Sunday, July 29, 2012

Mitochondrial Dysfunction and Cancer

The Warburg Effect

Cancer is well known for the Warburg effect, which is where tumor cells preferentially convert glucose to lactate, but even in the presence of oxygen (called ‘aerobic glycolysis’).  Even though this produces much less energy, it is advantageous for the tumour cell a number of reasons:

  • It reduces reactive oxygen species (ROS) produced by the tumour cell because a lot of ROS are a by-product of oxidative phosphorylation [1]*
  • It enables tumour cells to not require mitochondria and shut them down, thereby reducing/eliminating pro-apoptosis signalling from mitochondria [1]
  • The intermediary metabolites of glycolysis can be used as substrate for rapid growth [1]
  • Lactate and H+ ions** promote inflammation and metastasis [1] [2]

Aerobic glycolysis is not unique to tumour cells, as it’s a feature of rapidly dividing cells (immune cells and stem cells), that also use aerobic glycolysis to produce the intermediary metabolites of glycolysis as substrate for growth

The Warburg effect suggests a defect in oxidative phosphorylation, and the mitochondria in tumour cells are “structurally and functionally abnormal and incapable of generating normal levels of energy” [3].

* The story of oxidants and antioxidants in cancer is interesting.  ROS can be harmful for cancer: the immune system and the p53 genes use ROS to trigger apoptosis and tumour cells use antioxidants to protect themselves [4]; and Metformin protects against cancer and among other things, it selectively increases oxidative stress and apoptosis in tumour cells by promoting oxidative phosphorylation [5].  However, metastasis can be driven by ROS and mitochondrial catalase (an antioxidant enzyme) and other antioxidants reduce tumour progression and metastasis [6].

** Hydrogen ions are acidic and this is why proponents of dietary acid-base theories of cancer argue that an acidic environment promotes cancer growth.  However, our pH is tightly regulated and it's the elevated glycolysis in tumour cells that drives the local acidic environment rather than some dietary factor.

The Metabolic Theory of Cancer

The gene theory of cancer is the one most people are familiar with and it suggests that nuclear gene mutations initiates cancer, mitochondrial dysfunction, the Warburg Effect and other hallmarks of cancer; and that carcinogens primarily cause DNA damage that leads to the gene mutations.  An alternative theory, the ‘metabolic theory of cancer’, suggests that mitochondrial dysfunction is the primary cause of the gene mutations, Warburg effect and other hallmarks of cancer.  The evidence supporting mitochondrial dysfunction as a cause of cancer is described in detail in the paper Cancer as a Metabolic Disease, which I’ll summarise here:

  • Increased ROS production, as a result of mitochondrial dysfunction, can impair genome stability, tumor suppressor gene function, and control over cell proliferation
  • In humans, mitochondrial dysfunction precedes cancer an aerobic glycolysis
  • Mitochondrial dysfunction causes cancer cells to rely on aerobic glycolysis/substrate level phosphorylation, and a prolonged reliance on substrate level phosphorylation “produces genome instability, cellular disorder, and increased entropy i.e., characteristics of cancer”
  • The retrograde response is initiated in response to impaired mitochondrial functioning and “a prolonged RTG response, however, would leave the nuclear genome vulnerable to instability and mutability”
  • Mitochondrial dysfunction can increase HIF-1a (a signal of hypoxia), which increases angiogenesis and promotes inflammation, via impaired calcium homeostasis, ROS, inflammation (NFκB) and defects in oxidative phosphorylation increased
  • Some viruses can cause mitochondrial dysfunction and infected cells that survive a viral infection (don’t initiate apoptosis) upregulate substrate level phosphorylation
  • Transplanting healthy mitochondria into cybrid tumour cells suppresses tumour formation and growth, while transplanting nuclei from tumour cells into healthy cells has no effect
  • “Respiration is required for the emergence and maintenance of differentiation, while loss of respiration leads to glycolysis, dedifferentiation, and unbridled proliferation”.  (Remember that rapidly dividing cells use aerobic glycolysis for energy too.  Cells can be made stem-like by inhibiting oxidative phosphorylation)
  • Cells that can use substrate level phosphorylation for their energy needs can evade mitochondrial-induced apoptosis, but these may initiate apoptosis when their glucose supply is insufficient
  • Impaired calcium homeostasis, which can result from mitochondrial dysfunction, “could contribute to abnormalities in chromosomal segregation during mitosis”

The paper then discusses the ‘macrophage hypothesis of metastasis’: how metastatic tumour cells may be the result of a fusion between a tumour cell and a macrophage (tumour associated macrophages or TAMs), as metastatic tumour cells share many features with macrophages, in that they can infiltrate other tissues and promote inflammation.  The fusion with the macrophage also allows the metastatic tumour cell to evade the immune system

* In addition, it’s quite logical that mitochondrial dysfunction causes many of the hallmarks of cancer rather than the other way around seeing as mitochondria and mtDNA are more vulnerable to carcinogens, which also happen to damage/disrupt mitochondria

Other Mitochondrial Related Mechanisms


Mitochondrial dysfunction is strongly associated with and tends to cause ER stress (and vice versa) [7], which increases PTP1B [8].  PTP1B is a protein that is elevated in some cancers, seems to promote tumor growth and loss/inhibition of PTP1B is protective (but not completely) in animal models of cancer [9].

Insulin Resistance

Mitochondrial dysfunction can cause insulin resistance and insulin resistance increases both insulin and glucose levels in the blood.  Insulin resistance is associates with a poorer prognosis in breast cancer [10] and T2D, but not T1D, is associated with several cancers, likely due to hyperinsulinemia [11].  This might be because insulin resistance enhances sensitivity to the mitogenic (cell division promoting) effects of insulin [10].  High insulin and IGF-1 increase cell survival through the PI3K/Akt/mTOR pathway [12].  mTOR promotes cell proliferation and survival, is often elevated in cancer and correlates with cancer progression, adverse prognosis and resistance to chemotherapy [10]

* There are other causes of insulin resistance besides mitochondrial dysfunction (see here), but I included it here because it fits with the metabolic theme of the post

Further Reading:
(1) Cancer as a metabolic disease
(2) What Is The Origin of Cancer?
(3) Mitochondrial Dysfunction

Sunday, July 22, 2012


DNA Mutations 

DNA mutations are fairly synonymous with cancer, but it’s not DNA mutations per se, rather it’s mutations in very specific DNA that encode for genes that are related to apoptosis (programmed cell death), cell survival, cell proliferation, the cell cycle, etc.  These genes are called oncogenes 

The p53 gene inhibits cell proliferation and is involved in cell-cycle checkpoints, apoptosis and maintaining the genome following DNA damage [1] [2]

  • Tumor suppression genes like p53 are absent in mice with cancer [3]
  • p53 mutations correlate with decreases in apoptosis [1]
  • Disrupting activators of p53 promotes tumour development [1]
  • p53 mutations are common in tumours and are associated with advanced tumor stage and poor patient prognosis [1]
  • p53 mutations correlate with short remissions and drug resistance following therapy in some cancers [1] 

Mutations in other genes can lead to cancer as well:

  • Mutations in the gene encoding protein tyrosine kinase RET can leave the gene stuck in the ‘on’ position, resulting in unregulated cell growth.  Mutations are an early event of papillary thyroid carcinoma and are necessary to the cancer to develop [4]
  • The Fas/CD95 receptor controls cell numbers in the immune system and inducing apoptosis in immune cells when numbers get too high.  Disruption in this pathway can lead to cancers in the immune system [1]
  • PTEN (in the AMPK pathway) is a tumour suppressor gene that also inhibits PI3K (in the mTOR pathway), which promotes cell growth, proliferation and survival.  Mutations in PTEN are common in tumours [1] 

Viral Infection 

Infections are estimated to contribute to 17.8% of all cancers worldwide (26.3% in developing and 7.7% in developed countries).  H. pylori (bacteria) 5.5%, human papilloma virus (HPV) 5.2%, hepatitis B and C viruses (HBV, HBC) 4.9%, Epstein-Barr virus (EBV) 1%, HIV and herpes virus 0.9% (~12% from virus).  The contribution of viral infections can be as high as 100% of cervical cancer from HPV and as low as 0.4% of liver cancers from liver flukes [5]. 

A virus capable of causing cancer is called an oncovirus.  Most viruses inject their RNA into cells, which makes the infected cell replicate the virus.  Oncoviruses tend to replicate in a different way by first making the infected cell reverse transcribe their RNA into DNA, which then gets incorporated in the infected cell’s genome and is used to replicate the virus.  A virus that replicates in this manner is called a retrovirus. 

Oncoviruses have developed a number of oncogenes/oncoproteins for their replication that promote cell division and proliferation and influence genes such as p53 and NFκB [3] [6].  Cancer can be an accidental side effect of these viral replication strategies.  These can include:

  • The E1A oncoprotein that influences the cell cycle by inducing G0 cells (cells not going through cell division) enter the S phase (DNA replication) because viruses need cells to be in the S phase for replication.  Forcing cells into the S phase without cell cycle checkpoints and apoptosis increases the chance of DNA mutations [1] [6]
  • The LMP1 oncoprotein (from the EBV), that mimics a growth factor receptor and a TNF receptor, and so can promote cell proliferation and NFκB expression [3]
  • A growth factor (from the poxvirus), similar to epidermal growth factor, is secrete by infected cells and stimulates proliferation in neighboring cells [3] 

Viral oncoproteins are like an alternative to DNA mutations in regards to initiating cancer.  There is often a long latent period between the initial viral infection and tumour development and developing a viral related cancer generally requires one to be immunocompromised [3]. 

Cancer Progression 

DNA is continually damaged, but there are DNA repair mechanisms to protect against DNA mutations.  The problem arises when the rate of DNA damage exceeds the rate of repair.  Replicating damaged DNA can lead to DNA mutations or apoptosis.  DNA mutations can result in underexpression (sometimes no expression) or overexpression of genes [7]. 

Genetic mutations/alterations that promote cell growth, proliferation and survival may lead to a cell that grows and divides rapidly resulting in a tumour.  Initially tumour cells (the cells that make up a tumour) are not metastatic/malignant.  To become metastatic a tumour cell needs to develop further DNA mutations to survive in the bloodstream (block apoptosis) and invade other tissues.  Once a tumour cell becomes metastatic and then invades other tissues every tumour cell of the new tumour has invasive and metastatic abilities [1] [8]. 

Cancer treatments like chemotherapy and radiation therapy work by overwhelming the capacity of the cell to repair DNA, resulting in cell death.  Tumour cells are most vulnerable because of their high rate of cell division*.  .  But some anti-apoptotic mutations can select for chemoresistant cells.  Also if chemoresistant cells survive after chemotherapy and the tumour grows back then all the tumour cells will be chemoresistant [1] 

* A common side effect is that other non-cancerous but rapidly dividing cells such as stem cells in the bone marrow are also highly affected. 

** When cells divide telomeres shorten.  Tumour cells up-regulate the enzyme telomerase (synthesises telomeres), which kind of makes tumour cells immortal.

Sunday, July 15, 2012

Low Carb Diets, Mitochondrial Dysfunction and Weight Loss: A Theory

See Mitochondrial Dysfunction for background information

Let’s say we have an obese, insulin resistant person who barely make it from breakfast to lunch without a donut and a soft drink.  Being fed up with their weight and health prospects they turn to the internet, discover paleo/low carb and begin eating that way.  Carb consumption reduces to ~1/3 of the previous amount and they aren’t eating in between meals.  The first week or so is hell, they feel hungry, lethargic and light headed.  They once again consult the internet, are told you’re sugar detoxing so they decide to stick with it.  A few days later the weight is coming off from all the right places, they have more energy and don’t feel hungry.  They have no need of sugar or snacks, in fact because the weight is coming off so quickly now they don’t feel hungry enough for lunch.  A few months later when all the weight has come off their doctor reports the insulin resistance has gone and the triglycerides have come down from 2 to 1 mmol/l. 

The above scenario plays out quite often, especially to begin with, and I think mitochondrial dysfunction and biogenesis is the best explanation for it. 

The elevated ROS and mtDNA mutations that are related to mitochondrial dysfunction can reduce the activity of the electron transport chain (which is seen with obesity and type 2 diabetes [1]).  This in turn decreases PGC-1α, and then also fatty acid oxidation and oxidative phosphorylation [2].  The ratio of glycolytic:oxidative enzymes in muscle is associated with insulin sensitivity.  People who are insulin resistant (and to a lesser extent, the insulin sensitive obese) have higher glycolytic and lower oxidative enzymes than normal [3]. 

So people with insulin resistance, type 2 diabetes or obesity rely more on glycolysis and fermentation (pyruvate to lactate) for ATP and less on aerobic respiration.  Lactate is converted to lactate acid to enter the bloodstream and high lactic acid production is one cause of the metabolic acidosis seen with type 2 diabetes [4] [5]. 

Now, our hypothetical person goes low carb: 

Low carb diets more strongly rely on aerobic respiration.  There’s not as much glucose for glycolysis + fermentation and no easy way to metabolise fats with impaired oxidative phosphorylation from mitochondrial dysfunction, hence why symptoms of low carb flu include hunger and fatigue.  There’s a deficiency of ATP even though there might be plenty of calories.

But people on low carb diets don’t suffer forever.  Ketogenic diets and non-ketogenic low carb diets increase AMPK, which stimulates mitochondrial biogenesis.  One study found that when people with obesity calorie restict people on a low carb diet (20:50:30) their AMPK increased after 5 days whereas AMPK levels from those on a high carb diet (20:20:60) didn't increase [6] (reported in a post by Mike Eades).

Low carb flu lasts roughly one week then all is well: 

When fish are exposed to cold temperatures to stimulate mitochondrial biogenesis the activity of citrate synthase (a marker of mitochondrial respiration) in the liver increases rapidly, doubling in value after 9 days, and then plateaus.  Muscle is more flexible, but takes more time.  There is an initial increase of about 50% in a few days and then a further doubling takes about 2 months [7].  Following muscle injury in rats markers of mitochondrial respiration plummeted, but increased rapidly between day 5 and 10 and citrate synthase returned to baseline after day 10 [8]. 

These time periods in which mitochondrial respiration increases and plateaus (including the increase in AMPK after 5 days) are similar to the duration of low carb flu*.  Then the reduced triglycerides:

Some people may attribute the reduced triglycerides to less de novo lipogenesis of glucose and fructose.  But that’s just as bad as saying high fasting glucose is due to blood glucose spikes from last night’s dinner.  Both are homeostatically regulated.  Mitochondrial dysfunction increases insulin resistance, which causes elevated triglycerides.  After mitochondrial numbers and function return to normal insulin resistance improves, leaving you with normal triglycerides.

If the main benefit of low carb diets is mitochondrial biogenesis then the good news is that you don't have to go very low carb as 30% of calories from carbohydrate is enough to increase AMPK and you get of the benefit very quickly (~1-2 weeks) and pretty much all the benefit within less than 2 months.  Although this isn't an excuse to return to poor eating habits after 2 months

* I know they are animal studies and the circumstances may not apply to low carb diets.  Remember it’s just a theory, but one I believe that explains low carb flu and sone of the benefits of low carb diets quite well.

Sunday, July 8, 2012



Serotonin is considered to be the ‘happiness neurotransmitter’, so low serotonin is thought to cause depression and the treatment for depression is to increase serotonin signalling with drugs (SSRIs like Prozac and Zoloft) .  There is some evidence to support this idea, but serotonin isn’t a happiness neurotransmitter and SSRIs take many weeks to have an effect, suggesting acute changes are not responsible, and don’t have much effect in healthy people.

People with depression have lower brain volumes, particularly in the hippocampus.  One function of serotonin is to promote neurogenesis (the growth of new neurons), which seems to be how SSRIs have a beneficial effect.  Other things that increase neurogenesis, such as exercise, zinc and long chain omega 3s are also therapeutic for depression.  But why do people with depression have lower brain volumes?

Chronic inflammation decreases neurogenesis and promotes neurodegeneration.  Depression is associated with chronic inflammation and immune suppression, which are also two characteristics of autoimmune diseases and cancer, and helps to explain the association between depression and other chronic diseases.  Pro-inflammatory cytokines can induce depression or depressive symptoms and anti-inflammatory cytokines and drugs have antidepressant effects.  SSRIs also increase an anti-inflammatory cytokine called IL-10.  Sources of inflammation such as LPS, alcohol, ROS/mitochondrial dysfunction and homocysteine can promote depression/hippocampal atrophy.

Stressful life events often precede depression.  Glucocorticoids (such as cortisol) inhibit neurogenesis and promote apoptosis, and stress triggers an inflammatory response.  Chronic stress and inflammation promote glucocorticoid resistance – a state of elevated glucocorticoids and chronic inflammation, since glucocorticoids fail to adequate suppress inflammation and HPA axis activity.  40-60% of people with depression have glucocorticoid resistance.

Further Reading:
(1) Evolutionary Psychiatry
(2) Regulation of adult neurogenesis by stress, sleep disruption, exercise and inflammation

Some Strategies for Depression

This is for informational purposes only and is not meant to diagnose or treat any medical condition.

Biological Causes of Depression

For neurogenesis, ensure adequate:

  • Long chain omega 3’s (fish, grass-fed beef/lamb/dairy, pastured eggs, algal DHA (fish is generally better than fish oil))
  • Exercise*
  • Zinc (most foods, particularly animal foods and especially shellfish)
  • Sleep, while optimising circadian rhythms (sleep at night, in the dark, without much blue light or food prior to sleeping, and with light and exercise during the day.  See here)

For inflammation, reduce/manage:

* I don’t know which kind of exercise is best for neurogenesis.  Aerobic exercise is mentioned more often.  My speculation is that exercise done outdoors where you’re learning and performing more complex movement patterns is probably more effective.  However, for many people, particularly those with depression, finding an exercise program that you feel motivated to do and can stick to long term is probably be more important than doing the ‘right’ exercise for neurogenesis.

Psychological Causes of Depression

There’s often either a focus on the biological side of depression/mental illness or on the psychological side, but both are important.  (Although to simplify things, all thoughts are a result of neurons releasing neurotransmitters to other neurons.  Yes, I’m a reductionist)

Cognitive behavioural therapy (CBT) is used by psychologists, etc to treat depression and other mental illness.  CBT basically involves changing one’s thoughts about themselves, others and the world to change behaviour; and changing behaviour (completing appropriate and scaled challenges) to change one’s thinking (‘I can do this’ or ‘that wasn’t so bad after all’)

CBT is important because in mental illnesses such as depression there is often irrational thinking* (usually self-critical in depression) that underlies the thoughts/feelings and affects behaviour.  The effect on behaviour (avoidance strategies, etc) then reinforces the negative thoughts, which causes a vicious cycle.  Some of the irrational thoughts are briefly mentioned in this three part series (see part 1, 2 and 3).  A good book for CBT is Change Your Thinking.  However, books shouldn’t necessarily replace counselling of some sort they have qualities a book can’t give you such as insight into your individual needs, a person to talk to and accountability

Sunday, July 1, 2012

Inflammation and Neurodegeneration


Loss of the volume in brain areas in depression is related to both increased neurodegeneration and decreased neurogenesis.  In physiological levels pro-inflammatory cytokines promote neurogenesis, but pathological levels inhibit neurogenesis and initiate apoptosis, both processes leading to the reduction in brain volume seen in depression [1] [2] 

Depression is associated with inflammation (elevated pro-inflammatory cytokines: IL-1, IL-6, TNF-α,, IFN-γ, NFκB and inflammatory markers: elevated haptoglobin, PGE2, CRP, adhesion molecules – in both cerebral spinal fluid and serum) and also immune suppression (low natural killer cell cytotoxic capacity (NKCC) and low lymphocyte proliferation).  The co-occurrence of inflammation and immune suppression is also seen in cancer and can explain the associations between depression, cancer and autoimmune diseases (such as MS and RA) [1] [3] 

Chronic inflammation and immune suppression may promote each other.  Chronic exposure to TNF-α reduces T cells receptors and suppresses the function of T cells and NK cells.  This in turn impairs lymphocyte proliferation and lowers IL-2.  IL-2 operates in a positive feedback loop whereby proliferating lymphocytes secrete IL-2 and IL-2 promotes proliferation.  T cells normally switch to secreting IL-10 if there’s evidence (high IL-2) of an adequate immune response.  Depression is associated with impaired lymphocyte proliferation, low IL-2 production and consequently vulnerability to infection, low IL-10 and unresolved systemic inflammation [3]. 

Evidence to support the role of inflammation in depression:

  • IL-6 can induce depression in animal models [1]
  • TNF-α promotes glutamate neurotoxicity and inhibits cell survival [1] and inhibiting TNF-α reduces depressive symptoms [1] [3]
  • IL-1β production correlated with an earlier age of onset and longer duration of illness with dysthymia (like depression but milder and longer)* [3]
  • SSRIs reduce IFN-γ and increase IL-10 and revert immunological dysfunction described in depression and anxiety disorders [1]
  • Aspirin and other anti-inflammatory drugs enhances the effect of SSRIs [1] [3]
  • An immudomodulatory vaccine to increase Treg cells had an antidepressant effect similar to SSRIs [4] 

Sources of chronic inflammation are related to depression: 

LPS increases IL-1 and TNF-α in the hippocampus and decreases BDNF and neurogenesis.  Animal models of depression can be induced by LPS and in humans LPS acutely increases depressive and anxious symptoms, psychomotor retardation and cognitive disturbances.  People with depression tend to have elevated IgM and IgA against gut bacteria [1] and elevated MPO [3] 

Ethanol (alcohol) is highly inflammatory to the CNS in a similar manner as LPS [5] and it decreases neurogenesis [6].  Alcohol abuse is associated with depression [7] and lower brain volume [8]. 

Inflammation tends to increase ROS and oxidative stress, which may lead to mitochondrial dysfunction, then a positive feedback loop of oxidative stress, inflammation and apoptosis.  MDA and markers of DNA damage are associated with depression [1] and impaired energy metabolism can lead to dysfunctional neural communication [9] 

Homocysteine increases oxidative stress and also reases glutamate and stimulates the glutamate receptor (NMDA), promoting excitotoxicity, which inhibits repair and increases apoptosis in hippocampal neurons.  Elevated homocysteine is associated with a higher rate (150%) of hippocampal atrophy [10] and depression [11].

* Early age of onset for depression is associated with lower NKCC [3] 

** There are higher levels of inflammation in the post-partum period which may partly explain post-partum depression [1] 

*** In the previous post I mentioned exercise, zinc and LCO3 increased neurogenesis and are therapeutic for depression.  But they also reduce inflammation as well [1] [12] 

**** People with depression tend to have a higher CD4+:CD8+ ratio (helper:cytotoxic), which is also seen in autoimmune diseases and cancer, may represent inflammation + immune suppression, and is a consequence of zinc deficiency [12] 

Stress and Glucocorticoid Resistance 

Depression can occur with medical illness (such as autoimmune diseases, inflammatory diseases, infectious diseases, tissue damage or destruction) and in these cases the origin of inflammation makes sense [1].  But what about trigger for otherwise healthy people. 

Stressful life events often precede depression.  In acute stress, elevated levels of glucocorticoids (cortisol and related hormones) are neurotoxic to the hippocampus and decrease serotonin synthesis, reducing the concentration of the neurogenic growth factors, which ultimately decreases neurogenesis.  The negative effects of stressors are present even after cortisol levels normalise [2] [13] [14].  This can progress into a positive feedback loop whereby less neurogenesis produces fewer newly formed hippocampal neurons, and because these newly formed hippocampal neurons inhibit HPA axis activity, now there are even more glucocorticoids and even less neurogenesis* [15]. 

Psychological stressors also activate an inflammatory response (elevated pro-inflammatory cytokines in the blood and brain) [1], and newly diagnosed patients of depression exhibit a higher amount of inflammation [16].  Key inflammatory cytokines such as NF-κB impair glucocorticoid function and increase the production of less active glucocorticoid receptors.  Also during depression or chronic stress the glucocorticoids can lose their anti-inflammatory effects and not suppress the HPA axis (negative feedback).  These two mechanisms promote glucocorticoid resistance (GCR)** [1] [3]. 

GCR is associated with chronic inflammation (because glucocorticoids are anti-inflammatory), elevated serum cortisol (because of an overactive HPA axis).  Depression is association with chronic inflammation and elevated serum cortisol and 40-60% of people with depression have GCR.  Although there seems to be two types of depression: a melancholic/psychotic type with hyperactive HPA axis and GCR, and a non-melancholic/psychotic type with elevated inflammation [1] [3] [16] 

* Rats without an adrenal gland are protected against hippocampal atrophy during stress [1] 

** My guess is that a sign of GCR is having a cold and experiencing symptoms during stressful events (exams, public speaking, etc).  In which case the glucocorticoids released from stress aren’t adequately suppressing the immune response that produces symptoms. 

Further Reading:
(1) RHR: Chronic Stress, Cortisol Resistance, and Modern Disease