The Hallmarks of Cancer are ten anti-cancer defense mechanisms that are hardwired into our cells, that must be breached by a cell on the path towards cancer. (Part A) To these, I have added two more (Part B)
The Classical Ten Hallmarks
In 2000, Robert Weinberg and Douglas Hanahan published a review article in the journal Cell titled “The Hallmarks of Cancer“. It was a seminal paper in every sense of the word; downloaded 20,000 times a year between 2004 and 2007, with over 15,000 citations in other research papers. In 2011, Weinberg and Hanahan updated their list by proposing four more new Hallmarks of Cancer, bringing up the list to Ten Hallmarks of Cancer.
Why is this paper so important? Cancer, as we know by now, is an incredibly complicated disease. A single tumor sample could have over a hundred different mutations; nearly one in every two hundred genes in the human genome. If two breast cancer specimens are compared, the set of mutated genes are far from identical. Every tumor is unique. Weinberg and Hanahan simplified this dauntingly complex disease to six underlying principles. The hugely complicated beast that is cancer, so diverse that even the same organ can have many different tumor types, was reduced to just six common traits that every single cancer shares, to facilitate that transformation from a normal cell to a cancer cell. It answers the ‘how does cancer happen’ question very elegantly, and we gain insight into all the different things that go wrong in a cancer cell.
However, the science is not accessible to the public because the molecular mechanisms described require a specialist knowledge in the field of cell biology. Over the coming weeks, I will go through each of these Hallmarks in detail. I will explain the normal processes that occur inside the cell and then explain what goes wrong with this process in a cancer cell.
Cancer is so prevalent, and is a topic that we hear about on a daily basis. Cancer is also deeply personal; we know people who are directly affected by cancer, either themselves or a loved one. Not everyone has the knowledge to understand what is going on, and it can be very scary to hear the jargon from a doctor or an oncologist. To many people, cancer is the big scary ‘C-word’. Trying to research it online is not usually helpful, as even Wikipedia has a confusingly large amount of jargon. This series will address that knowledge gap. It will demystify cancer, and answer the question ‘why does cancer happen?’
The First Hallmark of Cancer is defined as “Self-Sufficiency in Growth Signals”. What does this mean? Before I explain how growth signals are intimately involved in the development of cancer, it is necessary to define and understand what growth factors are, and explain how they control normal cellular behavior.
Growth factors are, simply put, substances that control the multiplication of cells. There are many different types of growth factors, but they all have several characteristics in common. They are all proteins, and present at very low concentrations in tissues but with a high biological activity. They are responsible for controlling essential functions within the cell; growth, specialization and survival. Growth factors also do not circulate in the blood stream; instead, they act locally in areas near the cells that produce them. The image below shows a growth factor known as Vascular Endothelial Growth Factor (VEGF).
Growth Factors are beautiful! This is a 3D schematic representation (also known as a ribbon diagram) of the structure of a growth factor known as Vascular Endothelial Growth Factor (VEGF). VEGF stimulates the development of new blood vessels, a process known as angiogenesis. Many large tumors secrete their own supply of VEGF in order to generate a supply line of oxygen-rich blood for the growing tumor to feed on. Image credit: Gizmag
It is impossible to talk about growth factors and cancer without going over some of the basics of cell signaling. We are multi-cellular animals, and as such, our cells need to communicate with each other, so they can act in a coordinated manner in response to the environment. The basis of this communication comes from a process known as cell signaling.
Growth Factors fit perfectly into Growth Factor Receptor Binding Sites. Two different types of Fibroblast Growth Factor (FGF1 and FGF2, left) shown bound to its specific receptor (center) and separate (right). Image credit: Alexander Plotnikov.
The behavior of a cell depends on its immediate surrounding environment, known as the microenvironment. The assortment of growth factors in this microenvironment is the most important aspect regulating the behavior of that cell. All growth factors exert their effects by binding to a receptor. Receptors are proteins found on the surface of a cell that receive such chemical signals from the outside of the cell. Each growth factor has it’s own receptor; think of it as a key (the growth factor) fitting into a lock (the receptor). Growth factor receptors tend to be ‘transmembrane molecules‘; this means that one end of the receptor ‘sticks out’ through the cell membrane into the microenvironment while the other end projects inside the cell. By spanning across the cell membrane, growth factor receptors are able to communicate signals from outside the cell (e.g. presence of growth factors in the microenvironment) to the inside of the cell. Revisiting the lock and key analogy, think of it as a key that fits into a lock that protrudes through the door-frame, instead of being flush against the door.
The binding of the growth factor to its specific receptor triggers a phosphorylation reaction inside the cell. Phosphorylation, or the addition of a phosphate group to a protein molecule, is an important step in cell signaling. This is because many proteins exist in an ‘on’ or ‘off’ state that can be switched by phosphorylation. Therefore, phosphorylation is a key step in regulating their activity. The enzymes that add phosphate groups to proteins are known as kinases; enzymes that remove phosphates are known as phosphatases. The exterior end of the receptor protein (the bit that sticks out of the cell) carries the growth factor binding site; the other end which projects inside the cell carries a kinase site. Binding of growth factor to the receptor binding site activates the kinase domain on the interior end of the receptor protein. This activated kinase, true to it’s name, then goes on to add phosphate groups to other proteins inside the cell, which then activate more proteins downstream, triggering a signaling cascade that finally ends with the activation of genes that bring about….you guessed it, cellular growth, specialization, or survival! The image below illustrates this process – I couldn’t find a decent one online so I made my own!
Mode of action of a typical Growth Factor. Growth Factor (red) binds to specific Growth Factor Receptor Binding Site (dark blue) on cell surface, which activates the kinase region (light blue). Activated kinase region now adds a phosphate group (yellow) to Protein 1 (blue) which activates it. Activated Protein 1 now adds a phosphate group to Protein 2 (green) further down the pathway, which activates it. Activated Protein 2 subsequently adds a phosphate group to Protein 3 (orange) which activates it. Activated Protein 3 moves through the nuclear membrane into the cell nucleus where it physically binds to the DNA and activates genes that control cell growth, specialization and survival. Image credit: Buddhini Samarasinghe
The description above is an extremely simplified version of what happens inside a cell; in reality, it is not so much a linear signaling pathway as it is an interwoven, intricate signaling web, with promiscuous proteins from many different pathways activating and repressing one another. The image below is not meant to frighten you (!) but rather to give you an idea how truly complex just one such signaling pathway, known as the MAPK/Erk pathway is.
A truly complex web of cell communication! These are some of the proteins we know that are involved in a single pathway known as the MAPK/Erk pathway. Signals from the outside of the cell go through this web of signaling, ultimately ending up with the activation of genes involved in growth, specialization and survival of the cell. Image credit: Cell Signaling Technology.
So now that we understand the basics of the molecular mechanisms behind cell signaling, what happens in a cancer cell that turns this orderly process so horribly awry? As you may have figured out, normal cells cannot divide without the ‘go ahead’ from growth factors. Even normal cells growing on a petri dish need growth factors supplied from animal serum to divide; if not they enter a dormant state and eventually die. Cancer cells on the other hand, do not need this ‘go ahead’. This liberation from being dependent on externally supplied growth factors removes a very critical checkpoint on the path towards cancer.
Breaching the Defenses
How do cancer cells bypass this checkpoint? There are three common strategies;
First, they can alter the level of growth signal itself. Normally, growth factors are made by one type of cell in order to act on another type of cell. However many cancer cells acquire the ability to synthesize and secrete their own growth factors, stimulating others of their kind, which creates a feedback loop in which more cancer cells divide under the influence of growth factor to synthesize more growth factor and so on. If the key to the lock is typically provided by a caretaker, then this means having your own DIY key-cutting machine, so that dependence on a locksmith is eliminated.
Second, the cancer cell can tweak the growth factor receptor itself, so that a larger-than-normal number of these receptors are present on the surface of the cancer cell. This means that the cancer cell becomes hyper-responsive to ambient levels of growth factor that would normally not be enough to trigger cell division. Additionally in some tumors, the growth factor receptor is structurally altered, sometimes lacking the regulatory regions, which results in the switch being permanently stuck at ‘on’. There is no need for a key, the lock opens without one.
Finally, there are alterations further downstream of the signaling pathway, so that the requirement for growth factor and receptor are both bypassed. For example, one of the key downstream components of the growth factor signaling pathway is a protein known as Ras. Mutant Ras is permanently ‘switched on’. Mutant Ras is the most common gene in human cancer; 25% of all human tumors, and up to 90% of certain types of cancer such as pancreatic cancer have mutant Ras. Why bother with trying to unlock a door if the walls don’t exist?
It is worth remembering that cancer cells cannot do what they do in isolation. The apparently normal bystanders, such as cells of the nearby blood vessels and connective tissue must also play key roles in driving cancer cell growth. In normal tissues, cells are instructed to grow by their neighbors; this is true of the tumor microenvironment as well. A tumor is not only made of cancer cells. Tumors should be regarded as complex tissues in which the mutant cancer cells have co-opted and subverted normal neighbor cells by inducing them to release growth factors as well.
With these three strategies for achieving self-sufficiency in growth signals, cancer cells can successfully breach one of the ten anti-cancer defenses hardwired into our normal cells. The result is cells that are capable of growing uncontrollably, unstoppably and pathologically – in other words, cancer cells.
The Six Hallmarks of Cancer, as published in 2000 by Weinberg and Hanahan. Image credit: Weinberg and Hanahan.
How can we use this knowledge about cancer to fight back? This is where the magic words ‘targeted therapies’ come in. Since cancer cells hijack normal growth factor response pathways to become self sufficient, it is logical for us to target these errant pathways specifically. If a growth factor receptor is stuck on ‘always on’, for example as a perpetually active kinase, then finding a specific inhibitor to stop the activity of this kinase would starve the cancer cell of the signal it is dependent on for uncontrolled growth.
Two such drugs, discovered in the 1990s utilize this principle. Gleevec, used as treatment for chronic myelogenous leukemia and Herceptin, for the treatment of breast cancer, both inhibit specific components of growth factor response pathways to starve the cancer of this signal. Earlier in this article I mentioned the Ras protein, which is frequently mutated in cancers; work is currently underway to find small molecules that are capable of inhibiting Ras. An exciting era of targeted cancer therapies lie ahead of us, because we have a deeper understanding of how cancer happens.
This is the first of a series of articles that I will be writing about the Hallmarks of Cancer. This series is based on two review papers published by Robert Weinberg and Douglas Hanahan in the journal Cell. Both papers are Open Access, and you can download them here and here. The authors simplified cancer down to ten underlying principles, shared by every single cancer cell. Each of these ten changes in cell biology represent the successful breaching of an anticancer defense mechanism that is hardwired into our normal cells and tissues. These multiple defenses are the reason why we don’t all get cancer within hours of being born, or indeed being conceived!
The Second Hallmark of Cancer is defined as “Insensitivity to Antigrowth Signals”. Before I explain how failure to respond to antigrowth signals is closely involved in the development of cancer, it is useful to define and understand how cell division works. Cancer is, after all, the uncontrolled division of the cell; so we first need to understand how normal cell division is controlled through the cell cycle.
The Cell Cycle
Think of the Cell Cycle as the control system of a washing machine. A washing machine passes through several stages in a wash cycle; soaking the clothes, adding detergent at the correct time, rinsing the clothes for the appropriate duration to remove the detergent, adding the fabric softener at the correct time, a final rinse and then spinning the clothes to remove as much water as possible. In much the same way, the cell cycle is a series of tightly regulated events inside a cell that lead to its division into two daughter cells. In between these start and end states, the DNA inside the parent cell needs to double and then be divided equally between the two daughter cells. Intricate feedback loops of responsive proteins trigger events in the cell cycle, guiding the cell through checkpoints that lie between every stage. These checkpoints are important because they act as safety valves, ensuring that an errant, incorrectly dividing cell with damaged DNA is promptly whisked away and destroyed rather than allowed to continue its life.
The Four Stages
The cell cycle has four stages.
The Cell Cycle, showing G1, S, G2 and M stages. Cell growth occurs during G1 and G2, while DNA is synthesized in S stage. Cell division occurs during M stage (mitosis). G0 indicates quiescent stage where the cell exits the cell cycle but can re-enter if the signals from the environment are appropriate. Red dots indicate important Cell Cycle Checkpoints G1/S and G2/M. Image credit: Buddhini Samarasinghe
Gap 1 (G 1) – during this stage the cell is actively growing in size, and preparing the required components for DNA synthesis. During and at the end of this stage, the cell monitors its surrounding environment (the microenvironment) to make sure there aren’t any ‘stop!!’ signals. This is known as the G1/S Checkpoint.
Synthesis (S) – DNA replication takes place during this stage. If there is damaged DNA in the cell, it should not be replicated, and the G1/S checkpoint ensures this. The G1/S Checkpoint is extremely important to prevent the replication of damaged DNA.
Gap 2 (G2) – The cell continues to grow in the G2 stage following DNA replication. At this stage, it is vital to recognize any damaged DNA before proceeding with cell division. This constitutes the G2/M Checkpoint. If the cell detects damage to its DNA, it will not proceed to the fourth and final stage of the cell cycle. The first G1/S Checkpoint ensures that the template DNA prior to replication is undamaged, and the G2/M Checkpoint ensures that the newly replicated DNA is error-free before proceeding with cell division.
Mitosis (M) – This stage represents the culmination of the cell cycle. Cell growth stops and the cell divides into two equal daughter cells.
Some cells enter a G0 stage known as quiescence which can be considered a place of rest. A cell in G0 has exited the cell cycle, and is neither dividing nor preparing to divide; however the cell is still alive and actively metabolizing, it has simply stopped dividing. The cell may re-emerge from this quiescent stage back into the cell cycle, given the right signals from its microenvironment. Some cells permanently exit the cell cycle, moving to a post-mitotic state. There is no coming back from this path; it is usually associated with mature cells that have differentiated, i.e. taken their adult form. Therefore a cell, at the basic level, has three choices facing it; continue to grow and divide by staying in the cell cycle, take a temporary break by entering G0, or permanently exit the cell cycle into the post-mitotic state.
Controlling the Cell Cycle: Cyclins and CDKs
Two key classes of regulatory proteins control the checkpoints within the cell cycle. These proteins are known as Cyclins and Cyclin Dependent Kinases (CDKs). Cyclins and CDKs cannot work without each other; they need to ‘team up’. Cyclins provide the regulation for the team, and CDKs are the catalyst. CDKs are active only with their specific partner cyclins; this is why they are known as cyclin dependent kinases. In the previous Hallmark of Cancer article, I explained that many proteins exist in an ‘active’ or ‘inactive’ state that can be switched by phosphorylation (i.e. adding a phosphate group to the protein), and that kinases are enzymes that add such phosphate groups to proteins.
A 3D schematic representation of Cyclin A protein. This protein partners with CDK2, for S and G2 stages of the Cell Cycle. Image credit: Wikimedia Commons.
Each stage of the cell cycle has a specific cyclin/CDK pair that behaves as a single unit, called a complex. Each stage’s cyclin/CDK complex is required to allow the cell to commit to and proceed to the next stage of the cell cycle. When activated by a cyclin, a CDK is able to phosphorylate key proteins involved in the networks of chemical reactions in the cell, which in turn are activated and proceed to manufacture the cyclin required for proceeding into the next stage of the cell cycle. It is a beautifully elegant system of control, and it works to ensure that cells grow and divide when they’re supposed to, and remain quiescent at all other times.
Antigrowth signals are proteins, exactly like the growth factors that I mentioned in the previous Hallmarks of Cancer article, except that they inhibit growth rather than promote growth. Antigrowth factors can therefore block cell growth by the two mechanisms mentioned above; cellular quiescence through G0 and the post-mitotic state. A cancer cell must therefore evade these signals if it is to continue dividing uncontrollably. During the G1/S Checkpoint, cells monitor their microenvironment to choose whether to continue in the cell cycle, enter G0 or enter a post-mitotic state. At the molecular level, nearly all antigrowth signals are funneled through a protein known as the Retinoblastoma protein. Therefore, Retinoblastoma is classified as a tumor suppressor protein. Indeed, it was the very first tumor suppressor protein to be discovered in 1971 through an elegant process of deduction and statistical analysis of rare eye cancers.
If a cell detects that it has damaged DNA at the G1/S checkpoint, this damaged DNA should not be replicated. Therefore, at the G1/S checkpoint, prior to entry into S stage, the brakes are slammed and everything comes to a screeching halt. In this analogy, the Retinoblastoma protein acts as the brake. Retinoblastoma protein is active when it is not phosphorylated, meaning phosphorylation inactivates Retinoblastoma. A primary function of Retinoblastoma is to bind to and inactivate E2F transcription factors. These are extremely important proteins that bind to DNA and activate genes which…you guessed it, control the cell cycle and DNA replication, including the Cyclins and CDKs specific to G1 and S phases of the cell cycle. In other words, E2F transcription factors are controlled by their interaction with the Retinoblastoma protein, which acts as the brake in the cell cycle progression to the S Stage. During G1, Retinoblastoma protein binds to E2F and blocks the production of S stage Cyclins and CDKs. When cells are stimulated to divide by external signals in their microenvironment, active CDKs specific to the G1 stage accumulate and phosphorylate Retinoblastoma. The Retinoblastoma protein, now inactivated, moves away from E2F, allowing the cell cycle to proceed. The brake is disengaged (see diagram below).
Active Rb (Retinoblastoma protein) inactivates E2F protein. CDKs (cyclin-dependent kinases) from G1 stage of the cell cycle phosphorylate Rb, thereby inactivating it. E2F is now active and free to produce genes required for progression of cell cycle into S stage. The brake, from Rb, is disengaged. Image credit: Buddhini Samarasinghe
Without E2F transcription factors, cell division grinds to a halt, and the phosphorylation status of Retinoblastoma controls the activity of these E2F transcription factors! Think about it for a moment: the addition or removal of a tiny molecule of phosphate to a single Retinoblastoma protein within the cell is responsible for whether a cell divides or not!
What influences the activation (i.e. phosphorylation status) of the Retinoblastoma protein? Antigrowth signals from the surrounding microenvironment of the cell. Given how important a role Retinoblastoma plays in the control of cell division, it comes as no surprise that cancer cells have found a myriad ways to bypass these antigrowth signals. Disruption of the retinoblastoma pathway liberates E2F transcription factors to promote cell division, and thus cells become insensitive to antigrowth signals that normally control this process.
So what are these antigrowth signals? Probably the best documented is the signaling molecule TGF-beta. Recall from the previous paragraph that phosphorylated Retinoblastoma is inactive, and this ‘disengages’ the brakes on the cell cycle. TGF-beta has many different mechanisms for preventing the phosphorylation of Retinoblastoma (i.e. for preventing the disengagement of the brakes). Therefore, the presence of TGF-beta blocks the advancement of the cell cycle. Unsurprisingly, many cancers target this pathway for disruption. Some cancer cells stop responding to TGF-beta altogether, by producing less TGF-beta receptors on their cell surfaces. Other cancers produce mutated receptors that do not respond to the presence of TGF-beta. Some cancers get rid of downstream proteins that respond to TGF-beta. In many late-stage tumors, instead of functioning as an antigrowth signal, TGF-beta activates a cellular program known as Epithelial-to-Mesenchymal-Transition (EMT), which gives cancer cells stem-cell like abilities that is really bad news to a cancer patient. Finally, Retinoblastoma protein itself, the end target of this pathway, can be lost through mutation of its gene. Interestingly, certain cancer-promoting proteins (oncoproteins) can block the function of Retinoblastoma as well. For example, the human papillomavirus produces a protein known as E7, which binds to and inactivates Retinoblastoma.
The end result is that antigrowth signals, funneled through Retinoblastoma protein into the cell cycle, are, in one way or the other, disrupted in a majority of human cancers. Cancer cells with defects in the Retinoblastoma pathway are missing the services of a critical ‘gatekeeper’ of cell cycle progression; the absence of the Retinoblastoma gatekeeper permits persistent cell division. Therefore, the insensitivity to antigrowth signals represents a vital breach of an anti-cancer defense mechanism that is hardwired into our cells.
The Third Hallmark of Cancer is defined as “Evading Apoptosis”. Apoptosis is the opposite of cell growth; it is cell death. To divide and grow uncontrollably, a cancer cell not only has to hijack normal cellular growth pathways, but also evade cellular death pathways. Indeed, this acquired resistance to apoptosis is characteristic of all types of cancer. But before I explain how cancer cells do this, we need to understand how the process of cellular death occurs in a normal cell.
The apoptotic program is hardwired into every single cell in our body. It is like a cyanide capsule, swiftly delivering death if the circumstances require cellular suicide. If a cell detects that it has damaged DNA, it can activate apoptosis to remove itself from the population. Apoptosis, or cellular suicide, is an entirely normal function of cells. The same apoptotic program is activated when a tadpole changes into a frog; the cells in the tail die through apoptosis, and the tail disappears. The same is true for the webbing between our fingers in our early embryonic development. Apoptosis is an extremely tidy process; cellular membranes are disrupted, the chromosomes are degraded, the DNA breaks up into fragments, and the dying, shrinking cell is swallowed up by a neighboring cell or a patrolling immune cell, leaving no trace of the cellular suicide behind.
Regulators and Effectors
So how does apoptosis work at the molecular level? The apoptotic machinery can be divided into two broad categories; regulators and effectors. The regulators are responsible for monitoring the interior and exterior environment of the cell for conditions of abnormality in order to decide whether that cell should live or die. The possible abnormalities include DNA damage, signaling imbalance caused by the activation of cancer causing genes (oncogenes), lack of an oxygen supply or insufficient growth factors.
The Bcl-2 family of proteins can be divided into pro-apoptotic proteins and anti-apoptotic proteins, i.e. proteins that promote apoptosis and proteins that inhibit apoptosis. *For an explanation of the ‘interesting’ names for the pro-apoptotic proteins, see note at the end of this article! Image credit: Buddhini Samarasinghe
Apoptosis can therefore occur either through an intrinsic pathway, in which signals from within the cell activate the process, or through an extrinsic pathway where death signals from outside the cell are received and processed by the cell to activate apoptosis. It is thought that the intrinsic apoptotic pathway is more important in cancer prevention than the extrinsic pathway. Given how our cells carry machinery to destroy themselves with the precision of an executioner, it comes as no surprise that the process is tightly regulated.
The primary regulators of apoptosis are proteins belonging to a group known as the Bcl-2 family. These proteins can either be pro-apoptotic or anti-apoptotic; Bcl-2, Bcl-XL, Bcl-W, Mcl-1 and A1 proteins function as anti-apoptotic proteins that inhibit apoptosis, while Bax, Bad, Bid, Bok, Bik and Bak (I swear these names are not made up!*) are pro-apoptotic proteins that trigger apoptosis when activated. The anti-apoptotic proteins bind to and inactivate the pro-apoptotic proteins in a healthy cell that does not need to die. Apoptosis regulators also include death receptors on the cell surface which bind to death signaling molecules, as part of the extrinsic apoptotic pathway. This is similar to the way growth factors bind to and activate growth factor receptors, as I described previously, and this binding triggers the effectors of apoptosis.
The Suicide Machines
A brief overview of the extrinsic and intrinsic apoptotic pathways. In the extrinsic pathway, death signals from the surrounding environment of the cell bind to death receptors on the surface of the cell membrane. This causes the conversion of inactive pro-caspase-8 into active caspase-8. Caspase-8 then goes on to activate caspase-3, which begins the caspase cascade that leads to apoptosis. In the intrinsic pathway, typically initiated by DNA damage, P53 is activated. P53 then activates the pro-apoptotic protein Bax, which initiates the release of cytochrome c from the mitochondria. Apaf-1 and the released cytochrome c combine to form a complex known as the apoptosome. The apoptosome causes the conversion of inactive pro-caspase-9 into active caspase-9. Caspase-9 then goes on to activate caspase-3 in a similar manner to the extrinsic pathway. Activated caspase-3 then leads to the caspase cascade, resulting in apoptosis. Image credit: Buddhini Samarasinghe
Where is Mission Control for apoptosis? Many of the apoptotic signaling pathways converge at the mitochondria. Mitochondria are tiny organelles floating within a cell, and function as the cell’s energy factories. They contain a signaling molecule known as cytochrome c, which is bound to the mitochondrial membrane. In response to pro-apoptotic signals (from pro-apoptotic proteins such as Bax), cytochrome c is released into the cell by the mitochondria, and they bind to a protein known as Apaf-1. This results in the formation of the apoptosome. The apoptosome is an extremely beautiful structure resembling a wheel with seven spokes. Once formed, the apoptosome goes on to activate a group of proteins known as caspases.
All our cells contain the seeds of their own destruction; these come in the form of caspases. Caspases can be thought of as cellular executioners. They are proteins that degrade other proteins in our cells. Active caspases can wreak havoc within a cell and are therefore extremely dangerous, so they are produced in an inactive form by the cell (known as pro-caspases), like an executioner’s blade that is sheathed. Upon detecting an increase in the amount of cytochrome c, released from the mitochondria, the blades are unsheathed. There are 13 such caspase genes identified in the human genome so far. Two of the caspase proteins act as ‘gatekeeper’ caspases: caspase-8 and caspase-9. They are initiator caspases that, when activated by cytochrome c release, go on to activate the other caspases in a cascade of irreversible cellular protein degradation.
P53: The Guardian of the Genome
P53 is a vital protein; more than half of all cancers have inactive P53. When activated by either DNA damage or chromosome abnormalities, P53 can halt the cell cycle and initiate DNA repair. If repair is successful, the cell cycle is restarted, and cellular and genomic stability is re-established. If repair is not an option because the damage is too great, then P53 can facilitate apoptosis, leading to the death and elimination of the abnormal cell. Image credit: Buddhini Samarasinghe
How do cells detect the necessary conditions for triggering apoptosis? In the previous Hallmark of Cancer article, I explained the fundamental role of the Retinoblastoma protein in controlling cell division. Retinoblastoma, remember, is a vitally important brake on cell division. Damage to a cell’s Retinoblastoma gene releases this brake, leading to uncontrolled cell growth.
Similarly, P53 is an extremely important protein, dubbed ‘The Guardian of the Genome’. Amongst its many functions, it is responsible for detecting DNA damage, chromosome abnormalities and arresting the cell cycle to initiate repair; if repair is not possible then apoptosis is induced. P53 induces apoptosis by increasing the production of the pro-apoptotic protein Bax. Bax stimulates the mitochondria to release cytochrome c, which activates the caspase cascade which ultimately results in cell suicide. P53 is vital for maintaining the integrity of our genome at the most fundamental level.
So how do cancer cells escape death? The most common method is the loss of the apoptosis gatekeeper, the protein P53. More than half of all types of human cancers have a mutated or missing gene for p53, resulting in a damaged or missing P53 protein. As an alternative to achieving the loss of P53, cancer cells can compromise the activity of P53 by increasing the inhibitors of P53, or silencing the activators of P53. Previously I explained how the human papillomavirus produces a protein known as E7, which binds to and inactivates Retinoblastoma.
Similarly, another protein, E6, also produced by the human papilloma virus, binds to and inactivates P53. These two cancer-causing proteins, E6 and E7 (oncoproteins), therefore disable two vital gatekeepers, Retinoblastoma and P53, that control both cell division and cell death; the result is repeated uncontrolled cell division that manifests itself in warts, with strong associations with the development of cancer. Cancer cells can also produce excessive amounts of anti-apoptotic proteins such as Bcl-2, Bcl-XL etc. They can produce less of the pro-apoptotic proteins such as Bax and Bak. They can short-circuit the extrinsic death receptor apoptotic pathway.
It comes as no surprise that highly aggressive cancers often have both Retinoblastoma and P53 mutations. As a result, these rapidly growing tumors have extremely low levels of apoptosis and extremely high levels of cell division.
Like de-rigging the Sword of Damocles, cancer cells can inactivate the machineries of death; and the evasion of apoptosis by cancer cells therefore represents a key breach of an extremely important anti-cancer defense mechanism.
The Fourth Hallmark of Cancer is defined as “Limitless Replicative Potential”.
The first three Hallmarks of Cancer explain how independence from growth signals, insensitivity to antigrowth signals and resistance to apoptosis lead to the uncoupling of a cell’s growth program from the signals in its environment. However, cancer is not just a result of disrupted signaling. Our cells carry an in-built, autonomous program that limits their multiplication, even in the face of disrupted signals from their environment. For a single cancer cell to develop into a visible tumor, this program must also be disrupted.
The Cellular Timekeeper
Microscope image of normal cells (upper) compared to senescent cells (lower). Blue-green areas indicate expression of marker associated with senescence. Image credit: Wikimedia Commons.
Normal cells are hard wired with a timer that keeps track of their age; the number of times they divide and grow. Most cells in our body can only undergo a limited number of successive cell growth-and-division cycles. This limit is named the Hayflick Limit after its discoverer, Leonard Hayflick. After undergoing between 40 and 60 divisions, cell growth slows down and eventually stops altogether. This state is known as senescence, and it is irreversible; although the cell does not grow or divide, it remains alive. When normal human cells are cultured in the lab in a petri dish, we can observe this phenomenon, where cells grow and divide a fixed number of times and then enter senescence. Some cells are able to make it past the senescence barrier and continue dividing; however these cells then undergo a second phenomenon known as crisis, during which the ends of their chromosomes fuse with each other, and the cells all die on a massive scale via apoptosis.
How does a cell count its divisions? How does it ‘know’ when to stop? The answer is telomeres. Telomeres are regions of repetitive DNA, capping and protecting the ends of the chromosome from degrading or from fusing with another chromosome. Without telomeres, each time a cell divides our genomes would progressively lose information because the chromosomes would get shorter and shorter. A telomere is like the heat-shield of a spacecraft; it protects the actual spacecraft and absorbs the damage instead. With every replication of a cell, about 50-100 nucleotides of telomeric DNA is lost. This progressive loss eventually causes the telomeres to lose their ability to protect the ends of chromosomal DNA. Left unprotected, these exposed ends become damaged. The DNA damage response is activated, leading to growth arrest; senescence. When chromosome ends fuse with each other, this irreversible damage results in the activation of apoptosis; the cell enters crisis, and dies.
The End Replication Problem
Why do the ends of chromosomes shorten? To understand this, first we need to go over the basic mechanisms of DNA replication. A cell must replicate its DNA before it divides. DNA is a double-stranded molecule, and each strand of the original DNA molecule serves as a template for the production of a complementary strand. The two strands have a directionality; the two ends of a single strand are known as the 3′ end and the 5′ end. The numbers refer to the position of the carbon atom in the deoxyribose molecule at the end of the strand to which the next phosphate molecule in the DNA chain attaches. For a quick introduction into the structure of DNA, check out this YouTube video:
The chemical structure of DNA. The 5’ and 3’ ends are shown lying antiparallel to each other. Nitrogenous bases adenine (green), thymine (purple), guanine (blue) and cytosine (red). These bases are fixed to the phosphate-deoxyribose backbone. Image credit: Wikimedia Commons.
This directionality matters because DNA replication takes place under the direction of an enzyme known as DNA polymerase. This enzyme faithfully copies our genetic code letter by letter. However, DNA polymerase can only work in one direction; the 5′ to 3′ direction. Therefore, although DNA replication is straightforward for one of the DNA strands (the 5′ to 3′ strand), the other strand (the 3′ to 5′ strand) is more complicated. This strand is replicated in short fragments instead of one continuous strand of DNA. Replication begins with an enzyme known as primase, which reads the template DNA and initiates the synthesis of very short complementary RNA fragments. DNA polymerase is now able to use these RNA fragments as a starting point to synthesize complementary fragments of DNA in between the RNA fragments. The RNA fragments are then removed and replaced with DNA, and the fragments of DNA are joined together by another enzyme, DNA ligase. This solves the directionality problem of DNA polymerase, but now we run into another problem, known as the end replication problem. This is because although the 5′ to 3′ strand can be replicated to the very end, the 3′ to 5′ strand cannot; DNA polymerase enzyme requires RNA fragments to begin replication, and there is nothing for such a fragment to attach to at the very end of the DNA strand. Therefore, with every round of replication, a small fragment of the DNA would be lost from the end of the chromosome, since it cannot be replicated. The cell solves this problem by having telomeres at the ends of chromosomes, where they prevent the loss of valuable genetic information by acting as a disposable buffer. Over time, with each successive round of DNA replication, the telomeric DNA shortens until finally there is no more disposable buffer, at which point the cell stops dividing and enters senescence.
When cells are grown in petri dishes in the lab, repeated cycles of cell division lead first to senescence and then, for those cells that make it past this barrier, to crisis phase. Fascinatingly, in very rare instances (about 1 in 10X7) a cell can emerge from this ordeal exhibiting unlimited replicative potential. This cell is now said to be immortalized, and it is a trait that most cancer cells growing in labs exhibit, including the famous HeLa cells.
1. Double stranded DNA is made up of two DNA strands lying antiparallel to each other, with their 5′ and 3′ ends highlighted in red. 2. When DNA replicates, the two strands unwind. Replication of the Leading Strand is very straightforward, with the DNA polymerase enzyme copying from the purple template strand to form the new DNA strand, shown in orange. This new strand is formed in the 5′ to 3′ direction, since DNA polymerase only works in that direction. 3. For the synthesis of the Lagging Strand, the enzyme RNA Primase creates short complementary RNA fragments, shown in green. 4. DNA polymerase is now able to use these short RNA fragments as a starting point to synthesize DNA between the RNA gaps, shown in orange. The enzyme still works from 5’ to 3’, in a ‘backstitching’ manner. 5. The RNA primers are now removed, and the gaps are filled in by DNA polymerase to form a complete replicate of the lagging strand. 6. However, the extreme end of the DNA strand cannot be replicated because there is nowhere for the DNA polymerase enzyme to attach to. 7. The presence of a telomere, shown in pink, allows the DNA to be protected as the telomere gets shortened instead. Image credit: Buddhini Samarasinghe
Cancer cells have therefore not only uncoupled their growth program from the signals in their environment, they have also breached the in-built replication limit hard wired into the cell. How do they achieve this? All cancer cells maintain their telomeres. 90% of them do so by increasing the production of an enzyme known as telomerase. As its name implies, telomerase functions by adding telomeric DNA to the ends of chromosomes. Most normal cells do not divide frequently, and therefore are not in any danger of shortened telomeres; these cells can get away with having low telomerase activity. Indeed, most cells apart from fetal cells and stem cells show low telomerase activity levels. Many cancer causing proteins (oncoproteins) are able to activate the production of telomerase, while many cancer preventing proteins (tumor suppressors) such as P53 (see previous Hallmark) produce factors that inhibit the production of telomerase. The other 10% of cancers rely upon the activation of a pathway known as the Alternative Lengthening of Telomeres (ALT), which swaps around telomeres to lengthen them.
Human chromosomes, stained blue, with marker for telomere region in red. Image credit: Asako J. Nakamura (Wikimedia Commons).
Intriguingly, telomere length is also affected by oxidative stress. Oxidative stress, in the form of free radicals, damages DNA. This damage is usually repaired by DNA repair mechanisms, but these mechanisms are less effective on telomeric DNA than elsewhere on the chromosome. Telomeres are therefore highly susceptible to oxidative stress. It also explains the rate of telomere shortening observed; estimated loss per cell division because of the end replication problem has been estimated at 20 base pairs of DNA, yet the observed loss is much larger, between 50-100 base pairs of DNA. This difference shows that oxidative stress has a far greater impact on telomere length than the nuances of DNA replication. It is possible that cell senescence induced by telomere loss is therefore a stress response, evolved to block the growth and replication of cells that have been exposed to a high risk of DNA damage.
The defining feature of a cancer cell is its ability to divide endlessly, without exhaustion, generation after generation. They achieve this by destroying the cellular timekeeper, the telomere. Immortality comes at a price; the accumulation of damaging mutations only increases with time, which is why cancer is primarily a disease of an aging population. The immortalization of cancer cells by telomere maintenance therefore represents an essential step in tumor progression.