Role of Gap Junctions in Glioblastoma Multiforme Brain Tumors

MISSING IMAGE

Material Information

Title:
Role of Gap Junctions in Glioblastoma Multiforme Brain Tumors
Physical Description:
Book
Language:
English
Creator:
Oli, Muna
Affiliation:
Dr. brent Reynolds
Publication Date:

Subjects

Subjects / Keywords:
Glioblastoma Multiforme
Gap Junctions
Complex adaptive systems
dye-transfer assay
treatment

Notes

Abstract:
Cancer is the second leading cause of death in USA, yet little is known about how a collection of cancer cells form primary tumors. Recently, scientists have started studying tumors as a complex adaptive system. A key premise of this thinking is that cancer cells interact and communicate via gap junctions, and that disruption of cell-to-cell communication should adversely affect tumors, though this has not been well studied. This study aimed to assess the presence of gap junctions in primary Glioblastoma cells, and test the effect of the gap junction inhibition on tumor growth and treatment. Using Glioblastoma cell lines and in vitro dye transfer assays, Glioblastoma cells were shown to communicate via gap junctions. In vivo tumors models were used using to assess effect of gap junction inhibition on tumor volume. Gap junctions were found in Glioblastoma cells, and results showed that gap junction inhibition drugs reduced tumor volume and increased lifespans in mice that were treated with gap junction inhibitor and Temozolomide, a common chemotherapy agent compared to those that only received chemotherapy. These results show that interruption of cell-to-cell communication can potentially disrupt interactions and adaptations among tumor cells, making them more vulnerable to chemotherapy.

Record Information

Source Institution:
University of Florida
Rights Management:
Copyright Muna Oli. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
System ID:
AA00027017:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Role of Gap Junctions in Glioblastoma Multiforme Brain Tumors By Muna Oli Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of Bachelors of Science Psychology University of Florida August 2014

PAGE 2

Abstract Cancer is the second leading cause of death in USA, yet relatively little is known about how a collection of c ancer cells form primary tumors Recently, scientists have started studying tumors as a complex adaptive system in which a tumor is composed of heterogeneous collection of individual cancer cells that interact with neighboring cells and adapt to their local environments A key premise of this thinking is that cancer cells are able to interact and communicate with each other via gap junctions and that disruption of cell to cell communication should adversely affect tumors as a complex system. However, whether or not and to what extent Glioblastoma Multiforme (GBM) Brain Tumors act as complex adaptive systems or if cell to cell communication among tumor cells exist has not been well defined This study aimed to assess the presence of gap junctions in primary GBM cells, and subsequently test the effect of the gap junction inhibition o n tumor growth and treatment. Using GBM cell lines and in vitro dye transfer assays, G BM cells were shown to communicate via gap junctions. In vivo s ubcutaneous and int ra cranial tumors models were used using the same GBM cell lines as used in in vitro. Gap junctions were found in GBM cells, and r esults showed that gap junction inhibition drugs reduced tumor volume, sensitized cells to chem otherapy and increased lifespans in mice that were treated with gap junction inhibitor and Temozolomide, a common che motherapy agent compared to those that only received chemotherapy. These results show that interruption of cell to cell communication can potentially disrupt interactions and adaptations among tumor cells, making them more vulnerable to chemotherapy.

PAGE 3

Rol e of Gap Junctions in Glioblastoma Multiforme Brain Tumors by Muna Oli Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of Bachelors of Science Psychology University of Florida August 2014 Thesis committee: Dr. Brent Reynold s (Department of Neurosurgery) Dr. Darragh Devine (Department of Psychology) Dr. Vinata Vedam Mai (Department of Neurosurgery)

PAGE 4

Dedication The vast majority of my research has focused on cancer treatment. This is an area of science that is actively pursued, but often slow in moving ideas or new procedures to bench to and succumbed to various c ancers. It is these people who motivate me to do more experiments, study harder and ultimately achieve my career goals in the hope that, one day we will be able to provide personalized, highly effective and less toxic treatments for patients. Thus, I would like to dedicate this work, as well as all my future efforts to Opa, Tante Ruth, Sabina, Karen and all others who have been stricken with cancer.

PAGE 5

Acknowledgments I would like to thank my family, mentors and friends for all their help throughout this pro cess. First of all, my mentor, Dr. Brent Reynolds, who enabled me to work in his lab for the last six years, trusting me into his lab when I knocked at his door as high school student. He has fully supported my sometimes crazy ideas and has mentored me and challenging me throughout, to become a better scientist and person. He has taught me a lot about research, life and science. Dr. Vedam Mai has also been crucial in teaching me procedures and helping me to further understand the project. Dr. Devine has gra ciously been my psychology advisor and mentor and remembers me as a young science fair student many years ago. He was my faculty sponsor for my research internship last summer in Okinawa, Japan and is a great source of information and advice. I would also like to thank Dr. Ben Dunn and the Howard Hughes Medical Institute Science for Life program at UF who helped fund this research. My friends and family have been an integral part of this process as well. They have helped keep me motivated, supported me thro ughout and helped edit my paper. I would have not been able to accomplish all of this without everyone mentioned here. I truly appreciate each

PAGE 6

i T able of C ontents Chapter 1: Introduction ................................ ................................ ................................ ........ 1 Purpose of Study ................................ ................................ ................................ ............ 1 Cancer ................................ ................................ ................................ .................... 1 Gap junctions ................................ ................................ ................................ .......... 4 Role of gap junc tion in cancer ................................ ................................ ................ 4 Statement of Hypothesis ................................ ................................ ................................ 5 Significance of Study ................................ ................................ ................................ ..... 7 Limitations/Assumptions of study ................................ ................................ ................. 7 Chapter 2: Methods ................................ ................................ ................................ .............. 8 Passaging primary Glioblastoma Cell Lines ................................ ................................ .. 8 Immunostaining for Flow Cytometr y ................................ ................................ ............ 8 Western Blot ................................ ................................ ................................ .................. 9 Temozolomide solubility and Dose Response Experiments ................................ .......... 9 In vivo experiments ................................ ................................ ................................ ........ 9 Processing Tumors from mice ................................ ................................ ..................... 10 Chapter 3: Results ................................ ................................ ................................ .............. 11 Gap Junction Presence in vitro ................................ ................................ .................... 11 Toxicity levels Dose response in vitro ................................ ................................ ...... 11 Drug targets ................................ ................................ ................................ .................. 12 In vivo experiments ................................ ................................ ................................ ...... 12 In vivo s urvival curves ................................ ................................ ................................ 1 3

PAGE 7

ii Chapter 4: Conclusion ................................ ................................ ................................ ........ 14 Discussion ................................ ................................ ................................ .................... 14 Application ................................ ................................ ................................ ................... 15 Future Studies ................................ ................................ ................................ .............. 15 Tables and F igures ................................ ................................ ................................ ............. 17 Tables .. ................................ ................................ ................................ ......................... 17 Figures ................................ ................................ ................................ .......................... 18 References ................................ ................................ ................................ .......................... 24 Appendix A: Acronyms ................................ ................................ ................................ ..... 26 Appendix B: Detailed Methods ................................ ................................ ......................... 28 Passaging primary Glioblastoma Cell Lines ................................ ................................ 28 Dye Transfer Assay and Immunostaining for Fluorescent Activated Cell Sorting ................................ ................................ ................................ .............. 28 Temozolomide ................................ ................................ ................................ ............. 32 In vivo experiments ................................ ................................ ................................ ...... 33 Excising and Processing Tumors from mice ................................ ............................... 33

PAGE 8

1 Chapter 1 : Introduction Purpose of Study Cancer It is estimated that approximately 23,000 new cases of brain cancer are diagnosed e ach year, and about 14,000 deaths annually are due to brain cancer (Cahoon, Inskip, Gridley, & Brenner, 2014) Mortality rate due to gliomas is almost 100%, with most patients living less than a year. About 50% of glio mas are Glioblastoma M ultiforme (GBM) tumors, one of the most dangerous and aggressive types of tumors (Levin, Leibel, & Gutin, 2001) As the fight to cure cancer continues, more treatments have become available, and mortality rates have declined fo r some forms of cancer. However, brain cancer mortality has declined only by 5% in the last 50 years (Legler et al., 1999) Despite substantial research efforts over the years, surgery and systemic therapies such as radiation and chemotherapy are still the most common treatments used. These treatmen ts are painful, expensive, and have many side effects such as burns, hair loss, fatigue, nausea, and immunosuppression. Furthermore, systemic cancer therapies are designed to target fast growing cells, whereas cancer stem cells (CSC) (Reya, Morrison, Clarke, & Weissman, 2001) actual stem cells in the conventional sense, they mimic many stem cell properties and exhibit slow cell division rates. CSC s are resilient and are resistant to most systemic therapies that are typically designed target fast dividing cells (Vezzoni & Parmiani, 2008) The resistance of CSCs against traditional treatments has led scientists to believe that surviving CSCs are the cau se of cancer relapses and metastasis (Liu et al., 2014)

PAGE 9

2 Given the fact that systemic cancer treatments such as chemotherapy and radiation are generally ineffective in GBMs, and that patients receiving such treatments suffer many serious, often debilitating, side effects, managing GBM as a chronic illness becoming ever more popular (Witter & LeBas, 2008) Scientists and oncologists are also exploring the possibility of making cancer cells more susceptible to conventional treatments so as to make such treatments effective at lower doses, which can also potentially minimi ze the undesirable side effects. The effectiveness of both of these approaches rely on our understanding of how collection of cancer cells form primary tumors, and how tumors, which are composed of heterogeneous cells, act as single entities to cause the disease. A recent and unorthodox thinking of a small population of cancer biologists and ecologists considers primary tumors as a complex adaptive system in which a tumor is composed of heterogeneous collection of individual cancer cells that interact with neighboring cells and adapt to their local environments; the tumor and disease it causes are viewed as emergent properties of such interactions and adaptations (Micheli Tzanakou, 2006) ch anging collections of distributed interacting components that react to both their environments and to one (Macal & Hummel, 2014) Figure 1 shows examples of CAS in animal populations. CAS are characterized by the ability of individual components to work together to create something much larger and stronger than each individual member. For example, in an ant colony, each ant is an individual and follows a simple set of decisions

PAGE 10

3 based o n its job (find food, leave and/or follow a scent trail). There is no master plan or blue print for the entire colony and no one ant is in charge of the others. Yet, the colony as a whole tends to act as a single entity, as if it were a single organism. An t colonies are adaptable, robust, complex and quick respond to environmental perturbations; this complexity and adaptability partially explains why it is so difficult to get rid of an ant pile. Evidence suggests that collective animal behaviors such as sta rling murmuration and fish schooling emerge from each individual following a set of simple rules, communicating and interacting with neighbors and adapting to environmental perturbations (Cavagna et al., 2010; Hemelrijk & Hildenbrandt, 2011) A key premise of this thinking is that cancer cells are able to interact and communicate with each other and with their micro environments, and that disruption of cell to cell communication should adversely affect tumors as a complex system. However, wheth er or to what extent GBM act as complex adaptive systems or if cell to cell communication among tumor cells exist or if cell to cell communication is critical to structure and function of GBM tumors remains unknown. The overall goal of my research was to t est two hypotheses pertaining to the existence of cell to cell communication in GBM brain tumor cells, and how important such communications are to conferring robustness of tumors to systemic therapies. Scientists think that tumor cells may be communicati ng through gap junctions with their immediate neighbors primarily Gap junctions are specialized channels that directly couple adjacent cells and permit the bidirectional passage of small molecules between the cytoplasm (Chipman, Mally, & Edwards, 2003)

PAGE 11

4 Gap junctions Gap junctions are made of connexins, which are in turn comprised of transmembrane domains (Figure 3). There are many different types of connexins which by molecular weight. These different connexins can be found in different cells throughout the body For example, Connexin37 (Cx37), Cx40 and Cx45 are known to play a role in the heart, blood vessels and lymphatic vasculature and Cx37 and Cx40 are found in endothelial cells (Biervert & Steinlein, 1999) (Molica, Meens, Morel, & Kwak, n.d.) Some drugs tend to target one connexin more than another, and certain connexins that play an essential role in the body should not be targeted (Roell et al., 2007) The exchange of intercellu lar metabolites, secondary messengers, nutrients, ions, and other hydrophilic molecules under 1kDa allow neighboring cells to communicate via gap junctions and respond as a system (Chipman et al., 2003) Gap junction intercellular communication has been shown to play a role in the regulation of processes involving cell growth, tissue diffe rentiation, maintenance of homeostasis, and embryo development (Yamasaki, Mesnil, Omori, Mironov, & Krutovskikh, 1995) Role of gap junction in cancer T he role of gap junction coupling in Glioblastoma and cancer in general is not well understood. Increasing recognition of the importance of gap junction communication as a major cellular function has led to the need to establish reliable and accurate methods for measuring intercellular communication (Oliveira et al., 2005) The dye transfer assay used in this study demonstrates a method for the assessment of gap junction in tercellular communication by use of a modified version of the dye transfer assay developed by

PAGE 12

5 Goldberg et al (Goldberg, J, & Naus, 1995) The advantage of the dye transfer assay demonstrated herein is that i t allows for whole population analysis along with accurate quantification and high sensitivity measurements through the use of flow cytometry. Subtle effects that drugs may have on gap junction communication can be detected, whereas with visually quantifie d assays, e.g, microinjection, small differences in dye transfer cannot be perceived. In addition, this method allows for a minimal amount of cells to be used by adopting the assay to a 96 well plate. Thus, the possibility of screening for drugs that alter gap junction intercellular communication becomes much more practical (Lin et al., 2002) Investigations of gap junction communication within Glioblastoma have given varying results between tumor samples (Lin et al., 2002) As tumor heterogeneity is a leading concern within cancer, evaluation of the r ole of gap junction coupling in tumor progression, treatment resistance, and as avenues for therapy is important for furthering our understanding of tumor biology. Statement of Hypothesis First, I tested the hypotheses that GBM tumor cells communicate via gap junctions. Gap junctions are specialized channels that couple adjacent cells and permit bidirectional exchange of intercellular metabolites, secondary messengers, nutrients, ions, and other hydrophilic molecules under 1kDa. Although gap junction interc ellular communication has been shown to play a role in the regulation of processes involving cell growth, tissue differentiation, maintenance of homeostasis, and embryo development, their roles in Glioblastoma and cancer in general is not well understood (Dere, 2012)

PAGE 13

6 If GBM tumors f unction as complex adaptive systems, then it is reasonable to assume that cell to cell communication is important in structure and function of GBM tumors. Thus, I hypothesize that chemical inhibition of gap junctions should adversely ess and adaptability and make cancer cells more vulnerable to chemotherapy. Specifically, I hypothesize that treatment with the gap junction inhibitor drug and chemotherapy will cause a larger reduction in tumor size than chemotherapy alone. If so, then on e would expect longer lifespans in model animals treated with gap junction inhibitors (and chemotherapy) compared those treated only with chemotherapy. Determining the existence of cell to cell communication in GBM brain tumor cells was tested in vitro i n cell cultures using primary GBM cell lines cultured in Dr. Reynolds laboratory. Once proof of concept was established and toxicity levels of each gap junction inhibitor drug are determined in vitro I tested the second hypothesis can disruption of cell to cell communication in the tumor cause disruption of the tumor population, sensitizing the tumor to cancer treatment using in vivo experiments using NOD/SCID ( non obese diabetic mice with severe combined immunodeficiency disease) mice as model sys tems, either with a subcutaneous tumor in the right flank or intracranial tumors using the same GBM cell lines used in the aforementioned in vitro experiments Test of these two hypotheses is critical in determining the existence and importance of cell to cell communications via gap junctions, and the possibility of improving the efficacy of systemic therapies to cure GBM cancers.

PAGE 14

7 Significance of Study If it is found that gap junctions are essential in communication and for tumor growth and resistance, it can be assumed that decreasing communication will sensitize cells. This could revolutionize the way that we study and treat tumors. Furthermore, this concept has wide application as it can be applied to most solid tumor cancers. It could potentially improv e efficacy, reduce side effects and increase our capability to treat tumors and cancer. Evaluation of the role of gap junction coupling in tumor progression, treatment resistance, and as avenues for therapy is important for furthering our understanding of tumor biology. This research could also lead to advancements in personalized medicine and the ability to treat each patient based on their specific cancer with an individualized approach. Limitations/Assumptions of study Though the concept of gap junctions has been around for 50 years, the concept of studying gap junctions to influence cancer proliferation is new. Consequently, there are not many studies examining this hypothesis in vitro and in vivo A large portion of the project was spent creating and op timizing protocols, determining the best way to analyze experiments and overcoming difficulties regarding drug solubility and toxicity

PAGE 15

8 Chapter 2: Methods Passaging primary Glioblastoma Cell Lines Passaging GBM cell lines was used in all experiments and it is important to work with healthy cells at the right cell density Primary GBM cell lines were cultured as neurospheres following the protocol in (Deleyrolle et al., 2011) Briefly seven days after initial plating, neurospheres are incubated in trypsin to dissociate neurospheres. C ells are then spun down, viable cells counted using trypan blue and replated or used for experiments using fresh media and EGF. Immunostaining for Flow Cytometry Immunostaining can be used to look at a variety of signals. In this study, immunostaining was main ly used to look at apoptosis using a preconjugated Caspase 3 antibody and analyzing fluorescence of the cell through flow cytometry. Immunostaining procedure is described in detail in Appendix A and in Bossy (Bossy wetzel & Green, 2000) Briefly, cells were di ssociated from neurospheres into single cell suspension using trypsin. Cells were spun down and resuspended in 4% Paraformaldehyde (PFA) and incubated to fix cell s. After 20 minutes, cells were spun down and resusupended in desired cocktail with antibodies and detergent if needed Cells were then incubated, washed and run through the BD LSRII flow cytometer using standard protocols. Dye Transfer Assay To determine if GBM cells communicate via gap junctions we developed a protocol employing flow cytometric analysis loading method to evaluate gap junction communication (Goldberg et al., 1995) Cells are placed in single

PAGE 16

9 suspension as described previously. Half the cell population is treated with calcein AM (a green dye), and the other half of the cell population is treated with Cell Trace Violet (a blue dye). Cells are then incubated together for 24 hours at a bout two million cells per milliliter to allow for direct cell to cell communicatio n, and are analyzed using a BD LSRII, at various hourly intervals. Full methodology can be found in Appendix A. Western Blot Western blots were pe rformed on the three GBM cell lines cultured in the lab to determine the amount of connexin 43 expression, a key indicator to evaluate presence of gap junctions usin g a modified protocol described by Yu (Yu et al., 2014) Briefly, GBM cells were lysed, and loaded with a protein marker mixture and loaded on the gel. The gels were blotted and membranes were transferred to a blocking buffer and agitated during incubation. Blots were washed and images were taken (Taylor, Berkelman, Yadav, & Hammond, 2013) Temozolomide solubility and Dose Response Experiments Temozolomide (TMZ ) is the most common chemotherapy agent used to treat GBMs. It comes in a powder and is not soluble in water, so must be dissolved in DMSO. However, because DMSO is toxic, it is dissolved at the highest concentration possible (20mg/mL) and then diluted using other solvents. See Appendix A for further details. In vivo experiments To deter mine the effect of gap junction inhibitors on tum or growth and survival in mi ce, in vivo experiments on mice were conducted. A ll animal experiments were carried out in NOD/SCID mice, under IACUC protocol number 201101502. The in vivo

PAGE 17

10 experiments used mice (n=5) : control, DMSO, TMZ, Gap Junction Inhibitor Drug A (GJID A ), and Drug B (GJIDB), and mice treated with a combination of GJID A and TMZ Mice were treated three times a week (Monday, Wednesday and Friday) and tumor volumes were measured five times a wee k (Monday through Friday) using calipers in three different axes to calculate tumor volume. Mice in the control group were treated with PBS, the mice in the TMZ group (to represent current/common therapy) were treated at 5mg/kg of ? which is the average do se used for in vivo experiments in laboratories. Mice in the GJIDA alone were treated with the appropriate dose as determined by the dose response experiments (delivered through the food? Drinking water?) The fourth group of mice were treated with a combi nation of TMZ and the GJIDA to look at possible synergistic effects. I ntracranial m ice were treated Monday, Wednesday, and Friday through intraperitoneal injections once the tumor reached 100mm 2 for the subcutaneous tumors, and two weeks after surgery for the intracranial experiments. Mice were sacrificed when tumor exceeded 1000mm 2 or if mouse became lethargic or ill. Processing Tumors from mice If the mouse had a subcutaneous tumor, the mouse was sacrificed via decapitation, and then tumor was manually removed from right flank using scalpel and tweezers. If tumor was intracranial mouse was perfused and then brain extracted. Tumor or brain was placed in glucose or PBS until further analysis. Tu mors were processed and then fixed for further downstream analysis. Details can be found in Appendix A.

PAGE 18

11 Chapter 3: Results Gap Junction Presence in vitro The first goal of this project was to determine if primary GBM cells communicate via gap junction s using the dye transfer assay. Figure 4A shows the two individual cell populations at time 0 of the experiment. Half of the cells were labeled with calcein (green dye) and half were labeled with cell trace violet (blue dye). I t can be seen that when these two populations were combined and immediately run through flow cytometry without incubation time two clear populations can be visualized in the sample ( Figure 4A ) Over time, the cell trace violet dyed cells take up the calcein via gap junctions. As the b lue dyed cells retain the green dye, a shift in the cell population can be seen as the blue cells shift right, indicating green dye within the cell as well. In 4B, it can be seen that about 20% of the cells showed dye transfer after six hours of incubation However, when the gap junction inhibition drug was added, it decreased dye transfer by about 50%, as seen in 4C. This demonstrates the presence of drug inhibitable gap junctions in primary GBM cells. Toxicity levels Dose response in vitro Next, dose response curves were done to establish the toxicity of the GJID. Figure 5 shows one example of a cell count toxicity assay done to establish the dose response curve of the GJID A As can be seen, it follows a relatively scale. Th ese experiments were very important to assist in determining the doses to use in later experiments. Doses that retained 70 80% viability of cells were used for subsequent experiments.

PAGE 19

12 After toxicity of the GJID were established, we wanted to discern if th ere were differences on the effect of the GJID on different cell lines. Three different GJID were tested on three different cell lines. It was important to use different cells lines because each GBM is different and reacts differently to the drugs. Further more, the GBM Cell Line S3 is TMZ resistant, making it more aggressive and difficult to treat. Figure 6 shows the data for these experiments. The data labeled GBM L0, GBM L2, and GBM S3 are the untreated controls for each respective cell line. The data sho ws that exposure of cells to GJID inhibited dye transfer between 25% and 90%, further proving the effectiveness of the drugs used. Asterisks indicate *P<0.05. Drug targets Figure 7 shows that over 80% of all cells in all three cell lines expressed connexi n 43, which has a molecular weight of 43 47 kDa. IHC was also performed for visual confirmation. Cell nuclei were stained with DAPI (blue) and connexin 43 (Cx43) was stained green. Figure 8 shows a very prominent presence of Cx43 in the cells indicating t hat Cx43 is found widely among GBM cells. In vivo experiments NOD/SCID mice were used for both int ra cranial and subcutaneous experiments. Figures 9, 10 and 11 show data from in vivo experiments done using the subcutaneous tumor model. In figure 9 we can clearly see the additive effective of the combination of TMZ and GJID A It is also interesting to note that the GJID decreased tumor growth, even without the chemotherapy drug.

PAGE 20

13 To further investigate this theory another in vivo intracranial experiment w as perfor med using the GBM Line S3 cells using a similar experimental flow as previously. While the results shown in Figure 10 may not be as significant as those shown in figure 9 an additive effect was clearly shown when TMZ was used in conjunction with GJID. In vivo s urvival curves Figure 11 shows a survival curve from the experiment. This provides evidence that GJID not only sensitizes GBM cells and reduces tumor volume, but it also allows increases survival of the mice. Here it can be seen that the GJID drug increased the life span about 10 days. Another gap junction inhibitor, GJI drug B, increased the survival by almost a month. Translating this to patients, the use of GJI D could both reduce to rate of tumor expansion and allow the patient to live longer.

PAGE 21

14 Chapter 4: Conclusion Discussion The data presented in this paper support the hypotheses laid out at the beginning of the paper. I tested the hypotheses that GBM tumor cells communicate via gap junctions. Through the dye transfer assays the w estern blot experiment and the in vitr o assays it was clearly established that Glioblastoma cells communicate via connexin 43 containing gap junctions and that the process can be inhibited by specific GJIDs The second hypothesis tested was that i f Glioblastoma cancer cells communicate via gap junctions, chemical inhibition of gap junctions should adversely affect tumors and make cancer cells more vulnerable to chemotherapy. The in vivo experiments clearly sup port the notion that chemical inhibition of gap junctions adversely affect s tumor growth and progression and make s cancer cells more vulnerable to chemotherapy. This was seen through tumor volume reduction in animals treated with GJIDA and TMZ and increas ed life span of mice treated with GJID Th ese data provide the necessary preliminary evidence to continue to explore o ther gap junction inhibit or drugs with and without combination of standard chemotherapy agents. The data also provide evidence that it ma y be possible that cancer can modeled and treated as a complex adaptive syste m. Through the use of interdisciplinary research and collaborations between cancer cell biologists, physicians and ecologists, this approach could revolutionize the way we study a nd treat cancer.

PAGE 22

15 A pplication As can be seen by the minute decrease in GBM mortality in the last 50 years, there is an acute need to develop effective treatments to treat GBMs. Inhibition of gap junctions has a lot of potential to not only slow tumor grow th, but to also sensitize cells to conventional treatment effectively. The fundamental concept regarding gap junctions and CAS is applicable to most types of cancer and has the potential to help treat and manage many types of cancer. Although the possibili for cancer is very slim, better understanding and management of the disease is the immediate goal. If physicians focused on keeping a tumors growth at bay, and worked on improving the quality of life while managing the cancer, rathe r than trying to resect the whole tumor and kill every fast dividing cell in the body, including cancer cell s through intense and long sessions of chemotherapy and/or radiation, the patient may be able to lead a longer, healthier and happier life. Future Studies As can be seen by the minute decrease in GBM mortality in the last 50 years, there is an acute need to improve the treatment of GBMs. As sequencing ability improves and cost goes down, the possibility of sequencing patients genomes and tumor biopsi es microenvironment (glucose levels, pH, etc) to further sensitize cells to c onventional

PAGE 23

16 treatments. The ability to target specific connexins related to tumor cells, and target cancer stem cells will allow for targeted and personalized treatment for cancer patients.

PAGE 24

17 Tables and Figures Table s Table 1 : Characteristics shared by both animal populations and tumor populations, making it a complex adaptive system. Animal population Tumor population Nonlinear dynamics Small perturbation leads to large effect Many animal populations show drastic non linear population dynamics, but it is not always understood why Uncontrolled proliferation and down regulated apoptosis of cancer cells Self organizatio n Order from chaotic interactions of individuals The organization of a bee hive or an ant pile. Murmurations and baitballs are self organized and not directed by a dominant individual or practiced as a group Individual cells cause tumor formation and growth Emergent properties Properties not obvious from properties of individuals Each individual follows a set of simple rules, which emerges as a single behavior of the system Individual cells may be susceptible, yet in a tumor they are robust Adaptati on Environment becomes encoded in rules governing structure/behavior by selection Adaptation based on environment, predators or a change of dynamics Tumors adapt to treatment by becoming resistant

PAGE 25

18 Figures Figure 1 : Examples of complex adaptive systems in nature. Figure 2 : The relationship between a CAS in an animal population and a tumor. Here we see different types of fish (or different jobs in an ant colony) directly relates to the heterogeneity of cells within a tumor.

PAGE 26

19 Figure 3 : A) Gap junctions are made out of connexins, which are made out of (B) transmembrane domains. Notice how the gap junctions allow for direct communication between the cells (Shl & Willecke, 2004) Figure 4 : Proof of gap junctions in GBM cells. A) Donor cells stained with Calcein (green) dye. Acceptor cells stained with C ell Trace Violet Dye (blue). 1:1 ratio of the two populations analyzed after mixing. Assessment of dye transfer between the donor and acceptor populations over a six hour time period. B) Control cells shows 20% dye transfer over 6 hours. C) Gap junction in hibitor drug decreased dye transfer by approximately 50%.

PAGE 27

20 Figure 5 : Dose response curves were conducted using GJIDA to understand the toxicity of the drug, and determine what concentration of the drug should be used for future studies. The x axis shows the log dose and the y axis shows percent viability in the cells. The dose chosen varied based on the gap junction inhibition drug used, but a dose that had about 70% viability was chosen. Figure 6 : This shows the effect of the gap junction inhibition drugs on cell to cell communication. The x axis shows the different cell lines and gap junction inhibition drugs used, and the y axis shows the percent of dye transfer seen through flow cytometry, as c ompared to the control. Once toxicity and ideal dose were determined, the effectiveness of the GJID on three cell lines were tested to examine the effectiveness of different drugs on various types of cells. The asterisks indicate a p value of *P<0.05 as co mpared to control cells.

PAGE 28

21 Figure 7 : (A) This is a representative gel which demonstrates the size of the connexin (~43kDa) on a Western blot. (B) Western blot analysis of Cx43 in GBM. Cx43 antibody and secondary anti mouse horseradish peroxidase antibody were used. Experiment was performed on three cell lines. As can be seen, over 80% of all the cells expressed connexin 43. Figure 8 : A representative GBM neurosphere stained for connexin 43 (green) and cell nuclei stained with DAPI (blue). (B) shows a magnified view of (A). As can be visually seen, the majority of the cells seem to express connexin 43. A B 50kDa 30kDa B A

PAGE 29

22 Figure 9 : In vivo data showing tumor volume over time. T MZ is the most common chemotherapy drug used to treat GBMs. The x axis shows days post randomization, that is also defined as two weeks after implanting the GBM cells into the tumor, and also when tumor volume measurements began. The y axis is the average tumor volume in mm 3 for that group of mice. GJI Drug A is effective at tumor knockdown and TMZ+GJI Drug A had an additive effect on tumor volume as compared to the control. Figure 10 : This shows the tumor volume in mice that were transplanted with the S3 cell line, which is resistant to chemotherapy. The x axis shows days post randomization, that is also defi ned as two weeks after implanting the GBM cells into the tumor, and also when tumor volume measurements began. The y axis is the average tumor volume in mm 3 for that group of mice. GJI Drug A shown to have an effect on TMZ resistant cell. In combination wi th TMZ, GJI Drug A was capable of sensitizing the resistant cells to TMZ.

PAGE 30

23 Figure 11 : In a more clinically relevant model, GJI Drugs A and B extend the life of mice implanted with brain tumors intracranially. The x axis shows days post transplant of the GBM cells. The y axis is the percent of the group (n=5) who were still alive. As can be seen, the gap junction inhibition drug A and B increased the life of the mice by about ten to thirty days.

PAGE 31

24 References Biervert, C., & Steinlein, O. K. (1999). Structural and mutational analysis of KCNQ2, the major gene locus for benign familial neonatal convulsions. Human Genetics 104 (3), 234 40. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10323247 Bossy wetzel, E., & Green, D. R. (2000). Detection of Apoptosis by Annexin V Labeling Methods in Enzymology (Vol. 322, pp. 15 18). ACADEMIC PRESS. doi:10.1016/S0076 6879(00)22004 1 Cahoon, E. K., Inskip, P. D., Gridley, G., & Brenner, a V. (2014). Immune related conditions and subsequent risk of brain cancer in a cohort of 4.5 million male US veterans. British Journal of Cancer 110 (7), 1825 33. doi:10.1038/bjc.2014.97 Cavagna, A., Cimarelli, A., Giardina, I., Parisi, G., Sant agati, R., Stefanini, F., & Viale, M. (2010). Scale free correlations in starling flocks. Proceedings of the National Academy of Sciences of the United States of America 107 (26), 11865 70. doi:10.1073/pnas.1005766107 Chipman, J. K., Mally, A., & Edwards, G. O. (2003). Disruption of gap junctions in toxicity and carcinogenicity. Society of Toxicology 71 (2), 146 53. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/12563100 Deleyrolle, L. P., Harding, A., Reynolds, B. a. (2011). Evidence for label retaining tumour initiating cells in human glioblastoma. 134 (Pt 5), 1331 43. doi:10.1093/brain/awr081 Dere, E. (2012). Gap Jun ctions in the Brain: Physiological and Pathological Roles Oxford: Elsevier Inc. Goldberg, G., J, B., & Naus, C. (1995). A pre loading method of evaluating gap junctional communication by fluorescent dye transfer. BioTechniques 18 (3). Hemelrijk, C. K., & Hildenbrandt, H. (2011). Some causes of the variable shape of flocks of birds. PloS One 6 (8), e22479. doi:10.1371/journal.pone.0022479 Legler, J. M., Ries, L. a. G., Smith, M. a., Warren, J. L., Heineman, E. F., Kaplan, R. S., & Linet, M. S. (1999). Brain and Other Central Nervous System Cancers: Recent Trends in Incidence and Mortality. JNCI Journal of the National Cancer Institute 91 (16), 1382 1390. doi:10.1093/jnci/91.16.1382

PAGE 32

25 Levin, V., Leibel, S., & Gutin, P. (2001). Neoplasms of the central nervous s ystem. In Cancer: Principles and Practice of Oncology. (pp. 2100 60). M. (2002). Connexin 43 enhances the adhesivity and mediates the invasion of malignant glioma cell s. Society for Neuroscience 22 (11), 4302 11. doi:20026450 H. H. (2014). Survival strategy during crisis, 13 (4), 528 537. Macal, C., & Hummel, J. (2014). Complex Adaptive Systems Decision and Information Sciences Retrieved from http://www.dis.anl.gov/exp/cas/index.html Micheli Tzanakou, E. (2006). Complex Systems Science in Biomedicine (T. Deisboeck & Y. Kres h, Eds.) (1st ed.). New York: Springer. Molica, F., Meens, M. J. P., Morel, S., & Kwak, B. R. (n.d.). Mutations in cardiovascular connexin genes, 1 71. doi:10.1111/boc.201400038.This Oliveira, R., Christov, C., Guillamo, J. S., de Board, S., Palfi, S., Ve Peschanski, M. (2005). Contribution of gap junctional communication between tumor cells and astroglia to the invasion of the brain parenchyma by human glioblastomas. BMC Cell Biology 6 (1), 7. doi:10.1186/1471 2121 6 7 Reya, T., Morrison, S. J., Clarke, M. F., & Weissman, I. L. (2001). Stem cells, cancer, and cancer stem cells. Nature 414 (6859), 105 11. doi:10.1038/35102167 Roell, W., Lewalter, T., Sasse, P., Tallini, Y. N., Choi, B. Fleischmann, B. K. (2007). Engraftment of connexin 43 expressing cells prevents post infarct arrhythmia. Nature 450 (7171), 819 24. doi:10.1038/nature06321 Shl, G., & Willecke, K. (2004). Gap junctions and the connexin protein family. Cardiovascular Research 62 (2), 228 32. doi:10.1016/j.cardiores.2003.11.013 Taylor, S. C., Berkelman, T., Yadav, G., & Hammond, M. (2013). A defined methodology for reliable quantification of Western blot data. Molecular Biotechnology 55 (3), 217 26. doi:10.1007/s12033 013 9672 6 Vezzoni, L ., & Parmiani, G. (2008). Limitations of the cancer stem cell theory. Cytotechnology 58 (1), 3 9. doi:10.1007/s10616 008 9166 8 Witter, D., & LeBas, J. (2008). Cancer as a Chronic Disease Report to Physicians

PAGE 33

26 Yamasaki, H., Mesnil, M., Omori, Y., Mironov, N., & Krutovskikh, V. (1995). Intercellular communication and carcinogenesis. Mutation Research 333 (1 2), 181 8. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10403898 Taylor, R. N. (2014). Reduced connexin 43 in eutopic endometrium and cultured endometrial stromal cells from subjects with endometriosis. Molecular Human Reproduction 20 (3), 260 70. doi:10.1093/molehr/gat087 Appendix A: Acronyms ABX Antibiotics BD LSRII A flow cytometer equipped with 4 lasers Calcein AM A cell permeant dye that can be used to determine cell viability CAS Complex adaptive systems Casp3 Caspase 3 Apoptotic cell marker CO 2 Carbon dioxide CSC Cancer stem cells DMSO Dimethyl sulfoxide EDTA Ethylenediaminetetraacetic acid EGF Epidermal Growth factor ETOH Ethanol FBS Fetal bovine serum FSC Forward side scatter light in a flow cytometer GBM Glioblastoma multiforme GJID Gap junction inhibition drugs

PAGE 34

27 IHC Imm unohistochemistry K D a Kilo Dalton KI67 Proliferation cell marker L0 A primary Glioblastoma multiforme cell line L2 A primary Glioblastoma multiforme cell line MBI Cell culture media MCT Medium chain triglycerides oil NGS Normal goat serum NOD/SCID Mice N on obese diabetic mice with severe combined immunodeficiency disease NSA Serum free basal medium for the culture of human neural stem cells PBS Phosphate buffered saline PE Phycoerythrin, florescent dye PFA Paraformaldehyde RPM Rotation per minute S3 A primary Glioblastoma multiforme cell line, TMZ resistant SSC Side scattered light in a flow cytometer T Trypsin TI Trypsin Inhibitor TMZ Temozolomide an oral chemotherapy drug x g Centrifugal force, centrifuge speed

PAGE 35

28 Appendix B: Detailed Methods Passaging primary Glioblastoma Cell Lines Primary GBM cell lines were cultured as neurospheres following the a standard protocol (Deleyrolle et al., 2011) Briefly, seven days after initial plating, neurospheres are incubated in trypsin to dissociate neurospheres. Cells are then spun down, counted and replat ed using fresh media and EGF. GBM Cell lines L0 A primary GBM cell line. It reaches confluency after about seven days, and is not known to be resistant to TMZ. L2 A primary GBM cell line. It reaches confluency after about seven days, and is not known to be resistant to TMZ. S3 A primary GBM cell line. It reaches confluency after about five days, and is resistant to TMZ. GBM Treatment s TMZ Temozolomide is a common chemotherapy agent used to treat patients with GBMs GJI Drug A Gap Junction Inhibitor Drug A is a name that was assigned to one of the drugs we used as a gap junction inhibitor. Due to proprietary reasons we are unable to use the actual name of the drug. GJI Drug B Gap Junction Inhibitor Drug B is a name that was assigned to one of the drugs we used as a gap junction inhibitor. Due to proprietary reasons we are unable to use the actual name of the drug. Immunohistochemistry staining of neurospheres DAPI (blue) Annexin 43 (green) Dye Transfer Assay and Immunostaining for Fluorescent Activated Cell Sorting 1. Prepare Cells from incubator. a. Put trypsin and trypsin inhibitor in water bath to warm up. b. Work with cells only in tissue culture hood c. Take flask with neurosphere cells (GBML0, treated with 30 L 2mg/mL TMZ and 20 L DMSO) out of the incubator. d. Label tube. Transfer cells into 15ml tube. Spin at 500rpm for 5min. Aspirate using glass pipette. e. Resuspend in 1mL trypsin (0.25% trypsin) pipette up and down to mix well. Place in water bath at 37 0 C for 3min f. Add 1mL trypsin inhibitor. Pipette very well. Should be single cell. If not, spin and repeat trypsin step. g. Spin at room temperature at 500rpm at 5min. Aspirate using glass pipette. h. Take out of hood. Resuspend thoroughly in 1mL 4% PFA.

PAGE 36

29 i. Transfer to 1.5mL Eppend orf tube. Incubate at room temperature for 20min. j. Label tube. Spin at 1200rpm for 5min. Aspirate using glass pipette. Resuspend in 1mL PBS. k. Do cell count: Take a new Eppendorf tube. Add 90 L trypan blue (0.4%), and 10uL cells (mix by pipetting up and down). Inject 10 L into hemocytometer. Count. If 1 quadrant has less than 50 cells, count all four quadrants and divide that number by 4, than multiply by 10 5 (ex: if cell count for 4 quadrants is 107, then calculation is (107/4)*(1x10 5 ) = 2.675 x 10 6 (~2.6 million cells) and you will have the number of cells in your 1mL of PFA. l. Do a cell count for each cell line separately. 2. Staining of Cell lines a. Compare cell lines: L0 negative control and treated L0. b. Label tube. Prepare 1 million cells in 1mL: Take x L L from L0 and transfer to new tube. Resuspend in PBS to fill up to 1 ml. Split each cell line into 2 samples. c. Using 15ml tube, add 2mL NGS/PBS/Triton for each cell line. Label tube. d. Retrieve Casp3 an d KI67 from antibody fridge in box. Make sure to keep in dark. Casp3 and KI67 are used at 1:1000. e. Add 2uL of casp3 and Ki67 to NGS/PBS/Triton. MIX WELL. Keep in dark 3. Stain samples a. Take 0.5 ml cell samples from before. b. Spin at 1200rpm 5min. c. Aspirate using glass pipette d. Resuspend one of each cell line in 1000 L NGS/PBS/TritonX/Casp3/KI67. e. Incubate at RT or 37C for 1hr in the dark. f. While incubating, label 4 FLOW CYTOMETRY g. Turn on FLOW CYTOMETRY machine. If flow computer is empty, prepare new experiment and label each sample on the computer beforehand. h. Spin at 1200rpm for 5 min. Aspirate using pipette. Resuspend in 500uL PBS and transfer directly into correlating FLOW CYTOMETRY tube. 4. Run FLOW CYTOMETRY a. Set up experiment. b. Try to collect as many data points as possible (> 10,000 cells per sample) c. Run unstained untreated cells first as baseline. Adjust FSC/SSC/FITC/PE d. Run stained untreated cells. Use separate proto col to collect FLOW CYTOMETRY samples.

PAGE 37

30 e. Run stained treated cells and unstained treated cells DO NOT CHANGE SETTINGS. Dye Transfer Assay The assay can be performed in various different size cluster well plates and conducted through a variety of time courses. A 96 well plate format is demonstrated in this procedure to illustrate the use of a low number of cells. Six 1 hour intervals are evaluated in conjunction with a negative control gap junction inhibited series. Any variation in assay scale should m aintain the same ratios. 1. Dissociation of Gliomaspheres into Single Cell Suspension a. Human gliomasphere culture is maintained using the neurosphere assay. At the appropriate time for passaging the gliomaspheres, collect the culture and transfer it to a steri le tissue culture tube. b. Centrifuge cells at 110 x g for 5 minutes. c. After discarding the supernatant, resuspend the gliomasphere pellet in 1 mL of 0.05% trypsin EDTA and incubate in a 37 C water bath for 2 5 min. d. Add 1 mL of trypsin inhibitor and gently pi pette up and down to ensure homogeneity and complete neutralization of trypsin activity. e. Centrifuge cells at 110 x g for 5 minutes. f. Discard the supernatant then resuspend the cell pellet in 1 2 mL of NeuroCult NSA Basal Medium. g. Perform a cell count by mi xing 10l of the single cell suspension with 90l of trypan blue. 2. Staining Donor and Acceptor Cell Populations a. Transfer an equal number of cells from the single cell suspension to two tissue culture tubes for staining of the donor and acceptor cell populations. b. Prepare the donor and acceptor staining reagents in a volume that will be added to the donor and acceptor cell suspensions to give a concentration of 10 6 cells/mL. The donor staining solution is prepared by diluting 1L of 2mM Calcein AM dye s tock solution per 1 mL of NeuroCult NSA Basal Medium. The acceptor staining reagent is prepared by diluting 1L of 5mM Cell Trace TM Violet dye stock solution per 1 mL of NeuroCult NSA Basal Medium. c. Add the donor staining reagent to the tube containing t he donor cell suspension to give a final working concentration of 2 M Calcein AM. Then, add the acceptor staining reagent to the tube containing the acceptor cell suspension to give a final working concentration of 5 M Cell Trace TM Violet dye. Both donor and acceptor cell solutions should now be at a concentration of 10 6 cells/mL d. Incubate both cell solutions for 8 minutes in a 37 C water bath protected from light.

PAGE 38

31 e. Quench both staining solutions by adding 4 5 times the original staining volume of NeuroCu lt NSA Basal Medium and then centrifuge at 110 x g for 5 min. f. Discard the supernatant, then resuspend the cell pellets in 5 mL of NeuroCult NSA Basal Medium with NeuroCult NSA Proliferation Supplement. Incubate in a 37 C water bath for 30 minutes to al low any excess dye to purge from cells. g. Next, centrifuge the cells at 110 x g for 5 minutes, and subsequently discard the supernatant. h. Resuspend both the donor and acceptor populations in 0.5 1 mL of NeuroCult NSA Basal Medium containing NeuroCult NSA Pr oliferation supplement, 1% fetal bovine serum (FBS), and 20 ng/mL epidermal growth factor (EGF). i. Perform a cell count for each cell suspensions by diluting 10 L cells in 90 L of trypan blue. 3. Combining Donor and Acceptor Cell Populations to Assess Gap Jun ction Mediated Dye Transfer a. Combine 1.2 x 10 6 cells from the donor population and 1.2 x 10 6 cells from the acceptor population to a new tissue culture tube, i.e. 1 x 10 5 cells from the donor population and 1 x 10 5 cells from the acceptor population for eve ry well required in the 96 well plate. b. Add enough prepared NeuroCult NSA Basal Medium containing proliferation supplement, 1% FBS, and 20 ng/ml EGF so that the final volume contains 2.0 x 10 6 cells/ml. c. Add 100 L of the combined donor and acceptor cell so lution into six wells of a 96 well plate, i.e. one well for every 1 hour time interval. Cells should be at a density of approximately 3.1 x 10 4 cells/mm 2 well surface area. d. Add enough of a suitable gap junction inhibitor to the remaining combined donor and acceptor cell solution in order to bring the inhibitor to its working concentration. e. Add 100 L of the combined donor and acceptor cell solution with gap junction inhibitor into six wells of the 96 well plate. f. Place the cluster well plate in a 37 C, 5% C O 2 humidified incubator and begin recording the time. g. At the necessary time intervals, add 100 L of 0.05% trypsin EDTA to the appropriate wells and incubate at 37 C for 2 5 minutes. h. Add 100 L of soybean trypsin inhibitor and then harvest cells into a mi cro centrifuge tube. Centrifuge cells at 110 x g for 5 minutes and subsequently discard the supernatant. i. Resuspend cells in 0.5 mL of a 4% formalin solution and allow cells to fix for 15 minutes before centrifuging at 185 x g for 5 minutes. j. Finally, wash cells with 1 mL phosphate buffered saline (PBS), and store in PBS protected from light until ready for evaluation by flow cytometry.

PAGE 39

32 4. Detection of Gap Junction Intercellular Communication by Flow Cytometry a. Make sure suitable fluorescent detectors are activa ted on the flow cytometer to record the emission spectra for both Calcein and Cell Trace Violet dye. In this procedure, the 530/30 and 450/50 nm bandwidth detectors on a BD LSRII flow cytometer are used, shown in the video as Pacific Blue and CSFE detector s. In addition, activate Forward Scatter and Side Scatter Height and Width values to be recorded. b. Construct a plot of side scatter area (SSC A) versus forward scatter area (FSC A), then run a non stained sample of cells through the flow cytometer. Adjust SSC A and FSC A voltage settings as needed to center the cell population in the plot. c. Construct a plot of Cell Trace Violet dye versus calcein and run one of the 1:1 ratio samples through the flow cytometer. Adjust the voltage settings accordingly so that both the non stained cell population and stained donor and acceptor populations are visible within the plot. Bi exponential axes may need to be applied to the graph. d. Record flow cytometry events for non stained control cells, for the six hour time interva l series of donor and acceptor populations, and for negative control gap junction inhibited time trials. e. Analyze flow cytometry data by gating the main cell population. Next, select for singlet cell events within the main cell population by constructing a plot of Side Scatter Height (SSC H) versus Side Scatter Width (SSC W) and gating the singlet event population, see video. Analyze the selected population further by constructing a plot of Forward Scatter Height (FSC H) versus Forward Scatter Width (FSC W) and gating for the singlet event population, see video. f. Based on the first 1 hour time interval, construct a quadrant gate that excludes the donor and acceptor populations from the top right quadrant, as shown in Figure 3 B. Apply this gate to all other samples and controls throughout the six hour time period. Using the 1 hour time interval as the starting point for dye transfer allows for any initial dye present to be taken into account. g. Calculate the percent dye transfer to the acceptor population by di viding the number of acceptor cells obtaining calcein dye by the overall number of cells in the acceptor population. Temozolomide Caution: TMZ should be treated as a carcinogen use appropriate caution. Reconstitution should be done in the hood. The half life of TMZ is very short (less than an hour) so you must work quickly for this. 1. Carefully open the TMZ bottle. Try to make sure the powder is not stuck to the top cover. 2. Pipette 5mL of DMSO into the TMZ bottle. DO NOT PIPETTE or stick the pipette tip al l the way into the TMZ bottle.

PAGE 40

33 3. Sterilize a stir bar (using bleach/ethanol/DI water) and place into the TMZ bottle. 4. Replace cap of TMZ bottle 5. Place on spin plate for ~10min. 6. Ensure that the TMZ is completely dissolved. 7. Dilute (in medium, DMSO, MCT, etc) and aliquot. Label aliquots carefully. 8. Freeze in the 20 0 C as soon as possible. In vivo experiments All animal experiments were carried out in N on obese diabetic mice with severe combined immunodeficiency disease (NOD/SCID mice), approved by IACUC protocol number 201101502. The in vivo experiments used various treatment groups (n=5): control s DMSO, GJID, TMZ+GJID. Mice were treated Monday, Wednesday, and Friday through intraperitoneal injections once the tumor re ached 100mm 2 for the subcutaneous tumors, and two weeks after surgery for the intracranial intracranial experiments. Mice were sacrificed when tumor exceeds 1000mm 2 or if mouse became lethargic or ill. Tumors were processed and then fixed for further analy sis using above methods. Excising and Processing Tumors from mice Excising tumors 1. Kill mouse by appropriate method 2. One animal is sacrificed, it is extremely important to move quickly and efficiently. 3. Extract tumor using tweezers, scissors and scalpel. Sp ray down area around tumor with ethanol, ensuring the fur is wet. Make an incision rostral to the tumor. Cut the skin around the tumor until the tumor is fully exposed. Carefully using the scalpel or scissors, beginning cutting the connective tissue and fi laments between the tumor and skin, and tumor and muscle (it is easier to separate tumor from skin first). Once the tumor is completely cut away, make sure there are no hair on it. Cut the tumor in half. Take one half of the tumor and cut in half again, pu t 2 x pieces into labeled cryotubes. Take the other half of the tumor and place in a 50ml conical tube with 1ml PBS 4. Snap freeze tumor. Fill up a durable container with liquid nitrogen. Place cryotubes in liquid nitrogen for ~30 45 seconds. Remove cryotub es from liquid nitrogen with long tweezers and immediately place in 80 0 C freezer. 5. If processing more than 1 tumor, place cryotubes on dry ice until all tumors are processed, then transfer all samples to 80 0 C freezer at once. Label each tube carefully. Tumor processing 1. Process tumor into single cell suspension. Take a 50mL conical tube, put 2mL trypsin inhibitor in it. Put a 20 micron filter on the top. Put on ice. Place the petri dish on ice. Place tumor in petri dish (no PBS). Using scalpels, chop up t umor into pulp. After 5min of vigorous chopping, add 0.5mL trypsin. Keep chopping. Once in a fine pulp, cut the tip off of a 1mL pipette and transfer the tumor pulp to the filter. Agitate the tumor pulp and push through filter, adding ~1mL trypsin if neces sary. If tumor is still

PAGE 41

34 and keep cutting. 2. Transfer trypsin, trypsin inhibitor and cells into 15mL conical tube. 3. Spin at 1200rpm 5min. Remove supernatant and transfer in to new 15ml conical. 4. Spin at 1200rpm 5min. While spinning, add 1mL PFA to cell pellet and resuspend pellet. 5. Remove supernatant and add 1mL 4% PFA to second tube. Transfer the new pellet in PFA to the original tube (already containing the original cell pell et and PFA). There should now be single cells in 2mL of 4% PFA. 6. Place on rocker and incubate at room temperature for 30min. 7. Spin at 1200rpm 5min 8. Remove supernatant and resuspend 1mL PBS. Use cells for a variety of analyses.