Effects of Homeopathy in cell culture: O rly?
This post concerns this paper : “Dynamized Preparations in Cell Culture” — Sunila et al. 6 (2): 257 — Evidence-based Complementary and Alternative Medicine
I discovered this paper whilst taking a look at this website which is well worth a read, if ever feel you might want to learn about telepathic DNA (no. srsly) or absorb life threatening amounts of stupid.
Sunila et al want to look at the effect of homeopathic remedies on cells in a cell-culture setup. They also want to use homeopathic remedies rumoured to be effective in combating cancer. The hypothesis would seem to be that if the cells treated with homeopathic remedies die in cell culture then OMG! Homeopathy will cure cancer!
They use a set of homeopathic remedies, and compare mother tinctures (MTs), 30C preparations and 200C preparations.
So they take some cells, and add some homeopathic remedies to them. They then assay for stuff like cell viability (via MTT or tryptan blue exclusion method ) cell proliferation via tritiated thymidine uptake assay, and CHO cell colony formation, p53 message upregulation via RT-PCR and DNA damage via DNA laddering assay. Fine.
One of the first things that annoys the pants off me about this paper is that they seem to chop and change cells types for each different set of assays, seemingly arbitrarily. Why this is, I don’t know. If I was feeling uncharitable, I might say that this is because assay results they see only occur for certain cell types. If.
What they should have done (IMHO) is either do some experiments for all types of cells, or just use one cell type for all experiments. Chopping and changing like this makes comparison between the different assays dubious at best.
If you can’t get at the paper, or don’t want to plough through it, here are the results in bullet point form:
- Tryptan blue exclusion method: Dalton’s lymphoma ascites & Ehrlich ascites carcinoma cells – (figures 1 & 2 in the paper).
- “Results on DLA cells and EAC cells were almost similar.”
- Yes – “Almost similar” – I’ve never seen that phrase in a paper before, either.
- Negative controls: Cells live.
- Postive controls: There are no positive controls.
- 200C Conium, Carcosinium, MTs of Condurango, Hydrastis & Phytolacca and 200C preps of Thuja & Hydratis: Some cells Die.
- Thuja & Lycopodium mother tintcures (contain active ingredient). ~All cells die.
- Long term viability of L929 cells is quite drastically reduced in most MTs, and to a lesser extent in some of the 30C and 200C preps (fig3).
- They show that pretty much all their remedies, but not the controls reduce CHO cell colony formation (fig4).
- They show that their MTs significantly reduce tritiated thymidine uptake (fig5).
- They show that Thuja seems to induce DNA damage (fig6) and that Carcosinium seems to induce p53 expression (fig7).
- Table 1 shows that all but one of their MTs induces apoptosis, and fewer 30C and fewer again 200C preps cause apoptosis, as determined (subjectively) by morphology.
The dye exclusion method, CHO clustering & visual assessment of apoptotic phenotype are all subjective “looking down a microscope” assays, and *may* be subject to bias. Just saying, is all.
So, we’ve got a fair old batch of results to analyse. The assays are pretty standard, and so assuming they were done competently, methodology appears fine – although there is a caveat wrt the dye exclusion assays – which measure cell viability indirectly by measuring cell membrane integrity – but probably not an issue here.
However, they do not state how many times the experiments were repeated – n=2? n=3? n=400? who knows?
There is one teensy weensy little caveat in the paper, mentioned in the methods, tucked under the table of remedies used:
“Ethanol content in the preparations may vary and no attempt was made to determine the exact content. Maximum ethanol concentration used in the experiment was 2%, which will not produce any effect on cells. Moreover results were compared with dynamized vehicle control or non-dynamized alcohol control.”
The alcohol control was only used in one assay- the p53 RT-PCR assay. All the others were controlled with “dynamized vehicle control” – which sounds like some kind of groovy new steering feature on a formula 1 car – but is, as far as I can see, just shaken phosphate-buffered saline. Which won’t contain ethanol.
But hey! Remeber, that doesn’t matter, because “Maximum ethanol concentration used in the experiment was 2%, which will not produce any effect on cells.”
But they don’t provide a reference for this. This would appear to be an assumption made, and not supported, or checked, or referenced in any way.
To be fair, 2% Ethanol, doesn’t sound like much – however, a bit of high school chemistry can show you that 2% Ethanol solution corresponds to 435mM Ethanol.
Mwt of ethanol ~ 46g
100% (pure) ethanol = 1000/46 = 21.7M
2% Ethanol = 21.7*0.02 = 0.435M Ethanol.
Which is actually a pretty high concentration. Even if the remedies only contain 1% ethanol, we’d still have 4.35mM ethanol in our cell culture, which is more than enough of a compound to have an appreciable effect in cell culture.
But we don’t know what alcohol is in the remedies – The authors didn’t check. “no attempt was made to determine the exact content” – this is really bad science. They didn’t even bother to check.
I thought it would be fun to go to the Willmar Schwabe (suppliers of the remedies) website, and see if there is any information regarding alcohol content of their remedies.
This page reveals that their mother tinctures are “The therapeutically active ingredients in many cases are extracted in alcohol-water mixture of varying percentages of alcohol depending on the solubility, stability and extractability of the potential ingredients. Such alcoholic extracts are called “Mother Tinctures”.”
So the MTs will definately contain some ethanol.
What about the 30C and 200C remedies?
Well, according to this:
“The use of genuine raw materials, back potency and expensive and purest form of alcohol, namely Extra Neutral Alcohol (ENA) make Schwabe India dilutions superior to other dilutions available in the market. Extra Neutral Alcohol guarantees that the dilutions and mother tinctures are free from impurities.”
The 30C and 200C remedies will also definately contain some ethanol.
So, to sumarise:
- The MTs contain planet extract in an alcohol/water mix, of unknown proportions and composition.
- The 30C and 200C preps contain an unknown alcohol/water mix (+ water memory of original compounds, OBVIOUSLY)
I’ve e-mailed Willmar Schwabe asking for more specific information about the composition of these remedies… as and when I get a response, I’ll update this blog post.
Given that we have established that there is most likely *some* ethanol in all these remedies, and paying particular attention to the assays caried out in Sunila et al, just what are the effects of low concentrations of Ethanol in cell culture?
1) Effects of alcohol on cell viability.
Ethanol concentrations as low as 10mM lead to increased necrosis and apoptosis (i.e. cell death), and DNA fragmentation, another assay used in Sunila et al.
Castilla, R et al, DUAL EFFECT OF ETHANOL ON CELL DEATH IN PRIMARY CULTURE OF HUMAN AND RAT HEPATOCYTES (2004) Alcohol & Alcoholism Vol. 39, No. 4, pp. 290–296, 2004 doi:10.1093/alcalc/agh065,
2) Effects of alcohol on cell clustering
“Low concentrations of ethanol (5-10 mm) inhibited cell-cell adhesion“
An appreciable decrease in cell clustering is observed in low concentrations of alcohol – one of the measures of the effect of homeopathic remedies used in this paper.
Charness, ME et al. Ethanol Inhibits Neural Cell-Cell Adhesion, (1994), J. Biol. Chem, Vol. 269, No. 12, Issue of March 25, pp. 9304-9309
3) Effects of alcohol on thymidine uptake.
Suppressed levels of Thmidine uptake in the cells exposed to the MTs and remedies might be due to the fact that “alcohol levels greater than or equal to 50 mg/dL suppressed tritiated thymidine uptake of normal lymphocytes”
Glassman AB et al. Effects of ethyl alcohol on human peripheral lymphocytes. (1985) Arch Pathol Lab Med. Jun;109(6):540-2.
4) Effects of alcohol on p53 expression
Ethanol causes Oxidative stress – Sunila et al assayed for p53 expression – p53 expression known to be induced by oxidative stress.
Han, E.S. et al The in vivo gene expression signature of oxidative stress. (2008) Physiol Genomics. 12;34(1):112-26.
5) Effects of Alcohol on DNA fragmenting:
Castilla, R et al, again.
This paper shows that 10mM concentrations of Ethanol lead to increased necrosis and apoptosis, and DNA fragmentation, another assay used in Sunila et al.
Now. Given that there is published, peer-reviewed evidence that suggests that all the effects observed in Sunila et al can be explained by the presence of small amounts of ethanol, and that, the manufacturers state that all the remedies used in this Sunila et al will contain small amounts of ethanol, is it not more plausible and rational to conclude that the effects observed are caused by the small amounts of ethanol present?
Differences between different preps might be explained by different ethanol/water ratios. Differences between the effect of different MTs and MTs and 30C/200C preps can be explained by the presence in the MTs of actual, measurable amounts of chemicals that may have an effect upon cells in culture.
All this uncertainty could have been solved with a bit of Infrared spectroscopy, and better design of controls, to contain an appropriate amount ethanol.
However, such controls might well have just shown that these in vitro effects of homeopathic remedies are all down to the alcohol content. Just like Frenkel et al. (see Dr Rachie’s post and Orac’s post).
It would seem that when it comes to cell culture, homeopaths have some sort of cognitive blind spot which prevents them providing suitable controls for the solvents in which their homeopathic remedies are… remembered? One would hope that given the homeopathic preocupation with the memory of solvents, that they might remember to use the correct solvents in their controls.