DNA Aptamers for Biomarker Discovery in Ovarian Cancer

MISSING IMAGE

Material Information

Title:
DNA Aptamers for Biomarker Discovery in Ovarian Cancer
Physical Description:
1 online resource (148 p.)
Language:
english
Creator:
Van Simaeys, Dimitri O
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Chemistry
Committee Chair:
Tan, Weihong
Committee Members:
Smith, Benjamin W
Horenstein, Nicole A
Fanucci, Gail E
Fletcher, Bradley S

Subjects

Subjects / Keywords:
aptamer -- biomarkers -- cancer -- ovary -- stip1
Chemistry -- Dissertations, Academic -- UF
Genre:
Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Ovarian cancer is the most deadly gynecological malignancy.  Especially ovarian clear cell adenocarcinoma shows a particularly poor response to the standard treatment.  In general, the five-year survival rate of patients diagnosed in Stage I ovarian cancer is about 90%,however only 2% of all ovarian cancer is diagnosed at this stage.  The underlying reasons for this low number are the vague symptoms accompanied with this disease and the poor diagnostic tests available.  Furthermore, there is aneed for the differentiation between the two most prevalent types of ovarian cancer: clear cell adenocarcinoma and serous adenocarcinoma, which behave different clinically.  To enable physicians to diagnose ovarian cancer at an earlier stage and differentiate clear cell carcinoma from serous carcinoma, a panel of aptamers was selected against clear cell adenocarcinoma and ovarian cancer stem cells.  Aptamers are ssDNA or ssRNA based oligonucleotides that exhibit properties similar to antibodies, but these ligands can be generated in vitro.  A total of nine aptamers have been selected for TOV-21G, an ovarian clear cell adenocarcinoma cell line, and 5 aptamers from ovarian cancer stem cells have been selected.  The aptamers all show sub nanomolar to lower nanomolar affinities.  While the aptamers selected for TOV-21G show no binding to HeLa or the serous adenocarcinoma cell line CAOV3, there was binding to CEM, A172, HCT-116 and HL60 cells, which suggests shared molecular markers between these unrelated cell lines.  Furthermore, a strategy for aptamer target identification was outlined and used for the identification of the aptamer TOV6.  The protein found to bind to TOV6 was Stress induced Protein 1, a co-chaperone that is known to interact withheat shock protein 90.  The target was validated by siRNA silencing, a protein blot on the recombinant protein andantibody binding.  Stress-Induced Protein1 was expressed in several other ovarian cancer cell lines, including SKOV3,OVCAR3, TOV112D and C13. Finally, the role of Stress-Induced Protein 1 was studied on TOV-21G in relation with aptamer TOV6.  Results indicate that TOV6 is a potent inhibitor of invasion by binding to Stress induced protein.  The aptamer also has cytostatic properties.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Dimitri O Van Simaeys.
Thesis:
Thesis (Ph.D.)--University of Florida, 2012.
Local:
Adviser: Tan, Weihong.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-02-28

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2012
System ID:
UFE0044450:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 DNA APTAMERS FOR BIOMARKER DISCOVERY IN OVARIAN CANCER By DIMITRI VAN SIMAEYS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Dimitri Van Simaeys

PAGE 3

3 To my Family

PAGE 4

4 ACKNOWLEDGMENTS I would like to give special thank s to my advisor Dr. Weihong Tan of the department of Chemistry, University of Florida for the support and faith he has shown me and the chances he has provided me throughout my studie s at the University of Florida that lead to this dissertation and the development of my career. I also owe my gratitude to Dr. Rebecca Sutphen from the College of Medicine, University of South Florida for her support and knowledge in our collaboration. I wou ld like to thank my committee members Dr. Gail Fanucci, Dr. Nicole Horenstein, Dr. Ben Smith and Dr. Brad Fletcher for their input and constructive feedback that lead to the overall improvement of this dissertation I would like to thank the members from the Tan group, with special thanks to Dr. Kwame Sefah and Dr. Dalia Lpez C o l n for teaching me the ropes and SELEX to the cannot be underestimated, as she w as a defining factor in many aspects of the work I have done on ovarian cancer. I would also like to thank the current and past members of the Tan group, for the many meaningful discussions, friendship and support by Xiangling Xiong, Dr. Jennifer Martin, Elizabeth Jimenez, Dr. Meghan Dr. M. Carmen Estevez Jun Lui, Kejing Zhang, Sena Cansiz, Guizhi Zhu, Diane Turek, Danny Chang Dr. Tahir Bayrac, Jing Zheng, Carole Champanhac and Julia Vaizer. I would also like to give than ks to Dr. Pamela Havre for her friendship and guidance I would like to thank my parents Filip and Nadia Van Simaeys and my brother Mathieu Van Simaeys, for their support and encouragement and having me in their prayers. Finally I would like to thank al l my friends in both Belgium and the US for their support and encouragement.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURE S ................................ ................................ ................................ .......... 9 LIST OF SCHEMES ................................ ................................ ................................ ...... 12 LIST OF ABBREVIATIONS ................................ ................................ ........................... 13 ABSTRACT ................................ ................................ ................................ ................... 16 CHAPTER 1 BACKGROUND ................................ ................................ ................................ ...... 18 Cancer ................................ ................................ ................................ .................... 18 Ovarian Cancer ................................ ................................ ................................ 19 Diagnosis of Ovarian Cancer ................................ ................................ ............ 20 Biomarkers ................................ ................................ ................................ .............. 20 Biomarkers in Ovarian Cancer ................................ ................................ ......... 20 An Introduction to Biomarkers ................................ ................................ .......... 21 Membrane Proteins ................................ ................................ .......................... 22 Other Biomarker Discovery Techniques ................................ ........................... 23 Aptamers ................................ ................................ ................................ ................ 24 An Introduction to Aptamers ................................ ................................ ............. 24 SELEX ................................ ................................ ................................ .............. 26 Aptamer Target Recognition Interactions ................................ ......................... 27 Cell SELEX ................................ ................................ ................................ ...... 29 Principle ................................ ................................ ................................ ..... 29 Applications of aptamers derived from cell SELEX ................................ .... 29 Overview of the Dissertation ................................ ................................ ............. 31 2 SELECTION AND CHARACTERIZATION OF OVARIAN CLEAR CELL ADENOCARCINOMA CELL APTAMERS ................................ ............................... 37 Introduction: Ovarian Clear Cell Adenocarcinoma ................................ .................. 37 Materials and Methods ................................ ................................ ............................ 38 Instrumentation and Reagents ................................ ................................ ......... 38 Cell Culture and Buffers ................................ ................................ ................... 39 SELE X Library and Primer Design ................................ ................................ ... 39 In Vitro Cell SELEX ................................ ................................ .......................... 40 Sequencing and Selection of Putative Aptamers ................................ .............. 41

PAGE 6

6 Affinity Studies: Flow Cytofluorometric Analysis for the Determination of Binding Affinity ................................ ................................ .............................. 42 Selectivity and Specificity ................................ ................................ ................. 42 Effect of Temperature on Aptamer Bin ding ................................ ...................... 42 Protease Digestion Assay ................................ ................................ ................ 43 Results ................................ ................................ ................................ .................... 43 Monitoring of Pool E nrichment for TOV 21G against HeLa .............................. 43 Aptamer Characterization ................................ ................................ ................. 44 Concluding Remarks ................................ ................................ ............................... 45 3 METHOD DEVELOPMENT FOR APTAMER TARGET IDENTIFICATION ............. 63 Introduction ................................ ................................ ................................ ............. 63 Materials and Methods ................................ ................................ ............................ 65 Instrumentation and Reagents ................................ ................................ ......... 65 Cell Culture and Buffers ................................ ................................ ................... 66 Aptamer Target Purification for Protein Identification ................................ ....... 67 StIP1 siRNA Knockdown ................................ ................................ .................. 67 Antibody Biotinylation ................................ ................................ ....................... 68 Aptamer Blotting ................................ ................................ ............................... 68 Results ................................ ................................ ................................ .................... 68 Outline of the Aptam er Mediated Protein Identification Procedure ................... 68 SDS PAGE of the Bead Binding Fraction of Whole Cell Lysate ....................... 70 Confirmation of the Binding of aptTOV6 to StIP1 ................................ ............. 71 Discussion ................................ ................................ ................................ .............. 72 Conclusions ................................ ................................ ................................ ............ 74 4 THE FUNCTION OF StIP1 IN TOV 21G ................................ ................................ 87 Introduction ................................ ................................ ................................ ............. 87 Materials and Methods ................................ ................................ ............................ 88 Instrumentation, Cell Culture and Reagents ................................ ..................... 88 Proliferation Assay ................................ ................................ ........................... 89 Tumor Invasion Assay ................................ ................................ ...................... 89 Membrane Expression of StIP1 in Ovarian Cancer Cell Lines ......................... 8 9 Statistical Analysis ................................ ................................ ............................ 90 Results ................................ ................................ ................................ .................... 90 Growth Inhibitory Effects of TOV6 ................................ ................................ .... 90 The Non Proliferative Effects of 17AAG ................................ ........................... 90 The Effect of StIP1 siRNA Silencing on Invasion ................................ ............. 91 ................. 92 Discussion ................................ ................................ ................................ .............. 92 5 OVARIAN CANCER STEM CELL SELEX ................................ ............................ 106 Introduction ................................ ................................ ................................ ........... 106 Cell SELEX on an Ovarian CSC Cell Line ................................ ............................ 108

PAGE 7

7 Materials and Methods ................................ ................................ ................... 108 Instrumentation and reagents ................................ ................................ .. 108 Cell culture and buffers ................................ ................................ ............ 109 Cell SELEX library ................................ ................................ ................... 109 In vitro cell SELEX on ovarian CSCs ................................ ....................... 110 Specificity and affinity studies ................................ ................................ .. 110 Aptamer mediated cell sorting with magnetic beads ................................ 111 Results ................................ ................................ ................................ ........... 111 Monitoring the pool enrichment for undifferentiated OCSC vs differentiated ovarian stem cells ................................ ........................... 111 Binding assay of putative aptamers and determination of K d .................. 112 Discussion ................................ ................................ ................................ ............ 113 Future Work ................................ ................................ ................................ .......... 113 6 SUMMARY AND FU TURE DIRECTIONS ................................ ............................ 122 Summary ................................ ................................ ................................ .............. 122 Future Work ................................ ................................ ................................ .......... 125 APPENDIX THE ANALYSIS OF NEXT GEN SEQUENCING DATA FOR CELL SELEX ................................ ................................ ................................ .................. 129 Introduction ................................ ................................ ................................ ........... 129 The Trimming of the Sequencing Data With PERL ................................ ............... 130 Future Work ................................ ................................ ................................ .......... 133 LIST OF REFERENCES ................................ ................................ ............................. 135 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 148

PAGE 8

8 LIST OF TABLES Table page 1 1 The description of the different stages of ovarian cancer according to the AJCC ................................ ................................ ................................ .................. 36 2 1 Selected aptamer frequencies in the analyzed pools ................................ ......... 60 2 2 Compendium of the aptamers obtained by sel ection against TOV 21G ............. 61 2 3 Overview of the binding studies performed on multiple types of ce ll lines. ......... 62 5 1 The aptamers that show specificity towards OCSCs ................................ ........ 121 5 2 Binding assay of aptamers from OSCS with other ovarian cancer cell lines .... 121

PAGE 9

9 LIST OF FIGURES Figure page 1 1 Change of the US mortality rate by cause from 2005 to 2001 ............................ 33 1 2 Adenocarcinoma of the ovary (excluding borderline tumors): relative survival rates (%) by AJCC Stage, Ages 20+, 12 SEER Areas, 1988 2001. ................... 34 1 3 The basic scheme for cell SELEX ................................ ................................ ...... 35 2 1 Sample agarose gel of a semiquantitative evaluation of pool enrichment. ......... 47 2 2 Binding assay of the enriched pools with TOV 21G and HeLa cells. .................. 48 2 3 Binding profile of aptTOV1 ................................ ................................ ................. 49 2 4 Binding profile of aptTOV2 ................................ ................................ ................. 50 2 5 Binding profile of aptTOV3 ................................ ................................ ................. 51 2 6 Binding profile of aptTOV4 ................................ ................................ ................. 52 2 7 Binding profile of aptTOV5 ................................ ................................ ................. 53 2 8 Binding profile of aptTOV6 ................................ ................................ ................. 54 2 9 Binding profile of aptTOV7 ................................ ................................ ................. 55 2 10 Binding profile of aptTOV8 ................................ ................................ ................. 56 2 11 Binding profile of aptTOV9 ................................ ................................ ................. 57 2 12 K d determination for aptTOV1. Cells were incubated with varying concentrations of PE Cy5 labeled aptamer in duplicate. ................................ .... 58 2 13 Specificity assay for aptTOV1 on TOV 21G and HeLa. ................................ ...... 59 3 1 General procedure for protein identification with the use of aptamers. ............... 76 3 2 Study of the effect of formaldehyde on streptavidin binding and biotin elution of the desthiobiotin conjugated aptamer. ................................ ............................ 77 3 3 Silver staining of the material obtained from the aptamer mediated protein purification for aptTOV6. ................................ ................................ ..................... 78 3 4 Comparison of sample runs with insufficient or excessive crosslink times. ........ 79

PAGE 10

10 3 5 Proteins found in the aptamer TOV6 binding fractions. StIP1 was the only top protein found in both samples sent for analysis. ................................ ........... 80 3 6 Silencing of StIP1 in TOV 21G cells. ................................ ................................ .. 81 3 7 Silencing of StIP1 in A172 cells. ................................ ................................ ......... 82 3 8 Absence of PTK7 silencing with StIP1 siRNA treatment in A172 cells. .............. 83 3 9 Absence of PTK7 silencing with StIP1 siRNA treatment in TOV 21G cells. ....... 84 3 10 A: The binding of AptTOV6 with (blue) and without (black) StIP1 knock down; B: The binding of StIP1 antibody M33 with (blue) or without (black) StIP1 ......... 85 3 11 Chemi luminescent blot of rhStIP1. The aptamer is able to induce a strong chemiluminescent signal with rhSTIP1, but not with BSA. ................................ .. 86 4 1 Figure of a Boyden chamber for migration and invasion studies. ....................... 95 4 2 Microscopic images of TOV 21G cells treated with TOV6 for 3 days. ................ 96 4 3 MTT assay of TOV 21G cells incubated with library and TOV6. ......................... 97 4 4 Normalized cell count of TOV 21G after 3 days of incubation with TOV6. ......... 98 4 5 Normalized proliferation study of TOV 21G, after three days of incubation with 17AAG ................................ ................................ ................................ ......... 99 4 6 Microscopic images of TOV 21G cells treated with different levels of 17AAG in full media after 2 day s incubation. ................................ ................................ 100 4 7 Migration of TOV 21G across a microporous membrane. ................................ 101 4 8 Invasion assay of TOV 21G determining the effect of TOV6 on the ability of TOV 21G to cross a matrigel layer. ................................ ................................ .. 102 4 9 Combined TOV 21G cell invasion assay. The data suggests that the invasion is facilitated through a mechanism where StIP1 is involved. .............. 103 4 10 Binding assay of several ovarian cancer c ell lines with TOV6First column ...... 104 4 11 A proposed mechanism for TOV6 caused inhibition of cell invasion. ............... 105 5 1 Enrichment for ovarian cancer stem cells. The pools show binding to the undifferentiated cells. ................................ ................................ ........................ 115 5 2 Binding profile of the s elected pools on the negative cell line, ovarian differentiated cancer cells (ODC). ................................ ................................ ..... 116

PAGE 11

11 5 3 Binding assay of DOCSC 1 on undifferen tiated OCSC vs differentiated cells .. 117 5 4 Binding assay of DOCSC 2 on undifferentiated OCSC vs differentiated cells .. 117 5 5 Binding assay of DOCSC 3 on undifferentiated OCSC vs differentiated cells .. 118 5 6 Binding assay of DOCSC 4 on undifferentiated OCSC vs differentiated cells .. 118 5 7 Binding assay of DOCSC 5 on undifferentiated OCSC vs differentiated cells .. 119 5 8 Example of a K d determination for the aptamer DOCSC 3. .............................. 119 5 9 Cell sorting experiment with DOCSC 5. Red: Cells; Green: Library; Orange: DOCSC 5 unsorted; Black: DOCSC 5 after sorting ................................ .......... 120 5 10 Microscope images of one day old cells that have been sorted with aptamer functionalized magnetic beads. Left: DOCSC 2, Right: DOCSC 5 .................. 120 A 1 for sequencing. .............................. 134

PAGE 12

12 LIST OF SCHEMES Scheme page 3 1 The chemistry of formaldehyde mediated DNA Protein cross linking. ................ 75

PAGE 13

13 LIST OF ABBREVIATION S 17AAG 17 N allylamino 17 demethoxygeldanamycin 2D LC MS/MS 2 Dimensional liquid chromatography mass spectrometry/ mass spectrometry AJCC American joint committee on cancer ATCC American type cell culture ATP Adenosine tri phosphate BB Binding buffer BRCA Breast c ancer BSA Bovine serum albumin CA125 Carbohydrate antigen 125 CD Cluster of differentiation ChIP Chromatin immuno precipitation CPG Controlled porous glass CSC Cancer stem cell EDTA Ethylenediaminetetraacetic acid FACS Fluorescence assisted cell sorting FBS Fetal bovine serum FDA Food and drug administration FITC Fluorescein isothiocyanate Freq. Frequency HBSS HE4 Human epididymis 4 HOP Hsp70 hsp90 organizing protein HSP Heat shock protein

PAGE 14

14 IC50 Inhibition concentration were 50% of the maximum activity is reached LOD Limit of detection M CSF Macro phage colony stimulating factor MMP Matrix metalo protease MTT 3 (4,5 Dimethylthiazol 2 yl) 2,5 diphenyltetrazolium bromide NHS N Hydroxysuccinimide OCCA Ovarian clear cell adenocarcinoma OCSI Ovarian cancer symptom i ndex PBS Phosphate buffer solution PCR Polymerase chain r eaction PE Cy5.5 Phycoerythrin cyanine derivative 5.5 PERL Prac tical extraction and reporting l anguage PTK7 Protein tyrosin k inase 7 qPCR Qu antitative polymerase chain r eaction SDS PAGE Sodium d odec yl sulphate polyacrylamide gel e lectroph oresis SELEX Systematic evolution of l igands by ex ponential enrichment siRNA Small interfering RNA SOAC Serous ovarian adenocarcinoma ssDNA Single stranded deoxyribonucleic acid ssRNA Single stranded ribonucleic acid StIP1 Stress induced p rotein 1 Surfactant Surface active agents TCA Trichloro acetic acid TVU Trans vaginal u ltrasound USPSTF U.S. preventive services task f orce

PAGE 15

15 WB Washing b uffer

PAGE 16

16 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DNA APTAMERS FOR BIOMARKER DISCOVERY IN OVARIAN CANCER By Dimitri Van Simaeys August 2012 Chair: Weihong Tan Major: Chemistry Ovarian cancer is the most deadly gynecologic al malignancy. In particular, ovarian clear cell adenocarcinoma shows a n especially poor response to the standa rd treatment. In general, the five year survival rate of patient s diagnosed in Stage I ovari an cancer is about 90%; however only 2% of all ova rian cancer is diagnosed at this stage. The underlying reasons for this low number are the vague symptoms of this disease and the poor diagnostic tests available Furthermore, there is a need to differentiate the two most pr evalent types of ovarian cance r, clear cell adenocarcinoma and serous adenocarcinoma, which behave different clinically. To enable physicians to di a gnose ovarian cancer at an earlier stage and differentiate clear cell carcinoma from serous carcinoma, in this research a panel of aptamers was selected against clear cell adenocarcinoma and ovarian cancer stem cells. Aptamers are ssDNA or ssRNA oligonucleotides with selectivities and binding affinities similar to antibodies, but with the benefit of in vitro generation A total of nine aptamers have been selected for TOV 21G, an ovarian clear cell adenocarcinoma cell line, and 5 aptamers for ovarian cancer stem cells were selected. The aptamers all show s ubnanomolar to low nanomolar affinities. While the aptamers select ed for TOV

PAGE 17

17 21G show no binding to HeLa or the serous adenocarcinoma cell line CAOV3, there was binding to CEM, A172, HCT 116 and HL60 cells, suggesting that these unrelated cell lines have the same surface molecular markers Furthermore, a strategy for a ptamer target identification was outlined and used to determine that the molecular target of the aptamer TOV6 is Stress induced Protein 1, a co chaperone that is known to interact with heat shock protein 90. The target was validated by siRNA silencing, a protein blot on the recombinant protein and antibody binding. Stress I nduced Protein 1 is expressed in several other ovarian cancer cell lines, including SKOV3, OVCAR3, TOV112D and C13. Studies of TOV 21G and apta mer TOV6 indicate that TOV6 is a potent inhibitor of invasion of the basement membrane by binding to Stress i nduced protein 1 The aptamer also has cytostatic properties.

PAGE 18

18 CHAPTER 1 BACKGROUND Cancer Cancer is a group of diseases in which abnormal cells are growing and spreading in an abnormal fashion. Carcinogenesis, the initiation of can cer, can be caused by external factors, internal factors or a combination of both Predictions i n 2011 projected that 1,596,670 people would to be diagnosed with cancer and 571,950 people would die of cancer. I n 2011, one in four deaths will be caused by cancer 1 Over the past century, remarkable progress has been made in the fight against various disea ses, as can be seen in Figure 1 1. The leading cause of deaths in the US, heart disease, has seen over a 50% reduction from 1950 to 2001. Similarly, deaths due to cerebrovascular disease were reduced by two third s in the same period, and deaths caused by the flu and related diseases were redu ced by about one half. However, there has been no progress towards improving the survival rate for cancer. Progress has been made in the last ten years, but with no more then a few percent reduction per year, dependent on the type of cancer 2 There is a dire need for better treatment as well as better diagnostic methods 3 Cancers in which considerable progress has been made towards reducing mortality rates include lung and bronchial cancer (20% reduction 1990 2000), and prostate cancer (45% reduction 1990 2005) in men. Similar figures have been observed in women with regards to stomach and breast cancers. However pancreatic cancer liver cancer, ovarian cancer and leukemia have not experienced reductions in mortal ity rates over the past decades 1 T he absence of improvement in the statistics from ovarian cancer indicates that the current methods of fighting this deadly disease

PAGE 19

19 are inadequate and that more insight in the molecular biology of this type of cancer is needed. Ovarian Cancer Ovarian cance r is cancer that begins in the o varies with n inety percent attributed to epithelial origin. E pithelial ovarian cancer is classified by its histological subtypes, with serous and clear cell carcinoma being the most predominant subtypes of ovarian cancer 4 As mentioned above, there has been little progress in the treatment and survival rate of ovarian cancer, with 15,460 people expected to die from this disease in 2012 1 Symptoms that could indicate the presence of ovarian cancer are vague: pelvic or abd ominal pain, increased urinary urgency or frequency, bloating and difficulty eating all symptoms that can be misinterpreted with other common (non ) gynecological diseases. Symptoms of this type arise slow ly and sometimes go unnoticed until the disease ha s made severe progress Factors that increase the risk of ovarian disease include obesity, smoking, the use of estrogen as a postmenopausal hormone and certain mutation s in the BRCA1 and BRCA2 gene 1 Ovarian cancer is usually treated with a combination of surgery and chemotherapy. Surgical removal usually results in the complete removal of the ovaries and the fallopian tub es. As can be seen in Figure 1 2 the success of treatment is highly dependent on the stage in which the patient is diagnosed. Give n the same treatment, the 5 year survival rate of patients below 65 is nearly double the rate of patients over 65 (57% compared to 29% 4 ). Table 1 1 tabulates how ovarian cancer is staged by the American Joint Committee on Cancer (AJCC) 5 The five year s urvival

PAGE 20

20 rate of patients diagnosed at stage I is 90%, but only 2% of the patients with ovarian cancer are diagnosed at that stage 3 Diagnosis of Ovarian Cancer Early diagnosis of ovarian cancer is imperative towards the effective treatment of this disea se as illustrated in Figure 1 2 The vague symptoms of the disease make early diagnosis very difficult without additional aid, such as a transvaginal ultrasound (TVU) or serum levels of biomarkers assigned to the presence of ovarian cancer. Currently, patients complete a 27 item form over a period of several weeks as an initial diagnostic tool, called the Ovar ian Cancer Symptom Index (OCSI) 6,7 If symptoms persist, a TVU is then performed. An OSCI is considered to be positive when the listed symptoms appear more then 12 times in a single month and these symptoms first appeared within the previous year. In addition, t he use of longitudinal serum levels of various markers in combination with TVUs is currently b eing investigated 8,9 As previously above a transvaginal ultrasound is an important tool for early detection of tumors or benign swellings 10 However, i n 1996 the U.S. Preventive Services Task Force (USPSTF) recommended against the use of TVU for preventive screening as the test is not suf ficiently specific 11 and the cost paid ($550, national average 12 ) is not justified, considering the marginal reduction in mortality Biomarkers Biomarkers in Ovarian Cancer It is thought that the genetic changes causing carcinogenesis can be translated i n the release of several proteins, called biomarkers, for example: t he sensitivity of the OCSI (combined with transvaginal ultrasonography) can be enhanced when cross referenced with the patient CA125 blood levels. CA125 is the only FDA approved

PAGE 21

21 biomark er for ovarian cancer, but it is heterogeneously expressed: Some types of ovarian cancer do not express this glycoprotein, while in some other non cancerous situations like pregnancy or cirrhosis elevated levels of CA125 are known to exist in the blood o f the patient 13,14 The use of CA125 alone gives a sensitivity (m easures the proportion of truly positive test results ) of 48% at 98% specificity (m easures the proportion of truly negative test results ) When the test for early diagnosis was expanded wit h more markers, sensitivities of 75% were reached 15 Other possible ovarian cancer biomarkers include HE4, mesothelin, M CSF, osteopontin, kallikrein(s), and soluble EGF receptor 9, 15 An Introduction to Biomarkers a specific biological state, particularly 16 F rom a historical point of view, a biomarker could be a physical trait, a measurable biomolecule or a process. In recent years, the word biomarker has become synonymous with a molecular marker that indicates biological state. Biomarkers can take many forms, e.g., karyotype in a cell, methylation profile, the number of circulating cancer cells or copy num ber on a gene 17 20 Bi omarker can also be a metabolite related to the disease 21 Protein detection has become the gold standard in the biomarker field 16,22 Also, protein level s or presence can be easily measured in the blood, enabling researchers to directly correlate the efficacy of novel drugs to the biomarker level 24 Therefore, considerable effort in the field of proteomics has been focused on discovering novel protein biomarkers that are up or down regulated in cancer. However diagnostic prot ein measurements also face many challenges, such as heterogeneous protein expression, the wide possible concentration range of the putative biomarker in the

PAGE 22

22 blood, or the low abundance of the biomarker in the blood 16 These problems make protein biomarker discovery difficult and uncertain, and many putative biomarkers have been discarded due to poor analytical precision. Membrane Proteins One class of proteins that can prove outstanding as biomarkers are those associated with the membrane 25 This class of proteins is one of the most understudied and underestimated class studied to day. Membrane proteins can be divided in two main classes: 1) integral or transmembrane proteins that pass through the membrane multiple times and which are often found as por ins or signaling proteins ; and 2) peripheral or anchored proteins extend from the membrane, often extracellularly. The majority of this discussion will be on integral proteins, which can be further subdivided in two group s: the barrels and helic es helic es are the most prevalent group estimated to comprise 20 30% of most genomes 26 with the barrel group representing about 2 3% 27 Membrane proteins have a wide ra nge of functions: 1) transportation of ions or metabolites across the membrane; 2) transduction of chemical signals from their environment to initiate intracellular responses, 3) propagation of electrical signals, 4) cell attachment and 5) the control of membrane of lipid composition, vesic ular transpo rt, and organizing the shape of organelles within the cell itself 28 To become cancerous, cells often must change their membrane protein footprint to recruit growth factors, invade foreign tissues or evade apoptosis 29 31 Membrane proteins are in their n ative state s only when intercalated in the membrane and they are thus often insoluble in aqueous buffers due to sections of non polar lipid associating amino a cids. Thus, methods such as two dimensional gel

PAGE 23

23 electrophoresis or 2D LC MS /MS are required to s earch for potential biomarkers 32 33 As these proteins are predominantly found in the membrane, limitations of these analytical techniques make it difficult to obtain a good overview of the exact protein content of a membrane. There are three characteri stic problems associated with membrane protein analysis. First membrane proteins are general ly in low abundance, making it difficult to detect these proteins in standard gels with current extraction methods. This problem can be circumvented by various e nrichment techniques 34 Another problem is the limited aqueous solubility of membrane proteins In 2D gel electrophoresis, a technique in which a protein sample is separated by iso electrical focusing followed by SDS PAGE membrane proteins commonly precipitate at their iso electrical point s. Also, due to the limited solubility, surfactants are often used to increase th e amount of protein in solution, but these chemicals are not compatible with mass spectrometry, because some ionic surfactants or sal ts can suppress the ionization of peptides. T he final is the alkalinity of membrane proteins and the absence of ionic peptides in general 35 Since m ost 2D gels focus on analyzing proteins with a pI<8, proteins with higher pI are simply not detected. The LOD of mass spectrometry lays a thousand fold higher for unknown protein detection 40 Putting all these factors together, membrane proteins are underrepresented in global large scale proteomics studies 35 Other Biomarker Discovery Techniques Alterna tives to high throughput mass spectrometry techniques can also lead to the discovery of no vel biomarker leads such as immuno affinity capture techniques 36 (protein microarrays 37 ) or transcriptional profiling methods ( e. g., the analysis of transcriptomes in blood for pharmacological analysis 38 )

PAGE 24

24 Immuno affinity capture is currently considered as the most effective method for the detection and quantification of pu tative protein biomarkers. These techniques can be used for the detection of proteins at the ng/mL level commonly observed for biomarkers (with the help of signal amplification techniques through enzymatic amplification ) 39 New or unknown biomarkers are difficult to be detected with this approach as the detection of proteins through immune affi nity require knowledge about the analyte and as the production of antibodies is a lengthy process, many new promising proteins (which have a lo w incidence due to their novelty) do no t always have (good) antibodies available 41 Therefor e the search for al ternative capturing agents has been an important research area Some alternatives for antibodies can be found in peptoids 42 phage display 43 and in aptamers 44 ,45 Aptamers An Introduction to Aptamers Aptamers are oligo nucleotides usually ssDNA or ssR NA. Although r esearchers have developed some unusual aptamers with a peptide backbone 46 or with an extra unnatural base 47 The typical length of an aptamer is 20 100 nucleotides, which assume very distinctive and unique tertiary structure 48 Aptamer s w ere first described by Gold and Tuerk 49 and were also independently developed by Stozak and Ellington 50 who coined the aptamer The word is composed of two Greek Aptamers are developed from large random libraries of oligo nucleotides each containing 20 to 60 random nucleotides flanked by primers sequences (which a re needed for amplification by PCR) Target sequences are selected for a target of choice t hrough a pr ocess called Systematic Evolution of Ligands by EX ponential enrichment (SELEX). Their tertiary

PAGE 25

25 structure s allow them to bind to a wide variety of targets, from metal ions 51 to organic molecules 52 peptides 53 and proteins 54 Often, aptamers are selected f or the p roteins on the surface of viruses 55 bacteria 56 and eukaryotic cells 57 Aptamers bind to their targets with specificities and affinities comparable to antibodies. There are several other differences between antibodies and aptamers that change th e ways in which they are used For in vivo applications for example, antibodies have a relative ly long half life, while aptamers, when unaltered, are relatively quickly removed by DNases or RNases and the kidneys 58 59 However, because of the ease of chemical modification, the half life can by extended and fine tuned 59 60 By adding polyethylene glycol segments the size can also be fine tuned, ultimately affecting how the aptamer is cleared from the system by the first pass effect (renal filtration a nd liver filtration) 59 Other possib le modifications are to thiolation or methylat ion of certain hydroxi des on the ribose sugars to hamper the hydrolytic action of DNases or RNases 48,6 1 Another difference between antibodies and aptamers are in the way t hey are generated which allows selection of aptamer against i.e. toxins In order to make any antibody, the antigen of interest needs to be injected in an animal, after which the T cell that makes the antibody can be transformed into a hybridoma 62 SELE X is a procedure that is completely in vitro and automatable, which can reduce the development time of an aptamer in the hands of well trained selectors. This also allows easy scale up to the commercial level. The affinity of antibodies is also something that is hard to optimize (especially for low incidence antigens 41 which is why it may be challenging sometimes to find the right vendor for an antibody that can bind to the antigen of interest. Aptamers themselves can be designed to bind with a high dis sociation constant, and recently it is

PAGE 26

26 even possible to precisely design the selection process to match the affinity constant in 63 It is also assumed that aptamers are less prone to give immunologic responses in time as protei ns like antibodies are known to be doing 61 Many research groups and corporations in the world are actively investigating aptamers Currently, there is o ne aptamer on the market, Pegaptanib sodium, for the treatment of age related macular degeneration 6 4 and several other aptamers are now being in phase two and three clinical trails 45 It is estimated that the current aptamer therapeutic market was $10 million in 2009 I t is predicted that by this growth rate this number will grow to $1.2 billion in 2014 65 SELEX As stated in the introduction, aptamers are generated from large libraries that are prepared by solid phase chemistry 66 Because each library is consisted of a segment containing a string of 20 40 random nucleotides, flanked by primers for PCR amplification, the library contains in theory 4 20 4 40 ( 10 10 10 24 ) different possible sequences or femto to millimoles of unique molecules. Since each sequence form s its own unique tertiary structure, one library can be used to identify aptamer candidates for almost any target of interest. The basic scheme of SELEX is explained in Figure 1 3 In the initial round, usually 20nmol of nave library is melted at 95C and allowed to refold at the binding temperature (usually 4C). This ensure s that all strands fold in to their optimal tertiary structure s This mix is incubate d with the target of interest. Non binding sequences are removed by washing, after which the binding sequences are recovered by denaturing the target. In cell SELEX, o ne may choose to do a counter or negative selection against a non target cell line for additional stringency. However, usually this step is

PAGE 27

27 omitted in the first round in the SELEX process. When negative selection is introduced, only those seq uences that do not bind to the negative target are gathered and prepared for amplification. The pool that is retrieved after incubation with the target or counter target is amplified by PCR and prepared for the next round, leading towards the next generat by flow cytometry and when the desired enrichment is reached, the selection is stopped and one or a couple of pools are analyzed for sequence homology. The largest families obtained will be the primary aptamer candidates, which are further tested for binding with the target. Aptamer Target Recognition Interactions Aptamers bind to their targets via a combination of possible molecul ar interactions, resulting in affinities higher tha n their natural counterparts (riboswitches 68 ) and comparable to that of antibodies 45 There are in princip le two kinds of oligo nucleotides that can perform target recognition: naturally raised oligonucleotides or ribozymes and in vitro selected aptamers. The comparison between these two classes of ligands has shed light on the principles that underlie aptamer target recognition interactions 67 Since natural nucleic acids ne ed to perform a certain action as part of a larger network it is sometimes necessa ry for different structural motifs to be recognized. However, aptamers are selected to bind to one specific target though. For this reason, aptamers tend to have lower dissociation constant s (i.e., higher affinities) than their naturally raised counterpa rts. It is assumed that when aptamer s are freely flow ing in solution, they often exist as unpaired disordered loop regions 68 When the aptamer binds to its target, it assumes a very distinctive structure that wraps in a very specific way around

PAGE 28

28 its target. In some oligonucleotide peptide binding studies, it has been observed that unpaired nucleotides become flaps that cover the ligand 69 Thus, an aptamer can bind to its target with high affinity, and also with high specificity. A classic examp le of this is the inability of the theophylline aptamer to bind to caffeine, because the methyl group on caffeine induce s steric hindrance that prevents the apta mer from binding 70 With this example it is apparent that the binding of aptamers is not merel y dependent on t were true caffeine could very easily bind to the aptamer by intercalation ( or to any piece of oligonucleotide for that matt er ) as do common DNA imagining dyes do; e. g. ethidium bromide 71 In the e xample of theophylline and caffeine the binding of the aptamer with theophylline is governed by hydrogen bonds in a specific S loop, with further stabilization by 7 0 Aptamers fold into controlled conformations that allow binding to their se lection target s via a variety of possib le intermolecular interactions. H ydrogen bonding is a n important factor toward the selectivity of target specific aptamers, and has been illustrated in se veral examples of aptamer/ small molecules interactions (e.g ., arginine, a denosine mono pho sphate, flavin mononucleotide) 72 74 T he increased complexity of protein s allows for multiple stacking, shape conformations and electrostatic interactions 75 77 It is esp ecially important to consider electrostatic interactions when selecting for aptamers, as oligonucleotides have a phosphate backbone: it is important to introduce the right amount of competition from non specific poly anions during the selection process to increase the selective evolutionary pressure towards specific interactions 78

PAGE 29

29 Cell SELEX Principle In cell SELEX, a library is enriched for a cell line 79 xenograft 80 or tissue 81 As described above, i t is very common for a counter selection to be perfo rmed with healthy tissue or a non target cell line. In the end, the hypothesis is that the cells for positive selection have markers expressed that are not found on the negative (counter) cell line. T herefor e aptamers are also useful tool s for the study of membrane protein expression in cancer, as many cancers have membrane proteins that are uniquely expressed. Cell SELEX is performed in the same way as normal SELEX is performed, but the main difference is that libraries are e nriched for a complex targe t. Because of this, an enriched pool can have monoclonal aptamers for a wide array of membrane associated molecules, including proteins and their posttranslational modifications (lipids, poly or oligo saccharides). Similar studies have been performed on the proteome of human plasma 82 The enriched pool will thus contain a wide array of aptamers that can bind to different proteins. By analyzing the pools by cloning, or next generation sequencing, it is possible to find the aptamers that bind to the mo st significantly different target markers, result ing in a panel of aptamers that can distinguish the positive cell line from the negative cell line. Applications of aptamers derived from cell SELEX There are several possible applications many of which cor respond to antibody based applications, for aptamers selected for cells. One obvious application is the use of th is technique on cells that over express a certain type of membrane protein for drug delivery purposes and tissue profiling. The classic exampl e of such an aptamer is the tenascin c aptamer. In this example, the parallel selection of aptamers in compared

PAGE 30

30 between cell lines that are known to express tenascin c and a cell line that has been made to express the protein. The tenascin c aptamer that this study yielded could be used for cancer tissue targe ting, as tenascin c is an extra cellular matrix protein often over expressed in cancer 83 Although Cell SELEX is u sually performed in vitro in vivo selection is also possible In the work conducted by Mi et al. a mouse xenogra ft was targeted with a nuclease resistant RNA pool, which yielded an aptamer for p 68, an RNA helicase that allows for xenograft homing 81 The interesting observation is the unusual location of the se lected protein for this ap tamer. A lthough p68 has a known function in the nucleus of the cell, the elucidated protein was also found in the cell membrane C ell SELEX will become a very potent technique for the study of cell membrane s as current methodologies would discard the unexpected protein based on previous assumptions about membrane proteins, cytosolic proteins and their sub cellular location 84 A similar example can be found in nucleolin, another RNA helica se that is found on the surface of many cancer cell types 85 A nother application of cell SELEX is : the identification of protein associated at the membrane of cells. Researchers have demonstrated that it is possible to perform subtractive SELEX against virus infected versus uninfected cells 86 and different lineage s of hema topoietic cells or solid tumors 79 87 88 For some of these selections, the target identification of these aptamers can lead to the identification of prot eins that play a role in the differentiation state of cells, or in the infected state of cell s. The identification usually occurs through a method where the aptamer is used as the ligand

PAGE 31

31 in an affinity column, in the cell lysate or on the living cell 81 89 91 The captured protein is then analyzed by standard mass spectrometric protein identifica tion. The identification of proteins that these aptamers bind can have several possible appl ications. The classic example is the aptamer that is FDA ap proved for the treatment of age related macular degeneration, branded Macugen by Gilead Sc iences. Thi s aptamer has anti angiogenic properties as it binds to vascular endothelial growth factor (VEGF). It is an excellent example of what the possibilities are for aptamers towards changing their pharmacological profile by post SELEX modifications as the ap tamer is heavily modified by PEGylation 64 Several companies are developing aptamer based drugs that are now in clinical trails. Archemix is on the forefront by havi ng several aptamer in p hase II and phase III clinical trails. With their flagship aptamer, ARC1779, a phase II clinical trail was performed in patients with three types of Von Willebrand Factor Related Platelet Function Disorders 92 Ano ther phase II aptamer is ARC 19499, for the treatment of hemophilia 93 Archemix is al so aiding large pharmaceutical companies like GlaxoSmithKline (AS1411), Arca (anticoagulant aptamers) and Regalo Biosciences (anticoagulant aptamers) in developing aptamers for the treatment of several pathologies 94 Overview of the Dissertation In this di ssertation a panel of aptamers have been developed to bind to clear cell adenocarcinoma and ovarian cancer stem cells. Furthermore, a detailed descrip tion of each selection process, including aptamer characterization, is presented. The identity of the bi nding protein to one of the selected aptamers has been determined and the methodology for the identification process outlined. Finally, the oncological function of

PAGE 32

32 this protein is being investigated, along with the action of the aptamer on viability, prol iferation, migration and invasiveness of the cell line.

PAGE 33

33 Figure 1 1. Change of the US m ortality rate by cause from 1950 to 2001 The numbers represent the rate per 100,000. ( Figure based on data found in Vital Statistics of the United States, annual, Vol. I and Vol II I; 1971 2001 )

PAGE 34

34 Fig ure 1 2 Adenocarcinoma of the ovary (excluding borderline tumors): relative survival r ates (%) by AJCC Stage Ages 20+, 12 SEER Areas, 1988 2001 See Table 1.1 for the description of the stages (Adapted from Kosary C Cancer of the ovary. In: Ries L A G Young J L Keel G E Eisner M P Lin Y D ., et al. (editors) SEER survival monograph: cancer survival among adults: US SEER Program, 1988 2001, patient and tumor characteristics. Bethesda (Maryland): National C ancer Institute, SEER Program (2007) ) Copyright 2012. Kosary C. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, an d reproduction in any medium, provided the original work is properly cited

PAGE 35

35 Figure 1 3 The basic scheme for cell SELEX

PAGE 36

36 Table 1 1. The description of the different stages of ovarian cancer according to the American joint committee on cancer ( AJCC ) AJCC stage Description Stage I Ovarian cancer is growth limited to the ovaries. Stage IA Growth limited to one ovary; no ascites. No tumor on the external surface; capsule intact. Stage IB Growth limited to both ovaries; no ascites. No tumor on the external surfaces; capsules intact. Stage IC Tumor either stage IA or IB, but with tumor on the surface of one or both ovaries; or with capsule ruptured; or with ascites present containing malignant cells or with positive peritoneal washings. Stage II Ovarian cancer is growth involving one or both ovaries with pelvic extension. Stage IIA Extension and/or metastases to the uterus and/or tubes. Stage IIB extension to other pelvic tissues. Stage IIC tumor either stage IIA or stage IIB, but with tumor on t he surface of one or both ovaries; or with capsule(s) ruptured; or with ascites present containing malignant cells or with positive peritoneal washings. Stage III Ovarian cancer is tumor involving one or both ovaries with peritoneal implants outside the pelvis and/or positive retroperitoneal or inguinal nodes. Superficial liver metastasis equals stage III. Tumor is limited to the true pelvis but with histologically verified malignant extension to small bowel or omentum. Stage IIIA Tumor grossly limited to the true pelvis with negative nodes but with histologically confirmed microscopic seeding of abdominal peritoneal surfaces. Stage IIIB Tumor of one or both ovaries with histologically confirmed implants of abdominal peritoneal surfaces, none exceeding 2 centimeters in diameter. Nodes negative. Stage IIIC Abdominal implants greater than 2 centimeters in diameter and/or positive retroperitoneal or inguinal nodes. Stage IV Ovarian cancer is growth involving one or both ovaries with distant metastasis. If pleural effusion is present, there must be positive cytologic test results to allot a case to stage IV. Parenchymal liver metastasis equals stage IV. Copyright 2012. Kosary C. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited

PAGE 37

37 CHAPTER 2 SELECTION AND CHARACTERIZATION OF OVARIAN CLEAR CEL L ADENOCARCINOMA CELL APTAMERS Introduction: Ovarian Clear Cell Adenocarcinoma Ovarian cancer is the most lethal gynecological malignancy, and the ovarian clear cell adeno carcinoma subtype (OCCA) demonstrates a particularly poor response to standard treatment 95 OCCA is the second most prevalent histological subtype of ovarian adenocarcinoma (4 25% of ovarian adenocarcinomas) which accounts for almost 90% of all ovarian cancers T he largest histological group is serous ovarian adenocarcinoma (SOAC) ( 32,2% total) 4 Although ovarian cancer shows generally good response to standard treatment with platinum based drugs OCCA does not respond to these drugs, resulting in poor prognosis for OCCA patients 95 Clinically, OCCA and SOAC behave distinctly different Some notible fea tures of OCCA include it s rare bilateral occurrence its large pelvic mass presence and its association with endometriosis, thromboembolic complications and hyper calcemia. Li ke serous adenocarcinoma, early staged OCCA (e.g., stages I and II) exhibits a better survival rate then stage II I or IV OCCA. Of particular concern is the survival rate of OCCA, which is much lower than that of serous adenocarcinoma (median survival rate in serous adenocarcinoma is 3 to 4 years versus about a year in OCCA at stage I/II). Since OCCA hardly respond s to platinum based therapy different therapeutic targets are needed for OCCA in comparison to SOCA 95 97 Some potential candidates for the treatment of OCCA are tyrosine kinases, and inhibitors for the PI3K AKT mTOR pathway 98 Studies to make these inhibitors approved drugs for cancer therapy are underway, but still more drug candidates are needed as these approaches need to be proven safe 95

PAGE 38

38 With the current system of FDA approval through clinical trails a new and faster way to discover targets for cancer therapy in OCCA (and other diseases) is needed. It has been proposed that aptamers, or disease cell enriched libraries can lead towards the identification of new disease markers 9 9 In Cell SELEX, the majorit y of the possible targets are proteins, and when a library pool is generated against a disease cell, more information on the proteins expressed on the surface of that disease cell can be generated. Improvements in ovarian cancer outcomes, especially for O CCA, can be expected with a clearer understanding of the molecular pathology to guide strategies for earlier diagnosis and more effective treatment. This chapter describes the SELE X on the OCCA cell line TOV 21G 100 using the cervical cancer cell line HeLa 101 for counter selection Bi ochemical and biophysical properties of the selected aptamers are also studied. Materials and Methods Instrumentation and Reagents All oligonucleotides were syn thesized by standard phosphoro amidite chemistry using a 3400 DNA synthesizer (Applied Biosystems) and were purified by reversed phase HPLC (Varian Prostar), using a C18 column in 0,1M Triethylamonium acetate in water/acetonitrile gradient All PCR mixtures contained 50 mM KCl, 10 mM Tris HCl (pH 8.3), 2.0 mM MgCl 2 dNTP s (each at 2.5 mM), 0.5 mM of each primer, and Hot start Taq DNA polymerase (5 U /mL) (TaKaRa). PCR s were performed on a Biorad Thermocycler. The m onitoring of pool enrichment, characterization of the selected aptamers, and study of the target protein assays w ere performed by flow cytometry using a FACScan cytometer (BD Immunocytometry Systems). Trypsin and Proteinase K were purchased from Fisher Biotech. The DNA sequences in the pools were determined

PAGE 39

39 by the Genome Sequencing Serv ices Laboratory at the University of Florida with the use of 454 sequencing (Roche). Clustal X was used for analysis of sequence homology for sequence selection Cell Culture and Buffers All cell lines were obtained from the American Type Cell Culture (A TCC). The CAOV 3 and TOV 21G ovarian cancer cell lines where maintained in culture with MCBD 105: Medium 199 (1:1); the HeLa cell line was cultured in RPMI 1640; and the med ia where supp lemented with 10% FBS and 100 IU /mL Penicillin Streptomycin. Other cell lines used for selectivity assays included CEM (T cell acute lymphoblastic 116, DLD 1, HT 29 (colorectal adenocarcinoma), NCI H2 3 (non small cell lung cancer) and A172 (glioblastoma), all of which were cultured according to ATCC recommendations All cell lines where incubated at 37C in a 5% CO 2 atmosphere. During the selection, cells were washed before and after incubation with wa sh ing buffer (WB), containing 4.5 g/L glucose and 5 mM MgCl 2 CaCl 2 and MgCl 2 (Sigma). Binding buffer (BB) used for selection was prepared by adding yeast tRNA (0.1 mg /mL) (Sigma) and BSA (1 mg/mL Fisher) to th e wash ing buffer to reduce non specific binding. SELEX Library and Primer Design A typical DNA library for SELEX constitutes a random nucleotide region, flanked with primers that allow for amplification of the pool after incubation with the target. The pr imers are designed so that so called molecular parasites are reduced. Molecular parasites are found in unoptimized libraries, where the primers can interact with

PAGE 40

40 themselves in stead of the ir complement in the template 48 The p rimers in this thesis were o ptimized so that the T m differ less then 1C, and to minimize unwanted elongation, the annealing temperature maximum was 80% of the elongation temperature (55 60C). The primers were not allowed to form hairpins, and only three Watson Crick base pairs were permitted between the different primers. Finally, the complement of the reverse primer and the forward primer were not allow ed to form hairpins. The optimized primers were chosen using mfold software 102 The chosen primer pair was incorporated in to a library, with each component having a ssDNA strand that is synthesized as the forward primer, followed by the random sequences and ATC CAG AGT GAC GCA GCA (N) 40 TGG ACA CGG TGG CTT AG T synthesized using standard solid phase synthesis and HPLC purification as described above The forward and reverse primer were biotin respectively In Vitro Cell SELEX In this study, TOV 21G w as used as the target cell line and HeLa was used for counter selection. For the first round, the cells were incubated with 20 nmol of nave ssDNA library dissolved in BB. For later rounds, 50 pmol of enriched pool were used for incubation, also dissolved in BB. Before incubation, the ssDNA pool was denatured by heating at 95C for 5 minutes and was placed on ice for 5 minutes, allowing each sequence to form the most stable structure. The cells were washed twice (2 minutes) with WB and incubated with the DN A pool on ice in an orbital shaker for 30 minutes. In later selection rounds, the washing stringency was increased to remove weakly binding sequences (a larger number of washes and increased washing time, up to 5 minutes).

PAGE 41

41 The bound sequences were eluted i n 500 mL WB by heating at 95C for 10 15 minutes, cool ing on i ce for 5 minutes and centrifuging at 14,000 rpm for 2 minutes. The supernatant which contained the binding sequences was then incubated with a negative cell line to remove gener al sequences, as described in Figure 1.4 The remaining sequences were amplified by PCR using the FITC and biotin labeled primers. Amplifications were carried out at 95C for 30s, 60C for 30s, and 72C for 30s, followed by final extension for 3 minutes at 72C. The selected sense ssDNA was separ ated from the biotinylated antisens e ssDNA by streptavidin coated S epharose beads (Amersham Bioscience). The ssDNA was eluted from the streptavidin beads by melting the dsDNA in a 0.2M NaOH solution. The ssDNA is then desalted with the use of a G25 Sephad ex size exclusion column (GE healthcare). The enrichment of specific sequences was assayed using flow cytometry as explained below When the level of enrichment was satisfactory, pools of interest were submitted for sequencing. Sequencing and Selectio n of Putative Aptamers In this study, next generation sequencing was used for the analysis of enriched pools. Several pools were labeled with a specific sequence with the use of PCR, so enrichment of sequences could be monitored later o n. More specifically, 454 sequencing was used, which yielded about 6000 sequences per run. Since it was foreseen that the field of next gen sequencing will make drastic steps in the future 103 a script was written in PERL 104 (see appendix A) for the high throughput trimming of the primers prior to the processing of actual aptamer sequences for sequence homology. Sequences that fulfilled the requirements (data flanked with the primer pair an d minimum 70 nt long) were analyz ed for homology in ClustalX 105

PAGE 42

42 Affinity Studies: Flow Cytofluorometric Analysis for the Determination of Binding Affinity To determine the binding affinities of the aptamers and the enrichment of the pools the target cells (5 x 10 5 b iotin labeled aptamers or FITC labeled pools on ice for 30 minutes in 100 mL of BB. Cells were then washed twice with 500 mL of BB, and suspended in 100 m L of BB containing streptavidin PE Cy5.5 at appropriate dilution Cel ls were then washed twice with 5 00 mL of WB, and then suspended in 200 mL of BB for flow cytometry biotin labeled random sequence as the negative control. All the experiments for binding assays were repeated at least twice The specific binding intensity was calculat ed by subtraction of the mean fluorescence intensity of the background binding from the mean fluorescence intensity of the aptamers. The apparent equilibrium dissociation constant (K d ) of the fluorescent ligand was obtained by non linear regression of the specific binding intensity (Y) and the aptamer concentration (X) to the equation Y=B max X/(K d +X) using SigmaPlot. (Jandel, San Rafael, CA). Selectivity and Specificity To determine the cell specificity of the selected aptamers, cell lines including HeLa, K 562, H23, H69, A172, HL 60. HT 29, Ramos and CEM were used in binding assays by flow cytometry as described above. Effect of Temperature on Aptamer Binding The aptamer selection process and all of the binding assays were performed on ice. It has been obs erved that some (one out of dozens selected to date from the Tan lab) of the aptamers selected at lower temperatures may not bind well at 37C, leading to poor performance under physiological conditions 89 In order to verify binding stability,

PAGE 43

43 aptamers wer e incubated with the target at 37C, and fluorescence intensity was determined by flow cytometry. Aptamers incubated at 4C were used as the positive control. Protease Digestion Assay Target cells (5x10 5 ) were detached using a non enzymatic cell dissociati on solution (Sigma) After suspending the cells the y were washed with 3 mL of PBS and then incubated with 1 mL of 0.05% trypsin/0.53 mM EDTA in HBSS or 0.1 mg/mL proteinase K in PBS at 37C for 1, 5, 15, 30 and 60 minutes. Pure FBS w as added to quench the protease reactions After washing with 2 mL of BB, the treated cells were used for binding assays as described above. Results Monitoring of Pool Enrichment for TOV 21G against HeLa To start the s election process, 20 pmol of na ve library was enriched by sequential binding to TOV 21G cell monolayers. Sequences showing non specific binding to general cell surface markers were removed by incubating the enriched pool with HeLa cells (rounds 2, 4, 5, 7, 8, 9, 12, 20, 21, 22). The eluted pool for each round of SELEX was amplified through semi quantitative PCR. Special care was given to prevent unwanted amplification of molecular parasites A n example can be found in Figure 2 1, in the bands for 18 and 20 PCR cycles As the figure shows, after 16 rounds of PCR unwanted amplification occured Therefor e, a large r scale (1mL vs 50L) PCR was repeated at 16 rounds in order to maximize yields. This procedure was followed throughout the entire SELEX procedure When counter selection in rounds 13 to 19 was omitted, re enrichment for some HeLa binding sequences occurred (Figure 2 2) The sequences binding to HeLa cells were successfully removed by reintroducing

PAGE 44

44 counter selection in subsequent rounds, while the enrichment towards the target cell line was maintained A fter 22 rounds of SELEX, an enriched pool that specifically bound to the model OCCA cell line, but marginally to HeL a cells, was obtained (Figure 2 2). This procedure produced a pool enriched for sequences binding to surface markers expressed by the model OCCA cell line, but not by the cervical cancer cells. Aptamer Characterization A fter 22 rounds a pool with satisfactory ( i.e. specific binding to target) enrichment was obtained and pools were selected for sequencing after which aptamers were chosen fro m the sequencing data based on the homology and size s of the families As observed in a SELEX experiment on plasma ( another complex selection target ) observable enrichment is not always necessary to obtai n sufficiently large homologous families for aptam er choice 82, 106 It was hypothesized that 454 sequencing would give the resolution needed to find aptamers at an early stage. In other words, the information given by 454 sequencing of a pool was hypothesized to give a good representation of the actual pool under investigation. T he frequency of selected sequence fa milies is summarized in Table 2 1. Pool 13 was submitted for sequencing, as it was the first pool with a slight fluorescence intensity increase in flow cytometry Pool 22 was also submitted for analysis because it was the final pool and pool 21 was submitted to study the possible changes that can was only marginal. As can be seen in the po ol, only a small change was observed in the frequency of the presented aptamers between pool 13 and 22 and the frequency seemed to be independent of the binding (i.e. aptTOV6 does decrease, which could

PAGE 45

45 suggest that HeL a may have AptTOV6 binding motifs; ho wever the aptamer binds specifically to TOV 21G, but not to HeLa). From the alignment data, 9 sequences (and one extra sequence for aptTOV2 that chosen and characterized by protease digestion assays and affinity det ermination as described in the M ethods and M aterials section. Table 2 2 contains the sequences of the aptamers with their respective dissociation constant s (K d ) and frequencies in pool 22. The K d s for the selected aptamers were all in the picomolar to t he nano molar range. Especially aptTOV1 showed tight binding with the cells with a K d of 250 80pM (Figure 2 12). It is interesting to observe that the aptamer s with sub nano molar affinities were enriched relatively rapidly (Table 2 1). All the selected aptamers were found in all the pools and it was observed that they could bind to TOV 21G but not to HeLa, i.e., with t he same binding profile as their respective pool (Figure 2 13 illustrates this for aptTOV1) More binding studies were performed on an array of different cell lines from different types of cancer, and the results are tabulated in Table 2 3. Some basic characterization was performed by protease treatment studies, w h ere it was found that all the targets for the aptamers wer e cleaved of by proteases (Fi gure 2 3 to Figure 2 11). The protease mediated removal of the fluorescence implies that the target of the aptam er is associated with a protein. H owever, it does not specify the exact epitope of the target, which could be a p ept ide chain or a terminal oligosaccharide 89,91,107 Concluding Remarks A series of aptamers with high affinity for OCCA (TOV 21G) have been selected that can distinguish ovarian cancer from cervical cancer (Figure 2 13) In particular, AptTOV 1 showed ve ry high affinity towards TOV 21G, with a K d of 250 pM. Given the

PAGE 46

46 limited number of biomarkers for ovarian cancer currently available, the aptamers obtained from these selections have potential for improving diagnosis and treatment of this deadly disease. Because the aptamers also bind benign cysts (Table 2 3), the aptamers cannot be used to identify ovarian cancer per se (more binding studies are warranted) However, since the aptTOV aptamers do not bind to a cancer of similar etiology (CAOV3) and also no t to HeLa, they still have the potential to provide more insight into the pathology of ovarian cancer, and to become a valuable tool for distinguishing OCCA from other ovarian cancer subtypes. It has been observed that there are significant differences in the proteome of serous and clear cell ovarian cancer 108 and the need for oncologists to distinguish clear cell adeno carcinoma from serous ovarian adenocarcinoma was re cently unequivocally established to be a genuine need 95 The targets for thes e aptamers are most likely down regulated or silenced in these two cell models. Additionally, the a ptTOV aptamers show binding to cancer cell lines from differe nt non related cancers (Table 2 3) and some a ptTOV aptamers also bind CEM cells. This result suggests that the aptamers obtained from this SELEX can be used for profiling the expression of membrane proteins of different cancers. Identifying the targets of the selected aptamers is expected to shed light on the underlying mechanisms and pathways involved of the s e deadly disease s 109

PAGE 47

47 Figure 2 1. Sample agarose gel of a semiquantitative evalu ation of pool enrichment. L: 25bp ladder, : no template, the numbers stand for the number of PCR rounds

PAGE 48

48 Figure 2 2. B inding assay of the enriched pools with TOV 21G and HeLa cells. A ) E nrichment with TOV 21G cells B) M arginal binding of the respective pools to HeLa cells. By doing counter selection, sequences binding to HeLa were removed Copyright 2010 Van Simaeys. This is an open access article di stributed under the Creative Com mons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited

PAGE 49

49 Figure 2 3. Bi nding profile of aptTOV1 (Red: TOV21G; Green: Library ; Dark blue: 4C for 30min; Orange: 37C; Light blue: 37C after 30 minutes 0.05% trypsin treatment )

PAGE 50

50 Figure 2 4. Bi nding profile of aptTOV2 (Red: TOV21G; Green: Library ; Dark blue: 4C for 30min; Orange: 37C; Light blue: 37C after 30 minutes 0.05% try psin treatment )

PAGE 51

51 Figure 2 5. Bi nding profile of aptTOV3 (Red: TOV21G; Green: Library ; Dark blue: 4C for 30min; Orange: 37C; Light blue: 37C after 30 minutes 0.05% trypsin treatment )

PAGE 52

52 Figure 2 6. Bi nding profile of aptTOV4 (Red: TOV21G; Green: Library ; Dark blue: 4C for 30min; Orange: 37C; Light blue: 37C after 30 minutes 0.05% trypsin treatment )

PAGE 53

53 Figure 2 7. Bi nding profile of aptTOV5 (Red: TOV21G; Green: Library ; Dark blue: 4C for 30min; Orange: 37C; Light blue: 37C after 30 minutes 0.05% trypsin treatment )

PAGE 54

54 Figure 2 8. Bi nding profile of aptTOV6 (Red: TOV21G; Green: Library ; Dark blue: 4C for 30min; Orange: 37C; Light blue: 37C after 30 minutes 0.05% trypsin treatment )

PAGE 55

55 Figure 2 9. Bi nding profile of aptTOV7 (Red: TOV21G ; Green: Library ; Dark blue: 4C for 30min; Orange: 37C; Light blue: 37C after 30 minutes 0.05% trypsin treatment )

PAGE 56

56 Figure 2 10. Bi nding profile of aptTOV8 (Red: TOV21G; Green: Library ; Dark blue: 4C for 30min; Orange: 37C; Light blue: 37C after 30 minutes 0.05% trypsin treatment )

PAGE 57

57 Figure 2 11. Bi nding profile of aptTOV9 (Red: TOV21G; Green: Library ; Dark blue: 4C for 30min; Orange: 37C; Light blue: 37C after 30 minutes 0.05% trypsin treatment )

PAGE 58

58 Figure 2 12. K d determination for aptTOV1 C ells were incubated with varying concentrations of PE Cy5 labeled aptamer in duplicate. The fluorescence intensity originating from background binding at each concentration was subtracted from the mean fluorescence intensit y of the corresponding aptame r Copyright 2010 Van Simaeys. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited

PAGE 59

59 Figure 2 13. S pecificity assa y for aptTOV1 on TOV 21G and HeL a. The aptamer binds well to TOV 21G, but poorly to the negative SELEX cell line HeLa. A ptamers were labeled with PE cy5.5 119 Copyright 2010 Van Simaeys. This is an open access articl e distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited

PAGE 60

60 Table 2 1 Selected aptamer frequencies in the analyzed pools Pool 13 (%) Pool 21 (%) Pool 22 (%) aptTOV1 7.56 1.91 2.53 aptTOV2 6.43 10.84 18.62 aptTOV2 a 2.96 4.92 7.65 aptTOV2 total 9.30 15.76 26.27 aptTOV3 4.05 7.97 8.74 aptTOV6 0.30 0.92 0.58 Copyright 2010 Van Simaeys. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited

PAGE 61

61 Table 2 2. C ompendium of the aptamers obtained by selection against TOV 21G Name Sequence K d (nM) Freq (%) aptTOV1 ATC CAG AGT GAC GCA GCA GAT CTG TGT AGG ATC GCA GTG TAG TGG ACA TTT GAT ACG ACT GGC TCG ACA CGG TGG CTT A 3 0.25 0.08 2.52 aptTOV2 ATC CAG AGT GAC GCA GCA TAA TCT CTA CAG GCG CAT GTA ATA TAA TGA AGC CCA TCC ACC TGG ACA CGG TGG CTT A 3 0.90 0.25 18.62 aptTOV3 ATC CAG AGT GAC GCA GCA CTC ACT CTG ACC TTG GAT CGT CAC ATT ACA TGG GAT CAT CAG TCG ACA CGG TGG CTT A 3 30 9 8.73 aptTOV4 ATC CAG AGT GAC GCA GCA GGC ACT CTT CAC AAC ACG ACA TTT CAC TAC TCA CAA TCA CTC TCG ACA CGG TGG CTT A 3 20 5 0.52 aptTOV5 ATC CAG AGT GAC GCA GCA CAA CAT CCA CTC ATA ACT TCA ATA CAT ATC TGT CAC TCT TTC TCG ACA CGG TGG CTT A 3 4.5 1.2 0.82 aptTOV6 ATC CAG AGT GAC GCA GCA CGG CAC TCA CTC TTT GTT AAG TGG TCT GCT TCT TAA CCT TCA TCG ACA CGG TGG CTT A 3 29 7 0.58 aptTOV7 ATC CAG AGT GAC GCA GCA CCA ACT CGT ACA TCC TTC ACT TAA TCC GTC AAT CTA CCA CTC TCG ACA CGG TGG CTT A 3 6.6 2.3 0.19 aptTOV8 ATC CAG AGT GAC GCA GCA CCA GTC CAT CCC AAA ATC TGT CGT CAC ATA CCC TGC TGC GCC TCG ACA CGG TGG CTT A 3 17 3 0.76 aptTOV9 ATC CAG AGT GAC GCA GCA GCA ACA CAA ACC CAA CTT CTT ATC TTT TCG TTC ACT CTT CTC TCG ACA CGG TGG CTT A 3 26 10 0.06 Copyright 2010 Van Simaeys. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited

PAGE 62

62 Table 2 3 O verview of the binding studies performed on multiple types of cell lines ( : no binding; + : less then a 5 fold signal increase; ++ : between 5 to 10 fold signal increase; +++ : between 10 to 100 fold signal increase; ++++ : more then 100 fold signal increase) Aptamer TOV 21G CAOV3 HeLa BCC H23 HT 29 HCT 116 A172 Ramos CEM HL60 DLD 1 aptTOV1 +++ ++ ++ ++ + aptTOV2 +++ ++ ++ ++ aptTOV3 +++ ++ + ++ ++ aptTOV4 +++ + ++ + ++ +++ + aptTOV5 +++ + + +++ ++ + ++ +++ +++ aptTOV6 ++++ ++ ++ ++ +++ ++ aptTOV7 +++ ++ ++ ++ +++ ++ aptTOV8 +++ ++ ++ ++ +++ + aptTOV9 +++ ++ ++ ++ ++ Copyright 2010 Van Simaeys. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited

PAGE 63

63 CHAPTER 3 METHOD DEVELOPMENT F OR APTAMER TARGET IDENT IFICATION Introduction Ovarian cancer is the most lethal gynecological malignancy, with an annual mortality rate of around 140,000 per year globally 110 The average relative survival rate for ovarian cancer is 43.9%, a numb er that can dramatically change depending on the stage (Stage IA: 94.0%; Stage 4: 17.9%) 4 Unfortunately, only 2% of all ovarian cancer is detected at stage I and better methods are needed for early detection 3 A method that is showing great promise to facilitate ea rly cancer detection is biomarker detection in the peripheral fluids 16 A cancer biomarker is understood to be a protein that is circulating in the blood, allowing simple blood tests to determine whether a patient has cancer by its presence or absense 22 However, there is a lack of good biomarkers that can be used for the early detection of ovarian cancer 111 So far, only CA125 is recognized by the FDA as a marker that may indicate the presence of ovarian c ancer but with limited specificity 10 In ovarian cancer, there is therefor e a need for more biomarkers, in order to increase specificity and sensitivity towards early detection of this malignancy. As described in Chapter 1, Systematic Enhancement of Ligands through EXponential enrichment of apt amers on living ce lls (Cell SELEX) is touted as a potentially imp ortant tool for finding ligands that bind to biomarker leads 44,89 91 and is becoming mo re and more a routine procedure 79,86 88 The most important advantage of cell SELEX is that no prior kn owledge about the molecular profile of the cell is required 60, 112 T he selected aptamers can be used in applications without furth er knowledge when a good target/ non target model is chosen.

PAGE 64

64 W hen an interesting aptamer is found, the question about the ide ntity of the aptamer target presents itself. Since approximately 50% of the mass of a eukaryotic cell membrane consists of protein, proteins are the prime candidates as aptamer selected cell antigens 113 Several research groups have attempted to solve th e problem of aptamer target identification, but progress has been hampered, as the identification of individual aptamers target is finicky and intricate Thus, there is still a clear need for an easy and robust method for elucidating the protein targets of an aptamer generated from Cell SELEX. Successful ventures towards such a technique have been performed on aptamers 89,91 and whole pools 90 These methods cannot be applied to all aptamers 91 and are laborious and expensive, as they require many optimization steps. Some of these optimization steps include the insertion of nucleotides that can crosslink with the target without loss of binding, which requires e xtensive optimization to retain the aptamer s bindin g abilit y 89 Various biochemical techniques are used to study the interactions between DNA and proteins, including Chromatin immuno precipitation (ChIP). In this technique, f ormaldehyde fixes the protein and DNA within 2 of their respective binding gro up s at the primary amines of both moieties T he protein is then immuno precipitated and the DNA is further analyzed by PCR 114 The reaction mechanism for the p rotein DNA crosslinking is described in Scheme 3 1 115 The advantage s of formaldehyde crossl inking include: the small s ize of the resulting crosslink which allows only interacting DNA and protein to react with each other and the ability to reverse the crosslink on c e the molecule of interest (usually a short piece of DNA) is extracted 116 This a llows for easy analysi s after the extraction

PAGE 65

65 and precipitation step. ChIP is a valuable tool for various genetic studies, as transcription factors can only be retrieved from the pieces of DNA (e.g., promotor regions) with which they interact 117 A n innovative recent technique has used the hybri di zation of genomic DNA to elucidate the proteins involved in the binding of specific loci. S everal proteins that play a role in telomere interactions were identified leading to the identification of protei ns involved in the alternativ e lengthening of telomeres 118 This study le d us to the hypothesis that studies of interactions between aptamers and their target s are feasible with formaldehyde crosslinking, and the aptamer c ould serve as a valuable capturi ng agent for its target in this manner Thus, implementing the basic princi ples of ChIP on aptamer protein interactions can be a very attractive strategy for the elucidation of the target of an aptamer. As described in Chapter 2, a series of aptamers h ave been selected against ovarian cancer, and they bind to an ovarian clear cell carcinoma cell line (TOV 21G), but not to cervical canc er cells (HeLa) 119 In order to acquire a deeper understanding about the underlying molecular differences between these two types of malignancies, it is important that the targets for the aptamers be elucidated. This chapter describes the method that was used to elucidate the target for aptamer TOV6 The target determined to be StIP1 by mass spectrometry, and the assignm ent was further validated bysiRNA knockdown and antibody binding studies with the help of flow cytometry. Materials and Methods Instrumentation and Reagents ATC CAG AGT GAC GCA GCA CGG CAC TCA CTC TTT GTT AAG TGG TCT GCT TCT TAA CCT TCA TCG ACA C GG TGG CTT A

PAGE 66

66 ( N 76 ) were synthesized by standard phosphoroamidite chemistry using a 3400 DNA synthesizer (Applied Biosystems) on biotin or desthiobiotin CPG for protein capt ure (Glen R esearch) and were purified by reversed phase HPLC (Varian Prostar) as described in Chapter 2 StIP1 antibody M33 (clone 2E1) was purchased from Abnova (Taiwan), and biotinylated with the EZ link Sulfo NHS LC Biotin kit (Pierce) The capturing of the aptamer hybrids was done using Dynabeads M 280 Streptavidin (Invitrogen). Cell Culture and Buffers The TOV 21G cell line was purchased from the American Type Cell Culture (ATCC) and was maintained in culture with MCBD 105: Medium 199 (1:1), supp lemented with 10% FBS and 100 IU /mL Penicillin Streptomycin. TOV 21G was cultured at 37C in a 5% CO 2 atmosphere. For aptamer binding, cells were washed after non enzymatic dissociation buffer treatment (Sigma) and after incubation with washing buffer (WB ) containing 4.5 g/L glucose and 5 mM MgCl 2 buffered saline with CaCl 2 and MgCl 2 (PBS Sigma). Binding buffer (BB) used for aptamer binding was prepared by adding yeast tRNA (0.1 mg/mL, Sigma) and BSA (1 mg/mL, Fisher) to the washi ng buffer to reduce non specific binding. All chemicals used in the buffers were purchased from Sigma, unless otherwise specified. For crosslinking, a 1% formaldehyde (Fisher) in PBS solution was used. The cell lysis buffer contained 2% Triton 100X (Fis her), 1.5% Nonidet (Fisher) and 0.5% c holate in dideionized water. For the washing of the magnetic beads, a 10mM HEPES NaOH buffer (pH 7.8) was used with 100mM NaCl, 2mM ethylenediaminetetraacetic acid (EDTA), 1mM ethylene glycol tetraacetic acid (EGTA) 0.2% sodium dodecyl sulfate(SDS) and 0.1% sodium lauroyl sarcosinate (SLS)(All from Sigma) (PI washing buffer) As an elution buffer,

PAGE 67

67 12.5mM biotin in 7.5 mM HEPES NaOH in 75mM NaCl, 1.5 % EDTA, 0.5%EGTA 0.15% SDS and 0.075% SLS was used. As a crosslin k reversal solution, 250mM Tris buffer (pH 8.8) with 2% SDS and 0.2M mercaptoethanol was used. All solutions, except for WB and BB contain 0.1mM phenylmethanesulfonylfluoride ( PMSF ) For SDS PAGE, the SilverQuest staining kit was used to visualize the bands (Invitrogen). Aptamer Target Purification for Protein Identification TOV 21G cells (10 8 ) were incubated with 200pmol of desthiobiotin oligonucleotide according to the scheme in Figure 3 1 The cells were was hed and the aptamer was cross linked to the cells by incubating for one minute in 1% formaldehyde PBS. Washing the cells three times at 4C in PBS diluted the formaldehyde, as quenching the reaction with lysine (the usual method for quenching formaldehyde ) can make MS analysis nearly impossible 118 Subsequently, the cells were lysed in a dounce homogenizer for two minutes (75 strokes per minute) in lysis buffer. The lysate in lysis buffer was incubated overnight at room temperature in the presence of 200 g of magnetic beads. The beads were then washed with PI washing buffer on a magnetic stand until any remaining membrane was washed from the beads. Once the beads were clean, the protein aptamer hybrid was eluted by incubation for 1h with elution buffer. TCA precipitation and acetone washing were used for further purification of the eluate protein fraction The pellet was dissolved in crosslink reversal buffer and boiled for 1h, after which the sample was loaded on an SDS polyacrylamide gel Bands of i nterest were sent for MS analysis. St IP1 siRNA Knockdown Hs_St IP1 5, 6, 10 and 11 (QAIGEN) was used with the HiPerfect starte r kit (QAIGEN) as directed on 0.8 10 5 TOV 21G cells. Aptamer binding was verified 72

PAGE 68

68 hours past transfection. The t ransfecti on efficiency was tested by a cell death positive control and a scrambled siRNA negative control to ensure proper cell viability and delivery efficiency both provided by the supplier Antibody Biotinylation StIP1 antibody (50mg) was incubated with a 20 molar excess of sulfo NHS b iotin reagent, as described in (Pierce) Aptamer Blotting rhStIP1 and BSA was blotted as described 131 In brief, 20g of the respective protein is blotted in a nitrocel lulose membrane and blocked in 4 % non fat milk in P BS containing 0.05% (v/v) Tween 20 and 1 mM EDTA 250nM aptamer solution incubated on the membrane, after which streptavidin horseradish peroxidase is added. The complex can then be visualized with the ECL plus Western blot system (GE lifesciences). Results Outline of the Aptamer Mediated Protein Identification Procedure In order to determine the identity of the protein binding to aptTOV6, it was hypothesized that the interaction between the aptamer and the protein had to be fixed i n order to ensure easy extraction with detergents without loss of the binding between the aptamer and its target. Formaldehyde is an easy to use and proven cross linker which is resistant to surface active agents (surfactants) and thus maintains the inte raction between DNA and protein 1 20 To ensure efficient capture of the protein with the aptamer, the exposure time of the formaldehyde with the aptamer boun d cells was optimized. This le d to the procedure schematically shown in Figure 3.2.

PAGE 69

69 In essence, a ptamers were bound to the cells of interest and fixed in formaldehyde. After dilution of the cross linker the membranes were dissolved in lysis buffer and the prote in aptamer hybrid was recovered using streptavidin covered magnetic beads. These beads were washed 5 times with PI washing buffer to remove any remaining membranes The aptamer protein hybrid wa s then eluted and identified using mass spectrometry (MS). E lution of the hybrid was facilitated by desthiobiotin conjugated at the 3 ptamer. Since the K d of streptavidin is about 350 times higher (i.e., lower affinity) towards desthiobiotin (K d = 3. 5 10 13 ) tha n towards biotin(K d ~ 1 10 15 ) 121 the desthiobiotin conjugated protein aptamer hybrid elutes more rapidly then other biotin containing proteins in the cell lysate. To ensure that the sample would be MS compatible surfactants and salts that might cause ionization supression 122 were removed by TCA precipitation. The resulting pellet was dissolved and boiled for 2 hours to rev erse the crosslink betw een the aptamer and the protein 115 This sample is then separated by SDS PAGE, where the differential band was identified. Aptamer TOV6, which was selected from the ovarian clear cell adeno carcinoma cell line TOV 21G, was used for target identification The aptamer does not bind to the ovarian serous carcinoma cell line CAOV3 or to the cervical cancer cell line HeLa. It was first necessary to verify the binding of the desthiobiotin aptamer due to possible formaldehyde induced denaturation of the target protein or potential desthiobiotin reactions with formaldehyde Therefore, an elution study with streptavidin labeled PE cy5.5 was performed using flow cytometry. In this study, the labeling efficiency was studied by comparing the signal from desthiobiotinylated aptamer with the signal from aptamer bound cells after formaldehyde treatment and after incubation with biotin. If the

PAGE 70

70 biotin is able to compete with desthiobiotin, the cells lose their fluorescent signal. As can be f ound in Figure 3.2 the desthiobiotin can easily be eluted by biotin and formaldehyde does not prevent aptamer binding. SDS PAGE of t he Bead Binding Fraction of Whole Cell Lysate The protein fractions that bound to aptamer TOV6 with or without formaldehy de treatment were studied by SDS PAGE. As can be seen from the SDS polyacrylamide gel in Figure 3 3 a clear band at around 78kDa appears in the crosslinked sample and no bands are detected in the non crosslinked section This indicates that formaldehy de is effe ctive in main taining the aptamer protein interaction which would otherwise be lost in the extraction process Optimization of the crosslink time was imperative however, as excessive crosslinking can lead to significant back ground, while insufficient cro sslinking does no t yield any protein fraction (Figure 3 4). Because of the lower affinity of desthiobiotin towards streptavidin compared to biotin, the aptamer protein hybrid can be elute d prior to other biotin containing protein fraction s significantly reducing the background that originates from biotin containing enzymes or other non specific streptavidin inte ractions 91 As shown in Figure 3 3 a protein band (band 8 ) was effectively eluted of the beads as a result of the replacement of biotin by desthiobiotin After incubating longer at 65C (to increase the release of biotin containing protein), only trace amounts of protein can be observed (band 9 ). This indicates that the aptamer protein hybrid can easily be removed from of the c ell lysate and specifically recovered from the beads. There was one distinct band that could be eluted, but be detected in the non crosslinked sample. This band was exci sed and sent for MS analysis. Stress Induced Protein 1 (StIP1) 123 was the to p result for both samples sent for MS analysis

PAGE 71

71 The MS results from the service lab can be found in Fi gure 3 5 hsp90 Organizing Protein (HOP), a protein that is known to play a regulative role as a co chaperone in the heat shock p rotein (hsp) 90 chaperone complex 124 Recently, the protein was found in the membrane of various cell types, including ovarian cancer cell li nes (TOV 21G, ES2, SKOV3, OVCAR 3) 125 the pancreatic cell line Panc 1 126 and the glioblastoma cell line A172 127 Confirmation of the Binding of aptTOV6 to St IP1 In order to confirm the binding of AptTOV6 to the StIP1 TOV 21G was incubated with biotinylated St IP1 antibody and analyzed with flow cytometry The effect of AptTOV6 and StIP1 antibody binding to STIP 1 s iRNA treated TOV 21G cells was investigated with the help of flow cytometry. T he target of aptTOV6 was confirmed with the help of StIP1 siRNA knock down the effect of the binding of aptTOV6 has been investigated b y using predesigned and experimentally val idated siRNAs for St IP1 128 StIP1 is a known membrane protein in TOV 21G and also in A172 125,12 7,129 Figure 3 6 demo n strates that the binding of apt TOV 6 decreases when incubated with StIP1 siRNA treated cells, while the binding is unaffected against scr ambled siRNA treated TOV 21G cells. Figure 3 7 shows similar results for the glioblastoma cell line A172. In both experiments the expression of PTK7 91,130 was also tested to demonstrate that the general expression of membrane pr otein was unaffected (PTK7 is a protein that is generally expressed in many cell lines, including TOV21G) (Figure s 3.8 and 3.9 ) Finally, the binding of the StIP1 anti body M33 was abrogated by siRNA induced knockdown of StIP1 expression (Figure 3.10 B ). All these results pro ve that the aptamer TOV6 binds to the membrane protein StIP1. Furthermore, rhStIP1 was immobilized on a nitrocellulose membrane, and the TOV6

PAGE 72

72 biotin blot 13 1 was able to stain the membrane, while it could not st ain immobilized BSA (Figure 3.11 ). Discussi on An efficient method for the identification of an aptamer s binding protein has been developed. Formaldehyde in c ombination with a desthiobiotin conjugated aptamer was proven to be efficient and specific (i.e. only aptamer interacting proteins were extr acted) in order to identify the proteins that aptamers target The crosslinking of the DNA ap tamer protein interaction proved to be an efficient and straightforward way for the efficient extraction of m embrane proteins Without crosslinking, proteins that are imbedded in the lip id bilayer of the cell lose their tertiary structu re s when the lysis buffer is employed to remove the proteins from of the bilayer (Figure 3 3) 28,113 Therefore, in order to be able to extract an unknown protein with an aptamer and keep the tertiary structure intact, a crosslinker is warranted The use of s urfactants in the lysis buffer can also disrupt the interaction between the aptamer and the target, as the interaction between protein and tar get are governed by Van der W aals interactions and hydrogen bonds. Thus, in order to prevent loss of binding between the aptamer and the unknown target, it is imperative that the interaction be fixed. Our lab has previously developed a method that is ba sed on a similar concept, but its ap plication requires the use of specialized nucleotides which when incorporated in to the aptamer at the key position s destroy the binding of the aptamer with the target. At the other extreme, if the crosslinking modali ty is bonded at the wrong location, no protein can be extracted, as the necessary fixation between aptamer and protein would be absent. Thus, by using formaldehyde, the re is no need for separate optimization of

PAGE 73

73 the probe for target identificati on. The onl y optimization that is needed is the reaction time between the aptamer and the protein (Figure 3 4). F ormaldehyde only crosslinks amines that are in close proximity of each other. In this manner, only molecules that truly interact with each other can be extracted with the aptamer 120 Further more, the crosslinked protein aptamer hybrid is considerably more soluble then the membrane protein in itself, as the phosphate backbone of the aptamer provides extra anionic groups tha t increase the solubility of the protein One of the biggest advantages of SELEX, the selection of ligands for membrane proteins in their native form s could have been problematic for aptamer target elucidation, as the lysis of the cells potentially leads to the refolding of the t arget, that may lead to the loss of interaction. The use of a crosslinker like formaldehyde has solved this problem. Once the aptamer is fixed to its target, the maintenance of the tertiary structure of the membrane protein become s irrelevant as identifi cation solely depends on the individual peptides analyzed Furthermore, the use of formaldehyde does not prevent the analysis of the protein, as the formaldehyde mediated crosslink is unstable above 72C and therefore reversible allowing the release of the aptamer at the point where th e protein needs to be analyzed 132 Furthermore, because the conventional biotin conjugation is replaced with a desthiobiotin conjugation, the protein aptamer hybrid can easily be removed from the streptavidin by fre e biot in competition (Figure 3 2 ). Th at StIP1 a membrane protein is in cancerous cells, including TOV 21G is support ed in the literature Wang et al. found StIP1 in a 2D gel that was run as a comparison between tumor interstitial fluids and non tumor inter stitial fluids from ovarian cancer patients. They confirmed the presence of StIP1 in the cell membrane of several

PAGE 74

74 ovarian cancer cell lines, including SKOV3 and TOV 21G. They also performed a pilot study, and the results suggest ed that StIP1 can be used in combination with CA125 for the early detection of ovarian cancer in patients 125 Shin et al. show ed that chaperone and co chaperone proteins can commonly be found in the membrane of cancerous cells 133 Although StIP1 was not found in their fi ndings, many other (co ) chaperone proteins were extracted from the membrane of ovarian cancer pa tient tissue, demonstrating that chaperone proteins can be retrieved from the membrane of cancerous cells. In ovarian cancer, St IP1 is found in the cytoplasmi c membrane, where it interacts with ERK and indu ces proliferation 125 In pancreatic cancer it is found to play a crucial role in metastasis 126,134 Conclusions This work demonstrates that protein identification of aptamer targets can become a powerful tool in the development of new cancer biomarkers. SELEX can be employed for the selection of aptamers that are spec ific for diseased cells. T he target identification of these proteins can lead to deeper insight of the mechanism that drives the pathology of a cancer. Due to the nature of the selection procedure (i.e., target cell is cancerous, negative cell is healthy) it is very likely that the targets binding these disease specific aptamers can lead to new biomarker s or at least to deeper insight in t h e oncopathology of the cancer cells studied These biomarkers can not only be used for early diagnosis, but can also be useful for the development of new patient care strategies, as new pathways can be identified that can be targeted with small molecule drugs for a more efficient form of therapy and drug discovery 135

PAGE 75

75 Scheme 3 1 The chemistry of formalde hyde mediated DNA Protein cross linking In t his example cytidine crosslinks to a lysine

PAGE 76

76 Figure 3 1 G eneral procedure for protein identification with the use of aptamers. The aptamer is bound to the cells. Washing the cells in WB removes non binding aptamer Then the 1% formaldehyde solution is added to the cell s and allowed to crosslink for 2 minute s after which the formaldehyde is diluted by washing in WB. The cross linked cells are homogenized in lysis buffer, magnetic beads are added, and the aptamer hyb rid is captured on the beads. The beads are further washed after which the hybrid is eluted by biotin elution The crosslink is reversed and the protein fraction is dissolved and separated by SDS PAGE, after which the resulting band is analyzed

PAGE 77

77 Figure 3 2 S tudy of the effect of formaldehyde on st reptavidin binding and biotin elution of the desthiobiotin conj ugated aptamer. Red : TOV 21G; Green: Library ; Dark blue: TOV6; Orange: TOV6 in 1% CH 2 O; Light blue: TOV6 after 30 min in 5mM biotin solution; Magenta: TOV6 in 1% CH 2 O after 30 min in 5mM biotin solution

PAGE 78

78 Figure 3 3 Silver staining of the material obtained from the aptamer me diated protein purification for aptTOV6. Stains for ce lls in presence or absence of formaldehyde are shown. 1 & 10 : bands from la dder; 2: unbound fraction from non crosslinked sample; 3: Fraction that bound to nave library fr om non crosslinked sample; 4 : f r action that bound to a ptTOV6, eluted with 1 h of 5mM biotin in PBS at 37C from non crosslinked sample; 5: f raction that remained on the beads after elution at 37C, obtained by eluting at 65C from non crosslinked sample; 6: unbound fraction from crosslinked sample; 7: f raction that bound to nave library from crosslinked sample; 8 : f r action that bound to a ptTOV6, eluted with 1 h of 5mM biotin in PBS at 37C from crosslinked sample; 9: Fraction that remained on the beads af ter elution at 37C, obtained by eluting at 65C. The crosslink time here was set at 2 minutes

PAGE 79

79 Figure 3 4. C omparison of sample run s with insufficient or excessive crosslink time s B ands 2 4 represent samples from 1 minute incubation with formaldehyde. Bands 5 9 are from samples after 10 minutes incubation. Band s 1 and 10 represent ladder; 2: the eluted sample after 1h of incubation at 65C after 37C incubation for TOV6; 3: the eluted sample at 37C after 1h for TOV6; 3: the eluted sample at 37C after 1h for Library; 4: bead fraction after elution at 37C and 65C 5: Unbound fraction; 6: bead fraction after elution at 37C and 65C ; 7: elution fraction after 1h at 37C for TOV6, 8 and 9: the eluted sample after 1h of incu bation at 65C after 37C incubation for TOV6 The samples in band s 3 and 7 show the presence of the protein of interest which was observed in Figure 3 3

PAGE 80

80 Figure 3 5 P roteins found in the aptamer TOV6 binding fractions. StIP1 was the only top protei n found in both samples sent for analysis. Each column represents the protein hits for each sampl e

PAGE 81

81 Figure 3 6 Silencing of StIP1 in TOV 21G cells. Red : Str. Alexa 488 only treated TOV 21G; Green: Library treated cells; Black : mo ck siRNA treated cells ; Dark blue: StIP1 siRNA 5 Orange: StIP1 siRNA 6; Light blue: StIP1 siRNA 10;Magenta: StIP1 siRNA 11. The cells were tested for TOV 6 binding after 72 h of siRNA treatment

PAGE 82

82 Figure 3 7 Silencing of StIP1 in A172 cells. Red : Str. Alexa 488 only treated A172; Green: Library treated cells; Black : mo ck siRNA treated cells ; Dark blue: StIP1 siRNA 5 Orange: StIP1 siRNA 6; Light blue: StIP1 siRNA 10; Magenta: StIP1 siRNA 11. The cells were tested for TOV 6 binding after 72 h of siRNA treatment

PAGE 83

83 Figure 3 8 A bsence of PTK7 silencing with StI P1 siRNA treatment in A172 cells. R ed: Str. Alexa 488 only treated A172; Green: Library treated cells; Black : mock siRNA treated cells ; Dark blue: StIP1 siRNA 5; Orange: StIP1 siRNA 6; Light blue: StIP1 siRNA 10; Magenta: StIP1 siR NA 11. The cells were tested for sgc8 binding after 72 h of siRNA treatment

PAGE 84

84 Figure 3 9 A bsence of PTK7 silencing with StIP1 si RNA treatment in TOV 21G cells. R ed: Str. Alexa 488 only treated TOV 21G ; Green: Library tre ated cells; Black mock siRNA treated cells ; Dark blue: StIP1 si RNA 5; Orange: StIP1 siRNA 6; Light blue: StIP1 siRNA 10; Magenta: StIP1 siRNA 11. The cells were tested for sgc8 binding after 72 h of siRNA treatment

PAGE 85

85 (A) (B) Figure 3 10 A : The binding of AptTOV6 with (blu e) and without (black) StIP1 knock down; B : The binding o f StIP1 antibody M33 with (blue) or without (black) StIP1 silencing

PAGE 86

86 Figure 3 11 Chemi luminescent blot of rhStIP1. The aptamer is able to induce a strong chemiluminescent signal with rhSTIP1, but not with BSA. When staining the protein with library, no luminescence can be observed

PAGE 87

87 C HAPTER 4 THE FUNCTION OF STIP1 IN TOV 21G Introduction As shown in Chapter 3, t he TOV6 aptamer binds to the membrane protein StIP1. StIP1 or heat shock protein organizing protein (hop) is known to interact with heat shock protein 70 (HSP70) and heat shock protein 90 (HSP90) 136 StIP1 found in the cytosol is a component of the chaperone complex HSP70, StIP 1 and HSP90 which play s a key role in retaining the stability and proper folding of numerous proteins involved in cell viability 137 HSP90 acts as a foldase and that needs ATP to adopt a conformation that folds client proteins back to their active form s A recent study on the structure of the HSP90 StIP1 HSP70 complex provided information about the fun ction of StIP1 in this co mplex. StIP1 plays an important role in stabilizing the client loading conformation of HSP90, and allows for easy inter actions bet ween HSP90 and client proteins such as HSP70 The interaction of StIP1 with HSP90 forces conformati onal changes that mimic the ATP bound HSP90 form, and allows HSP90 to interact more easy with client proteins by bringing the hydrophobic surfaces of HSP90 to gether in an ATP independent manner 138 Chaperone and co chaperone proteins have been identified in the cell membrane of several cancer cell lines by global profiling techniques 133 and the presence of StIP1 has been documented in several of these 125,133 By performing Western blots with cell surface isolation kits, StIP1 has been found to be overexpressed in the membrane s of TOV 21G, OVCAR3, SKOV3 and ES2. This study also performed bl ood sample tests on several ovarian cancer patient s (both at early and late stages of cancer progression ), which indicate that StIP1 shows promise as a potential biomarker for ovarian cancer 133

PAGE 88

88 However, t he test was too limited for conclusive results. T o determine the extent to which in w hat extent StIP1 is expressed, m ore ovarian cancer cell lines were tested. StIP1 is also found in the membrane of several other cancer cell lines, including A172, a gliob l astoma cell line 127, 139, 140 and Panc 1 a pancreatic cancer cell line 126,134 In pancreatic cancer, silencing of StIP1 results in a significant decrease in invasiveness. Walsh and collegues observed that StIP1 was secreted along with HSP90 and matrix metallo protease 2 ( MMP2 ) to facilitate invas ion. T his chapter describes studies of the role of StIP1 in TOV 21G cells as well the effect of TOV6 binding on proliferation or viability. We hypothesize that StIP1 has a similar function in TOV 21G as it has in pancreatic cancer and th us plays a role in the metastatic properties of TOV 21G 163 Materials and Methods Instrumentation Cell Culture and Reagents The TOV 21G cell line was purchased from the American Type Cell Culture (ATCC) Dr. Patricia Kruk from the University of South F lorida k indly donated the cell lines OVCAR3, OVCAR8, TOV112D, SKOV3, A2780s, A2780cp and C13 All cell lines were maintained in culture with MCBD 105: Medium 199 (1:1), supplemented with 10% FBS and 100 IU /mL Penicillin Streptomycin The cells were cultu red at 37C in a 5% CO 2 atmosphere. Aptamer TOV6 and library were modifications biotin as needed The HSP90 specific inhibitor 17 N Allylam ino 17 demethoxygeldanamycin (17 AAG) was purchased from Sigma.

PAGE 89

89 Proliferation Assay For proliferation studies, 100,000 cells were seeded in 24 well plates and allowed to grow for 3 days. The cell viability was tested in 0.05% trypan blue and MTT. The cells were counted in a hemocytometer. Tumor Invasion Assay The i nvasion assay s were performed as described elsewhere 14 1 In brief the direct invasiveness of the cells was evaluated with the BD Falcon FluoroBlok 24 Multiwell Insert S ystem (BD biosciences), precoated with matrigel ( Figure 4 1). A migration control was run on the same system, using a BD Falcon FluoroBlok 24 Multiwell without matrigel coating (BD biosciences) The top compartment was loaded with 60,000 cells/ well in minimal media (RPMI) the lower compartment was filled with RPMI with a chemo attractan t (10% FBS) added The 17AAG or aptamer TOV6 that was added in the minimal media was pre filtered with 0.2M syringe filters (Fisher) Percent invasion is calculated by #invading cells/#migrating cells *100%. Membrane Expression of StIP1 in Ovarian Cancer Cell Lines To determine the binding of the aptamers with different ovarian cell lines, the target cells (3 x 10 5 ) were incubated with varying biotin labeled TOV6 on ice for 30 minutes in 100 mL of BB. Cells were then washed twice with 500 mL of BB, and suspended in 100 mL of BB containing streptavidin PE Cy5.5 at an appropriate dilution. Cells were then washed twice with 5 00 mL of WB, and were biotin labeled random sequence as the negative control.

PAGE 90

90 Statistical Analysis All errors reported are the standard deviation obtained from three replicates unless o ther w ise repo rted. T he statistical significance between different invasion samples was determined with the student t test, treating p values below 0.05 as significant. The EC50 or IC50 of the tested drugs were calculated with JMP (SAS). Results Growth Inhib itory Effects of TOV6 The aptamer and the library were added to monolayers of cells, and the cell viability was tested with MTT and Trypan Blue. TOV6 did not induce any cell death as shown in Figure s 4 2 and 4 3 However, t he cell proliferation was ha mp ered at higher levels of TOV6, as can be seen in Figure 4 4 The I C50 of TOV6 in TOV 21G cells is 0.2 M. The effects of the TOV6 are cytostatic as the viabilit y of the cells was not hampered (Figure 4 3 ) and no cell death was observed visually (Figure 4 2 ) but the proliferation of the cells was affected (Figure 4 4) The Non Proliferative Effects of 17AAG Compound 17AAG is an efficient HSP90 specific inhibitor 142 StIP1 binds with HSP90 to form a complex that allows p roteins to bind and be refolded 138 Since HSP90 also regulates other crucial cellular mechanisms 142 a n IC50 study was performed on TOV 21G cell s in vitro in order to determine the inhibition of HSP90 before a decrease in proliferation can be observed. If the concentration of 17AAG can be minimized in order to maintain proliferation a more correct view about the role of HSP90 in its interaction with StIP1 can possibly be observed Figure 4 5 shows the effect of 17AAG on TOV 21G proliferation The IC50 of 17AAG in TOV 21G was 0.060M which is close the value found in the literature 0.10 0 M 167 Figure 4 6

PAGE 91

91 represents the visual control for cells that remained attached on the dish after 17AAG treatment The Effect of St IP1 siRNA Silencing on Invasion Walsh et al. showed that StIP1 is involved in the invasion process in vitro in pancreatic cancer 134 Since StIP1 is expressed on the cell membrane of TOV 21G an invasive cell line 143 inside the membrane is playing a role in metastasis The true invasive potential of a cell line is measured by the propensity to migration through a microporous membrane and the ability of the cell line to cross the basement membrane (modeled by a microporous membrane cover ed with matrigel) TOV 21G that had StIP1 silenced through siRNA showed a reduction in its propensity to migrate ( Figure 4 7 A versus B histograms ) As shown in Figure 4 7 neither 17AAG (2A and 3A) nor saturation concentrations of TOV6 (4) by themselves (without StIP1 silencing) inhibits migration. In Figure 4 7, histogram 3B, 0.10 M 17 AAG slowed the migration, but at this concentration, the observed decrease in migration rate could be attributed to cell death triggered by the drug 17AAG In a duplica te experiment with matrigel coated membranes (Figure 4 8 ) it can be obser ved that HSP90 inhibition strong ly redu ces the ability of the cells to invade even at a concentration where no loss in proliferation is observed over the tested time span The StIP 1 silenced cells also had a diminished ability to cross the membrane. As can be seen, the effect of the inhibition of invasion works in a cumulative fashion. Remarkably, TOV6 strongly inhibit s the invasion of the cells, indicating that the aptamer could be used in further studies to fight metastasis. Figure 4 9 ( % invasion ) shows that that treatment with TOV6 provides invasion inhibitory effects to the same degree as StIP1 silencing when cells are treated treated with 0.10 M 17 AA G.

PAGE 92

92 The Expression of StIP1 in StIP1, the target for aptamer TOV6, is expressed in TOV21G (OCCA ) but not in CAOV3 ( SO AC) Some binding tests have been performed for other cell lines in Chapter 2 where expression of S tIP1 was found in the membrane of A172, CEM, HCT 116, H23. To determine if the aptamer could be useful clinically for possible early detection and to what extent the marker could be used for cancer different than ovarian clear cell adenocarcinoma, a pane l of cell lines was selected that include d several examples of serous adenocarcinoma ( OVCAR3, OVCAR8, SKOV3 ). Other types of ovarian histological subtypes include d ovarian endometrioid carcinoma (TOV 112D), and in the ovarian cancer cell lines A2780 s A2 780cp and C13. Both A2780cp and C13 are cisplatin resistant serous adenocarcinoma cell lines 100, 16 2 The flow cytometry experiments (Figure 4 10) on these cells show that there is some heterogeneity in serous adenocarcinoma (binding to OVCAR3 and SKOV3) but that most resistant cell lines do not express StIP1, with the exception of C13 ( Figure 4 10 ) The rate of expression in StIP1 positive cells is variable, as some cell line show high expression (SKOV3, OVCAR3, TOV112D, TOV21G) of StIP1, while some si gnals from TOV6 binding to StIP1 is but marginal (A2780s, C13) The results suggest that StIP1 is a protein that can be found in other histological types than OCCA Discussion This chapter describes a study that tests the hypothesis that StIP1 plays a role in cell invasion, as an adapto r protein that helps in the activation mechanism of HSP90 has with of matrix metallo proteases ( MMP s ) Recent structural studies have shown that StIP1 is binds in the N terminal ATP binding pocket of HSP90 138 ,165 That HSP90 is

PAGE 93

93 involved in the invasion mechanism in TOV 21G has been demonstrated by the inhibition of invasion with the HSP90 specific inhibitor, 17AAG 142,144 An interesting result was observed when TOV6 was added to the media of TOV 21G. The invasion dropped down to almost the same levels that occurred with StIP1 silencing, combined with 17AAG in the media Although to more clearly elucidate this phenomenon, structural studies are need ed it is likely that the aptamer inhibit s the activation of pro MMP 2 For example, HSP90 is a known activator of MMP2 ( constitutively expressed, but needs to be activated for function) and the actions of HSP90 have been subject to extensive study 165,166 It may be possible to expand the curre nt understanding of MMP activation by HSP90 by considering that StIP1 plays a critical role in the activation of these important proteins involved in cell invasion perhaps by maintaining the folding of HSP90 that is required for its activation function It can be hypothesized that the aptamer sterically hinders the binding of proMMP2 to StIP1 in a complex with HSP90, thereby preventing proMMP2 from being cleaved to MMP2 I t is known for example that MMP 2 cannot be found in the conditioned media of pancreatic cells in wh ich StIP1 has been knocked down 134 Furthermore in TOV 21G have the levels of active MMP 2 been directly correlated with HSP90 inhibition by 17AAG 165 A proposed mechanism of activation is repre sented in Fig ure 4 11 Figure 4 11 described how proMMP2 binds to the StIP1 HSP90 complex to form active MMP2. It is hypothesized that the proMMP2 has to bind to both proteins in complex to be activated, as StIP1 silencing and HSP90 inhibition strongly r educe invasion. Since the aptamer TOV6 is as affective as the combination of StIP1 silencing and HSP90 inhibition, it is very likely that the aptamer blocks the epitope for proMMP2 that requires both proteins. To gain more detailed

PAGE 94

94 insight in this phenom enon, more study is definitely warranted The question can be asked for example whether the aptamer is inhibiting the complex on the cell surface, or the secreted complex of StIP1 and HSP90. The role of trans membrane (TM) MMP may also be needed to be investigated.

PAGE 95

95 Figure 4 1 Figure of a Boyden chamber for migrati on and invasion studies. For these studies, the top chamber was filled with a m inimal media and the cells of interest. The lower compartment contained a chemo attractant (10% FBS) by which migrating (no matrigel) or invading cells (mat rigel) can travel through the microporous membrane. The cells can be stained with a fluorophore by transferring the transwell insert in a fluoroph ore containing solution. The fluorescense of the cells was measured with a botto m plate reader. Non migrated or invaded cells were not detected due to an opt ical filter in the trans well insert

PAGE 96

96 Figure 4 2. Microscopic images of TOV 21G cells treated with TOV6 for 3 days The cells look more round with increasing levels of TOV6 In order to det ect dead cells, the cells were stained with 0.05% Trypan Blue (Lack of blue cells indicates that TOV6 is not cytotoxic)

PAGE 97

97 Figure 4 3. MTT assay of TOV 21G cells incubated with l ibrary and TOV6. There was no decrease of viability of the cells at the concentra tions of the aptamer tested. Error bars represent the standard deviation (n=3)

PAGE 98

98 Figure 4 4. N ormalized cell count of TOV 21G after 3 d ays of incubation with TOV6. The cells were normalized against un treated cells after 3 days of growth. The I C50 was determined to be 0.20 M by non linear regres sion in JMP. The error flags are the standard deviation (n = 3)

PAGE 99

99 Figure 4 5. N ormalized proliferation study of TOV 21G, a fter three days of incubation with 17AAG the IC50 was determined to be 0.060M by non linear regression in JMP The error bars give the standard deviation (n = 3)

PAGE 100

100 Figure 4 6. Microscopic images of TOV 21G cells treated with different levels of 17AAG in full media after 2 days incub ation The cells were stained with 0.05% Trypan Blue The overall cell density decreases with increasing 17AAG concentrations.

PAGE 101

101 Figure 4 7. M igration of TOV 21G across a microporous membrane. StIP1 siRNA treated cells migrate slower (histograms la beled with B) then scramble d siRNA treated cells (histograms labeled with A) The reduction of migration with 0.1M 17AAG can be explained by the reduced health of TOV 21G cells at this concentration. TOV6 and 17AAG do not affect migration by themselves Error flags represent the standard deviation (n = 3)

PAGE 102

102 Figure 4 8. I nvasion assay of TOV 21G determining the eff ect of TOV6 on the ability of TOV 21G to cross a matrigel layer. StIP1 si RNA treated cells are hampered in their ability to digest the matri gel layer (all B histograms) a n effect that is amplified by 17 AAG inhibition (1A 2A 3A) TOV6 (histogram 4) without any siRNA treatment is slows down the invasion at similar rates at StIP1 knocked down 17AAG treated samples (3B). Error flags represent the standard deviation (n = 3)

PAGE 103

103 Figure 4 9. Combined TOV 21G cell invasion assay. The dat a suggests that the invasion is facilitated through a mechanism where StIP1 and HSP90 are involved. The effect of HSP90 inhibition and StIP1 knock down seems to be cumulative. Furthermore, TOV6 is able to inhibit invasion as strong as StIP1 knockdown in combination with 17AAG

PAGE 104

104 Figure 4 10 Bi nding assay of several ovarian cancer cell lines with TOV6First column, fro top t o bottom: a) OVCAR3; b) TOV112D; c) OVCAR8; d) A2780cp; second column, a) SKOV3; b) TOV 21G; c)A2780s; d)C13. Red: Unlabeled cells; Green Library ; Dark blue: TOV6 d) is found on the next page

PAGE 105

105 Figure 4 10 Continued Figure 4 11 A proposed mechanism for TOV6 caused inhibition of cell invasion. MMP2 needs to be activated by cell membrane protein interactions possibly either by direct proMMP2/HOP HSP90 interaction or via complexation with MT MMP and HOP HSP90. Either of these po ssibilities could be blocked by direct competition with TOV6

PAGE 106

106 CHAPTER 5 OVARIAN CANCER STEM CELL SELEX Introduction Ovarian cancer tumors like all cancer types exist as heterogeneous cell clusters. N ot all cells in the tumor are the same but there is a hierarchical distribution of cells, making tumors caricature s to the normal tissue s of their origin The hierarchy in normal tissue is attributed to stem cells, a type of cell that can differentiate into several dif ferent tissues ( e. g. skin, bone marro w). The similarities between normal tissue and tumors have le d to the attractive model of cancer stem cells. This model considers a tumor to be a n agglomerate of differentiated, non mitotic cells that also contains another type of cells capable of self r enewal which can stay in the body for a life time and are resistant to chemical or electromagnetic (i.e. radiation) attack s. These properties go hand in hand with the observation that stem cells can be non dividing for a long time and that they have the ability to colon ize other parts of the body. T hese properties explain what is commonly observed in ovarian cancer: the patient show s an excellent response to chemo or radiation therapy, but cannot be considered cured 145 It is not rare for a patient wh o ha s been declared healthy to suffer a reoccurring cancer, that is even more aggressive and shows strong drug resistance, and these properties become more heinous as the length of the treatment free time interval increase 146 Basically, it is assumed tha t most of the cells die from a successful chemical or irradiative assault, but the cell s that show stem cell like properties can survive the assault and sustain in the tissue for a long time. This results in the relative enrichment of a chemical resistant cell. Metastatic relapse can occur many years after a patient has been declared healthy 147 Since these types of cancer including ovarian

PAGE 107

1 07 cancer, fit the stem cell model remarkably well, the cells responsible for these relapses have been 148,149 In ovarian cancer cancer stem cells have been acknowledged to be involved in cancer aggressiveness 150 from tumors and have shown to be very effective in the initiation of tumors in the xenograft model s; as few as a 100 cells are needed to initia te the growth of a tumor 151 from ovarian cancer by Fluorescence Assisted Cell Sorting ( FACS ) based on the efflux of an organic dye like Hoecht 33324 152 or on the expression of a combination of specific cell surface markers 151 The use of these properties is somewhat co ntroversial as many of the proposed markers have been shown not to be not absolute markers for CSC biology 153 154 In other words, the use of cell surface markers is highly usef ul to sort out cells that show stemness properties but the used markers are not defining the stemness properties of the extracted cells 155 An example of the limitations of the currently known stem cell surface markers can be found in CD133, or prominin, which was thought for many years to be a stem cell specific surface marker, but is expressed in many normal epithelial tissue s as well 156 Due to the limited understanding of these CSC markers, it is highly desirable to have more tools to extract CSC out of tumors and to gain deeper knowledge about the molecular mechanisms that underlie CSC emergence. In addition many stem cell markers are expressed in a plastic fashion, meaning that the expression of markers do e not remain constant. An illustrative example c an be found again with prominin; cells sorted for CD133 will return to the pre sorting levels of C D133 157 ,159 A recent publication has reported this problem for all 15 commonly used

PAGE 108

108 cell surface biomarkers 158 In order to get deeper insight in to the behavior that CSCs exhibit, more knowledge about the ir cell surface markers would be highly beneficial. Also, a stable, specific cancer stem cell line could be of great benefit for the treatment of cancer. Currently, companies and labs around the w orld are able to provide cell lines that can be passaged sufficient amounts for application of SELEX 168 169 S electing probes that bind to cells that show stemness, without the knowledge of the actual molecular target may be very advantageous By doing s o novel markers can be identified that play an important role in the biology of cancer stem cells. Such an approach has been demonstrated to be plausible for mammalian membrane proteins. The system that is curren tly used for phenotyping cells by their m embrane proteins (the Cluster of Differentiation system), contains 350 proteins. H owever it is estimated that around one third of the protein in the genome is found in the membrane 28,113 As described in previous chapters, Cell SELEX has been touted as a new method for the elucidation of cell surface markers 44 This chapter describes the application of cell SELEX to identify aptamers against ovarian CSCs. Cell SELEX on a n Ovarian CSC Cell Line Materials and Methods Instrumentation and reagents All oli gonucleotides were synthesized by standard phosphoroamidite chemistry using a 3400 DNA synthesizer (Applied Biosystems) and were purified by reversed phase HPLC (Varian Prostar). All PCR mixtures c ontained 50 mM KCl, 10 mM Tris HCl (pH 8.3), 2.0 mM MgCl 2 dNTPs (each at 2.5 mM), 0.5 mM of each primer, and Hot start Taq DNA polymerase (5units/mL) (TaKaRa). PCR was performed on a Biorad Thermocycler. Monitoring of pool enrichment, characterization of the selected

PAGE 109

109 aptamers, and quantitiation of the target pro tein assays were performed by flow cytometr y using a FACScan cytometer (BD Immunocytometry Systems). The DNA sequences were determined by the Genome Sequencing Services Laboratory at the University of Florida with th e use of iontorrent sequencing analysi s on sequence homology was performed with MAFFT 16 0 Cell culture and buffers The ovarian cancer stem cell line was purchased from Celprogen (San Pedro). specifications. In short, the cells were maintained beneath 50 60% confluence using differentiation buffer The stem cells were differentiated with specifically designed media, provided by the supplier. The TOV112D, CAOV 3 and TOV 21G ovarian cancer cell lines where maintained in culture with MCBD 105: Medium 199 (1:1), supplemented with 10% FBS and antibiotics. All the cell lines were maintained in a humidified incubator in a 5% CO 2 atmosphere. During the selection, cells were washed befor e and after incubation with wash buffer (WB), containing 4.5 g/L glucose and 5 mM MgCl 2 buffered saline with calcium chloride and magnesium chloride (Sigma). Binding buffer (BB) used for selection was prepared by adding yeast tRNA ( 0.1 mg/mL) (Sigma) and BSA (1 mg/mL) (Fisher) to the wash buffer to reduce background binding. A non enzymatic cell dissociation buffer was used for cell detachment (Sigma). Cell SELEX library A library was designed as described in Chapter 2 with the foll owing primer sequences: ATC CAG AGT GAC GCA GCA (N) 40 TGG ACA CGG TGG CTT AGT. The forward primer was labeled with FITC and the reverse primer was labeled with biotin for

PAGE 110

110 efficient separation of the forward sequence for ssDNA elution from streptavidin columns 88 In vitro cell SELEX on ovarian CSCs The commercially available ovarian cancer stem cell (OCSC) line was chosen as the target cell line. Differentiated ovarian cancer stem cells were used for the negative selection. Differentiation of cells was monitored under the microscope: undifferentiated cells were round, while differentiated cells were larger, spindle shaped and had a more complex cellular structure. The selection occurred in s imilar fashion as described in C hapter 2, except that : all the cells that were used during the SELEX proced ure were detached from the flask with non enzymatic dissociation buffer in order to ma ximize the possible contact surface between the cells and the library. This procedure also res ulted in increased the washing stringency, as the supernatant was easier to remov e compared to a selection for adherent cells on the plate 119 PCR amplifications were carried out at 95C for 30s, 60C for 30s, and 72C for 30s, followed by a final extensi on step for 3 minutes at 72C. Enrichment was monitored with flow cytometry. Due to the relative instability of the cell line, SELEX was stopped and submitted for sequencing as soon a s specific a pool was selected. Specificity and affinity studies Flow cytometry was used for the determination of the affinity constants of the found aptamers. T he ovarian CSCs (2 x 10 5 ) were incubated with various biotin labeled aptamers or FITC labeled pools (for enrichment asses s m ent ) on ice for 30 minutes in 100 mL of BB. Cells were then suspended in 100 mL of BB containing streptavidin PE Cy5.5 for the labeling of the cells After labeling, the c ells were washed twice with 500 mL of WB, and were suspended in 200 mL of WB

PAGE 111

111 for flow cytometric biotin labeled random sequence as the negative control. All the experiments for binding assays we re repeated three times The specific binding intensity was calculated by subtracting the mean fluorescence intensity of the background b inding from the mean fluorescence intensity of the aptamers. The equilibrium apparent dissociation constant (K d ) of the fluorescent ligand was obtained by non linear regression analysis of the specific binding intensity (Y) versus the aptamer concentratio n (X) fitted to the equation Y=B max X/(K d +X) using SigmaPlot (Jandel, San Rafael, CA). Aptamer mediated cell sorting with magnetic beads The streptavidin covered magnetic beads (Invitrogen) were incubated with aptamer solution for 10 minutes, after which they were washed twice with WB resulting in aptamer functionalization of the beads The beads were then incubated in non enzymatic dissociation buffer treated cells for 30 minutes. The beads were used to extract cells that bound to the aptamer The cel ls were released from the beads by a 30 minute incubation with DNase I (Invitrogen) and cultured for one day before analysis. Results Monitoring the pool enrichment for undifferentiated O CSC vs differentiated ovarian stem cells At the start of the select ion, 20 pmol of library was used. See Figure 5 1 for the progressive enrichment. Se quences that bound generally express ed surface markers for the ovarian cell line were removed for the most part by counter selecting with the dif ferentiated cell line (Fig ure 5 2) The selection was concluded after 15 rounds, as some amplification for differentiated cells was observed (blue in Figure s 5 1 Figure 5 2), while a good pool was obtained after 14 rounds ( light blue in Figure s 5 1 and 5 2).

PAGE 112

112 Furthermore, due to t he instability of the cells (the cells a re only undifferentiated for 6 to 50 passages 169 (Celprogen, personal correspondence)) it was decided that the enriched pools were to be analyzed by next gen sequencing In round 15, there was an increase in the binding with OCSCs observed. H owever, this pool also bound a little to the differentiated cells (Figure 5 2 ). Since pool 14 shows a good binding profile with the cells (better specificity), pool 14 was also included in the iontorrent analysis. As can b e seen in Figure 5.1, the cells that bound to the pools were split in two distinct populations, which could reflect the fashion by which stems cells split into differentiated daughter cells and in undifferentiated mother cells 161 Binding assay of putati ve aptamers and determination of K d In total, 5 aptamers chosen from the alignment data showed specificity for un differentiated ovarian cancer stem cells; they showed binding to undifferentiated cells while they did not bind to the dif ferentiated cells. Flow cytometry plots for binding assays are shown in Figures 5 3 to 5 7 and the dissociation constants are summarized in Tab le 5 1. As can be seen, the apparent K d s of the selected aptamers all lie in the lower nanomolar range, which is the usual value f or aptamers selected through cell SELEX. Figure 5 8 shows an example of the K d 3 The aptamers were tested against several ovarian cancer cell lines. As shown in Table 5 2, DOCSC 2 is binding to the OCCA cell line TOV 21G, ovarian endometrial adenocarcinoma TOV112D and not to the serous adeno carcinoma cell line CAOV3. Also DOCSC 4 and DOCSC 5 also bind to TOV 21G. DOCSC2 and DOCSC5 were used to enrich for cells that specifically bound to the aptamers (Figure 5 9) For this, the aptamers were immobilized on magnetic microbeads. The beads were successful in the enrichment of cells expressing the

PAGE 113

113 marker binding to the aptamers. The phenotype s of the two cells showed s ome differen ces from each other. As shown in Figure 5 10, the cells enriched from DOCSC 2 are small and round, while the cells from DOSCS 5 are larger and are spindle shaped. The cells were also tested with flow cytometry for DOSCS 5 in Figure 5 9 and here it can be observed that after one day of culturing the cells that bound to the beads there was some re enrichment of cells that showed strong bind ing to DOSCS 5 Discussion Five aptamers for OCSC were selected from the sequencing data. The aptamers selected show good binding properties, with K d s in the low to middle nanomolar range Two aptamer s ( DOCSC 2 and DOCSC 5) were tested for the enrichment purposes. Figure 5 9 shows flow cytometry results for DOCSC 5 sorting of cells that did not show clear binding to DOCSC 5 (passage number 58) The histogram in orange represents the unsorted cells, while the black histogram clearly shows two populations of cells, similar to results obtained for the enriched pools (Figure 5 1) This experiment shows that the aptamers can be used for the extraction of the cells out of the population of undifferentiated cells. When comparing the sorted cell s with each other under the microscope there is a difference observable in the cell shape (Figure 5 10) For example, sorting wit h DOCSC 5 leads to the growth of large spindle shaped cell s compared to small round cells sorted with D OCSC 3. This implies that the cell surface markers binding to the aptamers are related to a defined phenotype. Future Work Molecular profiling is requ ired to verify if these samples are truly binding to markers related to stem cells. As has been demonstrated with the flow data and microscope images it is possible to re enrich cells which can be further analyzed for

PAGE 114

114 stemness. There are several exper i ments that can b e conducted to verify if the aptamers bind to a marker for a CSC. The experiment with the aptamer functionalized microbeads shows that this is possible. A classic experiment is the analysis of the transcriptome of the cells according to t he different aptamer sorted cells. There are a few genes that are know n to be associated with the stemness properties of CSC s 164 For example, two the membrane markers for cells in the undifferentiated state are CD133 and CD44. Other stemness markers are telomerase and c kit. If the cells indeed show stemness, the expression of these genes can be verified with qPCR. The true experiment to verify for stemness however lie s in the use of the xenograft seeding model 150, 151 in which f e wer than 1 000 cells are subcutaneously injected in nude mice. By verifying th e tumor size every 3 days for t w o months, the tumor initiating ability of the sorted cells could be determined At the moment, the current model is the only accepted method to verify stem ness, or tumor initiating ability 145

PAGE 115

115 Figure 5 1 E nrichment for ovarian cancer stem cells. The pools show binding to the undifferentiated cells. The fluorescence profile shows a separation of the enrichment in to two distinct population s

PAGE 116

116 Figure 5 2. Bi nding profile of the selected pools on the negative cell line, ovarian differentiated cancer cells (ODC). Pool 14 shows less binding than the library; pool 15 shows some binding to ODC but relatively less than the binding with OCSC

PAGE 117

117 Fi gure 5 3. Binding assay of DOCSC 1 on undifferentiated OCSC vs differentiated cells Figure 5 4. Binding assay of DOCSC 2 on undifferentiated OCSC vs differentiated cells

PAGE 118

118 Figure 5 5. Binding assay of DOCSC 3 on undifferentiated OCSC vs differentiated cells Figure 5 6. Binding assay of DOCSC 4 on undifferentiated OCSC vs differentiated cells

PAGE 119

119 Figure 5 7 Binding assay of DOCSC 5 on undifferentiated OCSC vs differentiated cells Figure 5 8 Example of a K d determination for the aptamer DOCSC 3. The fluorescence is corrected against the signal for the random library. The error bars are the standard deviation for n = 3

PAGE 120

120 Figure 5 9. Cell sorting experiment with DOCSC 5. Red: Cells; Green: Library; Orange: DOCSC 5 uns orted; Black: DOCSC 5 after sorting Figure 5 10. Microscope images of one day old cells that have been sorted with aptamer functionalized magnetic beads. Left: DOCSC 2, Right: DOCSC 5

PAGE 121

121 Table 5 1. The aptamers that show specificity towards OCSCs Name Sequence DOCSC 1 ATC CAG AGT GAC GCA GCA TCA TAC CCG AGA TTC ATC ACC CTT ACC TGT CGC TCT GCC TGG ACA CGG TGG CTT AGT A ND DOCSC 2 ATC CAG AGT GAC GCA GCA CCC GAC ACA TCT CAT TCA ATT TCG CCT CTC TGG ACA CGG TGG CTT A 3.81.6 DOCSC 3 ATC CAG AGT GAC GCA GCA TCA CCA CAC TAC ACA AAT GAT ATT CTC CAA TCC CCC GGC TGG ACA CGG TGG CTT AGT A 59.0 5.9 DOCSC 4 ATC CAG AGT GAC GCA GCA CCA AAC ACA ACT CCG GAA ACG TCA CTA ATC TGC GCA CCT GGA CAC GGT GGC TTA 11.6 2.2 DOCS C 5 ATC CAG AGT GAC GCA GCA CAC CAC CTG ACT ACA TAC CGA ACA TTC GAC TGC TGC GCC TGG ACA CGG TGG CTT AGT A 1.84 0.47 Table 5 2. Binding assay of a ptamers from OSCS with other o varian cancer cell lines Aptamer OCSC dOCSC CAOV3 TOV 112D TOV 21 G DOCSC1 + DOCSC2 + + + DOCSC3 + DOCSC4 + + DOCSC5 + ND +

PAGE 122

122 CHAPTER 6 SUMMARY AND FUTURE DIRECTIONS Summary In this thesis, the use of cell SELEX was employed as an innovative way for the determination of a cell surface marker (StIP1) that shows excellent potential for its use as a biomarker for ovarian clear cell carcinoma This protein also seems to play an important role in the invasive properties of the model cell line TOV 21G. This thesis has outlined the current difficulties that exist in the field of biomarker discovery and explains how cell SELEX can be used as a complementary technology for rapid, comprehensive determination of biomarker lead s. Since the primairy goal of cell SELEX is selection of aptamers, the method developed in this research has an important advantage as new ligands with high binding a ffinities are developed towards these markers A pool of ssDNA was enriched against TO V 21G, an ovarian clear cell carcinoma cell line. From this enriched pool of ssDNA sequences, aptamer s were selected that show nM to pM range dissociation contants and did not bind to the negative selection cancer cell line HeLa More importantly, the aptamers also bound to other cell lines, that were not used at any point during the selection, including CEM, A172, H23, HCT 116. This behavior implies that the cell surface markers of these different types of cancers have commona lities or more importantly, that the molecular mechanisms of carcinogenesis could be shared between these different types of cancer. More insight toward these mechanisms can prove to be invaluable towards curing or treating the cancer. After the aptame rs were selected, the project turned to the identification of the cell surface markers that bind these aptamers An important advantage of cell SELEX is

PAGE 123

123 that aptamer selection does not require any knowledge about the cell surface profile of the cancerous cell studied. However, the selected aptamers can be key players in the identification process Previous methods toward elucidating an aptamers target have proven to be difficult and are not generally applicable M arkers of interest come with their intri nsic problems, such as low abundance or high hydrophobicity. Problems can also occur because the interaction of the aptamer with the target of interest can be lost when the target is extracted from the cell The need for a method that reduced these probl ems therefor e presented itself In this research t he u se of formaldehyde as a cross linker between the protein an d the aptamer proved to be an elegant solution for target elucidation for one aptamer of interest, TOV6 which has excellent binding affinity to TOV 21G Because TOV6 also binds to cancer cells like CEM and A172, identificating the target protein can have significance for leukemic leukemic lymphoblasts and glioblastoma. After crosslinking the aptamer with the target protein t he band to be rese cted was easily retrieved due to the specific elution from the cell lysate was facilitated by desthiobiotin, which was conjugated to the aptamer B ecause of the specific and easy elution using biotin a clear protein band could be resected from SDS PAGE a nd submitted for mass spectral analysis. The MS results indicated that the target protein was StIP1 (also known as HOP). The identity of the protein target for TOV6 was confirmed to be StIP1 by several methods The expression of StIP1 was hampered by t he use of StIP1 siRNA silencing This silencing experiment was repeated on A172 cells (which also bound TOV6), where A172 showed the same behavior as TOV 21G, the binding with TOV6 was reduced All

PAGE 124

124 the knockdown experiments hamper ed the expression of oth er non related membrane proteins as proven by the binding of aptamer SGC8 to its target (targeting PTK7, a protein involved in planar cell polarity) The experiment was repeated with the StIP1 antibody, giving confirmation of the expression of StIP1 in TOV 21G. Finally, the binding of TOV6 was confi r med by an aptamer blot on rhStIP1. The literature suggested that StIP1 may play a n important role in cancer biology, more specifically in the stabilization of certain proteins as co chaperone and a lso in the pericellular activation of MMPs The expression of StIP1 has been confirmed in the work of Walsh et al. and Wang et al. Wang and co workers show ed the potential of using StIP1 in combination with other ovarian cancer biomarkers for early detec tion of ovarian cancer Walsh et al. proved that StIP1 play s a crucial role in the invasive capability of Panc 1 a pancreatic cancer line. It has been shown in this thesis that the same mechanism governs TOV 21G invasion. Experiments for this showed th at aptamer TOV6 is cytostatic to TOV 21G and that the aptamer inhibits TOV 21G invasion. Further study is needed to elucidate the mechanisms of these phenomena Characterization of other TOV aptamers will most likely result in the identification of HSP 90 as the target as this protein is always associated with StIP1 and also plays a role in metastasis. In this work, aptamers were also selected for ovarian cancer stem cells. Even though there are no known stable cell surface markers on cancer stem c ells, cell SELEX was still able to select a pool of aptamers that show binding to undifferentiated cells, but not to differentiated cells. Binding studies on five aptamers from the pool yielded Kd values in the l ower to middle nano molar range. The results indicate good

PAGE 125

125 promise that some of these aptamers bind to cancer stem cells but further experimental validation is required. Future Work This thesis provides an example of the additional dimension that cell SELEX can add to the field of biomarke r discovery. Today, the search for biomarker discovery is hampered by the massive amounts of data generated in proteomic studies while some of the more interesting biomarkers may never be detected considering the limitations of current analytical methods Although the limit of detection for protein s can reach close to the attomolar region (in MS) this is only the case for known proteins. When it comes to the analysis of unknown samples (e.g., search for a biomarker in blood) it is common that the ioni zation of important peptides can be suppressed by more abundant peptides of less er significance M embrane protein s are especially victim of this, because they are usually comprised of low polarity amino acids that ionize poorly and have a relative ly low s olubility This research has demonstrated the value of cell SELEX in locating interesting surface proteins with the added benefits of mass spectroscopy to determine their identities. Furthermore, cell SELEX provides new ligands for the very targets that are identified The field of cell SELEX will gain in efficiency andcapability as methodologies develop: for example selection of aptamers without the primer regions; use of next gen sequencing with fewer rounds and the use of cell sorting to make the cell SELEX process faster. The insight that can be gained from further study of ca ncers through cell SELEX is tremendous The study of ovarian cancer, as outlined in this thesis has provided insight not only in the pathogenesis of this particular cancer, but also other malignancies, such as glioblastoma, leukemia and colon cancer. As this field gain s

PAGE 126

126 ground and become s more developed, the insight in how a cancer interacts with its environment will increase to provide the required knowledge for advances in the treatment of this collection of deadly diseases. The aptamers selected fro m cell Selex can be used for drug discovery via a number of approaches. With the right selection model, aptamers can be selected that bring drugs to the cancer cells, but not to healthy tissue. Also, the identification of new proteins that are expressed in these cancers will lead to improvements in drug discovery and development. Aptamers can be used as a tool for identifying unknown pathways which could then be targeted with novel small molecule drugs or by the aptamer itself The aptamer could also be designed to act as a precursor to a highly toxic drug. T he aptamer drug would be specific in the sense that the aptamer would only bind to the cancer, avoiding toxic effects the surrounding healthy tissue. This would also allow the effective dosage fo r the patient to be lowered, result ing in more cost effective treatment with fewer side effects Aptamers themse lves can also have therapeutic efficacy. This has been demonstrated in the Macugen aptamer (blocks angiogenesis in the eye) and also in the nu cleolin aptamer, that has shown to induce cell death. It should be mentioned that the selection of aptamers for cells should be undertaken with a word of caution. The work described in this dissertation used in vitro cultures, but of cancer is suggests that not all surface molecules are the same between in vitro cultures and xenografts or real tumors There is more and more evidence that the expression of these molecules may be partially defined by the ir biological niche s. T he caus e s for a cell to become cancerous are rarely found in the

PAGE 127

127 environmental changes, but due to genetic change within the cells. However, because genetic mutations allow cell s to adapt and grow uninhibited; when selecting aptamers for therapy i t is important to consider the cells in balance with the niche where they grow On the other hand selecting aptamers for cancer biomarkers may be to find the markers that define the mechanisms of the cancerous cell, and that do not respond to the niche environment. Thi s leads to the question of the the difference is between the membrane proteome of a xenograft derived from a cell line and the membrane proteome from an in vitro culture a study that could also be conducted with cell SELEX Aptamer TOV6 has identified S tIP1 as a protein that play s a role in metastasis. More importantly, the aptamer itself can significantly inhibit metastasis and potentially help prevent metastasis from occuring But, t he aptamer itself must be further optimized if it is to be used as a drug. For this, a structural optimization step will be need ed as 76 nucleotide long aptamer drugs are not economically feasible at the moment of writing Alternatively, the answer may also lie in the development of small molecules that directly inhibit the formation of the complex in which StIP1 is involved for Wang et al. have demonstrated that StIP1 can be used for diagnosis of ovarian cancer in combination with other biomarkers. An important improvement would be use of TOV6 for Enzyme Linked Aptamer Sorbent Assay s (ELASAs) or a hybrid form of this with an antibody especially since most rare proteins do not have the antibodies with the properti es required to work in ELISAs. Furthermore, the mechanism by which TOV6 inhibits the invasion of TOV 21G needs to be investigated. Is invasion prohibited by the inhibition of the protein at the

PAGE 128

128 membrane level, or by t he inhibition of secreted StIP1? A d eeper understanding about this mechanism will prove important towards the elucidation on how invasion is regulated in the complex biology of cancer. A hypothesis that can be made here is that StIP1 is plays an important role in the stabilization of HSP90 in the cancer cell membrane.

PAGE 129

129 APPENDIX THE ANALYSIS OF NEXT GEN SEQUENCING DATA FOR CELL SELEX Introduction Next generation sequencing or high throughput sequencing are techniques which that allow massive throughput and which generates usually thousand s to millions of DNA sequences. This thesis research has demonstrated the need for this information, as well as the advantage of having sequence data. In this thesis, the methods developed by 454 Life Technologies and IonTorrent technology were used. Thes e t echnologies are is based on the release of either light, or a proton as nucleotides are added to beads in an emulsion. Each of these beads contain s one sequence, and wit h the use of a sensitive charge coupled device (CCD) camera, the synthesis is followed as triphopho nucleotides are added Then sulfurylase converts pyrophosphate to ATP, which in turn activates luciferin, which emits a light that can be detected by the CCD 170 In IonT orrent, this light based system is rep laced by the detection of the pH change that accompanies the addition of a nucleotide to the strand 171 Pools from the selection can be analyz ed by the elongation of the primer sequences with adapter sequences These sequences allow the pool to be oriented in the right direction, in order to have the forward sequence of the dsDNA that needs to be analyz ed. The actual run and analysis is done by the 454 or IonT orrent instrument, and in the end the sequences i n the library are returned. If more pool s need to be sequenced, there is the option to include additional sequences between the templa te specific primer and the bead specific primer (Figure A 1)

PAGE 130

130 The T rimming of the Sequencing Data W ith PERL Sequencing data are provided as a FASTA file that contains the sequences of the actual pools, which are considered to be representative of the entire pool. However, if the sequences need to be ordered according to their homology, it is imperative that the template specific primer be trimmed fromdata, with the help of bioinformatics. A PERL script was written that can easily remove the sequences of the primers, remove sequences that are to o short, and order sequences according to words they may contain. In the f irst script the sequences are prepared for primer remova l. The script removes all the data in the title line of the fasta line and links all the sequence data in one long string. #!/usr/bin/perl w open(FASTA, ) || die "Sorry, I couldn't find that file... \ n"; open(OUT, ">OUTPUT.fna"); my $seq; my $prim1; my $prim2; while (my $line = ){ chomp ($line); if ($line =~ /^>/) { @fasta = split(/ /, $line); $string = join(" ", @fasta); print \ n", $string, \ n"; print OUT \ n", $string, \ n"; } else{ $seq = $line; print $seq; print OUT $seq; } } print \ n"; print OUT \ n"; close(FASTA); #!/usr/bin/perl w open(FASTA1, ) || die "Sorry, I couldn't find that file... \ n";

PAGE 131

131 open(OUT1, ">MINING.fna "); open(RY, ">RY.fna"); open(SW, ">SW.fna"); print "Enter your first primer sequence: \ n"; $prim1 = ; print "Now, please enter the length of the random sequence: \ n"; $random = ; $begin = length($prim1); while (my $line = ){ chomp ( $line); if ($line =~ /^>/) { @fasta = split(/ /, $line); print $fasta[0], \ n"; print OUT1 $fasta[0], \ n"; print RY $fasta[0], \ n"; print SW $fasta[0], \ n"; } else{ $seq = $line; $total = length($line); $subtot = $random + $begin; if ($total > $subtot ){ $offset = $total $begin; $seq = substr($seq, $begin, $random); $ry = $seq; $sw = $seq; $ry =~ tr/ACGTacgt/RYRYryry/; $sw =~ tr/ACGTacgt/WSSWwssw/; print $seq, \ n"; print OUT1 $seq, \ n"; } else{ print OUT1 FALSE," \ n"; } } } print \ n"; print OUT1 \ n"; print RY \ n"; print SW \ n"; close(FASTA1); In this the file that needs to be trimmed is prompte d. Following that, the program asks for the primer and the length of the random sequence. The output of this script is the sequence of the in put data, after the primer has been removed. also removes the sequences after the length specified by the user. If the sequence of this output is to o short, the sequence is allows for the

PAGE 132

132 sorting of interesting famili es after a preliminary multiple sequence alignment 160 The #!/usr/bin/perl w # use strict; open(FASTA1, ) || die "Sorry, I couldn't find that file... \ n"; open(OUT, ">aptamer2.fna"); open(OUT1,">newpool.fna"); print "Enter your putative aptamer sequence: \ n"; my $apt = ; chomp ($apt); @pool = ""; my $n = 0; my $total = 0; my $m = 0; while (my $line = ) { chomp ($line); $total++; # @pool = split(/ /, $line); if ($line =~ /^>/) { $pool[0] = $line; } if ($line =~ m/($apt)/i){ $n++; $pool[1] = $line; if (defined ($pool[1])){ print OUT $pool[0], \ n", $pool[1], \ n"; $pool[1] = ""; } } elsif (($line !~ m/($apt)/i) && ($line !~ m/^>/)){ $m++; $rest[1] = $line; if (defined ($rest[1])){ $rest[1] =~ tr/AGCT/agct/; print OUT1 $pool[0], \ n", $rest[1], \ n"; $rest[1] = ""; } } } $n *= 100; $total /= 2; # $remain = $total $m; $m *= 100; print $total, \ n"; print "The number of aptamers of this type is: ", $n/100, ". "; print "This is ", $n/$total, "%. \ n"; print $m/100, "sequences remaining. \ n \ n"; print "That's ",$m/$total, "%. \ n \ n"; exit;

PAGE 133

133 Future W ork Improvements in the data processing would be an incorporation of the thr ee scripts in one maser script Additionally the software could be expanded with MAFFT software, so sequence dat a can directly be generated by simply inserting the input file.

PAGE 134

134 Figure A 1. The scheme of a pool submitted for sequencing. The blue portions of the DNA are the section s of the primers that allow proper orientation with the beads. The MID sequences are sequences that can be inserted in case more then one sample is analysed. This is attached to the pools to analyse by PCR with the help of the template specific sequences

PAGE 135

135 LIST OF REFERENCES 1. American Cancer Society. Cancer Facts & Figures 2011. Atlanta: American Cancer Society (2011). 2. U.S. Public Health Service, Vital Statistics of the United States, annual, Vol. I and Vol II; (1900 1970) U.S. National Center for Health Statistics, Vital Statistics of the United States, annual; National Vital Statistics Report (NVSR) (formerly Monthly Vital Statistic s Report)( 1971 2001 ). 3. Lucy Gilbert, Olga Basso, John Sampalis, Igor Karp, Claudia Martins, Jing Feng, Sabrina Piedim onte, Louise Quintal, Agnihotram V Ramanakumar, Janet Takefman, Maria S Grigorie, Giovanni Artho, Srinivasan Krishnamurthy. Assessment of symptomatic women for early diagnosis of ovarian cancer: results from the prospective DOvE pilot project The Lancet Oncology Medical Journal. Lancet Oncol; 13 : 285 91 (2012). 4. Kosary C Cancer of the ovary. In: Ries LAG, Young JL, Keel GE, Eisner MP, Lin YD, et al., editors. SEER survival monograph: cancer survival among adults: US SEER Program, 1988 2001, patient and tumor characteristics. Bethesda (Maryland): National C ancer Institute, SEER Program (2007) 5. Beahrs, OH, Henson DE, Hutter RVP, Myers MH (eds). AJCC Cancer Staging Manual, Third edition. American Joint Committee on Cancer. Philadelphia: Lippincott (1988). 6. Goff, B., Mandel, L., Melancon, C., & Muntz, H. (2004). Frequency of symptoms of ovarian can cer in women presenting to pri mary care clinics. Journal of the American Medical Association, 291(22), 2705 2712. 7. Goff, B., Mandel, L., Drescher, C., Urban, N., G ough, S., Schur man, K., et al. (2007). Develop ment of an Ovarian Cancer Symp tom Index: Possibilities for earlier detection. Cancer, 109(2), 221 227. 8. Yurkovetsky, Z. Skates, S., Lomakin, A., Nolen, B., et al. Development of a multimarker assay for early detection of ovarian cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 28 2159 66 (2010). 9. Bast, R., Brewer, M., Zou, C., Hernandez, M. A., Daley, M., Ozols, R., Lu, K., et al Prevention and Early Detect (2007). 10. Cesario, S. Advances in the early detection of ovarian cancer: How to hear the whispers early. 14 222 34 (2010). 11. http://www.uspreventiveservicestaskforce.org/3rduspstf/ovariancan/ovcanrs.htm Last accessed June 14, 2012.

PAGE 136

136 12. http://www.newchoicehealth.com/Directory/Procedure/60/Transvaginal%20Ultraso und Last accessed June 14, 2012. 13. http://www.medicinenet.com/ca_125/page2.htm La st accessed June 14, 2012. 14. H aga Y. S akamoto K., E gami H ., Y oshimura R., A kagi M. Evaluation of serum CA125 values in healthy individuals and pregnant women. The American journal of the medical sciences 292 25 9 (1986). 15. Bast, R.C Badgwell, D., Lu Z., Marquez, R e t al. N ew tumor markers: CA125 and beyond. International journal of gynecological cancer : official journal of the International Gynecological Cancer Society 15 Suppl 3 274 81 16. Rifai, N., Gillette, M. a & Carr, S. A Protein biomarker discovery and validation: the long and uncertain path to clinical utility. Nature biotechnology 24 971 83 (2006). 17. Verhoef, G., Meeus, P., Stul, M., Mecucci, C., et al. Cytogenetic and molecular studies of the Philadelphia translocation in myelodysplastic syndromes. Cancer Genetics and Cytogenetics 59 161 166 (1992). 18. Sandoval, J. & Esteller, M. Cancer epigenomics: beyond genomics. Current opinion in genetics & development 22 55 50 (2012). 19. Mikolajczyk, S.D. Millar, L. S., Tsinberg P., Coutts, S. M. et al. Detection of EpCAM Negative and Cytokeratin Negative Circulating Tumor Cells in Peripheral Blood. Journal of Oncology 2011 1 10 (2011). 20. Nielsen, K.V., Ejlertsen, B., Mller, S., Jrgensen, J. T. et al. The value of TOP2A gene co py number variation as a biomarker in breast cancer: Update of DBCG trial 89D. Acta oncologica (Stockholm, Sweden) 47 725 34 (2008). 21. Huang, Z., Lin, L., Gao, Y., Chen, Y. e t al. Bladder cancer determination via two urinary metabolites: a biomarker pattern approach. Molecular & cellular proteomics : MCP 10 M111.007922 (2011). 22. Hanash, S.M., Pitteri, S.J. & Faca, V.M. Mining the plasma proteome for cancer biomarkers. Nature 452 571 9 (2008). 23. Haga, Y., Sakamoto, K., Egami, H., Yoshimura, R. & Akagi M. Evaluation of serum CA125 values in healthy individuals and pregnant women. The American journal of the medical sciences 292 25 9 (1986). 24. Frank, R. & Hargreaves, R. Clinical biomarkers in drug discovery and development. Nature reviews. Drug discovery 2 566 80 (2003). 25. Adachi, J., Kumar, C., Zhang, Y., Olsen, J.V. & Mann, M. The human urinary proteome contains more than 1500 proteins, including a large proportion of membrane proteins. Genome biology 7 R80 (2006).

PAGE 137

137 26. Almen, M.; Nordstrom, K.; Fredriksson, R.; Schioth, H. Mapping the human membrane proteome: a majority of the human membrane proteins can be classified according to function and evolutionary origin. BMC Biol 7 50 ( 2009). 27. Garrow, A., Agnew, A.; Westhead, D. TMB Hunt: an amino acid composition based method to screen proteomes for beta barrel transmembrane proteins. BMC Bioinform 6 56 (2005). 28. Luckey, M. Membrane Structural Biology With Biochemical and Biophysical Foundations; Cambridge University Press: New York (2008). 29. Klijn, J.G.M., Ber ns, P.M.J.J., Schmitz, P.I.M. & Foekens, J.A. The Clinical Significance of Epidermal Growth Factor Receptor (EGF R) in Human Breast Cancer: A Review on 5232 Patients. Endocrine Reviews 13 3 17 (1992). 30. Sato, H., Takino, T., Okada, Y., Cao, J. et al. A matr ix metalloproteinase expressed on the surface of invasive tumour cells. Nature 370 61 5 (1994). 31. Igney, F.H. & Krammer, P.H. Death and anti death: tumour resistance to apoptosis. Nature reviews. Cancer 2 277 88 (2002). 32. Srinivas, P.R., Verma, M., Zhao, Y. & Srivastava, S. Proteomics for Cancer Biomarker Discovery. Clin. Chem. 48 1160 1169 (2002). 33. Delahunty, C. & Yates, J.R. Protein identification using 2D LC MS/MS. Methods (San Diego, Calif.) 35 248 55 (2005). 34. Shin, B. K., Wang, H., Yim, A. M., Le Naour, F. et al. Global profiling of the cell surface proteome of cancer cells uncovers an abundance of proteins with chaperone function. The Journal of biological chemistry 278 7607 16 (2003). 35. Barrera, N.P. & Robinson, C.V. Advances in the Mass Spectrometry of Membrane Proteins: From Individual Proteins to Intact Complexes. Annu. Rev. Biochem. 80 247 71 (2011). 36. Tang, N., Tornatore, P. & Weinberger, S.R. Current developments in SELDI affinity technology. Mass spectrometry reviews 23 34 44 37. Targeted Therapies (Humana Press: Totowa, NJ, 2011). 38. Burczynski, M.E. & Dorner, A.J. Transcriptional profiling of peripheral blood cells in clinical pharmacogenomic studies. Pharmacogenomics 7 187 202 (2006). 39. Elshal, M.F. & McCoy, J.P. Multiplex bead array assays: performan ce evaluation and comparison of sensitivity to ELISA. Methods (San Diego, Calif.) 38 317 23 (2006).

PAGE 138

138 40. Domon, B. & Aebersold, R. Mass spectrometry and protein analysis. Science (New York, N.Y.) 312 212 7 (2006). 41. Garratty, G. How concerned should we be about missing antibodies to low incidence antigens? Transfusion 43 844 847 (2003). 42. Udugamasooriya, D.G. & Kodadek, T. On Bead Two Color (OBTC) Cell Screen for Direct Identification of Highly Selective Cell Surface Receptor Ligands. Current protocols in chemica l biology 4 35 48 (2012). 43. Wang, X., Yu, J., Sreekumar, A., Varambally, S. et al. Autoantibody Signatures in Prostate Cancer. New England Journal of Medicine 353 1224 1235 (2005). 44. Kim, Y., Liu, C. & Tan, W. Aptamers generated by Cell SELEX for biomarker d iscovery. Future Medicine 3 193 202 (2009). 45. Keefe, A.D., Pai, S. & Ellington, A. Aptamers as therapeutics. Nature reviews. Drug discovery 9 537 50 (2010). 46. Felix Hoppe Seyler, Irena Crnkovic Mertens, Evangelia Tomai & Karin Butz Peptide Aptamers: Specific Inhibitors of Protein Function. Current Molecular Medicine 4 10 (2004). 47. Vaught, J. D.,Bock, C., Carter, J.,Fitzwater, T. et al. Expanding the chemistry of DNA for in vitro selection. Journal of the American Chemical Society 132 4141 51 (2010). 48. Mayer, G. The chemical biology of aptamers. Angewandte Chemie (International ed. in English) 48 2672 89 (2009). 49. Tuerk, C. & Gold, L. Systematic evolution of ligands by exponential enrichment: RNA ligands to bacteriophage T4 DNA polymeras e. Science 249 505 510 (1990). 50. Ellington, A.D. & Szostak, J.W. In vitro selection of RNA molecules that bind specific ligands. Nature 346 818 822 (1990). 51. Ciesiolka, J. & Yarus, M. Small RNA divalent domains. RNA 2, 785 793 (1996). 52. Jiang, L., Majumdar, A., Hu, W., Jaishree T.J. et al. Saccharide RNA recognition in a complex formed between neomycin B and an RNA aptamer Structure 7 817 827 (1999). 53. Jiang, F., Gorin, A., Hu, W., Majumdar A. et al. Anchoring an extended HTLV 1 Rex peptide within an RNA major groove containing junctional base triples Structure 7 1461 (1999).

PAGE 139

139 54. Orito, N. et al. High affinity RNA aptamers to C reactive protein (CRP): newly developed pre elution methods for aptamer select ion. Journal of Physics: Conference Series 352 012042 (2012). 55. Zhang, G. & Simon, A.E. A Multifunctional Turnip Crinkle Virus Replication Enhancer Revealed by in vivo Functional SELEX. Journal of Molecular Biology 326 35 48 (2003). 56. Chen, F., Zhou, J., Luo F., Mohammed, A. B. & Zhang, X. L. Aptamer from whole bacterium SELEX as new therapeutic reagent against virulent Mycobacterium tuberculosis. Biochemical and biophysical research communications 357 743 8 (2007). 57. Graham, J.C. & Zarbl, H. Use of Cell SELE X to Generate DNA Aptamers as Molecular Probes of HPV Associated Cervical Cancer Cells. PLoS ONE 7 e36103 (2012). 58. Lee, J.F., Stovall, G.M. & Ellington, A.D. Aptamer therapeutics advance. Current opinion in chemical biology 10 282 9 (2006). 59. Healy, J.M. et al. Pharmacokinetics and Biodistribution of Novel Aptamer Compositions. Pharmaceutical Research 21 2234 2246 (2004). 60. Pestourie, C., Tavitian, B. & Duconge, F. Aptamers against extracellular targets for in vivo applications. Biochimie 87 921 30 (2005). 61. P endergrast, P.S., Marsh, H.N., Grate, D., Healy, J.M. & Stanton, M. Nucleic acid aptamers for target validation and therapeutic applications. Journal of biomolecular techniques : JBT 16 224 34 (2005). 62. Cole S.P., Campling B.G., Atlaw T., Kozbor D., Roder J .C. Human monoclonal antibodies Mol. Cell. Biochem 62 109 20(1984). 63. Drabovich, A.P., Berezovski, M., Okhonin, V. & Krylov, S.N. Selection of smart aptamers by methods of kinetic capillary electrophoresis. Analytical chemistry 78 3171 8 (2006). 64. Gragouda s, E. S., Adamis, A. P., Cunningham, E. T., Feinsod, M., & Guyer, D. R. Pegaptanib for Neovascular Age Related Macular Degeneration. New England Journal of Medicine 351 2805 2816 (2004). 65. http://www.bccresearch.com/report/BIO071A.html?utm_campaign=Global%20mar ket%20for%20aptamers%20worth%20%25241%252E8%20billion%20by%202014 &utm_content=tom@aptagen.com&utm_medium=Email&utm_source=VerticalRes ponse&utm_term=Global%20market%2 0for%20aptamerscampaign Last accessed June 14, 2012. 66. US Patent 5,141,813

PAGE 140

140 67. Breaker, R.R. Complex riboswitches. Science 319 1795 7 (2008). 68. Hermann, T. Adaptive Recognition by Nucleic Acid Aptamers. Science 287 820 825 (2000). 69. Baskerville, S., Zapp, M. & El lington, A.D. Anti Rex Aptamers as Mimics of the Rex Binding Element. J. Virol. 73 4962 4971 (1999). 70. Zimmermann, G.R., Jenison, R.D., Wick, C.L., Simorre, J. P. & Pardi, A. Interlocking structural motifs mediate molecular discrimination by a theophylline binding RNA. Nature Structural Biology 4 644 649 (1997). 71. Waring, M.J. Complex formation between ethidium bromide and nucleic acids. Journal of Molecular Biology 13 269 282 (1965). 72. Fan P., Suri A.K., Fiala R., Live, D. Patel D.J. (1996). Molecular recogn ition in the FMN RNA aptamer complex. J Mol Biol 258 480 500. 73. Jiang, F., Kumar, R.A., Jones, R.A. & Patel, D.J. Structural basis of RNA folding and recognition in an AMP RNA aptamer complex. Nature 382 183 6 (1996). 74. Lin, C.H. & Patei D.J. Structural basis of DNA folding and recognition in an AMP DNA aptamer complex: distinct architectures but common recognition motifs for DNA and RNA aptamers complexed to AMP. Chemistry & Biology 4 817 832 (1997). 75. Horn, W. T., Convery, M. A., Stoneh ouse, N. J., Adams, C J et al. The crystal structure of a high affinity RNA stem loop complexed with the bacteriophage MS2 capsid: further challenges in the modeling of ligand RNA interactions. RNA (New York, N.Y.) 10 1776 82 (2004). 76. Long, S.B., Long, M.B., White, R.R. & Sullenger, B.A. Crystal structure of an RNA aptamer bound to thrombin. RNA (New York, N.Y.) 14 2504 12 (2008). 77. Huang, D B ., Vu, D., Cassiday, L. A., Zimmerman, J. M. et al. Crystal structure of NF kappaB (p50)2 complexed to a high affi nity RNA aptamer. Proceedings of the National Academy of Sciences of the United States of America 100 9268 73 (2003). 78. Hopfield, J.J. Kinetic Proofreading: A New Mechanism for Reducing Errors in Biosynthetic Processes Requiring High Specificity. Proceeding s of the National Academy of Sciences 71 4135 4139 (1974). 79. Shangguan, D., Li, Y., Tang, Z., Cao, Z. C. et al. Aptamers evolved from live cells as effective molecular probes for cancer study. Proceedings of the National Academy of Sciences of the United St ates of America 103 11838 43 (2006).

PAGE 141

141 80. Mi, J., Liu, Y., Rabbani, Z. N., Yang, Z. et al. In vivo selection of tumor targeting RNA motifs. Nature chemical biology 6 22 4 (2010). 81. Noma T., Ikebukuro K., Sode K., Ohkubo T., et al. A screening method for DNA aptamers that bind to a specific, unidentified protein in tissue samples. Biotechnology letters 28 1377 81 (2006). 82. Fitter, S. & James, R. Deconvolution of a complex target using DNA aptamers. The Journal of biological chemistry 280 34193 201 (2005). 83. Hicke B.J., Marion C., Chang Y.F., Gould T., et al. Tenascin C aptamers are generated using tumor cells and purified protein. The Journal of biological chemistry 276 48644 54 (2001). 84. Washburn, M.P., Wolters, D. & Yates, J.R. Large scale analysis of the yeast proteome by multidimensional protein identification technology. Nature Biotechnology 19 242 247 (2001). 85. Bates, P.J. Antiproliferative Activity of G rich Oligonucleotides Correlates with Protein Binding. Journal of Biological Ch emistry 274 26369 26377 (1999). 86. Tang, Z., Parekh, P., Turner, P., Moyer, R.W. & Tan, W. Generating aptamers for recognition of virus infected cells. Clinical chemistry 55 813 22 (2009). 87. Bayrac A.T., Sefah K., Parekh P., Bayrac C., et al. In Vitro Select ion of DNA Aptamers to Glioblastoma Multiforme. ACS Chemical Neuroscience 2 175 181 (2011). 88. of DNA aptamers using Cell SELEX. Nature protocols 5 1169 85 (2010). 89. Mallikaratchy P., Tang Z., Kwame S., Meng L., et al. Aptamer directly evolved from lymphoma cells. Molecular & cellular proteomics : MCP 6 2230 8 (2007). 90. Berezovski, M.V., Lechmann, M., Musheev, M.U., Mak, T.W. & Krylov, S.N. Aptamer facilitated biomarker discovery (AptaBiD). Journal of the American Chemical Society 130 9137 43 (2008). 91. Shangguan D., Cao Z., Meng L., Mallikaratchy P., et al. Cell Specific Aptamer Probes for Membrane Protein Eluci dation in Cancer Cells research articles. Journal of Proteome Research 2133 2139 (2008). 92. http://clinicaltrials.gov/ct2/show/NCT00632242?term=ARC1779&rank=1 Last accessed Ju ne 14, 2012. 93. http://clinicaltrials.gov/ct2/show/NCT01191372?term=archemix&rank=6 Last accessed June 14, 2012.

PAGE 142

142 94. http://www.thepharmaletter.com/file/100055/baxter buys all hemophilia related assets of archemix for 30 million and 285 million in potential milestones.html Last accessed June 14, 2012. 95. Anglesio, M.S., Carey, M.S., Kbel, M., Mackay, H. & Huntsman, D.G. Clear cell carcinoma of the ovary: a report from the first Ovarian Clear Cell Symposiu m, June 24th, 2010. Gynecologic oncology 121 407 15 (2011). 96. Sugiyama T., Kamura T., Kigawa J., Terakawa N., et al. Clinical characteristics of clear cell carcinoma of the ovary: a distinct histologic type with poor prognosis and resistance to platinum bas ed chemotherapy. Cancer 88 2584 9 (2000). 97. Takano M., Kikuchi Y., Yaegashi N., Kuzuya K., et al. Clear cell carcinoma of the ovary: a retrospective multicentre experience of 254 patients with complete surgical staging. British journal of cancer 94 1369 74 (2006). 98. LoPiccolo, J., Blumenthal, G.M., Bernstein, W.B. & Dennis, P.A. Targeting the PI3K/Akt/mTOR pathway: effective combinations and clinical considerations. Drug resistance updates : reviews and commentaries in antimicrobial and anticancer chemotherap y 11 32 50 99. Kerschgens, J., Egener Kuhn, T. & Mermod, N. Protein binding microarrays: probing disease markers at the interface of proteomics and genomics. Trends in molecular medicine 15 352 8 (2009). 100. Provencher D.M., Lounis H., Champoux L., Ttrault M., et al. Characterization of Four Novel Epithelial Ovarian Cancer Cell Lines. In Vitro Cellular & Developmental Biology. Animal 36 357 36 (2000). 101. Scherer, W.F., Syverton, J.T. & Gey G.O. Studies on the propagation in vitro of poliomyelitis viruses. IV. Viral multiplication in a stable strain of human malignant epithelial cells (strain HeLa) derived from an epidermoid carcinoma of the cervix. The Journal of experimental medicine 97 695 710 (1953). 102. Zuker, M. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Research 31 3406 3415 (2003). 103. Kahn, S.D. On the future of genomic data. Science (New York, N.Y.) 331 728 9 (2011). 104. Wall L., Programming in Per l 3 ( O'Reilly Media 2000) 105. Larkin M.A., Blackshields G., Brown N.P., Chenna R., et al. Clustal W and Clustal X version 2.0. Bioinformatics 23 2947 2948 (2007). 106. Shamah, S.M., Healy, J.M. & Cload, S.T. Complex target SELEX. Accounts of chemical research 41 130 8 (2008).

PAGE 143

143 107. Svarovsky S.A. and Joshi L. Biocombinatorial Selection of Carbohydrate Binding Agents of Therapeutic Significance. Current Drug Discovery Technologies 5 20 28 (2008). 108. Faca V. M., Ventura A. P., Fitzgibbon M. P., Pereira Faca S. R., et al Proteomic Analysis of ovarian cancer cells reveals dynamic processes of protein secretion and shedding of extra cellular domains. PLoS ONE 3 : e2425(2008). 109. Vogelstein, B. & Kinzler, K.W. Cancer genes and the pathways they control. Nature medicine 10 789 99 (2004). 110. Garcia, M., Jemal, A., Ward, E., Center, M., Hao, Y., et al. Global cancer facts and figures. Atlanta, GA: American Cancer Society (2007). 111. Lawrenson, K. & Gayther, S.A. Ovarian cancer: a clinical challenge that needs some basic answers. PLoS me dicine 6 e25 (2009). 112. Cibiel, A., Dupont, D.M. & Ducong, F. Methods To Identify Aptamers against Cell Surface Biomarkers. Pharmaceuticals 4 1216 1235 (2011). 113. Hedin, L.E., Illerg, K. & Elofsson, A. An Introduction to Membrane Proteins. Journal of Proteome Research 10 3324 3331 (2011). 114. Collas, P. Chromatin Immunoprecipitation Assays. Methods in Molecular Biology 567 (2009). And 115. Solomon, M.J. Formaldehyde Mediated DNA Protein Crosslinking: A Probe for in vivo Chromatin Structures. Proceedings of the Natio nal Academy of Sciences 82 6470 6474 (1985). 116. Von Hippel, P.H. & McGhee, J.D. DNA protein interactions. Annual review of biochemistry 41 231 300 (1972). 117. Schmidt D., Wilson M.D., Spyrou C., Brown G.D., et al. ChIP seq: using high throughput sequencing to d iscover protein DNA interactions. Methods (San Diego, Calif.) 48 240 8 (2009). 118. Djardin, J. & Kingston, R.E. Purification of proteins associated with specific genomic Loci. Cell 136 175 86 (2009). 119. Van Simaeys D., Lpez Coln D., Sefah K., Sutphen R., et al. Study of the molecular recognition of aptamers selected through ovarian cancer cell SELEX. PloS one 5 e13770 (2010). 120. Orlando, V., Strutt, H. & Paro, R. Analysis of chromatin structure by in vivo formaldehyde cross linking. Methods (San Diego, Calif.) 11 205 14 (1997).

PAGE 144

144 121. Wortmann, A., Jecklin, M.C., Touboul, D., Badertscher, M. & Zenobi, R. Binding constant determination of high affinity protein ligand complexes by electrospray ionization mass spectrometry and ligand competition. Journal of mass spectrom etry : JMS 43 600 8 (2008). 122. Annesley, T.M. Ion Suppression in Mass Spectrometry. Clinical Chemistry 49 1041 1044 (2003). 123. da Silva, V.C.H. & Ramos, C.H.I. The network interaction of the human cytosolic 90kDa heat shock protein Hsp90: A target for cancer t herapeutics. Journal of proteomics 1 13 (2012).doi:10.1016/j.jprot.2011.12.028 124. Onuoha, S.C., Coulstock, E.T., Grossmann, J.G. & Jackson, S.E. Structural studies on the co chaperone Hop and its complexes with Hsp90. Journal of molecular biology 379 732 44 (2008). 125. Wang T. H., Chao A., Tsai C.L., Chang C.L., et al. Stress induced phosphoprotein 1 as a secreted biomarker for human ovarian cancer promotes cancer cell proliferation. Molecular & cellular proteomics : MCP 9 1873 84 (2010). 126. Walsh N., O'Donovan N., Kennedy S., Henry M., et al. Identification of pancreatic cancer invasion related proteins by proteomic analysis. Proteome science 7 3 (2009). 127. Zanata, S.M. et al. Stress inducible protein 1 is a cell surface ligand for cellular prion that triggers neurop rotection. The EMBO journal 21 3307 16 (2002). 128. https://www.qiagen.com/geneglobe/genesolutionview.aspx?ID=GS10963 Last accessed June 14, 2012. 129. Erlich R.B., Kahn S.A., Lima F.R., Muras A.G., et al. STI1 promotes glioma proliferation through MAPK and PI3K pathways. Glia 55 1690 8 (2007). 130. Lu X., Borchers A.G., Jolicoeur C., Rayburn H., et al. PTK7/CCK 4 is a novel regulator of planar cell polarity in vertebrates. Nature 430 93 8 (2004). 131. Noma T., Ikebukuro K., Sode K., Ohkubo T., et al. A screening method for DNA aptamers that bind to a specific, unidentified protein in tissue samples. Biotech nology letters 28 1377 81 (2006). 132. Jackson, V. Studies on histone organization in the nucleosome using formaldehyde as a reversible cross linking agent. Cell 15 945 954 (1978). 133. Shin B.K., Wang H., Yim A.M., Le Naour F., et al. Global profiling of the cell surface proteome of cancer cells uncovers an abundance of proteins with chaperone function. The Journal of biological chemistry 278 7607 16 (2003).

PAGE 145

145 134. Walsh N., Larkin A., Swan N., Conlon K., et al. RNAi knockdown of Hop (Hsp70/Hsp90 organising protein) dec reases invasion via MMP 2 down regulation. Cancer letters 306 180 9 (2011). 135. Frank, R. & Hargreaves, R. Clinical biomarkers in drug discovery and development. Nature reviews. Drug discovery 2 566 80 (2003). 136. Carrigan, P.E. Domain:domain interactions within Hop, the Hsp70/Hsp90 organizing protein, are required for protein stability and structure. Protein Science 15 522 532 (2006). 137. Buchner, H.W..L.M..J. Reviews of Physiology, Biochemistry and Pharmacology 151 1 44 (Springer Berlin Heidelberg: Berlin, Hei delberg, 2004). 138. Southworth, D.R. & Agard, D.A. Client loading conformation of the Hsp90 molecular chaperone revealed in the cryo EM structure of the human Hsp90:Hop complex. Molecular cell 42 771 81 (2011). 139. Americo, T. a, Chiarini, L.B. & Linden, R. Signa ling induced by hop/STI 1 depends on endocytosis. Biochemical and biophysical research communications 358 620 5 (2007). 140. Erlich R.B., Kahn S.A., Lima F.R., Muras A.G., et al. STI1 promotes glioma proliferation through MAPK and PI3K pathways. Glia 55 1690 8 (2007). 141. Partridge, J. & Flaherty, P. An in vitro FluoroBlok tumor invasion assay. Journal of visualized experiments : JoVE (2009).doi:10.3791/1475 142. Neckers, L. Hsp90 inhibitors as novel cancer chemotherapeutic agents. Trends in molecular medicine 8 S55 6 1 (2002). 143. Hagemann, T. & Lawrence, T. Inflammation and Cancer. Business 512 325 332 (2009). 144. Trepel, J., Mollapour, M., Giaccone, G. & Neckers, L. Targeting the dynamic HSP90 complex in cancer. Nature reviews. Cancer 10 537 49 (2010). 145. Clevers, H. The canc er stem cell: premises, promises and challenges. Nature medicine 17 313 9 (2011). 146. Agarwal, R. & Kaye, S.B. Ovarian cancer: strategies for overcoming resistance to chemotherapy. Nature reviews. Cancer 3 502 16 (2003). 147. Aguirre Ghiso J.A. Models, mechanisms and clinical evidence for cancer dormancy. Nat. Rev. Cancer 7 834 846 (2007). 148. Durante F. Nesso fisio pathologico tra la struttura dei nei materni e la genesi di alcuni tumori maligni. Arch Memor Observ Chir Pract 11 217 26 (18 74).

PAGE 146

146 149. Sell, S. Stem cell origin of cancer and differentiation therapy. Critical reviews in oncology/hematology 51 1 28 (2004). 150. Bapat, S.A., Mali, A.M., Koppikar, C.B. & Kurrey, N.K. Stem and progenitor like cells contribute to the aggressive behavior of hu man epithelial ovarian cancer. Cancer research 65 3025 9 (2005). 151. Zhang S., Balch C., Chan M.W., Lai H.C., et al. Identification and characterization of ovarian cancer initiating cells from primary human tumors. Cancer research 68 4311 20 (2008). 152. Szotek P .P., Pieretti Vanmarcke R., Masiakos P.T., Dinulescu D.M., et al. Ovarian cancer side population defines cells with stem cell like characteristics and Mullerian Inhibiting Substance responsiveness. Proceedings of the National Academy of Sciences of the Uni ted States of America 103 11154 9 (2006). 153. Lichtenauer U.D., Shapiro I., Geiger K., Quinkler M., et al. Side population does not define stem cell like cancer cells in the adrenocortical carcinoma cell line NCI h295R. Endocrinology 149 1314 22 (2008). 154. Gamb elli F., Sasdelli F., Manini I., Gambarana C., et al. Identification of cancer stem cells from human glioblastomas: growth and differentiation capabilities and CD133/prominin 1 expression. Cell biology international 36 29 38 (2012). 155. Pan, Y. & Huang, X. Ep ithelial ovarian cancer stem cells a review. International journal of clinical and experimental medicine 1 260 6 (2008). 156. Karbanov J., Missol Kolka E., Fonseca A.V., Lorra C., et al. The stem cell marker CD133 (Prominin 1) is expressed in various human gl andular epithelia. The journal of histochemistry and cytochemistry : official journal of the Histochemistry Society 56 977 93 (2008). 157. Quintana E., Shackleton M., Sabel M.S., Fullen D.R., et al. Efficient tumour formation by single human melanoma cells. Nature 456 593 598 (2008). Shackleton, M., Quintana, E., Fearon, E.R. & Morrison, S.J. Heterogeneity in cancer: cancer stem cells versus clonal evolu tion. Cell 138 822 829 (2009). 158. Quintana E., Shackleton M., Foster H.R., Fullen D.R., et al. Phenotypic he terogeneity among tumorigenic melanoma cells from patients that is reversible and not hierarchically organized. Cancer Cell 18 510 523 (2010). 159. Bidlingmaier, S., Zhu, X. & Liu, B. The utility and limitations of glycosylated human CD133 epitopes in defining cancer stem cells. Journal of molecular medicine (Berlin, Germany) 86 1025 32 (2008). 160. Katoh, K., Asimenos, G. & Toh, H. Multiple alignment of DNA sequences with MAFFT. Methods in molecular biology (Clifton, N.J.) 537 39 64 (2009).

PAGE 147

147 161. Scadden, D.T. Cancer s tem cells refined. Nature immunology 5 701 3 (2004). 162. http://www.cellbankaustralia.com/estore/productdetail.aspx?productid=245&catego ryid=133 Last acce ssed June 14, 2012. 163. Wang, Y., Sheng, Q., Spillman, M. a, Behbakht, K. & Gu, H. Gab2 regulates the migratory behaviors and E cadherin expression via activation of the PI3K pathway in ovarian cancer cells. Oncogene 1 9 (2011). doi:10.1038/onc.2011.435 164. http://www.celprogen.com/DS/36113 40%20 %20DS.pdf Last accessed June 14, 2012. 165. Eustace B.K., Sakurai T., Stewart J.K., Yimlamai D., et al. Functional proteomic screens reveal an essential extracellular role for hsp90 alpha in cancer cell invasiveness. Nature cell biology 6 507 14 (2004). 166. Koga F., Kihara K., Neckers L Inhibition of Cancer Invasion and Metastasis by Targeting the Molecular C haperone Heat shock Protein 90. Anticancer Res 29 797 807 (2009). 167. Banerji U., O'Donnell A., Scurr M., Pacey S., et al Phase I pharmacokinetic and pharmacodynamic study of 17 allylamino, 17 demethoxygeldanamycin in patients with advanced malignancies. J C lin Oncol 23 : 4152 4161, ( 2005 ) 168. Baba T., Convery P.A., Matsumura N., Whitaker R.S., et al. Epigenetic regulation of CD133 and tumorigenicity of CD133+ ovarian cancer cells. Oncogene 28 209 18 (2009). 169. http://www.celprogen.com/DS/M36113 40%20 %20DS%20 %20Human%20Ovarian%20Cancer%20Stem%20Cell%20Culture%20Serum%20 Free%20Media.pdf Last accessed June 14, 2012. 170. Englan d, R. & Pettersson, M. Pyro Q CpG TM : quantitative analysis of methylation in multiple CpG sites by Pyrosequencing. Nature Methods 2 (2005).

PAGE 148

148 BIOGRAPHICAL SKETCH Dimitri Van Simaeys was born in Tielt, Belgium. He earned his Industrial En gineering in chemistry, option b iochemistry degree with distinction at the Provicia le Industrile Hogeschool West Vlaan deren, Kortrijk, Belgium in 2004 He subsequently at the University of Ghent in 2005, also with distinction in molecular b iotechnology After this, he joined Procter and Gamble where he was working on global formulations for hand dish products, with a focus on strategic use of raw materials. In 2007, he joined the chemistry graduate program at the University of Florida where he obtained his PhD in Chemistry in 2012, under the tutelage of Dr. Weihong Tan.