papilloma. a wart; generally benign lesions that rarely progressed to squamous cell caricnomas of the skin
papova. a family of viruses signifying papilloma, polyoma, and vacuoles
T antigens. tumor-associated proteins
episomes. unintegrated genetic elements
provirus. the DNA version of a RNA viral genome
proto-oncogene. a precursor to an active oncogene
insertional mutagenesis. a mechanism of proto-oncogene activate in which a viral, constitutive promoter takes transcriptional control over a proto-oncogene.
SV40 and human AdV were lytic (and therefore non-tumorigenic) in permissive environments (i.e. human and monkey hosts), but (occasionally) were tumorigenic in non-permissive environments (i.e. rodent hosts).
Tumor viruses induce multiple changes in cell phenotype including acquisition of tumorigenicity
In 1968, it was shown that SV40 viral DNA was tightly associated with chromosomal DNA, indicating integration into the chromosome
cervical cancer
HPV
E2 gene. product represses transcription of viral oncogenes; it is disrupted or discarded when HPV integrates into the chromosome with tumorigenic effect
E6, E7 oncogenes drive transformation
Retroviral genomes become integrated into the chromosomes of infected cells
pol gene encodes an integrase along with reverse transcriptase
because RNA viruses have a dedicated integrase enzyme (in contrast with DNA viruses), they are completely integrated into the host genome, rather than partially (DNA viruses are rarely and randomly integrated into the chromosome).
A version of the src gene carried by RSV is also present in uninfected cells
when a virus integrated next to a proto-oncogene, the constitutive promoter of the virus could activate the proto-oncogene, turning it into an oncogene.
the same proto-oncogenes activated by viruses could also be activated via non-biological factors
erbB from avian erythroblastosis virus (AEV) had related gene erbB2/neu/HER2 in human breast cancer
gene amplification of HER2 correlated with lower survival rates
make takeaway: a common set of cellular proto-oncogenes might be activated either by retroviruses (in animals) or, alternatively, by nonviral mutational mechanisms operating during the formation of human cancers
Proto-oncogenes can be activated by genetic changes affecting either protein expression or structure
chromosomal translocation (e.g. chromosomes 14 and 8 in Burkitt's lymphoma)
oncogenesis could be due to regulated → constitutive promoter
oncogenesis could also occur because microRNA recognition sites are deleted
e.g. Let-7 miRNA recognition sites in 3'UTR of HMGA2 mRNA deleted ⟹HMGA2 mRNA not degraded ⟹ increased levels of HMGA2 ⟹ (still unknown) alters chromatin configuration to facilitate cell transformation
miRNA also affect cell differentiation; changes in miRNA expression due to chromosomal translocations can prevent cells from differentiating, keeping them in a cell state that easily becomes cancer
overall, mechanisms leading to overexpression of genes in cancer cells remains poorly understood
A diverse array of structural changes in proteins can also lead to oncogene activation
mitotic recombination (not meiotic recombination) occurs at rate 10−5 to 10−4 per cell generation, much more likely than direct mutational inactivation
loss of heterozygosity (LOH). a type of genetic abnormality in which one copy of an entire gene and surrounding chromosomal region is lost
mitotic recombination can be a cause (loss of one parent's allele)
gene conversion (another potential cause of LOH). DNA polymerase temporarily switches templates and replicates DNA belonging to the homologous chromosome
chromosomal nondisjunction. chromosomal region breaks off, leading to hemizygosity
The Rb gene often undergoes loss of heterozygosity in tumors
increased risk of gliomas, pheochromocytomas, and myelogenous leukemias
Nf1 has high sequence homology with Ras-GAPs (GTPase activating proteins)
Ras-GAPs negatively regulate Ras proteins by activating the GTPase activity of Ras, which stops Ras' function
evidence indicates that
Nf1 LOH primarily targets undifferentiated neural crest-derived cells related to Schwann cell precursors
once such cells lose all Nf1 function, they initiate development of other tumor growths by inducing coproliferation via paracrine signaling
Cre-Lox uses flanking LoxP sites and the Cre recombinase to knockout a gene conditionally (e.g. tissue-specific knockout). This can be done because the Cre recombinase is only expressed under a tissue-specific promoter.
The alleles that are modified to have flanking LoxP sites are called floxed alleles. They are often denoted GENEFl/Fl, e.g. NF1Fl/Fl. There is no specific allele type (e.g. +/−) since the allele is controllable by the experimenter.
recruitment of mast cells to incipient neurofibromas is critical to formation of tumors
when surrounding mast cells have Nf1+/− genotype, neurofibromas are likely to form. If Nf1+/+ (wild-type), neurofibromas are much less likely to form.
APC facilitates egress of cells from colonic crypts
TSGs can also be linked to processes related to disease progression
KEAP1 is a sensor for chemical and environmental stresses
KEAP1 normally targets NRF2 for ubiquitylation in the cytoplasm; NRF2 has lifetime of 20 minutes
ROS react with cysteines on KEAP1 → KEAP1 release NRF2 → NRF2 translocates to the nucleus, joins with MAF transcription factor family→ activate genes involved in cellular detoxification
when ROS are cleared, cysteine residues are restored and KEAP1 continues to sequester NRF2
some cancers disable KEAP1 so that NRF2 is constitutively activating ROS detoxification genes
cancer is more resistant to radio- and chemotherapy
mutations on NRF2 that prevent NRF2 from binding to KEAP1 also enable constitutive ROS detoxification
inactivating NRF2 can make cancer formation easier (since ROS can cause mutations) but slow down cancer progression since cancers are generally in high oxidative stress environments (and need NRF2 detoxification to continue to survive)
Not all familial cancers can be explained by inheritance of mutant TSGs
in hormone-responsive tissue, potential mechanism: obesity → increased hormone production → proliferation of epithelial cells in the endometrium
in non-hormone-responsive tissue, potential mechanism: hyperinsulinemia→ increased synthesis of IGF-1, reduced production of IGF-1 antagonists in the liver (IGF-binding protein 1, IGFBP-1) → IGF-1 binding activates Akt/PKB, providing strong anti-apoptotic signals
@weinberg2e/11.1 Most human cancers develop over many decades of time
treating mice with carcinogen, then treating them with 5-FU (kills actively dividing cells) didn't stop progression of cancer ⟹cancer initiator is a cell type that only divides occasionally
blocked differentiation is a common theme in the development of blood cancers
we don't know the precise identities of stem cells targets of transformation
could be puripotent hematopoietic stem cell
but could also be an early committed progenitor, and mutation is stored long-term via de-differentiation back to a stem cell
this is a hypothesis; it is not known
can happen if stem cells are killed and need to be resupplied via de-differentiation
Apoptosis, drug pumps, and DNA replication mechanisms offer tissues a way to minimize the accumulation of mutant stem cells
stem cells in mouse crypts, in response to DNA damage, are primed to activate apoptosis rather than risk replicating errors from repair
it is possible some stem cells have stronger DNA repair machinery; this is hypothesized since some cancer stem cells are resistant to apoptosis and have enhanced DNA repair mechanisms
stem cells have enhanced drug pumps
⇐ stem cells efficiently pump out certain fluorescent dye molecules
due to plasma membrane protein called Mdr1 (multi-drug resistance 1)
Cell genomes are threatened by errors made during DNA replication
hydrogen and hydroxyl ions in water can cause accidental DNA damage, such as depurination, in which the bond linking a purine base to the deoxyribose breaks
deamination of 5-methylcytosine can evade detection because its resultant base is thymine, a normal base (i.e. can cause C>T point mutations)
ROS generated from reduction of oxygen to water can also cause damage
estimated 1-2% oxygen molecules consumed by mitochondria end up as ROS
increased production in inflammation
Cell genomes are under occasional attack from exogenous mutagens and their metabolites
xeroderma pigmentosum is a result of defective NER
7/8 XP-associated genes: XP{A,B,C,D,E,F,G} encode components of the NER complex
XPV encodes error-prone DNA polymerase pol-η used when regular DNA polymerases (e.g. pol-δ) are unable to copy over unrepaired DNA lesions (e.g. pyrimidine dimers).
hereditary non-polyposis colon cancer (=HNPCC). another cancer susceptibility syndrome caused by inherited defects in DNA repair
mismatch repair genes are defective
85-90% cases: MSH2 and MLH1
15% cases: MSH6 and PMS2
leads to high mutation rates in genes with microsatellite repeats
e.g. type II TGF-β receptor (TGF-βRII)
A10→A8; inhibition by TGFβ growth factor no longer occurs
A variety of other DNA repair defects confer increased cancer susceptibility through poorly understood mechanisms
evidence for participating in genomic maintenance is guilt-by-association
found in complexes with RAD50/Mre11 and RAD51 proteins, which are homologs of proteins in yeast involved in repairing DNA breaks caused by x-rays
BRCA2 partial loss of function
→ high rates of illegitimate recombination (between chromosomal arms that are nonhomologous)
NHEJ takes over if HDR fails
NHEJ is actually plays a major role in VDJ recombination
also enables class switching between immunoglobulin types
→ deregulation of centrosome number
BRCA1/BRCA2 breast cancers that initially respond to cisplatin therapy develop resistance by developing back-mutations that restore the function of the BRCA1/2.
BRCA1 likely a scaffold for DNA repair complexes
BRCA1 loss → H2A not ubiquitylated → H2A no longer heterochromatin, variety of repeat sequences expressed → genetic destabilization
The karyotype of cancer cells is often changed through alterations in chromosome structure
chromosomal instability (CIN). phenomenon where a population of (usually cancer) cells have a large distribution of chromosome copy numbers
both CIN and microsatellite instability (MIN) provide mutability for cancer tumor growth
changes in chromosome number are usually consequences of mis-segregation of chromosomes during mitosis (nondisjunction)
can be caused by failure of quality control ensuring chromatids are attached to their spindle fibers
another source is kinetochores (the disk-shaped center located on the chromosomes that attaches to spindle fibers) containing too many spindle fibers (normal is 20-25 microtubules)
leads to merotely where a kinetochore is attached to spindle fibers on both sides of the dividing cell, resulting in the loss of the chromosome
spindle assembly checkpoint (SAC) fails to check correct attachment of kinetochores to chromatids
can also be caused by incorrect spindle assembly, such as supernumerary centrosomes at interphase
CIN continues even after tumor progression has completed (unlike breakage-fusion-bridge cycles, which happen during progression and then stop)
molecular defaults leading to CIN
duplication of centrosomes occurs at G1/S transition
HPV E7 protein can destabilize centrosome number via pRb loss-of-function
HPV E6 protein allows cell to tolerate centrosome abnormalities by disrupting p53
common cancers can be caused by inherited defects in caretaker genes
healthy cells taken from cancer patients (but not cancer cells) are more vulnerable to ionizing radiation than healthy cells taken from healthy control patients
What types of evidence suggest that karyotypic alterations of cell genomes are not absolutely essential for neoplastic transformation?
When calculating the rates of mutation required in order for multi-step tumor progression to reach completion, what parameters must one know in order for such a calculation to accurately describe the actual biological process?
How does our understanding of defective DNA repair processes in tumor cells make possible the development of new anti-cancer therapeutic strategies?
In which ways do the defectiveness of p53 function and resulting defects in apoptosis and DNA repair facilitate the forward march of tumor progression?
What types of tumor promotion, as described in Chapter 11, favor the genetic evolution of premalignant cell clones?
What evidence implicates mutagenic chemicals originating outside the body in the pathogenesis of human cancers? How can one gauge their contribution to human carcinogenesis compared with that of mutagens and mutagenic processes of endogenous origin?
How do defects in various cell cycle checkpoints allow for accelerated rates of the accumulation of mutations?
How do the biological properties of stem cells help to reduce the rate at which tissues accumulate mutant genes?
How does the existence of cancer stem cells affect the calculations of the rate at which mutations must be accumulated in order to allow multi-step tumor progression to advance?
How does the genetic heterogeneity in the human gene pool affect the functioning of various types of biological defenses that have been erected to prevent the accumulation of mutant alleles in human somatic cells?
matching mutational patterns induced by carcinogens and the mutations found in particular cancers can reveal potential causes of cancer (bridging molecular cancer genetics and epidemiology)
mutational patterns = mutational spectra
devised algorithm to decompose mutational spectra into multiple mutation processes (which can be correlated to a causative agent)
transcription factors bind to enhancers and silencers
activators⊂ transcription factors that bind to enhancers
repressors⊂ transcription factors that bind to silencers
RNA pol II generally synthesizes both strands, and then is paused in transcriptional pausing before other signals either allow or prevent it from resuming elongation
Promoter methylation represents an important mechanism for inactivating tumor suppressor genes
mammalian cells can have methylation on the cytosine in CpG sites (denoted MeCpG)
often causes repression of transcription
cytosine is the base that is methylated
mechanisms not entirely known
one mechanism: protein complex can bind MeCpGs and second subunit functions as histone
de novo methylation. where methylation of histones can lead to methylation of CpGs
cancer is associated with two patterns of methylation
global _hypo_methylation
_hyper_methylation at CpG islands
CpG islands are often affiliated with gene promoters (~70% genes have promoters with CpG islands)
some cancer cells in tumors can have ~5% genes hypermethylated
may be due overexpression of DNA methyltransferase 3B (DNMT3B)
example of physiological consequence of methylation
RARβ2 gene encodes a retinoic acid receptor that arrests the cell cycle in the presence of retinoic acid. methylation of this promoter prevents cells from responding to retinoic acid.
lots of communication between cells in a tissue is via growth factors
growth factor in serum that promotes clotting: platelet-derived growth factor (PDGF)
PDGF attracts fibroblasts and then stimulates their proliferation
many oncogenes encode growth factor receptors; mutations in them trick cells into believing they have encountered large concentrations of growth factor
independent of kinase function, Src is also a phosphoprotein (carries phosphate groups attached covalently to one or more of its amino acid side chains).
tyrosine phosphorylation is used largely by mitogenic signaling pathways
serine/threonine phosphorylation are used by other kinases (non-mitogenic signaling pathways)
the v-src lead to significant increase in concentration of phosphotyrosines
v-sis is very similar to the B chain of PDGF (platelet-derived growth factor)
Friend leukemia virus, env gene is actually encodes a EPO (erythropoietin) growth factor mimic
a cell that produces a growth factor it responds to leads to autocrine signaling of proliferation
common growth factors with autocrine signaling
TGF-α
stem cell factor (SCF)
insulin-like growth factor (IGF-1)
tumors with autocrine signaling are more likely to have in vivo results that match in vitro experiments, likely because they create their own growth factor environment
Transphosphorylation underlies the operations of receptor tyrosine kinases
low-molecular weight lipophilic ligands: steroids, retinoids, vitamin D
esterogen, progesterone, and androgen receptors
play key roles in development of breast, ovarian, and prostate carcinomas.
structure of nuclear receptor molecules
DNA-binding domain
recognizes hormone response elements (HREs) in the DNA
pair of hexanucleotides separated by a variable number of spacer sequences
hinge region
conserved ligand-binding domain
tamoxifen (a selective estrogen receptor modulator) works by binding nuclear receptor and causing conformation change so that the receptor cannot act as a co-activator.
Integrin receptors sense association between the cell and the extracellular matrix
diverse stromal cell types within tumors are all members of several mesenchymal cell lineages that generate both connective tissue and various types of immune cells
heterotypic signaling. signaling between dissimilar cell types
breast cancer: stromal cells release SDF-1/CXCL12 (chemokine) + HGF/SF (growth factor) → proliferation and survival of nearby epithelial cancer cells
anoikis. a form of apoptosis resulting from loss of anchorage to a solid substrate
angiogenic factors stimulate growth of capillaries (endothelial cells)
endothelial cells stimulate growth of pericytes and vascular smooth muscle cells (collectively called mural cells) by releasing PDGF and HB-EGF (heparin-binding EGF).
pericytes release VEGF and Ang-1 (angiopoietin-1)
ascitic tumors. tumors that are almost totally independent and can accumulate in various body fluids
The cells forming cancer cell lines develop without heterotypic interactions and deviate from the behavior of cells within human tumors
cancer cell lines are not always predictive of actual drug efficacy because stroma is not present; usually stromal cells can survive in tissue cultures
epithelial cells that can survive in vitro are selected for; not representative of the tumor's true population
PDX (patient derived xenografts) in immunocompromised mice is not necessarily predictive of eventual clinical responses to drugs
keratinocytes transfected with PDGF expression vector did not proliferation in vitro; when implanted into host mice, suddenly became tumorigenic because stromal cells had PDGF receptors
stromal cells then drove proliferation of PDGF-secreting keratinocytes
carcinoma-associated fibroblasts (CAFs) are much different than normal fibroblasts
mechanism still not entirely known
CAF myofibroblasts release SDF-1/CXCL12 (stroma-derived factor 1) that recruits circulating EPCs (endothelial precursor cells, come from bone marrow) + other myeloid cell types → leads to formation of neovasculature
neovasculature assumed to be formed by existing endothelial cells and EPCs
discovered that cancer cells can transdifferentiate into endothelial-like cells
Macrophages and myeloid cells play important roles in activating the tumor-associated stroma
plots of tumor volume vs. microvessel density suggest that angiogenesis acts as a rate-limiting determinant of tumor growth
angiogenesis burst usually stopped quickly
in healthy tissue, HIF-1 TF only assembles under hypoxic conditions
thombospondin-1 (Tsp-1) protein binds to CD36 receptor on endothelial cells and halts their proliferation
also leads endothelial cells to release FasL, a pro-apoptotic signaling protein
Fas receptor displayed on newly formed endothelial cells, and not on mature cells
TSP1 transcription is strongly induced by p53
Ras shuts down Tsp-1 production
other inhibitors exist
often, excision of tumor can also excise the inhibitors and increase production of wound-healing growth factors, which leads to development of already seeded metastases
inhibitors include angiostatin and endostatin
MMP antagonists called tissue inhibitors of metalloproteinases (TIMPs)
@research TIMP MoA still not elucidated
Anti-angiogenesis (AAG) therapies can be employed to treat cancer
local invasion depends invariably on the release of secreted proteases, such as MMP-2 and MMP-9
some cancer cells make their own proteases
some cancers co-opt stromal cells to release these enzymes
process of intravasation is less well studied
in breast cancers, triad of caricnoma cells, macrophages, and endothelial cells assembles to enable intravasation
macrophages release EGF that stimulate cancer cells to invade
cancers with high densities of these triads predicts eventual metastatic relapse
incomplete understanding of cancer cells in general circulation (circulating tumor cells, or CTCs)
cancer cells and cancer cell-platelet aggregates are much larger than capillary diameter ⟹ within minutes of entering venous circulation many cancer cells will lodge in the capillary beds of the lungs
these cells may then pinch off cytoplasm, freeing them to travel to other capillary beds (since mets exist beyond the lung)
cancers may also co-opt functions of macrophages to extravasate
Colonization represents the most complex and challenging step of the invasion-metastasis cascade
colonization. the growth from microscopic metastases into macroscopic metastases
detecting micromets
anti-cytokeratins for micromets in bone marrow and blood
antiEpCAM useful for micromets in lymph nodes
metastatic shower
the second wave of mets that arise after the first mets successfully colonize are much more deadly since they inherit the colonization phenotype
however, mets in terminal cancer patients are generally genetically diverse; do the second wave of mets diversify, or did many diverse mets undergo secondary showers?
THe epithelial-mesenchymal transition and associated loss of E-cadherin expression enable carcinoma cells to become invasive
expression of αvβ6 integrin is associated with the EMT
micromets undergo MET likely because new tumor microevironment has not yet been colonized and no EMT growth factors are present (yet)
mesenchymal cancer cells don't proliferate well; a MET is necessary for outgrowth of metastatic colonies
two possibilities for activating EMTs
by the stroma, providing heterotypic signals to the tumor (most evidence points to this cause)
by the tumor cells, who sense that they are not surrounded by other epithelial cells
stromal signals include
Wnts (both canonical and non-canonical)
TGFβ
TNFα
EGF (epidermal growth factor)
HGF (hepatocyte growth factor)
IGF1
PGE2 (prostaglandin E2)
genetic alterations in carcinoma cells further increase their responsiveness to these signals
TGFβ is typically anti-proliferative
most carcinomas lose the pRb pathway that executives TGFβ's cytostatic effects
TGFβ and canonical Wnt proteins are secreted by epithelial cells normally, but ability to activate EMT is blocked by secreted inhibitory proteins as well: BMPs, SFRP1, DKK1
Stromal cells contribute to the induction of invasiveness
ZEB1 and ZEB2 EMT-TFs determine whether cells stay in epithelial or mesenchymal state (a bistable switch)
ZEB1 + ZEB2 compete with miR-200 microRNAs
shutting down actions of EMT-TFs in in vivo tumors decreased number of metastases by ~85%; the remaining were due to tumor cells that did not have Twist (an EMT-TF) expression shut off
shutdown of Twist enabled faster growth of primary tumor; therefore prevention of metastases cannot be attributed to a simple cytostatic effect
Twist and Slug enable cells to resist apoptosis and anoikis
has not been shown systematically yet however
many factors involved in tumor microenvironment also are involved in metastasis; EMT may be a natural consequence of a growing tumor
summary
many malignant cell phenotypes are induced by heterotypic signals from the stroma (non-genetic changes)
because EMT signals come from the stroma, carcinoma cells may revert back to epithelial upon leaving the primary tumor
EMT-TFs often express together, leading to possibility that many malignant traits are acquired simultaneously
Extracellular proteases play key roles in invasiveness
matrix metalloproteinases (MMPs) allow cancer cells to excavate passageways through the ECM and remodel the nearby tissue environment
most MMPs are secreted by recruited stromal cells rather than the tumor cells themselves
macrophages
neutrophils
fibroblasts
MMPs can cleave
fibronectin
tenascin
laminin
collagens
proteoglycans
and liberate growth factors that were sequestered in the ECM
cancer cells express MT1-MMP (membrane type-1 MMP) on the surface of their plasma membranes
MT1-MMP can cleave pro-enzymes (e.g. pro-MMP-2) into active enzymes
cell nuclei are relatively rigid and cannot be compressed
MT1-MMP are concentrated at cancer cell invadopodia
the remodeling of ECM takes place continuously in mitotically active tissues
clinical trials targeting MMP were terminated because of effects on remodeling of cartilage and joints (led to high levels of joint stiffness and pain)
MMPs are negatively regulated by tissue inhibitors of metalloproteinases (TIMPs)
MMPs alone can allow cancer cells to progress through all stages of multi-step tumorigenesis
Small Ras-like GTPases control cellular processes such as adhesion, cell shape, and cell motility
the one distinction is that epigenetic factors play a role in cancer evolution, but epigenetic factors have not been show to play a role in species evolution
Tumor stem cells further complicate the Darwinian model of clonal succession and tumor progression