Reading Scientific Literature
Tips on Paper Writing and Submitting
1. Papers require abstracts, introductions, materials and methods, results and discussions.
2. Abstracts describe the point of the paper in a precise short paragraph.
3. Introductions are a summary of the data within the paper itself and tells what is known already. They contain all the material leading to the main question discussed in the paper. These should be written to reel readers in.
4. Materials and methods are there so that the work done can be duplicated elsewhere as a verification. Statistical validity retains the minor possibility of a reproducing error. Nonreproducibility is bad; however, clinical studies can necessarily involve that.
5. Results are largely factual and outline what was garnered from the experiments. They describe flaws and what is hiding clarity. It also includes why different experiments were done (which are usually leadoffs from one that didn�t quite work).
6. Discussions are interpretations of results and can include future experiments and other related thoughts. These are not overly comprehensive. Key word: future implications.
7. Proof is in the figures. If they cannot be deciphered or are unenlightening, than the paper�s thesis can�t really be proven or isn�t being proven. Figures are done first as a focus and then text is filled in later.
8. Papers are given peer reviews and then an editor review when being submitted for publication. Suggestions include: do more experiments, include more figures, rewrite the paper, or refer to certain sources.
9. High impact publications include Nature, Cell, Science, etc.
10. Journals have different emphases and use many differing and contrasting styles and angles. Also, publishing requires paying for color pictures, space and other assorted frills on the article.
11. Spell check. Make sure transitions work.
12. Critiques should be done according to information known at the time. These should describe not only what is wrong but what is missing.
13. Clinical papers address efficacy of treatments in patients. They ask if treatments work, why would one want to continue, where does the treatment fit, whether or not it can lead to cures, better approaches, believability, etc.
Cell Biology
1. During development, cells divide in a controlled manner. During cancer, tumors grow when cells start to proliferate without regulation.
2. Cell division cycles have 4 main phases in which there are checkpoints that regulate progression to the next phase. During S-phase, all the DNA in the cell is duplicated. Before going to the G2-phase, the cell makes sure that the DNA has been copied correctly; if there is an unrepairable mistake, cell division stops.
3. DNA is packaged in a double helix which is made up of paired bases (ATGC) which are either purine or pyrimidines. Looking closer, these lines are wrapped around histone proteins and crammed into tightly wound chromatin. One whole piece is called a chromosome. DNA exists in antiparalell double strands, of which one (5� � 3�) is called �sense� and the other direction �antisense�. In papers, 5� � 3� is assumed.
4. During mitosis, chromosomes line up on the metaphase plates, whereupon sister chromatids separate and move to opposite poles.
5. If DNA can�t be repaired or chromosomes do not separate properly, cells undergo apoptosis, or cell suicide. Mistakes are caused by point mutations, single substitutions (including missense, nonsense and splices), insertions, deletions, duplications, and translocation. These are important because mutations can change gene function, cause differential expression which can change regulation or can make a whole new hybrid gene. Chromosomes can also be altered or lost through nondisjunction, mitotic recombination and gene conversion.
6. So if these checkpoints exist, what causes cells to grow uncontrollably and form tumors?
7. Proto-oncogenes are proteins which normally induce growth and proliferation in response to various stimuli but can become mutated so that they become constitutively active. The most famous example is Ras.
8. In normal cells, Ras is activated in response to extracellular stimuli. In cancer cells, Ras is always on, meaning this growth pathway is unregulated.
9. Some proteins function to stop cell growth, mainly in the checkpoints, and if these are mutated, cells can no longer arrest cell proliferation. P53 and pRb, the retinoblastoma protein, are the most common examples.
Statistics
1. Validity: External: do the results apply to everyone?, and Internal: did the study reach the correct conclusions for the study population?
2. Bias: error in study design or execution; types are selection, spectrum, length/time; lowered through double masked placebo trials (investigator and testee don�t know who is getting the placebo); inclusion criteria selected beforehand to avoid selection bias; ethics
3. Confounding: effect from coming into the office or some other linkage between subjects; can be lowered through randomization of trial cohorts
4. �If this doesn�t help you, don�t worry, it�s a placebo.�
5. False positive form p-value and false negative from study power.
6. Probability: measure of chance; sharing a birthday with me (0.08) but sharing a birthday with anyone (0.50), and in a room of 50 (0.97)
7. Chance: inferences on whole group by studying sample; the smaller the group size though, the less likely to get a representative sample by chance alone.
8. Hypothesis testing: build hypothesis and rephrase as a null hypothesis, then quantify the degree to which the results are consistent with the null hypothesis.
9. Your results are correct if you repeat the experiment and you get the same answer. However, if you get a difference where it was supposed to be no difference, you have a Type I or false positive. If you get no difference where there should be a difference, you have a Type II or false negative.
10. P-value: probability of outcome being produced by Type I error. If p = 0.05, then if the null hypothesis were true, then there is a 5% chance of observing the study result or a more extreme result. This doesn�t mean that there is a 95% chance the null is false, or a 5% chance that it is true. P-values do not determine truth. P << .05 does not mean that the effect under study is very strong. P > .05 does not show significance.
11. Power: ability of study to actually see an effect if one exists. Sample size is predetermined to minimize Type I error and Type II error (lowered with higher power study).
12. P-values are proportional to both the size of the effect and the amount of variability in the study result. Large studies tend to have smaller P-values because they have smaller variability. Very large studies can find statistically significant results that are not applicable in any other setting.
13. P < .05 does not apply to the entire study, only to the individual result of the study. So, if you test 20 different outcomes in a study, 1 in 20 will be positive just by chance alone and therefore the probability of finding at least one positive result by chance alone is 1.
Leukemia and Cancer
1. Leukemia means �white� and �in blood�, and is a cancer of the white blood cells in the bone marrow, where they are made. This causes a dysfunction of the immune system and is symptomized by different infections.
2. Hodgkin�s Disease causes swollen lymph nodes (lymph is filtered from the blood by the spleen) which are filled beyond capacity with T cells and B cells. Acute leukemia won�t cause swollen lymph nodes but mid-level leukemias might. For chronic myelogenous leukemia, neutrophils are flooding the bone marrow.
3. Differentiation is important to remember. T cells and B cells (lymphocytes) share a common ancestor. Its progeny differentiate into white blood cells which then turn into neutrophils, T-cells, etc. Basophils and mast cells combat allergies.
4. Other symptoms include easy bruiseability. Clotting is done by platelets which come from the bone marrow. In leukemia there is a lack and thus bruises come fast. This is also called thrombocytophenia. Pancytophenia is the lack of all normal cells.
5. Also malaise, or fatigue. Oxygen problems caused by red blood cells and anemia brought on by bad tenants of the bad marrow can cause fatigue. Also, there is an increased energy requirement, put into repair of damage done by the influx of white blood cells, and the energy needed to make all the cells. This can by hypoglycemia and lead to weight loss and unexplainable fevers, which stem from an uncontrolled immune response.
6. To help with the fatigue and anemia, Procrit is usually given. It is an erythropoietin (poeisis is growth) which stimulates red blood cell growth. Chelation therapy is done to remove heavy metals from the blood but this leads to osteoporosis, because while iron and other metals are being removed, so is calcium.
7. Radiation is used to treat cancer. Hair, blood and tract lining cells divide rapidly in the normal state, so those die off with the cancerous dividing cells. That�s why hair falls out during chemotherapy. The GI tract cell multiply incredibly fast and after radiation, those are mainly killed, causing vomiting and loss of appetite. This also causes anemia because it kills red blood cells.
8. Granulocytic stands for the granulin cell which is basically a neutrophil.
9. CML stands for any chronic disease to bone marrow.
10. There is an overlap between leukemia and lymphoma.
11. A satellite is highly repetitive DNA found close to the centromeres and telomeres. They are put there for a purpose; probability of finding repeat DNA is 4 to the power x. For a short 45-base pair, the odds are 1 to 4 to the 45th power. Translocation between satellite arms are easy because of similar patterns and increased pairing.
12. A blast crisis occurs when there are too many neutrophils. This is the transformation from acute ML to chronic ML.
Nowell and Hungerford. �Chromosome Studies in Human Leukemia: Chronic Granulocytic Leukemia.� J. Natl. Cancer Inst. 27: 1013-35, 1961.
A chromosomal study of 10 patients with chronic granulocytic leukemia found that a majority of the cases contained the same similar chromosome abnormality, a reduction in size of one of the four smallest acrocentric chromosome in the leukemic cells, due to loss of half of an arm. The abnormality was given the designation Ph. The study was conducted by doing chromosome counts and karytotype analyses of short term cultures of peripheral blood cells, and in a few cases, direct preparation of bone marrow. The blood was collected from patients ages 8 and 90 who had had chronic granulocytic leukemia. A majority of the counts returned the optimum number of 46 chromosomes. The Ph chromosome was not present in Case 144T, who was in the midst of an acute �blast� crisis and thus not representative, and in a 72-hour culture of Case 139T, because dividing leukemic cells tend to be less frequent in 72-hour cultures. The Ph chromosome seems to be a chromosome 21 which lost a portion of its longer arm. The researchers have not yet discovered if the mechanism of this change is translocation or deletion; they seem to be more for the idea of translocation. They also discuss the idea that this abnormality may be a primary change rather than a secondary symptom. This aberration is of a type known to be produced by ionizing radiation or other mutagenic agents, which could be the cause of this disease. Another consistent chromosome abnormality was also discovered, probably caused by extensive radiation therapy. In some cases, other chromosome changes, apparently non-specific, were observed. Implications of this study include a setting apart of this form of leukemia from the others because of a demonstrable specific chromosome abnormality, as well as a reevaluation of theories of leukemogenesis.
1. Previously, in 1956, it was determined that there were 46 chromosomes to be mixed and matched.
2. This paper was at the start of a quest to find what caused cancer; Nowell presented a genetic link.
3. Chromosomes were karyotyped by centromere and by the p, q method. Errors usually stemmed from artifacts and higher magnification at the end of the slide.
4. In this paper, chronic granulocytic leukemia was connected with the Philly chromosome, thought to be a mutated chromosome 21. The Philly chromosome is linked to the creation of B cells which can flood the bone marrow (for leukemia). For breast cancer, the linked gene is BCRA1. For colon cancer, the linked gene is APC.
5. ABL is a tyrosine kinase that can turn into an oncogene with one base mutation. Signaling can be continued indefinitely and if it regulates mitosis, would cause unstopped mitosis.
Ben-Neriah et al. �The Chronic Myelogeneous Leukemia-Specific P210 Protein is the Product of the bcr/abl Hybrid Gene.� Science. 4760 (233): 212-14, 1986.
1. Chronic myelogenous leukemia is associated with a chromosomal translocation (entitled the Philadelphia Chromosome) that is from a fusion of the c-abl locus on chromosome 9 to the bcr locus on chromosome 22.
2. c-abl encodes for a tyrosine kinase, an enzyme that phosphorylates tyrosines. Bcr is an abbreviation of breakpoint cluster region.
3. Previously, they had concluded P210 was active in CML. Here, they conducted immunoprecipitation experiments to find out if the bcr/abl complex created P210 as one of its products.
Signaling and Signal Transduction
1. How do proteins talk to each other? How do signals from outside the cell lead to responses inside the cell?
2. Extracellular signals: scaffolding proteins: the body is held together by an extracellular matrix of collagen and other proteins � these can be signals depending on how tight they pull; ion gradient: pumps embedded in the cell membrane pass along Cl, K, H and Na as signals; surface receptors: these bind growth factors and other proteins floating around and wanting access; and G-protein coupled receptors: which are the most abundant in the cell and pass through the membrane 7 times (a 7-trans membrane protein).
3. Proteins that bind the surface can set up a signaling cascade which starts a branching effect. Advantages of a cascade include: routes can change if one route is faulty and the signal can be amplified by introducing more intermediaries. Binding of protein to one receptor can cause different cascades to start up; the conformational change of one specific protein binding can cause a protein-specific reaction.
4. A kinase phosphorylates and a phosphatase dephosphorylates (enzymes). Protein activation is most commonly done through phosophorylation. Its addition to serine, threonine or tyrosine can change protein conformation.
5. Amino acids are numerous and are the building blocks of protein. They are abbreviated with either a 1 or 3 letter abbreviation. Further addition of another amino acid abbreviation after the name of one signifies a mutation. Eg. S-173a, serine 173 mutated to alanine.
6. Plasmids are circular pieces of DNA that contain DNA encoding for a particular gene. Average length is 4 kb. They are engineered to have restriction sites where DNA can be inserted. A resistance to ampicillin is also added so that exposure to the antibiotic kills all other bacteria. Transfection is the �infection� of bacteria with the plasmid so that it can be copied, or amplified. Lysing and vortexing of the bacteria yield the wanted DNA.
7. Blotting is done so that the proteins or nucleic acids can be spread by electrophoresis. Identification is done by size. Pictures are then developed from the stains, which are usually pretty clear.
8. Southern and Northern blots are for DNA and RNA respectively. Agarose is used, a porous gel on which an electric field is applied. Location and presence of the nucleic acids are determined by application of a radio labeled probe (p32). The DNA travels to the positive end, lightweights moving faster and farther than the heavier ones. Initial direction is important; the wrong way will make your DNA slide right off. A membrane is then placed on top which gets the DNA trapped through capillary action. X-ray film then develops the radioactive signals.
9. RNA is easily degraded through RNAses which are present on human skin.
10. Positive controls are other stuff that could possibly bind, and negative controls include water, which doesn�t contain proteins or nucleic acids.
11. Western blots use acrylimide gels because proteins are smaller and would fall through agarose pores too fast. Antibodies are used to bind and identify proteins; these are collected from provoking an immune response to the protein and gathering antibodies made from an animal�s blood serum.
12. Positive or negative depends on the protein�s side chains and structure. SDS is used to turn the overall charge to a negative value, as it is a highly negatively charged detergent. This also denatures protein to protein bonds, losing all protein interactions.
13. Proteins are treated with the primary identifying antibody and a secondary dying antibody. Horseradish peroxidase is used a lot because it gives off light easily, exposing the film and making a band. Secondary antibodies are really helpful in amplifying signals and conjugating many different proteins that you are studying, thereby saving time and money. On an antibody, there is a constant region and a changing variable region. Secondary antibodies probe and id only the constant regions.
14. Non-specific binding can also occur.
15. Antibody against Protein Y: anti protein Y; or Made in Rabbit against Sheep: rabbit anti-sheep.
16. Blotting doesn�t show protein interactions though, so we turn to other methods.
17. Immunoprecipitation: To a mixture of different proteins, add a specific antibody to bind protein X. Beads coated with a protein that would bind the protein X-antibody complex are added, and then centrifuging is done. The beads are heavy enough to sink to the bottom so the superfluous proteins can be removed. What is left is either Protein X or Protein X combined with Y. If Y shows up after you run a gel, it is evidence that X had bonded with Y at some point (coimmunoprecipitation).
18. This isn�t fully conclusive. Weak interactions, or the fact that the antibody interacts with both proteins, or the presence of a go-between protein, or the presence of �sticky� protein intruders � these all can screw up certainties. This test does show interaction though, whether direct or indirect.
19. To test whether Protein Y does belong, one could test Protein Y with no antibody and just the bead, to test if the bead is capable of binding anything. If so, then Protein Y could be binding to the beads and maybe to Protein Y as well, skewing the numbers. One could also try coating the bead with a different antibody to test for specificity.
20. Reporter genes help measure transcriptional activity related to certain extracellular signals. Commonly, one could try activating transcription of glowing proteins (GFP, luciferase, Lac-Z). Chloramphenicol (CAT) is a common reporter gene.
21. Transcriptional Activation assay: plasmid has a reporter gene downstream of the transcription factor consensus sequence, which are usually glowing proteins or CAT. The activity of the gene is measured to determine transportation activity. Using a radioactive substrate, it would be easier to tell how much activity there was. Transfer is evident if a thin layer chromatograph revealed radioactive materials scattered.
22. The typical plasmid, along with the gene in question, will contain an AP-1 promoter, CAT and the ampicillin resistance gene.
Predergast et al. �BCR-ABL-Induced Oncogenesis is Mediated by Direct Interaction with the SH2 Domain of the GRB-2 Adaptor Protein.� Cell. 75:175-181, 1993.
BCR-ABL is an oncogenic protein that is derived from translocated tyrosine kinase sequences and the breakpoint cluster sequence from chromosome 9 and 22 respectively, and is involved in the pathogenesis of Ph human leukemias. Its existence in complex with an adaptor GRB-2 protein, results in an accumulation of an active form of Ras, which is required for mitogenic signal transmission as well as transformation by oncogenic kinases.
This paper demonstrates their relationship through coimmunoprecipitation of the proteins from immunoprecipitate mixtures of Ph-positive cell lysate and antibodies against those two proteins. Western blots show that GRB-2 does not interact with the oncogenic ABL kinase separately in vivo. In vitro, BCR-ABL only bound to full length GRB-2 and not to other proteins in a GST fusion protein � GRB-2 complex. Similarly, an in vivo experiment showed that c-ABL or c-BCR itself does not complex with GRB-2. An analysis of which regions of BCR-ABL participate in GRB-2 binding revealed two things: that the GRB-2 SH3 domain apparently plays no part, and that separately expressed c-ABL and c-BCR do not bind to GRB-2 SH2 domains as ABL-BCR does. However, transphosphorylation of c-BCR with c-ABL tyrosine kinase in vitro resulted in its binding to the GRB-2 SH2 domain.
Its relevance to BCR-ABL induced oncogenesis was shown by demonstrating that a point mutation of tyrosine to phenylalanine in the BCR first exon prevents binding of the GRB-2 protein to BCR, thus decreasing its transforming capacity of fibroblast cells. A removal of two SH2-binding sites that are important for BCR-ABL-mediated fibroblast transformation in the protein resulted in a more pronounced deficiency in transformation than the protein with only the point mutation. Finally, it was shown that BCR-ABL transcriptional activation was mediated by Ras, and that BCR-ABL wild type induced 8-fold jumps in transcriptional rates, which lowered upon transfection fo Ras(17N).
The implications of this study are that the BCR-ABL and GRB-2 interactions could be an important part in uncovering better therapeutic methods of disrupting oncogenesis through protein action. They showed that the complex existed and that the interaction leading to Ras activation only occurred when both were fully present; binding was limited only by a point mutation of a required tyrosine to phenylalanine. Later studies can look further into other methods of lowering Ras activation and/or transformation of fibroblasts and other such cells through BCR-ABL GRB-2 complexes, for use in cancer prevention/intervention research.
Overall, the authors presented a well-rounded paper. Their conclusions were well defined and supported by graphs that appropriately displayed controls and variables. They adequately described methods and were clear on explanations of all their figures. However, I felt that as they described their results, they wandered frequently, expressing information not necessarily needed to prove the point of the paragraph.
1. Chronic granulocytic leukemia is connected to the Ph1 chromosome (the Philadelphia chromosome) which was thought to have been chromosome 21 in 1961.
2. Scientists later concluded that this chromosome is actually a translocation of the abl region on chromosome 9 and the BCR region on chromosome 22.
3. A 1986 paper found that the BCR/ABL complex was linked to CML because it produced P210 (molecular weight: 210), a protein previously linked as specific to CML. BCR-ABL activates signaling pathways transducing oncogenic signals from the cytoplasm to the nucleus.
4. This paper found that the BCR-ABL protein binds to the SH2 domain of the GRB-2 Adaptor protein and by doing so, disrupts the Ras signaling pathway and causes heightened transformation, uncontrolled mitosis and cancer.
5. GRB-2 has no enzymatic purpose; it is merely a binding protein which can be used in the scaffolding of the cell or to bring free floating proteins from inside the cell to the scaffolding. The SH2 and SH3 domains on GRB-2 bind specific sequences on receptor tyrosine kinases and the Sos exchange factor. SH2 specifically binds tyrosine-phosphorylated sequences on the receptors of proteins like ABL.
6. The scientists looked for tyrosines within the Philadelphia Chromosome and saw if the sequences matched any part of the SH2 domain. Of 11 tyrosines, 1 matched.
7. For their beginning experiment, they wanted to make sure that BCR-ABL and GRB-2 indeed formed a complex � so they tried coimmunoprecipitation, in vivo. Their samples were taken from human fibroblast cells as well as samples of cells from people with CML and ALL (acute lymphatic leukemia). These cells were lysed and then mixed with the antibodies.
8. They used antibodies against ABL and GRB-2, and as a control, used pre-immune serum, or the animal serum before proteins were injected in them to generate and collect the specific antibodies.
9. To label, the scientists used autoradiography. Since ABL can phosphorylate itself, everything with ABL in it gets phosphorylated and can thus be seen. GRB-2 is pulled out with antiGRB-2.
10. Two different proteins, P210 and P185, exist because they are created by BCR-ABL.
11. Why didn�t they use an antiBCR antibody? Maybe because one didn�t exist at the time, or because they didn�t need for another antibody to pull down bcr-abl � one was enough, or maybe because phosphate hooks to abl and not to bcr.
12. This does not conclude that the protein does not hook to bcr; that must be explored.
13. Criticisms include: if the last lane wasn�t there (with no GRB-2 pulled down), then the whole experiment would be fishy, as there were no �mistakes� otherwise, and ABL might merely be sticky. To test that, combine pure ABL with antiGRB-2 to see if GRB-2 is pulled down at all.
14. BCR might mediate the interaction � the last lane has no BCR and got no GRB-2, meaning that BCR might need to be there for the interaction to occur.
15. Another part of this experiment found that GRB-2 can be precipitated by antiABL in cells where BCR-ABL is expressed. This doesn�t decisively show that BCR-ABL was directly associating with GRB-2; there might have been intermediaries.
16. In vitro makes it easy to isolate certain proteins and change the environment, but in vivo makes sure it can happen in nature and not just in the test tube.
17. The next experiment was for a similar goal, except in vitro. It involved the use of methionine (an S-containing R-group amino acid) to label P185 proteins, which were in Sf9 insect cells. They are routinely used to copy genes because it is easily infected by baculoviruses.
18. GST is a glutathione S-transferas fusion protein that serves as a connector or conjugator between GRB-2 and glutathione covered beads. GST was inserted into a vector containing the GRB-2 protein to be fused together.
19. A negative control was done with just GST to show that GST by itself was not going to pull down BCR-ABL, which turned out to be true. A positive control was done with antiABL just to show ABL was there. The P185 BCR-ABl protein, if the experiment was successful, would be pulled down when binded to the GST-GRB-2 complex. They also prepared runs with just the SH2 domain and the C and N-terminal ends of the SH3 domain. Each lane pulled down the p185, meaning that the GST-GRB-2 complex did bind with ABL-BCR.
20. However, the �purified proteins� from the insect cells might not have been so pure and could have pulled extra proteins along, making it possible for the interaction to be indirect. To test for this, an all-purpose dye should be used to color all the proteins and a blot done. If more than 2 proteins show up, the purification was faulty.
21. The scientists then tried to explore what the interaction between ABL and BCR was. They ran the GST-GRB-2 complexes with just cABL and just cBCR and then both of them together. They found that cABL pulled down the full length GRB-2 with the SH3 domain, not SH2, and the cBCR also pulled down the full length GRB-2. However, when both cABL and cBCR were present, the SH2 domain was suddenly involved.
22. Since cABL was a tyrosine kinase, they thought it was possible that cABL was phosphorylating cBCR on a tyrosine, which the SH2 domains were recognizing.
23. They added evidence to this contention by showing that treatment of the cBCR and cABL complex with phosphatase eliminated its ability to associate with the SH2 domain but not SH3. This shows that the SH2 interaction with a phosphorylated tyrosine was not happening because the tyrosine was not phosphorylated any longer. Also, Western blots using a-phosphotyrosine antibodies show cBCR was phosphorylated after being with cABL.
24. They then used immunoprecipitation to show, in vivo, that GRB-2 really does only interact with BCR-ABL and not just one of them. P185 cells and ABL cells were lysed and combined with normal rabbit serum (NRS), antiGRB-2 and antiABL. This revealed that BCR-ABL could be pulled down with antiGRB-2 but not ABL by itself. To verify, they ran BCR with the same antibodies and only received a pull from anti-BCR.
25. So they could conclude that BCR-ABL was directly involved, but why? Because ABL phosphorylates BCR when they�re together.
26. The tyrosine that matched the SH2 domain was known as Y177. The next experiment was to determine whether this tyrosine, that was most likely the phosphorylation site, was necessary or not for binding.
27. Cotransfections similar to Experiment 2 using Y177 and a phenylalanine mutant, Y177F, concluded that only P185 (BCR-ABL) was able to bind GST-SH2 domain and not the mutant. However, it could be that maybe other amino acids as the mutant would have done something different.
28. The same thing was done using binding studies with antibodies. Treatment of wildtype and mutant with antiGRB-2 only revealed, again, that only wild type could pull down GRB-2. Now, how does this mechanism affect CML?
29. During transformation, the cell becomes cancerous. When culturing human cells on agar, normal ones do not divide, but transformed, cancerous cells continually divide, anywhere. This experiment involved the culture of cells with P185 wildtype, mutant, and without the BCR-ABL gene. The mutant got a decreased amount of transformation and the one without BCR-ABL were not transformed, proof of a definite lowered binding capacity when a part of the BCR-ABL gene is mutated or gone.
30. Now, the crux of the paper was discussing how those proteins were involved with the Ras pathway. The scientists ending up charting the transcriptional activity of the genes using CAT. The mutant and the one missing the gene altogether charted low activity, but the wild type registered extremely high transcriptional activity. However, when it was combined with Ras, the transcription rate was dramatically dropped.
31. We can conclude that Ras is a dominant �inhibitory� mutant that is a mediator or has varied response.
32. The paper basically said that Tyrosine 177 on BCR was able to be phosphorylated with ABL in an ABL-BCR complex that can bind with GRB-2, which leads to heightened Ras activation and cancer.
Sanchez-Garcia and Grutz. �Tumorigenic Activity of the BCR-ABL Oncogenes is Mediated by BCL2.� Proceedings of the National Academy of Sciences of the United States of America (PNAS). 92 (12): 5287-91, 1995.
1. Since growth factors are required for cells to grow, an unchecked growth, as in cancer, means that some other signal is substituting for the normal growth factor � and which must be derived from some gene. In this case, it is BCR-ABL.
2. P210 and P190 differ only in the number of BCR-encoded amino acid residues. P190 is more associated with acute leukemias and has heightened tyrosine kinase activity.
3. They use transformed murine hematopoietic cells because they can be made growth factor independent and tumorigenic by action of the BCR-ABL oncogene, in this case, is Ba/F3, which is specific to Bcl-2.
4. MYC protein is essential for BCR-ABL transformation; its transcription is activated by a tyrosine kinase oncogene as well as by growth factor stimulation. Their findings implicate the activation of Bcl-2 as the means whereby BCR-ABL oncogenes synergize with MYC during transformation.
5. Burkitt�s Lymphoma: a solid tumor of B-lymphocytes from t(14:18) which has a dystregulation of apoptosis. Studies of it identified Bcl-2 as a gene intimately involved in apoptosis regulation.
6. Apoptosis: Greek, means �falling away from.� Named after 1972 study by Kerr et al. who noticed things falling off cells. It is programmed cell death and is a tightly regulated progress. Symptoms are cell shrinkage, chromatin condensation, cytoplasmic membrane blebbing, phagocytosis of apoptotic bodies. It is seen throughout the body and keeps homeostasis. It is seen in development, inflammation, immune system and aging. Triggers include: internal cell signals, death activator signaling on cell surface (TNF-alpha, Fast, etc.) or reactive oxygen species (oxidative enzymes, for ATP).
7. Bcl-2: is a protein present on the outer mitochondrial membranes in healthy cells. Binds to Apaf-1, which it releases when cell is damaged. This allows cytochrome c to leak into the cell. These two proteins bind to caspase 9 which cleaves proteins and activates proteolytic cascades which eventually lead to phagocytosis of the blebs of cells.
8. The fact that increased levels of Bcl-2 keep the cell from undergoing apoptosis has been shown to result in several cancers, and is a target for therapeutic research.
9. Why choose Bcl-2? CML is related to FL which Bcl-2 is involved in. Possible parallel?
10. IL-3: growth factor that stimulates Ba/F4, a murine hematopoietic cell. Without it, a cancerous signal would take its place.
11. Antisense DNA: complementary strand which attaches to the mRNA and prevents copying. When binded, the strands are very stable.
12. Part A of Figure I is a Northern blot of the Ba/F3 cells. They are transfected with plasmids expressing P210, P190 and a negative control. The ABL probe is used as part of the fusion protein to label where ABL shows up. The actin probe just shows something was loaded in the trays.
13. Fetal calf serum is the source for IL-3.
14. Part B of Figure 1 shows cells deprived of IL-3. The positive control is normal Ba/F3 cells which quickly die out because of normal apoptosis. Ba/F3 cells stably transformed with P190 and P210 BCR-ABL proteins exist at high viability for indefinite time. This shows the transformation allowed the cells to become IL-3 independent.
15. Part C of Figure 1 shows DNA nucleosome laddering obtained from the contents of the cell after it dies. Southern blots were done with ethidium bromide, which selectively stains DNA.
16. Part D of Figure 1 shows the same cells as Part B but in 1% FCS as compared to 10% for the other parts. This means there is a lower level of IL-3 available than normally. Normal growth continues for a little while but then drops. High viability continues for the BCR-ABL transformed cells, which are IL-3 independent.
17. Viability was determined by adding trypan blue. Live cells pump this dye out as they have mechanisms to do so but dead cells would retain the blue.
18. Bcl-2 is not present in normal Ba/F3 cells. However, it is present in P210, P190, the FL cells and the Bcl-2 expressing cells. Northern blots were done in Figure 2a using the Bcl-2 probe. An ABL probe was used in Figure 2b � it is only present in P210 and P190. Again, actin probes show all lanes were loaded. Western blots were done to show there were protein products too.
19. Hydrogen peroxidase at 0.25 and 0.5 mM did oxidative damage and thus killed the Ba/F3 cells. Transfected cells survived � which was a common Bcl-2 effect. At 1 mM, however, all died.
20. Figure 4a shows that the Bcl-2 probe matched in the antisense Bcl-2 (the control) and in the P190 and P210 fusions with Bcl-2. Figure 4b shows antisense Bcl-2 probe being used. It labels Bcl-2 in the transformed cells but not the transformed antisense Bcl-2 � they stop making proteins and thus there are none to be labeled. Potentially Bcl-2 RNA should be there, but if protein is hoarded, the RNA need not be there.
21. If one knows the half life of a certain protein, one can tell when the protein will disappear without having to do a Western blot to make sure.
22. Table 1 shows us that normal cells and BCR-ABL with antisense Bcl-2 both have no tumors. However, BCR-ABL transfected cells have tumors.
23. Nude mice are used because they have no thymus, meaning no T-cell immune system.
24. How do you know Bcl-2 only affects Bcl-2?
25. Figure 5 shows that Il-3 deprivation also kills antisense Bcl-2 with P190 and P210 combos. Full viability maintained for the BCR-ABL transformed Bcl-2 proteins.
26. IL-3 is not specific. It was used because they were using an IL-3 dependent cell line.
Skorski et al. �c-RAF-1 Serine/Threonine Kinase Is Required in BCR/ABL-dependent and Normal Hematopoiesis.� Cancer Research. 55:2275-78, 1995.
1. It was already established that BCR-ABL maintains active Ras, which is important for the proliferation of Ph1 positive and normal hematopoietic (blood growth) cells. C-RAF-1, or P74, is the direct Ras downstream effector, which is encoded by the c-raf-1 proto-oncogene. The c-RAF-1 serine/threonine kinase can be activated by tyrosine kinases.
2. The researchers wanted to see if c-RAF-1 associates with or is regulated by the BCR/ABL oncogenic tyrosine kinase. They found that c-RAF-1 activity was regulated by, but not associated with, BCR-ABL.
3. Their next step was to see if this interaction was significant for BCR-ABL-dependent growth/transformation of CML cells and NBMC (normal bone marrow cells). After downregulation and inhibition of c-RAF-1, they found that BCR-ABL�s functions were indeed inhibited.
These papers delve further into the mechanisms by which the Philadelphia chromosome is able to lead to oncogenesis. Skorski�s paper maintains that the Ras signaling mechanism leading to oncogenesis is activated by BCR-ABL products at the downstream effector site (c-RAF-1). Their analysis also found that a downregulation and inhibition of that same site inhibited the function of BCR-ABL itself. Sanchez-Garcia et al. found that the tumorigenic action of the BCR-ABL oncogene � which prevents apoptosis in cells � requires the mediation of the Bcl-2 expression pathway. These oncoproteins cooperate with the MYC protein, which induces cell proliferation, leading to the implication that cancer is caused by a combination of cell proliferation and non-death.
It was already established that BCR-ABL maintains active Ras, which is important for the proliferation of Ph1 positive and normal hematopoietic (blood growth) cells. c-RAF-1, or P74, is the direct Ras downstream effector, which is encoded by the c-raf-1 proto-oncogene. The c-RAF-1 serine/threonine kinase can be activated by tyrosine kinases. Skorski et al. wanted to see if c-RAF-1 associates with or is regulated by the BCR/ABL oncogenic tyrosine kinase. After immunoprecipitation with growth factor starved and BCR-ABL boosted cells, it was found that significant activity only happened in the BCR-ABL cells. In a further immunoprep study, it was found that both c-RAF-1 and BCR-ABL were present in the lysate of leukemic cells but c-RAF-1 had no ABL interaction. Western blots of cell lines treated with antisense oligodeoxynucleotides or transfected with c-RAF-1 antisense DNA showed that c-RAF-1 protein was down-regulated compared to controls. Inhibition studies demonstrated that c-RAF-1 is needed for proliferation of Ph1 and normal cells � otherwise the level of function of BCR-ABL was quite low.
Since growth factors are required for cells to grow, an unchecked growth, as in cancer, means that some other signal is substituting for the normal growth factor which, in this case, is BCR-ABL. Transfection of P190 and P210 BCR-ABL protein expression plasmids into the Ba/F3 cell line rendered them tumorigenic without the presence of normal growth factors in both normal and serum-deprived culture conditions, meaning that BCR-ABL is mostly if not wholly responsible for this inhibition of apoptosis. An analysis of mRNA was done to define the level of interaction with Bcl2; the probe detected major similarities in the transcript in both BCR-ABL cells and cancerous cells when Bcl-2 was also found to be present. Exposure to lethal amounts of hydrogen peroxidase showed no damage to BCR-ABL cells, a typical Bcl-2 effect � meaning that there was proof of a further interaction with Bcl-2. Sanchez-Garcia et al. suppressed the Bcl-2 pathway to study the effects of BCR-ABL on those cells. They found that the BCR-ABL oncogenes were unable to transform the cell lines into tumorigenic cells after Bcl-2 suppression.
These two studies clarified further the mechanisms by which BCR-ABL leads to Ph1 leukemogenesis. The proteins studied interact with BCR-ABL but in different pathways. I felt the Skorski paper was short and to the point. The Sanchez-Garcia paper was very informative but its ending felt rushed and misplaced � they mention MYC and its function within the BCR-ABL and Bcl-2 system, but it is purely a discussion matter � no laboratory techniques were done to prove the role of MYC.
Kantarjian, H. et al. �Imatinib mesylate therapy for Ph+ CML in blast phase.� Blood. 99: 3547-3553, 2002.
Philadelphia chromosome-positive chronic myelogenous leukemia (Ph+ CML) occurs in three stages: chronic phase, accelerated phase and blastic phase. Imatinib mesylate, a signal inhibitor specific to Bcr-Abl tyrosine kinases, has been shown to be effective for treatment of CML within the chronic and accelerated phases; complete hematologic responses (CHR) occurred for 98% and 40% of cases for chronic and accelerated phases respectively. This paper more specifically studies the effects of imanitib mesylate on CML patients in the blastic phase. During this phase, the percentage of blast cells � the very immature cells normally present in the bone marrow in low proportions � increases to over 30%. Normally, the blasts are from myeloid origin, but in 25% of cases, they will have a lymphoid origin. Past analyses using small numbers of patients have shown a 28% CHR for blast phase CML; this study uses a larger amount of patients from a single institution � namely, 75 patients from a Texas medical center. All the subjects had confirmed Ph+ blast-phase CML, were over 18 years of age, had normal renal and hepatic function and had given informed consent. They were each given daily of doses between 300 and 1000 mg; the majority received 600 mg. Responses were measured by granulocyte, white blood cell, platelet and blast cell count, as well as by presence of Ph+ metaphases. CHR required certain counts and 5% maximum blasts in the bone marrow. Cytogenetic responses were measured by the percentage of Ph+ metaphases. Treatment failure counted if the patient died or if the disease proved resistant. The historical controls were previously well-documented results of treatment with cytarabine-containing medicines for comparison.
Among the patients with a non-lymphoid blast-phase CML, 23% achieved CHR. Most of the rest had either improvement or a partial response, and some returned back to the chronic phase. A significant number of patients achieved a complete recovery (no Ph+) within the marrow. Overall response was 55% and 30% for both non-lymphoid and lymphoid blast-phase CML respectively. A follow-up after 11 months showed that the median survival was 6.5 to 7 months with a 1 year survival rate of 28%. Side effects included nausea and vomiting, liver dysfunction, rashes, cramps, fluid retention and edema. In some patients, febrile episodes and fevers occurred. Hematologic reactions included suppression of granulocyte and platelet counts. A total of 57 patients died, and 8 continued to take imatinib mesylate before receiving stem cell transplantation (SCT). Three of these had no evidence of disease after 17 months. Of the 57, 10 died during therapy and 47 died after it was stopped � but there were no imatinib mesylate-associated deaths.
Age, presence of splenomegaly and blast cell proportions were similar in this study and the previously done cytarabine studies. However, objective responses were 55% for the imatinib mesylate treatment compared to 29% for cytarabine. 4-week mortality rates were 4% for imatinib mesylate and significantly higher, 15%, for the cytarabine group. A further analysis for predictors found that peripheral blasts predicted for response, and only platelet count associated with survival. Their conclusion was that imatinib mesylate was the most active agent yet able to be used against blast-phase CML even though prognosis was relatively unsatisfactory. Their suggestion is to study imatinib mesylate in combination with other treatment regimens including other cytarabine compounds, allogeneic SCT and farnesyl transferase inhibitors. Recommendations for improvement to this study include a larger analysis of lymphoid blast morphology, though that drawback was noted in the paper. Also, the authors use cytarabine as a comparison that would show that imatinib mesylate, being better, is the best agent discovered so far � it is true that the standard of care is to use cytarabine in combination with chemotherapy, but their �comparison� is weak. They do not cite or show the historical control data for cytarabine compounds.
Dr. Frederic Barr, Assistant Professor of Pathology and Lab Medicine. �Chromosomal Translocations in Rhabdomyosarcomas: From Cloning Breakpoints to Clinical Practices.� [email protected], 505 Stellar-Chance 898-0884
1. During the late 80�s, scientists thought that translocation was the exclusive domain of hematopoietic cells. The problem was with technique. It was refined and by 1990, translocations were found in epithelial tissues � but the connective parts, not the common epithelial tissue.
2. These cancers were known as sarcomas, or cancers of the connective tissue, which includes the bones, the cartilage, the muscles (rhabdomyosarcoma) and other such parts. At least 12 sarcomas are caused by specific translocations and fusion products.
3. Translocations are consistent and specific to one type of tumor and cancer. This allows us the opportunity to explore the biology of the tumor, the diagnosis of it through markers and therapy, by offering a target which we can focus on in the hunt for cures.
4. Case 1: 13 year old female with a two month history of abdominal and back pain with weight loss. She had cervical/mediastinal lymphadenopathy (enlarging of the lymph nodes in the neck and chest) and ascites (fluid collection). Lymph node and bone marrow biopsies were done, which showed malignant neoplasms. She was treated with chemotherapy and showed a complete clinical response. However, physical tests missed a few cells, as she regained ascites and hepalomegaly after 6 months. After 12 months, a laparotomy showed extensive tumors, and at 15 months, she was dead.
5. Histopathology of the cells showed a round nucleus with low cytoplasm and a low amount of cell architecture or structure. This is known as the small round cell tumors of childhood. These show a uniform morphology with no cell differentiation.
6. Differential diagnoses include Ewing�s and neuroblastoma, both with neural lineages; rhabdomyosarcomas, desmophastic small round cell tumor, synovial sarcoma, and lymphoma.
7. The sarcoma will start in the primary focus where it will then spread into the bone marrow.
8. Immunohistochemistry with muscle specific actin labels can narrow it down to show it�s a sarcoma related to muscle cells.
9. Cytogenetics show a karyotype with an elongation of chr. 2 and a shortening of 13 � the chunk from 13 was on the bottom leg of 2.
10. Alveolar rhabdosarcoma (ARMS): It is a soft tissue tumor with striated muscle differentiation. It frequently occurs in extremities and trunk and in the pediatric population. It comes with an unfavorable prognosis.
11. Gene mapping finds the localized place where the gene lies. Further location is done by murine splotch phenotype.
12. The PAX3 gene looked like it was split by the chromosomal translocation. They check for rearrangement of PAX3 in ARMS tumors. They took PAX3 hybridized to RNA from ARMS and other cell lines. Northern blots show cancerous cells express PAX3 much more than normal cells do. Novel sequences in PAX3 clone hook to a part of chr. 13.
13. Northern blots also show FKHR (forkhead rhabdomyosarcoma) in cancerous cells.
14. t(2:13) generates PAX3�FKHR in ARMS Variations in genes are in the introns.
15. Now, would you look in fusion DNA next? There are too many variable breakpoints and thus assays would take too long. How about fusion proteins? Specific antibodies for the fusion junction are impossible to find. Then fusion RNA? Perfect, PCR can be done.
16. RT-PCR detection: chimeric transcript with oligo primer ? reverse transcription ? cDNA with gene specific primer ? PCR ? PCR product with hybridization probe.
17. RT-PCR should be able to pick up the wild-type or it�s useless. On the analysis will be a control line, a test line and then a test hybridization.
18. Case 2: 11 month old male with 4 month history of leg swelling, had a six centimeter mass in anterior medial compartment of right thigh. He had mediastinal lymphadenopathy (in chest cavity). An excisional and bone marrow biopsy showed malignant neoplasm. He had chemotherapy and a complete clinical response. He lived.
19. Cytogenetics found a t(1:13). 13 was cut in the same place, and this piece went to the top of 1.
20. PAX family genes are on chr. 1 � PAX7. It combined with FKHR again, producing a fusion of PAX7-FKHR.
21. In large trials, PAX3-FKHR makes up 55% of cases, PAX7-FKHR makes up 23.5% and fusion negative is 21.5%. All non-ARMS are fusion negative (including ERMS, Batyoid, spindle cell and undifferentiated sarcomas).
22. Are these fusion negatives right or a mistake? Low expressor (only in a few cells) take up 15%; variant fusion (not FKHR but AFX, etc.) make up 20%; cryptic fusion (there at genomic level but not RNA) make up 15%; and true fusion negatives take up half.
23. PAX3-FKHR v. PAX7-FKHR: with age, the first is for older children and younger adults and the latter is for younger kids; with local invasion of tumor, the first is more invasive and the latter is less; and as for frequency of metastasis, there is no difference.
24. PAX7-FKHR patients tend to live longer but not statistically significant. With overall survival and local regional spread, there is no difference. With overall survival and metastatic disease, PAX3-FKHR die within five years, while PAX7-FKHR live much longer and don�t get worse.
25. In distal nodes, the p-value for each is 1 meaning an equal frequency. In bone marrow, however, PAX3-FKHR metastasizes (p-value = 0.044) and PAX7-FKHR does NOT metastasize. Maybe because it doesn�t spread to the bone marrow, people survive longer.
26. In combination of PAX and FKHR, only the FKHR transcription regulation domain is cut, but nothing else is (the homeobox, promoter, etc.).
27. Transcriptional activity of either PAX is low, but PAX3-FKHR and PAX7-FKHR is high.
28. PAX3-FKHR and PAX7-FKHR expressed at higher levels than either PAX in Ribonuclease Proection Assay.
29. Fluorescent in situ hybridization should show, in cancerous cells, a red (PAX3) and green (FKHR) dot combining into a yellow one as it fuses � but with PAX7, there are many more. With overexpression, PAX3-FKHR is copy number independent for transcription and PAX7-FKHR is gene amplification.
30. FKHR is downstream of an important signal pathway. It activates apoptosis. AKT is the protein above it in the chain; it adds a P to FKHR and kicks it out into the cytoplasm. In ARMS, this pathway is still active, so what is taking place of the FKHR that is supposed to be fusing? Or does it still play a role in the pathway? But, PAX3-FKHR and PAX7-FKHR is resistant to control by AKT.
31. 100 cases of ARMS occur nationwide every year.
32. PAX3 and PAX7 consequences: potent transcription, high level expression, constitutive nuclear localization. Hypothesis: maxime transcription of PAX3/PAX7 target gene.
33. PAX3 is expressed in early muscle development (myogenic progenitors). It appears in the presomitic mesoderm and early somites and migrates to the limb buds. In a homozygous splotch mouse without PAX3, there is a malformation in axial musculature and in limb musculature, it fails to develop, thereis increased apoptosis, and there is no migration of cells. SO, PAX# is for proliferation, apoptosis, differentiation and motility.
34. PAX3-FKHR targets MET, which mediates growth and motility signaling, and BCL-XL, a member of the Bcl-2 family of proteins which conducts anti-apoptotic activity.
35. STI571 made possible by these studies.
36. PAX3-FKHR and PAX7-FKHR are very potent oncogenes and can�t go into animal models because they would kill them too fast. A way around this is to put it into the cells in an inactive state and turn it on later.
37. PAX7 made in satellite cells in muscles. It is harder to make because PAX7 must be realigned and flipped, unlike PAX3.
Drug Design
1. Rational drug design is fitting a drug to a certain site on a molecule. It is hard because protein structure can continually change, but computer protein mapping is improving.
2. Usually, they just throw all the compounds in the Merck Index or Sigma Catalog to see if any compound will bind with it and then from there, why.
3. Dosage is done through animal models first, with trial and error methods.
4. When testing the drug, Phase I is for figuring out maximum dose; Phase II is for figuring the right dose; and Phase III is for testing with lots of people.
5. Mice are used because they gestate in 3 weeks, are cheaper, are guarded less strictly by ethical guidelines, and are much easier to handle than primates.
6. Drugs working in mice and at certain doses sometimes don�t translate to humans.
7. Human trials are expensive. Drug companies pay for the person using the drug, pay for the hospital bills, pay for the insurance, and many other supplementary costs.