to overcome this problem by temporarily masking the offending functional

group. Since the prodrug is already more soluble in the skin than the parent drug, a much

lower concentration of the chemical enhancer would be needed to experience great

improvement in drug permeability.

Predictive models for optimizing topical delivery

As mentioned in previous sections of this chapter, the intercellular lipid matrix of

the SC is the rate-limiting barrier to the passive diffusion of drugs through skin. Due to

the particular arrangement of the intercellular lipid lamellae (Figure 1-3), permeating

compounds must pass through alternating polar and nonpolar layers within the SC. On

this knowledge alone one might expect percutaneous absorption to be positively









dependent on lipid and aqueous solubilities. Though such dependency is most clearly

seen in homologous series of prodrugs in which the homolog exhibiting the highest flux

also exhibited the best balance of high lipid and high aqueous solubilities.20, 38 Although

such qualitative relationships can serve as a general guide for optimizing topical delivery,

a mathematical model for accurately predicting permeation through skin based on easily-

determined physicochemical properties would be of even greater value as a tool for

quickly identifying lead compounds (i.e. those compounds expected to exhibit the highest

flux).

Mathematical modeling of diffusion through a complex heterogeneous membrane

like the skin can be a formidable challenge. However, the problem can be simplified by

assuming that the skin behaves like a homogeneous membrane. Once this assumption is

made, most quantitative treatments of skin permeability data begin by considering Fick' s

first and second laws of diffusion expressed by equations 1 and 2, respectively:

J= -D(8C/8x) (1)

dC/8t = D(82C/dX2) (2)

Fick' s first law (equation 1) states that the amount of material passing through a given

area of a homogeneous membrane over time (flux, J) is directly proportional to the

concentration gradient across the membrane where D (the diffusion coefficient) functions

as the proportionality coefficient. Fick's second law (equation 2) states that the rate at

which the concentration changes (8C/8t) at any point within the membrane is

proportional (again, D is the proportionality coefficient) to the rate of fluctuation in the

concentration gradient at that point (82C/dX2 .40 If the concentration of the permeant in

the first layer of skin does not change with time, equations 1 and 2 simplify to equation 3:









J = (D/L)(CMlEM 0O) (3)

where L is the distance traveled by the permeant on passage through the skin (note: this is

not the same as the thickness of the skin; see "Physicochemical Barrier" section above)

and CMlEM and Co are the concentrations of the permeant in the first and last layers of the

skin. For all practical purposes, the body functions as a limitless reservoir on one side of

the skin where the concentration of the permeant is essentially zero (i.e. sink conditions).

In this case, CMlEM 0>C and equation 3 reduces to

J =(D/L)(CMEM) = (D/L)(KMlEM:v)Cv (4)

where KMlEM:V is the partition coefficient between the membrane and the vehicle (solvent)

in which the permeant has been applied, and Cv is the concentration of the permeant in

the vehicle.41

In the development of the Kasting-Smith-Cooper (KSC) model,41 the authors noted

that in order to make reliable comparisons of flux the experimental conditions under

which flux was measured should ensure that each permeant exhibited the same

thermodynamic activity. To meet this requirement, Kastings and coworkers decided to

only consider those cases in which the permeant is applied as a saturated solution (Cv =

Sv, where Sv is the solubility in the vehicle). This approach ensures that each permeant

experiences the same thermodynamic driving force since each permeant is at its

respective maximum concentration (i.e. saturation) in the first layer of the skin. Under

these conditions, equation 4 becomes

Jh = (D/L)(SMlEM) = (D/L)(KMlEM:V) SV (5)

where Jhl is the maximum flux, and SMlEM iS the solubility in the skin. In order to arrive at

the diffusion coefficient D, Kasting et al assumed41 that diffusion through the









intercellular lipids of the SC can be approximated from similar models that describe

diffusion through polymer membranes. By this approach, D becomes

D = Do exp (-p MV) (6)

where Do is the diffusivity of a hypothetical molecule having zero molecular volume,42

is a constant that is specific to the skin,43 and MV is molecular volume. The value for

ShlENI in equation 5 was either calculated from ideal solution theory or was assumed to be

approximately equal to the solubility in a model lipid (SLIPID) Such as octanol (SOCT).41

The general form of the KSC model is shown below in logarithmic form:

log Jhl = log (Do/L) + log ShlENI (P/2.303) MV (7)

As noted by Potts and Guy,42 One of the weaknesses of the KSC model is the

assumption that SOCT Can apprOximate the solubilizing capacity of the intercellular lipids

of the SC (ShlENI). To account for the differences between ShlENI and SOCT, Potts and Guy

proposed that when the vehicle is water (Sv = SAQ), KhlEM:AQ and KOCT:AQ are related by

equation 8

KhlEM:AQ = (KOCT:AQ)Y (8)

in which the coefficient y is a measure of the similarities between the two partitioning

domains. Since

ShlEN = (KhlEM:AQ)(SAQ) (9)

substitution of equation 9 into equation 7 gives the following equation for flux

log Jhl = log (Do/L) + y log KOCT:AQ + l0g SAQ p' MW (10)

where molecular weight (MW) has been substituted for molecular volume and P' =

P/2.303 but also includes a conversion factor for using MW in place of MV.42 Whereas









the Potts-Guy model (PG)42 iS an expression of the permeability coefficient (log P = log J

- log Sv) equation 10 is a modified version of PG that describes flux.

Though equation 10 is an improvement over KCS, it suffers from the fact that it

only applies to aqueous vehicles. Furthermore, it offers little insight into the relative

impact of aqueous and lipid solubilites on flux since the SOCT term is "hidden" within

KOCT:AQ. In Order to address these issues, Roberts and Sloan43 were able to extend the

applicability of equation 10 to vehicles other than water in a model which clearly shows

the dependency of flux on aqueous and lipid solubilities. Using isopropyl myristate

(IPM) as example of when a lipophilic vehicle is applied, the following identity may be

used:43

KMlEM:IPM = KMlEM:AQ KIPM:AQ (1

Modification of equation 8 to include IPM gave equation 12

KMlEM:AQ = (KIPM:AQ)Y (12)

Substitution of equation 12 into equation 11 gave equation 13

KMlEM:IPM = (KIPM:AQ)Y KIPM:AQ (13)

The general form of the Roberts-Sloan (RS) equation43 (equation 14) followed from the

assumption that solubility ratios could be substituted for partition coefficients and that

equation 13 could be substituted into equation 10 to give (after collecting terms):

log Jhl = x + y log SIPM+ (l-y) l0g SAQ z MW (14)

where x = log (Do/L) and z = P'.

It is important to note that all three models predict a negative dependence of flux on

the size of the permeant (expressed as either molecular volume MV or molecular weight

MW). However, in contrast to KSC (equation 7 where SMlEM = SOCT 41 and PG (equation









10 where SMlEM = (KOCT:AQ)Y(SAQ),42 RS (equation 14 where SMlEM = (SIPM)Y(SAQ 1-y)

indicates that the intercellular matrix of the SC is a biphasic material consisting of

aqueous and lipid phases--a description which is consistent with electron micrographs of

normal24, 26 and hydrated44 human skin. A fit of the flux, molecular weight, and solubility

data from 61 prodrugs (in vitro mouse) to RS suggested that water solubility was nearly

as important as lipid solubility (0.52 SIPM, 0.48 SAQ, r2 = 0.91).45 When a similar analysis

was performed on a smaller dataset (n = 10) from the delivery of nonsteroidal anti-

inflammatory drugs from mineral oil (MO) through human skin in vivo, flux was again

positively dependent on solubilities in water (0.28 SAQ) and in a lipid (0.28 SAQ, 0.72

Shlo, r2 = 0.93). A recent analysis of a much larger database (n = 103) of in vitro human

skin data gave similar values for octanol and water solubilities (0.56 SOCT, 0.44 SAQ, r2

0.90).46

Prodrugs

By definition, an inactive derivative of an active drug that does not revert to the

parent compound in vivo can not be considered a prodrug, and more importantly, is not

therapeutically useful. For example, Billich and coworkers recently reported that certain

trimethylammonio-alkyl carbonyl derivatives of cyclosporin A (CsA) exhibited fluxes

that were 180-times greater than CsA.47 However, the authors were unable to detect any

CsA in the skin and only trace amounts (< 5% total CsA species as CsA) were found in

the receptor phases of the diffusion cells. In this case, since the derivative was inactive,47

the improvement in flux was therapeutically useless except as a demonstration of the

potential permeation-enhancing effect of a trimethylammonio-alkyl carbonyl group.

Most prodrugs are designed to be enzymatically labile in order to avoid chemical

stability problems that might arise during formulation. One maj or benefit of enzymatic









activation is the potentially greater tissue-specific delivery of the active drug.48 An

example for purely enzymatic activation is the conversion of minoxidil (6-(1-

piperidinyl)-2,4-pyrimidinediamine-3 -oxide) to minoxidil sulfate following topical

application of minoxidil to the scalp (Figure 1-4).49 At least four different

sulfotransferase enzymes are believed to be responsible for the bioactivation of

minoxidil.49, 50 Although it was originally given orally as an antihypertensive agent, it

was later found to stimulate hair growth and is now used as a treatment for alopecia.5o

While the benefits of enzymatic activation are clear, it is important to recognize that

enzyme-mediated reactions are subj ect to interspecies and inter-individual variation,

whereas chemical activation is largely under the control of the researcher--a situation

that results in more predictable rates of delivery of active drug. In the case of minoxidil,

there is evidence to suggest that the inefficacy of topical minoxidil in some individuals is

due to relatively low sulfotransferase activity in those patients."

The rationale for using prodrugs to overcome the skin barrier was briefly

mentioned in Section A-4. Although the most well-known and profitable prodrugs have

been developed for oral administration,2, 48, 52 many of the same types of prodrugs have

been evaluated as topical delivery agents as well.19, 20 A comprehensive review of all the

maj or classes of prodrugs evaluated to date in topical delivery investigations is beyond

the scope of this thesis. However, the interested reader may find such information in

several detailed reviews of the subject.19, 20, 38 In this section, only two of the major

classes of prodrugs, acyl and soft alkyl, will be discussed.












N2 Sulfotransferase .z H
PAPS





Minoxidil Minoxidil Sulfate



Figure 1-4: Bioconversion of Minoxidil to Minoxidil Sulfate by Scalp Sulfotransferase in
the Presence of 3 '-Phosphoadenosine-5 '-phosphosulfate (PAPS)

Acyl Prodrugs

The most common type of prodrug found on the market today is one in which a

heteroatom on the drug has been acylated to give the corresponding ester, carbonate,

amide, or carbamate.2 Most of these promoieties contain simple aliphatic groups in the

acyl chain such as the esters of captopril ((2S)-1-(3 -mercapto-2-methylpropionyl)-L-

proline) recently evaluated by Moss et al (Figure 1-5).53 Six esters of captopril were

synthesized in which only the length of the alkyl chain was varied from the methyl to the

hexyl ester. As expected, all of the prodrugs were less soluble in water (SAQ) than

captopril (range of SAQ = 0.03-0.58 times the SAQ Value for captopril). However, all were

much more soluble in octanol (SOCT) than the parent. Although solubilities in Octanol

(SOCT) WeTO HOt measured, they may be estimated from the calculated partition

coefficients (KOCT:AQ) TepOrted by the authors. By this approach, all of the ester prodrugs

were approximately 4- to 89-times more soluble in octanol. As a result of their higher

lipophilicity, five of the derivatives permeated porcine skin more effectively than

captopril. Within this series of more lipophilic homologs, the member that exhibited the

greatest increase in flux (40-fold) was also the second-most water soluble member of the









series. Thus these results agree with literature precedentl9, 20 and the RS model (equation

14),43 and they demonstrate the dependence of flux on biphasic solubility.

While most acyl-type prodrugs contain simple aliphatic groups in the acyl chain,

there are many reportsl9, 20 Of the benefits of incorporating other functional groups into

the acyl chain. Milosovich and coworkers54 have shown that in lieu of the aliphatic ester

approach that is typically used to deliver steroids,19 introduction of a tertiary amine into

the promoiety can lead to dramatic improvements in flux. To prove the usefulness of

such an approach, the authors reported that a 10% solution of the hydrochloride salt of

testosteronyl-4-dimethylaminobutyrate (TSBH) exhibited a 60-fold greater flux through

human skin in vitro than a 10% suspension of testosterone (TS) (Figure 1-5). The free

base of TSBH also exhibited a flux that was 35-times greater than TS.54 As noted by

Milosovich et al.,54 the relatively high fluxes of the prodrugs are likely the result of

increasing aqueous solubility without compromising lipophilicity. For instance, TSBH is

at least 340-times more soluble in pH 7 phosphate buffer than TS, yet the decrease in

partition coefficient (KOCT:AQ) On going from TS (log KOCT:AQ = 3.3) to TSBH (log

KOCT:AQ = 2.7) is minimal. Similar results were reported by Wasdo and Sloan45 in a study

of alkylcarbonyloxy (AOC) derivatives of acetaminophen (4-hydroxyacetanilde, APAP)

(Figure 1-5). In this case, the goal was to improve the biphasic solubility of the parent by

replacing a methylene group in the acyl chain with oxygen to give an ether. Thus, the

difference between this and the previous example is the absence of an ionizable group in

the acyl chain of the AOC promoiety. The effect of heteroatom substitution on the

physicochemical properties of the prodrugs is most apparent in a comparison of 4-

butyl oxycarb onyl-APAP (4-BuOC -APAP) with 4-(2'-methoxyethyl oxycarb onyl)-APAP









(4-MOC2-APAP). Although 4-MOC2-APAP was 0.74-times less soluble in isopropyl

myristate (IPM) than 4-BuOC-APAP, it was 81-times more soluble in water than 4-

BuOC-APAP and consequently exhibited 8-times the flux of 4-BuOC-APAP. Both

prodrugs were more soluble in IPM (5- to 7-fold) than APAP, but neither was more

soluble in water than the parent. However, since 4-MOC2-APAP exhibited better

biphasic solubility than APAP, its flux was 1.5-times higher than the flux of APAP.

OHO

N~ O NHCOCH,

HS ,,, Acetaminophen (APAP)

Captopril Testosterone (TS) OO

NHCOCH,

O O CI4-BuOC-APAP
N O H- O O
HS *"', O /NCO

Captopril Ethyl Ester TSBH 4-MlOC2-APAP


Figure 1-5: Structures of Acyl Prodrugs for the Topical Delivery of Captopril
Testosterone, and Acetaminophen

A variety of mechanisms have been identified for the conversion of acyl prodrugs

to their respective parent compounds.19, 20 However, simple aliphatic acyl prodrugs are

typically hydrolyzed by one of the mechanisms shown in Figure 1-6.2, 55 Although both

reactions are theoretically reversible, the base-catalyzed hydrolysis is usually driven to

completion by the formation of the carboxylate anion55 and is shown in Figure 1-6 as an

irreversible process.










Acid-Catalyzed Hydrolysis of Esters (AAC2)

slow
H+ + H20 *OH2
Drug-X-C--R Drug-X-C~-R Drug-X--C--R
O O'H O'H


OH slow OH H+
Drug-X-C--R +C--R + Drug-XH Drug--XH + HO- -R
HO OH O

Base-Catalyzed Hydrolysis of Esters (BAC2)

HO -
OH
slowI
Drug-X-C--R Drug-X-C--R



Drug-X +HO- C-R Drug--XHI + O- C-R
0 0



Figure 1-6: Most Common Mechanisms by which Acyl Prodrugs are Hydrolyzed
Chemically

Soft Alkyl Prodrugs

The term "soft alkyl" was first given56 to the alkylcarbonyloxymethyl (ACOM)

derivative of the amide-type compound shown in Figure 1-7 because it is an ester

derivative of the corresponding hydroxymethyl compound which is an alkyl derivative of

the parent drug. Whereas the hydroxymethyl prodrug requires chemical activation to

give the parent, the corresponding ACOM derivative generally undergoes a two-step

process involving an initial enzymatic (or chemical) hydrolysis followed by chemical

activation to give the parent.19 This is in contrast to the "hard alkyl" prodrug shown in

Figure 1-7 for which bioconversion is restricted to enzymatic oxidation.56 Although soft

alkyl derivatives cover a wide range of promoieties,19 Only ACOM and









alkyloxycarbonyloxymethyl (AOCOM) derivatives will be considered since they are the

focus of this thesis.

Much of the work on soft alkyl approaches to improve topical deliveryl9, 20, 38 has

focused on polar heterocycles such as theophylline (Th) and 6-mercaptopurine (6-MP).

Three of these examples are shown in Figure 1-8. In their report on the synthesis and in

vitro evaluation of a homologous series of 7-ACOM-Th derivatives, Kerr and

coworkers" noted that all of the homologs (R = CH3 to CsH11 and (CH3)3C) were

substantially more soluble in IPM (8- to 229-times) than Th. However, the maximum

flux exhibited by any of the prodrugs was only 2.2-times higher (for R = C3H7) than the

flux of Th. Such a modest increase in flux is probably due to the loss of water solubility

(SAQ = 0.04 to 0.27-times the SAQ Of Th) on going from the parent to the prodrug. This

situation is much different for the ACOM prodrugs of 6-MP (R = CH3 to C5H11 and

C7H15). The first three members of the 6-ACOM-6-MP series were 2 to 6-times more

soluble in water than the parent. As with the Th series, all of the 6-ACOM-6-MP

prodrugs were much more soluble (50 to 200-times) in IPM than the parent. In contrast

to the Th series, the 6-MP prodrugs permeated the skin much more effectively than 6-MP

(53 to 69-fold improvement in flux for the first four members of the series). The relative

ineffectiveness of the ACOM approach in the case of Th may be rationalized by

considering the fact that Th itself is 41-times more soluble in water and 15-times more

soluble in IPM than 6-MP. Consequently, Th is much more effective (126-times higher

flux) at penetrating the skin than 6-MP. These results demonstrate that it is easier to

improve the flux of a poorly soluble compound such as 6-MP with a prodrug approach.











,CH3 Enzyme
Hard Alkylated: Drug N/ Drug N OH HO-



NAOH1. HO .
Soft Alkylated: Drug N OH H, C,


H2- R'CO2


Soft Alkylated: Drug N R~~JG8


Figure 1 -7: Mechani sm of Hydroly si s of S oft Alkyl Prodrugs (Alkyl carb onyl oxymethyl
and Hydroxymethyl Derivatives are shown) and Comparison to Metabolism
of "Hard Alkyl" Derivatives (General Mechanism for an Enzymatic N-
Demethylation Reaction is given as an Example)

In spite of their proven effectiveness in oral drug delivery,2 AOCOM prodrugs have

received little attention in topical delivery. In fact, the 7-AOCOM derivative of Th

shown in Figure 1-8 appearsl9, 20 to be the only example of the use AOCOM prodrugs to

improve percutaneous absorption.' However, the authors of the study for which it was

synthesized were more interested in the hydrolytically more labile 7-ACOM-Th prodrugs

and chose not to evaluate this particular derivative in diffusion cells.'" The example of

bacampicillin, an orally administered prodrug of ampicillin, has been included in Figure

1-8 as a reminder of the potential usefulness of the AOCOM promoiety. In a

comparative study of the pharmacokinetics of orally administered pivampicillin (an

ACOM prodrug of ampicillin), bacampicillin and ampicillin, bacampicillin exhibited the

highest rate of absorption and shortest absorption lag time. Both prodrugs were equally

effective at improving the oral bioavailability of ampicillin.59












O O,, O R O R



Theophylline (Th) 7-ACOM-Th 7-AOCOM-Th




HN NS

H H
6-Mercaptopurine (6-MP) 6-ACOM-6-MP







Ampicillin Bacampicillin


Figure 1-8: Examples of Alkylcarbonyloxymethyl (ACOM) and
Alkyloxycarbonyloxymethyl (AOCOM) Prodrugs

Conclusions

Although oral drug delivery will likely remain the method of choice for drug

administration, it is not a suitable route for many different medications due to the

substantial biochemical barrier presented by the GI tract and liver. One of the main

advantages of transdermal delivery is the avoidance of first-pass metabolism that stems

from the relatively low enzymatic activity of the skin compared to the liver. As

illustrated in the case of transdermal versus oral estrogen, topical delivery is often a safer

alternative to the oral route. In addition, topical delivery provides a means for treating

local conditions without exposing the systemic circulation to high levels of the

therapeutic agent.

In contrast to the GI tract and liver, the skin functions mainly as a physical barrier

to drug absorption with the outermost layer, the stratum corneum, providing most of the









resistance to permeation. Electron micrographs of the stratum corneum have shown that

intercellular matrix through which a permeant must pass is composed of alternating

layers of polar and nonpolar material. Such evidence supports in vivo and in vitro skin

penetration experiments in which flux through skin was positively dependent on the

aqueous as well as lipid solubility. These qualitative observations were subsequently

used to develop a mathematical model (i.e. the Roberts-Sloan model, RS) for accurately

predicting flux through skin based on the solubility properties and molecular weight of

the permeant.

Among the many methods used to overcome the skin barrier, a prodrug/formulation

approach is one of the most attractive as it would likely increase permeation while

minimizing side effects. Two of the most successful promoieties used in topical delivery

are the acyl and soft alkyl-type. Of these two types, the acyl promoiety is the most

common perhaps by virtue of its relatively facile synthesis and generally low toxicity of

its hydrolysis byproducts. Though they are not as common, soft alkyl prodrugs have a

long history of improving oral bioavailability as well as topical delivery. AOCOM

derivatives are a sub-type of soft alkyl prodrugs that are underrepresented in topical

delivery and should be further investigated using skin permeation experiments.

Regardless of the promoiety, flux was shown to depend directly on the lipid and aqueous

solubilities of the prodrug.














CHAPTER 2
SPECIFIC OBJECTIVES

First Objective

The first obj ective of the present investigation was to synthesize a homologous

series of alkylcarbonyloxymethyl (ACOM) and alkyloxycarbonyloxymethyl (AOCOM)

derivatives of a model phenol. There are currently no examples of the topical delivery of

ACOM and AOCOM derivatives of phenols. This is in spite of their well-documented

effectiveness at improving the oral bioavailability2 Of phosphates and carboxylic acids,

and the topical deliveryl9 Of amides, imides, thioamide, and carboxylic acids. Most of the

previous work on the topical delivery of phenols via a prodrug approach has focused on

the corresponding acyl derivatives.19, 45, 60-65 One of the most studied classes of drug in

that respect is the narcotic analgesics (see Figure 2-1 for examples). Narcotic analgesics

are usually given intravenously, sublingually, or intramuscularly in order to avoid

extensive first-pass metabolism on oral administration, but the parenteral routes are also

associated with high peak plasma levels and require frequent dosing. In addition to its

avoidance of first-pass metabolism, transdermal administration is typically associated

with constant rates of delivery into the systemic circulation and has a relatively high

degree of patient compliance.' Thus, topical delivery is an attractive alternative to the

current methods by which these compounds are administered.

Most reports on the use of ester (alkylcarbonyl AC)60, 62-64 and carbonate

(alkyloxycarbonyl AOC)45, 61, 65 prOdrugs to increase the percutaneous absorption of

narcotic analgesics indicate that the improvement in flux is only modest (2-7 fold).









However, Sung et al.63 found that the decanoate ester of nalbuphine was 40-times more

permeable than the parent when delivered from pH 4 buffer, and Drustrup et al.60 found

that the 3-hexanoate ester of morphine was approximate 3500-times more permeable than

morphine when delivered from IPM. Although it is impossible to know whether ACOM

and AOCOM prodrugs of phenols will work better than the corresponding acyl

derivatives,19 there does not appear to be great differences in permeation enhancement

when an acyl promoiety is used in place of an ACOM in the same parent drug (compare

1-AC66 to 1 -ACOM-5-fluorouracil67 and 3-AC68 to 3 -ACOM-5-fluorouracil69). On the

other hand, since the carbonyl moiety of the prodrug is separated from the parent

compound by a methylene spacer, the physicochemical properties of soft alkyl

derivatives are governed less by the parent drug and more by the promoiety. The result is

that soft alkyl prodrugs such as ACOM and AOCOM are more easily customized to meet

the particular objectives (drug solubility, stability, etc.) of the investigator.38

One example where a soft alkyl prodrug may be more effective than the

corresponding AC derivative is a-tocopherol (Vitamin E). Vitamin E is one of several

key compounds responsible for maintaining an effective barrier against free-radical

damage in cellular membranes.70 In fact, it is the primary antioxidant for membranes and

lipids. Since the body does not synthesize vitamin E, it must be taken in through diet or

given as a supplement. However, there is currently no efficient way to administer

supplemental Vitamin E.70 Oral administration of Vitamin E suffers from slow

absorption rates" and generally provides inferior photoprotection compared to topically

applied Vitamin E.71, 72 Intravenous formulations of Vitamin E have also been

administered, but in some cases,73 life-threatening side effects have ensued. Part of the









difficulty in delivering Vitamin E is that it is practically insoluble in water74 and readily

oxidizes in air. The problem of instability has traditionally been solved by converting

Vitamin E to its acetate or succinate esters. However, this approach introduces a new

problem: the acetate and succinate esters do not readily revert to the active compound in

vivo.7 A similar problem has been addressed before in the case of p-lactam antibiotics.76

Alkyl derivatives of the carboxylic acid group of these drugs exhibit poor bioavailability

in vivo, but often see dramatic improvements in prodrug-to-drug conversion when a

ACOM or AOCOM approach is used.2 In the case of Vitamin E, nucleophilic attack at

the carbonyl carbon is limited due to the flanking methyl groups on the aromatic ring.

One potential solution to this problem is to move the site of hydrolysis away from the

sterically hindered chromanol head of Vitamin E by way of a soft alkyl (ACOM or

AOCOM) derivative.

Before applying the soft alkyl approach to the narcotic analgesics and Vitamin E, it

seemed prudent to first validate the strategy using a simple phenol. Acetaminophen (4-

hydroxyacetanilide, APAP) was selected as a model because its ACOM and AOCOM

derivatives were expected to be solids (APAP mp = 167-170) and hence more easily

characterized. Since a series of AOC derivatives of APAP had been previously evaluated

in diffusion cell experiments,45 it would also be possible to compare the effects of using

an acyl versus a soft alkyl promoiety.

Second Objective

The second obj ective of this proj ect was to determine whether the ACOM and

AOCOM prodrugs could improve the topical delivery of APAP. Hairless mouse skin in

vitro was selected as a model for human skin due to its relatively low cost and in order to










be consistent with all previous work by our lab. Mouse skin also has the advantage of

exhibiting less variation than human skin.




HO HO \ HOH


O H N--CH, O O. HJ N O OH NP

HO OHON
CH,
Morphine (MOR) Naltrexone (NTX) Nalbuphine (NA) Ketobemidone

HO




HO
MeO H p H
HO-CF-CH, O
C(CH,),

Buprenorphine alpha-Tocopherol (Vitamin E)




Figure 2-1: Phenol-Containing Therapeutic Agents that may benefit from Topical
Delivery via Alkylcarbonyloxymethyl (ACOM) or
Alkyloxycarbonyloxymethyl (AOCOM) Derivatization

Third Objective

The third obj ective of the present investigation was to improve the accuracy of the

Roberts-Sloan equation (RS)43 for predicting flux through hairless mouse skin. At

present, the database (n = 61) upon which the RS equation is based is heavily dependant

on data from heterocyclic compounds: 59% 5-fluorouracil related entries, 18% 6-

mercaptopurine related entries, and 10% Theophilline related entries in the database.

Only 8 of the 61 entries (13%) are of a phenolic compound (i.e. APAP). An earlier study

found that in general, the error in predicting flux using RS was greater for a phenolic









prodrug (4-AOC-APAP) than for a heterocyclic prodrug.45 In Order to extend the

applicability of RS to a wider range of drugs, the structural diversity of the database must

be expanded. Incorporation of the ACOM and AOCOM prodrugs into the database

would likely result in a more robust RS model.














CHAPTER 3
ALKYLCARB ONYLOXYMETHYL PRODRUGS OF ACETAMINOPHEN (APAP)

Synthesis of Alkylcarbonyloxymethyl (ACOM) lodides

A key feature of the Roberts-Sloan database (Chapter 1)20 iS that it is almost

entirely comprised of homologous series. Such homogeneity was intentional as it is

easier to determine the impact of physicochemical properties on flux when structural

differences are minimal. In keeping with that theme, synthetic routes to 3 that allowed R'

to be simple aliphatic groups was desired. Currently, there are three reported methods

for synthesizing such alkylating agents. In two of these procedures, ACOM chloride 4

functions as the intermediate from which the corresponding iodide is subsequently

generated via a Finkelstein-type halide exchange. Chloromethyl chlorosulfate has proven

to be a useful reagent for obtaining ACOM chloride from carboxylic acids under phase-

transfer conditions." 7 However, since this method fails for carboxylic acids with fewer

than 6 carbon atoms," it was not suitable for the present study. Compound 4 may also be

generated via the condensation of acid chlorides with aldehydes in the presence of a

Lewis acid.79, 80 However, this route to ACOM iodide frequently provides low yields of

the desired compound.81 A different approach was taken by Fleischmann and coworkers:

they synthesized pivaloyloxyethyl iodide directly from acetaldehyde and pivaloyl

chloride in the presence of Nal.82 In an effort to extend the applicability of this reaction

to 3 where R = H, it was found that trioxane la reacts with acid chlorides in the presence

of Nal to give predominately compounds 3a-f in one step (Figure 3-1). Paraldehyde lb

exhibited a similar reactivity with acid chlorides under the same conditions to give 3










where R = CH3. The structure of 3 was arrived at by comparison of its 1H NMR spectra

with 1H NMR spectra reported for 3 in the literature.8




O R~ + R'COCI R'CO2CH(R)I + R'CO2CH(R)CI + (R'CO2 2CH(R)
1 2 3 4 5
a: R =H a: R' = CH, a: R' = CHz, R =H a: R' = CHz, R =H a: R' = CHz, R= H
b: R=CH, b: R' = C2Hs b: R' =C2Hs, R= H b: R' =C2Hs R= H b: R' =C2Hs, R= H
c: R'= CH7 c: R'=CH7, R=H c: R'=CH7, R=H c: R'=CH7, R= H
d: R' = CsH;, d: R' = CsH,,, R =H d: R' = CsH,,, R =H d: R' = CsH,, R= H
e: R' = C7His e: R' = C7His, R =H e: R' = C7His, R =H e: R' = C7His, R= H
f: R'=(CH,),C f:R' =CHz, R= CH, g: R' =(CH,),CR= H g:R'=(CH,),C,R= H
g: R' =(CH,),C R= H
Figure 3-1: Reaction of Trioxane la and Paraldehyde lb with Acid Chlorides in the
Presence of Nal

Unfortunately, various amounts of 4 and 5 formed along with 3 as well. Byproduct

4 was identified as the chloride analogue of 3 based on the 1H NMR spectra reported for

this compound in the literature,s and by comparison to an authentic sample of 4 prepared

via a previously reported method.83 COmpound 5 was assigned the structure shown in

Figure 3-1 by comparison of its 1H NMR spectra with the product of the reaction of

acetic acid with 3a by a modification of the method of Folkmann and Lund.84 1H NMR

analysis of the product mixture revealed an upfield shift in the diagnostic methylene

singlet from 5.99 ppm in RCO2CH2I to 5.73 ppm in the product. Furthermore, the

product gave a spectrum that was consistent with bis(acetyloxy)methane. It should be

noted that others have observed the formation of bis(acetyloxy)methane in the reaction of

trioxane with acetyl mesylate.8 Thus, it is not surprising that 5 is also formed in the

present case. However, in reactions involving paraldehyde (R = CH3), 4 and 5 could not

be detected in the 1H NMR spectrum of the reaction mixture.

In an effort to optimize the reaction, various reaction conditions were employed;

the results from some of these experiments are listed in Table 3-1. Given that ACOM









iodides are relatively unstable above room temperature,s 25 oC was set as the upper

temperature limit for all ACOM iodide syntheses. No reaction occurs in the absence of

Nal, and a slight excess of Nal is necessary to achieve good conversion of the starting

materials, regardless of solvent. Similarly, pyridine was unable to catalyze the reaction

(entry 5) and only carboxylic acid, acid anhydride, and starting material was detected in

the product mixture. This result is interesting since French and Adams86 had previously

found that mixtures of pyridine and aromatic acid halides react with aromatic aldehydes

to yield the corresponding ACOM halides. Thus, in the present case depolymerization of

1 may be rate-determining. Yields of the desired compound 3 appear to be unaffected by

variations in temperature below 25 oC. For example, the yield of 3 does not change

substantially if the reactants are allowed to stir for 1 hour at 0 oC after initial mixing,

versus allowing the mixture to stir at room temperature immediately after all reactants

have been added (data not shown). Likewise, the yield of 3 is substantially unaffected by

the length of time over which 2 is added (entry 7 versus entry 9) and by the degree to

which 2 is converted to the acyl iodide before 1 is added (entry 8). On the other hand, the

formation of 3 appears to be more sensitive to the form of the aldehyde undergoing

conversion. This relationship is most apparent in entries 6 and 10. As shown in the

Table 3-1, trioxane reacts with octanoyl chloride to give octanoyloxymethyl iodide in 86

% yield (entry 6). In contrast, paraformaldehyde reacts under the same conditions to give

only 45% yield of the desired ACOM iodide (entry 10). Though the reaction was run

only once, byproduct 4 seems to be more favored when paraformaldehyde is used instead

of trioxane (entry 10 versus entry 6). As shown in entries 3 and 11, the reaction is also










able to accommodate a certain amount of steric hindrance in 1 and 2; however, the

reaction of lb with 2f was not attempted.

Table 3-1: Variation in Reaction Conditions, Crude Yielda of 3, 4, and 5, and Percentage
of 1 Remaining at the End of the End of the Experimentb


a Unless otherwise noted, entries represent a single experiment (n 1). b Reaction time was
usually 20-24 hours. c Molecular ratio shown is based on equivalents of formaldehyde or in the
case of paraldehyde, equivalents of acetaldehyde. d Crude yield determined using benzene as an
internal standard. e 2 is added to a mixture of 1 and Nal within 1-20 min. fCould not determine
from 'H NMR spectrum. g Present as the monomer, acetaldehyde. h 0.3 mol % pyridine added as a
catalyst. 2 is allowed to react with Nal for 1 h at 25 oC. After 1 h, a solution of 1 is added over
40-60 min at 0 oC. 2 is added over 2 h to a mixture of 1 and Nal. k Paraformaldehyde used
instead of trioxane. 1 Average & SD, n 3.


As this study progressed, it became apparent that the product distribution was

dependent on the type of Nal being used. Practically all of the ACOM iodides used in

this study (including those represented in Table 3-1) were prepared using Nal from three

different lots and purity grades purchased from Aldrich during the 1980s (see

Experimental). These particular batches of Nal were eventually consumed and additional

Nal of the same purity and catalog number was ordered from Aldrich. However, when

this new (purchased 2005) Nal was used as shown in Figure3-1, the reaction failed to


Entry R R' Molecular
Ratios
1: 2 :Nal
14 e H C2H5 1 :1 : 1
24 e H C2Hs :12
34 e CH3 CH3 :12
44 e H C2Hs 1:0
Sh H C7His 1 :0
6e H C7H15 1 :1 : 1.2
7e H C7H15 1 :1 : 1
8't, i H C3H7 1 :1 : 1
9' H C7H15 1 :1 : 1
10ek -- 7H15 1 :1 : 1.2
11e H (CH3 3C 1 :1 : 1.2
12e, H C7H15 1 :1 : 1.2
13e H C5H11 1 :1 : 1.2
14e, H C3H7 1 :1 : 1.2
15e, H C2Hs :12
16e, H CH3 1:1:12


Solvent


CD3CN
CD3CN
CDCl3
CDCl3
CH212
CH212
CH212
CH3CN
CH212
CH212
CH212
CH212
CH212
CH212
CH212
CH212


% Yield
3 4


% of 1
5 Remaining


46
54
83


86
74
33
70
45
70
87 & 2
89
82 & 4
80 & 6
72 & 2


11
7
0
0
10
11

16
40
24
6 &1
7
14 & 4
16 &4
19 &3


11
11



4
6

8
15
6
3 & 0.6
4
4 &1
5 &2
11 &2


18
7
17"
100
100
0
6

8
0
0
6 &3
2
0
0
0









reach completion even after 48 hours. Moreover, the mixtures resulting from such

reactions were always contaminated with a large amount of unwanted byproducts. In

other experiments, this "new" Nal was still able to convert alkyl chlorides to the

corresponding iodides as expected for a Finkelstien reaction. These divergent results

were rationalized by assuming that the older batches of Nal were contaminated with

traces of an unidentified catalyst. Subsequent experiments in which various transition

metals and Lewis acids were added to the reaction mixture indicated that this was indeed

the case. For example, zinc dusts7 (23 mol %) was found to catalyze the reaction by fully

converting 1 to products, but unfortunately compound 5 was the major product. Other

transition metal catalysts such as iron also failed to improve the yield of 3. Aluminum

metal, as well as AlCl3 (<; 23 mol %), suppressed the formation of 4 and 5, but failed to

fully convert 1 to products. However, if a combination of AlCl3 (<; 10 mol %) and I2 (<; 5

mol %) was used, total 4 and 5 were minimized (< 14% and < 15% of product mixture,

respectively) and 1 was completely consumed. It was further noted that aluminum metal

is completely consumed during the reaction and that AlCl3 giVCS the same results as

aluminum metal under identical reaction conditions. These results suggest a reaction

mechanism that involves Lewis acid (formed by traces of HCI in the acid chloride

reacting with traces of metal in the Nal) catalysis. Interestingly, in the one case where

All3 and AlCl3 WeTO allOwed to react separately under identical conditions, the resulting

product mixtures differed considerably. Conversion rates in those experiments differed

by 50% (though in neither case was 1 completely consumed), and in the All3 reaction,

several unidentifiable byproducts were formed as well. These results suggest that AlCl3,

and not All3 is the principle catalyst in this reaction.









Though a Lewis acid such as AlCl3 is (apparently) important for successful

formation of 3, other experiments indicate that iodide ion and I2 are needed for the

depolymerization of 1. If AlCl3 is replaced with an equivalent amount of It, 1 is

completely converted to products: molar ratio 3:4:5 = 1:1:1.5 plus an unidentifiable

byproduct. Also, in the absence of Nal, 1 reacts slowly with 2 and AlCl3 to give a

mixture of unidentifiable byproducts and only minor amounts of 4 (approximately 50%

conversion of 1 to these products after 24 hours). Thus, iodide ion likely aids in opening

compound 1, perhaps through an SN2 process similar to that proposed by Balme and

Gores for the cleavage of acetals by TiCl4/Li. Since 12 alSo increases the rate at which

compound 1 is converted to products, it may facilitate the cleavage of 1 by coordinating

with the oxygen atoms in the ring thereby polarizing the CH20--CH2 bond in the formal

moiety. Unfortunately, a catalyst system that consistently matched the reactivity of the

older batches of Nal could not be identified. However, it should be noted that when

crude reaction mixtures of 3 generated via the modified procedure (5 mol % AlCl3 and 2

mol % It included in reaction mixture) were allowed to react with 4-hydroxyacetanilide,

the product mixtures were no different than those obtained using 3 generated from the

older batches of Nal.

Coupling Reaction of ACOM Iodides with 4-Hydroxyacetanilide

It has long been known that ACOM halides 3 display ambident reactivity--

sometimes nucleophiles react at the carbonyl to give acylated products while at other

times the alkyl halide carbon is attacked to give alkylated products (Figure 3-2). Such

reactivity has been observed in reactions of ACOM halides with a variety of nucleophiles

including amines,89 phenols,90 and alcohols.90 In the initial report on the reactions of

ACOM halides with phenols,90 it was noted that the nucleophilicity of the phenol and the









nucleofugicity of the halide are key determinants of the product distribution. More

nucleophilic phenols tend to give acylated products, while better leaving groups and less

nucleophilic phenols shift the product distribution in favor of the alkylated phenol. It was

also suggested that 7 is favored by functional groups at the methylene spacer that are

capable of stabilizing a positive charge.


R OX HO R"R O O + R O R
3 6 7 8
X =CI, Br, I
R"= alkyl, aromatic, etc.

Figure 3-2: General Reaction of Alkylcarbonyloxymethyl (ACOM) Halide 3 with Phenol
6 to Give Aryl Acylal 7 and Aryl Ester 8

Recently, Ouyang and coworkers have suggested that the percentage of 7 in the

product mixture is also directly proportional to the degree of steric hindrance in 3 and in

6.91 According to Ouyang, compound 8 is the major product if 3 and 6 are relatively free

from steric hindrance, but as the degree of steric hindrance in 3 and 6 increase so does the

percentage of 7. Ouyang's conclusions were based on reactions between various phenols

and compounds 9 in which the size of the amino-protecting group was varied from the

relatively small allyloxycarbonyl to the bulky 9-fluorenylmethoxy carbonyl (Figure 3-3).

As shown in Table 3-2, the product distribution was shifted almost entirely toward

acylated phenol 8 when the protecting group was small (entry 13). As the steric bulk of

the protecting group increased, so did the percentage of alkylated product 7, reaching as

high as 15% of the product mixture (entry 15). Although higher yields of 7 were realized

if both 3 and 6 were sterically hindered (entry 16), 8 remained the maj or product in all

cases. Compound 7 became the maj or product (58%) only when the base was changed

from K2CO3 to Cs2CO3, and both 3 and 6 were sterically hindered (R' = Boc-D-Leu, R =























Figure 3-3: Structures of ACOM Derivative of a Protected Amino Acid 9 (R"' =
Protecting Group) and its Corresponding Aliphatic Derivative 10, and
Structure of Byproduct 11

Table 3-2: Product Distribution of the Reactiona of ACOM Halides 3 with Phenols 6:
Data Taken from the Literature
Entry R' R X Y Z Distribution (%)b c' Ref
7 8
1 (CH3)3C H Cl H H 0 100 1.24d 90
2 (CH3)3C H Cl OCH3 H 0 100 90
3 (CH3)3C H Cl NO2 H 50 50 90
4 (CH3)3C H I H H 100 0 90,91e
5 (CH3)3C H I OCH3 H 100 0 90
6 (CH3)3C H I NO2 H 100 0 90
7 CH3 H I H CONH, (25)f 0.52d 93
8 C3H7 H I H CONH, (47)f 0.68d 93
9 (CH3)3C H I H CONH, (29)' 93

10 CH3 H Br Ho oL% (25)' 92


11 C2H5 H Br Hoo(24)' 92


12 (CH3)2CH H I Ho oL% (30) 0.76d 92

Alloc-D- 1.75h
13 H I H H 5 95 ,91
Leu (0.69)"
F-moc-D- 1.75
14 H I H H 10 90 ,91
Leu (1.41)
1.75
15 Boc-D-Leu H I H H 15 85 12) 91

16 Boc-D-Leu H I H Ml 38 62 91

" For entries 1-9 and 13-16, base = K2CO3, solvent = acetone or acetonitrile. For entries 10-12, base =
NaH, solvent = THF. b Determined from 'H NMR spectrum of the crude reaction mixture. Charton's
steric parameter for R'. d Reference 94. e In this case, Cs2CO3 was used as a base in lieu of K2CO3. '
Isolated yield. gReference makes no mention of any products other than 7. Calculated as described in the
text. Steric parameter v of the ally group (reference 95). Steric parameter v of the 9-Methyl-9-fluorenyl
group (reference 95). k Steric parameter of the t-butdl group (reference 94).


H, Y = H, Z = CCCO2Pac). Based on these results, it was concluded that both 3 and 6

must be sterically hindered in order to shift the product distribution in favor of 7.


H ~
R"'O N O I


11













On the other hand, Bensel and coworkers92 have demonstrated that good yields of 7

may be obtained when neither 3 nor 6 is sterically hindered (entries 10-12). Bundgaard

and coworkers93 have also shown that if 6 is sterically hindered but 3 is not, 7 may still be

obtained in good yield (entries 7-9). Yet, it was unclear whether sterically unhindered

ACOM derivatives of 4-hydroxyacetanilide (APAP) could be synthesized given the prior

assertions of Ouyang on the importance of steric hindrance.91 The results from the

reactions of 3a-3f with APAP 6a, phenol 6b, and 2,2,5,7,8-pentamethyl-chroman-6-ol 6c

(Figure 3-4) are shown in Table 3-3.

K2CO,, CHCN
R'CO2CH(R)I + HO 6-9%

3 6
a: R' =CH, R= H
b: R' =C2HR= H a: Y= NHCOCHz, Z= H
c: R'=CH7, R=H b: Y= H,Z=H
d: R' =C;Hs, ~R =H c: 2,2,5,7,8-pentamethyl-
e: R' = C7Hi,, R =H chroman-6-ol
f : R' =CH3, R= CH,



R'CO2C() Y + RCOO

7 8

a: R'=CH3, R=H, Y=NHCOCH,,Z=H a: R' =CH,, Y=NHCOCHz, Z=H
b: R'=C2H, R= H, Y=NHCOCH3,Z= H b: R'=C2H,, Y =NHCOCHz, Z= H
c: R'=CH7, R= H, Y=NHCOCH3,Z= H c: R'=CH7, Y=NHCOCHz, Z= H
d: R'=CHys, R= H, Y=NHCOCHz, Z= H d: R'=C H,4, Y=NHCOCHz, Z= H
e: R'=C7His, R=H, Y=NHCOCHz, Z=H e: R'=C H,z, Y=NHCOCHz, Z=H
f: R'=CHz, R=CHz, Y=NHCOCHz, Z=H g: R'=CH 3,Y=H, Z=H
g: R' = CHz, ~R =H, Y =H, Z= H h: R' = CH,, (phenol = 2c)
h: R' = CHz, ~R =H, (phenol = 2c)


Figure 3-4: Reaction of ACOM Iodides 3a-f with Phenols 6a-c

As shown in Table 3-3, 7 was the major product in every case regardless of the

steric hindrance presented by the phenol 6 or the ACOM iodide 3 (entries 1-6, 8, 9)

despite the predictions of Ouyang.91 There did seem to be a vague relationship between










product distribution and alkyl chain length, however. For alkyl chain lengths longer than

propyl, the percentage of 7 remained close to 70% (entries 1-3). As the alkyl chain

length decreased from propyl to methyl, there was an incremental decrease in the ratio of

7/8 (entries 3-5). The only instance where 8 formed in preference to 7 was when chloride

was used as the leaving group X (entry 7). In addition, the reaction of the sterically

hindered phenol 6c with the relatively sterically unhindered 3a gives credence to the

idea91 that sterically hindered phenols give higher ratios of 7/8 than sterically unhindered

phenols (entry 9 versus entries 5 and 8). Further increases in the percentage of 7 were

realized by introducing a methyl group in place of hydrogen in the methylene linker R of

3 (entry 5 versus entry 6).

Table 3-3: Product Distribution of the Reactiona of ACOM Halides 3 with Phenols 6:
Data from the Present Work
Entry R' R X Y Z Distribution (%)b Vc
7 8 11
71 27 2
Id C7H15 H I NHCOCH3 H 0.73e
(1.7) (1.7) (1.2)
2' CsH11 H I NHCOCH3 H 66 27 7 0.68e
3' C3H7 H I NHCOCH3 H 73 24 3 0.68e
4# C2H5 H I NHCOCH3 H 59 (7) 31 (3) 11 (9) 0.56e
49 37 15
5d CH3 H I NHCOCH3 H 0.52e
(2.9) (4) (7.5)
6' CH3 CH3 I NHCOCH3 H 60 40 0
7' CH3 H Cl NHCOCH3 H 0 100 0
8' CH3 H IHH 63 37 h

9' CH3 H I68 3

"Base = K2CO3, solvent = acetone or acetonitrile. b Determined from 'H NMR spectrum of the crude
reaction mixture. Charton's steric parameter for R'. d Average (SEM, 3 experiments). e Reference 94. fn
=1. Average, 2 experiments: value in parenthesis is the range. Could not determine by 'H NMR.


It is important to recognize that compounds 3a-3f were not purified before they

were used in the coupling reactions with 6a-c. As such, they (with the exception of 3f)

contained various amounts of 4 (Table 3-1) which may or may not have influenced the










product distributions. In their initial report on the coupling reactions of 3 with 6, Sloan

and Koch observed that acylated products readily formed when X = Cl even though 4

was sterically hindered (entries 1-3, Table 3-2).90 Although 3 is contaminated with 4 in

the present study, 3 is much more reactive, and is in excess of 4 by at least 3-4 fold. Thus

6 is more likely to react with 3 than 4. If the reaction of byproduct 4 with 6 is significant

under the present conditions, then there should be a correlation between the ratio of 8/7

and the percentage of 4 in the crude product 3 (Table 3-1). Using entry 4 (Table 3-2) as a

reference point for when 3 is pure, a plot of the ratio of 8/7 versus the percentage of 4 in

crude 3 is shown in Figure 3-5. As shown in Figure 3-5, there does not appear to be a

strong relationship between the purity of 3 and the ratio of 8/7. It is therefore reasonable

to assume that the product distributions observed in the present investigation result solely

from the reaction of 3 with 6.


0.8-

0.7 -y = 0.0243x + 0. 1204
0.6 -1 R2 = 0.7207

0.5-


S0.3-
0.2-
0.1-


0 5 10 15 20
% RCO2CH2C

Figure 3-5: Plot of the Percentage of 4 (RCO2CH2C1) in Crude 3 Versus the Ratio of 8/7
(Acylated/Alkylated phenol) Resulting from the Reactions of 3a-3e with 6a
and 6b (Taken from Entry 4, Table 3-2 and Entries 1-4, and 8, Table 3-3 m;
and Entry 5 o; Note: Entry 5 not Included in Linear Regression Analysis as it
appears to be an Outlier)









When ascertaining the affect of steric hindrance on a given reaction, it is often

helpful to use a quantitative measure of steric hindrance. In the present work, this was

done by relating Charton's steric parameter v94 to the ratio of alkylated / acylated phenol

(Tables 3-2 and 3-3). Since most of the derivatives of 3 shown in Tables 3-2 and 3-3

contain simple aliphatic groups in the acyl portion R', the v values could be taken

directly from the literature.94-96 To our knowledge, v values for the R' groups in entries

13-16 (Table 3-2) have not been reported. Since it was desirable to make all comparisons

of steric effects using the same scale, the steric parameter v for these groups were

estimated by assuming that the van der Waals radius of the carbamate moiety in 9 is

approximately equal to the corresponding arrangement of methylene groups in 10.97

Using 10 as a surrogate for 9, v values were then calculated96 fTOm v = 0.497n, + 0.409np

+ 0.0608n, 0.309, where n,, np, and n,, are the number of carbon atoms attached to the

alpha, beta, and gamma carbon atoms, respectively, in 10. Alternatively, the steric effect

of R' in entries 13-16 (Table 3-2) may be evaluated by assuming that for this series, the

ratio 7/8 is determined primarily by the steric bulk of the amino protecting group R"''. In

this case, v values may be taken from the literature since the steric parameters of R"'' are

known (values in parentheses, entries 13-16, Table 3-2).94, 95

Since neither Bundgaard93 HOr Bensel92 mentioned product distributions in there

reports, entries 7-12 (Table 3-2) offer only indirect evidence of the effect of steric

hindrance on the formation of 7 and 8. What is clear from their findings is that good

yields of 7 may be obtained under essentially the same conditions used by Ouyang91 but

from a sterically unhindered ACOM halide (X = Br or I). For entry 4 (Table 3-2) and

entries 1-5, and 8 (Table 3-3), the variation in 7/8 appears to be directly related to the










variation in v. A plot of v versus the ratio of 8/7 for these entries (Figure 3-6) gave a

good correlation (r2 = 0.95). If these results are representative of all reactions of acyclic 3

(where R' is aliphatic) with 6, then the effect of R' on the product distribution is related

to its ability to discourage nucleophilic attack at the carbonyl. Such a finding should not

be surprising since nucleophilic substitution at a carbonyl carbon is known to be sensitive

to steric hmndrance."





0.80
0.70

0.60 *
0.50
S0.40 -y = -0.7693x + 0.9358
4 R2 = 0.951
0.30
0.20
0.10
0.00 *r
0.40 0.60 0.80 1.00 1.20 1.40
steric parameter




Figure 3-6: Plot of Charton' s steric parameter v for R' Versus the Ratio of 8/7
(Acylated/Alkylated Product) Resulting from the Reactions of 3a-3e with 6a
and 6b (Taken from Table 3-2: Entry 4, Table 3-3: Entries 1-4, and 8 and
Entry 5 o. Note: entry 5 not included in linear regression analysis as it appears
to be an outlier)

On the other hand, analysis of the steric effect in entries 13-16 (Ouyang' s data,

Table 3-2)) is more complicated. If one assumes that 9 (R' = protected amino acid) and 3

(R' = simple aliphatic chain) react with 6 by the same mechanism, and that the acyl group

of 10 can approximate the steric effect of the acyl group in 9, then variations in the amino










protecting group should have no effect on product distribution, contrary to the

conclusions of Ouyang.91 This follows from the work of Charton96 that showed that for

aliphatic acyl groups, substitution at the delta carbon contributes nothing to the effective

van der Waals radius of the acyl group. Indeed, the fact that the ratio of 7/8 increases on

going to bigger protecting groups (see v values in parentheses, Table 3-2) implies that 9

reacts with 6 by a different mechanism than that prescribed90 foT Simple derivatives of 1

(where R' is aliphatic). One potential mechanism for rationalizing the results of Ouyang

is shown in Figure 3-7. It may be possible for compounds such as 9 to cyclize to give 5-

oxazolidinone 12. 5-Oxazolidinones are known to undergo nucleophilic addition at the

carbonyl carbon to give 13, followed by loss of formaldehyde to give 8.98 In this

scenario, bulky protecting groups likely retard the conversion of 9 to 12 and thus permit 9

to exhibit a reactivity with phenols similar to that displayed by more conventional

derivatives of 3 (i.e. where R' = aliphatic).




R"'O O "O N OR' O R'


9 12 13 8

Figure 3-7: Speculative Mechanism for Reactions of Protected Amino Acid Derivatives 9
with Phenols 6

In addition to the expected products 7 and 8, there was also the unanticipated

formation of byproduct 11 (Figure 3-3) in reactions involving APAP 6a (entries 1-5

Table 3-2). Compound 11 was assigned the structure shown by comparison of its 1H

NMR to the corresponding derivatives 7 (compound 11 was also analyzed by IR, but no

useful structural information could be gleaned from the spectrum). At present, it is not









clear why 11 is generated in reactions involving APAP, but fails to form when other

phenols react with 3 (entries 1-16 Table 3-2 and entries 8 and 9 Table 3-3). An analysis

( H NMR) of the crude reaction mixture resulting from the synthesis of 3 showed no

evidence of alkylating agents such as R' CO2CH20CH2I Or bis(acetoxymethyl) ether99

which might react with 6 to give 11. Presumably, the formaldehyde generated during the

acylation of 6 by 3 goes on to react with another molecule of 3 to form

R' CO2CH20CH21.

Several reaction conditions were employed in an effort to maximize the yield of 7.

Methods such as solid-liquid phase-transfer catalysis or the use of a non-nucleophilic

organic base failed to improve the yield of 3. Interestingly, the use of 1,8-

diazabicyclo[5.4.0]undec-7-ene as a base resulted in an increase in the percentage of 7 by

approximately 20% for the least sterically hindered member of the series (3a). However,

since the conversion of 6 to 7 was lower is this case, this technique was not synthetically

useful. Replacing K2CO3 with Cs2CO3 aS recommended by Ouyang91 TOSulted in an

increase in the conversion of 6 to 7 (50% versus 40% when K2CO3 WAS used as a base)

when 3a was used but such effect was not observed with the longer alkyl chain

derivatives. Likewise, the use Cs2CO3 TOSulted in a slight increase in the ratio of 7/8

when 3a was used (59/32 versus 53/44), but had no effect on product distribution for a

longer alkyl chain derivative such as 3e. As it turns out, the original ACOM/phenol

coupling method of Sloan and Koch90 prOVed to be the most effective in the present case

as well.100

As mentioned above and shown in Tables 3-2 and 3-3, the mixtures resulting from

the coupling of 3 and 6 are frequently contaminated with a large percentage of 8,









especially when R' offers little steric hindrance. Unfortunately, isolated yields of 7 suffer

as a consequence (see Experimental). Compounds 7 and 8 could not be separated by

simple crystallization, and could only be isolated in poor to low yield (1-30%) by way of

a time-consuming chromatographic procedure involving multiple passes through a

column of silica gel. Reverse-phase chromatography failed to improve the separation.

However, other have reported that a phenolic ester can be selectively cleaved in the

presence of an aliphatic ester.101-105 Yet when these techniques were applied to mixtures

of 7 and 8, a large portion of 7 was destroyed along with 8. Aminolysis with hydrazinel06

and t-butylaminel07 proved ineffective as well. Selective cleavage of 8 was Einally

achieved by subj ecting the crude reaction mixture to a solution of imidazole in 30%

aqueous acetonitrile.10s In general, the selectivity for 8 varied with the steric hindrance in

R and R'. This trend is reflected in the differences in isolated yield of 7 discussed in the

Experimental Section. Even though a portion of 7 is cleaved via this procedure, it was

quite practical in that it simplified the purification of 7. For example, compound 7 is

easily separated from the product of the cleavage of 8 (parent phenol 6) via a single

elution from a column of silica gel. Interestingly, byproduct 11 appeared to be unaffected

by this procedure.

Conclusions

A new method has been developed for synthesizing ACOM iodides 3 in one step

and in good yield starting from trioxane or paraldehyde. This reaction was found to be

dependent on an unidentified catalyst that was present in older batches of Nal, but is

absent in newer, purer batches. Although an optimized procedure for synthesizing the 3

using the newer brands of Nal was not developed, potential catalysts were identified.

The coupling reaction of 3 with phenols 6 appears to be somewhat dependent on steric









hindrance as measured by Charton's steric parameters. In fact, the percentage of

alkylated phenol 7 in the product mixture increases with increasing steric hindrance in 3

and in 6. However, based on literature precedent92, 93 and new data from our lab (Table

3-3), alkylated phenol is favored over acylated phenol regardless of the steric hindrance

in 3 or 6, contrary to the findings of Ouyang.91 As Ouyang's is the only report where R'

is a protected amino acid, this particular acyl group may impart a unique reactivity to 3

not found in more common derivatives (i.e. where R' = hydrocarbon).

Experimental

Batches of sodium iodide designated as "old" in the text were purchased from

Aldrich (99+%, catalogue number 21763-8, lot numbers 1327 DK and 04229 CV; 99.5%,

catalogue number 38311-2, lot number 11717 MG). Batches of sodium iodide designated

as "new" in the text were purchased from Aldrich (99+%, catalogue number 21763 8, lot

number 05412 BC; 99.5% catalogue number 383112, lot number 07908 CC) and from

Fisher (Certified, catalogue number S324-500, lot number 037120). Thin layer

chromatography (TLC) plates (Polygram Sil G/UV 254) were purchased from Brinkman.

Spectra ( H NMR) were recorded on a Varian Unity 400 MHz spectrometer or on a

Varian EM-390 90 MHz spectrometer. Melting points were determined on a Meltemp

melting point apparatus. Sodium sulfate and all solvents were purchased from Fisher.

Trioxane and paraldehyde were purchased from Eastman Chemical Company. Iodine

(crystalline) was purchased from Mallinckrodt. All other reagents were from Aldrich.

Containers ofNal and Cs2CO3 were wrapped in parafilm and stored in a vacuum

desiccator. Solvents listed as "dry" below were obtained as such following storage over

4-angstrom molecular sieves. Microanalyses were performed by Atlantic Microlab, Inc.,

Norcross, GA.









General procedure for the synthesis of 3a-e and 3g--synthesis of 3a: Sodium

iodide (12 mmol, from any of the "old" batches listed above) was added to a stirred

solution of la (3.3 mmol) in 12 mL dichloromethane, and the suspension that resulted

was cooled to 0 oC. A solution of 2a (10 mmol) in 12 mL dichloromethane was then

added, and the resulting mixture was allowed to reach room temperature. The reaction

vessel was protected from light with aluminum foil while the contents were allowed to

continue stirring at room temperature for 20-24 hours. The reaction mixture was filtered

by vaccum followed by concentration of the filtrate at room temperature on a rotary

evaporator to give an orange-colored oil. A sample of this oil was dissolved in CDCl3

and analyzed by 1H NMR. The yield of 3a was then calculated on the basis of the molar

ratio of the products. No further effort was made to purify 3a, and it was used as such in

subsequent reactions with phenols. Representative spectrum ( H NMR, CDCl3) fTOm the

reaction of la with 2a to give 3a (R' = CH3): 6 5.90 (s, 2 H), 6 2.10 (s, 3 H).

Reaction of 1 with 2 by modified procedure using AICl3/I2--Synthesis of 3f:

Sodium iodide (15.2 mmol, 2.28 g, from Fisher) was added to a stirred solution of lb (R

= CH3) (4.2 mmol, 0.55 g) in 25 mL dichloromethane, and the suspension that resulted

was cooled to 0 oC. A solution of 2a (12.7 mmol, 1.00 g) in 10 mL dichloromethane was

then added. Subsequent addition of aluminum chloride (0.42 mmol, 0.056 g) and iodine

(0.084 mmol, 0.021 g) gave a mixture that was then allowed to warm to room

temperature. The reaction vessel was protected from light with aluminum foil while the

contents were allowed to continue stirring at room temperature for 20-24 hours. After

such time, the reaction mixture was filtered by vacuum, diluted with 25 mL

dichloromethane, then washed with 10 mL 10% aqueous Na2S203 followed by 10 mL









brine. The organic phase was then dried over Na2SO4, filtered, and concentrated at room

temperature on a rotary evaporator to give 10.2 mmol 3f in Cl2CH2 (80% yield).

Reaction of 6a with 3--the reaction of 6a with 3e: To a stirred suspension of 6a

(19.9 mmol, 3.01 g) and K2CO3 (39.8 mmol, 5.50 g) in 50 mL dry acetonitrile was added

a solution of 3e (as indicated above, this solution is actually a mixture of 87% 3e, 7%

C7HlsCO2CH2C1, 4% (C7HlsCO2)2CH2), 1% trioxane) in 15 mL dry acetonitrile. The

mixture that resulted was allowed to stir overnight at room temperature. The reaction

mixture was then filtered and concentrated in vacuo to give 10.82 g oily residue. 1H

NMR (DMSO-d6) analySis of the solid retained in the filter cake revealed only a trace

amount of unreacted APAP. 1H NMR (DMSO-d6) analySis of the oily residue showed 89

% conversion to products and the product distribution shown in Table 3-2. Column

chromatography (3 consecutive experiments) on silica gel (gradient =

hexane~dichloromethanewacetone) gave 2.37 g of 4-octanoyloxymethyloxyacetanilide

7e as an oil (39%). This oil was then triturated with pentane to give 1.89 g of 7e as

colorless crystals (31%); mp = 53-54 oC; one spot on TLC (CHCl3 : acetone, 97 : 3) Rf

0. 13; 1H NMR (CDCl3) 6 7.42 (d, J= 8 Hz, 2 H), 6 7.09 (brs, 1H), 6 6.99 (d, J = 8 Hz, 2

H), 65.73 (s, 2H), 62.35 (t, J= 7Hz, 2H), 62. 16(s, 3H), 61.62 (m, 2H), 61.26

(quint, J= 7 Hz, 8 H), 6 0.87 (t, J= 7, 3 H); Anal. Called for C17H25NO4: C, 66.43; H,

8.20; N, 4.56. Found: C, 66.51; H, 8.19; N, 4.55.

In addition to 7e, the chromatography procedure described above also gave 2.24 g

solid material composed of a mixture of 7e and 4-octanoyloxyacetanilide 8e in a ratio of

1.3 : 1.0. By way of simple crystallization (EtOAc : hexane), 0.48 g of 8e was isolated

from this mixture as colorless crystals (9%); mp = 106-108 oC (lit37 103-105 oC); one









spot on TLC (CHCl3 : acetone, 97 : 3) Rf 0. 10; Anal. Called for C16H23NO4: C, 69.29; H,

8.36; 5.05. Found: C, 69.06; H, 8.34; N, 5.04.

The reaction of 6a with 3d was carried out and processed as described above

for 3e, except that in this case, the scale was larger (52.1 mmol) and a different

solvent gradient (hexane~,dichloromethane4EtOAc) was used for column

chromatography (3 consecutive experiments). In this way, 2.96 g of 4-

hexanoyl oxymethyl oxy ac etanili de 7 d wa s i isolated as col orl e ss cry stals (2 0%); mp = 50O-

52 oC; one spot on TLC (Cl2CH2 : EtOAc, 85 : 15) Rf 0.20; 1H NMR (CDCl3) 6 7.42 (d, J

= 8 Hz, 2 H), 6 7.10 (brs, 1H), 6 6.99 (d, J= 8 Hz, 2 H); 6 5.73 (s, 2 H), 6 2.35 (t, J= 7

Hz, 2H), 62. 16(s, 3H), 61.63 (quint, J= 7Hz, 2H), 61.29 (m, 4H), 60.87 (t, J= 7

Hz, 3 H), Anal. Called for C15H21NO4: C, 64.50; H, 7.58; N, 5.01. Found: C, 64.54; H,

7.56; N, 4.97.

In addition to 7d, 4-hexanoyloxyacetanilide 8d was isolated in a fashion similar to

that described above for se: 0.30 g of pale blue crystals (2%), mp = 105-109 oC (lit37 107-

109 oC); one spot on TLC (Cl2CH2 : EtOAc, 85 : 15) Rf 0. 17; Anal. Calcd. for

C14H19NO4: C, 67.45; H, 7.68; N, 5.62. Found: C, 67.17; H, 7.64; N, 5.59.

The reaction of 6a with 3c was carried out as described above for 3e, except

that in this case, the scale was much larger (112 mmol). The corresponding compound

Sc was selectively destroyed as described below to give 51.24 g oil containing 7c, 11c,

and 6a in the ratio of 50 : 1 : 3. The oil was then subjected to column chromatography

(silica gel, EtOAc : hexane, 1 : 1) to give 12.51 g of 4-butryloxymethyloxyacetanile 7c as

an oil (44%). Crystallization from diethyl ether : 2-methyl-butane gave 7.03 g of 7c as

colorless crystals (25%); mp = 56-58 oC; one spot on TLC (EtOAc: hexane, 1 : 1) Rf










0. 16; 1H NMR (CDCl3) 6 7.42 (d, J= 8 Hz, 2 H), 6 7. 13 (brs, 1H), 6 6.99 (d, J = 8 Hz, 2

H), 65.74 (s, 2H), 62.34 (t, J= 7Hz, 2H), 62. 17(s, 3H), 61.65 (m, 2H), 60.94 (t, J=

7 Hz, 3 H); Anal. Called for C13H17NO4: C, 62.14; H, 6.82; N, 5.57. Found: C, 61.92; H,

6.85; N, 5.52.

The reaction of 6a with 3b was carried out and processed as described above

for 3c, except in this case, two consecutive column chromatography experiments

(acetone : hexane 3 : 7) were required to separate 7b from 11b. Following

crystallization from ether : pentane, 3.64 g of 4-propionyloxymethyloxyacetanilide 7b

was obtained as colorless crystals (15%); mp = 56-59 oC; one spot on TLC (acetone :

hexane, 35 : 65) Rf 0.26; 1H NMR (CDCl3) 6 7.42 (d, J= 8 Hz, 2 H), 6 7.10 (brs, 1 H) 6

6.99 (d, J= 8 Hz, 2 H), 6 5.74 (s, 2 H), 6 2.39 (quart, J= 8 Hz, 2 H), 6 2. 16 (s, 3 H), 6

1.15 (t, J= 8 Hz, 3 H); Anal. Called for C12H15NO4: C, 60.75; H, 6.37; N, 5.90. Found:

C, 60.85; H, 6.35; N, 5.84.

In addition to 7b, column chromatography gave 3.14 g oil composed of a mixture

of 4-propionyloxymethoxymethoxyacetanilide 11b, solvent, and an unidentified

compound. Crystallization from Cl2CH2 : hexane gave 1.05 g of 11b as colorless

crystals (4%); mp = 71-73 oC; one spot on TLC (acetone : hexane, 3 : 7) Rf 0. 18; 1H

NMR (CDCl3) 6 7.08 (d, J= 8 Hz, 2 H), 6 6.88 (d, J= 8 Hz, 2 H), 6 5.40 (s, 2 H), 6 5.19

(s, 2 H), 6 2.35 (quart, J= 8 Hz, 2 H), 6 1.91 (s, 3 H), 6 1.31 (t, J= 8 Hz, 3 H); Anal.

Called for C13H17N05: C, 58.42; H, 6.41; N, 5.24. Found: C, 58.45; H, 6.43; N, 5.24.

The reaction of 6a with 3a was carried out and processed as described above

for 3c, except that in this case, an aqueous workup was not performed on the

aminolysis reaction (reaction mixture was too complex to determine ratio 7a, Sa, 11a









and 6a). Instead, the crude mixture was subjected to three consecutive column

chromatography experiments (first two experiments used hexane4Cl2CH24acetone;

final experiment used EtOAc : hexane, 1 : 1). In this way, 4-

acetyloxymethyloxyacetanilde 7a was obtained as 1.81 g pale green crystals (6.5%,

crystallized from ether : 2-methylbutane); mp = 92-95 oC; one spot on TLC (Cl2CH2

acetone, 95 : 5) Rf 0.21; 1H NMR (CDCl3) 6 7.43 (d, J= 8 Hz, 2 H), 6 7. 14 (brs, 1 H), 6

7.00 (d, J= 8 Hz, 2 H), 6 5.73 (s, 2 H), 6 2.18 (s, 3 H), 6 2.12 (s, 3 H); Anal. Called for

CllH13NO4: C, 59.19; H, 5.87; N, 6.27. Found: C, 58.96; H, 5.84; N, 6.22.

In addition to 7a, column chromatography also gave 1.61 g of 4-

acetyloxymethoxymethoxyacetanilide 11a as an oil. Crystallization from diethyl ether :

2-methylbutane gave 0.40 g 11a as colorless crystals; mp = 91-93 oC; one spot on TLC

(acetone : hexane, 3 : 7) Rf 0. 15; 1H NMR (CDCl3) 6 7. 10 (d, J= 8 Hz, 2 H), 6 6.87 (d, J

= 8 Hz, 2 H), 6 5.38 (s, 2 H), 6 5.18 (s, 2 H), 6 2.07 (s, 3 H), 6 1.90 (s, 3 H); Anal. Called

for C12H15N05: C, 56.91; H, 5.97; N, 5.53. Found: C, 56.72; H, 5.96; N, 5.47.

The reaction 6a with 3f was carried out and processed as described above for

3c except that in this case, the scale was much smaller (8.5 mmol). Using this

procedure, 1.54 g of oil containing 7f : 6a in the ratio of: 16 : 1 was obtained. The oil

was subjected to column chromatography (silica gel, acetone : hexane, 3 : 7) to give 0.79

g 4-acetyloxyethyloxyacetanilide 7f as a colorless solid. This solid was recrystallized

from ether : 2-methylbutane to give 0.56 g 7f as colorless crystals (28%). Upon heating,

7f displayed an initial melting point of 82-92 oC. Once this material had cooled to room

temperature and solidified, it was heated again. This time, 7f displayed a sharp melting

point: 81-83 oC; one spot on TLC (acetone : hexane, 3 : 7) Rf 0.20; 1H NMR (CDCl3) 6










7.41 (d, J= 9 Hz, 2 H), 6 7.08 (brs, 1 H), 6 6.92 (d, J= 9 Hz, 2 H), 6 6.51 (quart, J= 5

Hz, 1H), 62. 16(s, 3H), 62. 10(s, 3H), 61.60 (d, J= 5Hz, 3H); Anal. Called for

C12H15NO4: C, 60.75; H, 6.37; N, 5.90. Found: C, 60.69; H, 6.40; N, 5.91.

General procedure for the selective aminolysis of 7 in the presence of 8. The

procedure described above for the reaction of 6a with 3 gave various mixtures of 7, 8, 11,

unreacted 3 and 6a (determined by 1H NMR as described above). The mixture was then

triturated in dichloromethane, filtered, and concentrated in vacuo to give an oil. The oil

was dissolved in 30% aqueous CH3CN (approx. 17 mL / 1 mmol 8), and imidazole was

added (10 equiv. based on mmol 8 present in the oil, as determined by 1H NMR). The

resulting mixture was allowed to reflux overnight. After such time, the solvent was

removed in vacuo. The residue was dissolved in dichloromethane, washed with 1 M HCI

(1/6 vol. of organic phase), and water (1/6 vol. of organic phase). The organic phase was

dried over Na2SO4, filtered, and concentrated in vacuo to give an oil containing various

ratios of 7 : 8 : 11 (determined by 1H NMR: specific ratios listed above).

In Vitro Determination of Flux of ACOM Prodrugs of APAP






4-Hydroxyacetanilide 4-ACOM-APAP
(APAP)
6a 7a, R= CHz
7b, R = C2H
7c, R = CH,
7d, R= CsH,,
7e, R = CHi
Figure 3-8. Structure of 4-Hydroxyacetanilide and Corresponding 4-ACOM Prodrugs









Materials and Methods

Melting points were determined on a Meltemp capillary melting point apparatus

and are uncorrected. Ultraviolet (UV) spectra were obtained on a Shimadzu UV-265 or

UV-2501 PC spectrophotometer. The vertical Franz diffusion cells (surface area 4.9 cm2,

20 ml receptor phase volume, 15 ml donor phase volume) were purchased from Crown

Glass (Somerville, NJ, USA). A Fisher (Pittsburgh, PA, USA) circulating water bath was

used to maintain a constant temperature of 32 oC in the receptor phase. Isopropyl

myristate (IPM) was purchased from Givaudan (Clifton, NJ, USA). Theophylline (Th)

was purchased from Sigma Chemical Co. (St. Louis, MO, USA); all other chemicals

were purchased from Fisher. The female hairless mice (SKH-hr-1) were obtained from

Charles River (Boston, MA, USA). All procedures involving the care and experimental

treatment of animals were performed by Professor K. B. Sloan of the department of

Medicinal Chemistry in agreement with the NIH "Principles of Laboratory Animal Care."

Physicochemical properties and analysis

The molar absorptivity of each prodrug at 240 nm (E240) in acetonitrile was

determined in triplicate by dissolving a known amount of prodrug in acetonitrile, and

analyzing the dilute solution by UV spectrophotometry. Since the concentration C was

known, 8240 COuld be calculated by way of Beer' s law:

A240 = 240 1 C, where l= cell length (1)

For each prodrug, the solubility in isopropyl myristate (IPM) was determined in triplicate

by crushing a sample of the prodrug into a fine powder. Excess powder was added to a

test tube containing 3 ml IPM. The test tube was then insulated and the suspension was

allowed to stir at room temperature (23 + 1 oC) for 24 hours on a magnetic stir plate. The










suspension was filtered through a 0.25 Clm nylon syringe fi1ter. A sample of the fi1trate

was diluted with acetonitrile and analyzed by UV spectrophotometry. In order to be

consistent with a previous investigation of acetaminophen prodrugs,45 the absorbance at

240 nm (A240) WAS used to calculate the prodrug concentration C in the IPM solution

using the Beer' s law relationship. In this case, since C is the concentration of a saturated

solution, C is the solubility in IPM (SIPM):

CSaturation = SIPhi = A240 / 240 (2)

Solubilities in water were also determined in triplicate using an identical protocol to the

one described above, except that the suspensions were only stirred for one hour before

filtering. This was done in order to make direct comparisons between the present

investigation and previous studies.45, 68 In each case, a sample of the filtrate was diluted

with acetonitrile and analyzed by UV spectrophotometry using 8240 in acetonitrile (Table

3-4).

Table 3-4: Molar Absorptivities (E) of APAP 6a and Prodrugs 7a-e
Compound E240 in ACNa, b E240 a, cufr~ E280 aDffr d
6a, APAP 1.36e 1.01 0.053f 0. 174 0.020f
7a 1.48 & 0.011
7b 1.64 & 0.067
7c 1.56 & 0.057 1.20 0.025g 0.119 0.0025g
7d 1.58 & 0.050
7e 1.46 & 0.044
" Units of 1 x 104 L molf b MOlar absorptivities at 240 nm acetonitrile (a SD, n = 3). Molar
absorptivities at 240 nm in pH 7.1 phosphate buffer with 0. 11% formaldehyde. d MOlar
absorptivities at 280 nm in pH 7.1 phosphate buffer with 0. 11% formaldehyde. e Taken from
Reference 45.fn = 5 (f SD). n =6 (f SD).

Partition coefficients were also determined in triplicate for each prodrug by using

the saturated IPM solutions obtained from the solubility determinations. Since solubility

in pH 4.0 buffer (S4.0) is a parameter in the Roberts-Sloan database,20 acetate buffer (0.01

M, pH 4.0) was used as the aqueous phase in the partition coefficient experiments. In









this way, S4.0 COuld be estimated as described previously109 and the values included in the

database. Thus, an aliquot of the saturated IPM solution was partitioned against pH 4.0

buffer using the following volume ratios (V4.0 / VIPM) for compounds 7a, 7b, 7c, and 7d:

0.5, 2, 10, and 20, respectively. The two phases were vigorously shaken for 10

seconds,109 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. The partition coefficient 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. Due to the high

solubility ratio exhibited by compound 7e, it was not possible to accurately determine its

partition coefficient using this procedure. Therefore, in this case KIPM:4.0 WAS estimated

from the average methylene XnK Obtained for compounds 7a-d according to the following

relationship

log Kn +m (RK)(m) + l0g Kn (4)

where n is the number of methylene units in the promoiety of one prodrug and m is the

number of additional methylene units in the promoiety with which it is compared.

UV spectrophotometry was also used to determine the amount of 6a and prodrug

present in the receptor phase of the diffusion cell. Since all the prodrugs in this study

were part of a homologous series, it was assumed that satisfactory results would attain for

the entire series from the use of the molar absorptivity of one homolog. Thus, the molar

absorptivities of compounds 7c and 6a were determined in pH 7.1 phosphate buffer (0.05

M, I = 0. 11 M) containing 0. 11% formaldehyde by first dissolving a known amount of









either compound in acetonitrile. An aliquot (0.500 mL) of the acetonitrile solution was

removed, diluted with buffer, and analyzed by UV spectrophotometry to obtain the molar

absorptivities shown in Table 3-4. Because there is considerable overlap between the

UV spectra of APAP and its ACOM prodrugs 7a-e, the relative concentrations of each

were determined using the following approach. The differences in absorption were found

to be greatest at 240 nm and at 280 nm. Therefore, considering the additive nature of

absorption, the absorbance at each wavelength (assuming constant cell length) is

A240 EP240CP EA240CA (5)

A280 EP280CP EA280CA (6)

where A is the absorbance at the respective wavelengths, E is the molar absorptivity of

either the prodrug (P) or APAP (A) at the respective wavelengths, and C is the

concentration of the respective compounds in the mixture. Solving the two simultaneous

equations gives the following solution for the prodrug concentration CP

CP (A280A240 E .':~-'I 2 A280EP240 EA240EP280) (7)

Once CP is known, it may be inserted into equation 5 to give the following solution for

the concentration of APAP CA:

CA = (A240 EP240 P)/ EA240 (8)

Solubility parameters. Solubility parameters were calculated by the method of

Fedorsllo as demonstrated by Martin and coworkers "l and Sloan and coworkers.112

Diffusion cell experiments

The flux of each prodrug was measured using skin samples from three different

mice. 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 (2.7 ml of 36% aqueous formaldehyde/liter) to inhibit microbial


growth and maintain the integrity of the skins113 throughout the experiment. A rubber O-


ring was placed on top of the skin to ensure a tight seal, and the donor and receiver


compartments were fastened together with a metal clamp (see Figure 3-9).




Open to
Atmosphere



Sampling
Arm
Suspension of
Donor Drug or Prodrug
Compartment
Rubber
O-ring
Skin


Buffer


Receiver
Compartment

Water Out
Water InWater Jacket



Magnetic
Stir Bar




Figure 3-9. Diagram of Franz Diffusion Cell (Metal Clamp Not Shown)

Prior to the application of the prodrug, the skins were kept in contact with buffer


for 48 h to allow any UV absorbing material to leach out. During this time, the receptor


phase was removed and replaced with buffer 3 times in order to facilitate the leaching


process. Twenty four hours before application of the prodrug, a suspension (0.09 M to


0.80 M, i.e. roughly 10 x SIPlu) Of the prodrug in IPM was prepared and allowed to mix









until it was needed in the diffusion cell experiments. 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 (Table 3-4, equations 7 and 8) to

determine the amounts of permeated APAP and prodrug. 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 skins were kept in contact with buffer for an

additional 24 h to allow all APAP species (i.e. APAP and prodrug) to leach from the skin.

Following this second leaching period, the receptor phase was replaced with fresh buffer

and an aliquot (0.5 ml) of a standard drug/vehicle (theophylline/propylene glycol) was

applied to the skin surface: the second application period. Samples of the receptor phase

were taken at 1, 2, 3, and 4 h and analyzed by UV spectrophotometry. The concentration

of theophylline in the receptor phase was determined by measuring its absorbance at 270

nm (8 = 10,200 L mol )~. 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 species (APAP plus prodrug) against time as shown by the example in Figure 3-

10. Flux could then be calculated by dividing the slope of the steady-state portion of the

graph by the surface area of the skin (4.9 cm2)











80-
a. 70-
60 -( y = 3.5788x 51.368
E 0 R2 =0.9943

E 40-
2u a 30-



10

0 10 20 30 40
Time (h)

Figure 3-10: Flux of Compound 7a through Hairless Mouse Skin

Results and Discussion

Physicochemical properties

The solubilities of compounds 7a-e in IPM (SIPM) and in water (SAQ) are Shown in

Table 3-4. The relative standard deviations were all I & 5% except for the SAQ ValUe

measured for compound 7e (A 9%). As expected, all the prodrugs were more soluble in

IPM than APAP (Table 3-5). Although there was a thirteen fold range in SIPM between

the first and last member of the series, there was little variation in SIPM between the

second and last member of the series. The biggest increase in SIPM (7 fold) occurred on

going from the first (C1) to the second member (C2) of the series. Beyond C2, SIPM

gradually increased until the fourth member of the series (CS), but began to decrease

thereafter. It is reasonable to anticipate a "point of diminishing returns" where no further

increases in lipid solubility are realized by extending the length of the alkyl chain.

Typically, the increase in lipid solubility exhibited by the first member of a series of

prodrugs or analogues results from masking a hydrogen bond donor in the parent

compound. Elimination of the offending functional group results in a compound with









lower crystal lattice energy than the parent, and is thus more easily solvated. For a

homologous series in which the only difference between members is the length of an

aliphatic chain, lipid solubility will increase as the chain is extended due to the

incorporation of lipophilic groups. However, at a certain point van der Waals

interactions between the aliphatic chains become dominant, causing an increase in

melting point and a decrease in lipid solubility. In general, the trends in SIPM for 7a-e

appear to follow the trends in melting point, though there was less variation in melting

point among 7b to 7e than there was in SIPM. It is important to note that the trends in SIPM

shown here were observed previously in other prodrug series including 1-ACOM-5-

fluorouracil (1-ACOM-5U),67 3-ACOM-5-FU,69 1-AOC-5-FU,114 and bis-6,9-ACOM-6-

mercaptopurine (6,9-ACOM-6-MP).39

In addition to the 4-ACOM-APAP series, physicochemical data from a recently

described series of alkyloxycarbonyl (AOC) derivatives of APAP (Figure 3-1 1) is also

listed in Table 3-5. If homologs of the same alkyl chain length are compared (7a to 7c

versus Si to 8k), the ACOM derivatives all exhibit lower melting points and, with the

exception of 7a (C1), are more soluble in IPM and water than the corresponding

members of the AOC series. However, comparisons such as this do not take into account

the structural differences between the promoieties in question. In order to make

comparisons between homologs of approximately equal size, it is perhaps more

appropriate to consider the fact that members of the ACOM series contain a CH20 spacer

between the phenoxy group of APAP and the carbonyl of the promoiety which extends

the alkyl chain further from the phenyl ring of the parent. Disregarding the differences in

size between a methylene unit and oxygen, the Cl member of the ACOM series should










be compared to the C2 member of the AOC series. If similar comparisons are made for

the remainder of the two series, the ACOM prodrugs are 4 to 17-times more soluble in

water and, with the exception of 7a, are 3 to 5-times more soluble in IPM than the

corresponding members of the AOC series.

Table 3-5: Physicochemical Properties of 4-Hydroxyacetanilide 6a, 4-ACOM-APAP
Prodrugs 7a-e and 4-AOC-APAPa Prodrugs 8i-m
Compound MWb mp oC' SIPhid, e, f SA~dJ; g S4.0d, h KIPM:4.0
6a, APAP 151 167-170 1.9" 100"
7a, C1 223 92-95 8.41 & 0.44 15.2 & 0.34 16.2 0.52 &
0.016
7b, C2 237 56-59 62.0 & 1.91 24.7 & 0.33 26.6 2.33 &
0.039
7c, C3 251 56-58 73.5 A 1.45 7.12 & 0.0073 8.26 8.90 & 1.00
7d, C5 279 50-52 109 & 1.48 0.597 & 0.018 0.90 121 + 19.1
7e, C7 307 53-54 98.7 & 3.77 0.0637 & 0.0060 0.048 2077 "
8i, C1 209 112-115 12.0 20.4 17.0 0.692
8j, C2 223 120-122 9.33 3.80 4.47 2.09
8k, C3 237 104-106 23.4 2.70 3.02 7.94
81, C4 251 118-120 13.8 0.427 0.447 31.6
8m, C6 279 108-110 16.7 0.0479 0.0324 513
SData from Reference 45. Molecular weight. Melting point (uncorrected). dUnits of mM.
Solubility in isopropyl myristate (IPM). Mrueasured at 23 A 1 oC. Solubility in water. Solubility
in pH 4.0 buffer estimated from SIPM/KeIPMn. Partition coefficient between IPM and pH 4.0
acetate buffer. Extrapolated from previous KeIPMn in the series as described in the text.


Although 7a-e were 4 to 60 times more lipid soluble than APAP, they were all

much less soluble in water than APAP. In fact, the most water soluble member of the

series, C2, exhibited only one-fourth the aqueous solubility of APAP (Table 3-5). SAQ

increased on going from Cl (7a) to C2 (7b), but dropped off quickly as the alkyl chain

length increased. Interestingly, the C2 member was also the most water soluble member

of the 1-ACOM-5-FU67 and 3-ACOM-5-FU69 prOdrug series. Contrary to its effect on

SIPlu, maSking a hydrogen bond donor in the parent compound can often lead to lower

SAQ relative to the parent. Such was the case in the present study and in previous prodrug









series including 7-ACOM-theophylline (7-ACOM-Th),5 1 -alkylaminocarbonyl-5-FU (1-

AAC-5-FU),"'S and 4-AOC-APAP.45



R O NI~

4-AOC-APAP

8 i, R= OCH3
8 j, R =OC2H,
8 k, R = OC3H,
8 I, R =OC4H,
8 m, R= OC,H13

Figure 3-1 1: Structure of 4-alkyloxycarbonyl (AOC) derivatives of APAP

In order to incorporate the physicochemical property data for 7a-e into the Roberts-

Sloan database,20 pH 4.0 buffer was used as the aqueous phase in determinations of

partition coefficients. Partition coefficients obtained in this manner were then used to

estimate the solubilities of 7a-e in pH 4.0 buffer (S4.0, Table 3-5). Partition coefficients

between IPM and pH 4.0 buffer (KIPM:4.0) were experimentally determined for all

compounds except for 7e (Table 3-5). The relative standard deviations in KIPM:AQ were

all less than & 10% except for 7c (a 11%) and 7d (A 16%). Although the average

methylene XnK for the 4-ACOM-APAP series (0.60 + 0.05) is somewhat higher than the 4-

AOC-APAP series (0.55 & 0.06), it is consistent with average methylene XnK ValUeS SOCH

in other ACOM prodrug series: 1-ACOM-5-FU,67 RnK = 0.60 + 0. 14; 3-ACOM-5-FU,69 RnK

= 0.59 & 0.01; 7-ACOM-Th,57 XK = 0.58 & 0.05). Since the partition coefficients and XnK

values for 7a-d (Table 3-6) were reasonably well-behaved, the average XnK ValUe WaS used

to estimate the partition coefficient for 7e (Table 3-5). Use of the solubility ratios

SRIPM:AQ aS a surrogate for KIPM:4.0, TOSulted in an average methylene EnSR value that was

slightly higher than XnK, but exhibited a smaller standard deviation (0.62 & 0.03). The









estimated solubilities in pH 4.0 buffer S4.0 WeTO Somewhat higher than SAQ for 7a-c (10 &

5%), while the calculated S4.0 for 7e was only 0.75 times the experimentally measured

SAQ for 7e. Due to the relatively large difference between SIPhi and SAQ Of 7d (SIPhf/SAQ =

182), it was difficult to experimentally determine KIPM:4.0 with reasonable precision. As a

consequence, S4.0 for 7d was 1.5 times higher than its SAQ, which is somewhat greater

than the largest variation observed previously in the 4-AOC-APAP series (S4.0 Was 0.59

times the experimentally measured SAQ in the case of 4-(2'-

methoxy ethyl oxy carb onyl oxy)acetanili de) .45

Table 3-6: Log Solubility Ratios (log SRIPM:AQ), Differences Between Log SRIPM:AQ
(EnSR), Log Partition Coefficients (log KIPM:4.0), Differences Between Log
KIPM:4.0 (HK), and Solubility Parameters (6i) for Prodrugs 7a-e
Prodrug log SRIPM:A~a RCSRb lOg KIPM:4.0c SCKd 6ie
7a -0.257 -0.285 12.04
7b 0.400 0.66 0.368 0.65 11.77
7c 1.01 0.61 0.949 0.58 11.54
7d 2.26 0.62 2.09 0.57 11.18
7e 3.19 0.57 3.32' 10.89
" Log of the ratio of the solubilities in IPM (SIPM) and water (SAQ). b ESR (l0g SRn.m log
SRn)/m: n is the number of methylene units in the promoiety of one prodrug and m is the number
of additional methylene units in the promoiety with which it is compared. Log of the partition
coefficient between IPM and pH 4.0 buffer. d Same definition as in b with the exception that log
KIPM 4On is used in place of log SRes AQ. e Calculated as described in Reference 1 12 (units (cal
cm ) /. f Extrapolated from previous KIPM 4On in the series as described in the text.


Diffusion cell experiments

To date, there has been only one report of the topical delivery of 4-

hydroxyacetanilide (APAP) by a homologous series of prodrugs.45 In Order to facilitate

comparisons between the results of the present investigation to those of the prior study of

4-alkyloxycarbonyloxyacetanilide derivatives (4-AOC-APAP), data from both prodrug

series are listed in Table 3-7. As shown in Table 3-7, the fluxes (A SD) of the ACOM

prodrugs with the exception of 7e (A 32%) were within the typical45 & 30% variation of in









vitro experiments with hairless mice. Three of the five members of the ACOM series

were more effective at delivering APAP through the skin than APAP itself. This is in

contrast to the AOC series in which only one member (C1) permeated the skin better than

APAP. If comparisons are made between members of the same alkyl chain length (7a to

7c versus Si to 8k), the ACOM derivatives are, with the exception of 7a, 2 to 11i-times

more permeable than the corresponding members of the AOC series. The flux of the

most permeable derivative 7b was 3.6 times greater than that of APAP. An improvement

of this magnitude is modest when compared to the results of other prodrug series. For

instance, 6-ACOM derivatives of 6-mercaptopurine (6-MP)116 and 1-ACOM derivatives

of 5-fluorouracil67 improve the flux of the parent by as much as 69 and 16 times,

respectively. The apparent ineffectiveness of the ACOM promoiety in the present case

may be explained by considering the differences in the physicochemical properties of the

parent compounds. Compared to APAP, 5-FU and 6-MP are much less soluble in IPM

and water. Thus it is not surprising to find that the flux of APAP is two fold higher than

the flux of 5-FU and 134 times greater than that of 6-MP. As a consequence of its

relatively high SIPlc and SAQ ValUeS, it is more difficult to improve the flux of APAP than

it is to improve the flux of polar heterocycles such as 5-FU and 6-MP. It is also worth

mentioning that the 7-ACOM derivatives of theophylline (Th),57 a polar heterocycle,

exhibited only modest (2 fold) improvements in flux. Though Th is less soluble in lipid

and aqueous solvents than APAP, it is 7 times more soluble in IPM than 5-FU while still

exhibiting 54% of the water solubility of 5-FU. Again, the better the biphasic solubility

of the parent compound, the more difficult it is to improve the flux via a prodrug

approach.









When the receptor phases from the application of 7a-e were analyzed during

steady-state flux conditions, only APAP was found. The exception was compound 7b in

which the intact prodrug accounted for 9% of the total APAP species in the receptor

phase (Table 3-8). Since this particular derivative was also the most permeable member

of the series, the system of cutaneous esterases in this case may have been overwhelmed

and unable to completely hydrolyze the prodrug on its way through the skin. A similar

phenomenon was observed in the 4-AOC-APAP series45 in which the derivative that

exhibited the highest flux also delivered the highest percentage of intact prodrug through

the skin (Table 3-8). Although no effort was made to determine the half-lives of 7a-e in

the receptor phase buffer, the aqueous stability may be estimated based on similar studies

by others.93, 117 For example, Bundgaard and coworkers93 found that the 2-

acetyloxymethyl and 2-butyrloxymethyl derivatives of salicylamide exhibit half-lives of

46 and 98 h, respectively at 37 oC in pH 7.4 buffer. Others have found that 4-

hexanoyloxyacetanilide displays an approximate half-life of 19 hours at 37 oC in pH 7.8

buffer." Given the generally higher pKa of an aryl hemiacetal compared to its

corresponding phenol, the ACOM derivatives 7a-e should exhibit half-lives greater than

19 hours under the present experimental conditions. Thus, it is reasonable to assume that

the absence of intact prodrug in the receptor phase is due to extensive enzymatic

hydrolysis in the skin and is not the result of substantial chemical hydrolysis in the

receptor phase.

Apparently, the fluxes of 7a-e are not artificially high due to damage sustained by

the skin over the course of the experiment. This assessment is based on control

experiments in which a suspension of theophylline in propylene glycol (Th/PG) was










applied to the skin following the removal of the prodrug donor phase. This second

application of Th/PG resulted in Th flux values that were not significantly different from

those through skins treated with IPM alone (Table 3-7). However, it is important to

recognize that IPM is a well-known penetration enhancer which can increase flux 50-fold

compared to experiments where water was the vehicle.ll Although the apparent flux

values of 7a-e are likely inflated due to IPM, this is not expected to change the rank order

of flux within or between series.ll

Table 3-7: Flux of Total APAP Species through Hairless Mouse Skin from Suspensions
of 4-ACOM-APAP and 4-AOC-APAPa Prodrugs in IPM (Jhl), Second
Application Flux of Theophylline through Hairless Mouse Skin from a
Suspension in Propylene Glycol (JJ), Error in Predicting Log Jhl using the
Roberts-Sloan Equation (A log Jpredlctea), Error in Calculating Log Jay using the
Roberts-Sloan Equation (A log Jcalculatea), and Ratio of the Flux of the Prodrug to
the Flux of APAP (Jnrodrug / APAP).
Compound J~lb bJ lOg 1sib A lOg A lOg Jprodrug
Jpredictedc Jcalculatedd JAPAP
6a, APAP 0.51" 0.74" -0.29" -0.496e -0.484
7a, C1 0.730 + 0.23 0.934 0.136 -0.136 -0.104 -0.0911 1.4
7b, C2 1.86 & 0.24 0.935 0.0764 0.270 -0.213 -0.197 3.6
7c, C3 0.777 & 0.20 0.780 0.224 -0.109 -0.350 -0.331 1.5
7d, C5 0.344 & 0.062 0.857 0.148 -0.464 -0.254 -0.231 0.67
7e, C7 0. 110 + 0.028 0.687 0.147 -0.957 -0.0366 -0.00703 0.22
8i, C1 1.00 1.12 0.00 -0.0953e -0.0794 2.0
8j, C2 0.174 0.64 -0.76 -0.482e -0.464 0.51
8k, C3 0.355 1.14 -0.45 -0.260e -0.240 0.69
81, C4 0.0977 0.85 -1.01 -0.264e -0.241 0.20
8m, C6 0.0324 0.76 -1.49 -0. 162e -0.133 0.063
Control' 1.02 & 0.13g
SFrom Reference 45. Units of Clmol cm h~ Predicted from log Jay = -0.497 + 0.519 log SIPM +
0.481 S On 0.00268 MW (coefficients from n = 61 database, Reference 45, were recalculated
using SAS 8.1). Error in prediction = log Jay predicted log JM. d Calculated from log Jay -0.545
+ 0.5 11 log SIPM + 0.489 S On 0.00253 MW (n = 61 + current data gives a new database of n
66 compounds). Error in calculation was from log Jay calculated log JM. e Already included in
the n = 61 database, so the value listed here is actually the difference between log Jay and a
calculated value for flux, log Jcalculate. fSkins were sequentially subjected to 48 h conditioning, 48
h contact with IPM, methanol wash, 24 h leaching. From Reference 112.

If the fluxes of 7a-e are normalized by their respective solubilities in IPM, the

corresponding permeability coefficients Phl are obtained (Table 3-8). Phl has units of










distance per time (usually cm h- ) and is thus a measure of how quickly a compound

diffuses through the skin. Because Phl gives no indication of the amount, or dose, of the

permeant that is entering the body, it is not clinically useful apart from the appropriate

solubility data. Nevertheless, Phl is frequently used in the literature to quantify the

permeation efficiency of a compound through skin.5~I One of the most popular

expressions of Phl, the Potts-Guy equation (9),42 Shows that Phl is positively dependent on

the octanol-water partition coefficient (KOCT:AQ) and negatively dependent on molecular

weight (MW):

log Phl = -6.3 + 0.71 log KOCT:AQ 0.0061 MW (9)

Table 3-8: Percent Intact Prodrug Detected in Receptor Phase during Steady-State (%
Intact), Log Permeability Coefficients (log Phl), Concentrations of APAP
Species in Skin (Cs), and Dermal/Transdermal Delivery Ratios for APAP 6a,
4-ACOM-APAP 7a-e, and 4-AOC-APAP Prodrugsa 8i-m
Compound % Intactb lOg P1Lc Cd D/Te
6a, APAP -0.571 2.74 0.70f 0.046
7a, C1 0 -1.06 2.67 0.572 0.031
7b, C2 9 -1.52 13.1 2.10 0.060
7c, C3 0 -1.98 5.56 0.535 0.061
7d, C5 0 -2.50 3.55 1.05 0.088
7e, C7 0 -2.95 2.72 1.55 0.21
8i, C1 64 -1.08 5.45 1.57f 0.046
8j, C2 14 -1.73 1.08 0.13f 0.053
8k, C3 25 -1.82 2.84 1.44" 0.068
81, C4 0 -2.15 1.91 0.08f 0.17
8m, C6 0 -2.71 1.79 0.43f 0.47
a From Reference 45. b Percent intact prodrug detected in the 31 h receptor phase sample.
Calculated from log JM log SIPM, units of cm h-'. d Amount of total APAP species (in units of
Clmol) in receptor phase after 24 hours following donor phase removal to allow APAP and
prodrug to leach out of skin. e Calculated from D/T = [(Cs/4.9 cm2 24 h)]/JM. fFrom Reference
119.
Such a relationship suggests that percutaneous absorption is positively dependant

on lipid solubility and negatively dependant on the water solubility of a permeant.

However, a plot of the log Phl values for 7a-e versus their respective log KIPM:4.0 ValUeS

gave a negative slope (-0.519, r2 = 0.975, plot not shown). Similarly, a plot of log Phl










versus the calculated solubility parameters of 7a-e gave a positive slope (Figure 3-12),

demonstrating an inverse relationship between log Phl and alkyl chain length (i.e. higher

SIPM, l0WeT F6i). These results are consistent with the findings of others45, 69, 118 and

support the idea20 that lipophilicity alone is not a good predictor of flux.


-1.0-

-1.5-

-2.0-

-25 y =1.6415x -20.853
o R2 = 0.9974
-3.0-

-3.5-

-4.0
10.8 11 11.2 11.4 11.6 11.8 12 12.2
Solubility Parameter


Figure 3-12: Plot of Solubility Parameter versus Log P for 4-ACOM-APAP Prodrugs
7a-e

To further illustrate the relatively weak dependence of flux on lipid solubility,

consider the SIPM and SAQ ValUeS for APAP prodrugs 7a-e and 8i-m (Table 3-5).

Although compound 7c is 6.1 times more soluble in IPM than Si, compound Si is 2.9

times more soluble in water than 7c. This increase in water solubility on going from 7c

to 8i, though modest, resulted in 1.3 fold greater flux for Si compared to 7c. The impact

of SAQ On flUX is more distinct when comparisons are made between individual members

of a series. For instance, 7e is 1.6 times more soluble in IPM than 7b, but 7b is 388 times

more soluble in water. As a result, the flux of 7b is 17 times greater than the flux of 7e.

In order to ascertain the relative impact of solubility in a lipid, solubility in water, and

partition coefficient on flux, the trends in SIPM, S4.0, KIPM:4.0, and Jhl for APAP 6a and its









prodrugs 7a-e and 8i-m are graphically represented in Figure 3-13 (a Wasdo plot).119

What is clear from such a representation is that KIPM:4.0 iS of little positive predictive

value in determining the rank order of flux. For each increase in alkyl chain length, there

is a corresponding increase in KIPM:4.0 regardless of the trends in Jhl. It is interesting to

note that while similar observations have been made by others,"" the idea that partition

coefficient is predictive of flux120 remains an erroneous yet persistent concept. A similar

conclusion may be reached by examining the trends in SIPM. Within the AOC series and

to a lesser extent in the ACOM series, the trends in SIPM are relatively flat across the

series despite the fact that Jhl grows progressively smaller. In contrast, the trends in S4.0

generally mirror the trends in Jhl across a series. Although such trends imply that water

solubility is a better predictor of flux than lipid solubility, the reality is that flux is best

predicted when both properties are considered.43 This is demonstrated in the present case

by the fact that the most permeable members of both series (7b and Si) exhibit the best

mixture of high SIPM and high S4.0. Such behavior is no doubt related to the biphasic

nature of the absorption barrier presented by the stratum corneum (see Chapter 1).

Although it is obvious that flux is positively dependent on lipid and aqueous

solubility, there is currently only one mathematical model available for quantifying such

a relationship (see Chapter 1):

log Jhl = x + y log SIPM + (1 y) l0g S4.0 z MW (10)

log Jhl = -0.491 + 0.520 log SIPM + 0.480 log S4.0 0.00271 MW (11)

Equation 10, or the Roberts-Sloan (RS) model,43 was originally based on a database (n =

42) of 7 different series of prodrugs of polar heterocycles. This database was recently

updated45 to include two new series of heterocyclic prodrugs and one new series of










phenolic prodrugs (4-AOC-APAP) resulting in a more structurally diverse database of 61

compounds. A fit of that data to equation 10 gave the form of RS expressed by equation

11.45 In its present state, the model is heavily dependent on data from heterocyclic

compounds: 59% 5-FU related entries, 18% 6-MP related entries, and 10% Th related

entries in the database. Only 8 of the 61 entries (13%) are of a phenolic compound (i.e.

APAP). Therefore, it was of interest to determine whether equation 11 could accurately

predict the flux of the 4-ACOM prodrugs 7a-e of APAP. Application of equation 11 to

prodrugs 7a-e resulted in predicted flux values (Jpredicted, data not shown) that were

consistently higher than the experimentally determined fluxes (Jhl). The differences

between log Jhl and log Jpredicted (A log Jpredicted) foT 73-e arT liSted in Table 3-7. On

average, the error in predicting log Jhl (A log Jpredicted) foT 73-e WaS 0.192 + 0.124 log

umits.














-1





6a 7a 7b 7c 7d 7e 8i 8j 8k 81 8m
Compound
Figure 3-13: Log SIPM (O), Log S4.0 (A), Log KIPM:4.0 (0), and Log Jhl (*) Values for
APAP 6a, 4-ACOM-APAP Prodrugs 7a-e, and 4-AOC-APAP Prodrugs 8i-m.









In order to increase the diversity of the database and improve the predictive power

of RS, prodrugs 7a-e were incorporated into the database. A fit of the SIPhi, S4.0, and MW

for the resulting n = 66 entries to equation 10 gave the following estimates for x, y, and z:

x = -0.545, y = 0.511, z = 0.00253, r2 = 0.915. These parameter estimates were then used

to calculate Jhl for all 66 compounds (data not shown). A plot of Jhl versus the calculated

flux values is shown in Figure 3-14. The differences between the experimental and

calculated fluxes (A log Jealculated) for APAP 6a and its prodrugs 7a-e and 8i-m are listed

in Table 3-7. As shown in Table 3-7, the A log Jealculated for 6a, 7a-e, and 8i-m decreased

with the incorporation of the 4-ACOM-APAP data into the database. On average, the A

log Jealculated for 7a-e (0.171 & 0.126 log units) was somewhat higher than the average A

log Jealculated for the entire n = 66 database (0. 155 & 0. 118 log units), but was much lower

than the average A log Jealculated for 8i-mI (0.231 f 0. 148 log units). Interestingly, APAP

and its prodrugs all exhibit lower than expected fluxes based on the present form of RS

(Figure 3-14). In addition, the average A log Jealculated for APAP and its prodrugs (6a plus

7a-e, plus 8i-m; 0.227 & 0.133 log units) is quite a bit higher than the average A log

Calculated for the database as a whole.

In order to determine whether 4-ACOM-APAP prodrugs function better as dermal

(delivery into the skin itself) or transdermal (delivery through the skin and into the

systemic circulation) delivery agents, the skins were kept in contact with buffer for 24

hours after removing the donor phase to allow APAP and prodrugs to leach out. The

amount of total APAP species leached from the skin (Cs) is shown in Table 3-8. As

shown in Table 3-8, the rank order of Cs generally follows the rank order of flux. In

other words, the most permeable members of the series were also the most effective at







78


increasing the concentration of APAP in the skin. Three out of the five ACOM

derivatives were able to deliver more APAP into the skin than suspensions of topically

applied APAP alone, with derivative 7b delivering up to 5-times more APAP. Using the

Cs values as an estimate for the amount of total APAP species delivered into the skin,

dermal/transdermal delivery ratios (D/T, Table 3-8) were calculated from equation 12:

D/T = [(Cs/4.9 cm2 24 h)]/JM (12)





1.5
1 o
0.5 o o&


S-0.5 -ooo o~~; o'


.~ -1


-2.5
-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5
log JNIIPM = 0.545 + 0.511 log S,,, + 0.489 Iog S4.0 0.00253

Figure 3-14: Plot of Experimental Versus Calculated Flux for 5-FU, 6-MP, and Th
Prodrugs (0, n = 53), APAP (m), 4-AOC-APAP Prodrugs (*, n = 5, plus two
additional compounds mentioned in Reference 1 to give n = 7), and 4-ACOM-
APAP Prodrugs (A, n = 5)

Most of the prodrugs exhibited D/T ratios that were higher than APAP. Thus,

compared with topically applied APAP alone, all but one of the ACOM prodrugs (7a)

were more effective at delivering APAP to the skin itself rather than through it. Among

7a-e, the prodrugs that preferentially delivered more APAP into the skin itself were also

the most lipophilic and least permeable members of the series. Thus, compounds such as

7d and 7e are best suited for a therapeutic regimen involving sustained delivery of low









levels of a drug, while the shorter chain derivatives would allow for maximum exposure

of the drug to the systemic circulation.

Conclusions

Despite the success of ACOM prodrugs in improving the transdermal delivery of

heterocyclic drugs, there are currently no examples of this approach being applied to a

phenol. The results presented here demonstrate for the first time that ACOM derivatives

are capable of improving the topical delivery of a phenol. In general, the ACOM

derivatives of acetaminophen (APAP) exhibited better biphasic solubility and lower

melting points than the previously studied45 AOC derivatives. As a result, the 4-ACOM-

APAP prodrugs were capable of improving the delivery of acetaminophen by 4-fold. The

trends in flux were found to depend on a balance between lipid and aqueous solubility.

Addition of the 4-ACOM-APAP prodrugs to the Roberts-Sloan database increased the

structural diversity of the current database and resulted in a more robust RS model.

Given that all of the 4-ACOM-APAP derivatives contained simple aliphatic groups in the

acyl chain, it is likely that even greater improvements in flux will be realized by

incorporating more hydrophilic functional groups into the acyl chain.20














CHAPTER 4
ALKYLOXYCARBONYLOXYMETHYL (AOCOM) PRODRUGS OF
ACETAMINOPHEN (APAP)

Synthesis of AOCOM Prodrugs of 4-Hydroxyacetanlide (APAP)

To date, there has been only one report in the literature of the synthesis of an

AOCOM derivative of a phenol.121 In that study, Seki and coworkers arrived at the target

AOCOM compound (4-ethyloxycarbonyloxymethyloxyacetanilide by way of a four-step

synthetic route starting from methyl chloroformate (Figure 4-1, R = C2H5). At the time

of Seki's investigation, one of the key reagents, chloromethylchloroformate 16, was

commonly synthesized via the chlorination of methyl chloroformate.84, 122 This method

requires fractional distillation of the product mixture to obtain pure 16 and often provides

low yields of the desired product. Currently, chloroformate 16 may be purchased from

several suppliers and it is no longer synthesized in the lab on a regular basis.123 Since the

AOCOM and ACOM promoieties are structurally similar, it was of interest to determine

whether the same strategy that was used to synthesize ACOM iodides (Chapter 3) could

be used to eliminate the use of 16 (and 4, R = Oalkyl) altogether (alternative routes

shown in Figure 4-1 starting from la). In keeping with this strategy, chloroformates were

allowed to mix with trioxane and Nal at room temperature. Unfortunately, no reaction

was observed at room temperature, and at higher temperatures the chloroformate

apparently underwent decarboxylation as indicated by the generation of gas. Various

Lewis acid / Nal mixtures also failed to result in product. If a catalytic amount of









pyridine was added, approximately 70% of the chloroformate was converted to alkyl

iodidel24 even at room temperature.



APAFO--CH20 R



RCOCI / Nal f

O~~ O t ICH20 RC.,O CI + SO2 2,
la 3 14 15


RCOCI zO


CICH20 R RH CIH0 C
4 16


Figure 4-1: Synthetic Routes to Alkyloxycarbonyloxymethyl (AOCOM, R = Oalkyl)
Prodrugs of 4-hydroxyacetanilide (APAP)

An alternative two-step route to AOCOM iodides involving an intermediate

AOCOM chloride 4 (R = Oalkyl) was also attempted in order to avoid purchasing the

relatively expensive 16 (Figure 4-1). There are a few reports in the literature that suggest

such an approach is feasible.81, 125 For example, ethyloxycarbonyloxyethyl chloride had

been synthesizedsl previously in good yield (48%) by reacting acetaldehyde with ethyl

chloroformate in the presence of a catalytic amount of ZnCl2. Yet this method failed to

work in the present case where the aldehyde is the formaldehyde trimer trioxane la.

Furthermore, although certain AOCOM alkyl halides can be synthesized from a

monomeric aldehyde and chloroformate in the presence of a pyridine catalyst, 125 this

method also failed in the present investigation.









As it was not possible to neither shorten the synthesis of AOCOM iodide 3 nor

make the corresponding chloride 4 (R = Oalkyl) parsimoniously, 3 (R = Oalkyl) was

obtained through a two-step process starting from 16 (Figure 4-1). With 3 (R = Oalkyl)

in hand, the target prodrugs could be obtained by coupling 3 (R = Oalkyl) with APAP.

The coupling reaction used by Sekil21 was an adaptation of the method of Sloan and

Koch for the synthesis of ACOM ethers of phenols.90 Under those conditions (K2CO3 aS

base, acetone as solvent), Seki and coworkers noted that 4-

ethyloxycarbonyloxymethyloxyacetanilide was obtained in 20 % yield from the coupling

of APAP with ethyloxycarbonyloxymethyl iodide following a reaction time of seven

days. In an effort to improve the yield and ascertain the reaction parameters by which

this reaction is governed, a series of AOCOM derivatives of phenols (with emphasis on

APAP) was synthesized by the method of Sekil121 and by a more efficient method

involving phase-transfer catalysis. In addition, the results were compared to those

obtained from the coupling reaction of ACOM halides with phenols--an analogous

system whose reaction parameters are known (see Chapter 3).90




R10 X +" HO-~ R'R OR'+ OR
3 6 7 8
X =CI, Br, I
R =Oalkyl, alkyl
R' = Alkyl, aromatic, etc.


Figure 4-2: Generalized Reaction of AOCOM halides (R = Oalkyl) and ACOM halides
(R = alkyl) 3 with phenols 6

In the present investigation, 4-hydroxyacetanilide (APAP) was chosen as a model

phenol in order to make a direct comparison between this work and the work of Seki.121

In addition, if the reactivity of AOCOM halides was found to parallel that of ACOM











halides,90 then the reaction mixtures were expected to contain various percentages of


acylated phenol 8 as a byproduct (Figure 4-2). In the present case, since the carbonate

derivatives of APAP had been characterized previously,45, 126 adoption of this particular


phenol as a model facilitated byproduct identification. As shown in Figure 4-3, AOCOM

iodides may be obtained from the corresponding chlorides via halogen exchange in


acetone, preferably in the presence of sodium bicarbonate to neutralize traces of HI


formed during the reaction.84 Subsequent reaction with phenols under the standard


conditions acetonitrilee or acetone as solvent, K2CO3 aS base)90, 121 Or in a biphasic system


in the presence of tetrabutylammonium hydrogen sulfate (Figure 4-3) gave mixtures of 7


and 8.


RH CI O CI 2ridie R O I

17 16 4
a: R =OCHz h: R =OCH, (89%)
b: R =OC2H, i: R =OC2H, (92%)
c: R= OCH7 j: R= OCH7 (93%)
d: R =OC,H17 k: R =OC,H17 (98%)
e: R =OCloH21 1: R =OCloH21 (90%)
f : R =Oi-Pr m: R =i-Pr (85%)
g: R =Ot-Bu n: R =t-Bu (59%)



K2COI Bu4NHSO4
CI2CH2 / Water O
70-100% a R O OY +

6a: Y =NHCOCH,
6b: Y =H 7
6c: 2,2,5,7,8-pentamethyl-
chroman-6-ol i: R= OCHz, Y =NHCOCHz
j: R =OC2H,, Y =NHCOCH3
k: R =OCzH7 Y =NHCOCH3
1: R =OC,H17 Y =NHCOCH3
m: R =OCloH21 Y =NHCOCH3
n: R = i-Pr, Y = H
o: R =t-Bu, Y = H
p: R =OCH, (phenol = 2c)


h: R =OCHz (86%)
i: R =OC2H, (90%)
j: R= OCzH7 (72%)
k: R =OC,H17 (96%)
1: R =OCloH21(9%
m: R =i-Pr (92%)
n: R =t-Bu (87%)


NaceINaneCO,


i: R= OCH,, Y =NHCOCH,
j: R =OC2H,, Y =NHCOCH3
k: R =OCH7 Y =NHCOCH3
n: R =OC,H17 Y =NHCOCH3
o: R =OCloH21 Y =NHCOCH3
p: R = i-Pr, Y= H
q: R =t-Bu, Y= H
r: R =OCHz, (phenol = 2c)


Figure 4-3: Reaction of AOCOM iodides with phenols under phase-transfer conditions









Previously, Sloan and Koch90 had shown that the coupling of ACOM halides with

phenols is sensitive to the nucleofugicity of X, with better leaving groups giving more

alkylated product 7. Recently, others91 have suggested that the ratio 7/8 is also dependent

on the steric hindrance of the acyl group (R group in 3). Although the data presented in

Table 4-1 is not exhaustive, it suggests that the trends observed in reactions of ACOM

halides with phenols are operative in the analogous reactions of AOCOM halides. For

example, if X is a poor leaving group, 8 is favored, but as the nucleofugicity of X

increases, the product distribution shifts toward 7 (compare entries 1 and 2 with 4). For

X = I, alkylated phenol 7 becomes the maj or product when the alkoxy chain length

extends beyond OCH3. Interestingly, the ratio 7/8 when R = OCH3 inCreaSes by more

than 3 fold when the reaction is carried out under phase-transfer conditions instead of the

standard protocol (entry 3 versus entry 6). Under these conditions, there is an

incremental increase in the percentage 7 with increasing steric hindrance (as measured by

Charton's steric parametersl27) in R (entries 6-8), but beyond propyloxy, the percentage

of 7 remains fairly constant for the straight chain derivatives studied. However, the

product distribution shifts entirely toward 7 on going to more bulky R groups (entries 11-

12). On the other hand, the percentage of 7 may be increased even for sterically

unhindered R if the phenol is sufficiently hindered (entry 6 versus entry 13). This

particular result (entry 13) is not without precedent since others91 have observed a similar

trend in reactions of ACOM halides with phenols. Aside from its effect on product

distribution, the advantages of the phase-transfer reaction include shorter reaction times

(one day) and higher overall yield compared to the method of Seki.121 Although no

mention was made of product distribution, it is also worth noting that Wolff and










Hoffmannl28 have used a similar reaction system to successfully alkylate phenols with

cyclic ACOM halides.

Table 4-1: Product Distribution of the Reaction of RCO2CH2X 3 with Phenols 6 Under


Various Reaction Conditions
Entry R X Phenol


Solvent


Base Distribution
(%)"
7 8


1 OCzH5 [MeNC4Hs]+


6a acetonitrile


MeNC4He

KzCO3
KzCO3
KzCO3

KzCO3
KzCO3

KzCO3

KzCO3

KzCO3

KzCO3

KzCO3
KzCO3
KzCO3


0 100
(28)"
3 58
36 64
57 43
(17)" (13)"
58 42
66 34
(18)" (6)"
74 26
(50)" (13)"
84 16
(43)" (6)"
82 18
(45)" (3)"
78 22
(41)" (6)"
100 0
100 0
90 10
(33)" (0)"


OCzH,
OCH3
OCzH,

OC4H9
OCH3


acetonitrile
acetonitrile
acetone

aCetOne
Cl:CH2/HzO


0.36d


0.58d




0.56

0.61

0.56'


7" OCzH,

8" OC3H,

9" OCSHI,

10" OCl0H21

11" O-i-Pr
12" O-t-Bu
13" OCH3


6a ClaCH- H-O

6a ClaCH- H-O

6a ClaCH- H-O

6a ClaCH- H-O


Cl:CH2/HzO
ClaCH- H-O
ClaCH- H-O


14# CH3 I 6b acetonitrile KzCO3 63 37 0.52h
15" CzH5 I 6a acetonitrile KzCO3 59 31 0.56h
16# C3H7 I 6a acetonitrile KzCO3 73 24 0.68h
17" C5H11 I 6a acetonitrile KzCO3 66 27 0.68h
18# C7H15 I 6a acetonitrile KzCO3 71 27 0.73h
a Determined from 'H NMR spectrum of the crude reaction mixture. b Charton's steric parameter
for R. Isolated yield. d Reference 127. e Reaction mixture includes 1 equivalent
tetrabutylammonium hydrogen sulfate. fEstimated from the relationship v = 0.406np + 0.108n, +
0.059ns 0.00839 in Charton, M. J Org. Chem., 1978, 43, 3995-4001. Data taken from Chapter
3. h Reference 94.



As discussed previously in Chapter 3, ACOM halides react with phenols under the

standard conditions to give mainly 7 as long as X is a good leaving group ( > Br). Thus,

the relatively low ratio 7/8 in the AOCOM series compared to the ACOM series

(compare entries 3-5 with entries 14-18) was unanticipated. Moreover, since the









carbonyl of a carbonate is usually less reactive than the carbonyl of the corresponding

ester,55 one might expect less acylation when R is alkyloxy (as in AOCOM) than when it

is alkyl (as in ACOM). Since the AOCOM iodides 3 (R = Oalkyl) were not purified

other than to filter off NaCl and unreacted Nal (see Experimental below), it is worth

considering whether any remaining AOCOM chloride in crude 3 affected the product

distribution. If 4 (R = Oalkyl) was reacting with 6 to any significant extent then the

percentage of acylated product 8 would have increased as the percentage of 4 increased.

In the case of entries 3, 4, and 5, the percentages of unreacted AOCOM chloride 4 in

crude 3 (R = Oalkyl) were 2%, 9%, and 9% respectively. Thus it does not appear that the

product distribution was affected by the presence of AOCOM chloride in crude 3 (R =

Oalkyl). On the other hand, analysis of the steric parameters for both series (ACOM and

AOCOM) suggests that differences in 7/8 between the series are directly related to

differences in the steric hindrance of R based on Charton' s steric parameters (compare

entries 3-5 to entries 14-18).94' 127 A plot of v versus the ratio of 8/7 for the entries 3-5

and entries 14-18 is shown in Figure 4-4. Although the plot of the AOCOM series

consists of only three data points, the trends in the data suggest that the coupling reaction

of AOCOM iodides with phenols is much more sensitive to steric effects than the

analogous reactions of ACOM iodides (slope = -4.9 versus slope = -0.77). A plot of v

versus 8/7 for entries 6-11 (Figure 4-5) demonstrates a much weaker dependence of

product distribution on steric effects when phase-transfer conditions are used in lieu of

the standard conditions (slope = -1.3 versus slope = -4.9).
















y = -4.9125x + 3.4107
R2 = 0.814




R2 = 0.9505


0 0.2 0.4 0.6 0.8 1 1.2 1.4
Steric Parameter


Figure 4-4: Plot of Charton' s Steric Parameter v for R Versus the Ratio of
Acylated/Alkylated Product (8/7) Resulting from the Reactions of 6 with
AOCOM Iodides (Entries 3-5 in Table 4-1, o) and ACOM Iodides (Entries
14-18 in Table 4-1, A) Under the Standard Reaction Conditions.


0.5

0.4



0.2

0.1

0.0


y = -1.2955x + 0.9766
R2 = 0.9608


0.4 0.5 0.6 0.7


Steric Parameter


Figure 4-5: Plot of Charton' s Steric Parameter v for R Versus the Ratio of
Acylated/Alkylated Product (8/7) Resulting from the Reactions of 6 with
AOCOM Iodides (Entries 6-11 in Table 4-1, o) Under Phase-Transfer
Conditions.

Conclusions

In conclusion, the data presented here suggests that steric hindrance plays a greater

role in the coupling reactions of AOCOM halides with phenols than in the analogous