Cancer Initiation: 2013

Tuesday, December 31, 2013

Mesothelin emerges as a new biomarker for early cancer detection

 This note is intended to serve as a scientific background to a remarkable story, the story as told by The Scientist  [1] of Jack Andraka,  "The Cancer-Test Kid". While TS covers one end of this remarkable story, there are actually quite a few ways to look at it.  The first is, how is it that a relative youngster  who just earned his way into the education system beat an entire industry of "translational medicine" research to complete his remarkable find?  Were there things that "grown-ups" could have done to at least facilitate what seems to be a single handed story?  Obviously, the answer to these questions is "complicated". In addition to the usual convoluted emergence of understanding from medical science experiments, the diagnostic story of the mesothelin bio-marker overlaps "causation" and therapeutics in many remarkable and non-traditional ways, at least in terms of our traditional approach to medical research which strongly parallels the foundations of the "central dogma".

Research History

The story  of mesothelin appears to begin in the laboratories of the NIH's national cancer institute in Bethesda Maryland.[2] The initial discovery that led to the isolation and characterization of mesothelin occured in 1992, with an experiment that consisted of inoculating mice with ovarian cancer cells (OVCAR-3), and isolating an antibody that had unique interaction with the ovarian cancer cell surface. This antibody was designated K1. The antibody was found to also be reactive with human mesothelium, the external lining of many organs including the peritoneum. Hence, the future name of the K1 associated  biomarker was an indication of where it would normally be found in human tissues. The significance of this antibody was recognized immediately as this quote from the abstact shows:

The K1 antibody may be useful as a targeting agent for therapy and in the diagnosis of ovarian carcinoma, as well as some other human cancers. ( [2] 1992 )
Even with the significance immediately recognized, it would not be until 1996 when the gene identified by the K1 antibody would be isolated and cloned by that same team[3]. The gene and protein product were the officially named mesothelin, again, for its normal location in mesothelium derived tissues.. Its function remained unknown.
  The next step was to produce a "mono-clonal" antibody which involved immunizing mice with mesothelin produced from their previously cloned gene.  Apparently, the antibody K1 was not what would be called a "mono-clonal" antibody, that is one which arose from a single gene sequence to target a single epitope, or "feature" on a biomarker. Such mono-clonal antibodies are the result of the construction of a hybridoma or immortalized cell line which produces the antibody. The hybridoma was sucessfully produced by the Chang lab in 1997 [4], after what ( from the abstract ) seems to have been technical difficulties.
Several attempts to produce anti-mesothelin hybridomas from spleen cells of mice immunized with recombinant mesothelin were unsuccessful. This report describes the isolation of anti-mesothelin scFvs from a phage display library made from the mRNA of the same spleens.[4]

The story then continues in 1999 in the laboratories of the Pacific Northwest Laboratory and the Fred Hutchinson Cancer center with a publication by Scholler et. al. [5] . What Scholler did was gather cells from ovarian cancer patients and immunize mice with them. The immune systems of the mice then generated antibodies that were specific to biomarkers that would be found on ovarian cancer cells. Among these was a mouse antibody ( mAb ) designated OV569. This antibody ( OV569 ) was found to bind selectively to ovarian cancer cells, as well as certain other cancer cells, while showing no interaction with normal tissue. In addition, a binding target of OV569 was found to be circulating in the blood serum of patients with ovarian cancer, but not in those that are free from cancer. The value of the "mesothelin" type target as a marker for cancer was immediately recognized by the group that published the initial findings of a circulating form of mesothelin, and was noted as follows:
 Thus, there is a need for improvement, either in the form of a more specific and/or sensitive assay or an assay that uses a different marker and can be used to complement CA125 toward the goal to improve patient survival by improving diagnosis. [5]
The current biomarker for ovarian cancer in CA-125, or cancer antigen 125, otherwise known as MUC16 ,  a mucin or carbohydrate rich protein found highly expressed on surface cells of the female reproductive tract. Of note here, the laboratory at Johns Hopkins and Jack Andraka himself were interested in biomarkers for pancreatic cancer, not ovarian cancer. It is not clear that the Scholler team immediately recognized the importance of a non-mucoid biomarker as a means of early diagnosis of other types of cancer.
     The importance of finding a circulating and specific biomarker cannot be overstated. For example, in the current top video under the search "cancer biomarker", Raj Krishnan relates that a current Nobel Prize in medicine holder regards this as a top priority in cancer research, and one itself worthy of a Nobel Prize. Once again, the magnitude of the Jack Andraka story cannot be overstated.

    But, it seems that MUC16, the current and traditional biomarker for ovarian cancer,  remains part of the mesothelin story. MUC16 is not expressed in normal pancreatic ducts, but becomes upregulated and is expressed at the advancing edges of pancreatic cancer[6].  It seems that MUC16 and mesothelin actually bind to each other, and after doing so up regulat Matrix Metaloprotein 7 ( MMP-7)  [6]
     In less aggressive cancer, extra cellular matrix (ECM)  provides a barrier to invasion by the cancer. Extra cellular matrix is a web of protein and polysaccarides that exists outside the cell wall of tissues. It is called "matrix" because it consists of a web like structure, made largely of collagen for structural support and several types of glyco-proteins for signaling to cells.  Cells that "remodel" ECM do so by cleaving it with the aid of "Matrix Metalo-proteins" or MMPs. These MMPs are enzymes that use a particular metallic ion as a "knife" to enigmatically cut through ECM. One that is of relevance to pancreatic cancer is MMP-7. It has been shown, that in addition to cutting through extra-cellular matrix proteins, MMP-7 disrupts tight junctions, or cell to cell bindings in pancreatic epithelium. [7].    As such, cells of the invasive growth not only cut through ECM tissue, but cells dissociate and become free to metastasize through the blood stream.
   Consequently, when MMP-7 expression is blocked, the tissue invasion by  the cancer is reduced [6].

  As mentioned, mesothelin was found to be circulating in the blood stream of patients as early as 1999[5].  As such, mesothelin should be considered among the "gold standard" of potential biomarkers, because, it is very specific for a few types of cancer, it circulates in the blood stream, and is itself a promoter of the invasive qualities of the particular cancer. As such, it should have ( in my opinion ) been advanced more quickly than the article in The Scientist implies.

 Cell surface biomarkers are the foundation of targeted therapy. The first was the combination of human epidermal growth factor receptor 2, (HER-2) and herceptin a humanized monoclonal antibody to HER-2 used for treatment of breast cancer and other HER-2 positive cancer.[10] Based on the model of HER-2 and herceptin, mesothelin provides a new target for immune system targeting of cancer by manufacture of mesothelin specific monoclonal antibodies, or else the use of antibodies to deliver cyto-toxic loads targeted at specific cancer cells.[11][12]

 The nature of cancer research is that it is a competitive endeavor, were proposals are reviewed and compared, and rated at the level of each grant, or round of research. The goal is to provide the maximum rational basis of fairness between competing medical centers and universities. Unfortunately, when looking back over published track record of a particular line of research, such as that of mesothelin, the system gives the appearance of having a hap-hazard quality. The NIH must find a way to provide a fair system of grant awards, while maintaining a focused record of progress on important projects. In other words, there should not be a large number  of Jack Andracka type stories out there, and the National Cancer Institute of the NIH is the organization that should make sure this is not a regular occurrence.


[1] Dan Cossins, The Cancer-Test Kid. The Scientist, Dec. ( 2013) [Full Text]

[2]  Chang K, Pastan I, Willingham MC. Isolation and characterization of a monoclonal antibody, K1, reactive with ovarian cancers and normal mesothelium Int J Cancer. 1992 Feb 1;50(3):373-81. [Abstract]

[3]Chang K, Pastan I Molecular cloning of mesothelin, a differentiation antigen present on mesothelium, mesotheliomas, and ovarian cancers.. Proc Natl Acad Sci U S A. 1996 Jan 9;93(1):136-40.  [PubMed Central]

[4] Chowdhury PS, Chang K, Pastan I Isolation of anti-mesothelin antibodies from a phage display library.  Mol Immunol. 1997 Jan;34(1):9-20. [Abstract]

[5] Nathalie Scholler*, Ning Fu*†, Yi Yang*, Zhengmao Ye*,Gary E. Goodman*‡, Karl Erik Hellström*, and Ingegerd Hellström*§ Soluble member(s) of the mesothelin/megakaryocyte potentiating factor family are detectable in sera from patients with ovarian carcinoma
PNAS September 28, 1999 vol. 96 no. 20 11531-11536 [Full Text]

[6] Shih-Hsun Chen, Wei-Chien Hung, Pu Wang, Colin Paul & Konstantinos Konstantopoulos
Mesothelin Binding to CA125/MUC16 Promotes Pancreatic Cancer Cell Motility and Invasion via MMP-7 Activation Scientific Reports 3,Article number: 1870 doi:10.1038/srep01870 [Full Text]

[7]  Tan X, Egami H, Abe M, Nozawa F, Hirota M, Ogawa M.  Involvement of MMP-7 in invasion of pancreatic cancer cells through activation of the EGFR mediated MEK-ERK signal transduction pathway.   J Clin Pathol. 2005 Dec;58(12):1242-8. [PubMed Central Full Text]

[8] Argani P, Iacobuzio-Donahue C, Ryu B, Rosty C, Goggins M, Wilentz RE, Murugesan SR, Leach SD, Jaffee E, Yeo CJ, Cameron JL, Kern SE, Hruban RH. Mesothelin is overexpressed in the vast majority of ductal adenocarcinomas of the pancreas: identification of a new pancreatic cancer marker by serial analysis of gene expression (SAGE). Clin Cancer Res. 2001 Dec;7(12):3862-8. [PubMed Abstract and Full Text]

[9] Wang Y, Wang L, Li D, Wang HB, Chen QF. Mesothelin promotes invasion and metastasis in breast cancer cells. J Int Med Res. 2012;40(6):2109-16.[Abstract]

[10] Verma S, Joy AA, Rayson D, McLeod D, Brezden-Masley C, Boileau JF, Gelmon KA.HER story: the next chapter in HER-2-directed therapy for advanced breast cancer.
Oncologist. 2013;18(11):1153-66. [PubMed]

[11] Hassan R, Bera T, Pastan I. Mesothelin: a new target for immunotherapy. Clin Cancer Res. 2004 Jun 15;10(12 Pt 1):3937-42. [PubMed]

[12] Haridas D, Chakraborty S, Ponnusamy MP, Lakshmanan I, Rachagani S, Cruz E, Kumar S, Das S, Lele SM, Anderson JM, Wittel UA, Hollingsworth MA, Batra SK. Pathobiological implications of MUC16 expression in pancreatic cancer.  PLoS One. 2011;6(10):e26839.[Full Text]

[12] Tang Z, Qian M, Ho M. The role of mesothelin in tumor progression and targeted therapy. Anticancer Agents Med Chem. 2013 Feb;13(2):276-80. [PubMed Central]

Sunday, December 22, 2013

Glycosylation patterns of extra-cellular matrix proteins as fossils of an earlier metabolic state

As mentioned in the previous post, regarding the loss of expression and predicted function of PKD 1, I feel it is probably time to make a note, that to my knowledge has not been made, and is now overdue. The question we propose to answer is "why to extra-cellular matrix proteins so often make use of carbohydrates, or polysaccharides for signaling, when in fact, the traditional view is that proteins do most of the cells signalling work?"
  In short, the answer is that these "sugar" molecules are remnants from a period in the history of evolution that existed before proteins were being manufactured by cells, an "earlier metabolic state" so to speak. This is the term chosen by Harold B. White in his pivotal paper [1] concerning the molecular structure of coenzymes that are used by the cell in another cellular process which dates from "archaic" times, the Krebs cycle and oxydative phosphorylation.
   Two somewhat well known enzymes, at least to students who have taken  a biology course in which ATP production has been  discussed ,  are NAD and FAD .  The D in these stands for dinucleotide. What the "molecular fossils" paper noted was that these prosthetic groups bore more similarity to nucleotides than proteins, and they were necessary for their respective proteins to be functional. The structure poses the question "Why would nature contrive a system like this?" Scientists seek to define systems in terms of rules and laws, and nature seems to randomly through chaos into our otherwise neat theories.  The answer Dr. White proposes is that it is a "molecular fossil".  It is an artifact from a time period before the emergence of proteins, an "earlier metabolic state."
   At the time that "molecular fossils" was published it received little interest. Appearing in 1976, was quite a bit ahead of its time. After a decade or so went by, more molecular evidence had accumulated and Walter Gilbert summed it up in his paper coining the term "RNA World" to further define White's "earlier metabolic state".[2]   Two of the most pivotal citations ( in my opinion ) were those of White and Thomas Cech, the discoverer of  small enzymatic RNA molecules, or "ribozymes" in his terminology.
   Much has been said about the RNA world hypothesis. It has created a virtual assembly line of Nobel Prizes, including one for Cech.
   Possibly more important - at least for the sake of this note, is that there are many bio polymers that appear to be "archaic" and  have potential to shed light on the creation of life in the manor of molecular fossils, and also have medical significance.
  A systematic approach would be to start with the question "What are the most primitive,  or archaic processes that could be defined as "metabolism",  or a "metabolic state?"  The first answer might be to harness a source of energy from the environment. For this system, we already have a consideration since the function of the citric acid, or Krebs cycle is to generate ATP, the energy currency of the cell. For the contribution of the description of an "archaic" system, we have Dr. White. to thank.
   Next, at the risk of being too obvious,  we must have a system to separate, or contain those precious little cell contents. As a review, the energy generation system implemented by the citric acid cycle is entwined with the cell membrane, and "cell wall" of the cells mitochondia. To function, a cell membrane must exist which is impermiable to electrons and ions, so that a concentration and charge gradient can be obtained. It is actually the gradient that drives the ATP synthetase, so an enclosure of some type is implied. Our most commonly known enclosure system is the phospho-lipid bilayer known as the cell membrane. I don't know that the cell membrane has been officially listed as archaic, perhaps because it seems so obvious, but there are alternative candidates. Many bacteria make use of "lipo-polysacarride" cell walls in addition to membranes. There are alternatives for a precise structure.
    Next, cellular information must be stored, and it must be passed from generation to generation. This function has been covered beyond what is even practical to consider. In short, RNA can be coerced into performing many functions related not only to information storage, but implementing functioning enzymes. For more info, search on the key word "RNA World".
   Once our primoridal cells accumulated information, and tucked it away in their RNA, the next thing they needed to do was protect it. There is no use accumulating cool information, only to loose it to a harsh sea environment. Presumably, if not obviously, they commenced experimenting with "picket fence" technology.
  Presumably, their experiments were along the lines of what they knew best, making polysacarrides. One sugar which seems to have been particularly successful since archaic times is fucose. Fucose also goes by the name fucose sulfate, or fucoid. It is known as an "exopolysaccharide" due to its primary location outside the cell membrane. It is a common foundation of marine algae such as kelp. Presumably, the molecular structure of fucoid interacts with the aquatic environment in a way which is protective. Not coincidentally, kelp and its exopolymers of fucose are a good source of industrial emulsifiers.
   So ... Nothing controversial yet....
   So now we must ask "did these primitive inhabitants of the early earth ever decide they had to get together and exchange genetic information with each other?" In other words, is swapping genes a valid behavior in an earlier metabolic state?"
   First we must consider the possiblity of "no". Hmm.
    On the other hand life as we know it seems to be the result if continuous hybridization among various environmentally  selected information accumulations. It seems unlikely that we could deny anything we could call "metabolic" the property of genetic hybridization with its contemporaries. Our whole concept of "advancement" seems to be based upon systematic hybridization.
   The presumption of systematic hybridization in an environment where cellular life must:
  1. Be able to establish and hold chemical, ionic and electrical gradients
  2. Defend themselves from the chemical forces of the environment
  3. Defend integrity of accumulated informational ( nucleotide ) material
seems to imply that that a lock and key recognition mechanism was a valid foundational function of our "archaic" life form. A "wet handshake" so to speak. So what evidence of this may we point to today?
  The point of contact in today's most common hybridization mechanism is the egg-sperm interaction.
 It has been shown that the "receptor for egg jelly"  (REJ) protein on the acrosome of the sperm is
"keyed" to intract with "fucoidin", or "fucose sulfate of egg jelly" in the terms of Vacquier and Moy [3].
  For the reasons mentioned, this is a "non-traditional" interaction in that it involves polysaccharides and their recognition, and it involves a polysaccharide which has a known archaic function ( cellular protection). It is also a reaction that is preserved in modern life, the REJ- fucose sulfate reaction seems to qualify as a "molecular fossil".
  As a generally accepted principle, new genes form by duplication of existing genes, and then evolving
 more slowly as a function of natural selection in their environment. It can be speculated ( hypothesized ) that the REJ gene of the sperm, our most primitive cell, has under gone one or many duplications to give rise to the PKD1 gene. We known form gene sequence analysis that PKD1 contains an extra-cellular "suREJ" or sea urchin receptor for egg jelly domain.  We apparently can thank Vacquier, Moy and their sea urchin laboratory for naming this domain, and identifying the associated duplications in the sea urchin as well as the human species [4].
   PKD1 is a gene that seems to be found in all "epithelium". Epithelium has the defining property that it is a sheet of cells which are bound to a "basal membrane". The basal membrane in turn contains highly glycosylated extra cellular matrix proteins. Most likely ( but unproven ) , fucoid type decorations in the
basal membrane are key players in cellular recognition of the membrane, and polarization development.
  In this case, PKD1 would be a second generation molecular fossil, that is a molecular fossil by duplication of the REJ gene. Likewise, the basement membrane of epithelium can be thought of as a "display decorated by polysacharide fossils for the purpose of large scale structural organization."
  PKD 1 has something that most research targets in the RNA world don't have. That is, a common severe medical condition. PKD, itself as an acronym for Polycystic Kidney Disease. Thus, spending large amounts of medical foundation money on its function would actually be justified.
  Conversely, a recent announcement by a South Korean group has made obsolete most of what was previously known of the pathological dynamics  of PKD 1.


[1] White H.B. (1976). Coenzymes as fossils of an earlier metabolic state. Journal of Molecular Evolution,7, 101-104 [Abstract]

[2]  Walter Gilbert  Origin of life: The RNA world Nature 319, 618 (20 February 1986); 

[3] Vacquier VD, Moy GW.
The fucose sulfate polymer of egg jelly binds to sperm REJ and is the inducer of the sea urchin sperm acrosome reaction.
Dev Biol. 1997 Dec 1;192(1):125-35.

[4] Gunaratne HJ, Moy GW, Kinukawa M, Miyata S, Mah SA, Vacquier VD.
The 10 sea urchin receptor for egg jelly proteins (SpREJ) are members of the polycystic kidney disease-1 (PKD1) family.  BMC Genomics. 2007 Jul 13;8:235. [PubMed Central]

Wednesday, December 18, 2013

Kidney cysts are shown to arise from an epigenetic "second hit"

Autosomal Dominant Polycystic Kidney Disease has always been observed to be an inherited disorder.  But, as the name implies, there has always been a puzzle as to the nature of onset of kidney cysts.
Recently, in the past five years or so, there has been a surge of interest in what has been called epigenetic contributions to cancer causation. Traditional cancer research has historically been almost completely based upon what is known as the "central dogma" of molecular biology, which implies that cellular information flows in one direction, DNA provides a template for messenger RNA, and mRNA is in turn translated into proteins, the actual workhorses of cellular metabolism.
  As a result of strict adherence by researchers to the central dogma, research has focused almost exclusively on DNA sequence as a causal foundation for loss of cell cycle control and neoplasm. In other words, somatic mutations, where somatic refers to cells of the body as differentiated from germ line or directly inherited sequences.
  In recent years, there has been comprehensive acknowledgement of the observation that in many cases, the somatic sequence of the causative gene is intact, but gene expression, the result of the previously mentioned central dogma, has been blocked. If the expression of a so called cancer suppressor gene is blocked, the cell is at high risk to become cancerous.
  Epigenetic modifications are numerous, but the primary method, and easiest to detect using laboratory methods is called changes to DNA methylation. Methylation applies to the portion of a gene called a promoter. In order to initiate transcription of DNA to RNA, a complex of proteins including a polymerase must assemble on the DNA strand upstream of the start of transcription.This location is called a promoter.
  A promoter for a particular gene may contain a large number of the dinucleotide CG. These are referred to as CpG islands, where the p refers to the phosphate that joins two nucleotides in DNA structure. As the result of a cellular differentiation process, the CpG islands can become methylated, meaning a methyl group has been has been added. Once a promoter has been methylated, a protein known as methyl CpG binding protein can bind to the promoter. At this stage, expression is suppressed.
 Methylation as a result of differentiation  or expression constitutes what is known as "paradigm change". This is because cellular information flow in an epigenetically regulated gene expression system is fundamentally different than that implied by the central dogma.
  In a typical gene that is regulated, or blocked by promoter methylation, the messenger RNA interacts with a complementary micro RNA, or short sequence with which it can combine to form a helix. That short helix can be processes by the cell into a RISC which will target the promoter of that specific gene for repression by promoter methylation. A RISC, or gene suppression complex makes use of a particular strand of RNA to identify the proper gene in the nucleus of the cell, and bind to and methylate the promoter of that gene.
  This is why the recent announcement by a team in South Korea qualifies as a "paradigm shift" in our understanding of the genesis of a particular medical condition, kidney cysts [1] .  It seems that epigenetic analysis, as opposed to traditional genetics [2]  provided the data to advance understanding of cyst causation and formation.
  What Bae and Woo and are saying then, is that an observed lack of expression of PKD1 is due to promoter methylation, presumably as a metabolic incident which triggered the formation of a RISC which targeted PKD1 expression at the DNA level, and blocked it.
   We should make a quick note here that this explanation is significantly different than that offered by current researchers sponsored by the NIH. Current views are summarized in an NIH sponsored view provided by Harris[2]. The first line of the abstract summarizes puzzling nature of PKD presentation. It is as follows:
    A 50% dosage reduction of gene product, known as haploinsufficiency, as found in heterozygotes with an inactivating mutation, is not usually associated with a detectable physiologic effect. 
 In other words, the "autosomal dominant" nature of PKD is in itself unusual, we usually expect genetic syndromes to occur in the homozygous case. In most cases, of a patient has one good allele, health is maintained.  This has inspired the "second hit" hypothesis, or loss of heterozygosity. Indeed, the two hit hypothesis is attractive because cysts are observed to be "clonal", or a large number of cells having decended from a single cell, as identified by genetic analysis.
   But, we should note here, that epigenetic changes such as promoter methylation are also passed from parent to progeny across mitosis by a process of methylation pattern copying. Clonality of cysts in no way excludes epigenetic considerations.
  Because, in a relatively high percentage of cysts,  heterozygosity seems to be preserved, Harris moves on to an alternative hypothesis, "haploinsufficiency". This if a large word to imply an affect which is actually almost never observed. That is, on allele is insufficiently "powerful" enough to produce enough protein to reach a threshold level to maintain healthy expression. Based upon this hypothesis, Harris concludes:
The emerging model in ADPKD is of loci that are dosage sensitive, which likely alone explains some disease phenotypes.
As such, they seem ready to close the book on ADPKD. Harris then acknowledges:

    This study was supported by National Institute of Diabetes and Digestive and Kidney Disease grant DK058816.
Given the recent publication year, One would assume that his views reflect the views of this division of the National Institutes of Health - NIH.

   So then, the questions of interest to researchers are fundamentally different in this new paradigm. The first question is a primarily a computational problem, at least initially. That is to find any and all micro-RNAs which regulate our gene of interest.
interaction of micro RNAs and their substrates is dependent upon the secondary structure of the substrate. This also is primarily a computational problem, and for large transcripts, that is, most of the genes we are interested in, it is considered an unsolved problem. In fact, for large transcripts, there is no single structure which best represents our target, but a family of structures referred to as a Boltzman ensemble by one of the prefered computational tools SFOLD.
  Renal cysts can be caused by many abnormalities, but two of the must common sources of cysts are abnormalities to the genes PKD1 and PKD2.   PKD 1 is known to be a trans membrane receptor which has a lectin domain on the extracellular portion of the transcript.  A lectin is a protein which recognizes and interacts with a polysaccaride or a "sugar" group. Many proteins, particularly those which make up the extracellular matrix, are "decorated" with sugar groups in a process called glycosylation, which takes place in the Golgi after the polypeptide exits the endoplasmic reticulum.
   In the case of PKD 1, the lectin domain is called a "sea urchin receptor for egg jelly" or an suREJ domain, after the model in which it was first identified. The polysaccaride substrate for suREJ is fucose, which is also found in all epithelium basement membranes.
    In the sea urchin sperm - egg interaction, recognition results in calcium release as a sperm activation signal. Perhaps not coincidentally, PKD2 is known to be a calcium channel.
   So then, we get, finally, to Woo and Bae [1] to discuss the impact of epigenetic paradigm shift on our understanding of kidney cyst formation. They report that using a laboratory method to sequence and obtain a promoter methylation profile for somatic sequences ( those cells associated with cyst development in patients ) They observed promoter methylation in those cells where expression of PKD1 was reduced or eliminated.

Further, they performed a cell culture using Madin Darby Canine Kidney (MDCK) cells, which simulated cyst formation in a laboratory model system, and observed a reduction in cyst formation tendency when the cells were treated with therapeutic agents which are known DNA methylation inhibitors. Somewhat shockingly, Bae and Woo show how a paradigm shift in viewing a pathology can lead to a proposed therapeutic regimen where there was none.
  Bae and Woo certainly does not provide a close to the issue of autosomal dominant polycystic kidney disease (ADPKD). It certainly provides a new avenue of research. An "avenue" of research, has in the past, been guided by the principles of the central dogma. In abbreviated terms, locate DNA mutations, map them on to their amino acid in each protein, engineer affected proteins, and evaluate their change of function in terms of binding to their substrate, or enzymatic activity.
   In the new paradigm, there is no further interest in protein function. What is of interest is making a working map of the genes messenger RNA secondary structure, and seeing if it can be differentiated from that of the mutated gene. This itself is a tall computational task. To my knowledge, success in differentiating wild type and mutant (WT) and (MT) secondary structure has been claimed once, in Alcorn.
  Beyond that, DNA promoter methylation is the result of the interaction of an mRNA and its antisense micro-RNA partner. As a measure of complete systemic knowledge, PKD1's antisense micro-RNA should be identified. This is initially a computational problem to identify candidates from genomics, and then a verification step to profile gene expression and see which micro-RNAs are actually expressed in affected kidney cysts.
  All of this leaves the question of the calcium channel, PKD2 wide open. All of Bae and Woo's work could be repeated in cysts affected by PKD2 mutations. Since Bae and Woo seem to have used general methylation inhibitors in their follow up lab experiment with MDCK cells, the same results might be expected from cysts that arise as a result of PKD1 or 2.


[1] Woo YM, Bae JB, Oh YH, Lee YG, Lee MJ, Park EY, Choi JK, Lee S, Shin Y, Lyu J, Jung HY, Lee YS, Hwa Genome-wide methylation profiling of ADPKD identified epigenetically
 regulated genes associated with renal cyst development.

Hum Genet. 2013 Oct 16.  [Abstract]

[2]  Peter C. Harris What Is the Role of Somatic Mutation in Autosomal Dominant Polycystic Kidney Disease?  JASN July 1, 2010 vol. 21 no. 7 1073-1076  [Full Text]

[3] Jjingo D, Conley AB, Yi SV, Lunyak VV, Jordan IK.
 On the presence and role of human gene-body DNA methylation.
Oncotarget. 2012 Apr;3(4):462-74.  [Abstract] [Full Text]

Wednesday, September 4, 2013

Description of the "Alcorn Prize" in computational biology

Molecular biology is undergoing a "paradigm shift" in the way it views and researches pathological medical conditions. The overall catchword in the shift is "epigenetics". In what we call the traditional view, or central dogma, as Francis Crick described it, describes all biological systems as having information encoded in DNA, transcribed to RNA, and translated to sequences of amino acids known as proteins. This view has held up remarkably well as an organizing foundation for research for quite a few decades.  A traditional research project in medical genomics assumes that any anomalies in DNA sequence will be mechanically transcribed to RNA and then on to protein. Thus, it is assumed that malfunctioning genes result from malformed proteins. This view works well in laboratory model systems. More recently, primarily within the last decade, it has been realized that information flows in the cell are quite a bit more sophisticated.
   An example is the well known "guardian of the genome" known as p53.  As one of the earlier genes found to be highly associated with cancer, there have been literally thousands of papers and studies analyzing the myriad of cellular functions associated with p53. More recently it has been found that in many cases, if not most cases, p53 expression is lost in (ovarian) cancer. What this abstract describes in a pattern where the promoter of the gene has become methylated by epigenetic processes, and as a result, there is no p53 being expressed at all, regardless of anomalies to the genomic sequence. In other words, there is only marginal utility to doing further laboratory research on the p53 protein, as in most as many as half the cases reported ( if this study is representative), there is no p53 actually in the cell, itself a serious problem. A similar situation exists for the most of the usual culprits in cancer research, pRB ( retinoblastoma ), BRCA1/2, and telomerase. A slight variation in the case of telomerase in that this is a gene that is thought to be best unexpressed in somatic cells as a defense against runaway proliferation known as the Hayflick limit.
   Regulation of a genomic transcript at the RNA level is known to be a function of a particular RNA's secondary structure.  RNA secondary structure, for large transcripts is not a fixed description but a probabilistic description in that the actual shape of a transcript changes continuously as a result of its size, Brownian motion, and interaction with other transcripts ( regulatory micro RNAs , miRNA ).
   How then should can we go about prediction structural changes associated with genomic anomalies, such as those seen in BRCA1/2? This is the unsolved problem which faces the entire research community, it that prediction of secondary structure for large transcripts has proven unsuccessful.
   With one small exception.
   In 2009 Alcorn et. al. reported that they had designed a ribozyme which differentially targets wild type (WT) and mutant (MT) transcripts. The excerpt below from the "Hammerhead ribozyme vectors" section of the methods clearly states that the investigators used S-fold to model the the predicted secondary structure of the normal ( WT ) and mutant ( MT ) transcripts. They discovered that each of the mutations associated with pseudoachondroplasia was also associated with a flip in the secondary structure of the mRNA. As such, they could design a ribozyme, or any antisense based tool to selectively targeted the MT transcript and had reduced activity associated with the WT transcript.

Excerpted from Alcorn:

 S-fold was used to compare the predicted secondary structures for normal and three mutant COMP sequences chosen for this study, all of which affected the predicted folding pattern of the target transcripts in a manner that could impact interactions with targeted ribozymes. Ribozyme 56 cuts at nucleotide 56 of the COMP mRNA: 5′-CUGCCCUCGGCGCGUCCGGACAGGGCCAG-3′. The cut site (underlined) lies in the region of the transcript encoding the C-terminal half of the signal sequence extending just into the coding sequence for the mature protein.

 It is known that interaction between a target transcript and it's antisense partner is dependent upon the degree of self-binding ( cis ) within the target. Each biologically relavant mRNA consists of a series of stems and loops, where the stems are double stranded RNA and the loops are single stranded RNA. In this case, the investigators chose a cut site (GUC) just upstream of the start codon.
  It should be noted that in terms of epigenetic regulation of expressed proteins, the fact that all three mutants were associated with structural modifications in the mRNA of COMP is in itself suspicious. This observation, associated only with computational preliminaries to the investigation should have opened the door epigenetic considerations in the investigation.

In the discussion section of Alcorn, the author discusses the significance of his discovery:

Most importantly, the ribozyme is more active in knockdown of three well-described MT-COMP species in chondrocytes when compared with normal WT-COMP (Fig. 4). Such preference presumably occurs because of differences in the secondary structure () or accessibility of target sequences in ribonucleoprotein complexes () of the transcripts encoding wild-type versus mutant COMP. Selective targeting of mutant COMP over WT-COMP by a ribozyme has never been reported.

   Thus, Alcorn is stating that he predicted the change in secondary structure is somewhat unique in that the transcript for COMP weighs in at 2471 base pairs. This is well beyond what is currently considered within todays current computational technology.
  Thus the  Alcorn Prize is completely defined. It is the team or person which develops a computational biology system that can replicate Alcorns achieviement for COMP for broadly relavant transcripts such as p53, retinoblastoma (pRB )  Breast Cancer ( BRCA1/2) and polycystin. Good luck to all. Ready GO!

Saturday, June 15, 2013

The emerging connection between cancer epidemiology and environmental toxcants

This note is largely a review, explanation, and extension of the review produced by Su et. al. [1]. Until recent advances in our understanding of cancer initiation, the connection between cancer and environmental contaminants remained elusive. The presumption for a proposed carcinogen was that it had the ability to integrate into, disrupt, or mutate DNA. As such, the carcinogen would create the heritable changes in cellular progeny that we observe in cancer. On the other hand, many suspected carcinogens such as methyl mercury, other heavy metals, and environmental chlorocarbons ( dioxin ) seemed to slide under the radar in terms of their potential role in cancer epidemiology.
  As described by Su et. al., the emerging paradigm change in toxicology falls under the keyword "epigenetics". As described in the reference, the actual mechanisms of epigenetics are very broad, but in the simplest overview, it suffices to say that each (or most) genes in the chromosome have a "switch" mechanism associated with them. Each gene that may be transcribed is associated with a "promoter" region which is upstream of the start of transcription. When a transcription factor specific for that promoter binds to it, a transcription complex is initiated and begins to produce a transcript.
   The process of transcription described above can be blocked. If the promoter, or the region around the TATA box in the animation has an abundance of  "CpG Islands", these dinucleotides can be methylated, and will attract a binding protein ( Methyl CpG Binding Protein, or MeCP2 ).  The pattern of promoter methylation, as well as other epigenetic changes associated with chromatin, are the drivers of cell differentiation as cells evolve from stem cells to differentiated somatic cells.
  New high speed laboratory tools that can rapidly access the methylation patterns of a cell samples DNA have given researchers a new view of cancer causes. One such gene is retinoblastoma. Rb, or pRb was one of the first known genes associated with cancer, and is known as a key regulator of the cell cycle. In a rare occurrence in science, a long accepted hypothesis may be changed. Initial study of  pRB led to what is known as the "two hit hypothesis". More recently, it has been observed that pRb may be more often associated with epigenetic changes than genetic changes[2]. Likewise, another newsworthy gene, BRCA1 has been shown to be a largely epigenetic problem associated with inappropriate methylation of a checkpoint gene.[3].
  Moving back to the original theme of the article, although  methyl mercury and heavy metals have been difficult to connect to DNA mutations, they have been fairly easy to connect with the maintenance of methylation patterns associated with promoters.[4].
  Although Su [1] and others describe "progressive global hypomethylation" as an epigenetic description of carcinogenesis I have not yet seen the presumptive cause speculated upon. At this juncture, it seems to be an increasingly small jump, so lets just go ahead and make it. The term global hypomethylation means that as the stage of a cancer progresses, the overall number of methylated genes declines. As mentioned in Su et. al.,  methylation of CpG islands is maintained by two DNA methyltransferases, DNMT1 and DNMT3.
When the cell divides, he duplicate copy of the DNA must have the methylation patterns copied from the template strand. This is the job of DNMT1. There must be a checkpoint system in place to make sure that progression through the cell cycle is halted until not only DNA synthesis, but methylation pattern duplication is complete. That simple sentence seems to never be said, as it is too speculative, but at this juncture, it is increasingly obvious. For the sake of education and communication, the term incomplete epigenetic pattern duplication due to check point loss is infinitely less cryptic than progressive global hypomethylation  as it is typically referred to in clinical literature.
  There is one last catch. As methylation patterns are lost globally, they can also be inappropriately gained,  by new or "de novo" methylation, New methylation is by DNMT3. Perhaps the confusion here, for the uninitiated, is that de novo methylation can occur as a result of global hypomethylation. Many, if not most transcripts have antisense partners, or "mirrors". They consist of short non coding RNAs called microRNAs or miRNA. The job of a micro rna appears to be to shut down a gene that is no longer appropriate as differentiation continues.  When a micro RNA is transcribed, it finds its complementary partner and forms a silencing complex. This is also called RNA interference. This process, mediated by DROSHA and DICER leads to a complex of  RNA and proteins that will to to the nucleus, find the appropriate gene and silence it by recruiting and mediating DNMT3.
  There is also quite probably the case where antisense microRNAs implement a "surveillance" system on transcripts. That is,  a transcript and its antisense partner are both transcribed, and they do not bind due to secondary structure considerations associated with the sense transcript. If the sense strand loses secondary structure stability due to mutations in the sequence, or un-natural contaminants in the system, the sense strand binds with the antisense strand and shuts down the gene by activating an RNA interference response.
  The "surveillance hypothesis" explains why either germline mutations or environmental toxins could lead to the inappropriate methylation and shut down of a tumor suppressor gene like pRB or BRCA1/2.

[1]  Su LJ, Mahabir S, Ellison GL, McGuinn LA, Reid BC
Epigenetic Contributions to the Relationship between Cancer and Dietary
 Intake of Nutrients, Bioactive Food Components, and Environmental Toxicants.
Front Genet. 2011;2:91.  [PubMed] [Full Text]

[2] Domenico Mastrangelo, Cosimo Loré, Giovanni Grasso
Retinoblastoma as an Epigenetic Disease: A Proposal
JCT> Vol.2 No.3, August 2011 [Full Text]

[3] Birgisdottir V, Stefansson OA, Bodvarsdottir SK, Hilmarsdottir H, Jonasson JG, Eyfjord JE.
Epigenetic silencing and deletion of the BRCA1 gene in sporadic breast cancer.
Breast Cancer Res. 2006;8(4):R38. [Full Text]

[4] Baccarelli and V. Bollati
Epigenetics and environmental chemicals
Curr Opin Pediatr. 2009 April; 21(2): 243–251. [Full Text]

Thursday, February 21, 2013

Bone Marrow Stem Cells Provide a micro-environment that evades the immune system

A recent scientific announcement not immediately related to cancer may in fact have profound implications related to our understanding of cancer initiation and metastasis. It has been shown that tuberculosis bacteria can evade the immune system by "hiding" in the stem cells of the bone marrow.[1] [2]. This has profound implications in cancer research becasue these "CD271 positive" cells have also been identified as primary correlates to nucleation of cancer in bone as a result of metastasis of breast cancer cells.[3]
  The significance of "Das" is that the mechanism of "DeGiorgi" has not been elucidated. With Das, we can now hypothesize that the observatons and correlatons in De Giorgi are due to protection from the normal operation of the immune system in the micro-environment of bone marrow, particularly in association with CD271 positive hematopoetic stem cells.
 Even more importantly,  down regulation of the immune system in bone marrow has been predicted based upon what has been known as the "rattlesnake hypothesis". Briefly stated, the rattlesnake hypothesis proposes that cells of the body which must for biological, reproductive and metabolic reasons, express telomerase must also express "cancer/testes antigens". (CT antigens)
  In other words, it has long been observed and proposed that the "Hayflick Limit" operates as a native defense against cancer. The Hayflick limit is imposed by the repression due to methylation in the promoter region of telomerase. Certain cells, such as germ cells ( testes) and stem cells must express telomerase. As such, they need a different primary defense against loss of control of the cell cycle and the imitation of cancer.  In these cells, antigens known as CT antigens provide a clue to the immune system that the cell posses a particular "danger". As such expression of telomerase represents the fangs , and expression of CT antigens represents the rattle.  Thus, they hypothesis of the function of CT antigens is to serve as a flag to the immune system that the cell is expressing telomerase, and if it is not in an immune protected compartment of the body, such as the tests, the brain or bone marrow, the immune system will recognize and attack the cell.
The hypothesis explains many long observed phenomenon such as why sperm cells generate an immune response in the host, and why the brain must have a barrier and its own immune system.
[1] Nsikan Akpan Stem Cells: Safe Haven For TB The Scientist
[2] Das et al., CD271+ bone marrow mesenchymal stem cells may provide a niche for dormant Mycobacterium tuberculosis, Science Translational Medicine, 2013. [Abstract]
[3] De Giorgi et. al, Cancer Biol Ther. 2011 May 1;11(9):812-5. Epub 2011 May 1. Mesenchymal stem cells expressing GD2 and CD271 correlate with breast cancer-initiating cells in bone marrow. [Abstract]