How antioxidants can accelerate cancers

[Ecoclimber]

How antioxidants can accelerate cancers, and why they don't protect against them

Date:
July 10, 2014
Source:
Cold Spring Harbor Laboratory
Summary:
Two cancer researchers have proposed why antioxidant supplements might not be working to reduce cancer development, and why they may actually do more harm than good. Their insights are based on recent advances in the understanding of the system in our cells that establishes a natural balance between oxidizing and anti-oxidizing compounds. These compounds are involved in so-called redox (reduction and oxidation) reactions essential to cellular chemistry. Drs. Tuveson and Chandel explain why eating foods rich in antioxidants, as well as taking antioxidant supplements, can actually promote cancer, rather than fight or prevent it, as conventional wisdom suggests.

For decades, health-conscious people around the globe have taken antioxidant supplements and eaten foods rich in antioxidants, figuring this was one of the paths to good health and a long life.
Yet clinical trials of antioxidant supplements have repeatedly dashed the hopes of consumers who take them hoping to reduce their cancer risk. Virtually all such trials have failed to show any protective effect against cancer. In fact, in several trials antioxidant supplementation has been linked with increased rates of certain cancers. In one trial, smokers taking extra beta carotene had higher, not lower, rates of lung cancer.

In a brief paper appearing in The New England Journal of Medicine, David Tuveson, M.D. Ph.D., Cold Spring Harbor Laboratory Professor and Director of Research for the Lustgarten Foundation, and Navdeep S. Chandel, Ph.D., of the Feinberg School of Medicine at Northwestern University, propose why antioxidant supplements might not be working to reduce cancer development, and why they may actually do more harm than good.

Their insights are based on recent advances in the understanding of the system in our cells that establishes a natural balance between oxidizing and anti-oxidizing compounds. These compounds are involved in so-called redox (reduction and oxidation) reactions essential to cellular chemistry.

Oxidants like hydrogen peroxide are essential in small quantities and are manufactured within cells. There is no dispute that oxidants are toxic in large amounts, and cells naturally generate their own anti-oxidants to neutralize them. It has seemed logical to many, therefore, to boost intake of antioxidants to counter the effects of hydrogen peroxide and other similarly toxic "reactive oxygen species," or ROS, as they are called by scientists. All the more because it is known that cancer cells generate higher levels of ROS to help feed their abnormal growth.

Drs. Tuveson and Chandel propose that taking antioxidant pills or eating vast quantities of foods rich in antioxidants may be failing to show a beneficial effect against cancer because they do not act at the critical site in cells where tumor-promoting ROS are produced -- at cellular energy factories called mitochondria. Rather, supplements and dietary antioxidants tend to accumulate at scattered distant sites in the cell, "leaving tumor-promoting ROS relatively unperturbed," the researchers say.

Quantities of both ROS and natural antioxidants are higher in cancer cells -- the paradoxically higher levels of antioxidants being a natural defense by cancer cells to keep their higher levels of oxidants in check, so growth can continue. In fact, say Tuveson and Chandel, therapies that raise the levels of oxidants in cells may be beneficial, whereas those that act as antioxidants may further stimulate the cancer cells. Interestingly, radiation therapy kills cancer cells by dramatically raising levels of oxidants. The same is true of chemotherapeutic drugs -- they kill tumor cells via oxidation.

Paradoxically, then, the authors suggest that "genetic or pharmacologic inhibition of antioxidant proteins" -- a concept tested successfully in rodent models of lung and pancreatic cancers -- may be a useful therapeutic approach in humans. The key challenge, they say, is to identify antioxidant proteins and pathways in cells that are used only by cancer cells and not by healthy cells. Impeding antioxidant production in healthy cells will upset the delicate redox balance upon which normal cellular function depends.

The authors propose new research to profile antioxidant pathways in tumor and adjacent normal cells, to identify possible therapeutic targets.

1. 
   
   • Elizabeth G. Phimister, Navdeep S. Chandel, David A. Tuveson. The Promise and Perils of Antioxidants for Cancer Patients. New England Journal of Medicine, 2014; 371 (2): 177 DOI: 10.1056/NEJMcibr1405701 http://www.nejm.org/doi/full/10.1056/NEJMcibr1405701

 

[adreno]

This does however not mean that antioxidants aren't helpful in ME/CFS, that seem to be characterized by oxidative stress.

Perhaps liposomal antioxidants could be more useful, as they can more readily penetrate the mitochondria.

Furthermore some antioxidants likely target mitochondria more than others, so perhaps we can not think of all antioxidants in the same way.

Finally, most compounds classified as "anti-oxidants" also has pro-oxidant mechanisms.

 

[MeSci]

This does however not mean that antioxidants aren't helpful in ME/CFS, that seem to be characterized by oxidative stress. Also, perhaps liposomal antioxidants could be more useful, as they can more readily penetrate the mitochondria. Furthermore some antioxidants likely target mitochondria more than others, so perhaps we can not think of all antioxidants in the same way.

although, again, the fact that we appear to have trouble making ATP through oxidative phosphorylation suggests that perhaps reducing (decreasing) oxidation might not be a good idea? The fact that we may have difficulty getting oxygen into mitochondria?

I have long questioned the modern obsession with anti-oxidation being the reason for the benefits of many compounds, such as fruit and veg and omega-3s. It seems too simplistic, and I had for some time had questions in my mind along the lines of exactly what these researchers are saying.

Very interesting - thanks, @Ecoclimber

 

[anciendaze]

My comments here are made based on the post above. I may change my mind after I've had a chance to read the whole paper.

By concentrating on cancer these researchers seem to be downplaying the normal role of ROS in disposing of foreign or imperfect proteins. These proteins are tagged by ligands which bind ATP to them, perhaps several times. These tags also help to power the lysosomes which degrade proteins. These are standard cellular defense mechanisms against infection inside the cell which involve ROS. They are used in defense against viruses, rickettsial infections and even prion diseases, to say nothing of the defects in gene transcription which can trigger cancers.

Aside: We have two contradictory facts in ME/CFS: oxydative stress is increased; ATP is less available. To me at least, this sounds like the pathways used to degrade useless proteins are dysfunctional and action is misdirected. Similar dysfunction shows up in cancer.

Keep in mind that use in treatment of cancer takes place very late in the pathological process, when a great many things have already gone wrong. This paper does not appear to take any position on the etiology of the cancers being treated.

There are correlations between a variety of infectious agents and incidence of cancer. There are correlations with oxydative stress. Simply adding fuel to the fire may not be desirable in cases where either a pathogen or a homegrown cancer has managed to co-opt cellular machinery for its own purposes.

 

[adreno]

Interesting, @anciendaze

I don't have the biochem background to fully understand these things, but lets take ALA as an example. It seems to me that the MOA go beyond ROS scavenging:

alpha-lipoic acid a scavenger of reactive oxygen species in vivo? Evidence for its initiation of stress signaling pathways that promote endogenous antioxidant capacity.
Petersen Shay K1, Moreau RF, Smith EJ, Hagen TM.
Author information

Abstract
The chemical reduction and oxidation (redox) properties of alpha-lipoic acid (LA) suggest that it may have potent antioxidant potential. A significant number of studies now show that LA and its reduced form, dihydrolipoic acid (DHLA), directly scavenge reactive oxygen species (ROS) and reactive nitrogen species (RNS) species and protect cells against a host of insults where oxidative stress is part of the underlying etiology. However, owing to its limited and transient accumulation in tissues following oral intake, the efficacy of nonprotein-bound LA to function as a physiological antioxidant has been questioned. Herein, we review the evidence that the micronutrient functions of LA may be more as an effector of important cellular stress response pathways that ultimately influence endogenous cellular antioxidant levels and reduce proinflammatory mechanisms. This would promote a sustained improvement in cellular resistance to pathologies where oxidative stress is involved, which would not be forthcoming if LA solely acted as a transient ROS scavenger.

It also looks like ALA has pro-oxidant properties:

Antioxidant and prooxidant activities of alpha-lipoic acid and dihydrolipoic acid.
Moini H1, Packer L, Saris NE.
Author information

Abstract
Reactive oxygen (ROS) and nitrogen oxide (RNOS) species are produced as by-products of oxidative metabolism. A major function for ROS and RNOS is immunological host defense. Recent evidence indicate that ROS and RNOS may also function as signaling molecules. However, high levels of ROS and RNOS have been considered to potentially damage cellular macromolecules and have been implicated in the pathogenesis and progression of various chronic diseases. alpha-Lipoic acid and dihydrolipoic acid exhibit direct free radical scavenging properties and as a redox couple, with a low redox potential of -0.32 V, is a strong reductant. Several studies provided evidence that alpha-lipoic acid supplementation decreases oxidative stress and restores reduced levels of other antioxidants in vivo. However, there is also evidence indicating that alpha-lipoic acid and dihydrolipoic acid may exert prooxidant properties in vitro. alpha-Lipoic acid and dihydrolipoic acid were shown to promote the mitochondrial permeability transition in permeabilized hepatocytes and isolated rat liver mitochondria. Dihydrolipoic acid also stimulated superoxide anion production in rat liver mitochondria and submitochondrial particles. alpha-Lipoic acid was recently shown to stimulate glucose uptake into 3T3-L1 adipocytes by increasing intracellular oxidant levels and/or facilitating insulin receptor autophosphorylation presumably by oxidation of critical thiol groups present in the insulin receptor beta-subunit. Whether alpha-lipoic acid and/or dihydrolipoic acid-induced oxidative protein modifications contribute to their versatile effects observed in vivo warrants further investigation

 

[adreno]

@anciendaze

You say that oxidative stress and low ATP is contradictory, but according to this article it is not:

AGING IS A MULTIFACTORIAL PROCESS that leads to loss of function and the inability to adequately respond to external stress. Mitochondrial dysfunction appears to contribute to some of the loss of function accompanying aging (1⇓, 2⇓). Mitochondria from aged tissue use oxygen inefficiently, which impairs ATP synthesis and results in increased oxidant production (3⇓, 4⇓). The high flux of oxidants not only damages mitochondria, but other important cell biomolecules as well. Antioxidant defenses also decline with age (5⇓, 6⇓), making mitochondria even more susceptible to oxidative injury (1)⇓. The resultant mitochondrial decay may eventually cause inadequate energy production and/or the loss of calcium homeostasis. Such changes could result in unwarranted cellular apoptosis and also lead to the general metabolic decline evident in aging.

And lipoic acid reverses this age related decline:

We demonstrate that lipoic acid supplementation of old rats markedly improves the average mitochondrial membrane potential and restores the cellular oxygen consumption (Table 1)⇓ in hepatocytes to that of young rats. Rats on this feeding regimen were significantly more active, which further shows that (R)-lipoic acid acts physiologically to increase general metabolic activity. While the underlying causes for this increased energy metabolism were not explored, it is plausible that lipoic acid improves mitochondrial function through a number of mechanisms. Administration of lipoic acid stimulates insulin-dependent and independent glucose uptake into cells (33)⇓ and also enhances nonoxidative and oxidative glucose metabolism. Reduced (R)-lipoic acid has also been shown to increase ATP synthase activity (16)⇓, which in combination with increased glucose utilization would be expected to enhance overall cellular metabolism. Finally, as a potent antioxidant, dihydrolipoic acid may also maintain critical thiol groups in a reduced state and allow mitochondrial protein carriers to function more effectively (16)⇓.

http://www.fasebj.org/content/13/2/411.long

 

[Marco]

@adreno

Looks like an interesting paper you posted which I'll have to read.

This :

AGING IS A MULTIFACTORIAL PROCESS that leads to loss of function and the inability to adequately respond to external stress.

strikes me as one of the better descriptions of ME/CFS.

 

[anciendaze]

@adreno

Perhaps you would like it better if I had said "apparently contradictory". At the moment I'm leaving the original post unchanged so I won't confuse readers.

@Marco

I have already said that one problem with current definitions of "normal" w.r.t. to such things as punctate lesions on high-resolution MRIs is that they lump people in very different age cohorts together. When people who were apparently in excellent health can suddenly become comparable to those who are 40 years older you have evidence of serious illness. It was no accident that PACE made such comparisons with a general population that was much older, and frequently included diagnosed health problems for which those ME/CFS patients were said to have been screened.

Unfortunately, the term "multifactorial" is often used in papers where nobody has a clue about what is going on. The problem here is distinguishing between multiple initial causes, diagnostic confusion between different conditions, and the inevitable accumulation of defects when a pathological process persists for a long time. Acute disease frequently has the simplifying characteristic that a single cause in the recent past can be isolated. This is like the single-fault hypothesis in troubleshooting complex machines, which is convenient even though it is not necessarily true.

When you can't count on problems being the result of a single fault, as in higher-level maintenance of aircraft, computers or the Space Shuttle Orbiters, maintenance becomes a process of testing every subsystem independently. It helps if you can tear the system down and test each subsystem separately before reintegrating them. This is not possible for humans. Even when it is possible for aircraft or spacecraft, it is very expensive.

At a level closer to common experience, there comes a point where it is cheaper to buy a new automobile, even though it is possible to keep fixing an old one. Not only is real expertise in that case expensive, the expertise you may actually have available is likely to create new problems faster than it fixes existing ones.

Aside: when I talk about troubleshooting computers, most people with recent experience will immediately think of swapping various boxes and chips, which allows you to avoid detailed troubleshooting. I would caution them that this is not at all the same as troubleshooting when you are dealing with unique devices with complex interactions. Bringing up a bit-slice processor at the same time we debugged the microcode was an experience which left scars. We couldn't compare it with a correctly functioning machine because we didn't have one. We also had to deal with errors in subcontracted multilayer printed circuit boards. (Ever patch an error on an inside layer?) It didn't help that some of the parts we used turned out to have different chips inside from the one indicated on the label. This was a little joke by suppliers who could not actually deliver working parts they promised on schedule. After that experience I lost some of my bashfulness about challenging medical expertise. I had been through a baptism of fire dealing with problems of similar complexity to some common medical problems.

 

[Marco]

@anciendaze

As someone who likes to view computers as domestic appliances you have my sympathies.

My point was that "loss of function and the inability to adequately respond to external stress" quite neatly sums up how we are affected on a daily basis. Its not just not being able to keep up that weekly squash game like we used to or finding it tough working a 40 hour week - more just standing, walking to the mailbox or enduring a 20 min telephone call. Add to this extreme food and chemical sensitivities, alcohol intolerance (in some) etc. So yes something is very wrong in often very young people but, as appears to be the case with the aged, there is an inability to maintain homestasis (sic) in response to 'stressors' that were previously trivial.

While the strength of evidence may vary it does appear that CFS patients lack or have acquired deficiencies in a range of basic cellular protection and repair mechanisms. For example elevated markers of oxidative stress, reduced ATP/Glutathione, impaired heat shock protein production in response to exercise, reduced NK cell function (although I'm not convinced this is particularly significant), and reduced BDNF production in association with findings of microglial activation.

Similar findings are associated with aging including microglial priming which is evidenced by the exaggerated and prolonged 'sickness behavior' response to common infections (some never recover from common infections or minor injuries).

RE : 'CFS', the onset of illness is often associated with a 'flu from hell' however prior microglial priming might result in a similarly exaggerated sickness reponse casting considerable doubt on the assumption of a 'viral onset'.

 

[adreno]

@anciendaze

My point was that there is apparently a connection between oxidative stress and low ATP, and that it seems (at least some) antioxidants ameliorate both conditions.

 

[MeSci]

@anciendaze

My point was that there is apparently a connection between oxidative stress and low ATP, and that it seems (at least some) antioxidants ameliorate both conditions.

But do they do it via antioxidation...?

 

[anciendaze]

@adreno & @MeSci

I long ago learned caution concerning biochemistry and pharmacology. I've concluded that most popular descriptions of the kinetics of biochemical reactions are "lies-for-students". The real story in the vast majority of cases involves multiple stages of metabolism, each of which is strongly non-linear. Most scientists aren't up to handling masses of non-linear differential equations, so they rather arbitrarily decide to ignore some parts of the problem because they are "too small to be important", a judgment in terms of proportional response. This reflects a deep-seated belief that everything rational in nature must be fundamentally linear, because this is so much easier than the alternatives.

My own experience also involves such nonlinear phenomena as "flutter" in airplanes, which is not the same as resonance, yet can be strong enough to tear the wings off, as happened in some Lockheed Electras. I have enough trouble with physics teachers showing the film of the collapse of the Tacoma Narrows bridge as an example of resonance, ignoring the fact that the design was analyzed for resonances before construction, and the incomplete structure was tested for these. It was tested again after completion. This has no effect on the vast majority of people aware of this failure. They just don't believe that an effect which was too small to be detected could build over a period of days into an oscillation which could destroy a large structure. Linearity implies that response is proportional to the cause, and this violates that assumption.

Another part of my checkered past involves detonation waves in explosives, which are strongly nonlinear. You don't live to a ripe old age in that field if you use assumptions of linearity. (I tell people to remember Emil Oskar Nobel. "Who?" They ask.)

A third example of nonlinear behavior was a ship which generated a "solitary wave of translation" large enough to stop it from going faster, getting "stuck on the hump". This was also such a minor influence initially that people who should know better have insisted it was not possible except in water so shallow the ship would be in danger of running aground. Nonlinear processes are quite capable of accumulating energy over time that linear processes would disperse.

When it comes to medicine this is only one aspect of the conceptual gulf. I've regularly had doctors express disbelief about the percentage of serotonin receptors outside the brain, and the percentage inside the brain which are inactive because of glial blockade. When SSRIs work they do so by operating on a tiny minority of receptors over a period of a week or more. Because chemical concentrations change in hours, the delay in action requires a much more sophisticated understanding. I have yet to encounter a doctor who has this, though I feel sure there must be such a paragon somewhere out there. Even after doctors have learned caution in dealing with me, I still hear that they have used the same old explanations with other patients. Convenient plausibility is more powerful than truth.

As something easier to understand than the full cycles involved in energy metabolism I suggest the reaction kinetics of hemoglobin, which has the peculiar property of being more likely to pick up a second oxygen molecule once it has picked up the first, and to dump both when it dumps one. This is a good litmus test for a physician's understanding of biochemistry. If they can't explain that fundamental reaction for oxygen metabolism they are unlikely to cope with the intricacies of the Krebs cycle and its modern extensions.

Elsewhere I've also investigated literature on unexpected actions of drugs, like antibiotics which also affect viral titers. Just because the label says one thing does not stop the molecules from doing something quite different. This can reach such extremes as the antidepressant fluoxetine (Prozac) strongly inhibiting replication of coxsackie B virus.

It is entirely possible that medical professionals don't even understand the reasons for many of their successes.

 

[adreno]

But do they do it via antioxidation...?

I don't know if that is relevant? The compounds classified as "anti-oxidants" that people supplement in order to treat cancer or ME have complex mechanisms. I don't even know if any molecules with purely ROS scavenging mechanism exist. Perhaps it does in a lab, but the relevant question here is whether the supplements (lipoic acid, curcumin, resveratrol, quercetin, vitamin C etc) that people take in real life are helpful or not.

 

[MeSci]

I absolutely agree that the way science is taught is excessively simplistic. But it would require a huge overhaul to change this. Maybe one day it will be possible with the aid of technology. Biological processes are of course not linear but a complex web of interactions. At least we are taught a bit about positive and negative feedback.

But doctors? I wouldn't expect them to understand any of this. They're not scientists - at least the vast majority are not. In the UK at least they just have Bachelors degrees, of which only a modest proportion is science. Their knowledge of drug activity is probably no more/little more/less than what can be found here for example.

 

[anciendaze]

@MeSci

The problem I have constantly run into in applied mathematics is people forgetting the difference between the model they are using and the reality being modeled. It is also important to keep in mind the extent to which a model has been validated. This is the point where medical and legal professionals fall into the error of mistaking social validation by peer groups for validation against nature.

This error is so popular it may be that many simply do not care. Peer support will preserve their relative advantage over people outside their profession, regardless. As an example I offer the history of Semmelweiss vs. Virchow. Semmelweiss had strong statistical evidence that doctors were killing patients, though without showing the mechanism. He also showed that having doctors wash their hands in a lime solution cut mortality dramatically. This did not move the consensus until after his death. I suppose response would have been different if the doctors were dying.

From my experience with the effect of complexity on computation and human problem solving, I've developed rules of thumb about limitations on human information processing. I've noticed that doctors do much better in catching interaction of two drugs than three. (Though I also know of a case in which doctors gave "just a little Demerol" to a patient wearing a Medicalert bracelet for MAOIs.) Beyond three I don't think any of them can cope. This is probably a factor in mortality of patients who are elderly or have been sick for a long time.

Even the most powerful reasoning ability will be defeated by bad data. (There's a reason Klonopin is now spelled with a "K".) This has even turned up in the surprising situation of errors in surgical procedures. Years ago there was a checklist to cut errors which involved such things as having members of the surgical team introduce themselves before they began, and tell why they were there. This actually saved lives. "Oh silly me! The pneumonectomy is next Tuesday."

Caution: Having worked in a field where errors cause people to vanish in a pink cloud I probably have a jaundiced and distorted view of the frequency and importance of errors. People who survive for long become convinced that Murphy was overly optimistic.

 

[MeSci]

@MeSci

The problem I have constantly run into in applied mathematics is people forgetting the difference between the model they are using and the reality being modeled. It is also important to keep in mind the extent to which a model has been validated. This is the point where medical and legal professionals fall into the error of mistaking social validation by peer groups for validation against nature.

This error is so popular it may be that many simply do not care. Peer support will preserve their relative advantage over people outside their profession, regardless. As an example I offer the history of Semmelweiss vs. Virchow. Semmelweiss had strong statistical evidence that doctors were killing patients, though without showing the mechanism. He also showed that having doctors wash their hands in a lime solution cut mortality dramatically. This did not move the consensus until after his death. I suppose response would have been different if the doctors were dying.

I think we are in agreement!

I have come across something similar in my study and work on the subject of animal 'models'.

Example: a human skin model was developed as an alternative to shaving lab animals and testing them.

The findings from animals correlated poorly with those from human volunteers.

Yet the new human skin model was validated, not against the human volunteer results, but against the animal findings! (It was thus found wanting.)

I am very alert to the likelihood of errors, and have suffered numerous errors at the hands of the medical profession, which is why I no longer trust doctors, and avoid them as much as I can.