THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINEVolume 12, Number 4, 2006, pp. 359365 Mary Ann Liebert, Inc.
Does Potentized HgCl2 (Mercurius corrosivus) Affect the Activity of Diastase and -Amylase?
CLAUDIA M. WITT, M.D.,1 MICHAEL BLUTH, M.D.,1 STEPHAN HINDERLICH, Ph.D.,2
HENNING ALBRECHT, Ph.D.,3 RAINER LDTKE, M.Sc.,3 THOROLF E.R. WEISSHUHN,1
and STEFAN N. WILLICH, M.D., M.P.H.1
Background: Homeopathic drugs even with dilutions beyond 1023 (high potencies) are frequently used,although their working mechanism is still unknown. Curative information preserved in solvent structure is pos-tulated to exert biologic effects.
Objective: The objective was to test for a stimulating or inhibiting effect of high potencies of the homeo-pathic remedy HgCl2 (Mercurius corrosivus) on two sugar hydrolases.
Methods: High potencies were produced using stepwise dilution plus shaking. Controls included potentizedsolvent (aqua bidestillata), equimolar dilutions without shaking, and enzyme-free references. Tested were po-tencies with dilution factors 1:200 (CC) on diastase extract from winter barley, and 1:100 (C) on -amylasefrom hog pancreas. Enzyme activity was colorimetrically determined by Lugols iodine-starch reaction.
Results: An inhibiting effect of HgCl2 on enzyme activities was observed only in low potencies and dilu-tions. Statistically significant differences between potencies and controls were not found in randomized andblinded experiments.
Conclusions: This experimental design provided independent reproducible results of cell-free in vitro as-says. However, it did not indicate an effect of potentized HgCl2 on hydrolases. Demonstrating potency effectsmay require additional experimental features.
One of the most frequently used systems of complemen-tary and alternative medicine is also one of the most con-troversialhomeopathy. The mechanisms of this therapy arepoorly understood, and yet its use is rising worldwide.14 Thestrongest criticism targets the preparation of drugs, in whicha starting substance, the mother tincture (a traditional termthat applies also to solids and dissolved gases), is subjectedto iterations of dilution and shaking. The resulting potencieswith calculatory dilutions below (low) and beyond (high)Avogadros number are clinically used. However, how can apotency be different from a corresponding dilution? Can thisbe demonstrated for high potencies?
The in vitro field is privileged in two ways for potencyresearch: It reduces the complexity of a living organism andthe number of possible confounders while investigating re-actions fundamental to living organisms. A number of ex-periments have been performed already that demonstrateddifferences between potencies and controls. Four experi-ments had reported a stimulating effect of high potencies ofmercury chloride (known homeopathically as the remedyMercurius corrosivus) on hydrolases. Two pioneeringteams5,6 had used malt diastase, a mixture of several hydro-lases, among them -amylase and limit dextrinase.7 Recentconceptual replications investigated -amylase from humansaliva8 or hog pancreas.9 The diastase and amylase modelsas performed in the ways described in the literature thus
1Institute for Social Medicine, Epidemiology and Health Economics, Charit University Medical Center, Berlin, Germany.2Institute for Biochemistry and Molecular Biology, Charit University Medical Center, Berlin, Germany.3Karl and Veronica Carstens-Foundation, Essen, Germany.
seemed suitable to investigate effects of potentized HgCl2and successful replications might become starting points fora better understanding of the mechanisms underlying home-opathic treatment. The authors objective was to investigatewhether they could be used for this purpose employing mod-ern biochemical laboratory techniques and equipment.
Comprehensive high-quality meta-analyses of controlledclinical studies using various methods1013 (only the mostrecent that cover all languages, all diagnoses, and all publi-cations cited) do not support the placebo hypothesis, indi-cating that a potency may have a (yet-unexplained) specificeffect. Curative information that is passed from the startingsubstance on to high potencies via persistent changes in sol-vent structure is discussed to explain the observed ef-fects.1420 Systematic quality assessments in basic researchreached similar conclusions: For experimental toxicology,strong evidence of a high potency effect was found,21 andalso for physical experiments22 and in vitro research.*
MATERIALS AND METHODS
Materials included -amylase from hog pancreas(A2771), mercury chloride (M1136), soluble starch (S9765),and Lugols solution (L6146) by Sigma-Aldrich (Seelze,Germany); as well as sodium phosphate, sodium hydrogenphosphate, potassium chloride, sodium acetate, and sodiumcitrate by Roth (Karlsruhe, Germany). Fine flour of an en-zyme-rich winter barley (Hordeolum vulgare) was providedby the Institute of Brewing Sciences of the Technical Uni-versity, Berlin (Germany).
Preparation of aqua bidestillata
Aqua bidestillata was prepared from deionized water bydestillation at 100C in a glass destillation apparatus (Schott,Mainz, Germany). A freshly prepared lot of aqua bidestil-lata was used for all potencies, dilutions, and controls of anexperiment.
Preparation of malt diastase extract
Fifty milligrams (50 mg) of fine ground winter barleymalt were stirred with 100 mL aqua bidestillata for 15 min-utes at 250 rpm (Vortex Genie, Scientific Industries, NewYork, NY). The resulting slightly viscous suspension wasfiltered through a folded filter (Selecta; Schleicher undSchuell, Dassel, Germany). Malt diastase extract was freshlyprepared on every experiment day.
Preparation of 1% starch solution
One gram (1 g) of starch was mixed with 10 mL of aquabidestillata and 80 mL of boiling aqua bidestillata wereadded. After 1 minute, the swollen starch was boiled in amicrowave for 3 minutes. After cooling the solution was re-filled to 100 mL and filtered three times through a roundfilter, diameter 55 mm (Schleicher und Schuell) using a vac-uum pump, resulting in a viscous, dimmish liquid. This 1%starch solution was freshly prepared on every experimentday.
Preparation of HgCl2 mother tincture
A mother tincture of 0.1 g HgCl2 was dissolved in 10mL of aqua bidestillata and stirred for 10 minutes at 200rpm. Five milliliters (5 mL) of the clear solution werepipetted into a 10-mL glass bottle of brown spessart glass(Aponorm, Lohr, Germany). The solution was labeledHgCl2 C 1.
Potencies, dilutions, and controls
Every step of potentizing consisted of dilution with sub-sequent shaking. For the first potency, a starting substancewas diluted and shaken, then, for every subsequent potencythe result of its previous potentizing step. A potency n thuswas obtained after n iterations of dilution and shaking, eachtaking about 1 minute. The starting substance for all potencysamples was HgCl2, and for all controls the potentizing sol-vent was (i.e., aqua bidestillata). The parameters for thesepreparations are listed in Table 1. All potencies and controlsare identified in this paper according to the conventionalnomenclature for homeopathic potencies; that is, with start-ing substance, dilution factor in roman numerals (C for 1:100,CC for 1:200), and the number of potentizing steps. For ex-ample, HgCl2 CC 3 indicates the result of mercury chloridethat underwent three steps of dilution 1:200, each time be-ing shaken afterward. Potentizing satisfied instruction 5b of German regulations for the preparation of homeopathicdrugs,23 except for a dilution factor of 1:200 (when used).
Potencies and controls were potentized in parallel, alter-nating at every step between both series for identical tim-ing. Fresh 10-mL Spessart glass bottles were used for everystep. They were shaken 10 times by hand, striking downagainst a hard but elastic body (a heavy paperback book),with frequency and force as constant as possible. A seriesof stepwise dilutions, labeled dil instead of C or CC,were produced like potencies, except that refilling and drain-ing the pipette five times effected mixing.
The activity of malt diastase (experiments 1 and 2) and-amylase (experiments 3 and 4) served as indicators for asuspected effect of homeopathic potencies of HgCl2, viacomparison with controls of likewise potentized aqua bides-
WITT ET AL.360
*Bluth M. In vitro research on homeopathic potencies: A sys-tematic review and own experiments with cell-free systems [in Ger-man]. Thesis. Berlin: Institut fr Sozialmedizin, Epidemiologie undGesundheitskonomie; Charit-Universittsmedizin Berlin, 2005.
tillata, stepwise diluted HgCl2 solution, untreated aquabidestillata, and an enzyme-free reference. See Table 2 foran overview.
Malt diastase assays contained 100 L of 1% starch, 50L of the enzyme preparation, and 30 L of potentized HgCl2or respective controls. -Amylase assays contained 50 L of1% starch, diluted 1:24 in 10 mM of NaH2PO4, pH 7.5, 50L of the enzyme preparation, and 50 L of potentized orstepwise diluted HgCl2 or controls. Samples were incubatedon 96-well plates for the indicated time at 37C. Then 50 Lof Lugols solution in the indicated dilution was added, thesamples were homogenized by gentle shaking for 10 seconds,and then extinction at 650 nm was simultaneously detectedfor all samples by an enzyme-linked immunosorbent assayreader (Rainbow, Tecan, Maennedorf, Switzerland). Moredetails about the assays are listed in Table 2.
Data were entered in a spreadsheet (Microsoft Excel;Redmont, WA) and analyzed with SAS (release 9.1) soft-ware. With experiments 1 and 3 a randomized block analy-sis of variance (ANOVA) was fitted to the data where the dayof preparation (replication number) was taken as the blockfactor. With experimens 2 and 4 the authors fitted simple one-way analyses of variances. All models assumed homogene-ity of variances. Calculated p-values and confidence intervals(CIs) were based on appropriate t-tests within these models.
Enzyme activity was colorimetrically measured, based onLugols iodine-starch reaction, in which iodine is embeddedin linear chains of the starch helix. The color of this com-plex correlates with the chain length of the starchabsorp-
tion maximum of 12 glucose residues is at 490 nm, 30residues at 537 nm, and more than 80 residues higher than610 nm. Diastase and amylase are both endohydrolases ofstarch. Therefore, their activities were detected by decreaseof absorption at 650 nm.
The amounts of the different assay compounds of the as-says were optimized in pilot studies (data not shown). Fordiastase experiments, it was found that the assay developedby Boyd6 also was suitable for the present experiments, butassays with -amylase required much lower starch concen-trations. Details about the assays are listed in Table 2.
Experiments 1 and 2, malt diastase
For this experimental series a dilution factor of 1:200(except for the initial HgCl2 C 1) was used for potentizedHgCl2 and aqua bidestillata, in accordance with Boyds ex-periments.6 The potencies CC 6CC 12 were tested becausein this dilution range an assumed potency effect should be-come predominant to a biochemical effect. Potencies CC26CC 30 were included for comparison with Boyd.6 Un-treated aqua bidestillata (stored in a closed potentizing bot-tle while potencies were prepared) was used as an addi-tional control. This would disclose effects of a possiblepotentization of contaminants from the potentizing bottle incontrols.22
In a first approach, four replications of the same experi-ment were performed (experiment 1; see Table 2). Statisti-cal analysis revealed significant differences between HgCl2potencies and controls (Fig. 1). Diastase activity was re-duced by the potencies CC 6 (p 0.001), CC 8 (p 0.003),CC 10 (p 0.04), CC 28 (p 0.04), and CC 30 (p 0.04),whereas for CC 12, CC 26, and CC 27 no effect was found.Surprisingly, potentized as well as unpotentized aqua bides-tillata also had an inhibiting effect (p 0.001). For thesereasons, it cannot be excluded that the authors experimen-tal procedure resulted in artifacts. Therefore, it was decidedto perform a further series of experiments in which HgCl2potencies and controls were tested blind after randomizationby a statistician (RL) and coding by the team supervisor
POTENCIES AND HYDROLASE ACTIVITY 361
TABLE 1. PARAMETERS OF SAMPLE PREPARATION
Tested potencies Controls/comparisons
Sample series HgCl2 C n HgCl2 CC n ABD C n ABD CC n ABD HgCl2 dil n
Starting substance HgCl2 HgCl2 ABD ABD ABD HgCl2Dilution factor 1100a 1200a 1100 1200 1100a
Dil. quantities (l) 50 4950 25 4975 50 4950 25 4975 50 4950
Shaking strokes 10 10 10 10
aAt first potentizing step: HgCl2 1100 w/v in ABD.Solvent for all preparations was aqua bidestillata (ABD).n Number of iterations of dilution plus shaking for each sample, see text.C potencies are clinically used as centesimal potencies, the CC series matched earlier research, dil n are stepwise dilutions to
test for the influence of shaking.
(SH) (experiment 2; see Table 2). In this experiment, no sig-nificant differences were found between potentized HgCl2and potentized aqua bidestillata (p-values, 0.20.9), nor be-tween potentized and unpotentized controls (Fig. 1).
Experiments 3 and 4, -amylase
In this set of experiments, the influence of selected sam-ples from a potency series C 1C 30 (with dilution factor1:100) was tested on -amylase activity. In experiment 3(see Table 2), a strong inhibition (p 0.001) of potentizedHgCl2 (compared to potentized aqua bidestillata) was foundfor C 2. Clearly, this can be attributed to the well-knownbiochemical effect at a molar Hg2 concentration of about106 M.24 The activities in all other potentized samples weresimilar, and no differences to the controls were found (Fig.2). A single stimulation of -amylase activity with C 12(p 0.008) in one of four experiments could not be repro-duced and was not significant (p 0.67) when data fromall replications were pooled.
In order to distinguish between the biochemical effect ofC 2 found in the experiments described in the preceding andpotential other effects, a C 2C 5 series of potentized anddiluted HgCl2 (experiment 4; Table 2) was analyzed. Thepotencies C 2 and C 3 and the corresponding dilutions HgCl2dil 2 and dil 3 inhibited the activity of -amylase in com-parison to potentized aqua bidestillata (p 0.001; Fig. 2).In potencies C 4 and C 5, neither differences to potentizedaqua bidestillata was found (p, 0.50.9), nor differences be-tween dilutions and potencies (p, 0.30.9). HgCl2 C 3 and
the respective dilution contained an Hg2 concentration inthe range of 108 M, which seems to be still able to inhibitthe enzyme. Further dilutions to Hg2 concentration of1010 M and lower (as in C 4 and C 5) did not inhibit -amylase activity biochemically further. Therefore, the pre-requisite for other effects were given in potencies of C 6and higher.
In a series of four experiments, the authors investigatedwhether potentized HgCl2 stimulates the activity of hydro-lases. No effects were found that can be attributed to a spe-cific property of potencies. Earlier research had reported astimulating effect of HgCl2 potencies on malt diastase5,6 and-amylase.8,9 Others had found no such effects on five cy-tosolic and cytoplasmatic enzymes beyond the biochemicaleffect in very low potencies,25 similar to the present results.
Unlike most of the other cell-free research on potency ef-fects, (see footnote on p. 360)*, the authors tried to main-tain a strict standard for experimentation. Potentized con-
POTENCIES AND HYDROLASE ACTIVITY 363
FIG. 1. Effect of potentized HgCl2 on malt diastase. Malt dias-tase was treated with homeopathic potencies (CC) of HgCl2, andcompared to controls of likewise potentized solvent (aqua bides-tillata, ABD), untreated ABD, and an enzyme-free reference (Ref).Enzyme activity was detected by the iodine-starch reaction, andextinction was determined at 650 nm. Results were given as ex-tinction differences (Ext Diff mean of sample type differenceswith 95% confidence interval). Data were calculated with t-tests(***p 0.001; **p 0.01; *p 0.05; data from all replicationsper experiment pooled). Experiment 1, not randomized setup; Ex-periment 2, completely randomized and blinded setup.
FIG. 2. Effect of potentized HgCl2 on -amylase. -Amylasewas treated with homeopathic potencies (C) or dilutions (dil) ofHgCl2 against controls of likewise potentized solvent (aqua bides-tillata). Enzyme activity was detected by the iodine-starch reac-tion, and extinction was determined at 650 nm. Results were givenas extinction differences (Ext Diff mean of sample type differ-ences with 95% confidence interval). Data were calculated with t-tests (***p 0.001; data from all replications per experimentpooled). Experiment 3, not randomized setup of potencies. Exper-iment 4, not randomized comparison of potencies and dilutions;Differences: potency control; dilution potency; dilution control.
trols equalized impurities from the vessel wall, experimentswere replicated several times and, at least in part, randomizedand blinded. This may explain some of the differences to theexperiments that reported a potency effect: From the men-tioned precautions, Persson and Ginsberg5 and Shabirs team8
performed only the replications, and Sukuls group9 only thecontaminent checkup controls. Contamination from HgCl2preparations to potentized controls cannot be fully excludedas an explanation for the difference from the authors results;the alternated potentizing steps appear especially vulnerable(e.g., to aerosols). In such a case, however, potencies of HgCl2and aqua bidestillata essentially would be the same; their ef-fects, however, still should differ from controls of untreatedaqua bidestillata that had been stored in a closed containerduring potentizing.
The authors found in pilot studies that small changes iniodine concentration can cause large extinction changes ifthe starch is not completely saturated with Lugols solution.Therefore, small variations in pipetted volumes are ampli-fied, and may cause irregular results. The ratio of starch toLugols solution in the diastase experiments were similar tothe studies that reported an effect of potentized HgCl2.5,6
Yet, the high viscosity of the concentrated starch solutionmay have caused less exact pipetting with the 200-L tipsused in this study compared to the 3-mL glass pipettes ofBoyds team.6 The authors can therefore not exclude thatthey did not reproduce results of the former studies becauseof these technical problems.
For tests with -amylase, the optimal starch concentration(determined in pilot studies) was much lower than in the di-astase experiments. Therefore, for this set of experiments itis unlikely that starch solutions with different viscosities arethe reason that the authors results contradict the studies ofShabir and Sukul.8,9 Otherwise, as Lugols solution is definedby the ratio of iodine to potassium iodide, the absolute con-centrations may have deviated from the experiments thatfound a potency effect, in which a more concentrated solu-tion or a more active diastase may have narrowed deviations.Therefore, it is recommended to work with starch in low con-centrations and suboptimal concentrations of Lugols solutionin order to avoid the possibility of missing minor but signif-icant effects of potentized HgCl2.
Why some experimenters found differences between highHgCl2 potencies and controls, whereas others did not re-mains unclear. It is unlikely that all of them produced arti-facts, considering the strict and detailed protocols that atleast some of them observed. But it is possible that effectsof potentized solutions require experimental setups that canbe designed intentionally only after the nature of potencieshas become clear, and without this knowledge observationsof potency effects will occur seemingly at random. There-fore, focusing on a general description of an enzyme modelmay be too simple an approach. Instead, the full body of ex-perimental circumstance under which a potency effect hadbeen seen before should be considered, so that necessary co-
factors for the potency effect can be isolated. Boyds groupworked 15 years and established a long list of details to ob-serve6 before they considered their results reliable. Withoutsuch a comprehensive inclusion of circumstantial details,many independent26 replications of a high methodologicalstandard will probably eliminate most of the results pub-lished so far. However, a prolonged issue-by-issue refine-ment of model and protocol could result in a setup capableof demonstrating an effect that must be of a nature that isnot easily detected with standard procedures.
The authors found no effect of high potencies on hydro-lase activity. Demonstrating the potency effect may requireadditional experimental circumstance that protocols shouldobserve and try to isolate.
This work was funded with a grant from the Karl andVeronica Carstens-Foundation, Essen, Germany.
1. Harris P, Rees R. The prevalence of complementary and al-ternative medicine use among the general population: A sys-tematic review of the literature. Compl Ther Med 2000;8:8896.
2. Wootton JC, Sparber A. Survey of complementary and alter-native medicine: Part I. General trends and demographicgroups. J Alt Compl Med 2001;7:195208.
3. Hrtel U, Volger E. Utilization and acceptance of classicalnaturopathies and alternative therapies in GermanyResultsof a representative population study [in German]. Forsch Kom-plementrmed Klass Naturheilkd 2004;11:327334.
4. Eisenberg DM, Davis RB, Ettner SL, et al. Trends in alterna-tive medicine use in the United States, 19901997: Results ofa follow-up national survey. JAMA 1998;280:15691575.
5. Persson WM, Ginsberg AS. The action of microdoses of ho-meopathic remedies (of chemicals and plants) on the enzymesurease and diastase [in German]. Dt Z Hom 1932;11:97106.
6. Boyd WE. Biochemical and biological evidence of the activ-ity of high potencies. Br Hom J 1954;44:744.
7. Narziss L. The Malt Production Technology [in German], 7thed. Stuttgart: Enke, 1999.
8. Shabir S, Khan S, Begum S, et al. Effect of homeopathic drugson the in vitro activity of alpha amylase from human saliva.Ind J Hom Med 1996;31:9398.
9. Sukul NC, Sukul A, Sinhababu SP. Potentized Mercuric chlo-ride and Mercuric iodide enhance alpha-amylase activity invitro. Homeopathy 2002;91:217220.
10. Kleijnen J, Knipschild P, ter Riet G. Clinical trials of ho-moeopathy. BMJ 1991;302:316323.
WITT ET AL.364
11. Linde K, Clausius N, Ramirez G, et al. Are the effects of ho-moeopathy placebo effects? A meta-analysis of randomized,placebo controlled trials. Lancet 1997;350:834843.
12. Cucherat M, Haugh MC, Gooch M, et al. Evidence of clini-cal efficacy of homeopathy: A meta-analysis of clinical trials.Eur J Clin Pharmacol 2000;56:2733.
13. Dean ME. The Trials of HomeopathyOrigins, Structure andDevelopment. Essen: KVC-Verlag, 2004.
14. Anagnostatos GS, Pissis P, Viras K. Physico-chemical studyof homeopathic solutions. In: Bornoroni C, ed. Omeomed 92.First International Congress. The Homeopathic Medicine inEurope, September 2427, 1993. Physical-Chemical-Biologi-cal and Clinical Research. Urbino, Italy: Editrice Compositori,1993:7786.
15. Boiron J, Luu C. The structure of water, and its relation to theactivity mechanism of the homeopathic remedy [in French].Ann Hom Franc 1981;23:5365.
16. Conte RR, Berliocchi H, Lasne Y, et al. Theory of High Di-lutions and Experimental Aspects. Huddersfield, UK: DynsolLtd., 1996.
17. Endler PC, Schulte J, eds. Ultra High DilutionPhysiologyand Physics. Dordrecht, Boston, London: Kluwer Academic,1994.
18. Zhalko-Tytarenko O, Lednyiczky G, Topping S. Towards abiophysics of homoeopathy: I. Conceptual approach. J AdvMed 1998;11:2733.
19. Zhalko-Tytarenko O, Lednyiczky G, Topping S. Towards abiophysics of homoeopathy: II. Experimental evidence. J AdvMed 1998;11:135140.
20. Zhalko-Tytarenko O, Lednyiczky G, Topping S. Towards abiophysics of homoeopathy: III. Hydrogen bonding networkas information carrier. J Adv Med 1998;11:263275.
21. Linde K, Jonas W, Melchart D, et al. Critical review and meta-analysis of serial agitated dilution in experimental toxicology.Human Exp Toxicol 1994;13:481492.
22. Becker-Witt C, Weisshuhn TER, Ldtke R, et al. Quality as-sessment of physical research in homeopathy. J Alt Comple-ment Med 2003;9:113132.
23. Bundesminister fr Gesundheit und Soziales. Homoeopathicpharmacopoeia. [in German]. Stuttgart: Deutscher Apotheker-Verlag, 2000.
24. Steer M, Levitzki A. The metal specificity of mammalian al-pha-amylase as revealed by enzyme activity and structuralprobes. FEBS Lett 1973;31:8992.
25. Petit C, Belon P, Got R. Effect of homoeopathic dilutions onsubcellular enzymatic activity. Hum Toxicol 1989;8:125129.
26. Vickers AJ. Independent replication of pre-clinical research inhomeopathy: A systematic review. Forsch Komplementrmed1999;6:311320.
Address reprint requests to:Claudia M. Witt, M.D.
Institute for Social Medicine, Epidemiologyand Health Economics
Charit University Medical CenterD-10098 Berlin, Germany
POTENCIES AND HYDROLASE ACTIVITY 365