<%BANNER%>

Comparison of conventional culture methods and the polymerase chain reaction for the detection of Shigella spp. on tomat...

University of Florida Institutional Repository
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 E20110109_AAAAYO INGEST_TIME 2011-01-09T23:05:26Z PACKAGE UFE0001186_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 1053954 DFID F20110109_AACQIH ORIGIN DEPOSITOR PATH warren_b_Page_018.tif GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
d3ca5a45d2487a9dadf884520525cbd9
SHA-1
8217162952b0ccec042c86e3dab134c6ef866fa6
97896 F20110109_AACQHS warren_b_Page_044.jpg
33205195f7506c3fd051fcb30a6a28e1
8fd1711edbbf22720aabdb3da0a1e566e7c4066f
615095 F20110109_AACQII warren_b_Page_054.jp2
bba4b4683c637bd03c5b42dea88ca47e
5dc1727affa56786c88b6bd56b4e67feb6568b8f
1662 F20110109_AACQHT warren_b_Page_064.txt
667fc9b206a7b3f0c69a34275f395eb0
98f37fe94d91f65e9cd0c8bc38fffe8ebc4d41d0
102860 F20110109_AACQHU warren_b_Page_043.jp2
99175a7930dd4b9dd13a00b03ed84169
ad934fcf251ce97b025ceab4208a58e83dd11e3c
95048 F20110109_AACQIJ warren_b_Page_057.jpg
466ed0cf95a9fde107e4ac138c8a7a7d
379e6f3c12c096c7d0d64cb8a94282be8ac7cdfb
100443 F20110109_AACQHV warren_b_Page_046.jpg
753cb1b8ddf4bee71d5a62671289dbe4
b4c5f81ec8f8f4a6994a81adcc40c93f7f8eec24
7754 F20110109_AACQIK warren_b_Page_096thm.jpg
8d5c25b31faf4f4c7fcf7900a5f99568
34f8bf17cfcbefd25dc431feafeeba3d2f634633
1975 F20110109_AACQHW warren_b_Page_040.txt
367b540cb60fbf85d45062e77b2365c4
4374475880d2aee78d3ab2affd7bfbf21bca41ed
7605 F20110109_AACQJA warren_b_Page_062thm.jpg
2a0d52d5451f2599cfb783b5afbf0e7b
0dd1a5063998337b2bd55b0fe11e18b566c261e7
F20110109_AACQIL warren_b_Page_081.tif
1336c0ae5a5601de81f7bf15d3f02ae9
00d1b6931e35fc9e44b730cd2dfd99ceb2b7c619
33841 F20110109_AACQHX warren_b_Page_020.QC.jpg
b148ec9db9c5b81c482caf2babd21320
1f597a422e1cc92327cd0baed427c5f8d00f650f
109956 F20110109_AACQJB warren_b_Page_070.jpg
782dcf4a0b19d62e3c4525f17c805ef8
bf8efad49317608bd1dc22ae56db61f3ecb3d489
48492 F20110109_AACQIM warren_b_Page_069.pro
d3fc367cb4acdbfdb49cb2312e3d1b17
a54166ec3c72761f5a114006cf4c98db44287408
7399 F20110109_AACQHY warren_b_Page_024thm.jpg
33beced16e4470aadf36ea4f0721aa06
5fdbc7181b6240db28a0ddb91067bd083e0b4238
36150 F20110109_AACQJC warren_b_Page_034.QC.jpg
3c36d0f820d9a30f2e892a0d9344f846
2d691e6ff125d62b5cea01173341d7939b5472ca
772216 F20110109_AACQIN warren_b_Page_061.jp2
08255f5ffa434202432be7d8d8801f2d
fa61b4a37f677ddd4d880fab075d813ff5648403
F20110109_AACQHZ warren_b_Page_046thm.jpg
703a208038ae02a619bf1ac897d464b8
0c09000b6bfe677efcd5cbfb7649e541ea40b501
51655 F20110109_AACQJD warren_b_Page_096.pro
574dd25c484c7a8b440aebd83eb1cba7
5ccda460c63f0df341cea3944228394cdd6b0ee4
5078 F20110109_AACQIO warren_b_Page_002.jpg
8bb0540a666cf147fbcefb1dc6ccf595
f889975d5a998a6e2aad432165e750531bbb6293
63898 F20110109_AACQJE warren_b_Page_054.jpg
a455b496db72e891efa2a45a8f80182f
12d7e88562bdc71626600d2e1be62610531fd69a
2554 F20110109_AACQIP warren_b_Page_102.txt
06842d2df21aabe222520ccc601765d5
7842324bae41b14cb4aedb0be09388f5a782f654
1256 F20110109_AACQJF warren_b_Page_008.txt
ed968859a0717bf5983521e86251695a
e89074bb71fb3567e75e07ddd3a5ee08b06eb125
8492 F20110109_AACQIQ warren_b_Page_026thm.jpg
1d723bdbb5bff39fbd877ff40207bdad
7de6cd7fa696d0fc5ac4b5ef8c579c2b8c078ce7
25832 F20110109_AACQJG warren_b_Page_041.QC.jpg
ced9e8e41e69715b7623ff74d38643f3
cc48e92567a95c0672e4f8a66eb9a61e90344ef7
7256 F20110109_AACQIR warren_b_Page_009thm.jpg
379cb51cab28fb4e0fb673467f04d4f4
86a1e3f871d54ba44a70816bcd45547182c4d3d0
2700 F20110109_AACQJH warren_b_Page_071.txt
21d61eff3bf59c8a17358d3d90520f5e
115522805ef071d537b19c29e6ad53026fcabc91
1944 F20110109_AACQIS warren_b_Page_025.txt
0071e935aba8b09d86688ef1ed1d91ee
ce144ff919ab3c5c900bd426f0aed6b79a0472d3
80549 F20110109_AACQJI warren_b_Page_058.jpg
16dacffbed0d427b3eee99a80041bcab
9f0a33b8c6e0837eeb32e93ce5dcab09be97c2b0
9312 F20110109_AACQIT warren_b_Page_098thm.jpg
c59b3fb6066ccd23afa2a70e5fb28d09
9be9a2fab0d6f4f014419c4ba08fac3f0b2bf9e3
F20110109_AACQJJ warren_b_Page_082.tif
9922281a4dcdc192be7ee62d80f296c1
207d1f71f3c1d150e4ceddf14544cb54dea84c3a
8536 F20110109_AACQIU warren_b_Page_048thm.jpg
c8892cc3fa8496c9a3ca3754ff44b51a
6214eedfb33f58a930f0f820f7773d47bfe068d1
F20110109_AACQIV warren_b_Page_092.tif
a6ba6a30e8290d0717e5e010205eee16
9e3f0df691b548fcbeaeacc050454e82975c6eba
26361 F20110109_AACQJK warren_b_Page_011.QC.jpg
c0b3252ae77b4fa4b06f04d9ac764981
2663cba03e502c1abd0e2ea610b0d1c06d6e00b0
8456 F20110109_AACQIW warren_b_Page_087thm.jpg
9407c9fac73e0856cbc9b24faf039c59
f27770f9785c8f8047ce26bcf1afababa1c172ab
2006 F20110109_AACQJL warren_b_Page_012.txt
0464bd3350ebae0e290bae16aa3e92b6
bc746da7a1d716266abfd206586060a6cac3a8fe
111189 F20110109_AACQIX warren_b_Page_032.jp2
c27232301d45471fb62655a70534c238
27a5346389f8969d8426f73beabfcb381b8800d5
F20110109_AACQKA warren_b_Page_078.tif
19ffe4eef361ff398d1e0151e2a64cea
e25e731ae052819fc2526160bd9fb3b9eaec377e
5101 F20110109_AACQJM warren_b_Page_003.pro
b0974411c2d0f357c71357f4c86e7336
ae881945bb81fdbfc89b2212b521dc784c1f8aeb
8301 F20110109_AACQIY warren_b_Page_077thm.jpg
9b3f3842eb4a53599126358182507db6
5909fa0f7998d774726cd9432935a952659e0fd1
106606 F20110109_AACQKB warren_b_Page_056.jpg
d8ea43a0465910c269102c7a6fe1b920
051512ccf66ab7178520b3c00fc0fbb346baeabe
25577 F20110109_AACQJN warren_b_Page_073.QC.jpg
a055c75a8022b43fbd3baab105a400bd
c0fcd7a3293cf9c01cd82fdbe46eedd5d344f508
F20110109_AACQIZ warren_b_Page_072.tif
e1a0e5026d58abea7898239ab3c6fe8a
f6096fd03c9a14f254a7f90ac74c640548e88ebb
110611 F20110109_AACQKC warren_b_Page_078.jp2
e279cf1a9da0882429cdda8838031018
9cd2206bf9dc85d49b59537ad1f6132358289535
87244 F20110109_AACQJO warren_b_Page_015.jp2
705c6dcba97cc56fac5bf4b95abe8e04
f7fd3ff13c3b7efccec5c5936492f64fc076e5ed
6368 F20110109_AACQKD warren_b_Page_015thm.jpg
2c4d446c9f4506831efa14bf9d066255
3443305a4397acd36609e798dbcdd03b41d54895
52247 F20110109_AACQJP warren_b_Page_070.pro
f5ed5110a77836cced295607167c4358
af136fafe5f21b7ce60fa99ce8e37deebb5b07de
957518 F20110109_AACQKE warren_b_Page_052.jp2
305c64c8a54025bd090fee5c00eb3528
7ac887cf438ca946197c988689d0d6b137cddb51
6442 F20110109_AACQJQ warren_b_Page_011thm.jpg
b095bba6eb02021adb6d16279093ad3b
0b255ec904df00e1ff67dee397a8d9be136cca77
8401 F20110109_AACQKF warren_b_Page_027thm.jpg
abd87e12df8d871d4162da1fa0bc9037
c066a5bfae1b21ff27a1a5fc1770bc7a942e15ea
46144 F20110109_AACQJR warren_b_Page_030.pro
badba474fa1f0a399b690a07a8183a3d
1fcb994ebdd8c8c384161c7521ca5537b439e344
25271604 F20110109_AACQKG warren_b_Page_100.tif
575d3fedae8fb0d2ad30da279ff24850
a1174920de098ae6908773b2ab49013716155385
52468 F20110109_AACQJS warren_b_Page_087.pro
d75db73e7c3fd0d43c7dd367ca858b63
f1351b4c2562a7f97a79e8729bae1ddcbc04fac3
1569 F20110109_AACQKH warren_b_Page_095.txt
78aa2849ca8756f75332626c952356d7
eb40f8c4a1697edc0a8875b33c75339304104121
131007 F20110109_AACQJT warren_b_Page_102.jp2
4dc87b42a1f58c700bf4103428a80aab
2e8e4098170b755a9a2ffc48fb5eaef5dbe5c9a1
34007 F20110109_AACQKI warren_b_Page_032.QC.jpg
8706f16fb08b42e1b56855645a8ec3aa
eda802e861b612776806d58f90918662112f1aa2
F20110109_AACQJU warren_b_Page_026.tif
f0f08812b7ea77f295d060cb947cb477
a47cf0136cc5be1262c9b41e8e74e0cdd95e5dc5
46266 F20110109_AACQKJ warren_b_Page_068.pro
41d64014e834c5d579eb012ee0e26cbf
b5701ca66cd6d85ac257d17f86e96d522c7a1256
32966 F20110109_AACQJV warren_b_Page_053.QC.jpg
a45f217a109e324c4e65874caf2151e3
5a827af6c5478909a49c5d32df86a07751f03605
49416 F20110109_AACQKK warren_b_Page_038.pro
4fbfc0f9a3e4a1b12dc9651d1c0ef91a
c197c9e03c1f4a0551380a9dfaf0cbd41de323d1
F20110109_AACQJW warren_b_Page_071.tif
dda99f7530e9d5d752f6780c8395afa7
c86be2c3329591d5e9e95ccf8c8cd4c5652ff2fe
52960 F20110109_AACQLA warren_b_Page_085.pro
1ede4745fd4a85970d7025b0ad401f48
2a1f1e5aa08a7245fc129cc965e118ab10fd9684
F20110109_AACQKL warren_b_Page_021.tif
ea78903b394310ec071316f1e1df1238
013e81951db5cc978e82f9b321b862e0ee8bebf3
33301 F20110109_AACQJX warren_b_Page_023.QC.jpg
304398fa868e37ce5b4c301954ee786a
de01db36d9167be6a93948e3ba0d6de02d2c824d
8057 F20110109_AACQLB warren_b_Page_021thm.jpg
eb012f3b3e45010d20b3c2e9f49040de
99fd444b760ccce450565ab876e1fe485389e969
104200 F20110109_AACQKM warren_b_Page_033.jpg
cdb9bd3f910b0994ab68219c7a7ac7da
59cb3eccbad3f8ee117702de51317320e8af964b
44450 F20110109_AACQJY warren_b_Page_024.pro
1118670ebc570725d3d3b02c73c9f3e0
a6568a0b82f9aaf23006823fbb4b5d086ee2ee07
23970 F20110109_AACQLC warren_b_Page_092.pro
aaacf6ed096b10913ad0ba231c8bc589
70000089d00d1f742f9b500c4a1bc1554abd7c89
1907 F20110109_AACQKN warren_b_Page_021.txt
dd0e1651dfadaa1e8ec0133f9efe7ba3
74aaef4f3842eebd3d06c30c62a3c083c5c185c6
103579 F20110109_AACQJZ warren_b_Page_032.jpg
30d8a582b13ffa1a5b08fdc99f1469b2
88646d161432580273d686ee523dbdf7c58c2e5f
1970 F20110109_AACQLD warren_b_Page_032.txt
1aa844ed21e029c344e008c9d609d01d
7cc027292f524ed9aaf44ca3c3a24dcd29237f64
56843 F20110109_AACQKO warren_b_Page_097.pro
705b91b7c970ccaa13a98377583bc20f
a2616d271566442fb8495cde012efdb5f04a70bd
24672 F20110109_AACQLE warren_b_Page_061.QC.jpg
fd71990d50e9c0079e2a3a14f4fb4195
7aca96e12ae7e9f5b34bb00aa1cf8ee6aeaf8712
1051986 F20110109_AACQKP warren_b_Page_005.jp2
2446b9e5d522c406a3ae855793333109
ae731422e126233734fcf7b98c4dc1ae956a7a11
618 F20110109_AACQLF warren_b_Page_010.txt
05522f27103b7844a3c06a680c6714af
bede27357ead8a067b8ae133e35b6d1634d4b6bd
F20110109_AACQKQ warren_b_Page_068.tif
9d8e649138e67087cfcd46ba876f8e79
933c2007e6fd5b9fab6364b93e92469a7fd8261e
95011 F20110109_AACQLG warren_b_Page_006.pro
f659ee3e0ddd40cb281d147175cff107
c4e0475284d2d9ceb68ce694abc965af5ed19508
114049 F20110109_AACQKR warren_b_Page_047.jp2
3a7b8cd1a563560cf687ce883cb25a69
b5342f6fec141ea49bdee72d01619c1fd6262f41
27098 F20110109_AACQLH warren_b_Page_074.pro
107926b85abc783ca7bfd8b79dec7d97
a5cbff402e61576d0ff359b8a09947e9c1dcdc0b
49431 F20110109_AACQKS warren_b_Page_039.pro
fbd8e29de624d237846448deaf258ffb
9a0f9d59d2bd63fb10cafe5aedf9e7ba6a05dd03
137493 F20110109_AACQLI warren_b_Page_006.jpg
a6d9b7632f266c1fff0a7cedd7541828
d8390ce79b9d50d632fdf68da7ea001ad005b037
1519 F20110109_AACQKT warren_b_Page_041.txt
65bd36b04b844bc42dfdefd49ffde970
8d9f5aa3a7712c5bb8a73e97a62f1a23307cab32
53004 F20110109_AACQLJ warren_b_Page_034.pro
eade2e05ac6ac0ccfb85c0d44c23151a
9be6d6c229662126f885d2153d310614171880ce
3324 F20110109_AACQKU warren_b_Page_003.QC.jpg
c8b5f83b45ed0eb523300c3b49e6c7a7
560fa5ffcb2aa14e0364c60460459b2e3d2965b1
5732 F20110109_AACQLK warren_b_Page_059thm.jpg
94fe40c1fad94e5868b5c4ef71d3f47c
681fc630f3842221f5d5db1442432ee76a874308
2246 F20110109_AACQKV warren_b_Page_103.txt
22f24fe59314d7ac12ba1eebd75fd1b1
c55d98d47173667ba86e458b40a0812029c9a3a4
1051946 F20110109_AACQLL warren_b_Page_098.jp2
7731460f44877d7e9f729d7df0201fa9
c24a6006118114b4b97003c4c369641fef7d49f3
2081 F20110109_AACQKW warren_b_Page_087.txt
79c7a714e9a41b69b3405de6ee8fb278
ca27a31f1e3f8c178eb3302e783bf83afab1b72b
13200 F20110109_AACQKX warren_b_Page_003.jp2
797cf22b0501ac369748320e17775471
bbee4d69cf97d75ab211982b6cb82662e21a4a26
1158 F20110109_AACQMA warren_b_Page_074.txt
4ac6ce842d68ea96ecdb154e40b0f388
ebf78694738de35eca18bf6ea253aafdcec98407
94619 F20110109_AACQLM warren_b_Page_030.jpg
311cf1cd01b9aec1fcd93089fdeefa01
e0bed3ae427b9cca21a6350dbfe7d3814eff273e
24522 F20110109_AACQKY warren_b_Page_060.QC.jpg
ffc365e5e53ca25cfa6fdb8cc9f72372
a6563e0d7ad7277a946c9ffe27f5095dc1caff31
1051975 F20110109_AACQMB warren_b_Page_009.jp2
e24f6d6a4b1394de74a30552353c95f0
c049009230717431ef4cb6c86bfe252a7cbc9cde
F20110109_AACQLN warren_b_Page_093.tif
19c7fc74de54fe3486abc4a3403691fd
007bf8ffab24adc73309dec9ea6970bfe2c7bc93
6404 F20110109_AACQKZ warren_b_Page_061thm.jpg
984e208274a49bff7ad0a417dfd75511
fd05f0526134211ea4ec4667365b79e718b950c1
7511 F20110109_AACQMC warren_b_Page_063thm.jpg
93c9750d683fd5919482c4931729a19f
40a72f6363b5be2b947eb709cb07dde85ee96d52
40454 F20110109_AACQLO warren_b_Page_075.pro
e3b54e3dd07fe8c19a0309eb85d90565
5a2df26749f53ce0ced89540aa667ce0bf1d719f
1990 F20110109_AACQMD warren_b_Page_088.txt
ec84e520f79e556f099196dd7a4226bf
e95b65a4d7988e843eea6201ecaeabab3ed4f8c1
F20110109_AACQLP warren_b_Page_005.tif
2732367a693058cc237ffa22fc244b18
7e8caf8182746964fbbd15cd88f37810312c96bd
6032 F20110109_AACQME warren_b_Page_002.jp2
a26ea6d9ea4ed4f1d00873d219a34dcc
340d69c3362f4bc14565c7ac89c0a8a4ea19b52f
1923 F20110109_AACQLQ warren_b_Page_042.txt
bb9ca116a0604788a686dfb52ddaf67b
0828f172d0fa33258de883e756709e3e6366b29b
107929 F20110109_AACQMF warren_b_Page_047.jpg
686d2bd98efdd01ba7458b3972026c66
10812b50a8d69005dc6b28974ba5ffab96f1d79d
32268 F20110109_AACQLR warren_b_Page_025.QC.jpg
65d421a725d9a745c835a5957eda199a
ed92409b3e3503b3321cf3f91a12e2477e47ac45
102186 F20110109_AACQMG warren_b_Page_044.jp2
5a48bff886a11a0c0adafe55ac2bd004
d3e127e8c6dc8669f0798a93dad95ea5648ab9ff
F20110109_AACQLS warren_b_Page_045.tif
ff3d1292a24ce06921262d08ba484841
830e31db02c14b777932e505f836ebde5ee463c7
F20110109_AACQMH warren_b_Page_075.tif
d2971f1f3f66db5844a49eab4671b50e
b43d2387f5ef975745f224c99bb0dd24c88fd24f
107733 F20110109_AACQLT warren_b_Page_012.jpg
88c29090aae01a095dfe27894a216760
3b3a2945de116944d0990b5a4ae5e7a3da6974b2
84423 F20110109_AACQMI warren_b_Page_011.jpg
a8e55d35c8c5b2e86b223cd26e4d3b6f
0fc62ede36fe5c94e61eadd267cbe68e4fb511f0
1051980 F20110109_AACQLU warren_b_Page_099.jp2
edd680561fced1faebe1dfd9590816f8
94c6decdc94794abe26033df34202d832856dc83
50636 F20110109_AACQMJ warren_b_Page_056.pro
35d79c6a4c18685353bc5fe30ba70371
dab98c26f80cc88102574b4fb27ed62fc0e27875
1949 F20110109_AACQLV warren_b_Page_053.txt
023bb2fcd093be974ae54810a15ca995
0256fb18db24c6ffc286bc6c857061c1ccdec698
39948 F20110109_AACQMK warren_b_Page_058.pro
f284c7bb461989359737869f5334346b
45dd3ee29b77e8fbf24ff1a0f8f8cc6d8f1672cb
109270 F20110109_AACQLW warren_b_Page_066.jp2
006f3d74fb9e6273179356f8816aa744
d1c41069aa985c95c7da8fc9e8420f8cd663141e
104188 F20110109_AACQML warren_b_Page_014.jpg
6964b57277b8728d0dfc56d748eb7ede
7e9767c84976908610dbddf31c3758d8f1c1516b
32031 F20110109_AACQLX warren_b_Page_044.QC.jpg
97c203a850ecb35978cb3fcd9977e3c5
464a9c3ac36e89dfc5f73a232aea2b7f747da44d
51568 F20110109_AACQNA warren_b_Page_007.pro
7b12362aa9290e4eaf9a37bf3e626e82
e3e909e02718398fe7283dbee28d676392e47789
46066 F20110109_AACQMM warren_b_Page_035.pro
26526b9b3ad0d2149333b5f0ece3694e
38a6d8a67a3cb33d1765d8b05f727c5073216d80
94998 F20110109_AACQLY warren_b_Page_035.jpg
dc9ef11604cfa85014a1ef4080b656b5
55e9cf50170c0f89285b608babfe3f4526c0576f
F20110109_AACQNB warren_b_Page_104.tif
2d23187530a7e5342dec042a6a6b10a9
3e9da9838da02277880b5b86042367b333a00d47
F20110109_AACQLZ warren_b_Page_014.tif
6b636fc83b6ecb311ab55573aeb1a868
ed0808f4195b858216504322794d39f7983875b5
37207 F20110109_AACQNC warren_b_Page_098.QC.jpg
d17347c77963ceb74c5d4cfd7a3f94fd
400f0ec643673a98e618db6ae8f266266c15829e
32541 F20110109_AACQMN warren_b_Page_006.QC.jpg
6a6db2d7faab367001f5a8830398c56f
c8993e0b979836b06b84195207e638c0f9fc8aca
F20110109_AACQND warren_b_Page_088.tif
810c25914809ebc7946fb86780b18191
9d0d199cc223c5cb95caef8030e67e884640c259
1927 F20110109_AACQMO warren_b_Page_082.txt
8bb69dce4b39b92ffc73835d1a5ba5e3
3d11e58ee30448ddf52d4d12017e5292e74d5682
F20110109_AACQNE warren_b_Page_066.txt
0c0969451a6201abb2ffd8fa22dfd7f7
e6dffd5d1c52228db21d15798886884e62e76eb1
34712 F20110109_AACQMP warren_b_Page_097.QC.jpg
f48c35547fb8892041452c0d02b042f6
f7b2d82f2ba790d7dd461dd8854e403f1ef34a4e
90631 F20110109_AACQNF warren_b_Page_016.jpg
2db0e44216e9a7c2c4ccf6bd2e8b5220
b9abca79fb5b403b6124ed8bb6b69e98b58f8bbb
52587 F20110109_AACQMQ warren_b_Page_077.pro
2c698689693e9474047f202870329761
8c3a3b397a493ecbb3e94306b5227f754209fabd
33440 F20110109_AACQNG warren_b_Page_039.QC.jpg
fdb045a7d54f5a5186f29aeddcc7df83
1fd55bd7d64f3056fa578873e16886bd6330ccd2
F20110109_AACQMR warren_b_Page_036.tif
b2ab4fd504f7917b3a307abc317266b4
f7cadb035b0b66473505e51ed78a5a4036ebaa94
F20110109_AACQNH warren_b_Page_043.tif
0d314fb9937155b93927bae4b6cf5dd0
54b9b05ce700fb59c962892888d8754ca395daea
1701 F20110109_AACQMS warren_b_Page_076.txt
48e62461236184dc66e5e34dc2bd0769
c6604cd538fccecb5a2dd946d24fcec240744ff8
35315 F20110109_AACQNI warren_b_Page_079.QC.jpg
a40ecff5435704661fe2783975614b44
2cb153ba9d7022a8b99bef77a77f28e48ec96e42
50981 F20110109_AACQMT warren_b_Page_020.pro
2677fdbed4b4f351dedd9f50bb004353
c0d972725747d887f3de97ad82e202795375424d
1995 F20110109_AACQNJ warren_b_Page_014.txt
ca0372543f28f2e4ea52dd36c2a1fa5c
62b685ae2cf8310ebe461b57c05e53fab386ffa2
F20110109_AACQMU warren_b_Page_060.tif
1f9d1e0fcf13bc5e9012cb62978fa089
e5d93882c6b97897e7ceca8106a65502192d9f80
1011346 F20110109_AACQNK warren_b_Page_051.jp2
d5c55d092e78f4afe24d12d5d17ed568
7c7cedea2f7becf3e774f2aea80c9cbb8aa8347c
6882 F20110109_AACQMV warren_b_Page_058thm.jpg
11c090461f2fc4d980a837ca9f73c9cc
d0a9a9f51d605afdc0852968d7ef27a995a37617
8236 F20110109_AACQNL warren_b_Page_019thm.jpg
7395619e6d12b1f16aead19913a9c3d7
da12d76566771a2ba057d102a5be110206cb4fe1
8408 F20110109_AACQMW warren_b_Page_040thm.jpg
20877c6c81b3c94d6e8c642e0402443a
4f8f6156867ec72548fe2e73e966fb78d7a6dd1d
1851 F20110109_AACQOA warren_b_Page_057.txt
ba0d4a147685fc52289e91e7a70d0cca
1c68e4e35adb61ad922e29c1ee9d1326a7a73803
7363 F20110109_AACQNM warren_b_Page_016thm.jpg
6349935516f54f462354fbcecbf3cf63
f180f79c94561f4c81aff12891044a41904ea28f
8612 F20110109_AACQMX warren_b_Page_102thm.jpg
a6a65bc2ec5225be832479ef70d473c3
a4ed036e519ad46cb3413bb00f7f133cc6737783
F20110109_AACQOB warren_b_Page_055.tif
84036fb63b4cf0f7b403bf400e464354
8519c02e5b2392e6aa1de646842c54d3309ea930
1835 F20110109_AACQNN warren_b_Page_043.txt
b34ae548d686ff66c5e0d1cc6a3cfadd
69b3d30d1a5ec283dbd13e481ebeef9ca9611ba3
84262 F20110109_AACQMY warren_b_Page_041.jp2
25b6eb8bed3f5f3aeab78fe5214be9de
a417b5597bd41451c8f4cc51786b641f3aa95bac
27915 F20110109_AACQOC warren_b_Page_001.jp2
06c45ba55382669fc115de7ce353e684
54b18cb377d38a265e8174c1ced91db1b09fa1e5
7813 F20110109_AACQMZ warren_b_Page_028thm.jpg
9ea87f63e9ffb68039ded22dceca4879
fb048e28fd1d39039390c19ae462bc724e1fa8b0
F20110109_AACQOD warren_b_Page_101.tif
e91e511cd6a754f391879b0c9e0f1cd0
9103c81f3ad02c8e7401b5aa76c93641894e38ec
2072 F20110109_AACQNO warren_b_Page_031.txt
63081cc597db498ecdb5e0735d5ace69
dd1e39adf8fb64f6d8041fe2993f1e9e41db97a8
31579 F20110109_AACQOE warren_b_Page_047.QC.jpg
11174d5f07dfd38e66ae6bbb5139f827
d814318a5a75151a75a523a7e8532f56bd07d53d
F20110109_AACQNP warren_b_Page_038.tif
a07e370911f7620c4033f8310f1af678
5da4b8883412f13a4b5be97c827e6e920c0d072e
2097 F20110109_AACQOF warren_b_Page_079.txt
7653bd57a2136ad7b98843a90c4cd3a8
a725d72d662bf271ca558bc5bffcb9d7e8fa2b9a
83334 F20110109_AACQNQ warren_b_Page_015.jpg
916a5576dff57457725721bb83328844
3f6a33cb8d7f9c5fa3e2ced6eee3d6d7cbbd034d
92264 F20110109_AACQOG warren_b_Page_064.jp2
7c34699c9ebf1f7b08c89ca9b7fd5870
c73bfba805f4f13e331cb7bb9c35ec2500223ad2
8368 F20110109_AACQNR warren_b_Page_088thm.jpg
bfefba55fa8fec13e6b7bd49e1b97e95
65e721cc45be13fcd1b81cd9c9321cb6f2456ae2
115544 F20110109_AACQOH warren_b_Page_103.jp2
307b8edd47608148331771c9db54446b
b7f8e2e1543c60aab1305dbb4f672fc78dc717d4
F20110109_AACQNS warren_b_Page_057.tif
61324a61a4a60856265de9b3a7c533d0
c42a9905b139aaca7e8c5e4fa1b0b5b0f10f78f9
637 F20110109_AACQOI warren_b_Page_002thm.jpg
d6731e8a6e4d1b061bfe86926262d567
9128360ee3117584f139c1eeb3d2b4d186048861
121583 F20110109_AACQNT warren_b_Page_101.jpg
5c27d230f3faca48c0888a28113b597e
e6d4d9581c701f721268b4ed0e1dfb12b7a29353
7679 F20110109_AACQOJ warren_b_Page_075thm.jpg
5afcdf9b170e4687481a8ba935f3ab21
205c1f4083c17585bd8b7bf80b38ec76fb9ab14d
1051971 F20110109_AACQNU warren_b_Page_083.jp2
be2903ebd1eb6b4a1d4b19f84a108176
88933cd26efd75547b39d21f0bd4de30511e7188
8682 F20110109_AACQOK warren_b_Page_097thm.jpg
fb2b3b8ad982f121f1c5d8cfc420666f
f2ce0942c29fd36e824cb8df3d7a8247db06c8da
38646 F20110109_AACQNV warren_b_Page_099.QC.jpg
6c71d937fc7fc99a338dd24654a57cca
6ac01aa55684211f034816912a3157fd17201d3d
28335 F20110109_AACQOL warren_b_Page_016.QC.jpg
26e1d4c2b3b3880c3a39b7eaea526376
2181a3473e6ea7d6cd2dea16b1b2fb8580ea9729
103465 F20110109_AACQNW warren_b_Page_023.jpg
b7fb72e77594a41c07b8c9b4dc06b33e
d54533b1e179e5250f8f1314c913931602e8b647
52372 F20110109_AACQOM warren_b_Page_086.pro
f0aae2ec270a47cdb146c61623fff930
ba739f22a2ae1f7e4d730d1cba7cbca3848d6382
F20110109_AACQNX warren_b_Page_062.txt
5b09f4ac9d7b6f715e23d32a55cd991d
9d3bf0cbce1967387c330084f80656e4e3f106dd
F20110109_AACQPA warren_b_Page_033.tif
a76c3b5f7f3a39fe1a908544f366d058
c3d1eea70247425a7d2d2a6ed96f0b13ddd76c5c
25747 F20110109_AACQON warren_b_Page_004.QC.jpg
d126dce36d8a97bc4aac4a861fe5d4b6
233fbad6942496007cea491d403a08b941f70263
1841 F20110109_AACQNY warren_b_Page_035.txt
b5fd2f64455613633cbc0ae4484bc7a3
209804547f973cbf8a46edb4e306745d5dd38274
8246 F20110109_AACQPB warren_b_Page_071thm.jpg
715df39de4dfac0b0125cbf92c4f983c
1d507cef0dd34dc2056887615774432e7144aa9f
21770 F20110109_AACQOO warren_b_Page_054.QC.jpg
4a70fb2aea6fbb5d8f37667d5d0deeb9
164856ebec349a28e6cef0d888ddf616f47e5686
7986 F20110109_AACQNZ warren_b_Page_103thm.jpg
eec6cf4822ae9de9ad52dbf5de030afe
7d688de2bd694eb8d556fc8d44739e2fa7d73486
112242 F20110109_AACQPC warren_b_Page_033.jp2
c259390a7a8e504cfca70c1441e8fb2f
615c8ec1d9d957c719539ac00c75491b41a51eab
26847 F20110109_AACQPD warren_b_Page_062.QC.jpg
68872a879249b4438c0950296d25b8ab
2d44c252d349debd35413dd0706eb32eccc17a6e
26910 F20110109_AACQOP warren_b_Page_052.QC.jpg
110dad7e769d8bd5afc42f511e44da78
21512afb3a1c153510820145a0aff5470c16e0b8
F20110109_AACQPE warren_b_Page_068.txt
7cda3e1d71036f7f97bb24ab02363d99
b75b5a3d6c69aca938302e06e69ca73b9856008b
23481 F20110109_AACQOQ warren_b_Page_054.pro
dd15b876d44fdfa52ddab7a1a344695a
f3cc2e57e294f6afdac0169792b53bb6c6e1e630
105495 F20110109_AACQPF warren_b_Page_021.jp2
c03626a84b24c5a6706044f5a6257a63
8145decd05b1b1e09409fc3e9c601f6cf0fc86f4
69747 F20110109_AACQOR warren_b_Page_074.jpg
9916d52c08bec6fc6956bdd96e1067de
2e052b02153e2dfecae5d1edec25cfa3937dbf8e
50738 F20110109_AACQPG warren_b_Page_023.pro
7a391d027af4f6692e54dd421a00dd77
2577576268dd1d556e911307193a7bc8132b7023
8958 F20110109_AACQOS warren_b_Page_069thm.jpg
87e40132f17ed316f11408a120efe874
3e160915969322b17b9c5b7d865cfb8336f18402
F20110109_AACQPH warren_b_Page_085.tif
13d6d7e47cac09c68817deeb3a9b3949
41b0dd817bf3d1da985ac8e668e771735ea1910c
6201 F20110109_AACQOT warren_b_Page_005thm.jpg
eec93bd49c61b24fba7e98d7a46b027d
90687f9f82fea8fdb412b2b8f423f34b9bfadbbf
33645 F20110109_AACQPI warren_b_Page_078.QC.jpg
756a15868a38007b30efd49b26294dd4
1872a8e4a278ddf36b029de6c50062bb4760b02b
86774 F20110109_AACQPJ warren_b_Page_076.jpg
88e39ef789d70aba3fb51c826b6dc0ef
ea55088a4b4065f58ee5631468a11ca953b7e48d
F20110109_AACQOU warren_b_Page_070.tif
da7a0dcd48c033ae5e8bd9681a67912b
c4bf1a90c81f11961b3e8fba161fb4ca989790ab
F20110109_AACQPK warren_b_Page_064.tif
b183434653ca8784c016a8ca987eae34
0cd479f7b5b9568f7b00c07932d124e732280c24
34776 F20110109_AACQOV warren_b_Page_077.QC.jpg
d5835b7f17634efa035d385c619a0e70
b54a2fcd4c0c7ee4840f7eda45e4b1dbdcf5a4f2
7953 F20110109_AACQPL warren_b_Page_037thm.jpg
594a739540857741db87475586e243f1
eab3a4d40882bd595a28a1c166b570977ff6224d
F20110109_AACQOW warren_b_Page_097.jp2
944cd54f164ded2f128d31f02434211a
4b1d8d0b2ddfc956501a57bfb30a3ca82c022ef1
108048 F20110109_AACQQA warren_b_Page_087.jpg
785d9bae5ea0917c030b5e39120f0f42
2799f0237b1661181bbd42a01401aaf7926280b5
34910 F20110109_AACQPM warren_b_Page_089.QC.jpg
2501d98a5809088ac21de5c8cbbe9d91
c4990d1a155e326b57131f23262098e9225b5cb6
51510 F20110109_AACQOX warren_b_Page_081.pro
f3d872aeb071e55f1d649a949e9f555a
a56672cdbf1b04b4d56ae7a73358b2ff48c16b01
114009 F20110109_AACQQB warren_b_Page_017.jp2
739e8d1ff88d1b4024b78e1fe53fc042
97af1820abf33bdf58e31226fd99bfe6c8cefc41
101817 F20110109_AACQPN warren_b_Page_042.jpg
dbd0c8c63d0bb250af09cc1fcce4987d
1270e5fefff76fe7d9a41b4139deee2f24a9775a
F20110109_AACQOY warren_b_Page_094.tif
512d990153f79ebe50ebf95d468a09f1
eeba271920506c00e74ccaedaba456865cb2cbc5
52656 F20110109_AACQQC warren_b_Page_047.pro
120ae6a6df5036d62b9eb9601fcc84d0
0780459dd39d394c1e07aed8466f36b2e327d20f
F20110109_AACQPO warren_b_Page_090.tif
81a40406f5426e68c9a52dbea9efc4e4
1f7e2a04668da664e408436987ddb06b3f309206
954 F20110109_AACQOZ warren_b_Page_092.txt
22177c2079cddb88c9d262244ee6848b
c1c522f0c18ec815852452b47205fab67f7b9034
48182 F20110109_AACQQD warren_b_Page_036.pro
b5479a63e7ae9e6563d6f4868df31f7b
603e1957706ae212bb1e358accc29b0e28a877cd
33792 F20110109_AACQPP warren_b_Page_040.QC.jpg
a96ee690828b8d132bffcc328e4622ea
aa13589401393b1778f01f460f08a77eebbb318f
F20110109_AACQQE warren_b_Page_004.tif
e08f30c9c1bee66e2522692fd75693d4
41d9c24cc006eaf370448bd2e02a99123cff7cdf
31160 F20110109_AACQQF warren_b_Page_065.QC.jpg
e77d12e52dbe5dac72139e40591426ac
094292b31f5a19846d0541cf5782810ed01bfbc6
1878 F20110109_AACQPQ warren_b_Page_028.txt
d2c123903cb47329bae88bd7b2f9fca7
e5586e12990d7ac77e68a59472dfc9b0f011958f
8670 F20110109_AACQQG warren_b_Page_056thm.jpg
4436013910817917ce086c9bfd6b4022
a1e44395a1d0274c614f6400447e36abd3d7f2ec
107045 F20110109_AACQPR warren_b_Page_039.jp2
37baf7d4a2544ebd4d1a8e19ee8c251b
89c4fa54143985d4bc41902024127a7dbc8fcbba
F20110109_AACQQH warren_b_Page_063.tif
68145ebda53fdef2a425acf2ee1a941e
d286a50165a58375cea9698d550c0903235e2056
87816 F20110109_AACQPS warren_b_Page_013.jpg
51e586cd1aca205ce37aa5cc6126877a
63790c4bb34ef03bd42e53f9d0074e9d0e046d82
1716 F20110109_AACQQI warren_b_Page_011.txt
f7923301411612766b24b1b86d707ad3
486f8c12c6104ace667476666a504ed0d0919c30
97633 F20110109_AACQPT warren_b_Page_021.jpg
25a622ace301dae1d52c83ed2d83fc94
15bea62217bb3de860aca6277e84c8e1f391c2bf
1858 F20110109_AACQQJ warren_b_Page_037.txt
8bb66d65e609d494e0a1ad724e445a6b
63edc0670809fb0458ccb5e2cfac0dd0608ba3ac
2029 F20110109_AACQPU warren_b_Page_033.txt
db589717b770b60bea3c86fcaa252ebb
0cf14915d3678b128159b20ad8932d2d62af9815
47867 F20110109_AACQQK warren_b_Page_049.pro
ca21c0f07c69f5bfeb17324719426c6a
2f45912e250cf0d286d0fc33dcd4fc1b40c43ffc
1754 F20110109_AACQPV warren_b_Page_090.txt
e2562bc6296a7bbd52607d0f1b95b024
0d81653ab24dc52a5265bde7d184f33ac1ef48d5
104710 F20110109_AACQQL warren_b_Page_081.jpg
89a61f6c6e13939bab6ce2f93673df54
9e68661635b6ff0f159fc4a5e05f57cb3d15bb3a
2513 F20110109_AACQPW warren_b_Page_100.txt
e6eb8100535fc93032794cd4c35f5087
552f09a8cbbc667e72c27978cf93b780df5fb817
34031 F20110109_AACQQM warren_b_Page_033.QC.jpg
3744a3a07546fcc0c5823e21faeab6ee
2c62cfda0568034767e673c1ef71ecffc66475c3
50068 F20110109_AACQPX warren_b_Page_027.pro
56f6eee797d839feb92f3685f56fe474
c32e3a77dea705cfb942262f3f4adbba82b388b3
49928 F20110109_AACQRA warren_b_Page_032.pro
079670cd011ce8db73303512cdbde956
2aa16382f6c91737c8ee2b84de1b2f858a6b1b6f
111271 F20110109_AACQQN warren_b_Page_086.jp2
fd726b371ccbad6b06c50e1797af66e9
c69e9555a0cbbdea7a299d0adc8b21eb37c4dfaf
48791 F20110109_AACQPY warren_b_Page_042.pro
dc1f9819bc7ad61dbefadfa264a15456
e5fcd81592e3ff0a67278f42850bee145be4c635
26132 F20110109_AACQRB warren_b_Page_095.QC.jpg
e9fb4a91001d5e4173b669269f1ed8ed
21e5824094a3a50284c12ea5dbd91436fec8eb72
101659 F20110109_AACQQO warren_b_Page_053.jpg
a4bcc107d56634223580a5d9dddc9a01
0e4d70dbe05f1f685d151bf69d8581e197fff32b
35996 F20110109_AACQPZ warren_b_Page_102.QC.jpg
a163acac3a06692d619a627d48a5aa91
cbf1e4cbfab8168f4ca703c8ec691651ee3ae0b1
32428 F20110109_AACQRC warren_b_Page_061.pro
a800ea4c44495cd144c92f007f9a06c3
33f5a46012e6ce8c002172a3417d26a5cd4f5465
97146 F20110109_AACQQP warren_b_Page_025.jpg
9347030eeb2096139377b4a7b9cc6294
0fc44229418966fe556f1effe31fe0bc8592d3a8
F20110109_AACQRD warren_b_Page_048.tif
1660557ccaa03253b7ac3a0584dd8d10
6ecdaa1900dc63a580a93a53b69bbef08ae345d2
1051979 F20110109_AACQQQ warren_b_Page_096.jp2
10675355e7d49bd7bb153f13c27b6269
609a91156650d2b4aef6c3ad9ce6293909901e9a
34785 F20110109_AACQRE warren_b_Page_071.QC.jpg
32ef5c763d8fd77619af5f5587f89c20
d67cd4ad396c79fbd1a4ae05d304332f0a778915
46023 F20110109_AACQRF warren_b_Page_104.jpg
b88dd05642a0cfdf3eae52154b875a47
9d4607e744d607ccdb07a43699fe8f9d12e2e2a4
7954 F20110109_AACQQR warren_b_Page_068thm.jpg
495e68ec2d1dc38e787f261ae9abe8ee
75306b5cf12ba7f508b56e20ef640d92c8fc4f0a
F20110109_AACQRG warren_b_Page_046.tif
0a55f44d6b3930e96e259233fdb4fe21
1e7790752f29a4d01a1b8a667c91113dc7d17aaf
876 F20110109_AACQQS warren_b_Page_104.txt
8a92d513fe5e572e28992bbc8d49012d
4aa981c892aed060e6657e70b8a6f8bf168a0706
8075 F20110109_AACQRH warren_b_Page_045thm.jpg
dec259578334f18a84a4c051cb6db852
2d80adc9e684890f20f6834ed24f6925a9000bc1
34215 F20110109_AACQQT warren_b_Page_026.QC.jpg
ffd7195c58366fc1edf7fe41d72bdc76
219187c8a8e0a26fd8de87ffc1a0397897082e3f
1536 F20110109_AACQRI warren_b_Page_094.txt
47cae17b058861d1735aa65fd24c1a7c
e8c90e5c7c03181819842185a8401c99fc6c8bac
107395 F20110109_AACQQU warren_b_Page_031.jpg
e3190ce714e8d8a62a0ba9f1d251aed1
4bc6dfe22c9d4f4e8646fa25e23fe0eeb3d69c76
123473 F20110109_AACQRJ warren_b_Page_071.jp2
3fd20cbb186c47b4903a5820f74c0d9f
d05fdd5593c157e43130ef99edc03f41459408b0
2065 F20110109_AACQQV warren_b_Page_017.txt
56630d74e83e08eda68ff732ed64f22b
04c0f20267e3da9c8c84dc6faeea1742f159929b
F20110109_AACQRK warren_b_Page_025.tif
2cc595847523a3acdb5675547eb2a9de
6936d7e5a9eb07b49f4fe781329e9d528931fffa
61927 F20110109_AACQQW warren_b_Page_100.pro
530b0e5838d295ce273269821878ddd2
c8475e2aba5f08d704968b129b93da9f888c8c72
F20110109_AACQRL warren_b_Page_079.tif
bf67e5cd5354b0f298fe75d7d4abbe98
e939182ec2c2d1f366a39a1bdb7b0ac3f9a61913
111198 F20110109_AACQQX warren_b_Page_012.jp2
49b48992f450d56011bea719de09e1c3
46c509aa5b88dd79988d412a6e70464a3f3a629b
907966 F20110109_AACQSA warren_b_Page_062.jp2
b0b15f5410e8aaace3294fadf13274fb
dd68d362405a74dd45aa8cb533835811d69c42e4
27301 F20110109_AACQRM warren_b_Page_059.pro
caa5bbc3e33a8e2c1f0ca8d9b5831fea
5a6e66a50c2f022104c223d83c8440c4a29ed493
104277 F20110109_AACQQY warren_b_Page_020.jpg
4e62b37e699a887f639f50deae52a27d
d3cdedb2686e51f2716644ed56ee4dd1d32284a6
8437 F20110109_AACQSB warren_b_Page_032thm.jpg
8f1ac25dc5aa8186e66d09357da9b817
d50745718c47e0d24095189b882aa5ea933bfa23
88492 F20110109_AACQRN warren_b_Page_058.jp2
46637e44830c27403534c6f468e88fb1
06dc438c462c5abf0bb24672103a785c2ffca17e
F20110109_AACQQZ warren_b_Page_051.tif
688c8d65887d733139980d6bf6debd2e
b61fe9a49b4ade2df15089f29df71283a8e6f40e
115944 F20110109_AACQSC warren_b_Page_079.jp2
31dd8de653313c399b5f0fd3af0138d0
3e9ac8311a32fccc4fa69a946f02b0b618e4f9e5
8957 F20110109_AACQRO warren_b_Page_101thm.jpg
00d34a9922992025b1ad6d7bbecda907
484b013302ac9645d6f7c11529abea4fdb59c398
1947 F20110109_AACQSD warren_b_Page_039.txt
54dcf2a5fec5a523ffef48ad8c559c0d
3a0ee2ddd4cc8349f15aae1dcb5cb149c7a694aa
33280 F20110109_AACQRP warren_b_Page_080.QC.jpg
a2bf025bea59558363c8ac927ee4f132
72c64afe11ac43713bb46719d49f6e9ff62e09d2
49941 F20110109_AACQSE warren_b_Page_040.pro
0c74c1b8bc6939cc954766bf78fd30b2
6dd5947a765635d8f9a524b91daa612452ccae69
37943 F20110109_AACQRQ warren_b_Page_062.pro
6181d8703cc9b9466c772fc3a19362d6
594066566ba59bb20348459f5f01ab3277b84f66
1875 F20110109_AACQSF warren_b_Page_063.txt
919f3f3ff35c7c3295f38cc6003d7b91
74b30e54175f8577f5b3c83f5d8ccab90f6b51c0
1007307 F20110109_AACQRR warren_b_Page_065.jp2
65cad66440719c58065229d8e95277e8
10fa852ce1becd514b4751291bab5db663bc2d3e
88487 F20110109_AACQSG warren_b_Page_062.jpg
d6b2a7202751b0fb8d8a4ba7266372c5
9a3dfec7917a8b11fbf8f07d33bddb5163fa38ce
964341 F20110109_AACQSH warren_b_Page_063.jp2
1a1e8d57df3f124ccc29c38565ab0302
1e4948394522e1fca05787192ec1a86f94689cd2
7775 F20110109_AACQRS warren_b_Page_035thm.jpg
632fe2234d576c9574f008f057e6a03d
ceffa3ea63ef710292c7f4b7237997cfa418067a
35564 F20110109_AACQSI warren_b_Page_041.pro
f17d774ee39da9dd045aa6d3a9dd0043
c9ed75227768920e8eedb9f506da885b74c143fa
4019 F20110109_AACQRT warren_b_Page_008thm.jpg
7b3cc93bc40cea499629c0e5c63cb71d
5680c8d8479d65db4b99b55a0de1cea8efebc257
81985 F20110109_AACQSJ warren_b_Page_004.jp2
6c830c64724669f062341e89a12d6949
f44b464822115c27fc1b47b8591adcdd0845cb5a
9190 F20110109_AACQRU warren_b_Page_001.QC.jpg
39e398ba81d0e5e70f683dcf8ff3d434
0aaf80fcc75e7fedbe1808c46bd917afafe35496
F20110109_AACQSK warren_b_Page_047.tif
6c4fdb1d3d330b4f8c821a9a54ab9bc1
9a191de5ba6c1b7d7251a4c109120909c664b1a9
33574 F20110109_AACQRV warren_b_Page_018.QC.jpg
898b3db1a6fb00074d4836c275177e32
0e7afe842b54f98f2810b697f4bd8dcca6114d07
113713 F20110109_AACQSL warren_b_Page_067.jpg
dab77e9841d85869acd6cd6e0bbefd93
865abdc92c218487af56deb4d8865c9c6092ff39
112311 F20110109_AACQRW warren_b_Page_014.jp2
ef2d083a4c82d7bb73e6b17395c043f1
8e433f0c8887ed86c31f06e4260e7a4e842134ae
2444 F20110109_AACQTA warren_b_Page_048.txt
d38233347e14053e680656e53dd7920c
2e7e95e0752e7695c197db3b7987843d885a0ea8
60423 F20110109_AACQSM warren_b_Page_048.pro
2cae52889bd3c07ab7d0b2cb708d97aa
6708b8cf87fd9286c469693c2cdf6135b7803a59
119171 F20110109_AACQRX warren_b_Page_048.jpg
155d1d5216da6f433ce842d41737a8f8
7731b85fde880b021d4ad63ea8abd26989a95b3a
8337 F20110109_AACQTB warren_b_Page_014thm.jpg
f3e3d284e89306fcf13166ecb3408c67
8addb26c046fc6e3a9bbcb82db8b901891e20a76
8596 F20110109_AACQSN warren_b_Page_079thm.jpg
a2e3951aff420005143126554f0f151d
6a11c2497b5d23a0ff077847ca3a603eb1d6e29c
F20110109_AACQRY warren_b_Page_004.txt
113ad826e9a4f07762b89f1510d8c094
ac8dac1b3eada7dc37dcf2309754111e1410be52
47566 F20110109_AACQTC warren_b_Page_028.pro
3801b3d5c88e339d6a7bc8243da76f10
7f751e77c025d1ffc5e0cdcf900e78be8d06783b
8087 F20110109_AACQSO warren_b_Page_044thm.jpg
92e2e6be1ed27165cb62fde97136f5c7
c9c469df537a7d1f83db4030cd079440469fb84f
1115169 F20110109_AACQRZ warren_b.pdf
b969d02bd5a256476162f8dec4e08cca
2a386ef4290a7b5e76f0c718eb41c45f1235b688
51581 F20110109_AACQTD warren_b_Page_033.pro
4dd675beed3755b1460787a79cd63f48
7ae8e25a8c62afbe84363a27868afec3babf3b1c
55530 F20110109_AACQSP warren_b_Page_092.jp2
e38f10e664230d490078d4129d0ae50a
1203f210a63d012b5b6e987b1dd406977394eadb
8952 F20110109_AACQTE warren_b_Page_100thm.jpg
074449f609e5a65ff041fa5d554ceee5
ef4edb33034a6ed034e17488f6cf5a1939a4e62e
30127 F20110109_AACQSQ warren_b_Page_024.QC.jpg
dba5cfef12b0f85340f2ac8379b4cadc
7cfd7808656c6b3441fa6c9f4dcd2f59b4d21a0e
F20110109_AACQTF warren_b_Page_089.tif
208aecd46a4e275248f6eadf85c3cc30
763ef90bb24f2ab9c25f8b940f73d4709dc1d2e6
32716 F20110109_AACQSR warren_b_Page_010.jpg
e12e18ad363ae928cbebe8ad806c9107
218fb597118a3da486783abe7564d1f77f06c5c4
8160 F20110109_AACQTG warren_b_Page_042thm.jpg
c69606705644597bbeba4e8f55ff3838
0b3671b556394fd53736d72b840dae7f15f6c26c
F20110109_AACQSS warren_b_Page_084.tif
39f5c5bfcca906b2fc2907e86214f81c
496cd50e4a3b6ee3a841ac6a292fa44a52bf8024
8003 F20110109_AACQTH warren_b_Page_080thm.jpg
a7850f0ad70fa79ccf3989cbb32ed6e3
df216ec70c83f3ea4006d106e3170ecd762822b9
31647 F20110109_AACQTI warren_b_Page_043.QC.jpg
3e9c04b26c256881669e2c48e9548927
91d71932eb807af0bd784809fe5ef947c13e3d9f
1280 F20110109_AACQST warren_b_Page_059.txt
9fe57aba020ccb657211e6a766259884
61b973227b2522161080fe3df1b52eeb31b2bf10
20069 F20110109_AACQTJ warren_b_Page_059.QC.jpg
abca9fb3674b93702ccbf5d2efc9ba10
956cd7010f18de94d313ebba167f9703e3018d5a
1199 F20110109_AACQSU warren_b_Page_051.txt
1b73b5ed74c1d32795abda779265a63c
c42f302a1e892673d4fb8983605b0b6074a71ce2
43347 F20110109_AACQTK warren_b_Page_067.pro
c9ad1752f07860dd6658d97d6f6a0fbb
19bbc0b5a4e9cae2f6ff3e1ac7a4fb396f9fd06c
102565 F20110109_AACQSV warren_b_Page_091.jpg
f9f81a93bfd1b48849a12d4cd09ff28e
52dfac7d1b9c489d0ca16f14eb8a7afa2093bb5c
42561 F20110109_AACQTL warren_b_Page_013.pro
d4bff5cf52bc0e40cc20d7816d016ec7
71e6268b8a2acc92a05f98b2fcd1feb15dbd4a4b
51803 F20110109_AACQSW warren_b_Page_017.pro
54d96ed3551dae03e53f8f2f84e01822
9abe09929b11b2b3ac11ffc608e8220ca9d4b1ac
F20110109_AACQTM warren_b_Page_023.tif
6159c440f0fa9081c63681a4efbb0350
1d2b660b29b365fac7c29ab886b4eaa873e364c5
1769 F20110109_AACQSX warren_b_Page_024.txt
fbe175412a9bcb9de5f9f73f517bc753
302cbdf16cf8d242bf492f52264889f72b6ff770
F20110109_AACQUA warren_b_Page_087.tif
6bfb46d2313056e690867437c4f5c5da
9a32e9ef4efa9f8e6195cedbd1ab9e05ea202026
804056 F20110109_AACQTN warren_b_Page_073.jp2
53e45205444173d4bcc06fb2130299f5
cfb7b856279ced2c6c6ec6c6e62183997f33e761
84619 F20110109_AACQSY warren_b_Page_094.jpg
773f71906229b481ee29decec70a1101
d24fe87d08ffc1ab5c2a78510e95ef0cca531c62
112837 F20110109_AACQUB warren_b_Page_069.jpg
18745ffdedf00334d145205b8f551dba
116af4552df1d59fa2689f6997fb05f17f902a2d
101600 F20110109_AACQTO warren_b_Page_037.jp2
9fb6df666a334970fafa2822bcd0fdf8
3b4e25b9abe5beb4a5ed73bb23887a61b7ddc1b4
3504 F20110109_AACQSZ warren_b_Page_005.txt
bc99cf713499c3c1a8100b5d5a756610
ce0a4333a2919276616049d30370b31f0ff9237a
6874 F20110109_AACQUC warren_b_Page_073thm.jpg
0f9bcabdefe9d986a21d66e2eaf7db00
77fb91db45443a5ab9530a113a7aa0bfd1c191b9
46683 F20110109_AACQTP warren_b_Page_057.pro
0358a1f4b5a18240ba16d4ac1b13d7ba
763455e9ddd32af35a7a0d8cb6df323938509772
2104 F20110109_AACQUD warren_b_Page_096.txt
3a8fe87c0a7a42c41d362c9c197559cd
eef0531331f34cd30d78f93a8b0d92e2eaaeab20
67733 F20110109_AACQTQ warren_b_Page_059.jpg
39c18ab760505820fd77ea3cc23d0e02
f0fc9fd7536f20a160bdab3597fe8cf2394fa5a3
41912 F20110109_AACQUE warren_b_Page_050.pro
b1e8f1cda1a839647e2c5b987b9addba
7bef3a965915a30cff0fdb620c26071c8060ad01
60917 F20110109_AACQTR warren_b_Page_072.pro
945a0fd7c6df8dfdd81be87a48cadc87
46d04e97936c3adbf174ade324833c85c88426e2
111024 F20110109_AACQUF warren_b_Page_089.jp2
150f8bd2ce345752478b1862e4b09579
75424db33fdb4298366d417b7d954de045b3c16d
96903 F20110109_AACRAA warren_b_Page_028.jpg
a0a52405807bc5cd0513ffc7f2e5eb87
209becc2cb2fe2eebaa8ba635ea43305ea2c2cc0
F20110109_AACQUG warren_b_Page_001.tif
bf48b55ca0d0c6925def871b95f21942
3f0d6fa21857103e8fdb3190dd45d81069b5c25e
112445 F20110109_AACQTS warren_b_Page_005.jpg
6b3a1152e832c1b8d821cae9abdc66f9
5ce64244132771ab4577ca411b84b44bb153db65
94726 F20110109_AACRAB warren_b_Page_029.jpg
b908fdd51889ca686afda46c03e817b9
991a1eaf8648aff7142174447e5573cae2caaecb
7792 F20110109_AACQUH warren_b_Page_057thm.jpg
210629e32919a1474d46c09ff87219e5
3a02dc4c9b1aac67cc090b19f7e394eb4c4daab1
2008 F20110109_AACQTT warren_b_Page_020.txt
de43c26d54673b2adae9184da2b85484
72ed6df4e358af338e64521bacd433ba0b3d3d10
107847 F20110109_AACRAC warren_b_Page_034.jpg
a4b632488ef6e8e7382b261f9dabbb61
d3cf95ef2795bab8d84800157dbd21d34fbc5238
1909 F20110109_AACQUI warren_b_Page_036.txt
ce4318347defbab587ca1dfb8f464ee4
df838e7b504877cf4a360f79d36a456ba9e8cc9b
102206 F20110109_AACRAD warren_b_Page_038.jpg
beb7bf83a266d7b8a071d1228c125731
8b13256ade92a92d4cafb30e9c6fd793cccad959
F20110109_AACQUJ warren_b_Page_028.tif
3cc8b46b64f51258a62803697a889ced
46533184e52d2302ff8eb0d31444c89e22fa06da
30965 F20110109_AACQTU warren_b_Page_037.QC.jpg
ca57a9027abad63bb1c372b61997c3da
5e80b02a336f75c12dc68c3dc7da7114257ca052
101628 F20110109_AACRAE warren_b_Page_039.jpg
8a8228f1ef798c6351297b36757271a7
349a41fe2318eb397ce28b359625054857a2eb9b
2452 F20110109_AACQUK warren_b_Page_001thm.jpg
9ae2d4ed8af8cdd80d7034b9a503003a
7fb75e1ffaae7c1011d37443edd54818d471c198
7878 F20110109_AACQTV warren_b_Page_083thm.jpg
3733619be44281744ef854609208711e
1684d75e3b185e7f25a26fe4fe5aabbed652ff93
103171 F20110109_AACRAF warren_b_Page_040.jpg
7facfc32cb5b4cb8462bb76bb6c3cfe1
81da465854bf10721c4982352b004f992cf41d09
92330 F20110109_AACQUL warren_b_Page_024.jpg
6f3ac09fad5aea8702e50b9b281c168b
66090ba110350e6d036eee7f1f02c2b4473556c4
101761 F20110109_AACQTW warren_b_Page_036.jpg
ce6a1e994ed00b36b75d4cd6ddb60231
acfc27b9edc61c66a6a763d8b38889a96022c394
97123 F20110109_AACRAG warren_b_Page_043.jpg
06a0a542879e7161c472f5377276431f
aa5904e71a40d9b614e13b24c3cc8333522b4299
40760 F20110109_AACQVA warren_b_Page_064.pro
9de5feb488258a3e134366a8061c821a
b93014690ea6fc6bd9f72fba7053cdda71466a7b
115738 F20110109_AACQUM warren_b_Page_084.jp2
130abb52b8ec3db6143a6c72778297ba
318858a45f28ca54b9e2ba5fea8e3d3a179f50f5
6360 F20110109_AACQTX warren_b_Page_093thm.jpg
4742c8da059919417f07dc161ac50678
edf5942243b591bb642d807ba1bf0dbaa0655c4c
84382 F20110109_AACRAH warren_b_Page_051.jpg
59dc80a3a4dc040ce383a9c3c5c3a917
43c6d93f804bade35e04c5c88fb5bf62a33e28ca
F20110109_AACQVB warren_b_Page_073.tif
c40f920fd8e3e14bae341b51f8f09092
af1f19c3073f32873518512566243c6bb618caa1
8358 F20110109_AACQUN warren_b_Page_039thm.jpg
8b71ea4de5b56f3ac81048c52ea32a76
eefcb04baad7972ce034fb78f5f03ba4d4d9f5f2
1866 F20110109_AACQTY warren_b_Page_044.txt
5df345f6c8ca9d9db13beacbfa8a8d94
17353c143b9877df26da0b74dc527428d69262d9
86795 F20110109_AACRAI warren_b_Page_055.jpg
e1944ca083e86f9b023c5d8510264642
b79531de6ef18820ffec55cc75de2587c3ef994e
104590 F20110109_AACQVC warren_b_Page_027.jpg
d9170f330fcfef69e99773ba905029e0
ae0f2d4891287e85190e7df18611f48e77d7438f
1751 F20110109_AACQUO warren_b_Page_065.txt
bcc908f3e06dbcbd6bf41825cbfe8611
ee317bc09951662dd9711a584240cb0b1218555d
1955 F20110109_AACQTZ warren_b_Page_038.txt
06d9932e1ae35a3abe49a18d52e772ba
596e7b690092a7cd373de5ae6ad797acfd1961c8
94601 F20110109_AACRAJ warren_b_Page_063.jpg
ff48776dff60b184ecf34a601485adf7
45bcfe81a1369d3b00a3eb6731273a4f681485bf
104248 F20110109_AACQVD warren_b_Page_017.jpg
df6647ac57843dd107f2ba8711f378f2
aef58862eb740fad9760c8b06fbef6cfba23165c
34783 F20110109_AACQUP warren_b_Page_056.QC.jpg
714b2b5ee450e0f554b985a63cfe14fa
340e63faaf0764e1a0c53ff4cc6e949d42bc1f45
103298 F20110109_AACRAK warren_b_Page_066.jpg
970419652a9bdee73bb4839110f0e4e4
7ace07be5696b2ce57eae92293206c1f9e8a10ae
8729 F20110109_AACQVE warren_b_Page_067thm.jpg
b9819b44bfe941cde6fd20b12af2367d
4e451df282a38597ac16591573027b0f53b876f6
5990 F20110109_AACQUQ warren_b_Page_054thm.jpg
023331d48c655d2543462249eafc54af
417797002d16b21b4a672e3bf9621f0d043ab27c
119103 F20110109_AACRAL warren_b_Page_072.jpg
1d9cab659f34666db9aea7dcd97abfde
ae79d5acf7bf7227bb3b35229bc5b88739b705fc
3943 F20110109_AACQVF warren_b_Page_006.txt
f0adc68375a1123c7bb6cf3cc8d34ab1
488a9a7648ed8abe7f0b721eab5e1ccfc46e62e5
50599 F20110109_AACQUR warren_b_Page_088.pro
7b50486e5c2d42d964aaa79f34ebf15a
0cb16a19cc6a202b7e1a671cbe0cf58f37d0a637
1051978 F20110109_AACRBA warren_b_Page_008.jp2
0e17977aba492554fa9a4cd2e5b96039
918b0e51165c5c9245267099d3876321fefe6f76
78952 F20110109_AACRAM warren_b_Page_073.jpg
727bee3d56d4b9251c3ce7f177d98f2f
79411943a0a75a28f6e086feb0353786260a30b9
6960 F20110109_AACQVG warren_b_Page_004thm.jpg
833ddcc2a6535d5fd0f2507c96019142
c1b63b79f4f22b5cd2d0e069588eb1aa5d782eea
46163 F20110109_AACQUS warren_b_Page_043.pro
fef88cc9e14ebd81c71ae99f0c762aea
85dc616b13d82251483c7d18274e666efff4db10
107868 F20110109_AACRAN warren_b_Page_077.jpg
29665bc8c824c0bc2a8365aa964a2fe0
019abd39a688f358666cc0246eb7bb66198a7660
20718 F20110109_AACQVH warren_b_Page_104.pro
0604b51422a535edad75d1b67a23485f
fb695d1a221ead773ceb90d303cbe392335ea13a
87321 F20110109_AACQUT warren_b_Page_095.jpg
89a6a72aa808fa2c49f96b1e3b67cbe0
1a8e87822e665c54841fdf9cafa8f926a62f5cde
576434 F20110109_AACRBB warren_b_Page_010.jp2
bf935945ff064dbad4dd87d98bce5380
cd93972c61e09a9b122c9f654636bc2d9b902561
103561 F20110109_AACRAO warren_b_Page_078.jpg
0be222054f6ed914a6ec9d246ffc6ce7
9474cd87944a478d2b453153e93ad2878af80e72
8315 F20110109_AACQVI warren_b_Page_081thm.jpg
02038efa75121e4262170028d7a8ce8c
12b212cbe238056e48aa77cf4145ba6bc7044478
9460 F20110109_AACQUU warren_b_Page_099thm.jpg
0c9e6ea638de9f03ef794df5d576dd68
26f280f8b374374ee066b9ef347b5ce84c953c58
87610 F20110109_AACRBC warren_b_Page_011.jp2
e15826d11e2482db36ea10eaaf589f0f
6432e07ac805511781117dcf0f607e7d13bd1f56
27001 F20110109_AACQVJ warren_b_Page_083.QC.jpg
1c68307cf8e301bade7ac9b29d8d8f0b
c2f9c3219bf85d93abf6e5021f64aeb7ff9a15a0
96202 F20110109_AACRBD warren_b_Page_016.jp2
aa65e8c063de35dc022150335b572911
df4e534bbf9c7ba5f1317dfad8b1ebab26ea7eb2
100929 F20110109_AACRAP warren_b_Page_080.jpg
bd2390b5bc7558a1ce78ce44ae156bbf
331a12babd19e8d557dc4b1fadf09d201b25f164
F20110109_AACQVK warren_b_Page_029.tif
f09b529c846f21f47a234d122f950a3b
2259e2bb0a7fb82a5bd091eed84c65b8274cefdc
13328 F20110109_AACQUV warren_b_Page_003.jpg
ee8ce0ab93e813c527b7438057a86b54
f299abe2756187d0e4dc43fe98bf2fe859543d41
112642 F20110109_AACRBE warren_b_Page_018.jp2
bb190d59b08cdfcb1cc5af0688d68272
7438524288a86cb21ac62d52ce91443f4c7b1a24
109191 F20110109_AACRAQ warren_b_Page_085.jpg
de3c7b4c9a0fec347d69d67433557f07
aea581f95725c929695eddce50118a6253650f5b
51897 F20110109_AACQVL warren_b_Page_046.pro
0f587a4dddc026919b7b7ac741ffb5f8
699ed05d1bb2362d3a9fe4dcfe4772aaaa76e539
30213 F20110109_AACQUW warren_b_Page_029.QC.jpg
799d7e29a962e1c062d331f775103217
414d48c1c03215a1b42525d16eebe6985f40f974
113081 F20110109_AACRBF warren_b_Page_020.jp2
27c325b71144960a94c709476e5cb582
d6afd991c5a390ef647a36a2d52b41ec75abf4b5
104503 F20110109_AACRAR warren_b_Page_086.jpg
341edbc9ee795e1f6bb18ea7457e0489
e4b73db5eb56165ca29ccc91906f7c302b4da762
F20110109_AACQVM warren_b_Page_010.tif
14940c77f7d262248d985cc70dfcb2df
14e9e68f14b34024cdd08545684c9bf41f1016bc
30574 F20110109_AACQUX warren_b_Page_030.QC.jpg
0258df19e270e1facc9f6a66dfe33d59
a03b6ff1e8240823f1b490009713acfe32281c53
112309 F20110109_AACRBG warren_b_Page_022.jp2
e6fab91693d6af48845f571a2ddfeea5
56394c0069f7d60dc63fa50ba4acacf1a308a75b
101613 F20110109_AACQWA warren_b_Page_035.jp2
d690f331a6c6adbf56dc4bc6225d0cfb
d6a7b5cacdf04c9dc9e7423ae41ae547bd9470b3
107326 F20110109_AACRAS warren_b_Page_089.jpg
b2467af45f9c1328d2a34039912780f1
174dcb0a116cb293dd50ebca489ea21b3a6ad0aa
1757 F20110109_AACQVN warren_b_Page_013.txt
20efea316afd9e67b8a11c675a274852
803a40b779b28ca64b64de3d3e875e1ea438f1f5
109398 F20110109_AACQUY warren_b_Page_038.jp2
4453a5ea302a0f62fbe1ccf37c655e81
fc4d529aa354f314619001f4c298d8cefb2d1f15
108414 F20110109_AACRBH warren_b_Page_023.jp2
9947cb99bd088e388b720498478d924c
53c2416fa5dbff4a3e8b85e83f7486ff2f7910f2
20305 F20110109_AACQWB warren_b_Page_007.QC.jpg
d61da8c1b578d6c1112525e1c1f7702a
d7ab22e772ec70c3cce650a0445c8aa16f62a4da
97111 F20110109_AACRAT warren_b_Page_090.jpg
cd85873e26a5e2d78e0361343a1201e8
37fa036753bbf116820221b29c5f8fa027b52c06
2806 F20110109_AACQVO warren_b_Page_099.txt
423924097d6be33f596c7f226a350baa
9bcbfd29cf42178fa2888f2300d5da0f4dc3d5f5
F20110109_AACQUZ warren_b_Page_007.jp2
27cff07ff2e77ece0f5507334cf777bb
e30a2504bb24e9c97b5a3cdc7c03514542df6b87
99165 F20110109_AACRBI warren_b_Page_024.jp2
3f195992d82cfdc394ddf46e89418525
ea6cd67e79877f467fc0dceb07dd0c4c23cdec7c
33782 F20110109_AACQWC warren_b_Page_069.QC.jpg
73b3582330b0312ccac4672bba6bed02
edaf9faa32af4b877142e8a85c8c26c5e252c5ca
108748 F20110109_AACRAU warren_b_Page_096.jpg
4ff66e9ae60b470591a2fd358032ca31
cec4cc3ef60334abfa66c07529def8541ddb8b5e
111432 F20110109_AACQVP warren_b_Page_081.jp2
3835699df6c264e61148dfe746ac1702
5ec66bef104eae36b2efb1064a295c7560416006
105169 F20110109_AACRBJ warren_b_Page_025.jp2
e1952532822e6a16ee42204be4c5a0d0
1daa7f256105ae91c8912fb74edeb8f404cd250f
94339 F20110109_AACQWD warren_b_Page_013.jp2
d9d7143fda9c7535ca0f30c2b83e1f91
38861a7abd8f65ab3f6f917a01a81363b9a7d124
140031 F20110109_AACRAV warren_b_Page_099.jpg
842fadc53fff2721dd1be9c2cb52183c
dc0fe4f7cd1cfc08031137d0de0935ecf13cd0a0
26787 F20110109_AACQVQ warren_b_Page_015.QC.jpg
cd35722effde15999184bfff20984b0b
a5b0f88af703a27fccfec852f278f8a804a1a2c8
111426 F20110109_AACRBK warren_b_Page_027.jp2
8b03c5616fc40a2d50cf08088547fa79
123f2a41d1803409da65dd4932bd1c56b56f828c
32667 F20110109_AACQWE warren_b_Page_042.QC.jpg
df8215334b02959d14b12632c568f541
438b19e4931591a06f3d6c719084304b12b6f5b7
133389 F20110109_AACRAW warren_b_Page_100.jpg
c52653390de3d8dd329cb3b80cd445a2
c281e06c22d6ea4b11d25d682d5d08c565d17694
78863 F20110109_AACQVR warren_b_Page_061.jpg
e24fbdc061939cb460cc150fa7e9fa6b
40b43f7ef34042a6ee334dc3b786cf566943caf2
658365 F20110109_AACRCA warren_b_Page_059.jp2
9331e5a0b6d1f58f58b4788752f0fb41
65aa05e09993f7a26b61275875394af12a031835
102894 F20110109_AACRBL warren_b_Page_028.jp2
376f0a3579a2a39b893132ca923b62aa
c0625f90600e126d347d6ece701ac13b7cd8d268
36113 F20110109_AACQWF warren_b_Page_070.QC.jpg
0d53513a868316922588254ead41521c
af7aaf508a8ce16bff8a7d7aedc39b3b985e4e7f
125762 F20110109_AACRAX warren_b_Page_102.jpg
baf9b2e62d49fac1d6283fa028cb2600
6738cda6b65479670f9d1f5b158b5ecbce59b4e6
48707 F20110109_AACQVS warren_b_Page_066.pro
ce483e1c70111d4a1148dfede05bb6f9
3daa61b93b3bf070644afc27e447ad107737c373
1051985 F20110109_AACRCB warren_b_Page_067.jp2
c13193bfc06ab3d77c3fb3c040795575
0906a9a56c462034056fd6374e3837187b8734a1
101113 F20110109_AACRBM warren_b_Page_029.jp2
2f5606773cadac4d5524e193374914e9
59efff20c926802a3053c44e156768248d039e39
87472 F20110109_AACQWG warren_b_Page_064.jpg
469073addeace988adcc31b631fa03aa
82d10edff75b8b71ef2751a4f06b66019707fdbb
107679 F20110109_AACRAY warren_b_Page_103.jpg
da3f2399c1eebb0768bc4d11902d79ad
8b26fe05973f8a1db07d18a38ca3746bb7b6694b
2056 F20110109_AACQVT warren_b_Page_086.txt
ef308e0d01c362a7aa26f87e2d311538
7577c01050ea77965968c80634cc4defd073fd50
102185 F20110109_AACRBN warren_b_Page_030.jp2
85c2e6988e44b6a9fe189e43c346e34e
5bb04a59a329078fdf80a07bbbfe816b5ddc2c17
111256 F20110109_AACQWH warren_b_Page_026.jp2
c630adbc3761e9b1b8be178247d493d1
b8492e0ff0afa33c2c13ab62412e3d7ec8feec3a
1051972 F20110109_AACRAZ warren_b_Page_006.jp2
f61e184b5f25efc757271f27b3accaa1
78f33b232a8c957d7ad08a9b4f2590bc94fda8fb
121096 F20110109_AACQVU warren_b_Page_071.jpg
3e729218f583d17121784400dc16ff78
afdc27fefe8d5a90d0882a0fb0f781dabd2f2467
1051936 F20110109_AACRCC warren_b_Page_069.jp2
dba78d6be8ed0126f4c48d5d3e19c433
8199c357a2c8dfb120eb599295472bdd4afc30c4
113873 F20110109_AACRBO warren_b_Page_031.jp2
71c2c256f8ed5be0101a7e4f5a4ae1d5
9d6526d8d9b0128072db9cbc3db0f66c5d2606f3
52391 F20110109_AACQWI warren_b_Page_089.pro
9165c867d51fa88c54f844bd7615e078
6869907699c99cedcbc5f70d4412f6b00707261d
119 F20110109_AACQVV warren_b_Page_002.txt
3daf97f5a368a6e9378842ea12397ac1
4c1583343b481f2bb5d14c41f0a2c278db219731
703662 F20110109_AACRCD warren_b_Page_074.jp2
765aab52c7e56ab74f6e03c6178d1e14
111223f9d6814737cdbff3dd756c98a86a5fae06
107156 F20110109_AACRBP warren_b_Page_036.jp2
0883baa5063b9e82bd2add3430b25dc4
00cf59be820e05fcf70d986979173c964a66c79f
799047 F20110109_AACQWJ warren_b_Page_060.jp2
644129f83acb402827ac713b7ea80d32
58833b30b7711e9aee826a5a46e3fdca03562f6d
1006519 F20110109_AACRCE warren_b_Page_075.jp2
418236215d1791aa0f9216248c0a8567
506c1a78e79821c5d518768692d560cd12dbaab2
110464 F20110109_AACRBQ warren_b_Page_040.jp2
1c484f161f2190b707ba1152ee7acd56
04e4d9d487810b7e8d52c8f3836f56b2f231b3df
8423998 F20110109_AACQWK warren_b_Page_054.tif
9b5326c903451e1ff007485f8b7df275
8c8b7ddb278fde7438e89a387fd0ee6a6d277880
1407 F20110109_AACQVW warren_b_Page_060.txt
41449afc95c67ab6300ee9b6a1aa8ae1
a11d9b3d02e5159689b4270da7d8f0480f790948
92542 F20110109_AACRCF warren_b_Page_076.jp2
e4c27b9eec220bc9806ad1631f7364d8
f6c86ad306041bb1e392f1e41813270766a2a3b1
109217 F20110109_AACRBR warren_b_Page_042.jp2
feb6ee265a6fad11f107db45759d3635
d9accf3e3f47dbd5ec639a9df283a678c20d589d
F20110109_AACQWL warren_b_Page_056.tif
3886628392b438b838f6460620b11d4a
73d0cd6eb62a9a42703da02bff3902e19b377f1a
1782 F20110109_AACQVX warren_b_Page_016.txt
fd578a6756c885eb002a4348eeb89fb9
dd4218138ef99060c8f1532171f70844f75af696
114996 F20110109_AACRCG warren_b_Page_077.jp2
137bd68c2f94e6e62c95e0f8f2174884
ae152976d3c0869943088f9e31dadf3e744dcd5a
110262 F20110109_AACQXA warren_b_Page_019.jp2
69771eb65fcf1726ea5b0d359f96daff
5562a542cffdd7e31b5d4d3ddfd9f982f79be360
100760 F20110109_AACRBS warren_b_Page_045.jp2
2f314f2fdd689659060c09287ced5ce5
4a45043a52d14a3b5b04a0d9bc38494d2a6d75fa
7687 F20110109_AACQWM warren_b_Page_029thm.jpg
89943480f6aed3896e8129217dc0ea00
f33c236d95c5c34ff14368228219d7a294a573ec
F20110109_AACQVY warren_b_Page_050.tif
ffe92187d1cd87b56c964a342f8ca6dc
6db7470d169dc3614fa1090565ea58b478cf8b62
107734 F20110109_AACRCH warren_b_Page_080.jp2
4ca7692377a7be8f1cc4b98c4eda310a
91ef65ac107b7cd60c881bae5f9fb05048d42b4c
16277 F20110109_AACQXB warren_b_Page_008.QC.jpg
74d5c66dcaf5db0eccf259d2df6bb9f1
8c77c31dd1db2f9f22f80dfd68ef462425495836
107543 F20110109_AACRBT warren_b_Page_046.jp2
2a8f03bfe21433bf6a3c831a47756967
fddf83ed2e5af31d7338237ee1729aa799610873
1364 F20110109_AACQWN warren_b_Page_052.txt
a566af1fc422092090fcc28f99af3e33
f1199cb5d47c893af03c24b433b5ba79323cbb46
99051 F20110109_AACQVZ warren_b_Page_049.jp2
defa70b9278512c9ece8083802c937f6
7bae60525121f051332e55d3ab6f4be59e2c6623
109400 F20110109_AACRCI warren_b_Page_082.jp2
8d079a123448fcf8e48d3e0d8837d8d8
cfbb241be315eb66464c8438c0da3ff96556a360
8414 F20110109_AACQXC warren_b_Page_066thm.jpg
78a4d52e2557c4dbb86feac3274d1dab
a2f25df17d73f50e665888c2c80f4a4b4272b70e
126968 F20110109_AACRBU warren_b_Page_048.jp2
515d898aa596fdfdd8a8e4ddac01f4e0
c8f69609e8bc145266159844da33bbeb3dc3e183
30579 F20110109_AACQWO warren_b_Page_045.QC.jpg
3fa1e6ff9b40fdab51b8262889e322e2
9ae3d654d924c3d7234c63df3a9dac5f57c21d7e
112962 F20110109_AACRCJ warren_b_Page_085.jp2
34f30d7e5a247ced86fe8fefdc58ac0c
531250eebf507dcaeb1826e2012b7f0b2290da65
7461 F20110109_AACQXD warren_b_Page_055thm.jpg
3b1ee1c279f116ed56884ebf6ae095a6
1fef554f27ecbd490489811695c7ff227433df42
93330 F20110109_AACRBV warren_b_Page_050.jp2
1dea20cc7c6b2aada94d07d628a0857f
b7a1b5b5665e777f8ce42c6d40181a8b22450a7a
522 F20110109_AACQWP warren_b_Page_001.txt
258694f499a3bc395f0db89c9dc5a162
dc36db81c0021207d9a63c42b9702556ac7bdb45
113228 F20110109_AACRCK warren_b_Page_087.jp2
52deb3b7031d2c2daa04825ec93ed995
143f5c83f01aeeab03ce2ac8717f7c0c28a280c0
F20110109_AACQXE warren_b_Page_019.tif
25542868077cff73fe2d89df89bec64a
d6f90b409d231e65f13f322862ed3642576ee16d
106427 F20110109_AACRBW warren_b_Page_053.jp2
820a7620a6eac32b4ae4e82fc55d6c47
76760eeef587fb89242153965a40a9fea7fcb058
1788 F20110109_AACQWQ warren_b_Page_067.txt
51b447dff8eb730a179b14bf81bfb388
d71404b961521f8ecb42194d8c05e786ef695084
109695 F20110109_AACRCL warren_b_Page_091.jp2
d605e06a4d94606e59a033268679dcd3
48fc5d64b6ca552e270acbc8e8c42e772c8870cb
F20110109_AACQXF warren_b_Page_097.tif
d014a80be8e2402fe17bc2f9e16f5d14
53bbc62b8778a7256d967972c0da14813cfe40b4
1014427 F20110109_AACRBX warren_b_Page_055.jp2
938639ba36eafaca3e1db7d1fc6e5f71
398bf4ac32805e3749251adc4b9eab77647ac91c
103697 F20110109_AACQWR warren_b_Page_088.jpg
648fb7a35ed367da4e0d8afe20e299cb
4a9cc09284208bd00e7b3a1d486ce6c04c10e0c2
F20110109_AACRDA warren_b_Page_031.tif
883d284c19d75bdb9cf1b9cd3d10e6cd
03d9b3e45fe1ba62efda12f2b57272152f758d30
803011 F20110109_AACRCM warren_b_Page_093.jp2
ca048ff9f810184174f0cef5d15280a5
7ca7086bba24ba0142984e6a25804fb44fffce34
49245 F20110109_AACQXG warren_b_Page_080.pro
3df8d6d565d0a01134d20db420d9580a
9863f8e37674b640a04c9d24601760dbde9bcf78
109508 F20110109_AACRBY warren_b_Page_056.jp2
1f5aa26fddc9eaeb4b727914cbc1b9bf
3b9afe350755d620d332062b97758d50b91d2647
33575 F20110109_AACQWS warren_b_Page_066.QC.jpg
4278c5b6c2c9a3b992bf1d8e8ac5b7f7
936786205e178f7716d4872ab0599fcf1eb98693
F20110109_AACRDB warren_b_Page_037.tif
93d8fab39b227b2d33b091d938537524
d515c193140adb76b689a14fb0b278af6c0255e4
843341 F20110109_AACRCN warren_b_Page_094.jp2
4bccf5f836a2ee7a7f8ee6fa7e546637
f7549e4b3b5cf9830bde0c2ca82340f92f293a95
120524 F20110109_AACQXH warren_b_Page_097.jpg
86a0c5570ad80f86371c64d218e88e91
1d1e3824f8a65a740e659d0da8404e7a69135733
103423 F20110109_AACRBZ warren_b_Page_057.jp2
60ee052a1c51cd7f9737270d2c485e0f
3aeabebc6104c05476e52fed1ed73ba9a14c0a4d
51943 F20110109_AACQWT warren_b_Page_092.jpg
7976997cd3b7330a61e02bdc083534da
45f515079205d4224119c2321e62aa46243d3282
F20110109_AACRDC warren_b_Page_039.tif
885972cca8209d341cc223d215d80b7d
563ab92d36d4cf8af0ea50650e7af212257f1aa1
1051973 F20110109_AACRCO warren_b_Page_101.jp2
b08636d28367ba92a7725818a565b3bd
4de22a48bd1d1ea1a1dc981faba8d032445aa426
105031 F20110109_AACQXI warren_b_Page_068.jp2
822d6577d70ceb31f354dbd9a6db934b
088ce783e3204da737d7a69e6052b13aa3acd43f
34682 F20110109_AACQWU warren_b_Page_095.pro
945475e38689518fc9a8a2948348ae76
b87d3f1d8563e4793e7d57dede7f6a055dd956fb
48267 F20110109_AACRCP warren_b_Page_104.jp2
a4fd541f06ff832460c872b5e1f9145a
e9ed9652b07c4d02666504a428ab5382abaafa13
2050 F20110109_AACQXJ warren_b_Page_026.txt
850dfeab0a7a97289c405eb3a017c4f5
bb33a41bee56f614018e25e6308fd2a6be141516
F20110109_AACQWV warren_b_Page_034.tif
d43015d477912dc6e76751287bed4215
ef35fd1116f792b767c000df200e49e34989a51f
F20110109_AACRDD warren_b_Page_041.tif
5acf246c91fa1204ecfc58b396096b39
c435e09c3ccabdf28735423c8581685bc2d9d86a
F20110109_AACRCQ warren_b_Page_002.tif
6601db872b3c1e84aff2fba14364c4fd
4d7430c8f001e71c8bd4582079512228abdc7388
8335 F20110109_AACQXK warren_b_Page_017thm.jpg
158ef918c6cea5aeb4d677f0c4a50953
91092b6a586c152cf69efede8f26aacbd57ae0a2
F20110109_AACQWW warren_b_Page_035.tif
fee92963ec9e50f55edc8d48c2544bb9
271053a9f4c054b899f2f2f0d51b52ce779cd48c
F20110109_AACRDE warren_b_Page_042.tif
e02ac7ea0db93d1ece7a732bc11edf3d
c52c5dca99cee3058d27a4e55569c13c87df1ef1
F20110109_AACRCR warren_b_Page_003.tif
b0baa17f6bd41b35b5d6750a6f954cb5
5fb26fb76f8ad865f5b530e1deb24b022f7f6738
1166 F20110109_AACQXL warren_b_Page_073.txt
7cfba230f85220a7a83ee42e1b24a7a2
22f976d8fc33c973be0abd846d43d5fce8600750
F20110109_AACRDF warren_b_Page_044.tif
de9f45804e0df3f7e59f57625a04aea8
b43884f94aee2f89f21eb7bbc57049bd407476eb
1036096 F20110109_AACQYA warren_b_Page_090.jp2
d0409ef4f6a83d160a0ef3cb31db3660
1f386ac5f14d2fe8b175f662ed52d0de4377aab4
F20110109_AACRCS warren_b_Page_006.tif
2d5564a30a75f9ddb850884edd03acb2
441829b6a4f21c167d441d7e9bd85e47a84cd277
34818 F20110109_AACQXM warren_b_Page_012.QC.jpg
aef8d53dd733a24b886eb3690db351b7
ac42a65810b9d56d42ff5a7c269b26320ab73929
F20110109_AACQWX warren_b_Page_032.tif
6cc422350e08165046bce1d5c1c40bd4
dbdf55e260f37e3234a4032061c7ca9258846e1e
F20110109_AACRDG warren_b_Page_049.tif
3bdfb68b3eb0510fc33e729768e34125
e431ecfe1075e1e8f579749d3c609acecc3c84b1
46915 F20110109_AACQYB warren_b_Page_044.pro
1ddceb72f51ebb3ac4aacf989ce24243
77df936eed2bb52027388a4e15efa148c1d77a08
F20110109_AACRCT warren_b_Page_007.tif
330742e5d5abd2e0f20fa51d42a2acac
2b36a317bf692fe7ed0b1d9de1cd7b9862fa49f0
8389 F20110109_AACQXN warren_b_Page_023thm.jpg
229b065bebe98abf38392d31d7c51bee
476aae68cbc67176943c00c2e921d13dfb281cae
1863 F20110109_AACQWY warren_b_Page_029.txt
0f65b6ce427b8084dd709bee1f08e534
254e4cfd3ad9b0b73ef35316a0cce77d5f8439fe
F20110109_AACRDH warren_b_Page_052.tif
75896b1552dd23b1deef0908e4d55ffc
ae4b68a5112e5e79ee378f169a8bdf9dd85873b0
7125 F20110109_AACQYC warren_b_Page_052thm.jpg
1f143cc6fc8b9d52c44e5837ccc7add2
0b9b7e3443fff46bb29b943966a1d38ad1f863c6
F20110109_AACRCU warren_b_Page_011.tif
44241d24bb49a5f17613270eaf218af0
69354991373f3df89813847121ae297d1f3f53de
32298 F20110109_AACQXO warren_b_Page_082.QC.jpg
5bf03ae204adf6b99b23e9eb2baab658
799a699b126c16dc000ee1f052071f4e5023eabb
127826 F20110109_AACQWZ warren_b_Page_098.jpg
72d137f6b05f93363635094660dda511
b53d360f741b27c0eb4bd2d12a84820c2db2620f
F20110109_AACRDI warren_b_Page_053.tif
252da9f8ccb5512dbc6952d56f616666
22238d4f8c0faaca9a35de4b147f2f70d29586d7
84901 F20110109_AACQYD warren_b_Page_052.jpg
71c9219b3f07d05e5b72e0cba60d1a0c
d3d15cd704103e4ce4f8ed25c1c0f66ede4e1a55
F20110109_AACRCV warren_b_Page_012.tif
8b703f80fce0ba9921b7523e034d2211
755e49b6f2d2b05ed6403f3aacc4be24108d63f9
89106 F20110109_AACQXP warren_b_Page_083.jpg
6abc2f03fe162d7f540b7bb15527a0db
a6435d5f2088971c5a873e4370faf0e6d38f1620
F20110109_AACRDJ warren_b_Page_061.tif
155cc4f70e092b71e5e2f74f35d42e59
632055b150c4f7d7df3b6f4173bc0fc8f5b9d37c
122398 F20110109_AACQYE warren_b_Page_072.jp2
7dfe68d56ffdbeb5be4dbf2842763027
7e67cf9ccce48877cfa09048568820cb9be1783f
F20110109_AACRCW warren_b_Page_015.tif
0534f24948a4d642e46d3117fd5656b3
71da5b5e3554a9135acd699439cb5eddb0b44f8e
8164 F20110109_AACQXQ warren_b_Page_036thm.jpg
c260228ed4258f20c9e682d523b9c25c
b134441aa51ab3e68772926a2d9e979552d21c97
F20110109_AACRDK warren_b_Page_062.tif
08ed0ab0026958186a3ea99827690c92
bb76013f20ccb9f759c620732f06c2d1d7ce5c70
1977 F20110109_AACQYF warren_b_Page_027.txt
e79231f0ae7a7c6253ece61f1bf7643f
e898f26fc9eb7099c992425e15115b9205173f1b
F20110109_AACRCX warren_b_Page_017.tif
0eefdcee471d19d37f1cae5d0c9f07a2
10d932a0915b9f3581304e2ad804d476bcd410ab
34184 F20110109_AACQXR warren_b_Page_019.QC.jpg
d480bd98aa2056ccf395526cb5f70f8c
73423b46c090f4e619fb344f5167fdd9d7dd38fe
36755 F20110109_AACREA warren_b_Page_004.pro
8cd963e89e95652ba9fa6e474ccbb480
0d8c4bd03b79dc2b210b3eef872f0401854c39df
F20110109_AACRDL warren_b_Page_065.tif
2a12dccf7d1fe5a293bd65e729440142
7bb22bbae7ae8bd4efabe8b0d8ca292a0d17bef6
2026 F20110109_AACQYG warren_b_Page_081.txt
c4afebfd77d6b3438d114201d4208b5f
90613732b86db4fb822a3f501c50361371397f8d
F20110109_AACRCY warren_b_Page_020.tif
956b06e97c311009988b153a30f06430
5e21569dbda38175edaf579d1309fe8ff67b6782
2228 F20110109_AACQXS warren_b_Page_069.txt
3a3b1688427f690001d03ebdefb056d6
b6bc004e86f4481a202dfcac56cb3db40b4038fd
82120 F20110109_AACREB warren_b_Page_005.pro
0ce4a6f524ca55f03c7c6af9648ec1e8
3027921a47db7124184ab7016133db3d9d13bf44
F20110109_AACRDM warren_b_Page_067.tif
bb8a0b86dc39d9f133b0761ef4fadc72
fdd5d13b9add954604b18842591a681e5cd46c46
107444 F20110109_AACQYH warren_b_Page_079.jpg
0db3a7ed9aad9096e0ffe2c5358910c5
6f3c645431944b8ee92cc9c6e5bbd310a0038192
F20110109_AACRCZ warren_b_Page_022.tif
c675d25d7987d2f81ce946646ae1e350
8d2baf8e769477bc0c3a2ddc17f52fc993f49008
79694 F20110109_AACQXT warren_b_Page_041.jpg
d785f58668cb5322cbbe5472639c3cc9
ff6b5672e1b72191b27e84862190b1e49e315772
30177 F20110109_AACREC warren_b_Page_008.pro
ed7d8e09187d7649478880a1e4621484
d26f6359801e395be64d6b317339afaf472e73ce
F20110109_AACRDN warren_b_Page_069.tif
ac6ce68217445a14c8a13f76c681496a
da7aba10b5fd998becf0f9be1e6921948cd6e47b
42575 F20110109_AACQYI warren_b_Page_016.pro
aa4a01b6bfaa7ee587008fa573750eb6
15f6121ae64f701f159b905a777f94e319227527
7201 F20110109_AACQXU warren_b_Page_064thm.jpg
95070e5ade50cdfca4c8358c52d242b6
a1ec7ccfa1f5518402e378440ef56de94649a445
60891 F20110109_AACRED warren_b_Page_009.pro
9a9fdae0f1d8f2e66013fcd51d28a0cf
ded07a2c3b66cb3889f2d22c1f8646943556e263
F20110109_AACRDO warren_b_Page_074.tif
a6243ab074fba4839e553744e18431a6
1a7831cf533a495fdb8e7e032f82fb3b502d700f
2141 F20110109_AACQYJ warren_b_Page_007.txt
d2f1fc4afe7684868fbdba94c6d336c4
59643e8c42626bde333111b72f5b0af0aeac76c3
F20110109_AACQXV warren_b_Page_059.tif
8211ab8e44edd5038d347ad6c506df1d
b0628dd5b2e6f609d31e9965943112dab095bdbf
F20110109_AACRDP warren_b_Page_076.tif
ffde3230c2619a5aa5a99ea4456bf1c8
35e7938a39a549624736a12fdc2bc3c4c50f4f8f
F20110109_AACQYK warren_b_Page_083.tif
76bd4b88e00100b6124ade157c7ec0e3
9f0e3abb2abad0db45b8189183bfc8905474dddd
31291 F20110109_AACQXW warren_b_Page_021.QC.jpg
250dcb708e12d3a96a3d40515edd1692
ad1afb34852f22b9a52934b40bd12ad50a8ac26e
15322 F20110109_AACREE warren_b_Page_010.pro
0ea2f76c36bafc6a56847d3d0ff62dce
b571cb13f218c3e462a922bb8aa3f13fa2c89133
F20110109_AACRDQ warren_b_Page_080.tif
91c13dab37546b34bb03ed26a9bf8c8d
8bd3662762c3de42a8853379d4d00038c240f489
31098 F20110109_AACQYL warren_b_Page_090.QC.jpg
0b6e6f9be2cbe68b76223c209fef3229
5ebb91306dc8c46b0dfb8d1dea67b3103d732dcb
F20110109_AACQXX warren_b_Page_009.tif
7d82435535d094a9b49166ce4927cbf1
4f6789f676977438395b98c693dfc6d650a5aea5
38632 F20110109_AACREF warren_b_Page_011.pro
8e2d94ec60ef42d505ab314594397e12
d04876c2579415199f4041f9da34b674f7f7bda1
F20110109_AACRDR warren_b_Page_086.tif
ecdef44fc5a9241b628b4747828fbc36
4ee4d15a9536faca625aabfeedaf56c98521b059
102497 F20110109_AACQYM warren_b_Page_009.jpg
3192bb7550159ec64bf3d1f240c24d73
36f0c2fee1eefdbeac9cc5b2be5c4a44d3b43d31
50953 F20110109_AACREG warren_b_Page_012.pro
ecdcb73e5714badff7d8c9089205c2af
9551960fb506506eab98a6c52f5f357eb163d870
30650 F20110109_AACQZA warren_b_Page_035.QC.jpg
eded8b64e679f71ebcd2f5e27ad4c988
f91987df17ba58481af94e8c33371b3e26356fb3
F20110109_AACRDS warren_b_Page_091.tif
339f78a37b8756827dbd9087d1434f14
c10184473510cbb462fe2ed91b2ca4ffac42a7ed
F20110109_AACQYN warren_b_Page_024.tif
3f18da90aa3d2936ab0cb57a0b22969e
e5ab0c05d577b3e3a8763562d0daecc2f6722293
F20110109_AACQXY warren_b_Page_008.tif
2ac27441a72c8858541907829fc821e4
562520dc3a2116e2498fc16d8c7f8f39e44fe55c
50595 F20110109_AACREH warren_b_Page_014.pro
719c60afbd3c0d8cc5ab5f03efe377cd
666672dbbbd6be89bbcde7b5e779904f87659377
2698 F20110109_AACQZB warren_b_Page_072.txt
f33cccad25247d275538cba0c7f186cc
9c11752b1dd0ebef8797e0a26d28a7c8ecd4ac2c
F20110109_AACRDT warren_b_Page_095.tif
4b9dfd066c94a208df4c3e15c7eda7b4
0cea066d96687398cb6e0429b64687749606f855
55078 F20110109_AACQYO warren_b_Page_103.pro
fc230618353660f6dd993be31ce7fc64
bea3600ee54c57cb4d2b57885098c08d65ade11b
96383 F20110109_AACQXZ warren_b_Page_037.jpg
0ccb51f226716a0bce4dc62fe3682959
7c4239a5fbb09646dec61a04c88f957a08669949
51393 F20110109_AACREI warren_b_Page_018.pro
2fc608952425a53c5bf00f579f468a3f
8fc222bfa51962d9fba9eaa7388aff2b151c321d
114216 F20110109_AACQZC warren_b_Page_070.jp2
ace48db45b7c016a8026397b4c41965a
22e35b4fcad5aa071110f051d1ca66615b6bc18c
F20110109_AACRDU warren_b_Page_096.tif
cf8dbb4c4861f4a67227ecb646515fe5
baf1460d4b9edd48f30e8f8c79a59a9b2e1021bd
49757 F20110109_AACREJ warren_b_Page_019.pro
6640466c85696f406b8904514e49ccf5
23305696136b2acbadbc82b3baca316c9d9ecae7
1591 F20110109_AACQZD warren_b_Page_015.txt
0d5deca38279c97bfcb9aa0c09cc8b63
b6df466c52843325790ebb77f34a245a9dd19d18
F20110109_AACRDV warren_b_Page_098.tif
2ce2e0f09e085009a1e65dbf963ad3fe
9e80b434234993cc5cd6aecd3bb2fcdc677e5cbe
8117 F20110109_AACQYP warren_b_Page_049thm.jpg
1658dae8d2876058647e7f63f18e198b
3f6942b58e85cd577b94b7f4cd6c35546b40a27b
48454 F20110109_AACREK warren_b_Page_021.pro
1b7268dc552c230866f93635e6512a4c
ae760a235a8c0a5a738cb32257e7738859861c5c
108438 F20110109_AACQZE warren_b_Page_088.jp2
2d23d8be6d80e4158273ff832325a087
3fa1c29df5efb8973e7f7bef19fbb94bd6b6ecc4
F20110109_AACRDW warren_b_Page_102.tif
4f8c868349db0d37ac860d3362602881
be624ff64417e41d45713bbe1c3a1ca0d2bfde98
57877 F20110109_AACQYQ warren_b_Page_008.jpg
ee75028601e5b6762274b45c08244228
409e04a7fafa88659f17ef0789ba8801a8603365
40480 F20110109_AACRFA warren_b_Page_076.pro
d29d1644d9f8b599e3a4e6694fac37c6
fe0c4039d6a56d49007c455a43c48ed927c2a1bc
52146 F20110109_AACREL warren_b_Page_022.pro
432c7431dd4e841b57baac3936abf39d
d1fe1076d72ef926b41b2d5b2e2e9a0cc9b9e059
104318 F20110109_AACQZF warren_b_Page_022.jpg
50f95dc182c16e3b24e043a52b402c28
f63b434820e31591d9a13f3f1d51029b977fd281
F20110109_AACRDX warren_b_Page_103.tif
77461aab0117b2009520bdd653313abe
d8d97baaf24c4a0a2788b897a0a7ad2971d76f42
101383 F20110109_AACQYR warren_b_Page_075.jpg
f2d7b2ab800821df34111f3f526e3c26
31fedd1cb37906b449b1013bee3e8d145e292c0a
51006 F20110109_AACRFB warren_b_Page_078.pro
2929223d21354490adb35819c6409925
21d0944b0a6ba5e69b2175a47918a6312114bbf4
48983 F20110109_AACREM warren_b_Page_025.pro
c9f3af223c2654973f47e1875df5edaf
87acf4c5cb6c934952da2aa3dcc93b2597c81190
7490 F20110109_AACQZG warren_b_Page_030thm.jpg
2b6b0cd333a21a4a92feb8def9e46058
fd7f606eccf568f62b149934f5aa3499fdb7bc40
9511 F20110109_AACRDY warren_b_Page_001.pro
1dc471a5336731d4b0b1f4ec8cbb9dc4
f0aa5e60743a686d23371f95915d9031a64d4481
1963 F20110109_AACQYS warren_b_Page_019.txt
2dad102910823fdb91734900bfbf724b
a7ee4af925b361ea430976df3154bc218d76196e
53430 F20110109_AACRFC warren_b_Page_079.pro
e6523593f8dafef306bb5019e692e074
14513f3b68d3f24d922b977189edf4258ebf9d64
52246 F20110109_AACREN warren_b_Page_026.pro
f34005561d30ff84e346d8cf498ffecc
021fa25b3e2f7be046b96780440183339341a186
2109 F20110109_AACQZH warren_b_Page_078.txt
8958c664ab83166c33ad7eedf28411be
4a3605552ecd9f77af8a5ab6133623837340592d
1279 F20110109_AACRDZ warren_b_Page_002.pro
ef5d102fc415804b7298b036f12b8ae8
4071b9a78aa8ab3a6affaa8303fef2a08c168c8a
17366 F20110109_AACQYT warren_b_Page_092.QC.jpg
ce3b86730608024cc98e139371893648
90c4f163f1447083303533331d0a026f85d86888
48980 F20110109_AACRFD warren_b_Page_082.pro
1ce90292486fc57870765742a21d67ad
2e5098ca1db4a3184415f0d92ba5a5655fd47a2b
46974 F20110109_AACREO warren_b_Page_029.pro
37bba87f9bcaef71e6dc8256b778c29f
419c78cbdfcdb230d57ffa9b8d807b8d9fde7e5b
869482 F20110109_AACQZI warren_b_Page_095.jp2
754c019b2e68dd77978a25a0b09ee2d0
0de49a4d9e3fbd3062df9f85057622646778a321
53262 F20110109_AACQYU warren_b_Page_084.pro
7cd9008e9c1d32c137188069c4e1d445
4ddc41fb7c9ace801e1aa5cad3302e1736fceb9c
F20110109_AACRFE warren_b_Page_083.pro
6a978d9e423f359fc5978289257e0ab0
2eb4ca0e310b2f78e116685b22a32c418932d60d
52856 F20110109_AACREP warren_b_Page_031.pro
f48edd1e3fcacc8b2585fdf4e269b9f6
ef972d3413240d5f508f43566626860a29503a32
1941 F20110109_AACQZJ warren_b_Page_075.txt
f7c85134c98fc874f383d0f2352aef47
050723dcc14adb9ca815e5140ec8705f294e3b8b
F20110109_AACQYV warren_b_Page_016.tif
a5452cb31bf85d74fdee8d5985dad356
cfa92f18d193de1bc2711f73003289e9fece3c41
47130 F20110109_AACREQ warren_b_Page_037.pro
0a40facc320fb195e27645f3caa77f1c
c614d279ed49aafa732a97800fcfc0cbcbbe355e
34165 F20110109_AACQZK warren_b_Page_088.QC.jpg
a72b53acbf58675771735697efbc7301
96afd4ff668727df2012e513fe7fa6e69fdc2103
33486 F20110109_AACQYW warren_b_Page_055.pro
88e30457ded39ab4c7f053550dd87b73
f722d9271b094c80febae719d2a1603261ca40e8
43402 F20110109_AACRFF warren_b_Page_090.pro
80a949646358cc225ab629f2690d1ed4
f7a3d480768e899e650b17077c1670ad8c1f9a16
45485 F20110109_AACRER warren_b_Page_045.pro
fc6ca409a1aa072acdf89438b958b2c8
d685d0dd5a8a96b2585b4181b6559a454b471eb2
114496 F20110109_AACQZL warren_b_Page_034.jp2
4c624ff46d17039b6211a48a4935137c
408aaaa65677d67aad521bfae5c8ec0139e55cd6
F20110109_AACQYX warren_b_Page_030.tif
4f62ff6bf06f2387d646553a4ee11aac
a7efd6d92cc2426fa7c1c172d693f8a7b6c17d96
32644 F20110109_AACRFG warren_b_Page_093.pro
5cbb1e86911cb998b705785ab537d5bc
7a7b2bd733588e50519fc4530dbcd9a425299b3a
28905 F20110109_AACRES warren_b_Page_051.pro
dd30720f9845cd2580559ebd4c10689d
a6c24a5398a935fd7fecded388b077a940b129b3
29795 F20110109_AACQZM warren_b_Page_063.QC.jpg
b4f408ac9ba6b5c259d1e4216a84bd9c
c2fcbfccab58e6b732b0f7127fe844274fcbbaf8
F20110109_AACQYY warren_b_Page_077.tif
a573d382b91425c44fce41ab343a3d34
1e4c8ba98257786a4aeecf409282c11533b0ed7c
34714 F20110109_AACRFH warren_b_Page_094.pro
dbd332936715dd58237980d338f4ae0d
84ff1e5763b363cee4c2d886bfceaffe0722ce40
32767 F20110109_AACRET warren_b_Page_052.pro
89c029d5543e7a1d411451abdc568843
974734cfe47a8c70db4fea3b57237e0911fffc25
59454 F20110109_AACQZN warren_b_Page_101.pro
54d77c0364c16a50f1492042cb344817
e845bbfd83fce15d487567132efd7b7cb3a913b2
64558 F20110109_AACRFI warren_b_Page_098.pro
258a746cbce9390021e8449a84a680d2
808feea59a3141f3424f2a221b852f867adf13e6
48733 F20110109_AACREU warren_b_Page_053.pro
77e34974fcfc2929e0791097c718960a
cd9c81372f24f56c3f50bbeba8b29ac5e116a6df
F20110109_AACQZO warren_b_Page_100.jp2
0ee0c60330e92a3121732f07973f35df
21126c839cdf6e0375d6a499ffdf7be3b8041b82
39247 F20110109_AACQYZ warren_b_Page_015.pro
f449c37d143aa374f4468ea2cfc9763b
b65e34a663eec106d7061bdc94f26102586b505b
69303 F20110109_AACRFJ warren_b_Page_099.pro
87897600b91a8f2782737f89f06c38b2
271992e7deea103a156704b604e4249e18a4d695
29946 F20110109_AACREV warren_b_Page_060.pro
21c047c7bd24e23fcb1f76edab6a056d
1595ea811f72d9988450916fedc2b5c4b0d2884e
F20110109_AACQZP warren_b_Page_027.tif
7da0f11c9995c2ab1f130c47d3b6c678
2bc4e4a9b12b0f3b75cec848b58e72c80b2c3be6
289 F20110109_AACRFK warren_b_Page_003.txt
bdbc50593129b54b1f294c35e499e2e0
e927b62af2c17d31604ac959750e3dd3b16a87a0
42354 F20110109_AACREW warren_b_Page_063.pro
4bbcc859893642345e73dd1e4abfea87
54fa511a88c52ee1b58db401adb230e6a1cd1865
99681 F20110109_AACQZQ warren_b_Page_082.jpg
b8dc5996d9783fad6449326bf7b71bfe
70fa40db79f65489b145e5b1af128a2b3cea8072
2028 F20110109_AACRFL warren_b_Page_018.txt
c52c21f69bf556d2d2820549bfd64580
6fa887ba42c5a615488eab5b44ddaa6dc8fc31a9
37249 F20110109_AACREX warren_b_Page_065.pro
f0eacb2553e0f726c76edc8d0c8aee51
aa95f05117aa42555c2cc4e53fd105d3aa19badf
156807 F20110109_AACQZR UFE0001186_00001.xml FULL
92bcb8c65d4b152d198b0ad364c6cd8c
2448b4ba5d5d1e6e53e0d5e0c5ae14524ad7d3c4
F20110109_AACRGA warren_b_Page_085.txt
a2ec03455b0927f2905e332122eb7d2c
81c05c782546d2fe9556f80b1cfc80bb9dfa2a00
61592 F20110109_AACREY warren_b_Page_071.pro
95a6459681bf2777fa8b8029f581c597
24821a2723979d0a84c8d8a5ca9cc07497c2bb99
F20110109_AACRGB warren_b_Page_089.txt
f41c86192a53c9a3e7c51a9c1b0afc13
34b4729f9ae580cc0b16ee1db26f84a27c20c436
2059 F20110109_AACRFM warren_b_Page_022.txt
54033039a9c36216a8b679217b3948e6
6913566bc9dc156e37264ebe3ed55fd00cea8b8f
27215 F20110109_AACREZ warren_b_Page_073.pro
1133362f7a1e7425950353c5e41bd3f8
9c4c52d92aafce162b29fb38f38fe4dda510656f
2047 F20110109_AACRGC warren_b_Page_091.txt
01c8b306828af3caebacd6760e2757b5
2916479987adc594b9241ad54e59907d572c2c91
F20110109_AACRFN warren_b_Page_030.txt
c2e0bb04a8e2bfc56be74ae10222b2e4
effbae139521ef1b6e29434e4dc896cab4950e63
28821 F20110109_AACQZU warren_b_Page_001.jpg
5180e6b523598a8743b783e5a98fadfc
26bedcc53b40b831f5667ccc6fd7485f970b6c3a
1517 F20110109_AACRGD warren_b_Page_093.txt
26ab49f5ae4af1df74733c90c42f1c6c
fd502f38202ad88c6855fd1da78a9b9b9ded5f70
2087 F20110109_AACRFO warren_b_Page_034.txt
71e87f0445ca729564d98d0bfd3a6ae1
a362e115c06fc6a34c742401c6a0f6a17d8498ce
77856 F20110109_AACQZV warren_b_Page_004.jpg
aa4de857673274c309c11f54ab9d9ebf
6541dc7c352aef8e9cbbc0418b60c9b292d863ea
2321 F20110109_AACRGE warren_b_Page_097.txt
d4ca7174986e731b25454117a5231b20
5c7cecff79abae7d5f190b1deb8dcb615157cb5e
1808 F20110109_AACRFP warren_b_Page_045.txt
f48138f00f3aa986ae9f1d1a4d172a84
05354c6c59f45819d3f8bd8fb80d0abcd74e7825
79987 F20110109_AACQZW warren_b_Page_007.jpg
1acb8a1ef7f31f4323ca8ce32883a56e
6450207b1703645be652af5e86b25925edc620aa
2606 F20110109_AACRGF warren_b_Page_098.txt
018c214e24ff0386c0dd1df300ecfffb
2f6a03bb7bb09a0ffc3b435cad1e0a5c171b13f1
2377 F20110109_AACRFQ warren_b_Page_047.txt
6c6a3e1b1ffcf0792a137b63a1d187ab
86b7f0d4bc6a6edddbf109a98c143ada9a77e9f1
103951 F20110109_AACQZX warren_b_Page_018.jpg
2d764aee8928e336d204082545136791
04f860304535ce3ac6e35ebda3147a3e080c8b65
1678 F20110109_AACRFR warren_b_Page_050.txt
6b38eadf505c382b866ae1d11483141c
cc6fc4d857774c4ed7fd8631397c7d2f0e69fe5f
102313 F20110109_AACQZY warren_b_Page_019.jpg
2878adba9a787265b2ebbdacdb654ebe
f63ab28019d38e8a82b79c2b2927b3e5a0973b5c
2409 F20110109_AACRGG warren_b_Page_101.txt
1186ad33372798b7aeceb21f063fc40d
80f1a91ddf3c471f1288253b94a2e541157ab103
F20110109_AACRFS warren_b_Page_054.txt
3012b6618fc699dfc153c18176ba69c6
3432831f9a406a51c5ae23e4770506c727f5f446
105925 F20110109_AACQZZ warren_b_Page_026.jpg
494a0642a6fb7b082f18f75082a6a7f1
31a15ff3c521d9392bed83b4bec2cd0487bfa0cd
121230 F20110109_AACRGH UFE0001186_00001.mets
08b3f8aa4d6e361d8ef72fd5c308322a
37e60c053df5c1e6397d6c7aaee8db61d3c316b6
1406 F20110109_AACRFT warren_b_Page_055.txt
16fa085c165fe77b058fe5ffa23da2bd
b7dfa92c32443af0ef4972b8499984c5841d9056
1713 F20110109_AACRGI warren_b_Page_002.QC.jpg
c375afd67dd048aea40ffba2250801f6
3e93063832680000b09abf73d15ab1774adbf54b
2004 F20110109_AACRFU warren_b_Page_056.txt
964619ce00bb92c715c3bb3368f8675b
1bf94c7ecccc65bb9f156e76fde0d46ee23fdbb4
1222 F20110109_AACRGJ warren_b_Page_003thm.jpg
c5bd877460ba65e30ce10dae3eee3cbd
4e2206e1b3c6f8559f24ef3ddbe7d7f7956da396
1707 F20110109_AACRFV warren_b_Page_058.txt
638e868a19abc325ff8d4615cc7c925a
ea0f88fc1c69e6c94d69de1d92460a02d4ef5377
26297 F20110109_AACRGK warren_b_Page_005.QC.jpg
837e290da932598d38af53e9fec2af14
e01bc7ce9333d86b4876f8ef29f455a3a43699b7
1485 F20110109_AACRFW warren_b_Page_061.txt
755076904a48fa978809986e6cded5ec
9e8246c553aed4415675322e1d41bf285f6787d8
34103 F20110109_AACRHA warren_b_Page_027.QC.jpg
579bfee34e8a84159633d52bb6364522
ca26b77d379126daf4ebf35c80ad1cdc321531f9
7553 F20110109_AACRGL warren_b_Page_006thm.jpg
4a7f7422f2e4b3bdf05501fd2ffec2f2
d3ade9ab8231f126321a8b8fc09423e05b473c56
2067 F20110109_AACRFX warren_b_Page_077.txt
2e53afb1877f0c14467f39045c9fc836
3145a09f580f7c9f93cf5c0d04887c1a81f6a51a
31559 F20110109_AACRHB warren_b_Page_028.QC.jpg
8d58ef4d01e297ca342366847de5035e
36c0f834e61923c56fef55a0a6bd1d5c19fc3385
4969 F20110109_AACRGM warren_b_Page_007thm.jpg
85752f313e26bbaeb123f6995335cf6d
c01f05f6b5c0019ff2d858cf468dc4b471840ce4
F20110109_AACRFY warren_b_Page_080.txt
f9e8b6621be96af924786dd436cd0c9c
2574329c8d96825c498fdc1683b6a6693b846ac9
34756 F20110109_AACRHC warren_b_Page_031.QC.jpg
8da1b0bf575bb909aeee6f4b04dab44e
272c2cdcd9026f79dbdf079aa2c3f3e6a1590f60
29148 F20110109_AACRGN warren_b_Page_009.QC.jpg
a4d5de3707f16461d23f4e52719fd829
acb72296a19971203043c736f04a3abf29720f76
430 F20110109_AACRFZ warren_b_Page_083.txt
794b23081cadf0f6cde05ac206280e99
1924ab07ccdd84bd513ab127a113cba8be2969ee
8560 F20110109_AACRHD warren_b_Page_031thm.jpg
7c236417a4540dd21985f73c21622a9b
6aff1221b935496605cf9aabe2055eee9270c20c
9620 F20110109_AACRGO warren_b_Page_010.QC.jpg
38df573f301fb389957cdc5f2d80a7c9
1096baa188dcf907e9b50cac08b7f5eff272a6af
8290 F20110109_AACRHE warren_b_Page_033thm.jpg
5dbb02873cd8d63ace9ea327a59d1aea
b0dbc4219b0cce98efa18056948fe18f05a8faff
2430 F20110109_AACRGP warren_b_Page_010thm.jpg
f1f18c492bbe6cbd1b4870f258cae6b8
14e94d5d1089025f96cbd350439fdb522fe56819
8675 F20110109_AACRHF warren_b_Page_034thm.jpg
49ab4542753d4756c00bd3573e5d84c6
d2c52e62500bbcfa7697daf8afaeae853abfb35a
8576 F20110109_AACRGQ warren_b_Page_012thm.jpg
9be1b60a72843a557c4d72acc8767532
36486d15ccef91de05ea622e8ffd24860fb73b31
33189 F20110109_AACRHG warren_b_Page_036.QC.jpg
390d68808ce440035f1a95f3f34b798d
10bd9efa64c70214f11e8cbe0226af15a82b1f77
28864 F20110109_AACRGR warren_b_Page_013.QC.jpg
d4670abf32bdb108b1f6325d3488625a
ab48764c069bd2a4e9709b377548250c0da5b153
7046 F20110109_AACRGS warren_b_Page_013thm.jpg
3dfc61b0a0eb9346c6ba71c8f7f4e021
b4e8838d3a3ec9f3a6e3fa8ca9aa434d96e78fad
32862 F20110109_AACRHH warren_b_Page_038.QC.jpg
7edfe57d012f5677b2c9564d62c8c886
3b8fa12249cb25a8c0cdd52c8db8fb3e0baf3666
33399 F20110109_AACRGT warren_b_Page_014.QC.jpg
a4a226654acbea2c539a09e6cc8bd556
9556776f52911d981ae20c98f8bb998b71688ea2
8002 F20110109_AACRHI warren_b_Page_038thm.jpg
8140f49fbd33e772b7acff917a24a694
2af7cb5f133cab7cbc8b26a1b98869d9ce0550f1
34467 F20110109_AACRGU warren_b_Page_017.QC.jpg
5a69ad0d503011f665cc1cf2fff311f7
8b73e2713fefc07a3d37da7b66389a941785ab72
6476 F20110109_AACRHJ warren_b_Page_041thm.jpg
28218c7b2e21c0626b0dc18dd7713d4e
7af0355d5a9e74400b24f1374f32ecad8ee9952a
8438 F20110109_AACRGV warren_b_Page_018thm.jpg
6e2e79d910098984dfb591d4b49b0b35
66fbfc3b2ff165e8e1539ae9aaade95282eb66cd
30715 F20110109_AACRHK warren_b_Page_046.QC.jpg
1e3036f87b92b0be71d10a5fa9b529f4
f58b5d63fd347a6f340642f0d9bb35cee0644fb0
8379 F20110109_AACRGW warren_b_Page_020thm.jpg
26b3ffb47be9c89f2eb3d6783e3650e8
9e2caf3f993b03c1b722a7c0e213eebef963edba
7465 F20110109_AACRHL warren_b_Page_047thm.jpg
db636ec525841ff3441464131eceafd5
63bb8498399d86440848ab917baf48377cf01e13
34650 F20110109_AACRGX warren_b_Page_022.QC.jpg
dcb124f0422b372136df8b43ec1f7332
4bc1fa95c280457c7e6e81e04f456ae35b875e2b
8558 F20110109_AACRIA warren_b_Page_070thm.jpg
98a0bf3dae274bfbf55a02713d593d13
2155bf567acb119c639db3d13e78c9e3f98d2ea7
34773 F20110109_AACRHM warren_b_Page_048.QC.jpg
e40ecaa5f36b7a8523f5d4601d2d0983
5a42a2bd9ee4b81d23cb3cb34a601053cd732dae
8583 F20110109_AACRGY warren_b_Page_022thm.jpg
bbc91e5751b9e42f25d7ba40450bf92c
fbe1f53b00cfac8d4cf17da1bc87dcb41f28d071
8298 F20110109_AACRIB warren_b_Page_072thm.jpg
375ed09cee78286ffba416c6426a82af
368a2973ed9a9eec311feaa40e9149b1573d2f03
31092 F20110109_AACRHN warren_b_Page_049.QC.jpg
ae8369a4506e3091bc9d5b79be7524d4
ef08257fea58acf5643e1a1fb660581f0d1887ff
7647 F20110109_AACRGZ warren_b_Page_025thm.jpg
057e3441a1aba25a892f9852a191780b
32d5f517c16a73ba3f8d7539bad52af3546c51f1
21990 F20110109_AACRIC warren_b_Page_074.QC.jpg
6839ac53c617f99235b41df415c98653
19e8302ac71fce626ced19196de7a547308ad323
28852 F20110109_AACRHO warren_b_Page_050.QC.jpg
25cc5a156ac69325104c3460be1aa972
e3de4b3ef10501e72ece0046891e1849dfe679ea
5936 F20110109_AACRID warren_b_Page_074thm.jpg
b463a560dc48433134cf1acf5c298bdf
d361fe983337a21955f5289abf294f90bc3d30f0
7396 F20110109_AACRHP warren_b_Page_050thm.jpg
b5bf6ea06c46a9aac90ad515d683f172
f6bd7e0a7961da03e8c923834692cfe5f248ea91
6924 F20110109_AACRIE warren_b_Page_076thm.jpg
e36dd0e1b091888c33a2d6341eb9f09c
da44bc59a1a20f811779ed7b45a766d0341062f8
26586 F20110109_AACRHQ warren_b_Page_051.QC.jpg
6849952cf2fa99cccc045c3c02de2e25
012fc62bd7bf2f1c14c82c2f278f342bee9285b8
8273 F20110109_AACRIF warren_b_Page_078thm.jpg
e7cfe44a3d1728d922e27cd526b5f591
4002dc354761c409ea0ac5c770b3d3c8cf542d9f
7312 F20110109_AACRHR warren_b_Page_051thm.jpg
45320183c7bc7c47014a2266bcfea0ea
a32eca0cfa2abde21f1ca8aa64090f3af7fd89d7
34523 F20110109_AACRIG warren_b_Page_081.QC.jpg
7f3fa231beeb19877c1ad1c397815f85
304f2ab74216ff43349e5ae86ab2e2949559d5a2
8334 F20110109_AACRHS warren_b_Page_053thm.jpg
83c63b50a56698e06e393572f98e64b9
3083e87f7e747e86f6e4f655e93a6206a9e65a7d
8094 F20110109_AACRIH warren_b_Page_082thm.jpg
bc853d21be33a84ec4e12466a25fad91
99a7dfe7ee12a45c81a595ed6b072f2f9bad7b65
26940 F20110109_AACRHT warren_b_Page_055.QC.jpg
368991feeecfb858a063ab81074ceb68
6832732fa0a388d3eaff907779727588ded239bd
25175 F20110109_AACRHU warren_b_Page_058.QC.jpg
e1119e5d8be03f1bb29a93de15fbbacc
bd38f1bcf42e494197d253d27cc4ae3111f4a736
36289 F20110109_AACRII warren_b_Page_084.QC.jpg
2df00dd5e39c83ef1e2b87ec2fe65f7c
35704a35a21de13b18b2eaa5bfdf765d4496920c
6664 F20110109_AACRHV warren_b_Page_060thm.jpg
f0c6cba4171be47b090a783f151cdb9e
8c0993919aae4b3eb8eecc130b06549884a2e1d0
35967 F20110109_AACRIJ warren_b_Page_085.QC.jpg
9a6799b992b36f1e0a2f7710fd7cafab
50bad1486055091883f88bc751f563d1fb6f155f
28466 F20110109_AACRHW warren_b_Page_064.QC.jpg
995e2becab71e69daaad100235b9ec47
883e96b07ed31073590ecc1b1b453647947b6843
F20110109_AACRIK warren_b_Page_085thm.jpg
07515d383ebc0724c497bd77a5c52c19
5a4f302500dd940d2ed141d65c259fbeb0461b6f
7849 F20110109_AACRHX warren_b_Page_065thm.jpg
b3f656728ae029ae1949b5df89cdd995
95a6dbaf1828a8b8c2718106772596f2598f50ec
33581 F20110109_AACRIL warren_b_Page_086.QC.jpg
7d0c558ab55e8ce35f9cd08759d7905b
bc20a8393c36d32d261cde5c06717eb6f652c838
34768 F20110109_AACRHY warren_b_Page_067.QC.jpg
b05067b7ee2614c0fa6e5b98caa2920c
a52c9106f2086ba35c26248ccb18d2aa728036bd
8406 F20110109_AACRIM warren_b_Page_086thm.jpg
431d2d1d9ea6a9afca706a9ae35eaabf
e3a4e13f58416eaddd35a313367a2a423b5960f4
31528 F20110109_AACRHZ warren_b_Page_068.QC.jpg
a91967abbc3578f02994f9f85dc6bb1b
e7306815194c0d18d42f9ba2a1b72a5501b92bc7
35707 F20110109_AACRIN warren_b_Page_087.QC.jpg
26713fa25073436405a0a08ca0a2005e
2ba8679743b53ac14fb2ae7fb1338e49bc88de62
8396 F20110109_AACRIO warren_b_Page_089thm.jpg
c36bfa9ac47368dfa7fb6db6246abc17
6a5691398e79555dc0440456a12e2fdbacd47980
8323 F20110109_AACRIP warren_b_Page_090thm.jpg
4bc817b9b0be8839a4f0f1d305c83698
4882317c57e79eacd61beb7980f1be9a696e66d9
34099 F20110109_AACRIQ warren_b_Page_091.QC.jpg
7a59bd7938ae5bd71481ee6719ac0cc0
43aec3ecaefa2ee7e0e4482ac13e3a1f1ae1afd0
8254 F20110109_AACRIR warren_b_Page_091thm.jpg
ee184a21857c35cb631250f96ec58d5e
3358fdabd6e4c268fa734781aa22ce653fce2865
4221 F20110109_AACRIS warren_b_Page_092thm.jpg
0fdd230752a53514d9184ef2e102688e
623a51f9fbe61cba5cc3962399c36a7b83b1203c
25317 F20110109_AACRIT warren_b_Page_094.QC.jpg
80265679eac93499fee6cdb67274239e
e525fd6779f61186514336b36b5fed33fdad956d
6770 F20110109_AACRIU warren_b_Page_094thm.jpg
8a2cfd91189958c3ff0019d8a32598f6
4a8839a3aa943dc34136908656dd567541339948
6996 F20110109_AACRIV warren_b_Page_095thm.jpg
8db48fbf7c148a2152e24ff17d699d4d
de372fca2de19498ba83db9ed8ebdd7181f68149
37409 F20110109_AACRIW warren_b_Page_100.QC.jpg
f2b95c5f5ae70582f513a3f06bf49813
332074ea3489fc3c99c0151f8f15f5d45dd529b9
31277 F20110109_AACRIX warren_b_Page_103.QC.jpg
587e506b6bf78870e35ab1f9b5e69b6f
4f93154651e12d253803ce08feef9734272fdd6b
3987 F20110109_AACRIY warren_b_Page_104thm.jpg
e0bc4eb7f29b461da2fefca1a3c6eaf3
fdaa2e2e7fe91ece0960d08fe16a59497fe4e27f
2225 F20110109_AACQHA warren_b_Page_046.txt
3313031e8030e814bd720ad51518c589
95139b3758045bd4dafba6389d297ba0a60b5add
99044 F20110109_AACQHB warren_b_Page_068.jpg
293cd07f1f0d2ca1959e2994e09ce21c
984dfc0eda145df2fd6c698b4819d509fc3959fb
2218 F20110109_AACQHC warren_b_Page_049.txt
ee51805ee0782d7544f46a8bc921424a
90d6b05ab497e19e91c9e33e8659152324539e20
F20110109_AACQHD warren_b_Page_040.tif
05313eb822e96a1a3b8876229565de19
37c8762f43244d488a4c6add1f381ef6110c364a
62947 F20110109_AACQHE warren_b_Page_102.pro
bf9119939212b50f5247592fb351f7f4
df914648b6d39570ed980d61301bb0bbbb5fd6d6
83553 F20110109_AACQHF warren_b_Page_060.jpg
00ec4116b052014868a2ac50069d5160
fa618a50c7a13d20d311e4459fe5f599bd98e833
F20110109_AACQHG warren_b_Page_013.tif
5400fe522048a616899dde28a633995a
b1e23056f1d7b6e00ae4ebd1b71c106d7cea542a
2045 F20110109_AACQHH warren_b_Page_070.txt
4d6d31a1ea7d7ee0b8c932f25b804909
82516654295abed5e6f4f58d885b8e4a21f9f860
F20110109_AACQGU warren_b_Page_058.tif
13347e2b1a5968e5818fbe00e003f960
17e0e3ac41df5aabdf8f161322e21ad923d54d06
15279 F20110109_AACQHI warren_b_Page_104.QC.jpg
1594e40c20f2336ea52253f184910170
ee170d20f5331fdd57f8e08e457c32a6bd3da848
F20110109_AACQGV warren_b_Page_066.tif
363575c329fcb7833c2e120225a2d6a5
42725663bb1272bf0f253ceed70d3b03a801aa1c
30980 F20110109_AACQHJ warren_b_Page_096.QC.jpg
07a8cc7ca629717d0b1071d8e69e708b
5d4377997d3538ee7448ade32d08062e810941ca
50938 F20110109_AACQGW warren_b_Page_091.pro
85acbd86f4079efcc9a4723255356c1f
a775765b6033e63bf2c3341e6ac2ce32dcf8e27c
29072 F20110109_AACQHK warren_b_Page_075.QC.jpg
f72b0800ee11cf9745cd90bb65ef2b8a
0af48d4b037fcc9e87a59758415326291696ea25
109893 F20110109_AACQGX warren_b_Page_084.jpg
da2b662ca7c8a1bad2a240d928f0b041
1c6bac16470c2e355981fff3858d2f5af21cd582
F20110109_AACQIA warren_b_Page_099.tif
697e000c5a4dd3745a37c18a9359494e
b2989d2c04867f32dbfbefbb266e29b53800f457
28245 F20110109_AACQHL warren_b_Page_076.QC.jpg
da156909eb9e46177e5e15f07af7e159
d07a90e35ce00e000e5eb2d00a963a27a95f0aaf
2427 F20110109_AACQGY warren_b_Page_009.txt
e25bcf8b824bf845e666af6ee75e9c4a
0ce95a690ddaddbfae1f3b5e76c2d5c00427a490
96688 F20110109_AACQIB warren_b_Page_049.jpg
a2db874b906c9b2bc686e1f6702a3c95
043daa672631ed1d029dc31fc4fbe97efa9d0ae6
8551 F20110109_AACQHM warren_b_Page_084thm.jpg
4b56e8930fd4714f0b540f883f53777e
4bf0c2ed537b6c9542ac85e97bd8e0bdeca10c4d
26240 F20110109_AACQGZ warren_b_Page_093.QC.jpg
c7ca533098d2e36ec86502ef4c3d62fb
5b8adebef5677501bae54e2096077041574ed13a
99817 F20110109_AACQIC warren_b_Page_065.jpg
c84533b18f04e3d1451ee950606e2b61
bd563e5234b46652ae7a509ca0b098649e4248fe
2019 F20110109_AACQHN warren_b_Page_023.txt
51a1870c4137009de3710cf675aa5a31
a9a606a16f0a7544826b46bbaf0693d1e930327b
88129 F20110109_AACQID warren_b_Page_050.jpg
960cc34e4a4f258f3691e90dc2ccd5c1
93816ada954e44b7fe91cd88d9341b3ed5264037
33955 F20110109_AACQHO warren_b_Page_072.QC.jpg
e50bf2d70fe2bc68edbbf798a04a03f6
44a0e8c5540450e4a6a690c46540784a8b0aa4cf
81123 F20110109_AACQIE warren_b_Page_093.jpg
7f1f059e016acbded679578bdf1cedc2
1eff0e79318d6ac4fb13635e38e15d721c6efffb
30835 F20110109_AACQHP warren_b_Page_057.QC.jpg
22421b0d5194181a559708be787268d0
ea03994048a3b500848391afcb24b7ffc886d8a6
94113 F20110109_AACQIF warren_b_Page_045.jpg
68689d0b32430e59c8e71ce4507c4136
f66700445f9276ea643cda9cc6d5fb092bbb3c1a
7751 F20110109_AACQHQ warren_b_Page_043thm.jpg
301bc20c5d62206e96c77da832ac108c
17cde684dc3be24475412dd73c8a6badda6dceed
35021 F20110109_AACQIG warren_b_Page_101.QC.jpg
83b2f98d10f9ba0e25476260da9a64df
da85e12599c7901d1742fe4982eb0b37a0bc345a
2092 F20110109_AACQHR warren_b_Page_084.txt
1766804a299364decd5905aea9f73763
9543f3d1cb1a406bdfbd68a0ae812df6afd65da4



PAGE 1

COMPARISON OF CONVENTIONAL CULTURE METHODS AND THE POLYMERASE CHAIN REACTION FOR THE DETECTION OF Shigella spp. ON TOMATO SURFACES By BENJAMIN RAY WARREN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2003

PAGE 2

Copyright 2003 by Benjamin Ray Warren

PAGE 3

To Nikki, for your undying love a nd support; to my parents, fo r never losing faith in me; and to all my friends along the way, for w ithout all of you this would not have been possible.

PAGE 4

ACKNOWLEDGMENTS I would like to thank my committee co-chairs, Dr. Mickey Parish and Dr. Keith Schneider, for all of their support and guidance during the past two years. I would like to especially thank Dr. Parish for seeing my potential and inspiring me to return to the University of Florida to pursue this masters degree and a future Ph.D., and Dr. Schneider for always having an open door and encouraging my questions and ideas. I would like to thank my parents, Dennis and Linda Warren, and my fiance, Nicole Sanson, for their never-ending love and support. I would further like to thank the members of my graduate committee, Dr. Douglas Archer, Dr. Renee Goodrich, and Dr. Steven Sargent, for all of their assistance with this project. For all of the laughs and good times, I would also like to thank my friends and colleagues Norm Nehmatallah, April Elston, Raina Allen, and Tomas Ballesteros; good luck to all. Statistical assistance was provided by University of Florida, IFAS Statistics, with special thanks to Jamie Jarabek, MSTAT. Assistance in acquiring tomatoes was provided by the laboratory of Dr. Jerry Bartz, Plant Pathology, especially Mike Mahovic. Finally, I would like to thank Dr. Keith Lampel, CFSAN, FDA, for sharing information and methods which made this project possible. This project was funded by the USDA-CSREES IFAFS Grant number 00-52102-9637. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT... ....................................................................................................................xi CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................4 General Tomato Background........................................................................................4 Use of Tomatoes as a Model for Detection of Shigella spp. from Fruits and Vegetables................................................................................................................5 General Characteristics of Shigella..............................................................................5 Recent Outbreaks Involving Shigella spp.............................................................7 Prevalence of Shigella spp. on Produce................................................................7 Survival Characteristics of Shigella......................................................................9 Intracellular Activity of Shigella spp..........................................................................11 Invasion of Epithelial Cells.................................................................................11 Gaining entry into epithelial cells................................................................12 Intracellular multiplication...........................................................................12 Intraand intercellular spreading (actin-based motility)..............................13 Early host cell apoptosis...............................................................................17 Toxins Produced by Shigella spp........................................................................17 Shiga toxin....................................................................................................17 Shigella enterotoxins 1 and 2.......................................................................18 Traditional Microbiological Media for the Isolation/Detection of Shigella...............18 U.S. Food and Drug Administration (1998), Bacteriological Analytical Manual.............................................................................................................21 Compendium of Methods for the Microbiological Examination of Foods.........22 Nucleic Acid-Based Detection of Shigella spp. in Foods...........................................23 Colony Hybridization Assays..............................................................................23 PCR: The Basics..................................................................................................24 Bacterial DNA Template Preparation for PCR...................................................24 Nested PCR.........................................................................................................25 v

PAGE 6

PCR for the Detection of Shigella sp..................................................................26 Real-time PCR: The Future for the Detection of Shigella spp. in Food..............28 3 METHODS AND MATERIALS................................................................................29 Preliminary Trials.......................................................................................................29 Preparation of Microbiological Media................................................................29 Enrichment media........................................................................................29 Solid media...................................................................................................29 Acquisition and Maintenance of Shigella cultures..............................................30 Adaptation of Cultures to Rifampicin.................................................................31 Preparation of Optical Density Standard Curve..................................................32 DNA Extraction of Stock Shigella Cultures........................................................33 Crude DNA Extraction of Stock Non-Shigella Cultures.....................................34 Acquisition and Maintenance of PCR Primers for the Detection of Shigella.....35 Specificity of Primers..........................................................................................36 Analysis of PCR Product by Gel Electrophoresis...............................................36 Inoculated Studies.......................................................................................................37 Acquisition of Tomato.........................................................................................37 Inoculum Preparation..........................................................................................38 Inoculation of Tomatoes and Subsequent Recovery...........................................38 Experimental Design...........................................................................................39 Confirmation of Typical Colonies on Tri-Plates.................................................41 Assembly of Tandem Filter Funnels...................................................................41 Nested PCR Amplification of ipaH gene for Detection of Shigella...................44 Recording of Data and Statistical Evaluation......................................................45 4 RESULTS...................................................................................................................46 Preliminary Trials.......................................................................................................46 Growth Curves and Optical Density Standard Curves........................................46 S. boydii UI02 wild strain..........................................................................46 S. sonnei UI05 wild strain.........................................................................47 S. sonnei 9290 rifampicin adapted strain..................................................49 Primer Specificity................................................................................................50 Inoculated Studies.......................................................................................................52 Detection of Shigella spp. by Conventional Culture Methods............................52 Detection of Shigella spp. by Conventional Culture Methods with Rifampicin Supplemented Enrichment...............................................................................54 Lowest Detection Levels of Conventional Culture Methods..............................58 Detection of Shigella spp. by FTA Filtration / Nested PCR.............................60 Lowest Detection Levels of the FTA Filtration/ Nested PCR Method.............62 vi

PAGE 7

5 DISCUSSION AND CONCLUSION........................................................................64 Preliminary Studies.....................................................................................................64 Growth Characteristics of S. boydii UI02 and S. sonnei UI05............................65 Evaluation of Primers Specific for Shigella spp..................................................67 Inoculation Studies.....................................................................................................68 Evaluation of Enrichment Protocols....................................................................68 Evaluation of Plating Media................................................................................69 Analysis of Lowest Detection Levels of Conventional Culture Methods...........72 Sources of Variation Among Trials.....................................................................73 Comparison of Conventional Culture Methods and FTA Filtration/ Nested PCR......................................................................................................75 Analysis of Lowest Detection Levels of FTA Filtration/ Nested PCR.............76 Optimization of the FTA Filtration/ Nested PCR Assay...................................77 Predictive Value of Testing for Shigella spp.......................................................79 Conclusions.................................................................................................................79 APPENDIX GROWTH CHARACTERISTICS OF NALIDIXIC ACID ADAPTED STRAINS.........81 S. boydii UI02 NA adapted strain............................................................................81 S. sonnei UI05 NA adapted strain............................................................................82 LIST OF REFERENCES...................................................................................................84 BIOGRAPHICAL SKETCH.............................................................................................92 vii

PAGE 8

LIST OF TABLES Table page 3-1. Stock Shigella cultures...............................................................................................34 3-2. Stock non-Shigella cultures........................................................................................35 3-3. Primers for the detection of Shigella spp....................................................................36 3-4. Temperature programs for PCR primers....................................................................37 4-1. Lowest detection levels (LDLs) of conventional culture methods.............................59 4-2. Lowest detection levels in which S. boydii UI02 or S. sonnei UI05 was isolated in 100% of replicates (LDL100s) of conventional culture methods............................60 4-3. Lowest detection levels and lowest detection levels in which S. boydii UI02 or S. sonnei UI05 was isolated in 100% of replicates of FTA filtration/ nested PCR....63 5-1. Doubling times associated with exponential growth phase of investigated strains of Shigella spp..........................................................................................................66 viii

PAGE 9

LIST OF FIGURES Figure page 3-1. Spot inoculation of tomatoes......................................................................................39 3-2. Stomacher bag with inoculated tomato......................................................................40 3-2. Assembly of tandem filter funnels..............................................................................42 3-3. Vacuum flask apparatus with tandem filter funnels...................................................43 4-1. Growth curve: S. boydii UI02 wild strain................................................................47 4-2. Optical density standard curve for S. boydii UI02 wild strain.................................48 4-3. Growth curve: S. sonnei UI05 wild strain...............................................................48 4-4. Standard curve of O.D. (600 nm) of S. sonnei UI05 wild strain compared to log plate count................................................................................................................49 4-5. Growth curve: S. sonnei 9290 rifampicin adapted strain.........................................50 4-6. Standard curve of O.D. (600 nm) of S. sonnei 9290 rifampicin adapted strain compared to log plate count.....................................................................................51 4-7. Recovery of S. sonnei UI05 by conventional culture methods...................................53 4-8. Recovery of S. boydii UI02 by conventional culture methods with rifampicin supplemented enrichment.........................................................................................55 4-9. Recovery of S. sonnei UI05 by conventional culture methods with rifampicin supplemented enrichment.........................................................................................57 4-10. Representative gels from the first step nested PCR..................................................61 4-11. Representative gel from the second step nested PCR...............................................62 4-12. Detection of S. boydii UI02 and S. sonnei UI05 by FTA filtration/ nested PCR...63 5-1. Differentiation of background microflora by isolation media....................................71 5-2. FTA punches in 0.5 ml microcentrifuge tubes.........................................................78 ix

PAGE 10

A-1. Growth curve: S. boydii UI02 nalidixic acid adapted strain...................................81 A-2. Standard curve of O.D. (600 nm) of S. boydii UI02 nalidixic acid adapted strain compared to log plate count...........................................................................82 A-3. Growth curve: S. sonnei UI05 nalidixic acid adapted strain..................................83 A-4. Standard curve of O.D. (600 nm) of S. sonnei UI05 nalidixic acid adapted strain compared to log plate count...........................................................................83 x

PAGE 11

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science COMPARISON OF CONVENTIONAL CULTURE METHODS AND THE POLYMERASE CHAIN REACTION FOR THE DETECTION OF Shigella spp. ON TOMATO SURFACES By Benjamin Ray Warren August 2003 Chair: Mickey E. Parish Cochair: Keith R. Schneider Department: Food Science and Human Nutrition Isolation of Shigella spp. from food is very difficult due to the lack of appropriate selective media and the fastidious nature of Shigellae. Nucleic acid-based detection methods such as the polymerase chain reaction (PCR) have recently been developed for the detection of Shigella spp. with greater specificity and sensitivity than conventional culture methods. In this study, artificially inoculated S. boydii UI02 or S. sonnei UI05 was recovered from tomato surfaces using a phosphate buffer rinse and vigorous shaking/hand manipulation. Detection of inocula was evaluated by enrichment protocols of the U.S. Food and Drug Administration's (1998) Bacteriological Analytical Manual (FDA BAM), the Compendium of Methods for the Microbiological Examination of Food (CMMEF), enrichment in Enterobacteriaceae Enrichment (EE) broth supplemented with 1.0 g/ml novobiocin and incubated at 42C), and FTA filtration/ nested PCR. Conventional culture enrichments were repeated using enrichments supplemented with xi

PAGE 12

50g/ml rifampicin (rif+) to exclude natural tomato microflora and rifampicin-adapted inocula. Additionally, enrichments were plated on Shigella Plating Medium (SPM), Salmonella-Shigella agar (SSA) and MacConkey agar (MAC) in order to compare isolation rates of S. boydii UI02 and S. sonnei UI05 among the three plating media. The lowest detection levels (LDLs) of enrichment procedures in the presence of natural tomato microflora were >5.3 x 10 5 CFU/tomato (all three methods) for S. boydii UI02; and 1.9 x 10 1 (FDA BAM), 1.5 x 10 3 (CMMEF), and 1.1 x 10 1 CFU/tomato (EE broth) for S. sonnei UI05. There were no significant differences ( = 0.05) between the FDA BAM and the CMMEF for the isolation of S. boydii UI02, and no significant differences ( = 0.05) among any of the enrichment methods for the isolation of S. sonnei UI05. LDLs from enrichment procedures where background microflora was excluded were 6.3 x 10 0 CFU/tomato (FDA BAM rif+ and CMMEF rif+) and >5.3 x 10 5 CFU/tomato (EE broth rif+) for S. boydii UI02; and 1.9 x 10 1 CFU/tomato (FDA BAM rif+ and CMMEF rif+), and 1.1 x 10 1 CFU/tomato (EE broth rif+) for S. sonnei UI05. The LDL of the FTA filtration/ nested PCR method was 6.2 x 10 0 CFU/tomato for S. boydii UI02 and 7.4 x 10 0 CFU/tomato for S. sonnei UI05. The FTA filtration/ nested PCR method was significantly better than enrichment protocols of the CMMEF (P = 0.010) and in EE broth (P < 0.001) for the detection of S. boydii UI02; however it was not significantly better than the FDA BAM (P = 0.177). The FTA filtration/ nested PCR method was significantly better than enrichment protocols of all three conventional culture methods (P < 0.001) for the detection of S. sonnei UI05. EE broth was found to be inhibitory to S. boydii UI02. Furthermore, there were no significant differences ( = 0.05) among SPM, SSA and MAC for the isolation of S. boydii UI02 or S. sonnei UI05. xii

PAGE 13

CHAPTER 1 INTRODUCTION Foodborne illness associated with the consumption of fresh produce has increased during recent years in the Unites States. Although this increase of illness can be partially attributed to increased consumption of fresh produce, the increased demand for minimally processed fruits and vegetables and the growth in global food trade have also contributed (Tauxe et al., 1997). Between 1987 and 1997, the total fresh produce market increased from $34.8 billion to $70.8 billion in retail and foodservice sales (Kaufman et al., 2000). During the same time period, U.S. imports of fruits and vegetables have grown from $2.0 billion in sales to $4.1 billion (Kaufman et al., 2000). Imported fruits and vegetables represent an increased potential for foodborne illness, especially if grown under poor production standards or mishandled during a long distribution cycle (Food and Drug Administration (FDA), 2001d). Coinciding with the increase of fresh produce consumption, the incidence of foodborne shigellosis has also increased. Previously thought to be primarily a water-borne pathogen, foodborne outbreaks of Shigella spp. are increasing, most recently involving fresh parsley, iceburg lettuce, and a bean salad containing parsley and cilantro. Of the pathogens under surveillance by the U.S. Centers for Disease Control and Preventions (CDC) Emerging Infections Program, Foodborne Diseases Active Surveillance Network (FoodNet), Shigella spp. accounted for 18.2% of laboratory confirmed cases during the year 2000. Shigella spp. are the leading causative agents of 1

PAGE 14

2 foodborne illness among children age 1 to 4 years at 29.1 cases per 100,000 (CDC, 2002b), with a national incidence rate of 3.8 cases per 100,000 (CDC, 2002a). Produce related illnesses caused by Shigella spp. can be reduced with the use of rapid detection methods and the proper sampling/testing of imported and domestic produce. Historically, conventional culture methods such as the U.S. FDAs (1998) Bacteriological Analytical Manual (BAM) and the Shigella culture method published in the Compendium of Methods for the Microbiological Examination of Foods have been employed to detect Shigella spp. in food products. These methods make use of selective media and biochemical tests, which require several days to complete. Furthermore, the selective media available are not able to eliminate closely related organisms such as Enterobacter spp., Klebsiella spp., and Citrobacter spp. which tend to out-compete Shigella spp. More modern microbiological methods, such as DNA hydridization and the polymerase chain reaction (PCR), have focused on the detection of DNA segments unique to Shigella spp. and enteroinvasive E. coli. These assays are usually completed in one or two working days, thereby allowing a more rapid determination of contaminated product over conventional culture methods. The early detection and identification of a contaminated product, especially during an outbreak, can significantly limit the number of illnesses. Furthermore, the elimination of background microflora is not necessary in DNA-based assays, allowing for the detection of low populations of Shigella spp. among high levels of potentially competitive background. The primary problem associated with PCR methods for the detection of Shigella spp. and other human pathogens is that no colony is isolated; therefore no further characterization can be performed. To overcome this problem, FDA is investigating a

PAGE 15

3 new method for the detection of Shigella spp. in foods that incorporates both conventional culture methods and PCR techniques (Dr. Keith Lampel, FDA, personal communication). In the proposed method, a conventional culture method involving an enrichment step followed by plating with selective media is initiated simultaneously with a nested PCR method. The nested PCR method involves two successive PCR reactions in which diluted product from the first reaction is used as template for the second reaction. Template preparation procedures for the nested PCR include a double filtration (size exclusion/FTA ) technique from which FTA filter punches are taken and used as template for the first PCR reaction. According to the proposed method, enrichment procedures to isolate Shigella spp. would only be carried forth if positive amplification were observed in the first PCR reaction. The proposed method would thereby incorporate the rapid detection capabilities of PCR with the ability to isolate a colony of conventional plating techniques. The objectives of this study were as follows: 1. To compare conventional culture methods and the FTA filtration/nested PCR method for the detection of Shigella spp. on tomato surfaces. 2. To compare enrichment in Enterobacteriaceae Enrichment broth to enrichment by the FDA BAM and CMMEF methods. 3. To compare Shigella Plating Medium to MacConkey and Salmonella-Shigella agar for the isolation of Shigella spp. on tomato surfaces.

PAGE 16

CHAPTER 2 LITERATURE REVIEW General Tomato Background Tomatoes, although commonly classified as a vegetable, are actually a fruit since they are the ripened ovary of the tomato plant. Tomatoes are grown for one of two distinct markets: the fresh market or the processing market. The fresh and processing tomato markets can be distinguished by four general characteristics. First, tomato varieties grown for the processing market tend to have higher percentages soluble solids in order to efficiently make products like tomato paste (Economic Research Service, U.S. Department of Agriculture (ERS), 2003). Second, most tomatoes grown for processing are produced under contract between the grower and processing firms (ERS, 2003). Third, fresh-market tomatoes are all hand picked while processing tomatoes are machine harvested (ERS, 2003). Finally, fresh market tomato prices are higher and more variable due to larger production costs and greater market uncertainty (ERS, 2003). Most of the fresh tomatoes produced in the United States are grown in Florida (~ 40%) and California (~ 27%) (Sargent, 1998). Fresh market tomatoes are available year round because imports from Mexico supplement winter decreases in domestic tomato production (ERS, 2003). In 1999, the Americans ate 4.8 billion pounds of fresh tomatoes, or 17.8 pounds per person (ERS, 2003). Tomatoes provide a rich source of vitamins minerals, carotenoids, and other phytochemicals (Florida Tomato Committee (FTC), 2003). One medium, fresh tomato (5.2 oz) provides 40% of the U.S. recommended daily allowance of vitamin C and 20% 4

PAGE 17

5 of the vitamin A (ERS, 2003). Lycopene, an antioxidant found almost exclusively in tomatoes, is under current investigation as a reducer of risk for several types of cancer (FTC, 2003). Additionally, current research is also investigating a link between lycopene-rich diets and reduced incidence of heart disease (FTC, 2003). Use of Tomatoes as a Model for Detection of Shigella spp. from Fruits and Vegetables When choosing the tomato as a model for recovery of Shigella spp. from fruits and vegetables, the surface characteristics must be considered. The waxy surface of tomatoes is very smooth, unlike the corky surfaces of cantaloupes or potatoes, and without the invaginations found on oranges. Such surface irregularities make the recovery of inouclum difficult and more inconsistent. Furthermore, tomatoes do not contain epidermal or peridermal pores, such as stoma or lenticels, respectively, which allow gas exchange and also may allow internalization of bacterial pathogens. For these reasons, bacteria used to artificially inoculate tomatoes may be recovered with greater efficiency than from other fruits and vegetables. For purposes of this study, the inoculation of waxy surfaces of tomatoes was performed to reduce the number of variables associated with other surfaces, such as bacterial internalization and inefficient surface removal, thus yielding a better comparison of methods. General Characteristics of Shigella Shigella, the causative agent of shigellosis or bacillary dysentery, was first discovered over 100 years ago by a Japanese scientist Kiyoshi Shiga (Anonymous, 2002). Shigellae are members of the family Enterobacteriaceae, and are nearly genetically identical to Escherichia coli (E. coli) and are closely related to Salmonella and Citrobacter spp. (American Public Health Association (APHA), 2001). Shigellae are characterized as Gram-negative, facultatively anaerobic, non-sporulating, non-motile

PAGE 18

6 rods. Typically, species of Shigella do not ferment lactose, are lysine-decarboxylase negative, are acetate and mucate negative, and do not produce gas from glucose, although several exceptions exist (Echeverria et al., 1991). There are four serogroups of Shigella: S. dysenteriae (serogroup A) serotype 1 15, S. flexneri (serogroup B) serotype 1 8 (9 subtypes), S. boydii (serogroup C) serotype 1 19, and S. sonnei (serogroup D) serotype 1. Serogroups of Shigella can be differentiated by their biochemical traits and antigenic properties (CDC, 2003); however they can also be differentiated by their epidemiology (Ingersoll et al., 2002). S. dysenteriae is the serogroup primarily associated with epidemics (Ingersoll et al., 2002); S. dysenteriae type 1 is associated with the highest case fatality rate of all Shigella serogroups at 5-15% (CDC, 2003). S. flexneri is the predominate group found in areas of endemic infection, while S. sonnei is the group implicated in source outbreaks in developed countries (Hale, 1991). S. boydii has been associated with food imported from Central and South America and is rarely isolated in North America. Shigella, although classically thought of as a waterborne pathogen, has been involved in an increasing number of food-borne outbreaks (Smith, 1987). Food products associated with Shigella outbreaks are most commonly subjected to hand processing or preparation, limited heat treatment, or served/delivered raw to the consumer (Wu et al., 2000). Examples of food products from which Shigella spp. have been isolated include potato salad, ground beef, bean dip, raw oysters, fish, and raw vegetables. The infective dose for Shigella spp. is reported to be very low; ingestion of 1.0 x 10 1 cells of S. dysenteriae is sufficient for infection, while the other serogroups require ingestion of 1.0 x 10 2 to 1.0 x 10 4 cells (cited in Muriana, 2002). The low infective dose

PAGE 19

7 associated with Shigella spp. results in common spread of the disease through person to person contact. Typical symptoms of infection include bloody diarrhea, abdominal pain, fever, and malaise. Seizures in children with shigellosis have been reported in 5.4% of cases (Galanakis, 2002). Late complications of S. dysenteriae serotype 1 infections can include hemolytic uremic syndrome (HUS), while S. flexneri infections can result in development of Reiters syndrome, especially in persons with the genetic marker HLA-B27 (CDC, 2003). Reiters syndrome is characterized by joint pain, eye irritation, and painful urination (CDC, 2003). Recent Outbreaks Involving Shigella spp. In recent years, there has been an increasing number of Shigella outbreaks involving produce and prepared foods. In 1989, German potato salad was implicated in an outbreak of a multi-antibiotic resistant strain of S. flexneri aboard a cruise ship (Lew et al., 1991). In 1994, iceburg lettuce was implicated in an outbreak of S. sonnei which affected people from six Northern European countries (Long et al., 2002). In 1998, uncooked, chopped, curly parsley was implicated in a multi-state outbreak of S. sonnei (Morbidity and Mortality Weekly Report (MMWR), 1999). The source of this outbreak was traced back to a Mexican farm. In 1999, bean salad which contained parsley and cilantro was implicated in a Chicago area foodborne outbreak of S. boydii serotype 18. In 2000, a nationally distributed five layer bean dip was implicated in another multi-state outbreak of S. sonnei (MMWR, 2000). Prevalence of Shigella spp. on Produce In response to President Clintons National Food Safety Initiative (January 1997) and Produce & Imported Foods Safety Initiative (October 1997), the FDA has begun investigating the presence of human pathogens on produce. In March 1999, FDA initiated

PAGE 20

8 a field assignment entitled FDA Survey of Imported Fresh Produce to collect data on the incidence and extent of pathogen contamination on selected imported produce (FDA, 2001b). The survey analyzed broccoli, cantaloupe, celery, cilantro, culantro, loose-leaf lettuce, parsley, scallions (green onions), strawberries, and tomatoes for E. coli O157:H7 and Salmonella. All commodities except cilantro, culantro, loose-leaf lettuce, and strawberries were analyzed for Shigella. For those commodities tested, contamination with Shigella spp. was observed at the following rates: 0.9% (9/1003) of all commodities tested, 2.0% (3/151) cantaloupe samples, 2.4% (2/84) celery samples, 0.9% (1/116) lettuce samples, 1.2% (1/84) parsley samples, and 1.1% (2/180) scallion samples. In May 2000, FDA initiated its Survey of Domestic Fresh Produce to focus on high-volume domestic produce that is generally consumed raw (FDA, 2001a). This survey included the following commodities: cantaloupe, celery, cilantro, green onions, loose-leaf lettuce, parsley, strawberries, and tomatoes. All commodities were to be analyzed for the presence of E. coli O157:H7 and Salmonella, while all except strawberries were to be tested for Shigella. At time of publication, only interim results from analysis of 767 of the required 1000 samples had been released. As with the 1999 Imported Fresh Produce Survey, 0.9% (6/646) of the commodities tested were contaminated with Shigella, 0.9% (1/115) cantaloupe samples, 1.6% (1/62) cilantro samples, 4.1% (3/73) green onion samples, and 1.6% (1/64) parsley samples. In January 2001, FDA announced another survey entitled FDA Survey of Imported Fresh Produce: Imported Produce Assignment FY 2001. The focus of the study was to examine further the presence of E. coli O157:H7, Salmonella, and Shigella, on cilantro, culantro, cantaloupe, and tomatoes based on high rates of pathogens from

PAGE 21

9 previous surveys (FDA, 2001c). At time of publication, results from this survey were not available. Survival Characteristics of Shigella The ability of Shigella to survive is dependent, in part, upon pH, temperature, and salt concentration of its environment. In a study using S. flexneri, Zaika (2001) demonstrated the effects of temperature and pH on survival. A strain of S. flexneri was cultured in brain heart infusion broth (BHI) and subjected to various incubation temperatures (4, 12, 19, 28, and 37C) and pH conditions (pH 2, 3, 4, 5). In general, survival was enhanced by lower temperatures and increased pH for all experiments. Results of this study indicated that S. flexneri has acid resistance and suggest that foods of pH 5 or lower stored at or below room temperature may permit survival of the organism over long periods of time in sufficient numbers to cause illness (Zaika, 2001). Zaika (2002a) also investigated survival characteristics of S. flexneri as affected by NaCl. S. flexneri was able to tolerate and survive in levels of NaCl (1 6%) commonly found in food items such as pickled vegetables, caviar, pickled herring, dry cured ham, and certain cheeses for two weeks to two months (Zaika, 2002a). Survival of S. flexneri in the presence of organic acids (citric, malic, and tartaric acid), commonly found in fruits and vegetables, and fermentation acids (acetic and lactic acid), commonly used as preservatives, was studied (Zaika, 2002b). S. flexneri was cultured with each acid (plus an HCl control) at 0.04 M in BHI adjusted to pH 4, and incubated at various temperatures (4, 19, 28, and 37C). As seen in other experiments, survival increased as temperature decreased (Zaika, 2002b). At 4C, S. flexneri survived in the presence of all the acids tested for > 55 days (Zaika, 2002b).

PAGE 22

10 In water alone, Shigella spp. can survive with little decline in population levels. Rafii and Lunsford (1997) inoculated S. flexneri into distilled water. The initial count of 2.8 x 10 8 CFU/ml was only decreased to 9.2 x 10 7 CFU/ml after storage at 4C for 26 days. The high survival rate of S. flexneri in water supports the historical association of shigellosis outbreaks with water sources. Shigella spp. can survive for extended periods of time on raw vegetable surfaces. Wu et al. (2000) studied survival of S. sonnei on whole and chopped parsley leaves. When held at 21C, S. sonnei was able to grow on chopped parsley at a rate similar to that which occurs in nutritious liquid medium (Wu et al., 2000). At 4C, populations declined on both chopped and whole parsley throughout the 14 day storage period, however the pathogen survived regardless of initial population (Wu et al., 2000). Rafii and Lunsford (1997) studied the survival of S. flexneri on raw cabbage, onion, and green pepper held at 4C. Although the population decreased, S. flexneri survived storage at 4C for 12 days (at which time sampling was terminated due to spoilage) on the onion and green pepper at levels of 2.10 x 10 5 and 2.2 x 10 4 CFU/g, respectively (Rafii and Lunsford, 1997). S. flexneri continued to survive on the cabbage after 26 days at 1.13 x 10 3 CFU/g. These studies demonstrate how Shigella spp. can survive on refrigerated raw vegetables for periods of time that exceed the expected shelf life (Wu et al., 2000). Several studies have demonstrated the ability of Shigella spp. to survive in low pH foods at low temperature storage. Bagamboula et al. (2002) demonstrated the ability of S. sonnei and S. flexneri to survive in apple juice (pH 3.3-3.4) and tomato juice (pH 3.9-4.1) held at 7C for 14 days. No reduction was noted in the tomato juice, while only 1.2 to 3.1 log 10 reduction was observed in the apple juice during the 14 day study. Rafii and

PAGE 23

11 Lunsford (1997) observed the ability of S. flexneri to survive in carrot salad (pH 2.7 2.9), potato salad (pH 3.3 4.4), coleslaw (pH 4.1 4.2), and crab salad (pH 4.4 4.5) held at 4C. Sampling was terminated at day 11 for the carrot and the potato salad, at which time S. flexneri counts decreased from an initial 4.3 x 10 6 to 4.2 x 10 2 CFU/g and from 1.32 x 10 6 to 8.5 x 10 2 CFU/g, respectively. Sampling of the coleslaw and the crab salads ceased due to product spoilage on days 13 and 20, respectively, while S. flexneri counts were 2.16 x 10 4 and 2.4 x 10 5 respectively. These studies indicate that inoculated Shigellae were not rapidly killed by normal microflora, low pH (Rafii and Lunsford, 1997), or low temperature. Intracellular Activity of Shigella spp. Invasion of Epithelial Cells Invasion of epithelial cells by Shigellae involves four steps: entry into epithelial cells, intracellular multiplication, intraand intercellular spreading, and killing of the host cell (Sansonetti, 1991). The invasion process is controlled by a 220-kDa plasmid. The plasmid contains invasion plasmid antigen (ipa) genes which encode four highly immunogenic polypeptides; IpaA, IpaB, IpaC, and IpaD. Studies in which Tn5 insertions affecting the expression of the ipa genes reveal that expression of ipaB, ipaC, and ipaD is strongly associated with entry, while ipaA is not (Sansonetti, 1991). Invasion is also mediated by the virF gene, which is located on the virulence plasmid, and the virR gene, which is located on the chromosome. The virF gene encodes a 30-kDa protein that positively regulates the expression of the ipa genes and a plasmid gene icsA (also known as virG), which encodes intraand intercellular spread. Environmental factors which affect the expression of virF are not known. The virR gene is a repressor of the plasmid invasion genes in a temperature-dependant manner

PAGE 24

12 (Sansonetti, 1991). When Shigellae are grown at 30C they do not express any of the Ipa polypeptides and are therefore non-invasive, however, Shigellae grown at 37C are fully invasive and all plasmid polypeptides are encoded (Sansonetti, 1991). Gaining entry into epithelial cells Once ingested, Shigellae move through the gastrointestinal tract to the colon, where they translocate the epithelial barrier via M cells that overlay the solitary lymphoid nodules (Suzuki and Sasakawa, 2001). Upon reaching the underlying M cells, Shigella infects the macrophages and induces cell death (Suzuki and Sasakawa, 2001). Infected macrophages release interleukin-1, which elicits a strong inflammatory response (Zychlinsky et al., 1994). Once released from the macrophage, Shigella will enter the epithelial cells, also called enterocytes, which predominately line the colon via membrane ruffling and macropinocytosis. Epithelial cells produce inflammatory cytokines in response to bacterial invasion, therefore increasing inflammation of the colon (Suzuki and Sasakawa, 2001). Macropinocytosis involves the cell extending its membrane as formations known as pseudopodia, which will engulf large volumes and close around them forming a vacuole. In order for macropinocytosis to occur, actin must be polymerized and myosin, an actin binding protein, must be present (Stendahl et al., 1980). Common stimuli that induce macropinocytosis include cytokines and bacterial antigens. Intracellular multiplication Shigellae immediately disrupt phagocytic vacuoles allowing entry into the host cell cytoplasm. Once in the cytoplasm, Shigellae multiply rapidly. Sansonetti et al. (1986)

PAGE 25

13 observed the generation time for S. flexneri in HeLa cells to be approximately 40 minutes. Sustaining efficient intracellular growth requires the acquisition of host cell nutrients. Although production of Shiga toxin facilitates the availability of host cell nutrients, no relation between its production and intracellular growth rates can be observed (Fontaine et al., 1988). Likewise, no relation between Shiga-like toxin production and intracellular growth rate can be observed (Sansonetti et al., 1986; Clerc et al., 1987). Since little free iron exists within mammalian host cells, Shigella spp. must also express high-affinity iron acquisition systems. In order to obtain iron, Shigella spp. synthesize and transport the siderophores aerobactin and enterobactin (Vokes et al., 1999) and utilize a receptor/transport system in which iron is obtained from heme. Siderophores (aerobactin and enterobactin) are low molecular weight iron binding compounds that remove iron from host proteins. Enterobactin is produced by some but not all Shigella spp. (Perry and San Clemente, 1979; Payne, 1984) while aerobactin is synthesized by S. flexneri and S. boydii (Lawlor and Payne, 1984) and some S. sonnei (Payne, 1988). Headley et al. (1997) demonstrated that aerobactin systems, although active in extracellular environments, are not expressed intracellularly. This suggests that siderophore-independent iron acquisition systems can provide essential iron during intracellular multiplication (Headley et al., 1997). Intraand intercellular spreading (actin-based motility) The capacity for Shigella to spread intracellularly and infect adjacent cells is critical in the infection process (Sansonetti, 1991). Intraand intercellular spreading is controlled by the icsA (virG) gene located on the virulence plasmid. The icsA gene

PAGE 26

14 encodes the protein IcsA, which enables actin-based motility (Bernardini et al., 1989) and intercellular spread (Makino et al., 1986). IcsA is a surface-exposed outer membrane protein consisting of three distinctive domains: a 52 amino acid N-terminal signal sequence, a 706 amino acid -domain, and a 344 amino acid C-terminal -core (Goldberg et al., 1993; Lett et al., 1989; Suzuki et al., 1995). The -domain is the exposed portion and the -core is embedded in the outer membrane (Suzuki et al., 1995). IcsA is distributed at one pole of the outer membrane surface. This asymmetrical distribution allows the polar formation of actin tails, and thus polar movement of Shigella within host cell cytoplasm. The polar localization of IcsA is primarily affected by two events: (i) the rate of diffusion of outer membrane IcsA (Sandlin et al., 1995; 1996; Sandlin and Maurelli, 1999; Robbins et al., 2001) and (ii) the specific cleavage of IcsA by the protease IcsP (SopA) (dHauteville et al., 1996; Egile et al., 1997; Steinhauer et al., 1999). Rate of diffusion of outer membrane IcsA is directly affected by the O side chains of the membrane lipopolysaccharide (LPS). Sandlin et al. (1996) demonstrated this relationship with a S. flexneri LPS mutant, BS520 which does not make any O-antigen. As compared with a wild-type strain of S. flexneri, which polymerized actin at one pole, the LPS mutant strain polymerized actin in a non-polar fashion. Expression of LPS that does not have any O side chains causes an even distribution of IcsA over the entire outer membrane (Sandlin et al., 1995; 1996; Monack and Theriot, 2001). Composition of the C-terminal one-third of the IcsA domain is also required for polar localization as well as polar movement of S. flexneri (Suzuki et al., 1996). A S. flexneri mutant, in which a segment of this section was deleted, was unable to polymerize actin in a polar fashion or move unidirectionally (Suzuki et al., 1996).

PAGE 27

15 The icsP gene encodes the outer membrane protease IcsP (also called SopA) which cleaves laterally diffused IcsA, thus promoting polar localization (Egile et al., 1997). An E. coli K-12 strain, engineered to express the icsA gene, was shown to diffuse IcsA along its outer membrane (Monack and Theriot, 2001). When the same E. coli K-12 strain was engineered to express the icsP gene with the icsA gene, the number of bacteria which polymerized actin at one pole increased (Monack and Theriot, 2001). The N-terminal two-thirds of the IcsA domain is essential for mediating actin assembly in Shigella host cells (Suzuki and Sasakawa, 2001). This portion of IcsA contains six glycine rich repeats which interact with the Wiskott-Aldrich syndrome protein (N-WASP). N-WASP is composed of distinct domains: PH, a pleckstrin homology domain; IQ, a calmodulin binding domain; GBD, a GTPase binding domain that binds Cdc42; PRR, a proline-rich region; V, a G-actin-binding veroprolin homology domain; C, cofilin homology domain; A, a C-terminal acidic amino acid segment (cited in Suzuki and Sasakawa, 2001; Miki et al., 1996). The VCA domain of N-WASP activates and interacts with the Arp2/3 complex. The IcsA--N-WASP--Arp2/3 complex mediates rapid actin filament growth at the barbed end, including cross-linking between the elongated actin filaments (Suzuki and Sasakawa, 2001). The resulting network of actin filaments allows Shigella to gain a propulsive force with which to move in the cytoplasm of host cells (Suzuki and Sasakawa, 2001). Cdc42 (an N-WASP activator), profilin (an actin monomer-binding protein), and cofilin (which depolymerizes actin) also have roles in efficient actin assembly. Cdc42 binds to the GBD domain of N-WASP preventing the intramolecular interaction between the C-terminal acidic amino acids and the basic amino acids of the GBD, thereby forcing

PAGE 28

16 the N-WASP complex to unfold to its activated form (Suzuki and Sasakawa, 2001). Cdc42 independent activation of the N-WASP complex by IcsA has been reported for actin-based motility in Shigella, however efficient entry into cells was reported as Cdc42 dependent (Shibata et al., 2002). The role of Cdc42 in invasion and motility is still somewhat controversial. Profilin delivers monomeric actin to sites of actin assembly (Goldberg, 2001). Although profilin has been shown not to be absolutely essential for actin based motility, it is required for maximum rates of movement (Loisel et al., 1999). Cofilin generates actin monomers from the filamentous actin. By disassembling unneeded actin filaments within the tail, cofilin might work to free up actin for incorporation into newly generated filaments (Goldberg, 2001). Intercellular spreading is dependent upon an actin-based motility mechanism (Fig. 2-2) (Monack and Theriot, 2001). Shigella cells first form a membrane bound protrusion into an adjacent cell. This protrusion must distend two membranes: one from the donor cell, and another from the recipient cell (Parsot and Sansonetti, 1996). As the protrusion pushes further into the recipient cell, it is taken up by the recipient cell resulting in the bacteria enclosed in a double-membrane vacuole (Monack and Theriot, 2001). Intercellular spread is completed when Shigella rapidly escapes from the double-membrane vacuole, releasing it into the cytosol of the secondary cell. Monack and Theriot (2001) observed intercellular spread of an E. coli K-12 strain expressing the icsA and icsP genes in HeLa cells. As expected, they found the E. coli K-12 strain spread to adjacent cells and enclosed in a double-membrane vacuole as well as free in the cytosol of the adjacent cells.

PAGE 29

17 Early host cell apoptosis The early killing of host cells by Shigella is mediated by the virulence plasmid. Sansonetti (1991) demonstrated the inability of non-invasive S. flexneri to kill host cells whereas the invasive species killed efficiently and rapidly. Non-invasive strains can not kill host cells, since this requires that the Shigella be intracellular. Early killing of host cells involves metabolic events which rapidly drop the intracellular concentration of ATP, increase pyruvate, and arrest lactate production (Sansonetti and Mounier, 1987). Interestingly, Shiga toxin, a potent cytotoxin produced by S. dysenteriae serotype 1, does not play a role in the early killing of host cells. Fontaine et al. (1988) constructed a Tox mutant strain of S. dysenteriae serotype 1 and found that the mutant killed as efficiently as the wild-type strain. Toxins Produced by Shigella spp. Shiga toxin S. dysenteriae type 1 strains produce a potent toxin known as Shiga toxin (STX). Although the toxin is not necessary to sustain an infection, its expression increases the severity of disease. Three biologic activities associated with STX are cytotoxicity, enterotoxicity, and neurotoxicity, while the one known biochemical effect is the inhibition of protein synthesis (Donohue-Rolfe et al, 1991). STX is considered the prototype to a family of toxins known as Shiga-like toxins (SLT), which are similar in structure and function, and share the same receptor sites. Perhaps the most widely known human pathogen that produces SLTs is E. coli O157:H7, which has two toxin variants (SLT I and SLT II). STX is composed of two polypeptides: an A subunit (32,225 MW) and five B subunits (7,691 MW, each) (Donohue-Rolfe et al., 1991). The B subunits mediate

PAGE 30

18 binding to cell surface receptors, which have been identified as glycolipids containing terminal galactose-(1-4)galactose disaccharides such as galabiosylceramide and globotriasylceramide (Gb3) (Brown et al., 1991; Keusch et al., 1991). The A subunit, once inside the host cell cytoplasm, acts enzymatically to cleave the N-glycosidic bond of adenine at nucleotide position 4324 in the 28S rRNA of the 60S ribosomal unit (Donohue-Rolfe et al., 1991). Shigella enterotoxins 1 and 2 Recently, two enterotoxins, shigella enterotoxin 1 (SHET 1) and shigella enterotoxin 2 (SHET 2), have been characterized and are believed to play a role in the clinical manifestation of shigellosis (Yavzori et al., 2002). SHET 1, which is chromosomally encoded, was only prevalent in isolates of S. flexneri 2a. SHET 2, however, is encoded on the large virulence plasmid, and was detected in all Shigella isolates tested except several isolates which lost their plasmid. Traditional Microbiological Media for the Isolation/Detection of Shigella Traditional microbiological techniques make use of selective media for the enrichment/isolation of Shigella spp. Many variants of enrichment and plating media have been investigated for optimal recovery, often with conflicting results between laboratories and sample types. Due to the lack of appropriate selective media and the presence of Shigella spp. in relatively low population, background microflora tends to out-compete Shigella spp. when isolation is attempted from a food product. Unless media can be developed that are both specific and sensitive for Shigella spp. regardless of potential background microflora, isolation by traditional microbiological methods will always be suspect.

PAGE 31

19 Early enrichment of Shigella spp. was attempted using Selenite-F (SF) or Tetrathionate (TT) broth. These broths were originally designed for the isolation of salmonellae, but due to the lack of specific enrichment media for Shigellae they were used as all-purpose enteric enrichment broths (Taylor and Schelhart, 1969). Sodium selenite, although selective for salmonellae, is toxic to Shigella spp. (and most enterics), therefore its use in enrichment procedures for Shigella spp. has been terminated. TT is a peptone base broth with bile salts and sodium thiosulfate, which inhibit most Gram-positives and Enterobacteriaceae. Gram-negative (GN) broth is a peptone-based broth with glucose and mannitol. The concentration of mannitol in GN broth is higher than glucose to promote mannitol fermentors, like Shigella spp. Both TT broth and GN broth contain bile salts, which can be inhibitory to stressed cultures. Furthermore, GN broth contains sodium deoxycholate, which has been shown to inhibit heat-stressed Shigellae (Uyttendaele et al, 2001). Currently, enrichment procedures use a low carbohydrate medium, Shigella broth (SB) with addition of novobiocin, for the detection/isolation of Shigella spp. (APHA, 2001; FDA BAM). Acids produced by other Enterobacteriaceae during the fermentation of carbohydrates have been reported to be toxic to Shigellae (Mehlman et al., 1985); however, other studies have shown the acid tolerance of Shigella spp. to grow at a pH of 4.5 to 4.75 (Bagamboula et al., 2002) and to survive at a pH of 4.0 (Zaika, 2002). Since SB contains very little carbohydrate, the effect of a low pH environment on the enrichment of Shigella spp is limited when SB is used. SB is also less stringent than TT broth and GN broth since it contains neither bile salts nor sodium deoxycholate. In a recent study investigating enrichment media for detection of Shigellae, SB, GN broth, tryptic soy broth, and Enterobacteriaceae Enrichment (EE) broth with the

PAGE 32

20 addition of novobiocin were compared (Uyttendaele et al., 2001). When incubated in GN broth, Shigella spp. were unable grow to comparable levels as observed in SB and EE broths, even though EE broth contains bile salts. In order to increase the specificity of enrichment media, elevated incubation temperatures and anaerobic atmospheric conditions are recommended by the FDA BAM. In a competitive inoculation study, cultures incubated in SB and EE broth at 42C eliminated competitors such as Aeromonas and Erwinia while incubation at 37C did not (Uyttendaele et al., 2001). Due to the fastidious nature of Shigellae, multiple plating media of different selectivity should be used to increase the chances of isolation. The most common low selectivity media used for plating Shigella spp. is MacConkey Agar (MAC); while Eosin methylene blue (EMB) or Tergitol-7 (T7) agars could also be used. Since differentiation is based on lactose fermentation, non-lactose competitors make detection of Shigella spp. on MAC very difficult (Uyttendaele et al., 2001). On MAC, Shigella spp. are translucent and slightly pink, with and without rough edges. Shigella spp. produce colorless colonies on EMB and bluish colonies on the yellowish-green T7 agar (APHA, 2001). A more recently developed low selectivity, differential medium is Shigella Plating Medium (SPM) (RF Laboratories, West Chicago, IL). Shigella spp. are white to clear on SPM. Intermediate selectivity media useful in isolating Shigella spp. are desoxycholate citrate agar (DCA) and xylose lysine desoxycholate agar (XLD). Shigella spp. produce colorless colonies on both DCA and XLD. Bhat and Rajan (1975) reported XLD superior to DCA for the isolation of Shigella spp. since DCA required a 48 hour incubation to show clear colony morphology as opposed to overnight incubation for XLD. A problem with XLD is

PAGE 33

21 that D-xylose, which serves as a differentiating agent, is fermented by some strains of S. boydii while most Shigella spp. do not ferment xylose (APHA, 2001). This can cause some strains of Shigella to be missed if only XLD is used as a plating media. Highly selective media for Shigella spp. include Salmonella-Shigella agar (SSA) and Hektoen Enteric agar (HEA). A problem associated with SSA and HEA is that they may be too stringent for some strains of Shigella spp., especially if the culture is stressed (APHA, 2001; Uyttendaele et al., 2001). Shigella spp. produce colorless, translucent colonies on SSA and green colonies on HEA. U.S. Food and Drug Administration (1998) Bacteriological Analytical Manual The FDA BAM outlines a conventional culture method for the isolation and detection of Shigella spp. from food. A 25 g sample is transferred to 225 ml of Shigella broth (SB) to which novobiocin (0.5 g/ml for S. sonnei; 3.0 g/ml for other Shigella spp.) has been added. Samples are held at room temperature for 10 minutes and periodically shaken. Sample supernatants are transferred to an Erlenmeyer flask and the pH adjusted to 7.0 0.2 with sterile 1 N NaOH or 1 N HCl. Flasks are incubated anaerobically for 20 hours (44C for S. sonnei; 42C for all other Shigella spp.). After incubation, enrichment culture is used to streak a MAC plate. Confirmation of suspicious colonies involves tests for motility, H 2 S, gas formation, lysine decarboxylase, and fermentation of sucrose or lactose. All isolates showing any of these characteristics are discarded. Isolates negative for all confirmatory tests are tested for further biochemical reactions including adonitol, inositol, lactose, potassium cyanide, malonate, citrate, salicin, and methyl red. Shigellae are negative for all except methyl red. Antisera agglutination is then used to identify any culture displaying typical Shigella characteristics.

PAGE 34

22 The effectiveness of the FDA (1992) BAM method for Shigella spp. was evaluated by June et al. (1993). The 1992 FDA BAM procedures were the same as the 1998 FDA BAM procedures with respect to the isolation and detection of Shigella spp. Two strains of S. sonnei, strains 9290 and 25931, were inoculated on potato salad, chicken salad, cooked shrimp salad, lettuce, raw ground beef, and raw oysters. The lowest number of unstressed cells of strain 9290 recovered from a 25 g food sample were: 1.0 x 10 0 for potato salad, 8.5 x 10 -1 for chicken salad, 8.8 x 10 -1 for cooked shrimp salad, 7.6 x 10 -1 for lettuce, 8.4 x 10 1 for raw ground beef, and 8.5 x 10 2 for raw oysters. The lowest number of unstressed cells of strain 25931 recovered from a 25 g food sample were: 5.3 x 10 -1 for potato salad, 6.6 x 10 -1 for chicken salad, 1.6 x 10 2 for cooked shrimp salad, 4.2 x 10 -1 for lettuce, 6.9 x 10 4 for raw ground beef, and 5.4 x 10 2 for raw oysters. The recovery of 8.4 x 10 0 CFU/25 g for strain 9290 versus 6.9 x 10 4 CFU/25 g for strain 25931 from raw ground beef samples suggests high strain variability, which can complicate recovery methods. Chilled stressed cells for both strains were recovered with similar results for all 6 foods. Since the infective dose of Shigella spp. is has low as 10 cells per person, the 1992 FDA BAM was considered ineffective for the evaluation of raw ground beef and raw oysters. Compendium of Methods for the Microbiological Examination of Foods Another method for the isolation/detection of Shigella spp. from foods is described in the Compendium of Methods for the Microbiological Examination of Foods (CMMEF) (APHA, 4 th Edition, Chapter 38). This method calls for the enrichment of 25 g sample in either 225 ml of SB or GN broth, both with the optional addition of novobiocin (0.3 g/ml for S. sonnei; 3.0 g/ml for all other Shigella spp.) After holding at room temperature for 10 minutes, enrichment samples are incubated for 16 to 20 hours

PAGE 35

23 at 37C. The CMMEF suggests that 2 to 3 plates of various selective media be used to streak the enriched cultures; MAC for low selectivity, XLD for intermediate selectivity, and HEA for high selectivity. Suspicious colonies are tested for motility, with all non-motile isolates identified by biochemical and serological tests. Nucleic Acid-Based Detection of Shigella spp. in Foods The essential principle of nucleic acid based detection methods is the specific formation of double stranded nucleic acid molecules from two complementary, single stranded molecules under defined physical and chemical conditions (Olsen et al, 1995). There are basically two types of nucleic acid assays: hybridization assays and the polymerase chain reaction (PCR). Colony Hybridization Assays In colony hybridization assays, a culture sample is spread-plated on appropriate media. Following incubation, the colonial pattern is transferred to a solid support (usually a membrane or paper filter) by pressing the support onto the agar surface (FDA, 1998). Cells are lysed in situ by a combination of high pH and temperature (0.5 M NaOH and/or steam or microwave irradiation), which also denatures and affixes the DNA to the support (FDA, 1998). Solid supports are incubated in solution containing labeled probes ( 32 Por enzyme-label) to allow the probes to attach to their target DNA. Unbound probe is removed by washing the probe-target complexes on the support at an appropriate temperature and salt concentration (FDA, 1998). A signal is then generated using the label attached to the probe to identify positive colonies. Colony hybridization assays are useful when further characterization of positive colonies is required (Olsen et al., 1995).

PAGE 36

24 PCR: The Basics PCR amplifies regions of DNA by annealing specific primers to single stranded DNA (ssDNA) and rebuilding the double stranded molecule using a polymerase enzyme. Typical reactions are performed in a mixture of water, dNTPs, PCR buffer, primers, taq polymerase, and a DNA template. After initial denaturation, PCR subjects reaction mixtures to approximately 30 cycles of denaturation, annealing, and extension. Denaturation occurs at 94C and involves the unwinding of double stranded DNA (dsDNA) to ssDNA. Annealing, typically at 50-70C, is where the single stranded primers attach to the ssDNA at their specific sites. Extension occurs at 72C which is the optimal temperature for most taq polymerases. During extension, the polymerase enzyme interacts with the primer/ssDNA complex and rebuilds the dsDNA molecule using the dNTPs. As the cycling continues, the number of dsDNA copies doubles with each cycle. After the final cycle, most PCR reactions include a final extension step which allows the completion of any incomplete reactions. In theory, PCR can amplify a single copy of target DNA to over a million copies in a 30 cycle reaction. Bacterial DNA Template Preparation for PCR The failure or success of PCR greatly depends on the effectiveness of the DNA extraction method to provide adequate amounts of purified DNA. Early PCR used crude extracts of DNA obtained by boiling a culture/sample, centrifuging the cell material, and using the supernatant as DNA template. DNA templates prepared as crude extracts are often contaminated with high amounts of protein and contain very low concentrations of target DNA. The most common way to purify and concentrate DNA samples is to perform a phenol, phenol-chloroform, or guanidine isothionate extraction (purification) followed by

PAGE 37

25 ethanol precipitation (concentration). Other methods for DNA purification/concentration include the use of anion exchange resin (DNA affinity) columns or various filtration techniques. DNA affinity columns require elution by high salt and large fragments of DNA >20,000 bp can stick to the column (Millipore, 2003). Several filtration techniques are available for isolating DNA including size exclusion, glass fiber filters, and FTA filter paper. FTA filtration involves a chemically treated filter which can trap bacterial cells, lyse the cellular membrane, and bind bacterial DNA. Several PCR methods have been developed which can amplify DNA directly off the filter paper (Lampel et al., 2000; Orlandi and Lampel, 2000). Nested PCR Nested PCR refers to a two-step PCR technique in which the PCR product from the first reaction is diluted and used as template in the second PCR reaction. In true nested PCR, the first reaction uses an external set of primers (P1 and P2) to amplify a target region in a gene of interest. The second PCR reaction uses an internal primer set, (P3 and P4) to amplify a region from within the product of the first reaction. Semi-nested PCR is the same as nested PCR with the exception that instead of using two internal primers in the second PCR reaction, one internal (P3) and one external primer (P1 or P2) are used. The advantage of nested or semi-nested PCR is that greater specificity and sensitivity can be achieved. Protocols for the analysis of Shigella spp. in development at FDA utilize a nested PCR (Dr. Keith Lampel, personal communication). In the protocol, positive amplification in the first reaction serves as presumptive detection while positive amplification in the second reaction serves as confirmation of Shigella spp.

PAGE 38

26 PCR for the Detection of Shigella sp. PCR assays for Shigella spp. have targeted the invasion associated locus (ial) (Islam and Lindberg, 1992; Lindqvist, 1999), the virA gene (Vantarakis et al., 2000; Villalobo and Torres, 1998) or the ipaH gene (Sethabutr et al., 1993; Sethabutr et al., 2000). All three of these targets detect all four serogroups of Shigella and enteroinvasive E. coli. The ial and virA gene are located on the virulence plasmid, while the ipaH is encoded multiple times on the plasmid and on the chromosome (Jin et al., 2002). Since detection of the ipaH gene is possible in the event of losing the plasmid, it is a very attractive target for PCR assays. Although few studies have used PCR methods in the examination of food for Shigella spp., they have demonstrated higher sensitivity than conventional culture methods. Vantarakis et al. (2000) developed a multiplex PCR method to detect both Salmonella spp. and Shigella spp. in mussels. Artificially inoculated S. typhimurium and S. dysenteriae were recovered by homogenizing mussel meat with peptone water. DNA from an aliquot of the homogenate was purified using a guanidine isothionate method and concentrated via ethanol precipitation. Amplification of Shigella spp. and Salmonella spp. DNA targeted the virA and invA genes, respectively. The multiplex PCR was able to detect S. dystenteriae at 1.0 x 10 3 CFU/ml homogenate with no pre-enrichment, and 1.0 x 10 1 -1.0 x 10 2 CFU/ml homogenate following 22 hour incubation in buffered peptone water. Villalobo and Torres (1998) investigated PCR for the detection of Shigella spp. in mayonnaise. S. dysenteriae type 1 DNA was isolated from artificially contaminated mayonnaise samples by homogenizing in buffered peptone water, lysing cells with detergent, extracting with phenol-chloroform, and precipitating with ethanol.

PAGE 39

27 Amplification targeted the virA gene and was multiplexed with a region of the 16srDNA. This detection method was able to detect S. dysenteriae type 1 at 1.0 x 10 2 -1.0 x 10 3 CFU/ml homogenate. In a study by Lindqvist (1999), a nested PCR method was compared to a conventional culture method (NMKL no. 151 1995) for detection of Shigella spp. Recovery of DNA from spiked lettuce, shrimp, milk, and blue cheese samples was accomplished by homogenizing with physiological saline, buoyant density centrifugation (to separate components based on density), and boiling at 96-98C for 10 minutes. Amplification targeted internal regions of the ial. Single PCR, using the external primer sets only, was only able to detect S. flexneri in aqueous solution at 0.5-1.0 x 10 5 CFU/ml, however the nested PCR was able to detect 1.0 x 10 3 CFU/ml. The nested PCR assay in combination with buoyant density centrifugation was able to detect S. flexneri inoculated onto all four foods at 1.0 x 10 1 CFU/g (Lindqvist, 1999). Theron et al. (2001) investigated a semi-nested PCR method for the detection of Shigella spp. in spiked environmental water samples. S. flexneri was inoculated into sterile and non-sterile environmental water samples. Dilutions of the water samples were made and bacterial cells from each dilution were harvested by centrifugation and resuspended in GN broth. After incubation at 37C for 6 hours, bacterial cells were washed twice in distilled water and lysed by heating at 100C for 10 minutes. Lysate supernatant was used as DNA template for semi-nested PCR. Amplification targeted the ipaH gene. The detection limits of the various environmental water samples were 2 x 10 3 cfu/ml for well water, 1.4 x 10 1 CFU/ml for lake water, 5.8 x 10 2 CFU/ml for river water, 6.1 x 10 2 CFU/ml for treated sewage water, and 1.1 x 10 1 CFU/ml for tap water.

PAGE 40

28 Variability in results among the water samples was attributed to the presence of humic substances which serve as PCR inhibitors. Pre-enrichment in GN broth served to dilute these PCR inhibitors while allowing the S. flexneri to multiply, thereby increasing the concentration of target DNA. Real-time PCR: The Future for the Detection of Shigella spp. in Food Real-time PCR utilizes a fluorescently labeled oligonucleotide probe or a nonspecifically binding intercalating dye which allows for detection of generated product after each cycle of the PCR reaction. Total assay time is greatly reduced as compared to conventional PCR as there is no need for post-reaction analysis. The ABI Prism 7000 Sequence Detection System (Applied Biosystems, Foster City, CA) runs in a 96-well format and ABI offers support in primer and probe development (Applied Biosystems, 2003). Furthermore, the Taqman Universal Master Mix can be utilized to reduce laboratory preparation, contamination risks, and assay to assay variation. The Smart Cycler II System (Cepheid, Sunnyvale, CA) offers more flexibility with its sixteen ICORE (Intelligent Cooling/Heating Optical Reaction) modules (Cepheid 2003). The Smart Cyclers software allows separate experiments with unique cycling protocols to be carried out and analyzed simultaneously. The LightCycler (Roche Applied Science, Indianapolis, IN) provides the fastest PCR results by use of its unique capillary sample tubes format. Air heating of the 32 capillary tubes, as compared to a thermal block in other units, allows for rapid PCR cycling and a faster overall assay (Roche Applied Science, 2003). A 30-40 cycle PCR reaction can usually be completed in 20 to 30 minutes with the LightCycler All real-time PCR systems provide quantitative results based on assay specific standard curves or the incorporation of internal controls.

PAGE 41

CHAPTER 3 METHODS AND MATERIALS Preliminary Trials Preparation of Microbiological Media Wild strain cultures were grown in Tryptic Soy Broth (TSB) (Difco, Sparks, MD) and maintained on Tryptic Soy Agar (TSA) (Difco) slants. Adapted strains were cultured and maintained using TSB and TSA supplemented with 80 g/ml rifampicin (TSB-R80 and TSA-R80, respectively). All dilutions and tomato rinses were performed using Phosphate Buffered Saline (PBS) prepared using PBS tablets (ICN Biomedicals Inc., Aurora, OH). Enrichment media Shigella broth was prepared according to the U.S. Food and Drug Administrations (1998) Bacteriological Analytical Manual (FDA BAM) and supplemented with novobiocin at 3.0 g/ml (SB3.0), 0.5 g/ml (SB0.5), or 0.3 g/ml (SB0.3). EE Broth (Difco) was prepared as directed by manufacturers instructions and supplemented with novobiocin at 1.0 g/ml (EE1.0). For trials involving antibiotic supplemented enrichment media, rifampicin was added at 50 g/ml (SB3.0-R50, SB0.5-R50, SB0.3-R50, and EE1.0-R50). When necessary, the pH was adjusted using filter sterilized 1N NaOH. Enrichment media were prepared fresh for each experiment. Solid media MacConkey Agar (MAC) (Difco), Salmonella-Shigella Agar (SSA) (Difco), Shigella Plating Medium (SPM) (RF Laboratories, West Chicago, IL), Triple Sugar Iron 29

PAGE 42

30 (TSI) (Difco), Lysine Iron Agar (LIA) (Difco) and Motility Medium (MM) (Difco) were all prepared according to manufacturers instructions. When necessary, the pH was adjusted using filter sterilized 1N NaOH. MAC, SSA, and SPM were each poured into one compartment of a three-compartment Petri dish (Tri-Plate). TSI and LIA were prepared as slants and MM was prepared according to the FDA BAM. Acquisition and Maintenance of Shigella cultures Outbreak strains of Shigella sonnei and Shigella boydii were obtained from Dr. Hans Blascheks Laboratory at the University of Illinois, Department of Food Science and Human Nutrition. The Shigella boydii serotype 18 strain, encoded UI02, was isolated from a person involved in a bean salad outbreak in Chicago in March of 1999. The Shigella sonnei strain, encoded UI05, was isolated from a patient during an outbreak. Both strains, UI02 and UI05, originated from the State of Illinois Department of Public Health, Chicago. In addition, a strain of Shigella boydii serotype 18 (ATCC 35966) (encoded UI01), a human isolate of Shigella sonnei (encoded UI03) from an outbreak involving bean dip, and a human isolate of Shigella sonnei (encoded UI04) from an outbreak involving cilantro were also obtained. Strains UI03 and UI04 originated from the Enteric Bacteriology Unit, Microbial Diseases Laboratory, State of California Department of Health Services. Additional Shigella cultures were obtained to provide additional positive controls and to facilitate representation of each of the four serogroups. Shigella dysenteriae serotype 1 (ATCC 9361) and Shigella sonnei (ATCC 9290) were purchased from the American Type Culture Collection (ATCC). A strain of Shigella flexneri was obtained from Dr. Linda Harris Laboratory at the University of California, Davis, Department of

PAGE 43

31 Food Science and Technology. This strain originated from Dr. Keith Lampel at the U.S. Food and Drug Administration. Upon receipt, each strain was grown in 10 ml TSB at 37C (30 rpm) overnight. Overnight cultures were plated for isolation onto MAC and incubated overnight at 37C. Overnight plates were examined for typical growth. One typical colony of each strain was transferred to a TSA slant and stored at 4C. Another typical colony was transferred per product instructions to Protect Bacterial Preservers (Scientific Device Laboratory, Inc., Des Plaines, IL) and stored at -76C. Adaptation of Cultures to Rifampicin Adaptation of Shigellae to the bactericidal agent rifampicin was accomplished by challenging cultures in enrichment broth with low doses and increasing the dosage with each successive 24 hr transfer. A 10,000 ppm (1%) stock solution of rifampicin was prepared by dissolving 2.0 g rifampicin (Fisher # BP267925, Fisher Scientific, Pittsburg, PA) in 200 ml deionized water. This stock solution was then filter sterilized and stored in the dark at room temperature. Stock cultures were grown overnight in 10 ml TSB (37C, 30 rpm). Overnight cultures were transferred to 10 ml TSB with the progression of 2.5, 5.0, 10, 25, 40, 60, and 80 ppm rifampicin (TSB-R2.5, TSB-R5.0, TSB-R10, TSB-R25, TSB-R40, TSB-R60, and TSB-R80, respectively) and grown overnight (37C, 30 rpm). Once the cultures were adapted to 80 ppm rifampicin, cultures were grown overnight (37C, 30 rpm) three consecutive times in TSB-R80 to ensure well adapted populations. Once adaptation was complete, the final overnight adapted culture was plated for isolation onto MAC-R80 and incubated overnight at 37C. One typical colony from the

PAGE 44

32 overnight MAC-R80 plate was transferred to a TSA-R80 slant and stored at 4C. Another typical colony was transferred per product instructions to a Protect Bacterial Preserver and stored at minus 76C. Preparation of Optical Density Standard Curve For all standard curve studies, non-adapted cultures were grown using TSB and TSA. All adapted cultures were grown using TSB-R80 and TSA-R80. Procedures given below are worded for un-adapted cultures. A sterile wooden stick was used to transfer stock culture from a TSA or TSA-R80 slant into a 10 ml TSB tube. The tube was incubated overnight (37C, 30 rpm). A 10 l aliquot of the first overnight culture was transferred to fresh 10 ml TSB and grown overnight (37C, 30 rpm). Also from the first overnight culture, 10 l was used to inoculate 100 ml TSB in an Erlenmeyer flask. This flask was incubated (37C, 30 rpm) and observed every 30 minutes to determine the lag time associated with each strain. End of lag phase was determined by visible growth (cloudiness) in the TSB flask. A 10 l aliquot of the second overnight culture was transferred to fresh 10 ml TSB and grown for 18 hours (37C, 30 rpm). A 10 L aliquot of this 18 hour culture was then transferred into each of three Erlenmeyer flasks containing 100 ml TSB to provide three replicates. Each flask was labeled 1, 2, or 3 and inoculated five minutes apart in numerical order and incubated (37C, 30 rpm). After the pre-determined lag phase was complete, sampling began in 30 minute intervals. At each sampling, 1.25 ml of culture was transferred to a disposable cuvette and the absorbance was read at 600 nm using a spectrophotometer (Shimadzu Scientific Instruments, model UV-1201). Additionally, 1 ml of the culture was used to prepare

PAGE 45

33 serial dilutions using PBS. Appropriate dilutions were pour plated with TSA and incubated 24 to 48 hours at 37C. TSA plates were analyzed for growth and plates containing colonies in the countable range (25-250) were counted. Replicate results were averaged yielding one count of colony forming units (CFU) per ml for each sampling. Replicate results of absorbance values were also averaged. Using Excel software (Microsoft, Redmon, WA), a graph of the absorbance versus log 10 CFU/ml was prepared. The linear range was identified and subjected to linear regression analysis to form a standard curve. This standard curve was then used to estimate CFU/ml by means of absorbance at 600 nm in TSB. DNA Extraction of Stock Shigella Cultures Stock Shigella cultures used in this study are listed in Table 3-1. DNA was extracted from stock Shigella cultures via the DNeasy Tissue Kit (Qiagen, Valencia, CA). Stock cultures were grown overnight in 10 ml TSB (37C, 30 rpm). Overnight cultures were plated for isolation onto MAC and incubated overnight at 37C. Overnight plates were examined for typical growth and one typical colony was transferred to 10 ml TSB and grown overnight (37C, 30 rpm). A 1 ml aliquot of overnight culture was transferred to a clean, sterile 1.5 ml microcentrifuge tube and centrifuged for 10 minutes at 7,500 rpm. Supernatant was discarded and the resulting pellet was re-suspended in 180 l Buffer ATL (supplied with DNeasy Tissue Kit). DNA extraction continued from this point from step 2 of the DNeasy Protocol for Animal Tissues as per product literature. Final DNA elution was performed twice; once with 200 l Buffer AE (supplied with DNeasy Tissue Kit), and a

PAGE 46

34 Table 3-1. Stock Shigella cultures. DNA from each stock Shigella spp. was extracted from a 1 ml aliquot of an overnight culture in TSB (37C, 30 rpm) using a DNeasy Tissue Kit. Extracted DNA was transferred to a clean, sterile 1.5 microcentrifuge tube and stored at minus 20C until use. Extracted DNA was used to evaluate specificity of each primer set. DNA Code Culture Origin 01 Shigella boydii serotype 18 ATCC 35966 02 Shigella boydii serotype 18 Outbreak isolate 03 Shigella sonnei Patient isolate 04 Shigella sonnei Patient isolate 05 Shigella sonnei Outbreak isolate 06 Shigella sonnei ATCC 9290 07 Shigella flexneri FDA, Dr. Keith Lampel 08 Shigella dysenteriae serotype 1 ATCC 9361 second time with 50 l Buffer AE. This yields a final elution of 250 l DNA template in one 1.5 ml microcentrifuge tube. DNA templates for stock Shigella cultures were stored at minus 20C. Crude DNA Extraction of Stock Non-Shigella Cultures Stock non-Shigella cultures used as negative controls in this study are listed in Table 3-2. DNA from non-Shigella stock cultures was extracted to provide negative control templates. Stock cultures frozen on Protect Bacterial Preservers were retrieved from minus 76C storage and thawed. One bead was aseptically transferred from the Protect Bacterial Preserver into 10 ml TSB and grown overnight (37C, 30 rpm). Overnight cultures were plated for isolation on an appropriate selective and differential medium and incubated overnight at 37C. Overnight plates were observed for typical colony morphologies. One typical colony was transferred to 10 ml TSB and grown overnight (37C, 30 rpm). A 1 ml aliquot of the overnight culture was transferred to a clean, sterile 1.5 ml microcentrifuge tube and centrifuged for 10 minutes at 7,500 rpm. Supernatant was

PAGE 47

35 Table 3-2. Stock non-Shigella cultures. DNA from each of the non-Shigella spp. cultures was extracted by boiling a 1 ml aliquot of an overnight culture for 10 minutes, then centrifuging away any cell wall material. Supernatants were transferred to clean, sterial 1.5 ml microcentrifuge tubes and used to test the specificity of each primer set. DNA Code Culture Origin 09 Salmonella Agona Dr. Harris, UC Davis 10 Salmonella Gaminara Dr. Harris, UC Davis 11 Salmonella Montevideo Dr. Harris, UC Davis 12 Salmonella Michigan Dr. Harris, UC Davis 13 Salmonella Poona Dr. Harris, UC Davis 14 Salmonella Enteritidis Dr. Rodrick, UF 15 Salmonella Typhimurium ABC Research 16 Escherichia coli O157:H7 Deibel Laboratories 17 Escherichia coli O157:H7 Deibel Laboratories 18 Escherichia coli O157:H7 Deibel Laboratories 19 Escherichia coli O157:H7 Deibel Laboratories 20 Escherichia coli JM109 Dr. Wright, UF 21 Escherichia coli K12 Dr. Wright, UF 22 Escherichia coli (ATCC 25922) ATCC 23 Citrobacter freundii Cantaloupe isolate 24 Klebsiella pneumoniae Soil isolate 25 Klebsiella ozoanae Cantaloupe isolate 26 Enterobacter cloacae Cantaloupe isolate discarded and the resulting pellet was re-suspended in 200 l double de-ionized, sterilized water (hereby referred to as PCR water). Samples were then boiled for 10 minutes in a dry bath incubator (Fisher Scientific, IsoTemp 125D). Supernatant (DNA template) was aseptically transferred to a clean, sterile 1.5 ml microcentrifuge tube and stored at minus 20C. Pellet was discarded. Acquisition and Maintenance of PCR Primers for the Detection of Shigella Primers were synthesized by Sigma Genosys (The Woodlands, TX). Upon receipt, primers were reconstituted with 10% TE Buffer (1.0 mM Tris-Cl, 0.1 mM EDTA pH 8.0) to yield a 100 mM stock solution. The working solution was a 1:10 dilution of the stock solution with PCR water. Primer sets investigated in this study are shown in Table 3-3.

PAGE 48

36 Table 3-3. Primers for the detection of Shigella spp. All primers sets below are specific for Shigella spp. and EIEC. Primer set 01-001 amplifies a 620 bp fragment of the ipaH gene. Primer set 01-002 amplifies a 215 bp fragment of the virA gene. Primer sets 01-003 and 01-004 are a nested primer set where 01-003 is the internal primer amplifying a 217 bp region of the invasion associated locus (ial) and 01-004 is the external primer amplifying a 320 bp fragment of the ial. Primer set 01-005 serves as an internal primer to set 01-001 and amplifies a 290 bp fragment of the ipaH gene. Primer Code Primer Sequence Source 01-001F 5 gtt cct tga ccg cct ttc cga tac cgt c 3 Sethabutr et al., 2000 01-001R 5 gcc ggt cag cca ccc tct gag agt ac 3 Sethabutr et al., 2000 01-002F 5 ctg cat tct ggc aat ctc ttc aca tc 3 Villalobo and Torres, 1998 01-002R 5 tga tga gct aac ttc gta agc cct cc 3 Villalobo and Torres, 1998 01-003F 5 ttt tta att aag agt ggg gtt tga 3 Lindqvist, 1999 01-003R 5 gaa cct atg tct acc tta cca gaa gt 3 Lindqvist, 1999 01-004F 5 ctg gta ggt atg gtg agg 3 Frankel et al., 1990 01-004R 5 cca ggc caa caa tta ttt cc 3 Frankel et al., 1990 01-005F 5 cca ctg aga gct gtg agg 3 Lampel (unpublished 2002) 01-005R 5 tgt cac tcc cga cac gcc 3 Lampel (unpublished 2002) Specificity of Primers The specificity of each primer set was tested against DNA from each strain of Shigella (Table 3-1) and a battery of non-Shigella DNA (Table 3-2). Amplification of target DNA sequences was performed in a 25 l reaction mixture in clean, sterile 0.2 ml polypropylene microcentrifuge tubes. The reaction mixture consisted of 15.75 l PCR water, 2.5 l 10x PCR Buffer (100 mM Tris-HCl, 500 mM KCl, 15 mM MgCl 2 pH 8.3), 2.5 l dNTP mixture (2.5 mM each), 1.0 l forward primer, 1.0 l reverse primer, 0.25 l taq polymerase (TaKaRa Taq Hot Start Version, Shiga, Japan), and 2 l DNA template. The thermocycler used to perform the PCR reaction was an Eppendorf Mastercycler gradient (Brinkmann-Eppendorf, Westbury, NY). Temperature programs used to evaluate each primer set are given in Table 3-4. Analysis of PCR Product by Gel Electrophoresis DNA amplicons were analyzed by gel electrophoresis. Gel used in all experiments

PAGE 49

37 Table 3-4. Temperature programs for PCR primers. All temperature programs were obtained from the sources of the primer sets listed above in Table 3-3. Primer Set Initial Heating Denaturation Annealing Extension Number of Cycles Final Extension Hold 01-001 94C 10 min 94C 15 sec 60C 30 sec 72C 45 sec 30 72C 2 min 4C 01-002 94C 10 min 94C 45 sec 65C 30 sec 72C 30 sec 35 72C 10 min 4C 01-003 94C 5 min 94C 30 sec 60C 1 min 72C 30 sec 30 72C 10 min 4C 01-004 94C 5 min 94C 30 sec 60C 1 min 72C 30 sec 30 72C 10 min 4C 01-005 94C 10 min 94C 15 sec 60C 15 sec 72C 15 sec 30 72C 2 min 4C was 3% NuSeive GTG Agarose (BioWhittaker Molecular Applications, Rockland, MA). Electrophoresis was performed on Thermo EC gel trays (model CSSU78115, Holbrook, NY) powered with a Thermo EC power supply (model EC105). Gel trays were leveled using a Thermo EC leveling platform (model CSSLP78). Gels were stained with ethidium bromide (FisherBiotech, BP1302-10). DNA amplicons and DNA markers (X174/Hinf I, Promega, Madison, WI) were visualized by UV transilluminator (Spectroline TE-312S, Westbury NY). Images of DNA amplicon bands were captured using a photodocumentation handheld camera (Fisherbiotech, FB-PDC-34) loaded with Polaroid Type 667 high-speed print film (Fisher Scientific, 04-441-91). Inoculated Studies Acquisition of Tomato Tomatoes of the Florida cultivar 47 were obtained from a nearby packinghouse (DiMare, Palmetto, FL). Tomatoes were pulled prior to the wash/wax line in order to retain the normal bacterial population. Upon arrival, tomatoes were stored at 13C until

PAGE 50

38 use. Only fully green or tomatoes with less than 50% red color were used. All tomatoes with greater than 50% red color were discarded. Inoculum Preparation Three days prior to each experiment, stock culture stored on TSA-R80 slants at 4C was grown (37C, 30 rpm) in a 10 ml tube of TSB-R80. Overnight transfers were performed using 10 ml tubes of TSB-R80 each day. On the day of the experiment, an 18-hour culture (late stationary phase) was centrifuged (4,000 x g for 10 minutes) and washed twice with PBS. The washed culture was serially diluted using 9 ml tubes of PBS. Appropriate dilutions were pour-plated with TSA-R80 to confirm cell titer. Inoculation of Tomatoes and Subsequent Recovery Prior to each experiment, tomatoes were removed from cold storage and allowed to warm to room temperature (~21C). Large plastic trays were sanitized using reagent alcohol, 70% v/v (ethanol 63% v/v, methanol 3.5% v/v, isopropanol 3.5% v/v, and water balance) (LabChem Inc., Pittsburg, PA) prior to each experiment. Tomatoes were placed on the tray stem scar down. Using an Eppendorf Repeater Micropipette, ten 10 L aliquots of the appropriate dilution were spot inoculated around the blossom scar of each tomato without inoculating directly on the blossom scar (Figure 3-1). S. sonnei was inoculated onto tomato surfaces at levels of 10 5 10 4 10 3 10 2 10 1 and 10 0 CFU/tomato. S. boydii was inoculated onto tomato surfaces at levels of 10 6 10 5 10 4 10 3 10 2 10 1 and 10 0 CFU/tomato. Inoculated tomatoes were allowed to air dry completely prior to continuing with recovery.

PAGE 51

39 Figure 3-1. Spot inoculation of tomatoes. Tomatoes were inoculated with ten 10 l spots around but not touching the blossom scar. Inocula were allowed to air dry prior to recovery. After inocula were completely dry, tomatoes were transferred aseptically to a sterile stomacher bag containing 100 ml PBS (Figure 3-2). Each bag was sealed using stomacher bag clips. Inoculum was recovered by methods similar to that of Beuchat et al. (2001), modified to include vigorous shaking and hand rub/manipulation (15 seconds shake/ 15 seconds rub/ 15 seconds shake). Experimental Design Thirty tomatoes were prepared at each inoculation level and rinsed as described above. Ten tomato rinses were analyzed for S. boydii UI02 and S. sonnei UI05 using standard enrichment media, ten were analyzed using enrichment media supplemented with rifampicin, and ten were analyzed using the polymerase chain reaction (PCR). For standard enrichment, 25 ml of the tomato rinse was transferred to each of three 18 oz Whirl-Pak bags (Nasco, Modesto, CA) containing 225 ml of appropriate enrichment

PAGE 52

40 Figure 3-2. Stomacher bag with inoculated tomato. Inoculated tomatoes were aseptically transferred to a sterile stomacher bag which contained 100 ml PBS. The bag was sealed using stomacher bag clips and the inoculum was recovered using a vigorous shaking and hand rub/manipulation method (15 seconds shake/ 15 seconds rub/ 15 seconds shake). media. Inoculation studies involving S. sonnei utilized enrichments in SB0.3 (44C, anaerobically) (FDA BAM), SB0.5 (37C) (CMMEF), and EE1.0 (42C) (EE Broth). Inoculation studies involving S. boydii utilized enrichments in SB3.0 (42C, anaerobically) (FDA BAM), SB3.0 (37C) (CMMEF), and EE1.0 (42C) (EE Broth). Anaerobic conditions were generated using the Pack-Anaero anaerobic gas generating system (Mitsubishi Gas Chemical Company, Inc. (MGC), Japan) and 7.0 Liter Pack-Rectangular Jars (MGC). For antibiotic supplemented enrichment, SB0.3-R50, SB0.5-R50, SB3.0-R50, and EE1.0-R50 were used in place of standard enrichment media and incubated at the same conditions. After 24 hours, each enrichment bag was mixed and sterile wooden sticks were used to streak the enrichment for isolation on each compartment of a Tri-Plate. Tri-Plates were incubated overnight (37C).

PAGE 53

41 Confirmation of Typical Colonies on Tri-Plates Typical Shigella isolates from all plating media on Tri-Plates were carried through the following confirmation process. TSI and LIA slants were inoculated with suspect colonies and incubated overnight (37C). TSI and LIA slants demonstrating typical reactions for Shigella were used to inoculate MM and a 10 ml tube of TSB which were then incubated overnight (37C). Growth in TSB from samples demonstrating no motility in MM was streaked for isolation on MAC and incubated overnight (37C). Biochemical reactions were tested using the BBL Enterotube II (Becton Dickinson, Sparks, MD). Enterotubes were incubated overnight (37C) and positive reactions were read according to manufacturers instructions. Assembly of Tandem Filter Funnels Two Whatman 25 mm disposable filter funnels with grade 4 filters (Whatman Clifton, NJ) were assembled in tandem (Figure 3-2) with the following modifications. In the top filter funnel, the grade 4 filter was aseptically lifted and a 25 mm diameter VWR 413 filter was placed underneath. The threading on the funnel was lined with a layer of Parafilm M laboratory film (American National Can, Chicago, IL). In the bottom filter funnel, the grade 4 filter was replaced with a 25 mm FTA filter, aseptically cut from a FTA Classic Card (Whatman Cat, No. WB12 0205). The FTA filter was covered with a 25 mm diameter plastic shield, which had a 6 mm hole punched in its center. The plastic shield was aseptically cut from a plastic weight boat. The top filter funnel was inserted into the top of the bottom filter funnel and the joint was sealed with Parafilm. Tomato rinses to be analyzed by PCR were transferred to Oxford 4 oz. specimen cups (Cat # OX-067, International BioProducts, Bothell, WA) to facilitate easy pouring.

PAGE 54

42 Leave lid on top of tandem filter assembly A Wrap threading with Parafilm Grade 4 filter ( 20 25 microns ) 25 mm VWR 413 filter ( 5 microns ) Joint sealed with Parafilm B 25 mm Plastic shield with 6 mm punch to channel filtrate to center of FTA filter 25 mm FTA filter Grade 4 filter from this filter f unnel i s d iscarded Figure 3-2. Assembly of tandem filter funnels. (A) Top filters are for size exclusion, (B) bottom FTA filter is for trapping bacteria, lysing bacterial cell walls, and binding bacterial DNA. A ring stand equipped with a clamp was used to secure a 500 ml Pyrex vacuum flask. The side arm of the vacuum flask was connected to a water trap consisting of a 500 ml side arm flask filled with Drierite (anhydrous calcium sulfate) (W.A. Hammond Drierite Company Ltd., Xenia, OH) which was then connected to a vacuum pump (Emerson,

PAGE 55

43 Figure 3-3. Vacuum flask apparatus with tandem filter funnels. Filter funnels are attached the top of a 500 ml filter flask via a plastic stem (supplied with the filter funnels) inserted through the center of a number 7 rubber stopper. The filter flask is attached via Tygon tubing to another filter flask which is set up as a water trap. The water trap filter flask is connected to a vacuum pump. Filtration of tomato rinses were facilitated by vacuum at 400 mm Hg. model SA55NX6TE-4870, St. Louis, MO) using Tygon tubing. The top of the flask was sealed with a No. 7 rubber stopper in which a 6 mm hole had been bored through the center, through which a 4 inch plastic stem (supplied with the filter funnels) had been inserted. Tomato rinse was filtered through the tandem filter funnel via vacuum pressure (400 mm Hg). After all of the rinse had passed through the FTA filter, the vacuum pressure was turned off and the tandem filter funnel removed from the plastic stem. The FTA filter and plastic shield was aseptically removed and placed in one compartment of a three compartment Petri dish. FTA filters and shields were allowed to air dry overnight with the shield faced down. Using the holes in the plastic shields as a guide, 6

PAGE 56

44 mm punches were aseptically taken from each FTA filter and placed in a clean, sterile 1.5 microcentrifuge tube. FTA punches were washed twice with 500 ml FTA Purification Reagent (Whatman Clifton, NJ), then twice with 500 ml TE Buffer. Washed punches were aseptically transferred to Petri dishes and dried in a 37C incubator with the lids slightly open. Nested PCR Amplification of ipaH gene for Detection of Shigella Dry, washed FTA punches were transferred to 0.5 ml microcentrifuge tubes and subjected to a nested, two step PCR reaction. The PCR step 1 reaction used FTA punches as DNA template submerged in 200 l of PCR step1 reaction mix. The PCR step 1 reaction mix consisted of 142 l PCR water, 20 l 10x HotMaster Taq Buffer (with 25 mM Mg 2+ pH 8.5), 20 l dNTP mixture (2.5 mM each), 8.0 l primer 01-001F, 8.0 l reverse primer, 2 l HotMaster Taq Polymerase (Eppendorf, Hamburg, Germany). Prior to PCR, 0.5 microcentrifuge tubes were centrifuged (30 seconds at 15,000 x g) to ensure the FTA punch was completely submerged in step 1 reaction mix. The cycling conditions for step 1 were: initial denaturation at 94C for 10 min; 30 cycles of denaturation at 94C for 15 sec, annealing at 60C for 30 sec, and elongation at 72C for 45 sec; and a final elongation at 72C for 2 minutes. PCR step 2 amplified a segment from within the target DNA segment of step 1. A 1 l aliquot from the PCR step 1 product was diluted using 99 l PCR water. Amplification in PCR step 2 was performed in 25 l reactions in clean, sterile 0.2 ml microcentrifuge tubes. The step 2 reaction mixture consisted of 15.75 l PCR water, 2.5 l 10x PCR Buffer (100 mM Tris-HCl, 500 mM KCl, 15 mM MgCl 2 pH 8.3), 2.5 l dNTP mixture (2.5 mM each), 1.0 l primer 01-005F, 1.0 l primer 01-005R, 0.25 l HotMaster taq polymerase, and 2 l DNA template (from the 1:100 dilution of step 1 product). The

PAGE 57

45 cycling conditions for step 2 were: initial denaturation at 94C for 10 min; 30 cycles of denaturation at 94C for 15 sec, annealing at 60C for 15 sec, and elongation at 72C for 15 sec; and a final elongation at 72C for 2 minutes. DNA amplicons were analyzed by gel electrophoresis as described above. The amplicon in step 1 was 620 bp and the amplicon from step 2 was 290 bp. Recording of Data and Statistical Evaluation All results from inoculation studies involving S. boydii UI02 and S. sonnei UI05 were recorded as either positive or negative for detection. Positive enrichment was scored based on the isolation of inoculated S. boydii UI02 or S. sonnei UI05 by at least one of the three plating media. Positive isolation by plating media resulted from typical reactions for Shigella spp. in all confirmation steps and identification from Enterotubes. Positive detection by PCR methods resulted from amplification of a single band of the appropriate base pairs. Logistical regression models were constructed using the R software (R Development Core Team, Version 1.7.0 Patched) to identify significant differences between isolation of Shigella spp. by PCR methods and enrichment methods and isolation of Shigella spp. on the three plating media. Models for evaluating PCR methods and enrichments were constructed (without an intercept) with covariate factors for enrichments/PCR and inoculation levels. Models for evaluating plating media were constructed (without an intercept) using plating media as covariates. Multiple comparisons were performed using the Bonferroni method since this method is conservative in its estimates of significance compared to other multiple comparison methods. P values of < 0.05 were considered significant.

PAGE 58

CHAPTER 4 RESULTS This study consisted of two phases of research. The first phase consisted of preliminary trials involving the preparation of growth curves, optical density standard curves, and testing the specificity of each set of primers against a DNA library of positive and negative controls. The second phase consisted of inoculation studies where tomatoes were spot inoculated with S. boydii UI02 and S. sonnei UI05 at pre-determined levels. Recovery/detection of the inocula was tested using conventional culture methods, conventional culture methods with rifampicin supplemented enrichment, and a newly developed FTA filtration/ nested PCR method. Preliminary Trials Growth Curves and Optical Density Standard Curves Growth curves and optical density (O.D.) standard curves were prepared for S. boydii UI02 wild strain, S. sonnei UI05 wild strain, and S. sonnei 9290 rifampicin adapted. Since the growth curve and O.D. standard curve for S. sonnei 9290 wild strain was similar to that of S. sonnei UI05 wild strain, results for the 9290 wild strain are not shown. Results from growth curve and O.D. standard curve data were used to cultivate consistent inocula, thereby reducing variability in later recovery studies on tomatoes. S. boydii UI02 wild strain The growth curve for S. boydii UI02 wild strain demonstrates that stationary phase was reached in approximately 8 hours (Figure 4.1). A lag phase of approximately 4 hours was observed prior to exponential growth. 46

PAGE 59

47 0.00.20.40.60.81.01.21.402468Time (hours)A (600 nm 10 ) Figure 4-1. Growth curve: S. boydii UI02 wild strain. Each of three 100 ml TSB microcosms were inoculated with 10 l aliquots of an 18-hour S. boydii UI02 culture. The microcosms were incubated (37C, 30 rpm) and sampled at appropriate time intervals. At each time of sampling, the absorbance at 600 nm (A 600 nm) was measured spectrophotometrically. Shown above is the average A 600 nm of the three trials plotted against time. An O.D. standard curve for S. boydii UI02 wild strain was prepared. The cell titer in which the relationship between log 10 CFU/ml and absorbance at 600 nm (A 600 nm) showed linearity (R 2 = 0.9926; Figure 4-2) was approximately 7.03 x 10 7 to 3.18 x 10 8 CFU/ml. S. sonnei UI05 wild strain The growth curve prepared for S. sonnei UI05 wild strain demonstrates stationary phase was reached in approximately 6 hours (Figure 4-3), with an initial lag phase of approximately 3 hours.

PAGE 60

48 R2 = 0.992600.20.40.60.811.27.77.98.18.38.58Log Count (CFU/ml)A (600 nm .7 ) Figure 4-2. Optical density standard curve for S. boydii UI02 wild strain. Each of three 100 ml TSB microcosms were inoculated with 10 l aliquots of an 18-hour S. boydii UI02 culture. The microcosms were incubated (37C, 30 rpm) and sampled at appropriate time intervals. At each time of sampling, appropriate serial dilutions in PBS were pour-plated using TSA and incubated overnight (37C). Shown above is the average absorbance at 600 nm (A 600 nm) of the three trials (from Fig 4-1) plotted against the log 10 CFU/ml for the data points which form a linear relationship. 0.00.20.40.60.81.01.21.401234567Time (hours)A (600 nm 8 ) Figure 4-3. Growth curve: S. sonnei UI05 wild strain. Each of three 100 ml TSB microcosms were inoculated with 10 l aliquots of an 18-hour S. sonnei UI05 culture. The microcosms were incubated (37C, 30 rpm) and sampled at appropriate time intervals. At each time of sampling, the absorbance at 600 nm (A 600 nm) was measured spectrophotometrically. Shown above is the average A 600 nm of the three trials plotted against time.

PAGE 61

49 An O.D. standard curve for S. sonnei UI05 wild strain was prepared. The cell titer in which the relationship between log 10 CFU/ml and absorbance at 600 nm (A 600 nm) showed linearity (R 2 = 0.9833; Figure 4-4) was approximately 4.13 x 10 7 to 6.43 x 10 8 CFU/ml. R2 = 0.986600.20.40.60.811.21.47.47.67.888.28.48.68.89Log Count (CFU/ml)A (600 nm ) Figure 4-4. Standard curve of O.D. (600 nm) of S. sonnei UI05 wild strain compared to log plate count. Each of three 100 ml TSB microcosms were inoculated with 10 l aliquots of an 18-hour S. sonnei UI05 culture. The microcosms were incubated (37C, 30 rpm) and sampled at appropriate time intervals. At each time of sampling, appropriate serial dilutions in PBS were pour-plated using TSA and incubated overnight (37C). Shown above is the average absorbance at 600 nm (A 600 nm) of the three trials (from Fig 4-1) plotted against the log 10 CFU/ml for the data points which form a linear relationship. S. sonnei 9290 rifampicin adapted strain The growth curve for S. sonnei 9290 rifampicin adapted strain demonstrates stationary phase was reached in approximately 8 hours (Figure 4-5), with an initial lag phase of approximately 5 hours.

PAGE 62

50 0.00.20.40.60.81.00246810Time (hours)A (600 nm 12 ) Figure 4-5. Growth curve: S. sonnei 9290 rifampicin adapted strain. Each of three 100 ml TSB-R50 microcosms were inoculated with 10 l aliquots of an 18-hour S. sonnei 9290 culture. The microcosms were incubated (37C, 30 rpm) and sampled at appropriate time intervals. At each time of sampling, the absorbance at 600 nm (A 600 nm) was measured spectrophotometrically. Shown above is the average A 600 nm of the three trials plotted against time. An O.D. standard curve for S sonnei 9290 rifampicin adapted strain was prepared. The cell titer in which the relationship between log 10 CFU/ml and absorbance at 600 nm (A 600 nm) showed linearity (R 2 = 0.9944; Figure 4-6) was approximately 3.26 x 10 7 to 8.20 x 10 8 CFU/ml. Primer Specificity Each primer set (Table 3-3) was tested against DNA templates from both stock Shigella spp. (Table 3-1) and closely related microorganisms (Table 3-2). Primer set 01-001, which amplifies a 620 bp region of the ipaH gene of Shigella spp. and enteroinvasive E. coli (EIEC), successfully amplified DNA from all eight strains of Shigella. Primer set 01-002, which targets a 215 bp region of the virA gene of Shigella spp. and EIEC, successfully amplified DNA from S. boydii UI02, S. flexneri, and S. dysenteriae ATCC 9361, however it did not amplify a product from DNA from any of the

PAGE 63

51 R2 = 0.994400.20.40.60.8177.588.599.5Log Count (CFU/ml)A (600 nm ) Figure 4-6. Standard curve of O.D. (600 nm) of S. sonnei 9290 rifampicin adapted strain compared to log plate count. Each of three 100 ml TSB-R50 microcosms were inoculated with 10 l aliquots of an 18-hour S. boydii UI02 culture. The microcosms were incubated (37C, 30 rpm) and sampled at appropriate time intervals. At each time of sampling, appropriate serial dilutions in PBS were pour-plated using TSA-R50 and incubated overnight (37C). Shown above is the average absorbance at 600 nm (A 600 nm) of the three trials (from Fig 4-1) plotted against the log 10 CFU/ml for the data points which form a linear relationship. four S. sonnei strains or S. boydii serogroup 18 (ATCC 35966). Primer set 01-004 and 01-003, which amplify a 320 bp region of the invasion associated locus (ial) and a 217 bp region from within that 320 bp region, respectively, successfully amplified a product from the DNA of S. boydii UI02 only, while producing no product from S. sonnei UI05, S. flexneri, and S. dysenteriae. Primer sets 01-004 and 01-003 were not tested against DNA from the other four strains of Shigella. Primer set 01-005, which amplifies a 290 bp region from within the region amplified by primer set 01-001, was successful in amplifying DNA from all eight strains of Shigella tested. Primer sets 01-001, 01-002, and 01-005 did not amplify DNA from any of the non-Shigella DNA (Table 3-2). Primer sets 01-004 and 01-003 were only tested against non

PAGE 64

52 Shigella DNA of Salmonella Gaminara and Salmonella Typhimurium, producing no amplification in either sample. Inoculated Studies Detection of Shigella spp. by Conventional Culture Methods Tomato rinses were enriched according to protocols of the FDA BAM, the CMMEF, and in EE broth as described by Uyttendaele et al. (2000). Overnight enrichments were then plated using SSA, MAC, and SPM. For all conventional culture trials, 10 replicates were analyzed at each inoculation level. Inoculation levels were verified by pour plating serial dilutions in triplicate with TSA-R50. Recovery results for all conventional culture method experiments are expressed as percent recovery, defined as the number of tomatoes which tested positive for Shigella spp. out of the total number tested. S. boydii UI02 was not recovered by conventional culture methods from any samples inoculated at 10 6 10 5 or 10 4 CFU/tomato. For this reason, trials at the lower inoculation levels were not performed for S. boydii UI02. Figure 4-7 demonstrates the percent recovery of S. sonnei UI05 by conventional culture methods. At inoculation levels of 10 5 10 4 and 10 3 CFU/tomato, 10% recovery of S. sonnei UI05 was observed by CMMEF enrichment and plating on SSA and MAC. When these enrichments were plated on SPM however, 20% recovery was observed from inoculation levels of 10 5 and 10 3 CFU/tomato, and 10% recovery from 10 4 CFU/tomato. No recovery of S. sonnei UI05 was observed with CMMEF enrichment of inoculation levels of 10 2 10 1 or 10 0 CFU/tomato.

PAGE 65

53 0%20%40%60%80%100%SSAMACSPMSSAMACSPMSSAMACSPM CMMEFFDA BAMEE Broth Percent Recover y 10E5 10E4 10E3 10E2 10E1 10E0 Figure 4-7. Recovery of S. sonnei UI05 by conventional culture methods. Tomatoes were inoculated with S. sonnei UI05 at levels of 10E5 to 10E0. Inocula was recovered via a 100 ml PBS rinse using a shake-rub-shake method as described in Chapter 3. Tomato rinses were enriched by Shigella culture methods of the Compendium of Methods for the Microbiological Examination of Foods (CMMEF), the FDA Bacteriological Analytical Manual (FDA BAM), and in EE broth with 1.0 g/ml novobiocin at 42C. Percent recovery is calculated as the number of tomatoes from which S. sonnei UI05 was isolated over the total number of tomatoes sampled for each enrichment. S. sonnei UI05 recovered by enrichment by the FDA BAM is also shown in Figure 4-7. S. sonnei UI05 inoculated at 10 5 CFU/tomato was recovered at 60%, 70%, and 70% when plated on SSA, MAC, and SPM, respectively. When inoculated at 10 4 CFU/tomato, S. sonnei UI05 was recovered at 10%, 10%, and 20% when plated on SSA, MAC, and SPM, respectively. At inoculation levels of 10 3 CFU/tomato, recovery rates of 20%, 30%, and 40% were observed when plated on SSA, MAC, and SPM, respectively. At an inoculation of 10 2 CFU/tomato, S. sonnei UI05 was recovered from 20% of tomatoes when plated on SSA, MAC, or SPM. S. sonnei UI05 was not recovered when inoculated at levels of 10 1 or 10 0 CFU/tomato.

PAGE 66

54 Results from enrichment of S. sonnei UI05 in EE broth as described by Uyttendaele et al. (2000) is also shown in Figure 4-7. S. sonnei UI05 inoculated at 10 5 CFU/tomato was recovered from 10%, 40%, and 50% of tomatoes when enrichments were plated on SSA, MAC, and SPM, respectively. With an inoculation of 10 4 CFU/tomato, S. sonnei UI05 was only recovered when plated on SPM (20%). When inoculated at 10 3 CFU/tomato, S. sonnei UI05 was recovered from 10% of tomatoes when plated on SSA, MAC, or SPM. When inoculated at 10 2 CFU/tomato, S. sonnei UI05 was recovered from 10% of tomatoes when plated on MAC and SPM, but in none when plated on SSA. No recovery of S. sonnei UI05 was observed when inoculated at levels of 10 1 or 10 0 CFU/tomato. Detection of Shigella spp. by Conventional Culture Methods with Rifampicin Supplemented Enrichment Conventional culture methods (FDA BAM, the CMMEF, and in EE broth as described by Uyttendaele et al. (2000)) were repeated using enrichments supplemented with 50g/ml rifampicin to exclude natural tomato microflora and rifampicin-adapted inocula. Overnight enrichments were plated using SSA, MAC, and SPM. For all conventional culture trials with rifampicin supplemented enrichment, 10 replicates were analyzed at each inoculation level. Inoculation levels were verified by pour plating serial dilutions in triplicate with TSA-R50. Recovery results for all conventional culture method experiments are expressed as percent recovery, defined as the number of tomatoes which tested positive for Shigella spp. out of the total number tested. Results for the recovery of S. boydii UI02 using conventional culture methods where enrichments were supplemented with 50 g/ml rifampicin (rif+) are shown in Figure 4-8. Supplemented enrichment according to the CMMEF (rif+) protocol resulted

PAGE 67

55 in 100% recovery of S. boydii UI02 from tomatoes inoculated at 10 6 10 5 and 10 3 CFU/tomato on all three plating media. S. boydii UI02 inoculated at 10 4 CFU/tomato was recovered from 40%, 80%, and 70% of tomatoes when plated on SSA, MAC, and SPM, respectively. At an inoculation level of 10 2 CFU/tomato, S. boydii UI02 was recovered from 60% of tomatoes when plated on SSA, and 50% when plated using MAC or SPM. When inoculated at 10 1 CFU/tomato, S. boydii UI02 was recovered from 10% regardless of which plating medium was used. 0%20%40%60%80%100%SSAMACSPMSSAMACSPMSSAMACSPM CMMEF (rif+)FDA BAM (rif+)EE Broth (rif+) Percent Recover y 10E6 10E5 10E4 10E3 10E2 10E1 Figure 4-8. Recovery of S. boydii UI02 by conventional culture methods with rifampicin supplemented enrichment. Tomatoes were inoculated with rifampicin adapted S. boydii UI02 at levels of 10E6 to 10E1. Inocula were recovered via a 100 ml PBS rinse using a shake-rub-shake method as described in Chapter 3. Tomato rinses were enriched by Shigella culture methods of the Compendium of Methods for the Microbiological Examination of Foods (CMMEF), the FDA Bacteriological Analytical Manual (FDA BAM), and in EE broth with 1.0 g/ml novobiocin at 42C. All enrichments were supplemented with 50 g/ml rifampicin (rif+)to screen out natural background tomato microflora. Percent recovery is calculated as the number of tomatoes from which S. boydii UI02 was isolated over the total number of tomatoes sampled for each enrichment. Results from the enrichment of S. boydii UI02 by the FDA BAM (rif+) protocol are shown in Figure 4-8. 100% recovery of S. boydii UI02 was achieved from tomatoes

PAGE 68

56 inoculated at 10 6 10 5 10 4 and 10 3 CFU/tomato on all three plating media, except for tomatoes inoculated at 10 4 CFU/tomato and plated on SSA, which were recovered at 90%. At an inoculation level of 10 2 CFU/tomato, S. boydii UI02 was recovered from 90% of tomatoes on all three plating media. When inoculated at 10 1 CFU/tomato, S. boydii UI02 was recovered from 10%, 20%, and 30% of tomatoes when plated using SSA, MAC, and SPM, respectively. Figure 4-8 shows that enrichment in EE broth (rif+) as described by Uyttendaele et al. (2000) resulted in almost no recovery of S. boydii UI02. Only one tomato from the 10 3 CFU/tomato inoculation tested positive for S. boydii UI02 when plated on SSA. Results for the recovery of S. sonnei UI05 using conventional culture methods where enrichments were supplemented with 50 g/ml rifampicin (rif+) are shown in Figure 4-9. Supplemented enrichment of S. sonnei UI05 according to the CMMEF (rif+) protocol resulted in 100% recovery when inoculated at 10 5 CFU/tomato, 90% recovery when inoculated at 10 4 CFU/tomato, and 80% recovery when inoculated at 10 3 CFU/tomato and plated on SSA, MAC, or SPM. When S. sonnei UI05 was inoculated at 10 2 CFU/tomato, recovery was 20%, 30%, and 50% when plated on SSA, MAC, and SPM, respectively. At an inoculation level of 10 1 CFU/tomato, S. sonnei UI05 was not recovered when plated on SSA, and recovered from 20% of tomatoes when plated on MAC or SPM. No S. sonnei UI05 was recovered from inoculation levels of 10 0 CFU/tomato. Results from the enrichment of S. sonnei UI05 by the FDA BAM (rif+) protocol are shown in Figure 4-9. 100% recovery was achieved from tomatoes inoculated at 10 5 CFU/tomato when plated SSA, MAC, or SPM. At an inoculation level of 10 4

PAGE 69

57 0%20%40%60%80%100%SSAMACSPMSSAMACSPMSSAMACSPM CMMEF (rif+)FDA BAM (rif+)EE Broth (rif+) Percent Recover y 10E5 10E4 10E3 10E2 10E1 10E0 Figure 4-9. Recovery of S. sonnei UI05 by conventional culture methods with rifampicin supplemented enrichment. Tomatoes were inoculated with rifampicin adapted S. sonnei UI05 at levels of 10E5 to 10E0. Inocula were recovered via a 100 ml PBS rinse using a shake-rub-shake method as described in Chapter 3. Tomato rinses were enriched by Shigella culture methods of the Compendium of Methods for the Microbiological Examination of Foods (CMMEF), the FDA Bacteriological Analytical Manual (FDA BAM), and in EE broth with 1.0 g/ml novobiocin at 42C. All enrichments were supplemented with 50 g/ml rifampicin (rif+) to screen out natural background tomato microflora. Percent recovery is calculated as the number of tomatoes from which S. sonnei UI05 was isolated over the total number of tomatoes sampled for each enrichment. CFU/tomato, S. sonnei UI05 was recovered from 80% of tomatoes when plated on SSA, and 90% of tomatoes when plated on MAC or SPM. When inoculated with 10 3 CFU/tomato, S. sonnei UI05 was recovered from 60% of tomatoes when plated on SSA, and 80% of tomatoes when plated on MAC or SPM. For inoculation levels of 10 2 and 10 1 CFU/tomato, S. sonnei UI05 was recovered from 50% and 10% of tomatoes, respectively, when plated on SSA, MAC, or SPM. No S. sonnei UI05 was recovered when inoculated at 10 0 CFU/tomato. Results from the enrichment of S. sonnei UI05 in EE broth (rif+) as described by Uyttendaele et al. (2000) are shown in Figure 4-9. 100% recovery was observed at

PAGE 70

58 inoculation levels of 10 5 CFU/tomato when plated on SSA, MAC, or SPM. At inoculation levels of 10 4 CFU/tomato, S. sonnei UI05 was recovered from 90% of tomatoes when plated on SSA, and 100% of tomatoes when plated on MAC or SPM. When inoculated at a level of 10 3 CFU/tomato, S. sonnei UI05 was recovered from 30% of tomatoes when plated on SSA and 80% of tomatoes when plated on MAC or SPM. At inoculation levels of 10 2 CFU/tomato, S. sonnei UI05 was recovered from 10% of tomatoes when plated on SSA and 20% of tomatoes when plated on MAC or SPM. No S. sonnei UI05 was recovered from tomatoes inoculated with 10 1 CFU/tomato when plated on SSA, however S. sonnei UI05 was recovered from 10% of tomatoes when plated on MAC or SPM. No S. sonnei UI05 was recovered when inoculated at 10 0 CFU/tomato. Lowest Detection Levels of Conventional Culture Methods Results were reported based on the initial inoculation, however only 25 ml of the 100 ml tomato rinse was enriched by each protocol, therefore only a fourth of the inocula could have theoretically been enriched by each protocol. When reporting the lowest detection level (LDL) for conventional enrichments, corrections for the distribution of inoculum have been made. For example, results reported for S. sonnei UI05 at 10 5 CFU/tomato were reported for an initial inoculation of 6.1 x 10 5 CFU/tomato, but each enrichment procedure actually reflected a theoretical cell titer of 1.5 x 10 5 CFU, assuming all of the inoculum was recovered and distributed evenly in the rinse. LDLs from the enrichment procedures are reported as the lowest inoculation that resulted in isolation of Shigella spp. in at least one out of the 10 replicates. Conventional enrichment procedures in the presence of natural tomato microflora resulted in no recovery of S boydii UI02 (LDL >5.3 x 10 5 CFU/tomato); however LDLs for S. sonnei UI05 were 1.9 x 10 2 (FDA BAM), 1.5 x 10 3 (CMMEF), and 1.1 x 10 2 CFU/tomato (EE broth). For

PAGE 71

59 enrichment procedures supplemented with rifampicin (rif+) to exclude natural tomato microflora, LDLs were: 6.3 x 10 0 CFU/tomato (FDA BAM rif+ and CMMEF rif+) and >5.3 x 10 5 CFU/tomato in EE broth rif+ for S. boydii UI02; and 1.9 x 10 1 (FDA BAM rif+ and CMMEF rif+) and 1.1 x 10 1 CFU/tomato (EE broth rif+) for S. sonnei UI05. Table 4-1. Lowest detection levels (LDLs) of conventional culture methods. Tomatoes were inoculated with various levels of S. boydii UI02 or S. sonnei UI05. Inocula were allowed to air dry before tomatoes were rinsed in 100 ml PBS using a shake-rub-shake method. A 25 ml aliquot of the tomato rinse was transferred to 225 ml of the appropriate enrichment broth and incubated according Shigella culture methods found in the Compendium of Methods for the Microbiological Examination of Foods (CMMEF), the U.S. FDAs Bacteriological Analytical Manual (FDA BAM) or in EE Broth as described by Uyttendaele et al. (2000). Enrichment procedures were repeated using rifampicin adapted strains and supplemented enrichment (rif+). LDLs were determined by calculating the inoculum applied to the respective tomato, assuming 100% of the inoculum was recovered in the PBS rinse, and adjusting for how much of that inoculum went into each enrichment. Enrichment Procedure S. boydii UI02 (CFU/tomato) S. sonnei UI05 (CFU/tomato) CMMEF >5.3 x 10 5 1.5 x 10 3 FDA BAM >5.3 x 10 5 1.9 x 10 2 EE Broth >5.3 x 10 5 1.1 x 10 2 CMMEF rif+ 6.3 x 10 0 1.9 x 10 1 FDA BAM rif+ 6.3 x 10 0 1.9 x 10 1 EE Broth rif+ >5.3 x 10 5 1.1 x 10 1 Additionally, the lowest detection level in which Shigella spp. was isolated in 100% of the replicates (LDL100) was also determined for each enrichment procedure. In studies involving conventional enrichment procedures in the presence of natural tomato microflora, LDL100s were not achieved in this study, however from the results it can be stated that the LDL100s of S. boydii UI02 and S. sonnei UI05 are >5.3 x 10 5 CFU/tomato and >1.5 x 10 5 CFU/tomato, respectively. For trials involving rifampicin supplemented enrichments, the LDL100s were: 6.3 x 10 2 CFU/tomato (FDA BAM and CMMEF) and >5.3 x 10 5 CFU/tomato (EE broth) for S. boydii UI02; and 1.5 x 10 5 CFU/tomato (FDA

PAGE 72

60 BAM and CMMEF), and 1.5 x 10 4 CFU/tomato (EE broth) for S. sonnei UI05. The high inoculation levels required to achieve the LDL100 in conventional culture methods demonstrates the need for better methods with which to evaluate food products for Shigella spp. Table 4-2. Lowest detection levels in which S. boydii UI02 or S. sonnei UI05 was isolated in 100% of replicates (LDL100s) of conventional culture methods. Tomatoes were inoculated with various levels of S. boydii UI02 or S. sonnei UI05. Inocula were allowed to air dry before tomatoes were rinsed in 100 ml PBS using a shake-rub-shake method. A 25 ml aliquot of the tomato rinse was transferred to 225 ml of the appropriate enrichment broth and incubated according Shigella culture methods found in the Compendium of Methods for the Microbiological Examination of Foods (CMMEF), the U.S. FDAs Bacteriological Analytical Manual (FDA BAM) or in EE Broth as described by Uyttendaele et al. (2000). Enrichment procedures were repeated using rifampicin adapted strains and supplemented enrichment (rif+). LDL100s were determined by calculating the inoculum applied to the respective tomato, assuming 100% of the inoculum was recovered in the PBS rinse, and adjusting for how much of that inoculum went into each enrichment. Enrichment Procedure S. boydii UI02 (CFU/tomato) S. sonnei UI05 (CFU/tomato) CMMEF >5.3 x 10 5 >1.5 x 10 5 FDA BAM >5.3 x 10 5 >1.5 x 10 5 EE Broth >5.3 x 10 5 >1.5 x 10 5 CMMEF rif+ 6.3 x 10 2 1.5 x 10 5 FDA BAM rif+ 6.3 x 10 2 1.5 x 10 5 EE Broth rif+ >5.3 x 10 5 1.5 x 10 4 Detection of Shigella spp. by FTA Filtration / Nested PCR FTA filtration/ nested PCR was used to detect inoculated S. boydii UI02 and S. sonnei UI05 on inoculated tomatoes. Positive detection was recorded if a single band of 290 bp was produced in the second step of the nested PCR. Inoculation levels were verified by pour plating serial dilutions in triplicate with TSA-R50. Figure 4-10 shows representative gels from the first step of the nested PCRs. Positive amplification in the first PCR produced a very faint band of 620 bp (Figure 4-10A) from samples inoculated at 10 5 and 10 4 CFU/tomato with either S. boydii UI02 or S.

PAGE 73

61 sonnei UI05 and no bands (Figure 4-10B) from samples inoculated at 10 3 10 2 10 1 and 10 0 CFU/tomato. 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 620 bp 13 14 15 16 17 18 19 20 21 22 23 24 13 14 15 16 17 18 19 20 21 22 23 24 A B Figure 4-10. Representative gels from the first step nested PCR. S. boydii UI02 inoculated at (A) 10 4 CFU/tomato and (B) 10 2 CFU/tomato. Lanes assignments for both gels are: 1, 12, 13, and 24 empty; 2, 11, 14, and 23 DNA ladder; 3-10, 15, and 16 sample lanes; 17 tomato control; 18 filter control; 19 PBS control; 20 water control; 21 negative control; 22 positive control. Figure 4-11 shows a representative gel from the second step of the nested PCRs. Positive amplification in the second step PCR resulted in a single band of 290 bp. The results shown in Figure 4-12 demonstrate that S. boydii UI02 was detected at rates of 100% regardless of inoculation level. S. sonnei UI05 was detected in 100% of samples inoculated at levels of 10 5 10 4 and 10 3 CFU/tomato. When inoculated at 10 2 10 1 and 10 0 CFU/tomato, S. sonnei UI05 was detected on 90%, 40%, and 30% of the tomatoes, respectively.

PAGE 74

62 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Figure 4-11. Representative gel from the second step nested PCR. S. boydii UI02 inoculated at 10 4 CFU/tomato. Lanes assignments are: 1, 12, 13, and 24 empty; 2, 11, 14, and 23 DNA ladder; 3-10, 15, and 16 sample lanes; 17 tomato control; 18 filter control; 19 PBS control; 20 water control; 21 negative control; 22 positive control. Lowest Detection Levels of the FTA Filtration/ Nested PCR Method Results were reported based on initial inoculation levels. Since the entire tomato rinse was evaluated by FTA filtration/ nested PCR, no corrections for the distribution of inoculum was required. Reported results assumed recovery of 100% of the inoculum from the tomato surface. LDLs of the FTA filtration/ nested PCR method for the detection of S. boydii UI02 and S. sonnei UI05 were 6.2 x 10 0 CFU/tomato and 7.4 x 10 0 CFU/tomato, respectively (Table 4-3). In comparison, the LDL100s for S. boydii UI02 and S. sonnei UI05 were 6.2 x 10 0 CFU/tomato and 6.1 x 10 3 CFU/tomato, respectively (Table 4-3).

PAGE 75

63 0%20%40%60%80%100%10E510E410E310E210E110E0Inoculation Level (CFU/tomato)Percent Recover y S. sonnei S. boydii Figure 4-12. Detection of S. boydii UI02 and S. sonnei UI05 by FTA filtration/ nested PCR. Tomatoes were inoculated with S. boydii UI02 or S. sonnei UI05 at levels of 10E5 to 10E0. Inocula were recovered via a 100 ml PBS rinse using a shake-rub-shake method as described in Chapter 3. Tomato rinses were filtered using tandem filter funnels as described in Chapter 3. FTA punches were used as DNA template in the first step of the nested PCR. Positive detection was based on visualization of a 290 bp band from the second step of the nested PCR. Percent recovery is calculated as the number of tomatoes from which S. boydii UI02 or S. sonnei UI05 was isolated over the total number of tomatoes sampled. Table 4-3. Lowest detection levels and lowest detection levels in which S. boydii UI02 or S. sonnei UI05 was isolated in 100% of replicates of FTA filtration/ nested PCR. Tomatoes were inoculated with S. boydii UI02 or S. sonnei UI05 at levels of 10E5 to 10E0. Inocula were recovered via a 100 ml PBS rinse using a shake-rub-shake method as described in Chapter 3. Tomato rinses were filtered using tandem filter funnels as described in Chapter 3. FTA punches were used as DNA template in the first step of the nested PCR. Positive detection was based on visualization of a 290 bp band in the second step of the nested PCR. S. boydii UI02 (CFU/tomato) S. sonnei UI05 (CFU/tomato) LDL 6.2 x 10 0 7.4 x 10 0 LDL100 6.2 x 10 0 6.1 x 10 3

PAGE 76

CHAPTER 5 DISCUSSION AND CONCLUSION Foodborne shigellosis has been on the rise in recent years. Several outbreaks involving fresh produce have demonstrated a need for a better method of evaluating produce for Shigella spp. This study investigated conventional culture methods and a newly developed FTA filtration/ nested PCR method for the detection of Shigella spp. on tomato surfaces. S. boydii UI02 and S. sonnei UI05 were inoculated onto tomatoes at levels of 10 6 through 10 0 CFU/tomato and 10 5 through 10 0 CFU/tomato, respectively. Recovery of Shigella from tomatoes was performed by placing inoculated tomatoes in 100 ml PBS and applying a shake/rub/shake method as previously described. Tomato rinses were then analyzed by enrichment procedures as found in the U.S. Food and Drug Administrations (1998) Bacteriological Analytical Manual (FDA BAM), the Compendium of Methods for the Microbiological Examination of Food (CMMEF), and by EE broth as described by Uyttendaele et al. (2000). Shigella Plating Medium (SPM) was evaluated for the isolation of Shigella spp. against Salmonella-Shigella agar (SSA), MacConkey agar (MAC). Furthermore, tomato rinses were evaluated using a newly developed FTA filtration/ nested PCR method. Preliminary Studies The two Shigella spp. isolates selected for this study were S. boydii UI02 and S. sonnei UI05. Both strains were obtained from the laboratory of Dr. Hans Blaschek, University of Illinois, and were involved in Chicago area outbreaks involving fresh 64

PAGE 77

65 produce or products containing produce. S. sonnei represents the serogroup of Shigella most common to North America, while S. boydii is rarely isolated in North America. In order to evaluate the capability of conventional culture methods given total specificity for Shigella spp., rifampicin resistance was generated in S. boydii UI02 and S. sonnei UI05 subcultures by spontaneous mutation. Initially, dimethyl sulfoxide (DMSO) was used to make stock solutions of rifampicin; however the use of DMSO resulted in the production of strong malodors determined to be from the metabolism of DMSO to dimethyl sulfide and methanethiol (data not shown). As the source of the malodor was being determined, cultures were adapted to nalidixic acid as an alternative to rifampicin; however in preliminary trials, natural microflora on tomatoes proved resistant of up to 400 ppm nalidixic acid. After the source of malodor was determined to be DMSO, methanol was used to make stock solutions of rifampicin, which did not result in the production of malodors. In a study on antimicrobial activities of aqueous and methanol extracts of Juniperus oxycedrus, Karaman et al. (2003) used the addition of methanol alone to various media as negative controls. Methanol was found to have no inhibitory effects to various strains of Enterobacter, Klebsiella, Acinetobacter, and Pseudomonas along with one strain of Escherichia coli. Based on these findings with closely related microorganisms, methanol was assumed to have negligible effect on cell viability of S. boydii UI02 and S. sonnei UI05. Growth Characteristics of S. boydii UI02 and S. sonnei UI05 Growth characteristics of S. boydii UI02 and S. sonnei UI05 were determined by performing growth curves and constructing optical density (O.D.) standard curves. In a preliminary trial, a growth curve of S. sonnei 9290 (data not shown) was performed and the absorbance was read at two wavelengths, 400 nm and 600 nm. When O.D. standard

PAGE 78

66 curves were constructed using data from both wavelengths, there was no difference in the range of linearity between absorbance and log 10 CFU/ml. For all the remaining growth curves a wavelength of 600 nm was used. Using growth curve and O.D. standard curve data, the exponential growth phase doubling times for each strain investigated was calculated (Table 5-1). Table 5-1. Doubling times associated with exponential growth phase of investigated strains of Shigella spp. Strain S. boydii UI02 S. sonnei 9290 S. sonnei UI05 Wild type 43.8 min 22.9 min 30.0 min Nalidixic acid adapted 47.4 min N/A 45.0 min Rifampicin adapted N/A 30.5 min N/A It was observed that S. sonnei UI05 entered exponential phase growth and achieved stationary phase faster than S. boydii UI02. This can be explained by the faster growth rate calculated for S. sonnei UI05 and an extension in the lag phase associated with S. boydii UI02. The wild strain of S. sonnei UI05 doubled every 30 minutes compared to every 43.8 minutes for S. boydii UI02 (Table 5-1). The lag phase prior to exponential growth observed for the wild strain of S. boydii UI02 was approximately 4 hours (Figure 4-1) compared to only approximately 3 hours for the wild strain of S. sonnei UI05 (Figure 4-3), with the onset of stationary phase at approximately 8 hours and 6 hours, respectively. Growth curve data from the wild strain S. sonnei 9290 was comparable to that of the wild strain S. sonnei UI05. The time before exponential growth observed in the wild strain S. sonnei 9290 was approximately 3 hours with the onset of stationary phase at approximately 6 hours (data not shown). The S. sonnei 9290 wild strain doubled every 22.9 minutes compared to every 30.0 minutes for the S. sonnei UI05 wild strain When compared with data collected for a rifampicin adapted strain of S. sonnei 9290 (Figure 4

PAGE 79

67 5), entry into exponential growth of rifampicin-adapted culture was delayed by approximately 2 hours. Furthermore, the doubling time in exponential growth phase increased to 30.0 minutes with the rifampicin adapted strain. For rifampicin adapted strains of S. boydii UI02 and S. sonnei UI05, similar growth characteristics to the S. sonnei 9290 rifampicin adapted strain were assumed. These assumptions included a 2 hour extension of stationary phase between the wild strain and rifampicin-adapted strain and slightly slower growth rates during exponential growth phase. This assumption was supported by growth curve data collected for strains of S. boydii UI02 and S. sonnei UI05 adapted and grown in the presence of nalidixic acid (data shown in Appendix). Exponential growth rates of nalidixic acid-adapted strains grown in the presence of nalidixic acid for both S. boydii UI02 and S. sonnei UI05 were slower than the observed rates of the wild strains (Figure 5-1). Both nalidixic acid-adapted strains experienced 4 hour extensions in lag phase compared to that of the wild strain, which supports the assumption of a 2 hour extension for both strains adapted to rifampicin. Assumptions for rifampicin adapted strains of S. boydii UI02 and S. sonnei UI05 maintained entry into stationary phase at no later than 12 hours. Growth curve data concluded that 18-hour cultures used in this study were late stationary phase cultures. Evaluation of Primers Specific for Shigella spp. Several sets of primers were investigated for the detection of Shigella spp. by PCR. Results indicated that primer sets 01-001 and 01-005 were the only sets specific for all four serogroups of Shigella. None of the primers amplified DNA from the non-Shigella library. Primer sets 01-001 and 01-005 targeted 620 and 290 bp regions, respectively, of the ipaH gene of Shigella spp. and enteroinvasive E. coli. When Jin et al. (2002) sequenced the genome of S. flexneri 2a, the ipaH gene was located seven times on the

PAGE 80

68 chromosome and five more times on the plasmid. Primer sets 01-002 and the nested set 01-003 and 01-004 target regions of the virA gene and the invasion associated locus (ial), respectively. Unlike the ipaH gene, the virA gene and the ial are only located on the large virulence plasmid of Shigella spp. Negative results observed with primer sets 01-002, 01-003, and 01-004 in this study could have been due to loss of plasmid during storage or repeated transfers of the Shigella strains not amplified. No studies were conducted to determine if plasmids were intact in these strains. The potential loss of plasmid does yield an advantage to using primers which amplify the ipaH gene since detection of non-pathogenic, plasmid-less strains would still be possible. Inoculation Studies Evaluation of Enrichment Protocols Three enrichment protocols were investigated for the isolation of S. boydii UI02 and S. sonnei UI05 from inoculated tomatoes in the presence of natural background microflora. When the S. sonnei UI05 data (Figure 4-7) were analyzed using a logistic regression model, no significant difference (P = 1.000) among enrichment procedures was observed. S. boydii UI02, however, was never recovered regardless of inoculation level in any of the three investigated enrichment procedures. The inability for conventional enrichment protocols to recover inoculated S. boydii UI02 demonstrates the difficulty of isolating Shigella spp. by traditional methods and the need for more selective and sensitive media. Conventional culture enrichment was repeated with background contamination eliminated using enrichments supplemented with 50 g/ml rifampicin (rif+) and rifampicin-adapted inocula. In these studies, S. boydii UI02 was recovered by CMMEF rif+ and FDA BAM rif+ protocols but not from the EE broth rif+ enrichment (Figure 4

PAGE 81

69 8), while S. sonnei UI05 was recovered by all three enrichments (Figure 4-9). Logistic regression modeling revealed no significant difference (P = 1.000) between the CMMEF rif+ and the FDA BAM rif+ enrichment methods for the recovery of S. boydii UI02, and no significant differences (P = 1.000) among the CMMEF rif+, FDA BAM rif+, and EE broth rif+ enrichments for the recovery of S. sonnei UI05. It was observed that although S. boydii UI02 was unable to compete amid natural tomato microflora, it was recovered at higher percentages than S. sonnei UI05 at each inoculation level in the CMMEF rif+ and FDA BAM rif+ enrichments. In contrast, S. boydii UI02 was rarely recovered when enriched in EE broth rif+ (1 out of 60 samples; Figure 4-8). EE broth contains bile salts which have been reported to be inhibitory to some Shigella spp. particularly stressed cells (Tollison and Johnson, 1985). The S. boydii strain used in this study may have been inhibited by the bile salts in EE broth. When Uyttendaele et al. (2000) investigated EE broth among other enrichment broths for the recovery of stressed and un-stressed Shigellae, only strains of S. sonnei and S. flexneri were used. These results stress the importance of including all four serogroups of Shigella when evaluating enrichment procedures. These results do not suggest EE broth may be used in place of Shigella broth for the enrichment of Shigella spp. since a known pathogenic strain, S. boydii UI02, may be missed. Evaluation of Plating Media Using logistic regression models, the isolation of Shigella spp. among isolation media was compared. In all studies involving plating media, there was no significant difference ( = 0.05) among isolation rates on the three media. It can be noted, however, that differentiation of S. boydii UI02 and S. sonnei UI05 colonies from those of background contaminants was far easier on SPM than on MAC or SSA. The most

PAGE 82

70 common contaminants observed in this study were species of Enterobacter, Citrobacter, and Klebsiella. These background contaminants were also identified in other studies in which Shigella spp. isolation was attempted by conventional culture methods (Uyttendaele et al., 2000; Schneider et al., unpublished data). Figure 5-1 demonstrates the ability of SPM to allow easier differentiation of various contaminants as compared to MAC and SSA. Colonies of Shigella spp. on SPM are white while colonies of Enterobacter, Citrobacter, Acenitobacter, and Klebsiella are either bluish or greenish. On MAC, Shigella spp. are translucent and slightly pink, with and without rough edges. Enterobacter, Citrobacter, and Klebsiella spp. all make pink colonies on MAC which may or may not be easily differentiated from colonies of Shigella spp. Since differentiation on MAC is based on lactose fermentation, lactose negative strains of Enterobacter, Citrobacter, and Klebsiella will produce colonies that are more translucent, resembling those of Shigella spp. Furthermore, Acinetobacter produces tiny translucent to pink colonies on MAC which are difficult to differentiate from Shigella spp. On SSA, Enterobacter and Klebsiella spp. produce pink colonies however Shigella, Citrobacter and Acinetobacter spp. produce colonies that are translucent to light pink. Isolation of Shigella spp. on MAC and SSA became increasingly difficult as colony morphologies of Shigella spp. and/or several contaminant colony morphologies were present on the same plate. While plating on SPM allowed greater differentiation between Shigella spp. and contaminant colonies, the results support previous suggestions that several different media with varying selectivity should be used in order to increase the chance of isolating Shigella spp. If only one plating

PAGE 83

71 3 1 2 3 1 2 A B 3 1 2 3 1 2 C D 3 1 2 3 1 2 E F Figure 5-1. Differentiation of background microflora by isolation media. Tri-Plates contain: 1) SSA, 2) MAC, and 3) SPM, streaked with: A) Enterobacter cloacae, B) Klebsiella ozanae, C) Citrobacter freundii, D) Acinetobacter anitratus, E) Shigella boydii UI02, and F) Shigella sonnei UI05.

PAGE 84

72 media is to be used, the results in this study suggest SPM can be used with equivalent isolation rates of Shigella spp. as compared to MAC and SSA. Analysis of Lowest Detection Levels of Conventional Culture Methods The lowest detection levels (LDLs) of conventional culture methods for S. boydii UI02 and S. sonnei UI05 were determined (Table 4-1). In a similar study, Jacobson et al. (2002) evaluated the FDA BAM Shigella method using two strains of S. sonnei on selected types of produce. LDLs were determined using unstressed, chill-stressed, and/or freeze-stressed cells. LDLs with unstressed cells were less than 1.0 x 10 1 CFU/25g for all produce types, while LDLs with chill-stressed and freeze-stressed cells were less than 5.2 x 10 1 CFU/25g for all produce types tested (Jacobson et al., 2002). LDLs with the FDA BAM enrichment in this study for S. sonnei UI05 (1.9 x 10 2 CFU/tomato; Table 4-1) were higher than LDLs reported by Jacobson et al. for S. sonnei strains 357 and 20143. Variation between strains of S. sonnei could explain the difference in reported LDLs with the FDA BAM method. Based upon its prevalence in outbreaks of shigellosis in the U.S., Jacobson's group chose to evaluate only strains of S. sonnei; however, results of this study demonstrate the importance of including other serogroups as well. While the FDA BAM appears to be quite effective based on evaluations with strains of S. sonnei (present study and Jacobson et al., 2002) the LDLs observed for S. boydii UI02, a strain responsible for a Chicago area outbreak, was >5.3 x 10 5 CFU/tomato. This indicates more effective methods are needed for the detection of Shigella spp. in food. The lowest detection levels in which the inoculum was recovered from 100% of the replicates (LDL100s) of the conventional culture methods were also determined for S. boydii UI02 and S. sonnei UI05 (Table 4-2). LDL100s of >5.3 x 10 5 CFU/tomato and >1.5 x 10 5 CFU/tomato were observed for S. boydii UI02 and S. sonnei UI05,

PAGE 85

73 respectively, by conventional culture methods in the presence of natural tomato microflora. The inability of the conventional culture methods to detect S. boydii UI02 or S. sonnei UI05 at high levels of contamination demonstrates the need for a more sensitive and specific conventional culture method for the detection of Shigella spp. on tomato surfaces. When rif+ enrichments were used to select rifampicin-adapted inocula, LDL100s were 6.3 x 10 2 CFU/tomato (FDA BAM rif+ and CMMEF rif+) and >5.3 x 10 5 CFU/tomato (EE Broth rif+) for S. boydii UI02 and 1.5 x 10 5 CFU/tomato (FDA BAM rif+ and CMMEF rif+) and 1.5 x 10 4 CFU/tomato (EE Broth rif+) for S. sonnei UI05. The exclusion of natural tomato microflora by rif+ enrichment did not facilitate the consistent detection of inoculated S. boydii UI02 or S. sonnei UI05 at levels sufficient to cause disease, further demonstrating the need for a more sensitive and specific conventional culture method for the detection of Shigella spp. on tomato surfaces. Sources of Variation Among Trials In Figure 4-8, inconsistent recovery results were obtained from S. boydii UI02 data sets at the 10 4 CFU/tomato inoculation levels. Recovery by conventional culture methods with rifampicin supplemented enrichment at the 10 5 CFU/tomato and 10 3 CFU/tomato both resulted in 100% recovery, while the 10 4 CFU/tomato was less than 100% for the CMMEF enrichment. In this study, trials involving 10 6 10 5 and 10 4 CFU/tomato inoculations were performed on one day, while the trials involving 10 3 10 2 and 10 1 CFU/tomato inoculations were performed on another day. This gap in recovery could possibly be explained by variations in attachment and/or potential biofilm formation. In support of this theory, the 10 4 CFU/tomato set was run last on its day of sampling, while the 10 3 CFU/tomato was run first on its day of sampling. This would allow a significantly greater amount of time for the10 4 CFU/tomato inoculum to attach and/or form biofilms

PAGE 86

74 while the 10 6 and 10 5 CFU/tomato data sets were run that day. Unpublished data using the S. boydii UI02 strain (Blaschek et al., University of Illinois) has suggested the formation of biofilms on the surfaces of produce, which resulted in difficult removal and resistance to produce washes. Another factor was that tomatoes used in different trials were harvested and shipped in different lots. On some trial days the tomatoes available were visibly cleaner than for other trials. The presence of excess filth on the surface of tomatoes might have contributed to sub-optimal attachment. The potential for S. boydii UI02 and S. sonnei UI05 to form biofilms can be tested using several methods. To simply determine whether these strains produce biofilms, two traditional tests could be performed: the tube test (Christensen et al., 1982) or the microtiter-plate test (Christensen et al., 1982; Christensen et al., 1985). The tests involve staining the bacterial film with a cationic dye and reading the degree of film formation either visually (tube test) or by optical density (microtiter-plate assay). The major drawbacks of these tests are in their qualitative nature; the tube test is visual, while the microtiter-plate assay only measures the film formation on the bottom of the well (Stepanovi et al., 2000). In order to better quantify biofilm formation, Stepanovi et al. (2000) describe a modified microtiter-plate assay which involves fixing of bacterial films with methanol, staining with crystal violet, releasing the bound dye with 33% glacial acetic acid, and measuring the optical density with a plate reader. This assay could be performed with S. boydii UI02 and S. sonnei UI05 strains using several time periods for attachment/formation to determine the rate of biofilm formation, if any. Conversely, biofilm formation on tomato surfaces could be directly observed by environmental scanning electron microscopy (Blaschek et al., unpublished material).

PAGE 87

75 The effect of filth on tomato surfaces on S. boydii UI02 and S. sonnei UI05 attachment and subsequent recovery could be tested as well. Tomatoes with visible filth could be obtained and a portion washed and rinsed to remove any visible filth. Using rifampicin adapted strains, dirty and clean tomatoes could be used in inoculation/recovery procedures as described in Chapter 3, Materials and Methods. Enumeration of recovered S. boydii UI02 and S. sonnei UI05 from both clean and filthy tomatoes should provide insight as to their effect on attachment/recovery characteristics. Comparison of Conventional Culture Methods and FTA Filtration/ Nested PCR Logistic regression analysis was used to compare isolation rates of the conventional culture methods to the detection rates of FTA filtration/ nested PCR method for S. boydii UI02 and S. sonnei UI05. For studies involving conventional culture methods and S. sonnei UI05, the FTA filtration/ nested PCR method was significantly better than the CMMEF (P = 0.007), the FDA BAM (P = 0.003), and the EE broth enrichment (P = 0.001). For studies involving rifampicin supplemented enrichments and S. boydii UI02, the FTA filtration/ nested PCR method was significantly better than enrichment by the CMMEF (P = 0.010) or enrichment in EE broth (P < 0.001), however it was not significantly different then enrichment by the FDA BAM (P = 0.177). For studies involving rifampicin supplemented enrichments and S. sonnei UI05, the FTA filtration/ nested PCR method was significantly better than the CMMEF (P = 0.007), the FDA BAM (P = 0.003), and enrichment in EE broth (P = 0.001). When tomatoes were analyzed by FTA filtration/ nested PCR method, much faster results were obtained than with conventional culture methods. The time required for a confirmed result by conventional culture methods was 4 days at best. Since positive confirmation was based on a positive result in the second step PCR, the FTA filtration/

PAGE 88

76 nested PCR could produce a confirmed result in as quickly as 2 days. It should be noted that until a tandem filter funnel assembly comparable to that used in this study can be commercially purchased, the assembly process of such tandem filters requires significant time, energy, and attention to aseptic techniques. The assembly process of tandem filter funnels alone provides a risk of contamination that needs to be addressed by proper laboratory practices. Contamination problems experienced in this study were attributed to the assembly process of tandem filter funnels. Analysis of Lowest Detection Levels of FTA Filtration/ Nested PCR The LDLs and LDL100s were determined for S. boydii UI02 and S. sonnei UI05 for FTA filtration/ nested PCR (Table 4-3). Other studies which investigated the detection of Shigella spp. in food by PCR techniques reported LDLs of: 1.0 x 10 1 -1.0 x 10 2 CFU/ml (Vantarakis et al., 2000), 1.0 x 10 2 -1.0 x 10 3 CFU/ml (Villalobo and Torres, 1998), 1.0 x 10 1 CFU/ml (Lindqvist, 1999), and 1.1 x 10 1 CFU/ml (Theron et al., 2001). Vantarakis et al. (2000) were only able to achieve detection of S. dysenteriae type 1 from homogenized mussel samples at 1.0 x 10 1 -1.0 x 10 2 CFU/ml after a 22 hour pre-enrichment step in peptone water was employed. Without the pre-enrichment a LDL of 1.0 x 10 3 CFU/ml was observed. In a similar study, Villalobo and Torres (1998) observed LDLs of S. dysenteriae type 1 to be 1.0 x 10 2 -1.0 x 10 3 CFU/ml in homogenized mayonnaise samples. Lindqvist (1999) utilized buoyant density centrifugation food inoculated with S. flexneri to increase sensitivity of a nested PCR assay from 1.0 x 10 3 CFU/ml to 1.0 x 10 1 CFU/ml. Finally, a 6 hour pre-enrichment step in GN broth was necessary for Theron et al. (2001) to detect S. flexneri at 1.1 x 10 1 CFU/ml in environmental water samples by semi-nested PCR. While the pre-enrichment increased

PAGE 89

77 sensitivity through multiplication of S. flexneri, it also served to dilute PCR inhibitors found in several of the environmental water samples. In comparison to these previous studies investigating the detection of Shigella spp. in food with PCR techniques, the FTA filtration/ nested PCR method detects S. boydii UI02 and S. sonnei UI05 with equivalent sensitivity. Tandem filter funnels were able to filter large volumes of tomato rinse, concentrating bacteria in the center of the FTA filter thereby increasing the ability to detect Shigella spp. present at low levels. In addition, no pre-enrichment step was required to increase sensitivity in FTA filtration/ nested PCR, therefore faster results can be obtained. Furthermore, washing protocols for FTA punches serve to remove any potential PCR inhibitors. Optimization of the FTA Filtration/ Nested PCR Assay Further development of the FTA filtration/ nested PCR method is required to obtain the desired amplification on the first step PCR. Figure 4-10A and 4-10B show electrophoresis gels obtained with step 1 PCR amplification of sample tomatoes inoculated with S. boydii UI02 at 10 4 CFU/tomato and 10 2 CFU/tomato, respectively. At the 10 4 CFU/tomato inoculation level only very faint bands can be seen upon close observation, while at 10 2 CFU/tomato no bands can be observed. In step 2 PCR however, all inoculation levels were successfully amplified as shown in Figure 4-11. If negative amplification in step 1 of the FTA filtration/ nested PCR method is to be used to determine a sample negative for Shigella spp., further optimization of this step is required as Shigella spp. present at levels 10 3 CFU/tomato would be missed. Preliminary optimization trials were performed using inoculation levels of 10 2 CFU/tomato in order to visualize bands from the step 1 PCR by adjusting the ramp rate of the annealing step from 3/sec to 1/sec, increasing the number of PCR cycles from 30 to

PAGE 90

78 40, sectioning the FTA punches, and adjusting the temperature control on the thermocycler from tube to block. The ramp rate and temperature control settings were adjusted to account for variation between our Eppendorf Mastercycler gradient and thermocyclers used in previous studies with FTA punches (Orlandi and Lampel, 2000; Lampel et al., 2000). Sectioning FTA punches into two or four equal parts with a sterile scalpel was attempted to allow the punch to remain completely submerged in PCR reaction mix for the duration of the PCR (Figure 5-2). Even though high speed centrifugation was used to force the FTA punches down into the reaction mix, the 6 mm diameter of the punch forced them back to the surface during PCR. FTA punch sections sank to the bottom of reaction tubes and remained there for the duration of the PCR. A B C Figure 5-2. FTA punches in 0.5 ml microcentrifuge tubes. A) Un-manipulated punches held at top of reaction mix by diameter; B) punches sectioned into halves; C) punches folded with forceps. Both changing the ramp rate for the annealing step and adjusting the temperature control from tube to block resulted in non-specific amplification (data not shown). Increasing PCR cycles from 30 to 40 alone, and increasing the number of cycles in combination with sectioning FTA punches into halves or quarters both resulted in observable bands from step 1 PCR (data not shown). As previously mentioned, the FTA punches were sectioned using a sterile scalpel and forceps. Although positive results were obtained from sectioning FTA punches, this practice is not recommended for future

PAGE 91

79 studies due to the increased potential for cross-contamination. Instead, FTA punches in future optimization studies will be folded during transfer using forceps against the inside wall of the microcentrifuge reaction tube prior to step 1 PCR. Folding allows the punches to remain completely submerged (Figure 5-2) in the PCR reaction mix to allow more optimal amplification without adding the potential chance of contamination associated with sectioning. Predictive Value of Testing for Shigella spp. It should be noted that the predictive value (PV) of any microbiological assay for the detection of Shigella spp. in food is affected by both the prevalence of Shigella spp. associated with that food product and the specificity and sensitivity of the testing procedure. In the Institute of Food Technologists (IFT) Expert Report on Emerging Microbiological Food Safety Issues, Implications for Control in the 21 st Century, PV of testing is explained as a function of prevalence, specificity, and sensitivity. According to the PV model, if the pathogen is prevalent at a high frequency the PV of a test with high specificity and sensitivity will be quite high. Conversely, if a pathogen is prevalent at very low frequency, then the PV of a test with high specificity and sensitivity remains quite low. Such is the case with Shigella spp. whose prevalence on produce has been demonstrated at 4.1% (FDA 2001a; FDA 2001b); therefore the PV of the sampling is very low. Although the conversion from conventional culture methods to nucleic-acid assays such as PCR will allow for greater test specificity and sensitivity, the PV of these tests to screen produce for potential Shigella spp. contamination will remain low. Conclusions The results of this study demonstrate the superiority of the FTA filtration/ nested PCR method over conventional culture methods for the detection of Shigella spp. from

PAGE 92

80 inoculated tomato surfaces. The FTA filtration/ nested PCR method was successful in detecting as few as 6.2 S. boydii UI02 cells and 7.2 S. sonnei UI05 cells amid high background contamination. If future proposed methods for the isolation/detection of Shigella spp. from food are to include a conventional culture analog, the results of this study suggest that either the enrichment protocols described by the FDA BAM or the CMMEF should be considered. EE broth, although adequately effective for S. sonnei UI05, did not recover S. boydii UI02 from inoculated tomato surfaces. Previous suggestions of using several isolation media of varying selectivity were supported by results of this study. SPM is capable of isolating S. boydii UI02 and S. sonnei UI05 at rates equivalent to MAC and SSA, while providing a greater differentiation between colonies of Shigella spp. and closely related contaminants.

PAGE 93

APPENDIX GROWTH CHARACTERISTICS OF NALIDIXIC ACID ADAPTED STRAINS Growth curves and optical density (O.D.) standard curves were prepared for S. boydii UI02 and S. sonnei UI05 strains adapted to nalidixic acid (NA). S. boydii UI02 NA adapted strain The growth curve for S. boydii UI02 NA adapted strain demonstrates that stationary phase was reached in approximately 13 hours (Figure B-1). A lag phase of approximately 7 hours was observed prior to exponential growth. 0.0000.2000.4000.6000.8001.0001.2003579111315Time (hours)A (600nm ) Figure A-1. Growth curve: S. boydii UI02 nalidixic acid adapted strain. Each of three 100 ml TSB (200 ppm NA) microcosms were inoculated with 10 l aliquots of an 18-hour S. boydii UI02 culture. The microcosms were incubated (37C, 30 rpm) and sampled at appropriate time intervals. At each time of sampling, the absorbance at 600 nm (A 600 nm) was measured spectrophotometrically. Shown above is the average A 600 nm of the three trials plotted against time. An O.D. standard curve for S. boydii UI02 NA adapted strain was prepared. The cell titer in which the relationship between log 10 CFU/ml and absorbance at 600 nm (A 81

PAGE 94

82 600 nm) showed linearity (R 2 = 0.9764; Figure B-2) was approximately 9.17 x 10 6 to 4.90 x 10 7 CFU/ml. R2 = 0.97640.10.20.30.40.50.66.97.17.37.57.77Log Counts (CFU/ml)A (600nm .9 ) Figure A-2. Standard curve of O.D. (600 nm) of S. boydii UI02 nalidixic acid adapted strain compared to log plate count. Each of three 100 ml TSB (200 ppm NA) microcosms were inoculated with 10 l aliquots of an 18-hour S. boydii UI02 culture. The microcosms were incubated (37C, 30 rpm) and sampled at appropriate time intervals. At each time of sampling, appropriate serial dilutions in PBS were pour-plated using TSA (100 ppm NA) and incubated overnight (37C). Shown above is the average absorbance at 600 nm (A 600 nm) of the three trials (from Fig 4-1) plotted against the log 10 CFU/ml for the data points which form a linear relationship. S. sonnei UI05 NA adapted strain The growth curve prepared for S. sonnei UI05 wild strain demonstrates stationary phase was reached in approximately 10 hours (Figure B-3), with an initial lag phase of approximately 7 hours. An O.D. standard curve for S. sonnei UI05 wild strain was prepared. The cell titer in which the relationship between log 10 CFU/ml and absorbance at 600 nm (A 600 nm) showed linearity (R 2 = 0.9821; Figure B-4) was approximately 4.13 x 10 7 to 6.43 x 10 8 CFU/ml.

PAGE 95

83 0.0000.2000.4000.6000.8001.0001.20056789101112Time (hours)A (600nm ) Figure A-3. Growth curve: S. sonnei UI05 nalidixic acid adapted strain. Each of three 100 ml TSB (200 ppm NA) microcosms were inoculated with 10 l aliquots of an 18-hour S. sonnei UI05 culture. The microcosms were incubated (37C, 30 rpm) and sampled at appropriate time intervals. At each time of sampling, the absorbance at 600 nm (A 600 nm) was measured spectrophotometrically. Shown above is the average A 600 nm of the three trials plotted against time. R2 = 0.98210.10.30.50.70.91.17.27.47.67.888.28.4Log Count (CFU/ml)A (600nm ) Figure A-4. Standard curve of O.D. (600 nm) of S. sonnei UI05 nalidixic acid adapted strain compared to log plate count. Each of three 100 ml TSB (200 ppm NA) microcosms were inoculated with 10 l aliquots of an 18-hour S. sonnei UI05 culture. The microcosms were incubated (37C, 30 rpm) and sampled at appropriate time intervals. At each time of sampling, appropriate serial dilutions in PBS were pour-plated using TSA (100 ppm NA) and incubated overnight (37C). Shown above is the average absorbance at 600 nm (A 600 nm) of the three trials (from Fig 4-1) plotted against the log 10 CFU/ml for the data points which form a linear relationship.

PAGE 96

LIST OF REFERENCES [Anonymous]. 2002. About-Shigella.com. www.about-shigella.com Accessed 2002 Nov 21. [APHA] American Public Health Association. 2001. Compendium of Methods for the Microbiological Examination of Foods. 4 th Edition. Chapter 38:Shigella. Applied Biosystems. 2003. ABI Prism 7000 online information. http://www.appliedbiosystems.com/products/productdetail.cfm?prod_id=641 Accessed 2003 May 23. Bagamboula, C., M. Uyttendaele and J. Debevere. 2002. Acid Tolerance of Shigella sonnei and Shigella flexneri. J. Appl. Microbiol. 93:479-486. Bernardini, M.L., J. Mounier, H. dHauteville, M. Coquis-Rondon and P.J. Sansonetti. 1989. Indentification of icsA, a Plasmid Locus of Shigella flexneri that governs Bacterial Intraand Intercellular Spread Through Interaction with F-actin. Proc. Natl. Acad. Sci. U S A. 86:3867-3871. Beuchat, L.R., L.J. Harris, T.E. Ward and T.M. Kajs. 2001. Development of a Proposed Standard Method for Assessing the Efficacy of Fresh Produce Sanitizers. J. Food Prot. 64(8):1103-1109. Bhat, P. and D. Rajan. 1975. Comparative Evaluation of Desoxycholate Citrate Medium and Xylose Lysine Desoxycholate Medium in the Isolation of Shigellae. Am. J. Clin. Pathol. 64:399-403. Brown, J.E., P. Echeverria and A.A. Lindberg. 1991. Digalactosyl-Containing Glycolipids as Cell Surface Receptors for Shiga Toxin of Shigella dysenteriae 1 and Related Cytotoxins of Escherichia coli. Rev. Infect. Dis. 13(Suppl 4):S298-303. [CDC] Centers for Disease Control and Prevention. 2002a. Shigella: Annual Summary, 2001. www.cdc.gov/ncidod/dbmd/phlisdata/shigella.htm Accessed 2002 Oct 11. [CDC] Centers for Disease Control and Prevention. 2002b. Preliminary FoodNet Data on the Incidence of Foodborne Illnesses Selected Sites, United States, 2001. Morbidity and Mortality Weekly Report. April 19,2002. 51(15):325-329. http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5115a3.htm Accessed 2002 Oct 11. 84

PAGE 97

85 [CDC] Centers for Disease Control and Prevention. 2003. Shigellosis. Technical Information. http://www.cdc.gov/ncidod/dbmd/diseaseinfo/shigellosis_t.htm Accessed 2003 May 22. Cepheid. 2003. Smart Cycler online information. http://cepheid.com/pages/smart_cycler.html Accessed 2003 May 23. Christensen, G.D., W.A. Simpson, A.L. Bisno and E.H. Beachey. 1982. Adherence of Slime-Producing Strains of Staphylococcus epidermidis to Smooth Surfaces. Infect. Immun. 37:318-326. Christensen, G.D., W.A. Simpson, J.J. Younger, L.M. Baddour, F.F. Barrett, D.M. Melton and E.H. Beachey. 1985. Adherence of Coagulase-Negative Staphylococci to Plastic Tissue Culture Plates: a Quantitative Model for the Adherence of Staphylococci to Medical Devices. J. Clin. Microbiol. 22: 996-1006. Clerc, P., A. Ryter, J. Mounier and P.J. Sansonetti. 1987. Plasmid-mediated Early Killing of Eukaryotic Cells by Shigella flexneri as Studied by Infection of J774 Macrophages. Infect. Immun. 55:521-527. Donohue-Rolfe, A., D.W.K. Acheson and G.T. Keusch. 1991. Shiga Toxin: Purification, Structure, and Function. Rev. Infect. Dis. 13(Suppl 4):S293-297. Echeverria, P., O. Sethabutr and C. Pitarangsi. 1991. Microbiology and Diagnosis of Infections with Shigella and Enteroinvasive Escherichia coli. Rev. Infect. Dis. 13(Suppl 4):S220-5. Egile, C., H. dHauteville, C. Parsot and P.J. Sansonetti. 1997. SopA, the Outer Membrane Protease Responsible for Polar Localization of IcsA in Shigella flexneri. Mol. Microbiol. 23:1063-1073. [ERS] Economic Research Service, U.S. Department of Agriculture. 2003. Briefing Room: Tomatoes. http://www.ers.usda.gov/Briefing/Tomatoes/background.htm Accessed 2003 July 10. [FDA] Food and Drug Administration. 1998. Bacteriological Analytical Manual. 8 th Edition. AOAC International, Arlington, VA. [FDA] Food and Drug Administration. 2001a. Survey of Domestic Fresh Produce: Interim Results. http://www.cfsan.fda.gov/~dms/prodsur9.html Accessed 2002 Nov 21. [FDA] Food and Drug Administration. 2001b. FDA Survey of Imported Fresh Produce FY 1999 Field Assignment. http://www.cfsan.fda.gov/~dms/prodsur6.html Accessed 2002 Nov 21.

PAGE 98

86 [FDA] Food and Drug Administration. 2001c. FDA Survey of Imported Fresh Produce: Imported Produce Assignment FY 2001. http://www.cfsan.fda.gov/~dms/prodsur7.html Accessed 2002 Nov 21. [FDA] Food and Drug Administration. 2001d. Analysis and Evaluation of Preventive Control Measures for the Control and Reduction/Elimination of Microbial Hazards on Fresh and Fresh-Cut Produce. http://www.cfsan.fda.gov/~comm/ift3-1.html Accessed 2002 Dec 5. Fontaine, A., J. Arondale and P.J. Sansonetti. 1988. Role of Shiga Toxin in the Pathogenesis of Bacillary Dysentery Studied Using a ToxMutant of Shigella dysenteriae 1. Infect. Immun. 56:3099-3109. [FTC] Florida Tomato Committee. 2003. Tomato 101: Health Information and Research. http://www.floridatomatoes.org/education.htm Accessed 2003 July 10. Galanakis, E., M. Tzoufi, M. Charisi, S. Levidiotou and Z. Papadopoulou. 2002. Rate of Seizures in Children with Shigellosis. Acta Paediatrica. 91(1):1001-2. Goldberg, M.B., O. Barzu, C. Parsot and P.J. Sansonetti. 1993. Unipolar Localization and ATPase Activity of IcsA, a Shigella flexneri Protein Involved in Intracellular Movement. J. Bacteriol. 175:2189-2196. Hale, T. 1991. Genetic Basis of Virulence in Shigella Species. Microbiol. Rev. 55:206-224. dHauteville, H., R. Dufourcq Lagelouse, F. Nato and P.J. Sansonetti. 1996. Lack of Cleavage of IcsA in Shigella flexneri Causes Aberrant Movement and Allows Demonstration of a Cross-Reactive Eukaryotic Protein. Infect. Immun. 64:511-517. Headley, V., M. Hong, M. Galko and S.M. Payne. 1997. Expression of Aerobactin Genes by Shigella flexneri During Extracellular and Intracellular Growth. Infect. Immun. 65:818-821. Ingersoll, M, E.A. Groisman and A. Zychlinsky. 2002. Pathogenicity Islands of Shigella. Curr. Topics Microbiol. Immunol. 264(1):49-65 Institute of Food Technologists. 2003. Expert Report. Emerging Microbiological Food Safety Issues. Implications for the 21 st Century. Islam, D. and A.A. Lindberg. 1992. Detection of Shigella dysenteriae Tyoe 1 and Shigella flexneri in Feces by Immunomagnetic Isolation and Polymerase Chain Reaction. J. Clin. Microbiol. 30(11):2801-2806. Jacobson, A.P., M.L. Johnson, T.S. Hammack and W.H. Andrews. 2002. Evaluation of the Bacteriological Analytical Manual (BAM) Culture Method for the Detection of Shigella sonnei in Selected Types of Produce. Poster presented at 2002 FDA Science Forum. Washington, D.C.

PAGE 99

87 Jin, Q., Z. Yuan, J. Xu, Y. Wang, Y. Shen, W. Lu, J. Wang, H. Liu, J. Yang, F. Yang, X. Zhang, J. Zhang, G. Yang, H. Wu, D. Qu, J. Dong, L. Sun, Y. Xue, A. Zhao, Y. Gao, J. Zhu, B. Kan, K. Ding, S. Chen, H. Cheng, Z. Yao, H. Bingkun, R. Chen, D. Ma, B. Qiang, Y. Wen, Y. Hou and J. Yu. 2002. Genome Sequence of Shigella flexneri 2a: Insights into Pathogenicity Through Comparison with Genomes of Escherichia coli K12 and O157. Nucleic Acids Res. 30(20):4432-4441. June, G.A., P.S. Sherrod, R.M. Amaguana, W.A. Andrews and T.S. Hammack. 1993. Evaluation of the Bacteriological Analytical Manual Culture Method for the Recovery of Shigella sonnei from Selected Foods. J. AOAC Int. 76(6):1240-1248. Karaman, ., F. ahin, M. Gllce, H. t, M. engl and A. Adgzel. 2003. Antimicrobial Activity of Aqueous and Methanol Extracts of Juniperus oxycedrus L. J. Ethnopharm. 85:231-235. Kaufman, P.R., C.R. Handy, E.W. McLaughlin, K. Park and G.M. Green. 2000. Understanding the Dynamics of Produce Markets: Consumption and Consolidation Grow. USDA, Economic Research Service. Agriculture Information Bulletin No. 758. http://www.ers.usda.gov/publications/aib758/aib758.pdf Accessed 2002 Dec 5. Keusch, G.T., M. Jacewicz, M. Mobassaleh and A. Donohue-Rolfe. 1991. Shiga Toxin: Intestinal Cell Receptors and Pathophysiology of Enterotoxic Effects. Rev. Infect. Dis. 13(Suppl 4):S304-310. Lampel, K.A., P.A. Orlandi and L. Kornegay. 2000. Improved Template Preparation for PCR-Based Assays for the Detection of Food-Borne Bacterial Pathogens. Appl. Environ. Microbiol. 66(10):4539-4542. Lawlor, K.M. and S.M. Payne. 1984. Aerobactin Genes in Shigella spp. J. Bacteriol. 160:266-272. Lett, M.-C., C. Sasakawa, N. Okada, T. Sakai, S. Makino, M. Yamada, K. Komatsu and M. Yoshikawa. 1989. virG, a Plasmid-Coded Virulence Gene of Shigella flexneri: Identification of the virG Protein and Determination of the Complete Coding Sequence. J. Bacteriol. 171:353-359. Lew, J.F., D.L. Swerdlow, M.E. Dance, P.M. Griffin, C.A. Bopp, M.J. Gillenwater, T. Mercantante and R.I. Glass. 1991. An Outbreak of Shigellosis Aboard a Cruise Ship Caused by a Multiple-Antibiotic-Resistant Strain of Shigella flexneri. American J. Epidemiol. Aug 15;134(4):413-420. Lindqvist, R. 1999. Detection of Shigella spp. in Food with a Nested PCR Method Sensitivity and Performance Compared with a Conventional Culture Method. J. Appl. Microbiol. 86:971-978. Loisel, T.P., R. Boujemaa, D. Pantaloni and M.-F. Carlier. 1999. Reconstitution of Actin-Based Motility of Listeria and Shigella Using Pure Proteins. Nature. 401:613-616.

PAGE 100

88 Long, S.M., G.K. Adak, S.J. OBrien and I.A. Gillespie. 2002. General Outbreaks of Infectious Intestinal Disease Linked with Salad Vegetables and Fruit, England and Wales, 1992-2000. Common Dis. Public Health. 5(2):101-105. Makino, S., C. Sasakawa, T. Kamata and M. Yoshikawa. 1986. A Genetic Determinate Required for Continuous Reinfection of Adjacent Cells on a Large Plasmid of Shigella flexneri 2a. Cell. 46:551-555. Mehlman, I.J., A. Romero and B.A. Wentz. 1985. Improved Enrichment for Recovery of Shigella sonnei from Foods. J. AOAC. 68:552-555. Miki, H., K. Miura and T. Takenawa. 1996. N-WASP, a Novel Actin-Depolymerizing Protein, Regulates the Cortical Cytoskeletal Rearrangement in a PIP2-Dependent Manner Downstream of Tyrosine Kinases. EMBO J. 15:5326-5335. Millipore. 2003. Sample Preparation Methods for DNA Analysis. Online information. http://www.millipore.com/catalogue.nsf/docs/C7489?open&lang=de Accessed 2003 May 23. [MMWR] Morbidity and Mortality Weekly Report. 1999. Outbreaks of Shigella sonnei Infection Associated with Eating Fresh Parsley -United States and Canada, July Agust 1998. April 16, 1999 / 48(14);285-289. [MMWR] Morbidity and Mortality Weekly Report. 2000. Public Health Dispatch: Outbreak of Shigella sonnei Infections Associated with Eating a Nationally Distributed Dip -California, Oregon, and Washington, January 2000. January 28, 2000 / 49(03);60-61. Monack, D.M. and J.A. Theriot. 2001. Actin-Based Motility is Sufficient for Bacterial Membrane Protrusion Formation and Host Cell Uptake. Cell. Microbiol. 3(9):633-647. Muriana, P.M. 2002. Shigella Pathogenesis and Genetic Basis of Virulence. Online Lecture. http://www.okstate.edu/OSU_Ag/fapc/fsw/shigella/shigpm.htm Accessed 2003 June 18. Olsen, J.E., S. Aabo, W. Hill, S. Notermans, K. Wernars, P.E. Granum, T. Popovic, H.N. Rasmussen and Olsvik. 1995. Probes and Polymerase Chain Reaction for Detection of Food-Borne Bacterial Pathogens. Int. J. Food Microbiol. 28:1-78. Orlandi, P.A. and K.A. Lampel. 2000. Extraction-Free, Filter Based Template Preparation for Rapid and Sensitive PCR Detection of Pathogenic Parasitic Protozoa. J. Clin. Microbiol. 38(6):2271-2277. Parsot, C. and P.J. Sansonetti. 1996. Invasion and the Pathogenesis of Shigella Infections. Curr Topics Microbiol. Immunol. 209:25-42.

PAGE 101

89 Payne, S.M. 1988. Iron and Virulence in the Fampily Enterobacteriaceae. CRC Crit. Rev. Microbiol. 16:81-111. Payne, S.M., D.W. Nielsen, S.S. Peixotto and K.M. Lawler. 1983. Expression of Hydroxamate and Phenolate Siderophores by Shigella flexneri. J. Bacteriol. 155:949-955. Perry, R.D. and C.L. San Clemente. 1979. Siderophore Synthesis in Klebsiella pneumoniae and Shigella sonnei During Iron Deficiency. J. Bacteriol. 140:1129-1132. Rafii, F. and P. Lunsford. 1997. Survival and Detection of Shigella flexneri in Vegetables and Commercially Prepared Salads. J. AOAC Int. 80(6):1191-1197. Robbins, J.R., D.M. Monack, S.J. McCallum, A. Vegas, E. Pham, M.B. Goldberg and J.A. Theriot. 2001. The Making of a Gradient: IcsA (VirG) Polarity in Shigella flexneri. Mol. Microbiol. 41:861-872. Roche Applied Science. 2003. LightCycler online information. http://www.roche-applied-science.com/lightcycler-online/ Accessed 2003 May 23. Sandlin, R.C. and A.T. Maurelli. 1999. Establishment of Unipolar Localization of IcsA in Shigella flexneri 2a is not Dependent on Virulence Plasmid Determinants. Infect. Immun. 67:350-356. Sandlin, R.C., K.A. Lampel, S.P. Keasler, M.B. Goldberg, A.L. Stolzer and A.T. Maurelli. 1995. Avirulence of Rough Mutants of Shigella flexneri: Requirement of O Antigen for Correct Unipolar Localization of IcsA in the Bacterial Outer Membrane. Infect. Immun. 63:229-237. Sandlin, R.C., M.B. Goldberg and A.T. Maurelli. 1996. Effect of O Side-Chain Length and Composition on the Virulence of Shigella flexneri 2a. Mol. Microbiol. 22:63-73. Sansonetti, P.J. 1991. Genetic and Molecular Basis of Epithelial Cell Invasion by Shigella Species. Rev. Infect. Dis. 13(Suppl 4):S285-292. Sansonetti, P.J. and J. Mounier. 1987. Metabolic Events Mediating Early Killing of Host Cells Infected by Shigella flexneri. Microbial Path. 3:53-61. Sansonetti, P.J., A. Ryter, P. Clerc, A.T. Maurelli and J. Mounier. 1986. Multiplication of Shigella flexneri within HeLa Cells: Lysis of the Phagocytic Vacoule and Plasmid-Mediated Contact Hemolysis. Infect. Immun. 1:461-469. Sargent, S.A. 1998. Handling Florida Vegetables Tomato. Vegetable Crops Fact Sheet. SS-VEC-928. Univ. of Florida, Gainesville, FL 32611.

PAGE 102

90 Sethabutr, O., M. Venkatesan, G.S. Murphy, B. Eampokalap, C.W. Hoge and P. Echeverria. 1993. Detection of Shigellae and Enteroinvasive Escerichia coli by Amplification of the Invasion Plasmid Antigen H DNA Sequence in Patients with Dysentery. J. Infect. Dis. 167:458-461. Sethabutr, O., M. Venkatesan, S. Yam, L.W. Pang, B.L. Smoak, W.K. Sang, P. Echeverria, D.N. Taylor and D.W. Isenbarger. 2000. Detection of PCR Products of the ipaH Gene from Shigella and Enteroinvasive Escherichia coli by Enzyme Linked Immunosorbent Assay. Diagn. Microbiol. Infect. Dis. 37:11-16. Shibata, T., F. Takeshima, F. Chen, F.W. Alt and S.B. Snapper. 2002. Cdc42 Facilitates Invasion but Not the Actin-Based Motility of Shigella. Curr. Biol. February 19; 12:341-345. Smith, J.L. 1987. Shigella as a Foodborne Pathogen. J. Food Prot. 50:788-801. Steinhauer, J., R. Agha, T. Pham, A.W. Varga and M.B. Goldberg. 1999. The Unipolar Shigella Surface Protein IcsA is Targeted Directly to the Bacterial Old Pole: IcsP Cleavage of IcsA Occurs Over the Entire Bacterial Surface. Mol. Microbiol. 17:945-951. Stendahl, O.I., J.H. Hartwig, E.A. Brotschi and T.P. Stossel. 1980. Distribution of Actin-Binding Protein and Myosin in Macrophages During Spreading and Phagocytosis. J. Cell. Biol. 84:215-224. Stepanovi, S., D. Vukovi, I. Daki, B. Savi and M. vabi-Vlahovi. 2000. A Modified Microtiter-Plate Test for Quantification of Staphylococcal Biofilm Formation. J. Microbiol. Meth. 40:175-179. Suzuki, T. and C. Sasakawa. 2001. Molecular Basis of the Intracellular Spreading of Shigella. Infect. Immun. 69(10):5959-5966. Suzuki T., M.-C. Lett and C. Sasakawa. 1995. Extracellular Transport of VirG Protein in Shigella. J. Biol. Chem. 270:30874-30880. Suzuki T., S. Saga and C. Sasakawa. 1996. Functional Analysis of Shigella VirG Domains Essential for Interaction with Vinculin and Actin-Based Motility. J. Biol. Chem. 271:21878-21885. Tauxe, R., H. Kruse, C. Hedberg, M. Potter, J. Madden and K. Wachsmuth. 1997. Microbial Hazards and Emerging Issues Associated with Fresh Produce: a Preliminary Report to the National Advisory Committee on Microbiological Criteria for Foods. J. Food Prot. 60:1400-1408. Taylor, W.I. and D. Schelhart. 1969. Isolation of Shigellae VII. Comparison of Gram-Negative Broth with Rappaports Enrichment Broth. Appl. Microbiol. 18(3):393-395.

PAGE 103

91 Theron, J., D. Morar, M. Du Preez, V.S. Brzel and S.N. Venter. 2001. A Sensitive Seminested PCR Method for the Detection of Shigella in Spiked Environmental Water Samples. Wat. Res. 35(4):869-874. Tollison, S.B. and M.G. Johnson. 1985. Sensitivity to Bile Salts of Shigella flexneri Sublethally Heat Stressed in Buffer or Broth. Appl. Environ. Microbiol. 50:337-341. Uyttendaele, M., C.F. Bagamboula, E. De Smet, S. Van Wilder and J. Debevere. 2000. Evaluation of Culture Media for Enrichment and Isolation of Shigella sonnei and S. flexneri. Int. J. Food Microbiol. 70:255-265. Vantarakis, A., G. Komninou, D. Venieri and M. Papapetropoulou. 2000. Development of a Multiplex PCR Detection of Salmonella spp. and Shigella spp. in Mussels. Lett. Appl. Microbiol. 31:105-109. Villalobo, E. and A. Torres. 1998. PCR for the Detection of Shigella spp. in Mayonnaise. Appl. Environ. Microbiol. 64(4):1242-1245. Vokes, S.A., S.A. Reeves, A.G. Torres and S.M. Payne. 1999. The Aerobactin Iron Transport System Genes in Shigella flexneri are Present within a Pathogenicity Island. Mol. Microbiol. 33(1):63-73. Wu F.M., M.P. Doyle, L.R. Beuchat, J.G. Wells, E.D. Mintz and B. Swaminathan. 2000. Fate of Shigella sonnei on Parsley and Methods of Disinfection. J. Food Prot. 63:568-572. Yavzori, M., D. Cohen and N. Orr. 2002. Prevalence of the Genes for Shigella Enterotoxins 1 and 2 Among Clinical Isolates of Shigella in Isreal. Epidemiol. Infect. 128:533-535. Zaika, L. 2001. The Effect of Temperature and Low pH on Survival of Shigella flexneri in Broth. J. Food Prot. 64(8):1162-1165. Zaika, L. 2002a. The Effect of NaCl on Survival of Shigella flexneri in Broth as Affected by Temperature and pH. J. Food Prot. 65(5):774-779. Zaika, L. 2002b. Effect of Organic Acids and Temperature on Survival of Shigella flexneri in Broth at pH 4. J. Food Prot. 65(9):1417-1421. Zychlinsky, A., C. Fitting, J.M. Cavaillon and P.J. Sansonetti. 1994. Interleukin 1 is Released by Murine Macrophages During Apoptosis Induced by Shigella flexneri. J. Clin. Invest. 94:1328-1332.

PAGE 104

BIOGRAPHICAL SKETCH Benjamin Ray Warren was born in Brandon, Fl on June 29, 1975. In 1998, he received his Bachelor of Science from the University of Florida in food science. After graduation he took the position of Food Safety and Product Development Manager with Bloods Hammock Groves, Inc., a grower/shipper/processor of fresh Florida citrus. In 2001, he returned to the University of Florida where he began work towards his masters degree in food science, specializing in food microbiology. The author worked part-time at Deibel Laboratories of Illinois, Inc. during his masters research. Upon completion of his masters degree, the author will remain at the University of Florida where he will begin his doctoral research and pursue a research career in food microbiology. 92


Permanent Link: http://ufdc.ufl.edu/UFE0001186/00001

Material Information

Title: Comparison of conventional culture methods and the polymerase chain reaction for the detection of Shigella spp. on tomato surfaces
Physical Description: Mixed Material
Creator: Warren, Benjamin Ray ( Author, Primary )
Publication Date: 2003
Copyright Date: 2003

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: UFE0001186:00001

Permanent Link: http://ufdc.ufl.edu/UFE0001186/00001

Material Information

Title: Comparison of conventional culture methods and the polymerase chain reaction for the detection of Shigella spp. on tomato surfaces
Physical Description: Mixed Material
Creator: Warren, Benjamin Ray ( Author, Primary )
Publication Date: 2003
Copyright Date: 2003

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: UFE0001186:00001


This item has the following downloads:


Full Text












COMPARISON OF CONVENTIONAL CULTURE METHODS AND THE
POLYMERASE CHAIN REACTION FOR THE DETECTION OF \/nge//l spp. ON
TOMATO SURFACES

















By

BENJAMIN RAY WARREN


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2003

































Copyright 2003

by

Benjamin Ray Warren

































To Nikki, for your undying love and support; to my parents, for never losing faith in me;
and to all my friends along the way, for without all of you this would not have been
possible.















ACKNOWLEDGMENTS

I would like to thank my committee co-chairs, Dr. Mickey Parish and Dr. Keith

Schneider, for all of their support and guidance during the past two years. I would like to

especially thank Dr. Parish for seeing my potential and inspiring me to return to the

University of Florida to pursue this master's degree and a future Ph.D., and Dr. Schneider

for always having an open door and encouraging my questions and ideas.

I would like to thank my parents, Dennis and Linda Warren, and my fiancee,

Nicole Sanson, for their never-ending love and support. I would further like to thank the

members of my graduate committee, Dr. Douglas Archer, Dr. Renee Goodrich, and Dr.

Steven Sargent, for all of their assistance with this project.

For all of the laughs and good times, I would also like to thank my friends and

colleagues Norm Nehmatallah, April Elston, Raina Allen, and Tomas Ballesteros; good

luck to all.

Statistical assistance was provided by University of Florida, IFAS Statistics, with

special thanks to Jamie Jarabek, MSTAT. Assistance in acquiring tomatoes was provided

by the laboratory of Dr. Jerry Bartz, Plant Pathology, especially Mike Mahovic. Finally, I

would like to thank Dr. Keith Lampel, CFSAN, FDA, for sharing information and

methods which made this project possible.

This project was funded by the USDA-CSREES IFAFS Grant number 00-52102-

9637.
















TABLE OF CONTENTS
page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES .................................................... ............ .............. viii

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

ABSTRACT .............. .................. .......... .............. xi

CHAPTER

1 IN TR O D U C TIO N ................. .................................. .... ........ .................

2 LITER A TU RE R EV IEW ............................................................. ....................... 4

General Tomato Background...................... ... .. .. ..................... 4
Use of Tomatoes as a Model for Detection of .\/ige//t spp. from Fruits and
V vegetables ....................................................... ................... ......... ..... 5
G general C characteristics of h/ige//t ....................................................... ................ .....
Recent Outbreaks Involving ge// spp. ..................................... ............ 7
Prevalence of.\/i/ge/l t spp. on Produce .................... ..... .... ....................
Survival Characteristics of ./ge/ t ..................................... ...............9
Intracellular Activity of.\/'igel//t spp ......................... ...... ....... .................. 11
Invasion of E pithelial C ells ........................ ............................................... 1
G aining entry into epithelial cells ..................................... ............... ..12
Intracellular m ultiplication ...................................... ............. .............. 12
Intra- and intercellular spreading (actin-based motility).............................13
E arly host cell apoptosis........................................ ........................... 17
Toxins Produced by g / spp. ....................................................... 17
Shiga toxin .......................................... .............................. ......... 17
\/nhge//At enterotoxins 1 and 2 ..................................... .. ...............18
Traditional Microbiological Media for the Isolation/Detection of \l/.ge//at...............18
U.S. Food and Drug Administration (1998), Bacteriological Analytical
M anu al ............................... .... .............. ....... ........ .................. 2 1
Compendium of Methods for the Microbiological Examination of Foods .........22
Nucleic Acid-Based Detection of.\/ige//t spp. in Foods................ ..................23
Colony H ybridization A ssays........................................ .......................... 23
PCR: The Basics ................... ....... ..... ...... ...... ....... .......... ....24
Bacterial DNA Template Preparation for PCR ................................................24
N e sted P C R ..................................................... ................ 2 5


v









PCR for the D election of /l gel/h sp. ............................... ......................... 26
Real-time PCR: The Future for the Detection of .\/ge//A spp. in Food .............28


3 METHODS AND MATERIALS.............................. ..............29

Prelim inary Trials ............. .................... ...... ................................ .............. 29
Preparation of Microbiological Media ................. ............. ............... 29
E nrichm ent m edia ............................................... ............................. 29
S olid m edia .................... ........ .................... 2 9
Acquisition and Maintenance of .\/ge//At cultures............................................30
Adaptation of Cultures to Rifampicin ...................................... ............... 31
Preparation of Optical Density Standard Curve ...............................................32
DNA Extraction of Stock \ligel//A, Cultures............................................ 33
Crude DNA Extraction of Stock Non-,'\/lgel//, Cultures.....................................34
Acquisition and Maintenance of PCR Primers for the Detection of ,/Nlge//a. ..... 35
Specificity of Prim ers ...................... .. ................................... ........ ........ ......36
Analysis of PCR Product by Gel Electrophoresis.............................................36
Inoculated Studies................. ...................... ......... 37
Acquisition of Tomato................................ ............... ............... 37
Inoculum P reparation ...................................................... .................. 38
Inoculation of Tomatoes and Subsequent Recovery .......................................38
Experim mental D design ................................................................ ............... 39
Confirmation of Typical Colonies on Tri-Plates..............................................41
Assembly of Tandem Filter Funnels ........................... ......................41
Nested PCR Amplification of ipaH gene for Detection of .\/lge//ll ...................44
Recording of Data and Statistical Evaluation........................ ............... 45


4 R E S U L T S .............................................................................4 6

Prelim inary T rials ................. ......... ............................ ...... .... ............... 46
Growth Curves and Optical Density Standard Curves.................. ................46
S. boydii U I02 w ild strain...................................... ........................ 46
S. sonnei U 05 w ild strain ............................. .................................. 47
S. sonnei 9290 rifampicin adapted strain ..............................................49
Prim er Specificity .................. .......................... .. ....... ................. 50
Inoculated Studies......................................................... .... .. ........ .. ........... .. 52
Detection of \/'igell// spp. by Conventional Culture Methods............................52
Detection of \/'igell// spp. by Conventional Culture Methods with Rifampicin
Supplem ented Enrichm ent..................................................... ............... 54
Lowest Detection Levels of Conventional Culture Methods ...........................58
Detection of .\ /ge//t spp. by FTA Filtration /Nested PCR ...........................60
Lowest Detection Levels of the FTA Filtration/ Nested PCR Method .............62











5 DISCUSSION AND CONCLUSION ............................................. ............... 64

P relim in ary Stu dies............................................. .. .. ......... ......... ..... ........ .... 64
Growth Characteristics of S. boydii UI02 and S. sonnei UI05 ............................65
Evaluation of Primers Specific for .\ligel// spp................................................67
Inoculation Studies ................................. ......... ................... 68
Evaluation of Enrichment Protocols....................................................68
Evaluation of Plating M edia.......................................................................... 69
Analysis of Lowest Detection Levels of Conventional Culture Methods...........72
Sources of Variation Among Trials................................................ ................ 73
Comparison of Conventional Culture Methods and FTA Filtration/
N ested P C R ................................... .... ............ ..... .............. .....................7 5
Analysis of Lowest Detection Levels of FTA Filtration/ Nested PCR .............76
Optimization of the FTA Filtration/Nested PCR Assay..............................77
Predictive Value of Testing for .\ge//At spp................................................. 79
C o n clu sio n s..................................................... ................ 7 9


APPENDIX

GROWTH CHARACTERISTICS OF NALIDIXIC ACID ADAPTED STRAINS.........81

S. boydii U I02 N A adapted strain ........................................................................... 81
S. sonnei UIO5 NA adapted strain ................................................................. 82


L IST O F R E FE R E N C E S ...................................... .................................. ..................... .. 84

BIO GRAPH ICAL SK ETCH .................................................. ............................... 92
















LIST OF TABLES


Table p

3-1. Stock ,/ige l//, cultures. ....................................... ................... .. ...... 34

3-2. Stock non-. g t cultures ............................................................ ............... 35

3-3. Primers for the detection of.\hi/ge//At spp...................................... ............... 36

3-4. Temperature programs for PCR primers. ...................................... ............... 37

4-1. Lowest detection levels (LDLs) of conventional culture methods................................59

4-2. Lowest detection levels in which S. boydii UI02 or S. sonnei UI05 was isolated in
100% of replicates (LDL100s) of conventional culture methods ...........................60

4-3. Lowest detection levels and lowest detection levels in which S. boydii UI02 or S.
sonnei UI05 was isolated in 100% of replicates of FTA filtration/ nested PCR....63

5-1. Doubling times associated with exponential growth phase of investigated strains
of / / spp. ........................................................................................ .......66
















LIST OF FIGURES


Figure pge

3-1. Spot inoculation of tom atoes. ............. ................ ............... .................................. 39

3-2. Stomacher bag with inoculated tomato. ........................................ ............... 40

3-2. A ssem bly of tandem filter funnels....................................... ........................... 42

3-3. Vacuum flask apparatus with tandem filter funnels .............................................43

4-1. Growth curve: S. boydii UI02 wild strain........................................................47

4-2. Optical density standard curve for S. boydii UI02 wild strain..............................48

4-3. Growth curve: S. sonnei UI05 wild strain. ....................................................48

4-4. Standard curve of O.D. (600 nm) of S. sonnei UI05 wild strain compared to log
p late cou nt. ........................................................ ................ 4 9

4-5. Growth curve: S. sonnei 9290 rifampicin adapted strain............... ... ...............50

4-6. Standard curve of O.D. (600 nm) of S. sonnei 9290 rifampicin adapted strain
com pared to log plate count. ............................................ ............................ 51

4-7. Recovery of S. sonnei UI05 by conventional culture methods.............................. 53

4-8. Recovery of S. boydii UI02 by conventional culture methods with rifampicin
supplem ented enrichm ent .......................................................................... ... ...... 55

4-9. Recovery of S. sonnei UI05 by conventional culture methods with rifampicin
supplem ented enrichm ent .......................................................................... ... ...... 57

4-10. Representative gels from the first step nested PCR....................... ............... 61

4-11. Representative gel from the second step nested PCR.............................................62

4-12. Detection of S. boydii UI02 and S. sonnei UI05 by FTA filtration/ nested PCR. ..63

5-1. Differentiation of background microflora by isolation media............................... 71

5-2. FTA punches in 0.5 ml microcentrifuge tubes. ............. ........... .....................78









A-1. Growth curve: S. boydii UI02 nalidixic acid adapted strain.................................81

A-2. Standard curve of O.D. (600 nm) of S. boydii UI02 nalidixic acid adapted
strain com pared to log plate count. ........................................ ....... ............... 82

A-3. Growth curve: S. sonnei UI05 nalidixic acid adapted strain ...............................83

A-4. Standard curve of O.D. (600 nm) of S. sonnei UI05 nalidixic acid adapted
strain com pared to log plate count. ........................................ ....... ............... 83















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

COMPARISON OF CONVENTIONAL CULTURE METHODS AND THE
POLYMERASE CHAIN REACTION FOR THE DETECTION OF hige/,t spp. ON
TOMATO SURFACES

By

Benjamin Ray Warren

August 2003


Chair: Mickey E. Parish
Cochair: Keith R. Schneider
Department: Food Science and Human Nutrition

Isolation of,\hige//t spp. from food is very difficult due to the lack of appropriate

selective media and the fastidious nature of Shigellae. Nucleic acid-based detection

methods such as the polymerase chain reaction (PCR) have recently been developed for

the detection of .h/i/ge/t spp. with greater specificity and sensitivity than conventional

culture methods. In this study, artificially inoculated S. boydii UI02 or S. sonnei UI05

was recovered from tomato surfaces using a phosphate buffer rinse and vigorous

shaking/hand manipulation. Detection of inocula was evaluated by enrichment protocols

of the U.S. Food and Drug Administration's (1998) Bacteriological Analytical Manual

(FDA BAM), the Compendium of Methods for the Microbiological Examination of Food

(CMMEF), enrichment in Enterobacteriaceae Enrichment (EE) broth supplemented with

1.0 [g/ml novobiocin and incubated at 420C), and FTA filtration/ nested PCR.

Conventional culture enrichments were repeated using enrichments supplemented with









50g/ml rifampicin (rif+) to exclude natural tomato microflora and rifampicin-adapted

inocula. Additionally, enrichments were plated on .,1hge/l t Plating Medium (SPM),

Salmonella-.\/i/ge/h/a agar (SSA) and MacConkey agar (MAC) in order to compare

isolation rates of S. boydii UI02 and S. sonnei UI05 among the three plating media.

The lowest detection levels (LDLs) of enrichment procedures in the presence of

natural tomato microflora were >5.3 x 105 CFU/tomato (all three methods) for S. boydii

UI02; and 1.9 x 101 (FDA BAM), 1.5 x 103 (CMMEF), and 1.1 x 101 CFU/tomato (EE

broth) for S. sonnei UI05. There were no significant differences (a = 0.05) between the

FDA BAM and the CMMEF for the isolation ofS. boydii UI02, and no significant

differences (a = 0.05) among any of the enrichment methods for the isolation of S. sonnei

UI05. LDLs from enrichment procedures where background microflora was excluded

were 6.3 x 100 CFU/tomato (FDA BAM rif+ and CMMEF rif+) and >5.3 x 105

CFU/tomato (EE broth rif+) for S. boydii UI02; and 1.9 x 101 CFU/tomato (FDA BAM

rif+ and CMMEF rif+), and 1.1 x 101 CFU/tomato (EE broth rif+) for S. sonnei UI05.

The LDL of the FTA filtration/ nested PCR method was 6.2 x 100 CFU/tomato for S.

boydii UI02 and 7.4 x 100 CFU/tomato for S. sonnei UI05. The FTA filtration/ nested

PCR method was significantly better than enrichment protocols of the CMMEF (P =

0.010) and in EE broth (P < 0.001) for the detection of S. boydii UI02; however it was

not significantly better than the FDA BAM (P = 0.177). The FTA filtration/ nested PCR

method was significantly better than enrichment protocols of all three conventional

culture methods (P < 0.001) for the detection of S. sonnei UI05. EE broth was found to be

inhibitory to S. boydii UI02. Furthermore, there were no significant differences (a = 0.05)

among SPM, SSA and MAC for the isolation of S. boydii UI02 or S. sonnei UI05.














CHAPTER 1
INTRODUCTION

Foodborne illness associated with the consumption of fresh produce has increased

during recent years in the Unites States. Although this increase of illness can be partially

attributed to increased consumption of fresh produce, the increased demand for

minimally processed fruits and vegetables and the growth in global food trade have also

contributed (Tauxe et al., 1997). Between 1987 and 1997, the total fresh produce market

increased from $34.8 billion to $70.8 billion in retail and foodservice sales (Kaufman et

al., 2000). During the same time period, U.S. imports of fruits and vegetables have grown

from $2.0 billion in sales to $4.1 billion (Kaufman et al., 2000). Imported fruits and

vegetables represent an increased potential for foodborne illness, especially if grown

under poor production standards or mishandled during a long distribution cycle (Food

and Drug Administration (FDA), 2001d).

Coinciding with the increase of fresh produce consumption, the incidence of

foodborne shigellosis has also increased. Previously thought to be primarily a water-

borne pathogen, foodborne outbreaks of .\/ige/lt spp. are increasing, most recently

involving fresh parsley, iceburg lettuce, and a bean salad containing parsley and cilantro.

Of the pathogens under surveillance by the U.S. Centers for Disease Control and

Prevention's (CDC) Emerging Infections Program, Foodborne Diseases Active

Surveillance Network (FoodNet), ,\/Vige/l spp. accounted for 18.2% of laboratory

confirmed cases during the year 2000. \/nhgel//t spp. are the leading causative agents of









foodbome illness among children age 1 to 4 years at 29.1 cases per 100,000 (CDC,

2002b), with a national incidence rate of 3.8 cases per 100,000 (CDC, 2002a).

Produce related illnesses caused by .\ligel/h spp. can be reduced with the use of

rapid detection methods and the proper sampling/testing of imported and domestic

produce. Historically, conventional culture methods such as the U.S. FDA's (1998)

Bacteriological Analytical Manual (BAM) and the .\nge//At culture method published in

the Compendium of Methods for the Microbiological Examination of Foods have been

employed to detect .\nge//At spp. in food products. These methods make use of selective

media and biochemical tests, which require several days to complete. Furthermore, the

selective media available are not able to eliminate closely related organisms such as

Enterobacter spp., Klebsiella spp., and Citrobacter spp. which tend to out-compete

.\ngel//t spp. More modem microbiological methods, such as DNA hydridization and the

polymerase chain reaction (PCR), have focused on the detection of DNA segments

unique to .\/nge//t spp. and enteroinvasive E. coli. These assays are usually completed in

one or two working days, thereby allowing a more rapid determination of contaminated

product over conventional culture methods. The early detection and identification of a

contaminated product, especially during an outbreak, can significantly limit the number

of illnesses. Furthermore, the elimination of background microflora is not necessary in

DNA-based assays, allowing for the detection of low populations of.\//lge/lt spp. among

high levels of potentially competitive background.

The primary problem associated with PCR methods for the detection of.\/ige//a

spp. and other human pathogens is that no colony is isolated; therefore no further

characterization can be performed. To overcome this problem, FDA is investigating a









new method for the detection of.\//nge/lt spp. in foods that incorporates both

conventional culture methods and PCR techniques (Dr. Keith Lampel, FDA, personal

communication). In the proposed method, a conventional culture method involving an

enrichment step followed by plating with selective media is initiated simultaneously with

a nested PCR method. The nested PCR method involves two successive PCR reactions in

which diluted product from the first reaction is used as template for the second reaction.

Template preparation procedures for the nested PCR include a double filtration (size

exclusion/FTA) technique from which FTA filter punches are taken and used as

template for the first PCR reaction. According to the proposed method, enrichment

procedures to isolate \/ngel//t spp. would only be carried forth if positive amplification

were observed in the first PCR reaction. The proposed method would thereby incorporate

the rapid detection capabilities of PCR with the ability to isolate a colony of conventional

plating techniques.

The objectives of this study were as follows:

1. To compare conventional culture methods and the FTA filtration/nested PCR
method for the detection of.\/l/ge/lt spp. on tomato surfaces.

2. To compare enrichment in Enterobacteriaceae Enrichment broth to enrichment by
the FDA BAM and CMMEF methods.

3. To compare i.\igell/, Plating Medium to MacConkey and Salmonella-I.\lgell/ t agar
for the isolation of.\h/ge//At spp. on tomato surfaces.














CHAPTER 2
LITERATURE REVIEW

General Tomato Background

Tomatoes, although commonly classified as a vegetable, are actually a fruit since

they are the ripened ovary of the tomato plant. Tomatoes are grown for one of two

distinct markets: the fresh market or the processing market. The fresh and processing

tomato markets can be distinguished by four general characteristics. First, tomato

varieties grown for the processing market tend to have higher percentages soluble solids

in order to efficiently make products like tomato paste (Economic Research Service, U.S.

Department of Agriculture (ERS), 2003). Second, most tomatoes grown for processing

are produced under contract between the grower and processing firms (ERS, 2003).

Third, fresh-market tomatoes are all hand picked while processing tomatoes are machine

harvested (ERS, 2003). Finally, fresh market tomato prices are higher and more variable

due to larger production costs and greater market uncertainty (ERS, 2003). Most of the

fresh tomatoes produced in the United States are grown in Florida (- 40%) and California

(- 27%) (Sargent, 1998). Fresh market tomatoes are available year round because imports

from Mexico supplement winter decreases in domestic tomato production (ERS, 2003).

In 1999, the Americans ate 4.8 billion pounds of fresh tomatoes, or 17.8 pounds per

person (ERS, 2003).

Tomatoes provide a rich source of vitamins, minerals, carotenoids, and other

phytochemicals (Florida Tomato Committee (FTC), 2003). One medium, fresh tomato

(5.2 oz) provides 40% of the U.S. recommended daily allowance of vitamin C and 20%









of the vitamin A (ERS, 2003). Lycopene, an antioxidant found almost exclusively in

tomatoes, is under current investigation as a reducer of risk for several types of cancer

(FTC, 2003). Additionally, current research is also investigating a link between lycopene-

rich diets and reduced incidence of heart disease (FTC, 2003).

Use of Tomatoes as a Model for Detection of .\/Ne/lt spp. from Fruits and Vegetables

When choosing the tomato as a model for recovery of.\/ige//At spp. from fruits and

vegetables, the surface characteristics must be considered. The waxy surface of tomatoes

is very smooth, unlike the corky surfaces of cantaloupes or potatoes, and without the

invaginations found on oranges. Such surface irregularities make the recovery of

inouclum difficult and more inconsistent. Furthermore, tomatoes do not contain

epidermal or peridermal pores, such as stoma or lenticels, respectively, which allow gas

exchange and also may allow internalization of bacterial pathogens. For these reasons,

bacteria used to artificially inoculate tomatoes may be recovered with greater efficiency

than from other fruits and vegetables. For purposes of this study, the inoculation of waxy

surfaces of tomatoes was performed to reduce the number of variables associated with

other surfaces, such as bacterial internalization and inefficient surface removal, thus

yielding a better comparison of methods.

General Characteristics of \/hle, t//

.h/gell/i, the causative agent of shigellosis or "bacillary dysentery," was first

discovered over 100 years ago by a Japanese scientist Kiyoshi Shiga (Anonymous, 2002).

Shigellae are members of the family Enterobacteriaceae, and are nearly genetically

identical to Escherichia coli (E. coli) and are closely related to Salmonella and

Citrobacter spp. (American Public Health Association (APHA), 2001). Shigellae are

characterized as Gram-negative, facultatively anaerobic, non-sporulating, non-motile









rods. Typically, species of.\hige//At do not ferment lactose, are lysine-decarboxylase

negative, are acetate and mucate negative, and do not produce gas from glucose, although

several exceptions exist (Echeverria et al., 1991).

There are four serogroups of h\/ngel,/ S. dysenteriae (serogroup A) serotype 1 -

15, S. flexneri (serogroup B) serotype 1 8 (9 subtypes), S. boydii (serogroup C) serotype

1 19, and S. sonnei (serogroup D) serotype 1. Serogroups of ,/hlgel,// can be

differentiated by their biochemical traits and antigenic properties (CDC, 2003); however

they can also be differentiated by their epidemiology (Ingersoll et al., 2002). S.

dysenteriae is the serogroup primarily associated with epidemics (Ingersoll et al., 2002);

S. dysenteriae type 1 is associated with the highest case fatality rate of all .\nge//lt

serogroups at 5-15% (CDC, 2003). S. flexneri is the predominate group found in areas of

endemic infection, while S. sonnei is the group implicated in source outbreaks in

developed countries (Hale, 1991). S. boydii has been associated with food imported from

Central and South America and is rarely isolated in North America.

,h\ge/lli, although classically thought of as a waterborne pathogen, has been

involved in an increasing number of food-borne outbreaks (Smith, 1987). Food products

associated with ,\lge/llt outbreaks are most commonly subjected to hand processing or

preparation, limited heat treatment, or served/delivered raw to the consumer (Wu et al.,

2000). Examples of food products from which .\ligel // spp. have been isolated include

potato salad, ground beef, bean dip, raw oysters, fish, and raw vegetables.

The infective dose for .\ligel// spp. is reported to be very low; ingestion of 1.0 x

101 cells of S. dysenteriae is sufficient for infection, while the other serogroups require

ingestion of 1.0 x 102 to 1.0 x 104 cells (cited in Muriana, 2002). The low infective dose









associated with .\/nge//l spp. results in common spread of the disease through person to

person contact. Typical symptoms of infection include bloody diarrhea, abdominal pain,

fever, and malaise. Seizures in children with shigellosis have been reported in 5.4% of

cases (Galanakis, 2002). Late complications of S. dysenteriae serotype 1 infections can

include hemolytic uremic syndrome (HUS), while S. flexneri infections can result in

development of Reiter's syndrome, especially in persons with the genetic marker HLA-

B27 (CDC, 2003). Reiter's syndrome is characterized by joint pain, eye irritation, and

painful urination (CDC, 2003).

Recent Outbreaks Involving .\',ilgel/l spp.

In recent years, there has been an increasing number of .\/ge//At outbreaks

involving produce and prepared foods. In 1989, German potato salad was implicated in

an outbreak of a multi-antibiotic resistant strain of S. flexneri aboard a cruise ship (Lew et

al., 1991). In 1994, iceburg lettuce was implicated in an outbreak of S. sonnei which

affected people from six Northern European countries (Long et al., 2002). In 1998,

uncooked, chopped, curly parsley was implicated in a multi-state outbreak of S. sonnei

(Morbidity and Mortality Weekly Report (MMWR), 1999). The source of this outbreak

was traced back to a Mexican farm. In 1999, bean salad which contained parsley and

cilantro was implicated in a Chicago area foodborne outbreak of S. boydii serotype 18. In

2000, a nationally distributed five layer bean dip was implicated in another multi-state

outbreak of S. sonnei (MMWR, 2000).

Prevalence of.\llti.e/,t spp. on Produce

In response to President Clinton's National Food Safety Initiative (January 1997)

and Produce & Imported Foods Safety Initiative (October 1997), the FDA has begun

investigating the presence of human pathogens on produce. In March 1999, FDA initiated









a field assignment entitled "FDA Survey of Imported Fresh Produce" to collect data on

the incidence and extent of pathogen contamination on selected imported produce (FDA,

2001b). The survey analyzed broccoli, cantaloupe, celery, cilantro, culantro, loose-leaf

lettuce, parsley, scallions (green onions), strawberries, and tomatoes for E. coli 0157:H7

and Salmonella. All commodities except cilantro, culantro, loose-leaf lettuce, and

strawberries were analyzed for .\/Vlgel// For those commodities tested, contamination

with .\Nltgel/A spp. was observed at the following rates: 0.9% (9/1003) of all commodities

tested, 2.0% (3/151) cantaloupe samples, 2.4% (2/84) celery samples, 0.9% (1/116)

lettuce samples, 1.2% (1/84) parsley samples, and 1.1% (2/180) scallion samples.

In May 2000, FDA initiated its "Survey of Domestic Fresh Produce" to focus on

high-volume domestic produce that is generally consumed raw (FDA, 2001a). This

survey included the following commodities: cantaloupe, celery, cilantro, green onions,

loose-leaf lettuce, parsley, strawberries, and tomatoes. All commodities were to be

analyzed for the presence of E. coli 0157:H7 and Salmonella, while all except

strawberries were to be tested for ./Nige/l/ At time of publication, only interim results

from analysis of 767 of the required 1000 samples had been released. As with the 1999

Imported Fresh Produce Survey, 0.9% (6/646) of the commodities tested were

contaminated with .lnge//li, 0.9% (1/115) cantaloupe samples, 1.6% (1/62) cilantro

samples, 4.1% (3/73) green onion samples, and 1.6% (1/64) parsley samples.

In January 2001, FDA announced another survey entitled "FDA Survey of

Imported Fresh Produce: Imported Produce Assignment FY 2001." The focus of the

study was to examine further the presence of E. coli 0157:H7, Salmonella, and .\l/ge/,i,

on cilantro, culantro, cantaloupe, and tomatoes based on high rates of pathogens from









previous surveys (FDA, 2001c). At time of publication, results from this survey were not

available.

Survival Characteristics of .\/ ige/l

The ability of.\/i/gie// to survive is dependent, in part, upon pH, temperature, and

salt concentration of its environment. In a study using S. flexneri, Zaika (2001)

demonstrated the effects of temperature and pH on survival. A strain of S. flexneri was

cultured in brain heart infusion broth (BHI) and subjected to various incubation

temperatures (4, 12, 19, 28, and 370C) and pH conditions (pH 2, 3, 4, 5). In general,

survival was enhanced by lower temperatures and increased pH for all experiments.

Results of this study indicated that S. flexneri has acid resistance and suggest that foods

of pH 5 or lower stored at or below room temperature may permit survival of the

organism over long periods of time in sufficient numbers to cause illness (Zaika, 2001).

Zaika (2002a) also investigated survival characteristics of S. flexneri as affected by NaC1.

S. flexneri was able to tolerate and survive in levels of NaCl (1 6%) commonly found in

food items such as pickled vegetables, caviar, pickled herring, dry cured ham, and certain

cheeses for two weeks to two months (Zaika, 2002a). Survival of S. flexneri in the

presence of organic acids (citric, malic, and tartaric acid), commonly found in fruits and

vegetables, and fermentation acids (acetic and lactic acid), commonly used as

preservatives, was studied (Zaika, 2002b). S. flexneri was cultured with each acid (plus

an HC1 control) at 0.04 M in BHI adjusted to pH 4, and incubated at various temperatures

(4, 19, 28, and 370C). As seen in other experiments, survival increased as temperature

decreased (Zaika, 2002b). At 40C, S. flexneri survived in the presence of all the acids

tested for > 55 days (Zaika, 2002b).









In water alone, .\l/ngel/ spp. can survive with little decline in population levels.

Rafii and Lunsford (1997) inoculated S. flexneri into distilled water. The initial count of

2.8 x 108 CFU/ml was only decreased to 9.2 x 107 CFU/ml after storage at 40C for 26

days. The high survival rate of S. flexneri in water supports the historical association of

shigellosis outbreaks with water sources.

./lgel/t spp. can survive for extended periods of time on raw vegetable surfaces.

Wu et al. (2000) studied survival of S. sonnei on whole and chopped parsley leaves.

When held at 210C, S. sonnei was able to grow on chopped parsley at a rate similar to

that which occurs in nutritious liquid medium (Wu et al., 2000). At 40C, populations

declined on both chopped and whole parsley throughout the 14 day storage period,

however the pathogen survived regardless of initial population (Wu et al., 2000). Rafii

and Lunsford (1997) studied the survival of S. flexneri on raw cabbage, onion, and green

pepper held at 40C. Although the population decreased, S. flexneri survived storage at

4C for 12 days (at which time sampling was terminated due to spoilage) on the onion

and green pepper at levels of 2.10 x 105 and 2.2 x 104 CFU/g, respectively (Rafii and

Lunsford, 1997). S. flexneri continued to survive on the cabbage after 26 days at 1.13 x

103 CFU/g. These studies demonstrate how .'/ngel//t spp. can survive on refrigerated raw

vegetables for periods of time that exceed the expected shelf life (Wu et al., 2000).

Several studies have demonstrated the ability of.\/ /ge//At spp. to survive in low pH

foods at low temperature storage. Bagamboula et al. (2002) demonstrated the ability of S.

sonnei and S. flexneri to survive in apple juice (pH 3.3-3.4) and tomato juice (pH 3.9-4.1)

held at 70C for 14 days. No reduction was noted in the tomato juice, while only 1.2 to 3.1

logo1 reduction was observed in the apple juice during the 14 day study. Rafii and









Lunsford (1997) observed the ability of S. flexneri to survive in carrot salad (pH 2.7 -

2.9), potato salad (pH 3.3 4.4), coleslaw (pH 4.1 4.2), and crab salad (pH 4.4 4.5)

held at 40C. Sampling was terminated at day 11 for the carrot and the potato salad, at

which time S. flexneri counts decreased from an initial 4.3 x 106 to 4.2 x 102 CFU/g and

from 1.32 x 106 to 8.5 x 102 CFU/g, respectively. Sampling of the coleslaw and the crab

salads ceased due to product spoilage on days 13 and 20, respectively, while S. flexneri

counts were 2.16 x 104 and 2.4 x 105, respectively. These studies indicate that inoculated

Shigellae were not rapidly killed by normal microflora, low pH (Rafii and Lunsford,

1997), or low temperature.

Intracellular Activity of.\h/,ne//h spp.

Invasion of Epithelial Cells

Invasion of epithelial cells by Shigellae involves four steps: entry into epithelial

cells, intracellular multiplication, intra- and intercellular spreading, and killing of the host

cell (Sansonetti, 1991). The invasion process is controlled by a 220-kDa plasmid. The

plasmid contains invasion plasmid antigen (ipa) genes which encode four highly

immunogenic polypeptides; IpaA, IpaB, IpaC, and IpaD. Studies in which Tn5 insertions

affecting the expression of the ipa genes reveal that expression of ipaB, ipaC, and ipaD is

strongly associated with entry, while ipaA is not (Sansonetti, 1991).

Invasion is also mediated by the virF gene, which is located on the virulence

plasmid, and the virR gene, which is located on the chromosome. The virF gene encodes

a 30-kDa protein that positively regulates the expression of the ipa genes and a plasmid

gene icsA (also known as virG), which encodes intra- and intercellular spread.

Environmental factors which affect the expression of virF are not known. The virR gene

is a repressor of the plasmid invasion genes in a temperature-dependant manner









(Sansonetti, 1991). When Shigellae are grown at 300C they do not express any of the Ipa

polypeptides and are therefore non-invasive, however, Shigellae grown at 370C are fully

invasive and all plasmid polypeptides are encoded (Sansonetti, 1991).

Gaining entry into epithelial cells

Once ingested, Shigellae move through the gastrointestinal tract to the colon, where

they translocate the epithelial barrier via M cells that overlay the solitary lymphoid

nodules (Suzuki and Sasakawa, 2001). Upon reaching the underlying M cells, .\/nge/lt

infects the macrophages and induces cell death (Suzuki and Sasakawa, 2001). Infected

macrophages release interleukin-13, which elicits a strong inflammatory response

(Zychlinsky et al., 1994). Once released from the macrophage, \l/igel // will enter the

epithelial cells, also called enterocytes, which predominately line the colon via membrane

ruffling and macropinocytosis. Epithelial cells produce inflammatory cytokines in

response to bacterial invasion, therefore increasing inflammation of the colon (Suzuki

and Sasakawa, 2001).

Macropinocytosis involves the cell extending its membrane as formations known as

pseudopodia, which will engulf large volumes and close around them forming a vacuole.

In order for macropinocytosis to occur, actin must be polymerized and myosin, an actin

binding protein, must be present (Stendahl et al., 1980). Common stimuli that induce

macropinocytosis include cytokines and bacterial antigens.

Intracellular multiplication

Shigellae immediately disrupt phagocytic vacuoles allowing entry into the host cell

cytoplasm. Once in the cytoplasm, Shigellae multiply rapidly. Sansonetti et al. (1986)









observed the generation time for S. flexneri in HeLa cells to be approximately 40

minutes.

Sustaining efficient intracellular growth requires the acquisition of host cell

nutrients. Although production of Shiga toxin facilitates the availability of host cell

nutrients, no relation between its production and intracellular growth rates can be

observed (Fontaine et al., 1988). Likewise, no relation between Shiga-like toxin

production and intracellular growth rate can be observed (Sansonetti et al., 1986; Clerc et

al., 1987).

Since little free iron exists within mammalian host cells, .l/igell/t spp. must also

express high-affinity iron acquisition systems. In order to obtain iron, ./lgel//t spp.

synthesize and transport the siderophores aerobactin and enterobactin (Vokes et al., 1999)

and utilize a receptor/transport system in which iron is obtained from heme. Siderophores

(aerobactin and enterobactin) are low molecular weight iron binding compounds that

remove iron from host proteins. Enterobactin is produced by some but not all \/lge//it

spp. (Perry and San Clemente, 1979; Payne, 1984) while aerobactin is synthesized by S.

flexneri and S. boydii (Lawlor and Payne, 1984) and some S. sonnei (Payne, 1988).

Headley et al. (1997) demonstrated that aerobactin systems, although active in

extracellular environments, are not expressed intracellularly. This suggests that

siderophore-independent iron acquisition systems can provide essential iron during

intracellular multiplication (Headley et al., 1997).

Intra- and intercellular spreading (actin-based motility)

The capacity for .\l/nge// to spread intracellularly and infect adjacent cells is

critical in the infection process (Sansonetti, 1991). Intra- and intercellular spreading is

controlled by the icsA (virG) gene located on the virulence plasmid. The icsA gene









encodes the protein IcsA, which enables actin-based motility (Bemardini et al., 1989) and

intercellular spread (Makino et al., 1986). IcsA is a surface-exposed outer membrane

protein consisting of three distinctive domains: a 52 amino acid N-terminal signal

sequence, a 706 amino acid a-domain, and a 344 amino acid C-terminal P-core (Goldberg

et al., 1993; Lett et al., 1989; Suzuki et al., 1995). The a-domain is the exposed portion

and the P-core is embedded in the outer membrane (Suzuki et al., 1995).

IcsA is distributed at one pole of the outer membrane surface. This asymmetrical

distribution allows the polar formation of actin tails, and thus polar movement of .\l/ge//ll

within host cell cytoplasm. The polar localization of IcsA is primarily affected by two

events: (i) the rate of diffusion of outer membrane IcsA (Sandlin et al., 1995; 1996;

Sandlin and Maurelli, 1999; Robbins et al., 2001) and (ii) the specific cleavage of IcsA

by the protease IcsP (SopA) (d'Hauteville et al., 1996; Egile et al., 1997; Steinhauer et

al., 1999). Rate of diffusion of outer membrane IcsA is directly affected by the O side

chains of the membrane lipopolysaccharide (LPS). Sandlin et al. (1996) demonstrated

this relationship with a S. flexneri LPS mutant, BS520 which does not make any 0-

antigen. As compared with a wild-type strain of S. flexneri, which polymerized actin at

one pole, the LPS mutant strain polymerized actin in a non-polar fashion. Expression of

LPS that does not have any O side chains causes an even distribution of IcsA over the

entire outer membrane (Sandlin et al., 1995; 1996; Monack and Theriot, 2001).

Composition of the C-terminal one-third of the IcsA a domain is also required for polar

localization as well as polar movement of S. flexneri (Suzuki et al., 1996). A S. flexneri

mutant, in which a segment of this section was deleted, was unable to polymerize actin in

a polar fashion or move unidirectionally (Suzuki et al., 1996).









The icsP gene encodes the outer membrane protease IcsP (also called SopA) which

cleaves laterally diffused IcsA, thus promoting polar localization (Egile et al., 1997). An

E. coli K-12 strain, engineered to express the icsA gene, was shown to diffuse IcsA along

its outer membrane (Monack and Theriot, 2001). When the same E. coli K-12 strain was

engineered to express the icsP gene with the icsA gene, the number of bacteria which

polymerized actin at one pole increased (Monack and Theriot, 2001).

The N-terminal two-thirds of the IcsA a domain is essential for mediating actin

assembly in .\l/lge/lt host cells (Suzuki and Sasakawa, 2001). This portion of IcsA

contains six glycine rich repeats which interact with the Wiskott-Aldrich syndrome

protein (N-WASP). N-WASP is composed of distinct domains: PH, a pleckstrin

homology domain; IQ, a calmodulin binding domain; GBD, a GTPase binding domain

that binds Cdc42; PRR, a proline-rich region; V, a G-actin-binding veroprolin homology

domain; C, cofilin homology domain; A, a C-terminal acidic amino acid segment (cited

in Suzuki and Sasakawa, 2001; Miki et al., 1996). The VCA domain of N-WASP

activates and interacts with the Arp2/3 complex. The IcsA--N-WASP--Arp2/3 complex

mediates rapid actin filament growth at the barbed end, including cross-linking between

the elongated actin filaments (Suzuki and Sasakawa, 2001). The resulting network of

actin filaments allows .\/gel//t to gain a propulsive force with which to move in the

cytoplasm of host cells (Suzuki and Sasakawa, 2001).

Cdc42 (an N-WASP activator), profilin (an actin monomer-binding protein), and

cofilin (which depolymerizes actin) also have roles in efficient actin assembly. Cdc42

binds to the GBD domain of N-WASP preventing the intramolecular interaction between

the C-terminal acidic amino acids and the basic amino acids of the GBD, thereby forcing









the N-WASP complex to unfold to its activated form (Suzuki and Sasakawa, 2001).

Cdc42 independent activation of the N-WASP complex by IcsA has been reported for

actin-based motility in .\/ige//i, however efficient entry into cells was reported as Cdc42

dependent (Shibata et al., 2002). The role of Cdc42 in invasion and motility is still

somewhat controversial. Profilin delivers monomeric actin to sites of actin assembly

(Goldberg, 2001). Although profilin has been shown not to be absolutely essential for

actin based motility, it is required for maximum rates of movement (Loisel et al., 1999).

Cofilin generates actin monomers from the filamentous actin. By disassembling

unneeded actin filaments within the tail, cofilin might work to free up actin for

incorporation into newly generated filaments (Goldberg, 2001).

Intercellular spreading is dependent upon an actin-based motility mechanism (Fig.

2-2) (Monack and Theriot, 2001). ./nge//it cells first form a membrane bound protrusion

into an adjacent cell. This protrusion must distend two membranes: one from the donor

cell, and another from the recipient cell (Parsot and Sansonetti, 1996). As the protrusion

pushes further into the recipient cell, it is taken up by the recipient cell resulting in the

bacteria enclosed in a double-membrane vacuole (Monack and Theriot, 2001).

Intercellular spread is completed when .\/ilgle/ rapidly escapes from the double-

membrane vacuole, releasing it into the cytosol of the secondary cell. Monack and

Theriot (2001) observed intercellular spread of an E. coli K-12 strain expressing the icsA

and icsP genes in HeLa cells. As expected, they found the E. coli K-12 strain spread to

adjacent cells and enclosed in a double-membrane vacuole as well as free in the cytosol

of the adjacent cells.









Early host cell apoptosis

The early killing of host cells by .\/Nge//At is mediated by the virulence plasmid.

Sansonetti (1991) demonstrated the inability of non-invasive S. flexneri to kill host cells

whereas the invasive species killed efficiently and rapidly. Non-invasive strains can not

kill host cells, since this requires that the .\llge/lt be intracellular. Early killing of host

cells involves metabolic events which rapidly drop the intracellular concentration of

ATP, increase pyruvate, and arrest lactate production (Sansonetti and Mounier, 1987).

Interestingly, Shiga toxin, a potent cytotoxin produced by S. dysenteriae serotype 1, does

not play a role in the early killing of host cells. Fontaine et al. (1988) constructed a Tox-

mutant strain of S. dysenteriae serotype 1 and found that the mutant killed as efficiently

as the wild-type strain.

Toxins Produced by \/n.lg//At spp.

Shiga toxin

S. dysenteriae type 1 strains produce a potent toxin known as Shiga toxin (STX).

Although the toxin is not necessary to sustain an infection, its expression increases the

severity of disease. Three biologic activities associated with STX are cytotoxicity,

enterotoxicity, and neurotoxicity, while the one known biochemical effect is the

inhibition of protein synthesis (Donohue-Rolfe et al, 1991). STX is considered the

prototype to a family of toxins known as Shiga-like toxins (SLT), which are similar in

structure and function, and share the same receptor sites. Perhaps the most widely known

human pathogen that produces SLT's is E. coli 0157:H7, which has two toxin variants

(SLT I and SLT II).

STX is composed of two polypeptides: an A subunit (32,225 MW) and five B

subunits (7,691 MW, each) (Donohue-Rolfe et al., 1991). The B subunits mediate









binding to cell surface receptors, which have been identified as glycolipids containing

terminal galactose-a(1-4)galactose disaccharides such as galabiosylceramide and

globotriasylceramide (Gb3) (Brown et al., 1991; Keusch et al., 1991). The A subunit,

once inside the host cell cytoplasm, acts enzymatically to cleave the N-glycosidic bond of

adenine at nucleotide position 4324 in the 28S rRNA of the 60S ribosomal unit

(Donohue-Rolfe et al., 1991).

\l/tige'// enterotoxins 1 and 2

Recently, two enterotoxins, shigella enterotoxin 1 (SHET 1) and shigella

enterotoxin 2 (SHET 2), have been characterized and are believed to play a role in the

clinical manifestation of shigellosis (Yavzori et al., 2002). SHET 1, which is

chromosomally encoded, was only prevalent in isolates of S. flexneri 2a. SHET 2,

however, is encoded on the large virulence plasmid, and was detected in all .\/lge//t

isolates tested except several isolates which lost their plasmid.

Traditional Microbiological Media for the Isolation/Detection of.\/,l,//at

Traditional microbiological techniques make use of selective media for the

enrichment/isolation of .\/ige//At spp. Many variants of enrichment and plating media

have been investigated for optimal recovery, often with conflicting results between

laboratories and sample types. Due to the lack of appropriate selective media and the

presence of .\/1/ge//t spp. in relatively low population, background microflora tends to

out-compete ,/Nigel/h spp. when isolation is attempted from a food product. Unless media

can be developed that are both specific and sensitive for ./Vge//At spp. regardless of

potential background microflora, isolation by traditional microbiological methods will

always be suspect.









Early enrichment of .\h/ge//t spp. was attempted using Selenite-F (SF) or

Tetrathionate (TT) broth. These broths were originally designed for the isolation of

salmonellae, but due to the lack of specific enrichment media for Shigellae they were

used as all-purpose enteric enrichment broths (Taylor and Schelhart, 1969). Sodium

selenite, although selective for salmonellae, is toxic to ,.higel/h spp. (and most enterics),

therefore its use in enrichment procedures for .Shge/Alt spp. has been terminated. TT is a

peptone base broth with bile salts and sodium thiosulfate, which inhibit most Gram-

positives and Enterobacteriaceae. Gram-negative (GN) broth is a peptone-based broth

with glucose and mannitol. The concentration of mannitol in GN broth is higher than

glucose to promote mannitol fermentors, like ./gel//At spp. Both TT broth and GN broth

contain bile salts, which can be inhibitory to stressed cultures. Furthermore, GN broth

contains sodium deoxycholate, which has been shown to inhibit heat-stressed Shigellae

(Uyttendaele et al, 2001). Currently, enrichment procedures use a low carbohydrate

medium, .\//ge/lit broth (SB) with addition of novobiocin, for the detection/isolation of

\ligell,// spp. (APHA, 2001; FDA BAM). Acids produced by other Enterobacteriaceae

during the fermentation of carbohydrates have been reported to be toxic to Shigellae

(Mehlman et al., 1985); however, other studies have shown the acid tolerance of.\/1ige//A

spp. to grow at a pH of 4.5 to 4.75 (Bagamboula et al., 2002) and to survive at a pH of

4.0 (Zaika, 2002). Since SB contains very little carbohydrate, the effect of a low pH

environment on the enrichment of .\hge//At spp is limited when SB is used. SB is also less

stringent than TT broth and GN broth since it contains neither bile salts nor sodium

deoxycholate. In a recent study investigating enrichment media for detection of Shigellae,

SB, GN broth, tryptic soy broth, and Enterobacteriaceae Enrichment (EE) broth with the









addition of novobiocin were compared (Uyttendaele et al., 2001). When incubated in GN

broth, ./nlg//At spp. were unable grow to comparable levels as observed in SB and EE

broths, even though EE broth contains bile salts.

In order to increase the specificity of enrichment media, elevated incubation

temperatures and anaerobic atmospheric conditions are recommended by the FDA BAM.

In a competitive inoculation study, cultures incubated in SB and EE broth at 42C

eliminated competitors such as Aeromonas and Erwinia while incubation at 37C did not

(Uyttendaele et al., 2001).

Due to the fastidious nature of Shigellae, multiple plating media of different

selectivity should be used to increase the chances of isolation. The most common low

selectivity media used for plating .,/lge/la spp. is MacConkey Agar (MAC); while Eosin

methylene blue (EMB) or Tergitol-7 (T7) agars could also be used. Since differentiation

is based on lactose fermentation, non-lactose competitors make detection of.\/i/ge//t spp.

on MAC very difficult (Uyttendaele et al., 2001). On MAC, ,'/nlg//At spp. are translucent

and slightly pink, with and without rough edges. .\/lnge// spp. produce colorless colonies

on EMB and bluish colonies on the yellowish-green T7 agar (APHA, 2001). A more

recently developed low selectivity, differential medium is .\llgl//At Plating Medium

(SPM) (RF Laboratories, West Chicago, IL). ./nlge//t spp. are white to clear on SPM.

Intermediate selectivity media useful in isolating .,/lge/la spp. are desoxycholate citrate

agar (DCA) and xylose lysine desoxycholate agar (XLD). .\/nlgl t spp. produce colorless

colonies on both DCA and XLD. Bhat and Rajan (1975) reported XLD superior to DCA

for the isolation of.\/'/ge//t spp. since DCA required a 48 hour incubation to show clear

colony morphology as opposed to overnight incubation for XLD. A problem with XLD is









that D-xylose, which serves as a differentiating agent, is fermented by some strains of S.

boydii while most ./n,/ge/lt spp. do not ferment xylose (APHA, 2001). This can cause

some strains of .\/ige//t to be missed if only XLD is used as a plating media. Highly

selective media for ./ ge//At spp. include Salmonella-.\ llgel/i agar (SSA) and Hektoen

Enteric agar (HEA). A problem associated with SSA and HEA is that they may be too

stringent for some strains of.\,//ge//At spp., especially if the culture is stressed (APHA,

2001; Uyttendaele et al., 2001). ,/Nigel/lt spp. produce colorless, translucent colonies on

SSA and green colonies on HEA.

U.S. Food and Drug Administration (1998) Bacteriological Analytical Manual

The FDA BAM outlines a conventional culture method for the isolation and

detection of .\/'gel//t spp. from food. A 25 g sample is transferred to 225 ml of .\/hlge//l

broth (SB) to which novobiocin (0.5 pg/ml for S. sonnei; 3.0 pg/ml for other .\//ge/lt

spp.) has been added. Samples are held at room temperature for 10 minutes and

periodically shaken. Sample supernatants are transferred to an Erlenmeyer flask and the

pH adjusted to 7.0 + 0.2 with sterile 1 N NaOH or 1 N HC1. Flasks are incubated

anaerobically for 20 hours (44C for S. sonnei; 42C for all other '.\/lgel/l spp.). After

incubation, enrichment culture is used to streak a MAC plate. Confirmation of suspicious

colonies involves tests for motility, H2S, gas formation, lysine decarboxylase, and

fermentation of sucrose or lactose. All isolates showing any of these characteristics are

discarded. Isolates negative for all confirmatory tests are tested for further biochemical

reactions including adonitol, inositol, lactose, potassium cyanide, malonate, citrate,

salicin, and methyl red. Shigellae are negative for all except methyl red. Antisera

agglutination is then used to identify any culture displaying typical ,.\/gel/l

characteristics.









The effectiveness of the FDA (1992) BAM method for .\/1/ge/ll spp. was evaluated

by June et al. (1993). The 1992 FDA BAM procedures were the same as the 1998 FDA

BAM procedures with respect to the isolation and detection of.\,/ige//At spp. Two strains

of S. sonnei, strains 9290 and 25931, were inoculated on potato salad, chicken salad,

cooked shrimp salad, lettuce, raw ground beef, and raw oysters. The lowest number of

unstressed cells of strain 9290 recovered from a 25 g food sample were: 1.0 x 100 for

potato salad, 8.5 x 101 for chicken salad, 8.8 x 101 for cooked shrimp salad, 7.6 x 101 for

lettuce, 8.4 x 101 for raw ground beef, and 8.5 x 102 for raw oysters. The lowest number

of unstressed cells of strain 25931 recovered from a 25 g food sample were: 5.3 x 10-1 for

potato salad, 6.6 x 10-1 for chicken salad, 1.6 x 102 for cooked shrimp salad, 4.2 x 10-1 for

lettuce, 6.9 x 104 for raw ground beef, and 5.4 x 102 for raw oysters. The recovery of 8.4 x

10 CFU/25 g for strain 9290 versus 6.9 x 104 CFU/25 g for strain 25931 from raw

ground beef samples suggests high strain variability, which can complicate recovery

methods. Chilled stressed cells for both strains were recovered with similar results for all

6 foods. Since the infective dose of.\/Nige/lt spp. is has low as 10 cells per person, the

1992 FDA BAM was considered ineffective for the evaluation of raw ground beef and

raw oysters.

Compendium of Methods for the Microbiological Examination of Foods

Another method for the isolation/detection of.\//ge,/ht spp. from foods is described

in the Compendium of Methods for the Microbiological Examination of Foods

(CMMEF) (APHA, 4th Edition, Chapter 38). This method calls for the enrichment of 25

g sample in either 225 ml of SB or GN broth, both with the optional addition of

novobiocin (0.3 tg/ml for S. sonnei; 3.0 tg/ml for all other .\l/tgel// spp.) After holding

at room temperature for 10 minutes, enrichment samples are incubated for 16 to 20 hours









at 37C. The CMMEF suggests that 2 to 3 plates of various selective media be used to

streak the enriched cultures; MAC for low selectivity, XLD for intermediate selectivity,

and HEA for high selectivity. Suspicious colonies are tested for motility, with all non-

motile isolates identified by biochemical and serological tests.

Nucleic Acid-Based Detection of.\h/ge//, spp. in Foods

The essential principle of nucleic acid based detection methods is the specific

formation of double stranded nucleic acid molecules from two complementary, single

stranded molecules under defined physical and chemical conditions (Olsen et al, 1995).

There are basically two types of nucleic acid assays: hybridization assays and the

polymerase chain reaction (PCR).

Colony Hybridization Assays

In colony hybridization assays, a culture sample is spread-plated on appropriate

media. Following incubation, the colonial pattern is transferred to a solid support (usually

a membrane or paper filter) by pressing the support onto the agar surface (FDA, 1998).

Cells are lysed in situ by a combination of high pH and temperature (0.5 M NaOH and/or

steam or microwave irradiation), which also denatures and affixes the DNA to the

support (FDA, 1998). Solid supports are incubated in solution containing labeled probes

(32P- or enzyme-label) to allow the probes to attach to their target DNA. Unbound probe

is removed by washing the probe-target complexes on the support at an appropriate

temperature and salt concentration (FDA, 1998). A signal is then generated using the

label attached to the probe to identify positive colonies. Colony hybridization assays are

useful when further characterization of positive colonies is required (Olsen et al., 1995).









PCR: The Basics

PCR amplifies regions of DNA by annealing specific primers to single stranded

DNA (ssDNA) and rebuilding the double stranded molecule using a polymerase enzyme.

Typical reactions are performed in a mixture of water, dNTPs, PCR buffer, primers, taq

polymerase, and a DNA template. After initial denaturation, PCR subjects reaction

mixtures to approximately 30 cycles of denaturation, annealing, and extension.

Denaturation occurs at 940C and involves the unwinding of double stranded DNA

(dsDNA) to ssDNA. Annealing, typically at 50-70C, is where the single stranded

primers attach to the ssDNA at their specific sites. Extension occurs at 720C which is the

optimal temperature for most taq polymerases. During extension, the polymerase enzyme

interacts with the primer/ssDNA complex and rebuilds the dsDNA molecule using the

dNTPs. As the cycling continues, the number of dsDNA copies doubles with each cycle.

After the final cycle, most PCR reactions include a final extension step which allows the

completion of any incomplete reactions. In theory, PCR can amplify a single copy of

target DNA to over a million copies in a 30 cycle reaction.

Bacterial DNA Template Preparation for PCR

The failure or success of PCR greatly depends on the effectiveness of the DNA

extraction method to provide adequate amounts of purified DNA. Early PCR used

"crude" extracts of DNA obtained by boiling a culture/sample, centrifuging the cell

material, and using the supernatant as DNA template. DNA templates prepared as crude

extracts are often contaminated with high amounts of protein and contain very low

concentrations of target DNA.

The most common way to purify and concentrate DNA samples is to perform a

phenol, phenol-chloroform, or guanidine isothionate extraction (purification) followed by









ethanol precipitation (concentration). Other methods for DNA purification/concentration

include the use of anion exchange resin (DNA affinity) columns or various filtration

techniques. DNA affinity columns require elution by high salt and large fragments of

DNA >20,000 bp can stick to the column (Millipore, 2003). Several filtration techniques

are available for isolating DNA including size exclusion, glass fiber filters, and FTA

filter paper. FTA filtration involves a chemically treated filter which can trap bacterial

cells, lyse the cellular membrane, and bind bacterial DNA. Several PCR methods have

been developed which can amplify DNA directly off the filter paper (Lampel et al., 2000;

Orlandi and Lampel, 2000).

Nested PCR

Nested PCR refers to a two-step PCR technique in which the PCR product from the

first reaction is diluted and used as template in the second PCR reaction. In true nested

PCR, the first reaction uses an external set of primers (P1 and P2) to amplify a target

region in a gene of interest. The second PCR reaction uses an internal primer set, (P3 and

P4) to amplify a region from within the product of the first reaction. Semi-nested PCR is

the same as nested PCR with the exception that instead of using two internal primers in

the second PCR reaction, one internal (P3) and one external primer (P1 or P2) are used.

The advantage of nested or semi-nested PCR is that greater specificity and sensitivity can

be achieved. Protocols for the analysis of ./igel/ht spp. in development at FDA utilize a

nested PCR (Dr. Keith Lampel, personal communication). In the protocol, positive

amplification in the first reaction serves as presumptive detection while positive

amplification in the second reaction serves as confirmation of\,h/ge//At spp.









PCR for the Detection of .\/nge/ht sp.

PCR assays for .\/nge/lt spp. have targeted the invasion associated locus (ial)

(Islam and Lindberg, 1992; Lindqvist, 1999), the virA gene (Vantarakis et al., 2000;

Villalobo and Torres, 1998) or the ipaH gene (Sethabutr et al., 1993; Sethabutr et al.,

2000). All three of these targets detect all four serogroups of .\/ge//t and enteroinvasive

E. coli. The ial and virA gene are located on the virulence plasmid, while the ipaH is

encoded multiple times on the plasmid and on the chromosome (Jin et al., 2002). Since

detection of the ipaH gene is possible in the event of losing the plasmid, it is a very

attractive target for PCR assays. Although few studies have used PCR methods in the

examination of food for .\lnge/lt spp., they have demonstrated higher sensitivity than

conventional culture methods.

Vantarakis et al. (2000) developed a multiplex PCR method to detect both

Salmonella spp. and ,l/nge/la spp. in mussels. Artificially inoculated S. typhimurium and

S. dysenteriae were recovered by homogenizing mussel meat with peptone water. DNA

from an aliquot of the homogenate was purified using a guanidine isothionate method and

concentrated via ethanol precipitation. Amplification of.\h/ge//At spp. and Salmonella

spp. DNA targeted the virA and invA genes, respectively. The multiplex PCR was able to

detect S. dystenteriae at 1.0 x 103 CFU/ml homogenate with no pre-enrichment, and 1.0 x

10 -1.0 x 102 CFU/ml homogenate following 22 hour incubation in buffered peptone

water.

Villalobo and Torres (1998) investigated PCR for the detection of.\/ige//t spp. in

mayonnaise. S. dysenteriae type 1 DNA was isolated from artificially contaminated

mayonnaise samples by homogenizing in buffered peptone water, lysing cells with

detergent, extracting with phenol-chloroform, and precipitating with ethanol.









Amplification targeted the virA gene and was multiplexed with a region of the 16srDNA.

This detection method was able to detect S. dysenteriae type 1 at 1.0 x 102 -1.0 x 103

CFU/ml homogenate.

In a study by Lindqvist (1999), a nested PCR method was compared to a

conventional culture method (NMKL no. 151 1995) for detection of.\h/ge//At spp.

Recovery of DNA from spiked lettuce, shrimp, milk, and blue cheese samples was

accomplished by homogenizing with physiological saline, buoyant density centrifugation

(to separate components based on density), and boiling at 96-980C for 10 minutes.

Amplification targeted internal regions of the ial. Single PCR, using the external primer

sets only, was only able to detect S. flexneri in aqueous solution at 0.5-1.0 x 105 CFU/ml,

however the nested PCR was able to detect 1.0 x 103 CFU/ml. The nested PCR assay in

combination with buoyant density centrifugation was able to detect S. flexneri inoculated

onto all four foods at 1.0 x 101 CFU/g (Lindqvist, 1999).

Theron et al. (2001) investigated a semi-nested PCR method for the detection of

. iige/ll, spp. in spiked environmental water samples. S. flexneri was inoculated into

sterile and non-sterile environmental water samples. Dilutions of the water samples were

made and bacterial cells from each dilution were harvested by centrifugation and

resuspended in GN broth. After incubation at 370C for 6 hours, bacterial cells were

washed twice in distilled water and lysed by heating at 1000C for 10 minutes. Lysate

supernatant was used as DNA template for semi-nested PCR. Amplification targeted the

ipaH gene. The detection limits of the various environmental water samples were 2 x 103

cfu/ml for well water, 1.4 x 101 CFU/ml for lake water, 5.8 x 102 CFU/ml for river water,

6.1 x 102 CFU/ml for treated sewage water, and 1.1 x 101 CFU/ml for tap water.









Variability in results among the water samples was attributed to the presence of humic

substances which serve as PCR inhibitors. Pre-enrichment in GN broth served to dilute

these PCR inhibitors while allowing the S. flexneri to multiply, thereby increasing the

concentration of target DNA.

Real-time PCR: The Future for the Detection of \l/,ge//u spp. in Food

Real-time PCR utilizes a fluorescently labeled oligonucleotide probe or a

nonspecifically binding intercalating dye which allows for detection of generated product

after each cycle of the PCR reaction. Total assay time is greatly reduced as compared to

conventional PCR as there is no need for post-reaction analysis. The ABI Prism 7000

Sequence Detection System (Applied Biosystems, Foster City, CA) runs in a 96-well

format and ABI offers support in primer and probe development (Applied Biosystems,

2003). Furthermore, the Taqman Universal Master Mix can be utilized to reduce

laboratory preparation, contamination risks, and assay to assay variation. The Smart

Cycler II System (Cepheid, Sunnyvale, CA) offers more flexibility with its sixteen

ICORE (Intelligent Cooling/Heating Optical Reaction) modules (Cepheid 2003). The

Smart Cycler's software allows separate experiments with unique cycling protocols to

be carried out and analyzed simultaneously. The LightCycler (Roche Applied Science,

Indianapolis, IN) provides the fastest PCR results by use of its unique capillary sample

tubes format. Air heating of the 32 capillary tubes, as compared to a thermal block in

other units, allows for rapid PCR cycling and a faster overall assay (Roche Applied

Science, 2003). A 30-40 cycle PCR reaction can usually be completed in 20 to 30

minutes with the LightCycler. All real-time PCR systems provide quantitative results

based on assay specific standard curves or the incorporation of internal controls.














CHAPTER 3
METHODS AND MATERIALS

Preliminary Trials

Preparation of Microbiological Media

Wild strain cultures were grown in Tryptic Soy Broth (TSB) (Difco, Sparks, MD)

and maintained on Tryptic Soy Agar (TSA) (Difco) slants. Adapted strains were cultured

and maintained using TSB and TSA supplemented with 80 tg/ml rifampicin (TSB-R80

and TSA-R80, respectively). All dilutions and tomato rinses were performed using

Phosphate Buffered Saline (PBS) prepared using PBS tablets (ICN Biomedicals Inc.,

Aurora, OH).

Enrichment media

\/nge//At broth was prepared according to the U.S. Food and Drug Administration's

(1998) Bacteriological Analytical Manual (FDA BAM) and supplemented with

novobiocin at 3.0 tg/ml (SB3.0), 0.5 tg/ml (SB0.5), or 0.3 tg/ml (SB0.3). EE Broth

(Difco) was prepared as directed by manufacturer's instructions and supplemented with

novobiocin at 1.0 tg/ml (EE1.0). For trials involving antibiotic supplemented enrichment

media, rifampicin was added at 50 tg/ml (SB3.0-R50, SB0.5-R50, SB0.3-R50, and

EE1.0-R50). When necessary, the pH was adjusted using filter sterilized IN NaOH.

Enrichment media were prepared fresh for each experiment.

Solid media

MacConkey Agar (MAC) (Difco), Salmonella-.\lV/gel h Agar (SSA) (Difco),

,/l1/ge/ht Plating Medium (SPM) (RF Laboratories, West Chicago, IL), Triple Sugar Iron









(TSI) (Difco), Lysine Iron Agar (LIA) (Difco) and Motility Medium (MM) (Difco) were

all prepared according to manufacturers instructions. When necessary, the pH was

adjusted using filter sterilized IN NaOH. MAC, SSA, and SPM were each poured into

one compartment of a three-compartment Petri dish (Tri-Plate). TSI and LIA were

prepared as slants and MM was prepared according to the FDA BAM.

Acquisition and Maintenance of \/1g'e//At cultures

Outbreak strains of .\/ige//At sonnei and .higell // boydii were obtained from Dr.

Hans Blaschek's Laboratory at the University of Illinois, Department of Food Science

and Human Nutrition. The .\/igel/h boydii serotype 18 strain, encoded UI02, was isolated

from a person involved in a bean salad outbreak in Chicago in March of 1999. The

,hige/llt sonnei strain, encoded UI05, was isolated from a patient during an outbreak.

Both strains, UI02 and UI05, originated from the State of Illinois Department of Public

Health, Chicago. In addition, a strain of ./h/ge/ht boydii serotype 18 (ATCC 35966)

(encoded UI01), a human isolate of .\//ge//t sonnei (encoded UI03) from an outbreak

involving bean dip, and a human isolate of.h/ige//t sonnei (encoded UI04) from an

outbreak involving cilantro were also obtained. Strains UI03 and UI04 originated from

the Enteric Bacteriology Unit, Microbial Diseases Laboratory, State of California

Department of Health Services.

Additional \,i/ge/ll cultures were obtained to provide additional positive controls

and to facilitate representation of each of the four serogroups. .\ngel//l dysenteriae

serotype 1 (ATCC 9361) and .Nhgelt// sonnei (ATCC 9290) were purchased from the

American Type Culture Collection (ATCC). A strain of .\ hge//Atflexneri was obtained

from Dr. Linda Harris' Laboratory at the University of California, Davis, Department of









Food Science and Technology. This strain originated from Dr. Keith Lampel at the U.S.

Food and Drug Administration.

Upon receipt, each strain was grown in 10 ml TSB at 370C (30 rpm) overnight.

Overnight cultures were plated for isolation onto MAC and incubated overnight at 370C.

Overnight plates were examined for typical growth. One typical colony of each strain was

transferred to a TSA slant and stored at 40C. Another typical colony was transferred per

product instructions to Protect TM Bacterial Preservers (Scientific Device Laboratory,

Inc., Des Plaines, IL) and stored at -760C.

Adaptation of Cultures to Rifampicin

Adaptation of Shigellae to the bactericidal agent rifampicin was accomplished by

challenging cultures in enrichment broth with low doses and increasing the dosage with

each successive 24 hr transfer. A 10,000 ppm (1%) stock solution of rifampicin was

prepared by dissolving 2.0 g rifampicin (Fisher # BP267925, Fisher Scientific, Pittsburg,

PA) in 200 ml deionized water. This stock solution was then filter sterilized and stored in

the dark at room temperature.

Stock cultures were grown overnight in 10 ml TSB (370C, 30 rpm). Overnight

cultures were transferred to 10 ml TSB with the progression of 2.5, 5.0, 10, 25, 40, 60,

and 80 ppm rifampicin (TSB-R2.5, TSB-R5.0, TSB-R10, TSB-R25, TSB-R40, TSB-R60,

and TSB-R80, respectively) and grown overnight (370C, 30 rpm). Once the cultures were

adapted to 80 ppm rifampicin, cultures were grown overnight (370C, 30 rpm) three

consecutive times in TSB-R80 to ensure well adapted populations.

Once adaptation was complete, the final overnight adapted culture was plated for

isolation onto MAC-R80 and incubated overnight at 370C. One typical colony from the









overnight MAC-R80 plate was transferred to a TSA-R80 slant and stored at 40C. Another

typical colony was transferred per product instructions to a Protect TM Bacterial Preserver

and stored at minus 760C.

Preparation of Optical Density Standard Curve

For all standard curve studies, non-adapted cultures were grown using TSB and

TSA. All adapted cultures were grown using TSB-R80 and TSA-R80. Procedures given

below are worded for un-adapted cultures.

A sterile wooden stick was used to transfer stock culture from a TSA or TSA-R80

slant into a 10 ml TSB tube. The tube was incubated overnight (37C, 30 rpm). A 10 pl

aliquot of the first overnight culture was transferred to fresh 10 ml TSB and grown

overnight (370C, 30 rpm). Also from the first overnight culture, 10 [l was used to

inoculate 100 ml TSB in an Erlenmeyer flask. This flask was incubated (370C, 30 rpm)

and observed every 30 minutes to determine the lag time associated with each strain. End

of lag phase was determined by visible growth (cloudiness) in the TSB flask.

A 10 pl aliquot of the second overnight culture was transferred to fresh 10 ml TSB

and grown for 18 hours (37C, 30 rpm). A 10 [L aliquot of this 18 hour culture was then

transferred into each of three Erlenmeyer flasks containing 100 ml TSB to provide three

replicates. Each flask was labeled 1, 2, or 3 and inoculated five minutes apart in

numerical order and incubated (370C, 30 rpm).

After the pre-determined lag phase was complete, sampling began in 30 minute

intervals. At each sampling, 1.25 ml of culture was transferred to a disposable cuvette

and the absorbance was read at 600 nm using a spectrophotometer (Shimadzu Scientific

Instruments, model UV-1201). Additionally, 1 ml of the culture was used to prepare









serial dilutions using PBS. Appropriate dilutions were pour plated with TSA and

incubated 24 to 48 hours at 370C.

TSA plates were analyzed for growth and plates containing colonies in the

countable range (25-250) were counted. Replicate results were averaged yielding one

count of colony forming units (CFU) per ml for each sampling. Replicate results of

absorbance values were also averaged. Using Excel software (Microsoft, Redmon, WA),

a graph of the absorbance versus loglo CFU/ml was prepared. The linear range was

identified and subjected to linear regression analysis to form a standard curve. This

standard curve was then used to estimate CFU/ml by means of absorbance at 600 nm in

TSB.

DNA Extraction of Stock l/ge/Ali Cultures

Stock ./lge//At cultures used in this study are listed in Table 3-1. DNA was

extracted from stock .\/Nige/l cultures via the DNeasy Tissue Kit (Qiagen, Valencia,

CA). Stock cultures were grown overnight in 10 ml TSB (370C, 30 rpm). Overnight

cultures were plated for isolation onto MAC and incubated overnight at 370C. Overnight

plates were examined for typical growth and one typical colony was transferred to 10 ml

TSB and grown overnight (370C, 30 rpm).

A 1 ml aliquot of overnight culture was transferred to a clean, sterile 1.5 ml

microcentrifuge tube and centrifuged for 10 minutes at 7,500 rpm. Supernatant was

discarded and the resulting pellet was re-suspended in 180 [il Buffer ATL (supplied with

DNeasy Tissue Kit). DNA extraction continued from this point from step 2 of the

DNeasy Protocol for Animal Tissues as per product literature. Final DNA elution was

performed twice; once with 200 pl Buffer AE (supplied with DNeasy Tissue Kit), and a









Table 3-1. Stock .ngel//, cultures. DNA from each stock .\/ngel// spp. was extracted
from a 1 ml aliquot of an overnight culture in TSB (37C, 30 rpm) using a
DNeasy Tissue Kit. Extracted DNA was transferred to a clean, sterile 1.5
microcentrifuge tube and stored at minus 200C until use. Extracted DNA was
used to evaluate specificity of each primer set.
DNA Code Culture Origin
01 .lhge/lht boydii serotype 18 ATCC 35966
02 .lgeAllt boydii serotype 18 Outbreak isolate
03 \/lge//lt sonnei Patient isolate
04 \1hlge/l t sonnei Patient isolate
05 iihlgellt sonnei Outbreak isolate
06 hlgel t sonnei ATCC 9290
07 .\lhell/,tflexneri FDA, Dr. Keith Lampel
08 .\/1e/A ll dysenteriae serotype 1 ATCC 9361


second time with 50 [l Buffer AE. This yields a final elution of 250 [l DNA template in

one 1.5 ml microcentrifuge tube. DNA templates for stock .lhige/ll cultures were stored

at minus 200C.

Crude DNA Extraction of Stock Non-,'\ ge/lAlt Cultures

Stock non-.\/hge/lt cultures used as negative controls in this study are listed in

Table 3-2. DNA from non-,'/hge//t stock cultures was extracted to provide negative

control templates. Stock cultures frozen on Protect TM Bacterial Preservers were retrieved

from minus 76C storage and thawed. One bead was aseptically transferred from the

Protect TM Bacterial Preserver into 10 ml TSB and grown overnight (370C, 30 rpm).

Overnight cultures were plated for isolation on an appropriate selective and differential

medium and incubated overnight at 370C. Overnight plates were observed for typical

colony morphologies. One typical colony was transferred to 10 ml TSB and grown

overnight (370C, 30 rpm).

A 1 ml aliquot of the overnight culture was transferred to a clean, sterile 1.5 ml

microcentrifuge tube and centrifuged for 10 minutes at 7,500 rpm. Supernatant was









Table 3-2. Stock non-.\lige/l/ cultures. DNA from each of the non-.\'lge//A spp. cultures
was extracted by boiling a 1 ml aliquot of an overnight culture for 10 minutes,
then centrifuging away any cell wall material. Supernatants were transferred
to clean, sterial 1.5 ml microcentrifuge tubes and used to test the specificity of


each primer set.
Code Culture
Salmonella Agona
Salmonella Gaminara
Salmonella Montevideo
Salmonella Michigan
Salmonella Poona
Salmonella Enteritidis
Salmonella Typhimurium
Escherichia coli 0157:H7
Escherichia coli 0157:H7
Escherichia coli 0157:H7
Escherichia coli 0157:H7
Escherichia coli JM109
Escherichia coli K 12
Escherichia coli (ATCC 25922)
Citrobacter feundii
Klebsiella pneumoniae
Klebsiella ozoanae
Enterobacter cloacae


Origin
Dr. Harris, UC Davis
Dr. Harris, UC Davis
Dr. Harris, UC Davis
Dr. Harris, UC Davis
Dr. Harris, UC Davis
Dr. Rodrick, UF
ABC Research
Deibel Laboratories
Deibel Laboratories
Deibel Laboratories
Deibel Laboratories
Dr. Wright, UF
Dr. Wright, UF
ATCC
Cantaloupe isolate
Soil isolate
Cantaloupe isolate
Cantaloupe isolate


discarded and the resulting pellet was re-suspended in 200 pl double de-ionized,

sterilized water (hereby referred to as PCR water). Samples were then boiled for 10

minutes in a dry bath incubator (Fisher Scientific, IsoTemp 125D). Supernatant (DNA

template) was aseptically transferred to a clean, sterile 1.5 ml microcentrifuge tube and

stored at minus 200C. Pellet was discarded.

Acquisition and Maintenance of PCR Primers for the Detection of.\l/,,i//a

Primers were synthesized by Sigma Genosys (The Woodlands, TX). Upon receipt,

primers were reconstituted with 10% TE Buffer (1.0 mM Tris-C1, 0.1 mM EDTA pH 8.0)

to yield a 100 mM stock solution. The working solution was a 1:10 dilution of the stock

solution with PCR water. Primer sets investigated in this study are shown in Table 3-3.


DNA
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26









Table 3-3. Primers for the detection of .\/hge//At spp. All primers sets below are specific
for ./nge//At spp. and EIEC. Primer set 01-001 amplifies a 620 bp fragment of
the ipaH gene. Primer set 01-002 amplifies a 215 bp fragment of the virA
gene. Primer sets 01-003 and 01-004 are a nested primer set where 01-003 is
the internal primer amplifying a 217 bp region of the invasion associated locus
(ial) and 01-004 is the external primer amplifying a 320 bp fragment of the ial.
Primer set 01-005 serves as an internal primer to set 01-001 and amplifies a
290 bp fragment of the ipaH gene.


Primer Code
01-001F
01-001R
01-002F
01-002R
01-003F
01-003R
01-004F
01-004R
01-005F
01-005R


Primer Sequence
5' gtt cct tga ccg cct ttc cga tac cgt c 3'
5' gcc ggt cag cca ccc tct gag agt ac 3'
5' ctg cat tct ggc aat ctc ttc aca tc 3'
5' tga tga gct aac ttc gta age cct cc 3'
5' ttt tta att aag agt ggg gtt tga 3'
5' gaa cct atg tct acc tta cca gaa gt 3'
5' ctg gta ggt atg gtg agg 3'
5' cca ggc caa caa tta ttt cc 3'
5' cca ctg aga gct gtg agg 3'
5' tgt cac tec cga cac gcc 3'


Source
Sethabutr et al., 2000
Sethabutr et al., 2000
Villalobo and Torres, 1998
Villalobo and Torres, 1998
Lindqvist, 1999
Lindqvist, 1999
Frankel et al., 1990
Frankel et al., 1990
Lampel (unpublished 2002)
Lampel (unpublished 2002)


Specificity of Primers

The specificity of each primer set was tested against DNA from each strain of

\litge//At (Table 3-1) and a battery of non-,'\/ige//a DNA (Table 3-2). Amplification of

target DNA sequences was performed in a 25 .il reaction mixture in clean, sterile 0.2 ml

polypropylene microcentrifuge tubes. The reaction mixture consisted of 15.75 pl PCR

water, 2.5 p l 10x PCR Buffer (100 mM Tris-HC1, 500 mM KC1, 15 mM MgC12, pH 8.3),

2.5 .il dNTP mixture (2.5 mM each), 1.0 [il forward primer, 1.0 [il reverse primer, 0.25 [il

taq polymerase (TaKaRa Taq Hot Start Version, Shiga, Japan), and 2 pl DNA template.

The thermocycler used to perform the PCR reaction was an Eppendorf Mastercycler

gradient (Brinkmann-Eppendorf, Westbury, NY). Temperature programs used to evaluate

each primer set are given in Table 3-4.

Analysis of PCR Product by Gel Electrophoresis

DNA amplicons were analyzed by gel electrophoresis. Gel used in all experiments









Table 3-4. Temperature programs for PCR primers. All temperature programs were
obtained from the sources of the primer sets listed above in Table 3-3.
Primer Initial Denaturation Annealing Extension Number Final Hold
Set Heating of Extension
Cycles
01-001 940C 940C 600C 720C 30 720C 40C
10 min 15 sec 30 sec 45 sec 2 min
01-002 940C 940C 650C 720C 35 720C 40C
10 min 45 sec 30 sec 30 sec 10 min
01-003 940C 940C 600C 720C 30 720C 40C
5 min 30 sec 1 min 30 sec 10 min
01-004 940C 940C 600C 720C 30 720C 40C
5 min 30 sec 1 min 30 sec 10 min
01-005 940C 940C 600C 720C 30 720C 40C
10 min 15 sec 15 sec 15 sec 2 min

was 3% NuSeive GTG Agarose (BioWhittaker Molecular Applications, Rockland,

MA). Electrophoresis was performed on Thermo EC gel trays (model CSSU78115,

Holbrook, NY) powered with a Thermo EC power supply (model EC 105). Gel trays were

leveled using a Thermo EC leveling platform (model CSSLP78). Gels were stained with

ethidium bromide (FisherBiotech, BP1302-10). DNA amplicons and DNA markers

(QX174/Hinf I, Promega, Madison, WI) were visualized by UV transilluminator

(Spectroline TE-312S, Westbury NY). Images of DNA amplicon bands were captured

using a photodocumentation handheld camera (Fisherbiotech, FB-PDC-34) loaded with

Polaroid Type 667 high-speed print film (Fisher Scientific, 04-441-91).

Inoculated Studies

Acquisition of Tomato

Tomatoes of the Florida cultivar 47 were obtained from a nearby packinghouse

(DiMare, Palmetto, FL). Tomatoes were pulled prior to the wash/wax line in order to

retain the normal bacterial population. Upon arrival, tomatoes were stored at 130C until









use. Only fully green or tomatoes with less than 50% red color were used. All tomatoes

with greater than 50% red color were discarded.

Inoculum Preparation

Three days prior to each experiment, stock culture stored on TSA-R80 slants at 4C

was grown (37C, 30 rpm) in a 10 ml tube of TSB-R80. Overnight transfers were

performed using 10 ml tubes of TSB-R80 each day. On the day of the experiment, an 18-

hour culture (late stationary phase) was centrifuged (4,000 x g for 10 minutes) and

washed twice with PBS. The washed culture was serially diluted using 9 ml tubes of

PBS. Appropriate dilutions were pour-plated with TSA-R80 to confirm cell titer.

Inoculation of Tomatoes and Subsequent Recovery

Prior to each experiment, tomatoes were removed from cold storage and allowed to

warm to room temperature (-21C). Large plastic trays were sanitized using reagent

alcohol, 70% v/v (ethanol 63% v/v, methanol 3.5% v/v, isopropanol 3.5% v/v, and

water balance) (LabChem Inc., Pittsburg, PA) prior to each experiment. Tomatoes were

placed on the tray stem scar down. Using an Eppendorf Repeater Micropipette, ten 10 [L

aliquots of the appropriate dilution were spot inoculated around the blossom scar of each

tomato without inoculating directly on the blossom scar (Figure 3-1). S. sonnei was

inoculated onto tomato surfaces at levels of 105, 104, 103, 102, 101, and 100 CFU/tomato.

S. boydii was inoculated onto tomato surfaces at levels of 106, 105, 104, 103, 102, 101, and

100 CFU/tomato. Inoculated tomatoes were allowed to air dry completely prior to

continuing with recovery.




























Figure 3-1. Spot inoculation of tomatoes. Tomatoes were inoculated with ten 10 pl spots
around but not touching the blossom scar. Inocula were allowed to air dry
prior to recovery.

After inocula were completely dry, tomatoes were transferred aseptically to a sterile

stomacher bag containing 100 ml PBS (Figure 3-2). Each bag was sealed using stomacher

bag clips. Inoculum was recovered by methods similar to that of Beuchat et al. (2001),

modified to include vigorous shaking and hand rub/manipulation (15 seconds shake/ 15

seconds rub/ 15 seconds shake).

Experimental Design

Thirty tomatoes were prepared at each inoculation level and rinsed as described

above. Ten tomato rinses were analyzed for S. boydii UI02 and S. sonnei UI05 using

standard enrichment media, ten were analyzed using enrichment media supplemented

with rifampicin, and ten were analyzed using the polymerase chain reaction (PCR). For

standard enrichment, 25 ml of the tomato rinse was transferred to each of three 18 oz

Whirl-Pak bags (Nasco, Modesto, CA) containing 225 ml of appropriate enrichment




























Figure 3-2. Stomacher bag with inoculated tomato. Inoculated tomatoes were aseptically
transferred to a sterile stomacher bag which contained 100 ml PBS. The bag
was sealed using stomacher bag clips and the inoculum was recovered using a
vigorous shaking and hand rub/manipulation method (15 seconds shake/ 15
seconds rub/ 15 seconds shake).

media. Inoculation studies involving S. sonnei utilized enrichments in SB0.3 (44C,

anaerobically) (FDA BAM), SB0.5 (37C) (CMMEF), and EE1.0 (42C) (EE Broth).

Inoculation studies involving S. boydii utilized enrichments in SB3.0 (42C,

anaerobically) (FDA BAM), SB3.0 (37C) (CMMEF), and EE1.0 (42C) (EE Broth).

Anaerobic conditions were generated using the Pack-Anaero anaerobic gas generating

system (Mitsubishi Gas Chemical Company, Inc. (MGC), Japan) and 7.0 Liter Pack-

Rectangular Jars (MGC). For antibiotic supplemented enrichment, SB0.3-R50, SB0.5-

R50, SB3.0-R50, and EE1.0-R50 were used in place of standard enrichment media and

incubated at the same conditions. After 24 hours, each enrichment bag was mixed and

sterile wooden sticks were used to streak the enrichment for isolation on each

compartment of a Tri-Plate. Tri-Plates were incubated overnight (37C).









Confirmation of Typical Colonies on Tri-Plates

Typical .\/lgel//t isolates from all plating media on Tri-Plates were carried through

the following confirmation process. TSI and LIA slants were inoculated with suspect

colonies and incubated overnight (37C). TSI and LIA slants demonstrating typical

reactions for .\ligell// were used to inoculate MM and a 10 ml tube of TSB which were

then incubated overnight (37C). Growth in TSB from samples demonstrating no motility

in MM was streaked for isolation on MAC and incubated overnight (37C). Biochemical

reactions were tested using the BBL EnterotubeTM II (Becton Dickinson, Sparks, MD).

Enterotubes were incubated overnight (37C) and positive reactions were read according

to manufacturer's instructions.

Assembly of Tandem Filter Funnels

Two Whatman 25 mm disposable filter funnels with grade 4 filters (Whatman,

Clifton, NJ) were assembled in tandem (Figure 3-2) with the following modifications. In

the top filter funnel, the grade 4 filter was aseptically lifted and a 25 mm diameter VWR

413 filter was placed underneath. The threading on the funnel was lined with a layer of

Parafilm "M" laboratory film (American National CanTM, Chicago, IL). In the bottom

filter funnel, the grade 4 filter was replaced with a 25 mm FTA filter, aseptically cut

from a FTA Classic Card (Whatman*, Cat, No. WB 12 0205). The FTA filter was

covered with a 25 mm diameter plastic shield, which had a 6 mm hole punched in its

center. The plastic shield was aseptically cut from a plastic weight boat. The top filter

funnel was inserted into the top of the bottom filter funnel and the joint was sealed with

Parafilm.

Tomato rinses to be analyzed by PCR were transferred to Oxford 4 oz. specimen

cups (Cat # OX-067, International BioProducts, Bothell, WA) to facilitate easy pouring.










S Leave lid on top of tandem
filter assembly
A


Wrap threading with Parafilm



Grade 4 filter
S(20 25 microns)

25 mm VWR 413 filter
(5 microns)




B 1 l Joint sealed with Parafilm







25 mm Plastic shield
with 6 mm punch to
channel filtrate to
center of FTA filter

25 mm FTA filter


Grade 4 filter from this
filter funnel is discarded




Figure 3-2. Assembly of tandem filter funnels. (A) Top filters are for size exclusion, (B)
bottom FTA filter is for trapping bacteria, lysing bacterial cell walls, and
binding bacterial DNA.

A ring stand equipped with a clamp was used to secure a 500 ml Pyrex vacuum flask. The

side arm of the vacuum flask was connected to a water trap consisting of a 500 ml side

arm flask filled with Drierite (anhydrous calcium sulfate) (W.A. Hammond Drierite

Company Ltd., Xenia, OH) which was then connected to a vacuum pump (Emerson,






























Figure 3-3. Vacuum flask apparatus with tandem filter funnels. Filter funnels are attached
the top of a 500 ml filter flask via a plastic stem (supplied with the filter
funnels) inserted through the center of a number 7 rubber stopper. The filter
flask is attached via Tygon tubing to another filter flask which is set up as a
water trap. The water trap filter flask is connected to a vacuum pump.
Filtration of tomato rinses were facilitated by vacuum at 400 mm Hg.

model SA55NX6TE-4870, St. Louis, MO) using Tygon tubing. The top of the flask was

sealed with a No. 7 rubber stopper in which a 6 mm hole had been bored through the

center, through which a 4 inch plastic stem (supplied with the filter funnels) had been

inserted.

Tomato rinse was filtered through the tandem filter funnel via vacuum pressure

(400 mm Hg). After all of the rinse had passed through the FTA filter, the vacuum

pressure was turned off and the tandem filter funnel removed from the plastic stem. The

FTA filter and plastic shield was aseptically removed and placed in one compartment of

a three compartment Petri dish. FTA filters and shields were allowed to air dry

overnight with the shield faced down. Using the holes in the plastic shields as a guide, 6









mm punches were aseptically taken from each FTA filter and placed in a clean, sterile

1.5 microcentrifuge tube. FTA punches were washed twice with 500 ml FTA

Purification Reagent (Whatman, Clifton, NJ), then twice with 500 ml TE Buffer.

Washed punches were aseptically transferred to Petri dishes and dried in a 37C incubator

with the lids slightly open.

Nested PCR Amplification of ipaH gene for Detection of.\/nge//,

Dry, washed FTA punches were transferred to 0.5 ml microcentrifuge tubes and

subjected to a nested, two step PCR reaction. The PCR step 1 reaction used FTA

punches as DNA template submerged in 200 tl of PCR step reaction mix. The PCR step

1 reaction mix consisted of 142 tl PCR water, 20 p l 1Ox HotMaster Taq Buffer (with 25

mM Mg2+, pH 8.5), 20 pl dNTP mixture (2.5 mM each), 8.0 tl primer 01-001F, 8.0 pl

reverse primer, 2 tl HotMaster Taq Polymerase (Eppendorf, Hamburg, Germany). Prior

to PCR, 0.5 microcentrifuge tubes were centrifuged (30 seconds at 15,000 x g) to ensure

the FTA punch was completely submerged in step 1 reaction mix. The cycling

conditions for step 1 were: initial denaturation at 940C for 10 min; 30 cycles of

denaturation at 940C for 15 sec, annealing at 600C for 30 sec, and elongation at 720C for

45 sec; and a final elongation at 720C for 2 minutes.

PCR step 2 amplified a segment from within the target DNA segment of step 1. A 1

pl aliquot from the PCR step 1 product was diluted using 99 tl PCR water. Amplification

in PCR step 2 was performed in 25 pl reactions in clean, sterile 0.2 ml microcentrifuge

tubes. The step 2 reaction mixture consisted of 15.75 tl PCR water, 2.5 itl 1Ox PCR

Buffer (100 mM Tris-HC1, 500 mM KC1, 15 mM MgCl2, pH 8.3), 2.5 pl dNTP mixture

(2.5 mM each), 1.0 tl primer 01-005F, 1.0 pl primer 01-005R, 0.25 pl HotMaster taq

polymerase, and 2 tl DNA template (from the 1:100 dilution of step 1 product). The









cycling conditions for step 2 were: initial denaturation at 940C for 10 min; 30 cycles of

denaturation at 940C for 15 sec, annealing at 600C for 15 sec, and elongation at 720C for

15 sec; and a final elongation at 720C for 2 minutes.

DNA amplicons were analyzed by gel electrophoresis as described above. The

amplicon in step 1 was 620 bp and the amplicon from step 2 was 290 bp.

Recording of Data and Statistical Evaluation

All results from inoculation studies involving S. boydii UI02 and S. sonnei UI05

were recorded as either positive or negative for detection. Positive enrichment was scored

based on the isolation of inoculated S. boydii UI02 or S. sonnei UI05 by at least one of

the three plating media. Positive isolation by plating media resulted from typical

reactions for .\higel/h spp. in all confirmation steps and identification from Enterotubes.

Positive detection by PCR methods resulted from amplification of a single band of the

appropriate base pairs.

Logistical regression models were constructed using the R software (R

Development Core Team, Version 1.7.0 Patched) to identify significant differences

between isolation of .\hge//At spp. by PCR methods and enrichment methods and

isolation of.\h/ige//t spp. on the three plating media. Models for evaluating PCR methods

and enrichments were constructed (without an intercept) with covariate factors for

enrichments/PCR and inoculation levels. Models for evaluating plating media were

constructed (without an intercept) using plating media as covariates. Multiple

comparisons were performed using the Bonferroni method since this method is

conservative in its estimates of significance compared to other multiple comparison

methods. P values of < 0.05 were considered significant.














CHAPTER 4
RESULTS

This study consisted of two phases of research. The first phase consisted of

preliminary trials involving the preparation of growth curves, optical density standard

curves, and testing the specificity of each set of primers against a DNA library of positive

and negative controls. The second phase consisted of inoculation studies where tomatoes

were spot inoculated with S. boydii UI02 and S. sonnei UI05 at pre-determined levels.

Recovery/detection of the inocula was tested using conventional culture methods,

conventional culture methods with rifampicin supplemented enrichment, and a newly

developed FTA filtration/ nested PCR method.

Preliminary Trials

Growth Curves and Optical Density Standard Curves

Growth curves and optical density (O.D.) standard curves were prepared for S.

boydii UI02 wild strain, S. sonnei UI05 wild strain, and S. sonnei 9290 rifampicin

adapted. Since the growth curve and O.D. standard curve for S. sonnei 9290 wild strain

was similar to that of S. sonnei UI05 wild strain, results for the 9290 wild strain are not

shown. Results from growth curve and O.D. standard curve data were used to cultivate

consistent inocula, thereby reducing variability in later recovery studies on tomatoes.

S. boydii UI02 wild strain

The growth curve for S. boydii UI02 wild strain demonstrates that stationary

phase was reached in approximately 8 hours (Figure 4.1). A lag phase of approximately 4

hours was observed prior to exponential growth.










1.4
1.2
L1.0
C 0.8
0.6
< 0.4
0.2
0.0 -
0 2 4 6 8 10

Time (hours)


Figure 4-1. Growth curve: S. boydii UI02 wild strain. Each of three 100 ml TSB
microcosms were inoculated with 10 pl aliquots of an 18-hour S. boydii UI02
culture. The microcosms were incubated (37C, 30 rpm) and sampled at
appropriate time intervals. At each time of sampling, the absorbance at 600
nm (A 600 nm) was measured spectrophotometrically. Shown above is the
average A 600 nm of the three trials plotted against time.

An O.D. standard curve for S. boydii UI02 wild strain was prepared. The cell titer

in which the relationship between logo CFU/ml and absorbance at 600 nm (A 600 nm)

showed linearity (R2 = 0.9926; Figure 4-2) was approximately 7.03 x 10 to 3.18 x 108

CFU/ml.

S. sonnei UI05 wild strain

The growth curve prepared for S. sonnei UI05 wild strain demonstrates stationary

phase was reached in approximately 6 1/2 hours (Figure 4-3), with an initial lag phase of


approximately 3 hours.























Log Count (CFU/ml)


Figure 4-2. Optical density standard curve for S. boydii UI02 wild strain. Each of three
100 ml TSB microcosms were inoculated with 10 pl aliquots of an 18-hour S.
boydii UI02 culture. The microcosms were incubated (37C, 30 rpm) and
sampled at appropriate time intervals. At each time of sampling, appropriate
serial dilutions in PBS were pour-plated using TSA and incubated overnight
(37C). Shown above is the average absorbance at 600 nm (A 600 nm) of the
three trials (from Fig 4-1) plotted against the logo CFU/ml for the data points
which form a linear relationship.


0 1 2 3 4 5 6 7 8
Time (hours)


Figure 4-3. Growth curve: S. sonnei UI05 wild strain. Each of three 100 ml TSB
microcosms were inoculated with 10 pl aliquots of an 18-hour S. sonnei UI05
culture. The microcosms were incubated (37C, 30 rpm) and sampled at
appropriate time intervals. At each time of sampling, the absorbance at 600
nm (A 600 nm) was measured spectrophotometrically. Shown above is the
average A 600 nm of the three trials plotted against time.


R2 = 0.9926









An O.D. standard curve for S. sonnei UI05 wild strain was prepared. The cell titer

in which the relationship between logo CFU/ml and absorbance at 600 nm (A 600 nm)

showed linearity (R2 = 0.9833; Figure 4-4) was approximately 4.13 x 10 to 6.43 x 108

CFU/ml.


1.4
1.2

H R2 = 0.9866
S0.8
0.6
0.4
0.2
0


7.4 7.6


8 8.2 8.4 8.6


Log Count (CFU/ml)


Figure 4-4. Standard curve of O.D. (600 nm) of S. sonnei UI05 wild strain compared to
log plate count. Each of three 100 ml TSB microcosms were inoculated with
10 [l aliquots of an 18-hour S. sonnei UI05 culture. The microcosms were
incubated (37C, 30 rpm) and sampled at appropriate time intervals. At each
time of sampling, appropriate serial dilutions in PBS were pour-plated using
TSA and incubated overnight (37C). Shown above is the average absorbance
at 600 nm (A 600 nm) of the three trials (from Fig 4-1) plotted against the
logo CFU/ml for the data points which form a linear relationship.

S. sonnei 9290 rifampicin adapted strain

The growth curve for S. sonnei 9290 rifampicin adapted strain demonstrates

stationary phase was reached in approximately 8 12 hours (Figure 4-5), with an initial lag


phase of approximately 5 hours.










1.0

0.8

? 0.6

< 0.4

0.2

0.0
0 2 4 6 8 10 12
Time (hours)


Figure 4-5. Growth curve: S. sonnei 9290 rifampicin adapted strain. Each of three 100
ml TSB-R50 microcosms were inoculated with 10 pl aliquots of an 18-hour S.
sonnei 9290 culture. The microcosms were incubated (37C, 30 rpm) and
sampled at appropriate time intervals. At each time of sampling, the
absorbance at 600 nm (A 600 nm) was measured spectrophotometrically.
Shown above is the average A 600 nm of the three trials plotted against time.

An O.D. standard curve for S sonnei 9290 rifampicin adapted strain was

prepared. The cell titer in which the relationship between logo CFU/ml and absorbance at

600 nm (A 600 nm) showed linearity (R2 = 0.9944; Figure 4-6) was approximately 3.26 x

10 to 8.20 x 108 CFU/ml.

Primer Specificity

Each primer set (Table 3-3) was tested against DNA templates from both stock

\/Ngel//t spp. (Table 3-1) and closely related microorganisms (Table 3-2). Primer set 01-

001, which amplifies a 620 bp region of the ipaH gene of ./nge//lt spp. and

enteroinvasive E. coli (EIEC), successfully amplified DNA from all eight strains of

\/ngel//l Primer set 01-002, which targets a 215 bp region of the virA gene of.\/igel//A

spp. and EIEC, successfully amplified DNA from S. boydii UI02, S. flexneri, and S.

dysenteriae ATCC 9361, however it did not amplify a product from DNA from any of the










1
I -I----------------

0.8
0.8 -R2 = 0.9944
S0.6

| 0.4

0.2

0
7 7.5 8 8.5 9 9.5
Log Count (CFU/ml)


Figure 4-6. Standard curve of O.D. (600 nm) of S. sonnei 9290 rifampicin adapted
strain compared to log plate count. Each of three 100 ml TSB-R50
microcosms were inoculated with 10 pl aliquots of an 18-hour S. boydii UI02
culture. The microcosms were incubated (37C, 30 rpm) and sampled at
appropriate time intervals. At each time of sampling, appropriate serial
dilutions in PBS were pour-plated using TSA-R50 and incubated overnight
(37C). Shown above is the average absorbance at 600 nm (A 600 nm) of the
three trials (from Fig 4-1) plotted against the logo CFU/ml for the data points
which form a linear relationship.

four S. sonnei strains or S. boydii serogroup 18 (ATCC 35966). Primer set 01-004 and 01-

003, which amplify a 320 bp region of the invasion associated locus (ial) and a 217 bp

region from within that 320 bp region, respectively, successfully amplified a product

from the DNA of S. boydii UI02 only, while producing no product from S. sonnei UI05,

S. flexneri, and S. dysenteriae. Primer sets 01-004 and 01-003 were not tested against

DNA from the other four strains of.\/igel//t Primer set 01-005, which amplifies a 290 bp

region from within the region amplified by primer set 01-001, was successful in

amplifying DNA from all eight strains of .hige//At tested.

Primer sets 01-001, 01-002, and 01-005 did not amplify DNA from any of the non-

\ligell,// DNA (Table 3-2). Primer sets 01-004 and 01-003 were only tested against non-









.hge//Al DNA of Salmonella Gaminara and Salmonella Typhimurium, producing no

amplification in either sample.

Inoculated Studies

Detection of .\nge//At spp. by Conventional Culture Methods

Tomato rinses were enriched according to protocols of the FDA BAM, the

CMMEF, and in EE broth as described by Uyttendaele et al. (2000). Overnight

enrichments were then plated using SSA, MAC, and SPM. For all conventional culture

trials, 10 replicates were analyzed at each inoculation level. Inoculation levels were

verified by pour plating serial dilutions in triplicate with TSA-R50. Recovery results for

all conventional culture method experiments are expressed as "percent recovery," defined

as the number of tomatoes which tested positive for .\nge//lt spp. out of the total number

tested.

S. boydii UI02 was not recovered by conventional culture methods from any

samples inoculated at 106, 105, or 104 CFU/tomato. For this reason, trials at the lower

inoculation levels were not performed for S. boydii UI02.

Figure 4-7 demonstrates the percent recovery of S. sonnei UI05 by conventional

culture methods. At inoculation levels of 105, 104, and 103 CFU/tomato, 10% recovery of

S. sonnei UI05 was observed by CMMEF enrichment and plating on SSA and MAC.

When these enrichments were plated on SPM however, 20% recovery was observed from

inoculation levels of 105 and 103 CFU/tomato, and 10% recovery from 104 CFU/tomato.

No recovery of S. sonnei UI05 was observed with CMMEF enrichment of inoculation

levels of 102, 101, or 100 CFU/tomato.










100%

S80% 10E5
> 0 10E4
0 14
S60% 10
0 10E3
40% 10E2
20% 10E1



SSA MAC SPM SSA MAC SPM SSA MAC SPM







described in Chapter 3. Tomato rinses were enriched by \/nge//it culture
methods of the Compendium of Methods for the Microbiological Examination
of Foods (CMMEF), the FDA Bacteriological Analytical Manual (FDA
BAM), and in EE broth with 1.0 cg/ml novobiocin at 420C. Percent recovery
is calculated as the number of tomatoes from which S. sonnei UI05 was
isolated over the total number of tomatoes sampled for each enrichment.

S. sonnei UI05 recovered by enrichment by the FDA BAM is also shown in Figure

4-7. S. sonnei UI05 inoculated at 105 CFU/tomato was recovered at 60%, 70%, and 70%

when plated on SSA, MAC, and SPM, respectively. When inoculated at 104 CFU/tomato,

S. sonnei UI05 was recovered at 10%, 10%, and 20% when plated on SSA, MAC, and

SPM, respectively. At inoculation levels of 103 CFU/tomato, recovery rates of 20%, 30%,

and 40% were observed when plated on SSA, MAC, and SPM, respectively. At an

inoculation of 102 CFU/tomato, S sonnei UI05 was recovered from 20% of tomatoes

when plated on SSA, MAC, or SPM. S. sonnei UI05 was not recovered when inoculated

at levels of 101 or 100 CFU/tomato.









Results from enrichment of S. sonnei UI05 in EE broth as described by Uyttendaele

et al. (2000) is also shown in Figure 4-7. S. sonnei UI05 inoculated at 105 CFU/tomato

was recovered from 10%, 40%, and 50% of tomatoes when enrichments were plated on

SSA, MAC, and SPM, respectively. With an inoculation of 104 CFU/tomato, S. sonnei

UI05 was only recovered when plated on SPM (20%). When inoculated at 103

CFU/tomato, S. sonnei UI05 was recovered from 10% of tomatoes when plated on SSA,

MAC, or SPM. When inoculated at 102 CFU/tomato, S. sonnei UI05 was recovered from

10% of tomatoes when plated on MAC and SPM, but in none when plated on SSA. No

recovery of S. sonnei UI05 was observed when inoculated at levels of 101 or 100

CFU/tomato.

Detection of \l/igel/h, spp. by Conventional Culture Methods with Rifampicin
Supplemented Enrichment

Conventional culture methods (FDA BAM, the CMMEF, and in EE broth as

described by Uyttendaele et al. (2000)) were repeated using enrichments supplemented

with 50.g/ml rifampicin to exclude natural tomato microflora and rifampicin-adapted

inocula. Overnight enrichments were plated using SSA, MAC, and SPM. For all

conventional culture trials with rifampicin supplemented enrichment, 10 replicates were

analyzed at each inoculation level. Inoculation levels were verified by pour plating serial

dilutions in triplicate with TSA-R50. Recovery results for all conventional culture

method experiments are expressed as "percent recovery," defined as the number of

tomatoes which tested positive for ./hge//At spp. out of the total number tested.

Results for the recovery of S. boydii UI02 using conventional culture methods

where enrichments were supplemented with 50 .g/ml rifampicin (rif+) are shown in

Figure 4-8. Supplemented enrichment according to the CMMEF (rif+) protocol resulted









in 100% recovery of S. boydii UI02 from tomatoes inoculated at 106, 105, and 103

CFU/tomato on all three plating media. S. boydii UI02 inoculated at 104 CFU/tomato was

recovered from 40%, 80%, and 70% of tomatoes when plated on SSA, MAC, and SPM,

respectively. At an inoculation level of 102 CFU/tomato, S. boydii UI02 was recovered

from 60% of tomatoes when plated on SSA, and 50% when plated using MAC or SPM.

When inoculated at 101 CFU/tomato, S. boydii UI02 was recovered from 10% regardless

of which plating medium was used.


100%

80%

60%

40%

20%


* 10E6
* 10E5
O 10E4
O 10E3
* 10E2
* 10E1


FDA BAM (rif+)


Figure 4-8. Recovery of S. boydii UI02 by conventional culture methods with rifampicin
supplemented enrichment. Tomatoes were inoculated with rifampicin adapted
S. boydii UI02 at levels of 10E6 to 10E1. Inocula were recovered via a 100 ml
PBS rinse using a shake-rub-shake method as described in Chapter 3. Tomato
rinses were enriched by .\lg/ni// culture methods of the Compendium of
Methods for the Microbiological Examination of Foods (CMMEF), the FDA
Bacteriological Analytical Manual (FDA BAM), and in EE broth with 1.0
pg/ml novobiocin at 420C. All enrichments were supplemented with 50 pg/ml
rifampicin (rif+)to screen out natural background tomato microflora. Percent
recovery is calculated as the number of tomatoes from which S. boydii UI02
was isolated over the total number of tomatoes sampled for each enrichment.

Results from the enrichment of S. boydii UI02 by the FDA BAM (rif+) protocol are

shown in Figure 4-8. 100% recovery of S. boydii UI02 was achieved from tomatoes









inoculated at 106, 105, 104, and 103 CFU/tomato on all three plating media, except for

tomatoes inoculated at 104 CFU/tomato and plated on SSA, which were recovered at

90%. At an inoculation level of 102 CFU/tomato, S. boydii UI02 was recovered from 90%

of tomatoes on all three plating media. When inoculated at 101 CFU/tomato, S. boydii

UI02 was recovered from 10%, 20%, and 30% of tomatoes when plated using SSA,

MAC, and SPM, respectively.

Figure 4-8 shows that enrichment in EE broth (rif+) as described by Uyttendaele et

al. (2000) resulted in almost no recovery of S. boydii UI02. Only one tomato from the 103

CFU/tomato inoculation tested positive for S. boydii UI02 when plated on SSA.

Results for the recovery of S. sonnei UI05 using conventional culture methods

where enrichments were supplemented with 50 .g/ml rifampicin (rif+) are shown in

Figure 4-9. Supplemented enrichment of S. sonnei UI05 according to the CMMEF (rif+)

protocol resulted in 100% recovery when inoculated at 105 CFU/tomato, 90% recovery

when inoculated at 104 CFU/tomato, and 80% recovery when inoculated at 103

CFU/tomato and plated on SSA, MAC, or SPM. When S. sonnei UI05 was inoculated at

102 CFU/tomato, recovery was 20%, 30%, and 50% when plated on SSA, MAC, and

SPM, respectively. At an inoculation level of 101 CFU/tomato, S. sonnei UI05 was not

recovered when plated on SSA, and recovered from 20% of tomatoes when plated on

MAC or SPM. No S. sonnei UI05 was recovered from inoculation levels of 100

CFU/tomato.

Results from the enrichment of S. sonnei UI05 by the FDA BAM (rif+) protocol are

shown in Figure 4-9. 100% recovery was achieved from tomatoes inoculated at 105

CFU/tomato when plated SSA, MAC, or SPM. At an inoculation level of 104










100%

80% --- 10E5
S10oE4
60% 10E3

S40%- 10E2
SI 10El
S20% O0EO

0%
SSA MAC SPM SSA MAC SPM SSA MAC SPM

CMMEF (rif+) FDA BAM (rif+) EE Broth (rif+)


Figure 4-9. Recovery of S. sonnei UI05 by conventional culture methods with rifampicin
supplemented enrichment. Tomatoes were inoculated with rifampicin adapted
S. sonnei UI05 at levels of 10E5 to 10E0. Inocula were recovered via a 100 ml
PBS rinse using a shake-rub-shake method as described in Chapter 3. Tomato
rinses were enriched by .\iige/ll culture methods of the Compendium of
Methods for the Microbiological Examination of Foods (CMMEF), the FDA
Bacteriological Analytical Manual (FDA BAM), and in EE broth with 1.0
pg/ml novobiocin at 420C. All enrichments were supplemented with 50 pg/ml
rifampicin (rif+) to screen out natural background tomato microflora. Percent
recovery is calculated as the number of tomatoes from which S. sonnei UI05
was isolated over the total number of tomatoes sampled for each enrichment.

CFU/tomato, S. sonnei UI05 was recovered from 80% of tomatoes when plated on SSA,

and 90% of tomatoes when plated on MAC or SPM. When inoculated with 103

CFU/tomato, S. sonnei UI05 was recovered from 60% of tomatoes when plated on SSA,

and 80% of tomatoes when plated on MAC or SPM. For inoculation levels of 102 and 101

CFU/tomato, S. sonnei UI05 was recovered from 50% and 10% of tomatoes, respectively,

when plated on SSA, MAC, or SPM. No S. sonnei UI05 was recovered when inoculated

at 100 CFU/tomato.

Results from the enrichment of S. sonnei UI05 in EE broth (rif+) as described by

Uyttendaele et al. (2000) are shown in Figure 4-9. 100% recovery was observed at









inoculation levels of 105 CFU/tomato when plated on SSA, MAC, or SPM. At inoculation

levels of 104 CFU/tomato, S. sonnei UI05 was recovered from 90% of tomatoes when

plated on SSA, and 100% of tomatoes when plated on MAC or SPM. When inoculated at

a level of 103 CFU/tomato, S. sonnei UI05 was recovered from 30% of tomatoes when

plated on SSA and 80% of tomatoes when plated on MAC or SPM. At inoculation levels

of 102 CFU/tomato, S. sonnei UI05 was recovered from 10% of tomatoes when plated on

SSA and 20% of tomatoes when plated on MAC or SPM. No S. sonnei UI05 was

recovered from tomatoes inoculated with 101 CFU/tomato when plated on SSA, however

S. sonnei UI05 was recovered from 10% of tomatoes when plated on MAC or SPM. No

S. sonnei UI05 was recovered when inoculated at 100 CFU/tomato.

Lowest Detection Levels of Conventional Culture Methods

Results were reported based on the initial inoculation, however only 25 ml of the

100 ml tomato rinse was enriched by each protocol, therefore only a fourth of the inocula

could have theoretically been enriched by each protocol. When reporting the lowest

detection level (LDL) for conventional enrichments, corrections for the distribution of

inoculum have been made. For example, results reported for S. sonnei UI05 at 105

CFU/tomato were reported for an initial inoculation of 6.1 x 105 CFU/tomato, but each

enrichment procedure actually reflected a theoretical cell titer of 1.5 x 105 CFU,

assuming all of the inoculum was recovered and distributed evenly in the rinse. LDLs

from the enrichment procedures are reported as the lowest inoculation that resulted in

isolation of.\//ge//l spp. in at least one out of the 10 replicates. Conventional enrichment

procedures in the presence of natural tomato microflora resulted in no recovery of S

boydii UI02 (LDL >5.3 x 105 CFU/tomato); however LDLs for S. sonnei UI05 were 1.9 x

102 (FDA BAM), 1.5 x 103 (CMMEF), and 1.1 x 102 CFU/tomato (EE broth). For









enrichment procedures supplemented with rifampicin (rif+) to exclude natural tomato

microflora, LDLs were: 6.3 x 100 CFU/tomato (FDA BAM rif+ and CMMEF rif+) and

>5.3 x 105 CFU/tomato in EE broth rif+ for S. boydii UI02; and 1.9 x 101 (FDA BAM

rif+ and CMMEF rif+) and 1.1 x 101 CFU/tomato (EE broth rif+) for S. sonnei UI05.

Table 4-1. Lowest detection levels (LDLs) of conventional culture methods. Tomatoes
were inoculated with various levels of S. boydii UI02 or S. sonnei UI05.
Inocula were allowed to air dry before tomatoes were rinsed in 100 ml PBS
using a shake-rub-shake method. A 25 ml aliquot of the tomato rinse was
transferred to 225 ml of the appropriate enrichment broth and incubated
according .\nge//At culture methods found in the Compendium of Methods for
the Microbiological Examination of Foods (CMMEF), the U.S. FDA's
Bacteriological Analytical Manual (FDA BAM) or in EE Broth as described
by Uyttendaele et al. (2000). Enrichment procedures were repeated using
rifampicin adapted strains and supplemented enrichment (rif+). LDLs were
determined by calculating the inoculum applied to the respective tomato,
assuming 100% of the inoculum was recovered in the PBS rinse, and
adjusting for how much of that inoculum went into each enrichment.
Enrichment Procedure S. boydii UI02 S. sonnei UI05
(CFU/tomato) (CFU/tomato)
CMMEF >5.3 x 105 1.5 x 103
FDA BAM >5.3 x 105 1.9 x 102
EE Broth >5.3 x 105 1.1 x 102
CMMEF rif+ 6.3 x 100 1.9 x 101
FDA BAM rif+ 6.3 x 100 1.9 x 101
EE Broth rif+ >5.3 x 105 1.1 x 101


Additionally, the lowest detection level in which .\ igel/h spp. was isolated in

100% of the replicates (LDL100) was also determined for each enrichment procedure. In

studies involving conventional enrichment procedures in the presence of natural tomato

microflora, LDL100s were not achieved in this study, however from the results it can be

stated that the LDL100s of S. boydii UI02 and S. sonnei UI05 are >5.3 x 105 CFU/tomato

and >1.5 x 105 CFU/tomato, respectively. For trials involving rifampicin supplemented

enrichments, the LDL100s were: 6.3 x 102 CFU/tomato (FDA BAM and CMMEF) and

>5.3 x 105 CFU/tomato (EE broth) for S. boydii UI02; and 1.5 x 105 CFU/tomato (FDA









BAM and CMMEF), and 1.5 x 104 CFU/tomato (EE broth) for S. sonnei UI05. The high

inoculation levels required to achieve the LDL100 in conventional culture methods

demonstrates the need for better methods with which to evaluate food products for

\/nge//At spp.

Table 4-2. Lowest detection levels in which S. boydii UI02 or S. sonnei UI05 was isolated
in 100% of replicates (LDL100s) of conventional culture methods. Tomatoes
were inoculated with various levels of S. boydii UI02 or S. sonnei UI05.
Inocula were allowed to air dry before tomatoes were rinsed in 100 ml PBS
using a shake-rub-shake method. A 25 ml aliquot of the tomato rinse was
transferred to 225 ml of the appropriate enrichment broth and incubated
according .\ngel//t culture methods found in the Compendium of Methods for
the Microbiological Examination of Foods (CMMEF), the U.S. FDA's
Bacteriological Analytical Manual (FDA BAM) or in EE Broth as described
by Uyttendaele et al. (2000). Enrichment procedures were repeated using
rifampicin adapted strains and supplemented enrichment (rif+). LDL100s
were determined by calculating the inoculum applied to the respective tomato,
assuming 100% of the inoculum was recovered in the PBS rinse, and
adjusting for how much of that inoculum went into each enrichment.
Enrichment Procedure S. boydii UI02 S. sonnei UI05
(CFU/tomato) (CFU/tomato)
CMMEF >5.3 x 10 >1.5 x 105
FDA BAM >5.3 x 105 >1.5 x 105
EE Broth >5.3 x 105 >1.5 x 105
CMMEF rif+ 6.3 x 102 1.5 x 105
FDA BAM rif+ 6.3 x 102 1.5 x 105
EE Broth rif+ >5.3 x 105 1.5 x 104

Detection of.V\/ie/hi spp. by FTA Filtration / Nested PCR

FTA filtration/ nested PCR was used to detect inoculated S. boydii UI02 and S.

sonnei UI05 on inoculated tomatoes. Positive detection was recorded if a single band of

290 bp was produced in the second step of the nested PCR. Inoculation levels were

verified by pour plating serial dilutions in triplicate with TSA-R50.

Figure 4-10 shows representative gels from the first step of the nested PCRs.

Positive amplification in the first PCR produced a very faint band of 620 bp (Figure 4-

10A) from samples inoculated at 105 and 104 CFU/tomato with either S. boydii UI02 or S.









sonnei UI05 and no bands (Figure 4-10B) from samples inoculated at 103, 102, 101 and

100 CFU/tomato.






















Figure 4-10. Representative gels from the first step nested PCR. S. boydii UI02
inoculated at (A) 104 CFU/tomato and (B) 102 CFU/tomato. Lanes
assignments for both gels are: 1, 12, 13, and 24 empty; 2, 11, 14, and 23 -
DNA ladder; 3-10, 15, and 16 sample lanes; 17 tomato control; 18 filter
control; 19 PBS control; 20 water control; 21 negative control; 22 -
positive control.

Figure 4-11 shows a representative gel from the second step of the nested PCRs.

Positive amplification in the second step PCR resulted in a single band of 290 bp.

The results shown in Figure 4-12 demonstrate that S. boydii UI02 was detected at

rates of 100% regardless of inoculation level. S. sonnei UI05 was detected in 100% of

samples inoculated at levels of 105, 104, and 103 CFU/tomato. When inoculated at 102,

101, and 100 CFU/tomato, S. sonnei UI05 was detected on 90%, 40%, and 30% of the

tomatoes, respectively.





























Figure 4-11. Representative gel from the second step nested PCR. S. boydii UI02
inoculated at 104 CFU/tomato. Lanes assignments are: 1, 12, 13, and 24 -
empty; 2, 11, 14, and 23 DNA ladder; 3-10, 15, and 16 sample lanes; 17 -
tomato control; 18 filter control; 19 PBS control; 20 water control; 21 -
negative control; 22 positive control.

Lowest Detection Levels of the FTA Filtration/ Nested PCR Method

Results were reported based on initial inoculation levels. Since the entire tomato

rinse was evaluated by FTA filtration/ nested PCR, no corrections for the distribution of

inoculum was required. Reported results assumed recovery of 100% of the inoculum

from the tomato surface. LDLs of the FTA filtration/ nested PCR method for the

detection of S. boydii UI02 and S. sonnei UI05 were 6.2 x 100 CFU/tomato and 7.4 x 100

CFU/tomato, respectively (Table 4-3). In comparison, the LDL100s for S. boydii UI02

and S. sonnei UI05 were 6.2 x 100 CFU/tomato and 6.1 x 103 CFU/tomato, respectively

(Table 4-3).










100%

80%

8 60%








Inoculation Level (CFU/tomato)


Figure 4-12. Detection of S. boydii UI02 and S sonnei UI5 by FTAs filtration/ nested
PCR. Tomatoes were inoculated with S boydii UI02 or S. sonnei UI5 at
levels of 10E5 to 10E0. Inocula were recovered via a 100 ml PBS rinse using
a shake-rub-shake method as described in Chapter 3. Tomato rinses were
filtered using tandem filter funnels as described in Chapter 3. FTA punches
were used as DNA template in the first step of the nested PCR. Positive
detection was based on visualization of a 290 bp band from the second step of
the nested PCR. Percent recovery is calculated as the number of tomatoes
from which S. boydii UI02 or S. sonnei UI05 was isolated over the total
number of tomatoes sampled.

Table 4-3. Lowest detection levels and lowest detection levels in which S. boydii UI02 or
S. sonnei UI05 was isolated in 100% of replicates of FTA filtration/ nested
PCR. Tomatoes were inoculated with S. boydii UI02 or S. sonnei UI05 at
levels of 10E5 to 10EO. Inocula were recovered via a 100 ml PBS rinse using
a shake-rub-shake method as described in Chapter 3. Tomato rinses were
filtered using tandem filter funnels as described in Chapter 3. FTA punches
were used as DNA template in the first step of the nested PCR. Positive
detection was based on visualization of a 290 bp band in the second step of
the nested PCR.
S. boydii UI02 S. sonnei UI05
(CFU/tomato) (CFU/tomato)
LDL 6.2 x 100 7.4 x 100
LDL100 6.2 x 100 6.1 x 103














CHAPTER 5
DISCUSSION AND CONCLUSION

Foodborne shigellosis has been on the rise in recent years. Several outbreaks

involving fresh produce have demonstrated a need for a better method of evaluating

produce for .\nge//At spp. This study investigated conventional culture methods and a

newly developed FTA filtration/ nested PCR method for the detection of .\/hgel/l spp.

on tomato surfaces. S. boydii UI02 and S. sonnei UI05 were inoculated onto tomatoes at

levels of 106 through 100 CFU/tomato and 105 through 100 CFU/tomato, respectively.

Recovery of./hige//At from tomatoes was performed by placing inoculated tomatoes in

100 ml PBS and applying a shake/rub/shake method as previously described. Tomato

rinses were then analyzed by enrichment procedures as found in the U.S. Food and Drug

Administration's (1998) Bacteriological Analytical Manual (FDA BAM), the

Compendium of Methods for the Microbiological Examination of Food (CMMEF), and

by EE broth as described by Uyttendaele et al. (2000). .\/lge/ll, Plating Medium (SPM)

was evaluated for the isolation of,\/higel//t spp. against Salmonella-.\Igell/t agar (SSA),

MacConkey agar (MAC). Furthermore, tomato rinses were evaluated using a newly

developed FTA filtration/ nested PCR method.

Preliminary Studies

The two .\nge//At spp. isolates selected for this study were S. boydii UI02 and S.

sonnei UI05. Both strains were obtained from the laboratory of Dr. Hans Blaschek,

University of Illinois, and were involved in Chicago area outbreaks involving fresh









produce or products containing produce. S. sonnei represents the serogroup of.\/nge/lt

most common to North America, while S. boydii is rarely isolated in North America.

In order to evaluate the capability of conventional culture methods given total

specificity for .\/nlge/t spp., rifampicin resistance was generated in S. boydii UI02 and S.

sonnei UI05 subcultures by spontaneous mutation. Initially, dimethyl sulfoxide (DMSO)

was used to make stock solutions of rifampicin; however the use of DMSO resulted in the

production of strong malodors determined to be from the metabolism of DMSO to

dimethyl sulfide and methanethiol (data not shown). As the source of the malodor was

being determined, cultures were adapted to nalidixic acid as an alternative to rifampicin;

however in preliminary trials, natural microflora on tomatoes proved resistant of up to

400 ppm nalidixic acid. After the source of malodor was determined to be DMSO,

methanol was used to make stock solutions of rifampicin, which did not result in the

production of malodors. In a study on antimicrobial activities of aqueous and methanol

extracts of Juniperus oxycedrus, Karaman et al. (2003) used the addition of methanol

alone to various media as negative controls. Methanol was found to have no inhibitory

effects to various strains of Enterobacter, Klebsiella, Acinetobacter, and Pseudomonas

along with one strain of Escherichia coli. Based on these findings with closely related

microorganisms, methanol was assumed to have negligible effect on cell viability of S.

boydii UI02 and S. sonnei UI05.

Growth Characteristics of S. boydii UI02 and S. sonnei UI05

Growth characteristics of S. boydii UI02 and S. sonnei UI05 were determined by

performing growth curves and constructing optical density (O.D.) standard curves. In a

preliminary trial, a growth curve of S. sonnei 9290 (data not shown) was performed and

the absorbance was read at two wavelengths, 400 nm and 600 nm. When O.D. standard









curves were constructed using data from both wavelengths, there was no difference in the

range of linearity between absorbance and logo CFU/ml. For all the remaining growth

curves a wavelength of 600 nm was used.

Using growth curve and O.D. standard curve data, the exponential growth phase

doubling times for each strain investigated was calculated (Table 5-1).

Table 5-1. Doubling times associated with exponential growth phase of investigated
strains of./i/ge/hl spp.
Strain S. boydii UI02 S. sonnei 9290 S. sonnei UI05
Wild type 43.8 min 22.9 min 30.0 min
Nalidixic acid adapted 47.4 min N/A 45.0 min
Rifampicin adapted N/A 30.5 min N/A


It was observed that S. sonnei UI05 entered exponential phase growth and achieved

stationary phase faster than S. boydii UI02. This can be explained by the faster growth

rate calculated for S. sonnei UI05 and an extension in the lag phase associated with S.

boydii UI02. The wild strain of S. sonnei UI05 doubled every 30 minutes compared to

every 43.8 minutes for S. boydii UI02 (Table 5-1). The lag phase prior to exponential

growth observed for the wild strain of S. boydii UI02 was approximately 4 hours (Figure

4-1) compared to only approximately 3 hours for the wild strain of S. sonnei UI05 (Figure

4-3), with the onset of stationary phase at approximately 8 hours and 6 hours,

respectively.

Growth curve data from the wild strain S. sonnei 9290 was comparable to that of

the wild strain S. sonnei UI05. The time before exponential growth observed in the wild

strain S. sonnei 9290 was approximately 3 hours with the onset of stationary phase at

approximately 6 hours (data not shown). The S. sonnei 9290 wild strain doubled every

22.9 minutes compared to every 30.0 minutes for the S. sonnei UI05 wild strain When

compared with data collected for a rifampicin adapted strain of S. sonnei 9290 (Figure 4-









5), entry into exponential growth of rifampicin-adapted culture was delayed by

approximately 2 hours. Furthermore, the doubling time in exponential growth phase

increased to 30.0 minutes with the rifampicin adapted strain.

For rifampicin adapted strains of S. boydii UI02 and S. sonnei UI05, similar growth

characteristics to the S. sonnei 9290 rifampicin adapted strain were assumed. These

assumptions included a 2 hour extension of stationary phase between the wild strain and

rifampicin-adapted strain and slightly slower growth rates during exponential growth

phase. This assumption was supported by growth curve data collected for strains of S.

boydii UI02 and S. sonnei UI05 adapted and grown in the presence of nalidixic acid (data

shown in Appendix). Exponential growth rates of nalidixic acid-adapted strains grown in

the presence of nalidixic acid for both S. boydii UI02 and S. sonnei UI05 were slower

than the observed rates of the wild strains (Figure 5-1). Both nalidixic acid-adapted

strains experienced 4 hour extensions in lag phase compared to that of the wild strain,

which supports the assumption of a 2 hour extension for both strains adapted to

rifampicin. Assumptions for rifampicin adapted strains of S. boydii UI02 and S. sonnei

UI05 maintained entry into stationary phase at no later than 12 hours. Growth curve data

concluded that 18-hour cultures used in this study were late stationary phase cultures.

Evaluation of Primers Specific for .\/lge//At spp.

Several sets of primers were investigated for the detection of.\/ige//t spp. by PCR.

Results indicated that primer sets 01-001 and 01-005 were the only sets specific for all

four serogroups of.\ /ge//,t None of the primers amplified DNA from the non-.\ligel//

library. Primer sets 01-001 and 01-005 targeted 620 and 290 bp regions, respectively, of

the ipaH gene of .\lige//At spp. and enteroinvasive E. coli. When Jin et al. (2002)

sequenced the genome of S. flexneri 2a, the ipaH gene was located seven times on the









chromosome and five more times on the plasmid. Primer sets 01-002 and the nested set

01-003 and 01-004 target regions of the virA gene and the invasion associated locus (ial),

respectively. Unlike the ipaH gene, the virA gene and the ial are only located on the large

virulence plasmid of .\ligel//t spp. Negative results observed with primer sets 01-002, 01-

003, and 01-004 in this study could have been due to loss of plasmid during storage or

repeated transfers of the .\nge//At strains not amplified. No studies were conducted to

determine if plasmids were intact in these strains. The potential loss of plasmid does yield

an advantage to using primers which amplify the ipaH gene since detection of non-

pathogenic, plasmid-less strains would still be possible.

Inoculation Studies

Evaluation of Enrichment Protocols

Three enrichment protocols were investigated for the isolation of S. boydii UI02

and S. sonnei UI05 from inoculated tomatoes in the presence of natural background

microflora. When the S. sonnei UI05 data (Figure 4-7) were analyzed using a logistic

regression model, no significant difference (P = 1.000) among enrichment procedures

was observed. S. boydii UI02, however, was never recovered regardless of inoculation

level in any of the three investigated enrichment procedures. The inability for

conventional enrichment protocols to recover inoculated S. boydii UI02 demonstrates the

difficulty of isolating .\,ge//At spp. by traditional methods and the need for more selective

and sensitive media.

Conventional culture enrichment was repeated with background contamination

eliminated using enrichments supplemented with 50 pg/ml rifampicin (rif+) and

rifampicin-adapted inocula. In these studies, S. boydii UI02 was recovered by CMMEF

rif+ and FDA BAM rif+ protocols but not from the EE broth rif+ enrichment (Figure 4-









8), while S. sonnei UI05 was recovered by all three enrichments (Figure 4-9). Logistic

regression modeling revealed no significant difference (P = 1.000) between the CMMEF

rif+ and the FDA BAM rif+ enrichment methods for the recovery of S. boydii UI02, and

no significant differences (P = 1.000) among the CMMEF rif+, FDA BAM rif+, and EE

broth rif+ enrichments for the recovery of S. sonnei UI05. It was observed that although

S. boydii UI02 was unable to compete amid natural tomato microflora, it was recovered at

higher percentages than S. sonnei UI05 at each inoculation level in the CMMEF rif+ and

FDA BAM rif+ enrichments. In contrast, S. boydii UI02 was rarely recovered when

enriched in EE broth rif+ (1 out of 60 samples; Figure 4-8).

EE broth contains bile salts which have been reported to be inhibitory to some

,hlgel/t spp. particularly stressed cells (Tollison and Johnson, 1985). The S. boydii strain

used in this study may have been inhibited by the bile salts in EE broth. When

Uyttendaele et al. (2000) investigated EE broth among other enrichment broths for the

recovery of stressed and un-stressed Shigellae, only strains of S. sonnei and S. flexneri

were used. These results stress the importance of including all four serogroups of.\h/ige//A

when evaluating enrichment procedures. These results do not suggest EE broth may be

used in place of .\higell// broth for the enrichment of .\/hge//A, spp. since a known

pathogenic strain, S. boydii UI02, may be missed.

Evaluation of Plating Media

Using logistic regression models, the isolation of.\//ge//At spp. among isolation

media was compared. In all studies involving plating media, there was no significant

difference (a = 0.05) among isolation rates on the three media. It can be noted, however,

that differentiation of S. boydii UI02 and S. sonnei UI05 colonies from those of

background contaminants was far easier on SPM than on MAC or SSA. The most









common contaminants observed in this study were species of Enterobacter, Citrobacter,

and Klebsiella. These background contaminants were also identified in other studies in

which .\/ige/ll spp. isolation was attempted by conventional culture methods

(Uyttendaele et al., 2000; Schneider et al., unpublished data).

Figure 5-1 demonstrates the ability of SPM to allow easier differentiation of

various contaminants as compared to MAC and SSA. Colonies of.\hige//At spp. on SPM

are white while colonies of Enterobacter, Citrobacter, Acenitobacter, and Klebsiella are

either bluish or greenish. On MAC, ./gel//At spp. are translucent and slightly pink, with

and without rough edges. Enterobacter, Citrobacter, and Klebsiella spp. all make pink

colonies on MAC which may or may not be easily differentiated from colonies of

,\hgel//t spp. Since differentiation on MAC is based on lactose fermentation, lactose

negative strains of Enterobacter, Citrobacter, and Klebsiella will produce colonies that

are more translucent, resembling those of.\lhige//t spp. Furthermore, Acinetobacter

produces tiny translucent to pink colonies on MAC which are difficult to differentiate

from ,.s/gel//A spp. On SSA, Enterobacter and Klebsiella spp. produce pink colonies

however .\nge//li, Citrobacter and Acinetobacter spp. produce colonies that are

translucent to light pink. Isolation of ./hige//t spp. on MAC and SSA became

increasingly difficult as colony morphologies of ,/lhge//At spp. and/or several contaminant

colony morphologies were present on the same plate. While plating on SPM allowed

greater differentiation between .nge//, t spp. and contaminant colonies, the results

support previous suggestions that several different media with varying selectivity should

be used in order to increase the chance of isolating .\ngel//t spp. If only one plating






















































Figure 5-1. Differentiation of background microflora by isolation media. Tri-Plates
contain: 1) SSA, 2) MAC, and 3) SPM, streaked with: A) Enterobacter
cloacae, B) Klebsiella ozanae, C) Citrobacterfreundii, D) Acinetobacter
anitratus, E) .\l/ge/lt boydii UI02, and F) .\ige/llt sonnei UI05.









media is to be used, the results in this study suggest SPM can be used with equivalent

isolation rates of .\hge//At spp. as compared to MAC and SSA.

Analysis of Lowest Detection Levels of Conventional Culture Methods

The lowest detection levels (LDLs) of conventional culture methods for S. boydii

UI02 and S. sonnei UI05 were determined (Table 4-1). In a similar study, Jacobson et al.

(2002) evaluated the FDA BAM .\nge//At method using two strains of S. sonnei on

selected types of produce. LDLs were determined using unstressed, chill-stressed, and/or

freeze-stressed cells. LDLs with unstressed cells were less than 1.0 x 101 CFU/25g for all

produce types, while LDLs with chill-stressed and freeze-stressed cells were less than 5.2

x 101 CFU/25g for all produce types tested (Jacobson et al., 2002). LDLs with the FDA

BAM enrichment in this study for S. sonnei UI05 (1.9 x 102 CFU/tomato; Table 4-1)

were higher than LDLs reported by Jacobson et al. for S. sonnei strains 357 and 20143.

Variation between strains of S. sonnei could explain the difference in reported LDLs with

the FDA BAM method. Based upon its prevalence in outbreaks of shigellosis in the U.S.,

Jacobson's group chose to evaluate only strains of S. sonnei; however, results of this

study demonstrate the importance of including other serogroups as well. While the FDA

BAM appears to be quite effective based on evaluations with strains of S. sonnei (present

study and Jacobson et al., 2002) the LDLs observed for S. boydii UI02, a strain

responsible for a Chicago area outbreak, was >5.3 x 105 CFU/tomato. This indicates more

effective methods are needed for the detection of.\higel//t spp. in food.

The lowest detection levels in which the inoculum was recovered from 100% of the

replicates (LDL100s) of the conventional culture methods were also determined for S.

boydii UI02 and S. sonnei UI05 (Table 4-2). LDL100s of>5.3 x 105 CFU/tomato and

>1.5 x 105 CFU/tomato were observed for S. boydii UI02 and S. sonnei UI05,









respectively, by conventional culture methods in the presence of natural tomato

microflora. The inability of the conventional culture methods to detect S. boydii UI02 or

S. sonnei UI05 at high levels of contamination demonstrates the need for a more sensitive

and specific conventional culture method for the detection of.\'hige//t spp. on tomato

surfaces. When rif+ enrichments were used to select rifampicin-adapted inocula,

LDL100s were 6.3 x 102 CFU/tomato (FDA BAM rif+ and CMMEF rif+) and >5.3 x 105

CFU/tomato (EE Broth rif+) for S. boydii UI02 and 1.5 x 105 CFU/tomato (FDA BAM

rif+ and CMMEF rif+) and 1.5 x 104 CFU/tomato (EE Broth rif+) for S. sonnei UI05. The

exclusion of natural tomato microflora by rif+ enrichment did not facilitate the consistent

detection of inoculated S. boydii UI02 or S. sonnei UI05 at levels sufficient to cause

disease, further demonstrating the need for a more sensitive and specific conventional

culture method for the detection of.\hige//At spp. on tomato surfaces.

Sources of Variation Among Trials

In Figure 4-8, inconsistent recovery results were obtained from S. boydii UI02 data

sets at the 104 CFU/tomato inoculation levels. Recovery by conventional culture methods

with rifampicin supplemented enrichment at the 105 CFU/tomato and 103 CFU/tomato

both resulted in 100% recovery, while the 104 CFU/tomato was less than 100% for the

CMMEF enrichment. In this study, trials involving 106, 105, and 104 CFU/tomato

inoculations were performed on one day, while the trials involving 103, 102, and 101

CFU/tomato inoculations were performed on another day. This gap in recovery could

possibly be explained by variations in attachment and/or potential biofilm formation. In

support of this theory, the 104 CFU/tomato set was run last on its day of sampling, while

the 103 CFU/tomato was run first on its day of sampling. This would allow a significantly

greater amount of time for the 104 CFU/tomato inoculum to attach and/or form biofilms









while the 106 and 105 CFU/tomato data sets were run that day. Unpublished data using

the S. boydii UI02 strain (Blaschek et al., University of Illinois) has suggested the

formation of biofilms on the surfaces of produce, which resulted in difficult removal and

resistance to produce washes. Another factor was that tomatoes used in different trials

were harvested and shipped in different lots. On some trial days the tomatoes available

were visibly cleaner than for other trials. The presence of excess filth on the surface of

tomatoes might have contributed to sub-optimal attachment.

The potential for S. boydii UI02 and S. sonnei UI05 to form biofilms can be tested

using several methods. To simply determine whether these strains produce biofilms, two

traditional tests could be performed: the tube test (Christensen et al., 1982) or the

microtiter-plate test (Christensen et al., 1982; Christensen et al., 1985). The tests involve

staining the bacterial film with a cationic dye and reading the degree of film formation

either visually (tube test) or by optical density (microtiter-plate assay). The major

drawbacks of these tests are in their qualitative nature; the tube test is visual, while the

microtiter-plate assay only measures the film formation on the bottom of the well

(Stepanovic et al., 2000). In order to better quantify biofilm formation, Stepanovic et al.

(2000) describe a modified microtiter-plate assay which involves fixing of bacterial films

with methanol, staining with crystal violet, releasing the bound dye with 33% glacial

acetic acid, and measuring the optical density with a plate reader. This assay could be

performed with S. boydii UI02 and S. sonnei UI05 strains using several time periods for

attachment/formation to determine the rate of biofilm formation, if any. Conversely,

biofilm formation on tomato surfaces could be directly observed by environmental

scanning electron microscopy (Blaschek et al., unpublished material).









The effect of filth on tomato surfaces on S. boydii UI02 and S. sonnei UI05

attachment and subsequent recovery could be tested as well. Tomatoes with visible filth

could be obtained and a portion washed and rinsed to remove any visible filth. Using

rifampicin adapted strains, dirty and clean tomatoes could be used in inoculation/recovery

procedures as described in Chapter 3, Materials and Methods. Enumeration of recovered

S. boydii UI02 and S. sonnei UI05 from both clean and filthy tomatoes should provide

insight as to their effect on attachment/recovery characteristics.

Comparison of Conventional Culture Methods and FTA Filtration/ Nested PCR

Logistic regression analysis was used to compare isolation rates of the conventional

culture methods to the detection rates of FTA filtration/ nested PCR method for S.

boydii UI02 and S. sonnei UI05. For studies involving conventional culture methods and

S. sonnei UI05, the FTA filtration/ nested PCR method was significantly better than the

CMMEF (P = 0.007), the FDA BAM (P = 0.003), and the EE broth enrichment (P =

0.001). For studies involving rifampicin supplemented enrichments and S. boydii UI02,

the FTA filtration/ nested PCR method was significantly better than enrichment by the

CMMEF (P = 0.010) or enrichment in EE broth (P < 0.001), however it was not

significantly different then enrichment by the FDA BAM (P = 0.177). For studies

involving rifampicin supplemented enrichments and S. sonnei UI05, the FTA filtration/

nested PCR method was significantly better than the CMMEF (P = 0.007), the FDA

BAM (P = 0.003), and enrichment in EE broth (P = 0.001).

When tomatoes were analyzed by FTA filtration/ nested PCR method, much faster

results were obtained than with conventional culture methods. The time required for a

confirmed result by conventional culture methods was 4 days at best. Since positive

confirmation was based on a positive result in the second step PCR, the FTA filtration/









nested PCR could produce a confirmed result in as quickly as 2 days. It should be noted

that until a tandem filter funnel assembly comparable to that used in this study can be

commercially purchased, the assembly process of such tandem filters requires significant

time, energy, and attention to aseptic techniques. The assembly process of tandem filter

funnels alone provides a risk of contamination that needs to be addressed by proper

laboratory practices. Contamination problems experienced in this study were attributed to

the assembly process of tandem filter funnels.

Analysis of Lowest Detection Levels of FTA Filtration/ Nested PCR

The LDLs and LDL100s were determined for S. boydii UI02 and S. sonnei UI05

for FTA filtration/ nested PCR (Table 4-3). Other studies which investigated the

detection of.\/nge///,i spp. in food by PCR techniques reported LDLs of: 1.0 x 101 -1.0 x

102 CFU/ml (Vantarakis et al., 2000), 1.0 x 102 -1.0 x 103 CFU/ml (Villalobo and Torres,

1998), 1.0 x 101 CFU/ml (Lindqvist, 1999), and 1.1 x 101 CFU/ml (Theron et al., 2001).

Vantarakis et al. (2000) were only able to achieve detection of S. dysenteriae type 1 from

homogenized mussel samples at 1.0 x 10 -1.0 x 102 CFU/ml after a 22 hour pre-

enrichment step in peptone water was employed. Without the pre-enrichment a LDL of

1.0 x 103 CFU/ml was observed. In a similar study, Villalobo and Torres (1998) observed

LDLs of S. dysenteriae type 1 to be 1.0 x 102 -1.0 x 103 CFU/ml in homogenized

mayonnaise samples. Lindqvist (1999) utilized buoyant density centrifugation food

inoculated with S. flexneri to increase sensitivity of a nested PCR assay from 1.0 x 103

CFU/ml to 1.0 x 101 CFU/ml. Finally, a 6 hour pre-enrichment step in GN broth was

necessary for Theron et al. (2001) to detect S.flexneri at 1.1 x 101 CFU/ml in

environmental water samples by semi-nested PCR. While the pre-enrichment increased









sensitivity through multiplication of S. flexneri, it also served to dilute PCR inhibitors

found in several of the environmental water samples.

In comparison to these previous studies investigating the detection of.\h/ige//a spp.

in food with PCR techniques, the FTA filtration/ nested PCR method detects S. boydii

UI02 and S. sonnei UI05 with equivalent sensitivity. Tandem filter funnels were able to

filter large volumes of tomato rinse, concentrating bacteria in the center of the FTA

filter thereby increasing the ability to detect .\/ige, / spp. present at low levels. In

addition, no pre-enrichment step was required to increase sensitivity in FTA filtration/

nested PCR, therefore faster results can be obtained. Furthermore, washing protocols for

FTA punches serve to remove any potential PCR inhibitors.

Optimization of the FTA Filtration/ Nested PCR Assay

Further development of the FTA filtration/ nested PCR method is required to

obtain the desired amplification on the first step PCR. Figure 4-10A and 4-10B show

electrophoresis gels obtained with step 1 PCR amplification of sample tomatoes

inoculated with S. boydii UI02 at 104 CFU/tomato and 102 CFU/tomato, respectively. At

the 104 CFU/tomato inoculation level only very faint bands can be seen upon close

observation, while at 102 CFU/tomato no bands can be observed. In step 2 PCR however,

all inoculation levels were successfully amplified as shown in Figure 4-11. If negative

amplification in step 1 of the FTA filtration/ nested PCR method is to be used to

determine a sample "negative" for .\nge//At spp., further optimization of this step is

required as ,\hgel/ht spp. present at levels < 103 CFU/tomato would be missed.

Preliminary optimization trials were performed using inoculation levels of 102

CFU/tomato in order to visualize bands from the step 1 PCR by adjusting the ramp rate of

the annealing step from 3/sec to l/sec, increasing the number of PCR cycles from 30 to









40, sectioning the FTA punches, and adjusting the temperature control on the

thermocycler from "tube" to "block." The ramp rate and temperature control settings

were adjusted to account for variation between our Eppendorf Mastercycler gradient and

thermocyclers used in previous studies with FTA punches (Orlandi and Lampel, 2000;

Lampel et al., 2000). Sectioning FTA punches into two or four equal parts with a sterile

scalpel was attempted to allow the punch to remain completely submerged in PCR

reaction mix for the duration of the PCR (Figure 5-2). Even though high speed

centrifugation was used to force the FTA punches down into the reaction mix, the 6 mm

diameter of the punch forced them back to the surface during PCR. FTA punch sections

sank to the bottom of reaction tubes and remained there for the duration of the PCR.











Figure 5-2. FTA punches in 0.5 ml microcentrifuge tubes. A) Un-manipulated punches
held at top of reaction mix by diameter; B) punches sectioned into halves; C)
punches folded with forceps.

Both changing the ramp rate for the annealing step and adjusting the temperature

control from "tube" to "block" resulted in non-specific amplification (data not shown).

Increasing PCR cycles from 30 to 40 alone, and increasing the number of cycles in

combination with sectioning FTA punches into halves or quarters both resulted in

observable bands from step 1 PCR (data not shown). As previously mentioned, the FTA

punches were sectioned using a sterile scalpel and forceps. Although positive results were

obtained from sectioning FTA punches, this practice is not recommended for future









studies due to the increased potential for cross-contamination. Instead, FTA punches in

future optimization studies will be folded during transfer using forceps against the inside

wall of the microcentrifuge reaction tube prior to step 1 PCR. Folding allows the punches

to remain completely submerged (Figure 5-2) in the PCR reaction mix to allow more

optimal amplification without adding the potential chance of contamination associated

with sectioning.

Predictive Value of Testing for .,/ntlgel'/ spp.

It should be noted that the predictive value (PV) of any microbiological assay for

the detection of .,/ige//t spp. in food is affected by both the prevalence of ./Nigel,// spp.

associated with that food product and the specificity and sensitivity of the testing

procedure. In the Institute of Food Technologists (IFT) Expert Report on Emerging

Microbiological Food Safety Issues, Implications for Control in the 21st Century, PV of

testing is explained as a function of prevalence, specificity, and sensitivity. According to

the PV model, if the pathogen is prevalent at a high frequency the PV of a test with high

specificity and sensitivity will be quite high. Conversely, if a pathogen is prevalent at

very low frequency, then the PV of a test with high specificity and sensitivity remains

quite low. Such is the case with ,\/ige/ll spp. whose prevalence on produce has been

demonstrated at < 4.1% (FDA 2001a; FDA 2001b); therefore the PV of the sampling is

very low. Although the conversion from conventional culture methods to nucleic-acid

assays such as PCR will allow for greater test specificity and sensitivity, the PV of these

tests to screen produce for potential ,\/ige/ll spp. contamination will remain low.

Conclusions

The results of this study demonstrate the superiority of the FTA filtration/ nested

PCR method over conventional culture methods for the detection of.\h/ge//At spp. from









inoculated tomato surfaces. The FTA filtration/ nested PCR method was successful in

detecting as few as 6.2 S. boydii UI02 cells and 7.2 S. sonnei UI05 cells amid high

background contamination. If future proposed methods for the isolation/detection of

.hgel//At spp. from food are to include a conventional culture analog, the results of this

study suggest that either the enrichment protocols described by the FDA BAM or the

CMMEF should be considered. EE broth, although adequately effective for S. sonnei

UI05, did not recover S. boydii UI02 from inoculated tomato surfaces. Previous

suggestions of using several isolation media of varying selectivity were supported by

results of this study. SPM is capable of isolating S. boydii UI02 and S. sonnei UI05 at

rates equivalent to MAC and SSA, while providing a greater differentiation between

colonies of.\/i/ge/lt spp. and closely related contaminants.















APPENDIX
GROWTH CHARACTERISTICS OF NALIDIXIC ACID ADAPTED STRAINS

Growth curves and optical density (O.D.) standard curves were prepared for S.

boydii UI02 and S. sonnei UI05 strains adapted to nalidixic acid (NA).

S. boydii UI02 NA adapted strain

The growth curve for S. boydii UI02 NA adapted strain demonstrates that

stationary phase was reached in approximately 13 hours (Figure B-l). A lag phase of

approximately 7 hours was observed prior to exponential growth.


1.200
1.000
0.800
0 0.600
< 0.400
0.200
0.000
3 5 7 9 11 13 15
Time (hours)


Figure A-1. Growth curve: S. boydii UI02 nalidixic acid adapted strain. Each of three
100 ml TSB (200 ppm NA) microcosms were inoculated with 10 tl aliquots
of an 18-hour S. boydii UI02 culture. The microcosms were incubated (37C,
30 rpm) and sampled at appropriate time intervals. At each time of sampling,
the absorbance at 600 nm (A 600 nm) was measured spectrophotometrically.
Shown above is the average A 600 nm of the three trials plotted against time.

An O.D. standard curve for S. boydii UI02 NA adapted strain was prepared. The

cell titer in which the relationship between logo CFU/ml and absorbance at 600 nm (A









600 nm) showed linearity (R2 = 0.9764; Figure B-2) was approximately 9.17 x 106 to 4.90

x 107 CFU/ml.


7.1 7.3 7.5 7.7 7.9


Log Counts (CFU/ml)


Figure A-2. Standard curve of O.D. (600 nm) of S. boydii UI02 nalidixic acid adapted
strain compared to log plate count. Each of three 100 ml TSB (200 ppm NA)
microcosms were inoculated with 10 pl aliquots of an 18-hour S. boydii UI02
culture. The microcosms were incubated (37C, 30 rpm) and sampled at
appropriate time intervals. At each time of sampling, appropriate serial
dilutions in PBS were pour-plated using TSA (100 ppm NA) and incubated
overnight (37C). Shown above is the average absorbance at 600 nm (A 600
nm) of the three trials (from Fig 4-1) plotted against the logo CFU/ml for the
data points which form a linear relationship.

S. sonnei UI05 NA adapted strain

The growth curve prepared for S. sonnei UI05 wild strain demonstrates stationary

phase was reached in approximately 10 hours (Figure B-3), with an initial lag phase of

approximately 7 hours.

An O.D. standard curve for S. sonnei UI05 wild strain was prepared. The cell titer

in which the relationship between logo CFU/ml and absorbance at 600 nm (A 600 nm)

showed linearity (R2 = 0.9821; Figure B-4) was approximately 4.13 x 10 to 6.43 x 108


CFU/ml.










1.200
1.000
0.800
0.600
0.400
0.200
0.000


5 6 7 8 9 10 11


Time (hours)


Figure A-3. Growth curve: S. sonnei UI05 nalidixic acid adapted strain. Each of three
100 ml TSB (200 ppm NA) microcosms were inoculated with 10 tl aliquots
of an 18-hour S. sonnei UI05 culture. The microcosms were incubated (37C,
30 rpm) and sampled at appropriate time intervals. At each time of sampling,
the absorbance at 600 nm (A 600 nm) was measured spectrophotometrically.
Shown above is the average A 600 nm of the three trials plotted against time.


S0.7

0.5

0.3

0.1


Log Count (CFU/ml)


Figure A-4. Standard curve of O.D. (600 nm) of S. sonnei UI05 nalidixic acid adapted
strain compared to log plate count. Each of three 100 ml TSB (200 ppm NA)
microcosms were inoculated with 10 tl aliquots of an 18-hour S. sonnei UI05
culture. The microcosms were incubated (37C, 30 rpm) and sampled at
appropriate time intervals. At each time of sampling, appropriate serial
dilutions in PBS were pour-plated using TSA (100 ppm NA) and incubated
overnight (37C). Shown above is the average absorbance at 600 nm (A 600
nm) of the three trials (from Fig 4-1) plotted against the logo CFU/ml for the
data points which form a linear relationship.


R2 = 0.9821















LIST OF REFERENCES


[Anonymous]. 2002. About-Shigella.com. www.about-shigella.com. Accessed 2002 Nov
21.

[APHA] American Public Health Association. 2001. Compendium of Methods for the
Microbiological Examination of Foods. 4th Edition. Chapter 38 .\h/gel//

Applied Biosystems. 2003. ABI Prism 7000 online information.
http://www.appliedbiosystems.com/products/productdetail.cfm?prodid=641.
Accessed 2003 May 23.

Bagamboula, C., M. Uyttendaele and J. Debevere. 2002. Acid Tolerance of.\Nihgell//
sonnei and .\nge/ll tflexneri. J. Appl. Microbiol. 93:479-486.

Bemardini, M.L., J. Mounier, H. d'Hauteville, M. Coquis-Rondon and P.J. Sansonetti.
1989. Indentification of icsA, a Plasmid Locus of .\rhge/ltflexneri that governs
Bacterial Intra- and Intercellular Spread Through Interaction with F-actin. Proc.
Natl. Acad. Sci. US A. 86:3867-3871.

Beuchat, L.R., L.J. Harris, T.E. Ward and T.M. Kajs. 2001. Development of a Proposed
Standard Method for Assessing the Efficacy of Fresh Produce Sanitizers. J. Food
Prot. 64(8):1103-1109.

Bhat, P. and D. Rajan. 1975. Comparative Evaluation of Desoxycholate Citrate Medium
and Xylose Lysine Desoxycholate Medium in the Isolation of Shigellae. Am. J.
Clin. Pathol. 64:399-403.

Brown, J.E., P. Echeverria and A.A. Lindberg. 1991. Digalactosyl-Containing
Glycolipids as Cell Surface Receptors for Shiga Toxin of.\/ige//At dysenteriae 1
and Related Cytotoxins of Escherichia coli. Rev. Infect. Dis. 13(Suppl 4):S298-
303.

[CDC] Centers for Disease Control and Prevention. 2002a. .\/nge//l Annual Summary,
2001. www.cdc.gov/ncidod/dbmd/phlisdata/shigella.htm. Accessed 2002 Oct 11.

[CDC] Centers for Disease Control and Prevention. 2002b. Preliminary FoodNet Data on
the Incidence of Foodborne Illnesses Selected Sites, United States, 2001.
Morbidity and Mortality Weekly Report. April 19,2002. 51(15):325-329.
http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5115a3.htm. Accessed 2002
Oct 11.









[CDC] Centers for Disease Control and Prevention. 2003. Shigellosis. Technical
Information. http://www.cdc.gov/ncidod/dbmd/diseaseinfo/shigellosist.htm.
Accessed 2003 May 22.

Cepheid. 2003. Smart Cycler online information.
http://cepheid.com/pages/smartcycler.html. Accessed 2003 May 23.

Christensen, G.D., W.A. Simpson, A.L. Bisno and E.H. Beachey. 1982. Adherence of
Slime-Producing Strains of Staphylococcus epidermidis to Smooth Surfaces. Infect.
Immun. 37:318-326.

Christensen, G.D., W.A. Simpson, J.J. Younger, L.M. Baddour, F.F. Barrett, D.M.
Melton and E.H. Beachey. 1985. Adherence of Coagulase-Negative Staphylococci
to Plastic Tissue Culture Plates: a Quantitative Model for the Adherence of
Staphylococci to Medical Devices. J. Clin. Microbiol. 22: 996-1006.

Clerc, P., A. Ryter, J. Mounier and P.J. Sansonetti. 1987. Plasmid-mediated Early Killing
of Eukaryotic Cells by .\/nlge/tflexneri as Studied by Infection of J774
Macrophages. Infect. Immun. 55:521-527.

Donohue-Rolfe, A., D.W.K. Acheson and G.T. Keusch. 1991. Shiga Toxin: Purification,
Structure, and Function. Rev. Infect. Dis. 13(Suppl 4):S293-297.

Echeverria, P., O. Sethabutr and C. Pitarangsi. 1991. Microbiology and Diagnosis of
Infections with .\/n/ge/lt and Enteroinvasive Escherichia coli. Rev. Infect. Dis.
13(Suppl 4):S220-5.

Egile, C., H. d'Hauteville, C. Parsot and P.J. Sansonetti. 1997. SopA, the Outer
Membrane Protease Responsible for Polar Localization of IcsA in .\/hge//ltflexneri.
Mol. Microbiol. 23:1063-1073.

[ERS] Economic Research Service, U.S. Department of Agriculture. 2003. Briefing
Room: Tomatoes. http://www.ers.usda.gov/Briefing/Tomatoes/background.htm.
Accessed 2003 July 10.

[FDA] Food and Drug Administration. 1998. Bacteriological Analytical Manual. 8th
Edition. AOAC International, Arlington, VA.

[FDA] Food and Drug Administration. 2001a. Survey of Domestic Fresh Produce:
Interim Results. http://www.cfsan.fda.gov/-dms/prodsur9.html. Accessed 2002
Nov 21.

[FDA] Food and Drug Administration. 2001b. FDA Survey of Imported Fresh Produce
FY 1999 Field Assignment. http://www.cfsan.fda.gov/-dms/prodsur6.html.
Accessed 2002 Nov 21.









[FDA] Food and Drug Administration. 2001c. FDA Survey of Imported Fresh Produce:
Imported Produce Assignment FY 2001.
http://www.cfsan.fda.gov/-dms/prodsur7.html. Accessed 2002 Nov 21.

[FDA] Food and Drug Administration. 2001d. Analysis and Evaluation of Preventive
Control Measures for the Control and Reduction/Elimination of Microbial Hazards
on Fresh and Fresh-Cut Produce. http://www.cfsan.fda.gov/-comm/ift3-1.html.
Accessed 2002 Dec 5.

Fontaine, A., J. Arondale and P.J. Sansonetti. 1988. Role of Shiga Toxin in the
Pathogenesis of Bacillary Dysentery Studied Using a Tox- Mutant of.\/ige//A
dysenteriae 1. Infect. Immun. 56:3099-3109.

[FTC] Florida Tomato Committee. 2003. Tomato 101: Health Information and Research.
http://www.floridatomatoes.org/education.htm. Accessed 2003 July 10.

Galanakis, E., M. Tzoufi, M. Charisi, S. Levidiotou and Z. Papadopoulou. 2002. Rate of
Seizures in Children with Shigellosis. Acta Paediatrica. 91(1): 1001-2.

Goldberg, M.B., O. Barzu, C. Parsot and P.J. Sansonetti. 1993. Unipolar Localization and
ATPase Activity of IcsA, a /ige/lliflexneri Protein Involved in Intracellular
Movement. J. Bacteriol. 175:2189-2196.

Hale, T. 1991. Genetic Basis of Virulence in ./nge//At Species. Microbiol. Rev. 55:206-
224.

d'Hauteville, H., R. Dufourcq Lagelouse, F. Nato and P.J. Sansonetti. 1996. Lack of
Cleavage of IcsA in .liigell/iflexneri Causes Aberrant Movement and Allows
Demonstration of a Cross-Reactive Eukaryotic Protein. Infect. Immun. 64:511-517.

Headley, V., M. Hong, M. Galko and S.M. Payne. 1997. Expression of Aerobactin Genes
by .l/igelltflexneri During Extracellular and Intracellular Growth. Infect. Immun.
65:818-821.

Ingersoll, M, E.A. Groisman and A. Zychlinsky. 2002. Pathogenicity Islands of.\/nge//ll
Curr. Topics Microbiol. Immunol. 264(1):49-65

Institute of Food Technologists. 2003. Expert Report. Emerging Microbiological Food
Safety Issues. Implications for the 21st Century.

Islam, D. and A.A. Lindberg. 1992. Detection of /hlge//lt dysenteriae Tyoe 1 and
\//lge/lltflexneri in Feces by Immunomagnetic Isolation and Polymerase Chain
Reaction. J. Clin. Microbiol. 30(11):2801-2806.

Jacobson, A.P., M.L. Johnson, T.S. Hammack and W.H. Andrews. 2002. Evaluation of
the Bacteriological Analytical Manual (BAM) Culture Method for the Detection of
Shigella sonnei in Selected Types of Produce. Poster presented at 2002 FDA
Science Forum. Washington, D.C.









Jin, Q., Z. Yuan, J. Xu, Y. Wang, Y. Shen, W. Lu, J. Wang, H. Liu, J. Yang, F. Yang, X.
Zhang, J. Zhang, G. Yang, H. Wu, D. Qu, J. Dong, L. Sun, Y. Xue, A. Zhao, Y.
Gao, J. Zhu, B. Kan, K. Ding, S. Chen, H. Cheng, Z. Yao, H. Bingkun, R. Chen, D.
Ma, B. Qiang, Y. Wen, Y. Hou and J. Yu. 2002. Genome Sequence of.\l/ge/lt
flexneri 2a: Insights into Pathogenicity Through Comparison with Genomes of
Escherichia coli K12 and 0157. Nucleic Acids Res. 30(20):4432-4441.

June, G.A., P.S. Sherrod, R.M. Amaguana, W.A. Andrews and T.S. Hammack. 1993.
Evaluation of the BacteriologicalAnalytical Manual Culture Method for the
Recovery of .\/nge/lt sonnei from Selected Foods. J. AOAC Int. 76(6):1240-1248.

Karaman, I., F. Sahin, M. Gillice, H. Oguttg, M. Sengul and A. Adiguzel. 2003.
Antimicrobial Activity of Aqueous and Methanol Extracts ofJuniperus oxycedrus
L. J. Ethnopharm. 85:231-235.

Kaufman, P.R., C.R. Handy, E.W. McLaughlin, K. Park and G.M. Green. 2000.
Understanding the Dynamics of Produce Markets: Consumption and Consolidation
Grow. USDA, Economic Research Service. Agriculture Information Bulletin No.
758. http://www.ers.usda.gov/publications/aib758/aib758.pdf Accessed 2002 Dec
5.

Keusch, G.T., M. Jacewicz, M. Mobassaleh and A. Donohue-Rolfe. 1991. Shiga Toxin:
Intestinal Cell Receptors and Pathophysiology of Enterotoxic Effects. Rev. Infect.
Dis. 13(Suppl 4):S304-310.

Lampel, K.A., P.A. Orlandi and L. Kornegay. 2000. Improved Template Preparation for
PCR-Based Assays for the Detection of Food-Borne Bacterial Pathogens. Appl.
Environ. Microbiol. 66(10):4539-4542.

Lawlor, K.M. and S.M. Payne. 1984. Aerobactin Genes in Shigella spp. J. Bacteriol.
160:266-272.

Lett, M.-C., C. Sasakawa, N. Okada, T. Sakai, S. Makino, M. Yamada, K. Komatsu and
M. Yoshikawa. 1989. virG, a Plasmid-Coded Virulence Gene of .\l/gell/ flexneri:
Identification of the virG Protein and Determination of the Complete Coding
Sequence. J. Bacteriol. 171:353-359.

Lew, J.F., D.L. Swerdlow, M.E. Dance, P.M. Griffin, C.A. Bopp, M.J. Gillenwater, T.
Mercantante and R.I. Glass. 1991. An Outbreak of Shigellosis Aboard a Cruise
Ship Caused by a Multiple-Antibiotic-Resistant Strain of .\S/gel/, flexneri.
American J. Epidemiol. Aug 15;134(4):413-420.

Lindqvist, R. 1999. Detection of.\/nge/lt spp. in Food with a Nested PCR Method -
Sensitivity and Performance Compared with a Conventional Culture Method. J.
Appl. Microbiol. 86:971-978.

Loisel, T.P., R. Boujemaa, D. Pantaloni and M.-F. Carlier. 1999. Reconstitution of Actin-
Based Motility of Listeria and ./nge,//At Using Pure Proteins. Nature. 401:613-616.









Long, S.M., G.K. Adak, S.J. O'Brien and I.A. Gillespie. 2002. General Outbreaks of
Infectious Intestinal Disease Linked with Salad Vegetables and Fruit, England and
Wales, 1992-2000. Common Dis. Public Health. 5(2):101-105.

Makino, S., C. Sasakawa, T. Kamata and M. Yoshikawa. 1986. A Genetic Determinate
Required for Continuous Reinfection of Adjacent Cells on a Large Plasmid of
.\nlgehll flexneri 2a. Cell. 46:551-555.

Mehlman, I.J., A. Romero and B.A. Wentz. 1985. Improved Enrichment for Recovery of
.\ligellt sonnei from Foods. J. AOAC. 68:552-555.

Miki, H., K. Miura and T. Takenawa. 1996. N-WASP, a Novel Actin-Depolymerizing
Protein, Regulates the Cortical Cytoskeletal Rearrangement in a PIP2-Dependent
Manner Downstream of Tyrosine Kinases. EMBO J. 15:5326-5335.

Millipore. 2003. Sample Preparation Methods for DNA Analysis. Online information.
http://www.millipore.com/catalogue.nsf/docs/C7489?open&lang=de. Accessed
2003 May 23.

[MMWR] Morbidity and Mortality Weekly Report. 1999. Outbreaks of .\/Nigel/t sonnei
Infection Associated with Eating Fresh Parsley -- United States and Canada, July -
Agust 1998. April 16, 1999 / 48(14);285-289.

[MMWR] Morbidity and Mortality Weekly Report. 2000. Public Health Dispatch:
Outbreak of .\/i/ge//t sonnei Infections Associated with Eating a Nationally
Distributed Dip -- California, Oregon, and Washington, January 2000. January 28,
2000 / 49(03);60-61.

Monack, D.M. and J.A. Theriot. 2001. Actin-Based Motility is Sufficient for Bacterial
Membrane Protrusion Formation and Host Cell Uptake. Cell. Microbiol. 3(9):633-
647.

Muriana, P.M. 2002. .\l/nge'// Pathogenesis and Genetic Basis of Virulence. Online
Lecture. http://www.okstate.edu/OSU Ag/fapc/fsw/shigella/shigpm.htm. Accessed
2003 June 18.

Olsen, J.E., S. Aabo, W. Hill, S. Notermans, K. Wernars, P.E. Granum, T. Popovic, H.N.
Rasmussen and 0. Olsvik. 1995. Probes and Polymerase Chain Reaction for
Detection of Food-Borne Bacterial Pathogens. Int. J. Food Microbiol. 28:1-78.

Orlandi, P.A. and K.A. Lampel. 2000. Extraction-Free, Filter Based Template
Preparation for Rapid and Sensitive PCR Detection of Pathogenic Parasitic
Protozoa. J. Clin. Microbiol. 38(6):2271-2277.

Parsot, C. and P.J. Sansonetti. 1996. Invasion and the Pathogenesis of.\/ilgel//t Infections.
Curr Topics Microbiol. Immunol. 209:25-42.