Isotopic Resonance Hypothesis: Experimental Verification by Escherichia coli Growth Measurements

Isotopic composition of reactants affects the rates of chemical and biochemical reactions. As a rule, enrichment of heavy stable isotopes leads to progressively slower reactions. But the recent isotopic resonance hypothesis suggests that the dependence of the reaction rate upon the enrichment degree is not monotonous. Instead, at some “resonance” isotopic compositions, the kinetics increases, while at “off-resonance” compositions the same reactions progress slower. To test the predictions of this hypothesis for the elements C, H, N and O, we designed a precise (standard error ±0.05%) experiment that measures the parameters of bacterial growth in minimal media with varying isotopic composition. A number of predicted resonance conditions were tested, with significant enhancements in kinetics discovered at these conditions. The combined statistics extremely strongly supports the validity of the isotopic resonance phenomenon (p ≪ 10−15). This phenomenon has numerous implications for the origin of life studies and astrobiology, and possible applications in agriculture, biotechnology, medicine, chemistry and other areas.


Appendix 1. Early studies of unusual biological effects of small concentrations of deuterium.
The unusual biological effects of dilute heavy water were first reported by a Yale researcher T.
C. Barnes. In his experiments, mass cultures of Spirogyra exhibited much less abscission or cell disjunction (sign of growing at favourable conditions) and greater longevity in 0.06 % D water compared to ordinary water 19 . The positive effect of low deuterium concentration was further demonstrated in experiments confirming an increased longevity in Spirogyra 20 . In flatworms, longevity in the heavy water media appeared at deuterium concentrations of 0.06% and 0.07%, while at 0.13% and up to 0.47% D the effect became progressively obscure 21 .
Increased cell division was observed in Euglena kept for forty-five days in 0.06 % heavy water 22 . Richards has repeated Barnes' tests with yeast and confirmed that a slight excess of deuterium in water is biologically significant. He observed a 26% increase in dry weight of yeast grown in 0.06% D water 23 . The strongest effect of 0.06% deuterium was observed during fast cellular proliferation 24 . The flatworm Phagocatagracilis left in normal water gradually shrank after a few months time to one-fifth or less of their original body size while in those left in 0.06 %D water showed only a slight diminution 21 . Lockemann and Leunig studied the effect of heavy water with ≤0.54% D upon Escherichia coli and Pseudomonasa eruginosa. Concentrations of as low as 0.04% D were found to favour survival at adverse conditions 25 .
Not all groups have confirmed the effect of diluted heavy water. Macht and Davis reported no difference between the growth in ordinary water and that with 0.06% D 26 . In fact, their data did imply a 10% faster growth, but the result was not statistically significant due to a large experimental uncertainty. Curry et al. repeated the experiments by Barnes and others but could not confirm the previously reported effects of dilute heavy water 27 . In general, early reports on the effects of diluted heavy water can be criticized for insufficient statistics and the small number of concentration points (often just two) used in the studies.
In 1970s, Lobyshev et al. have realized the deficiencies of previous efforts and studied the effects of low deuterium concentrations on biological and biochemical systems with much greater rigor. They found that the Na,K-ATPase activity increases at low deuterium concentrations, reaching maximum (+50% compared to normal water) at 0.04-0.05% D 14,15 .
They also studied regeneration of hydroid polyps Obelia geniculata in a wide range of D 2 O added to sea water. They found strong inhibition at high deuterium concentrations, as well as activation of regeneration by small (≤0.1%) deuterium concentrations 28 .
In 1990s, Somlyai et al. have also shown that 0.06% deuterium in tissue culture activated the growth of L 929 fibroblast cell lines 29 . In contrast, water with deuterium depleted below the normal levels suppressed fast-growing cells 29-31 . Sample preparation workflow is shown in Figure S5.  Figure S6 33 ). Second, 30 µL (151 µL for 50% of 15 N) aliquot of stock A was dispensed into each "sample" well on plate P A (marked with purple in Figure S6) to prepare 32 replicates of sample S A . Third, 30 µL (151 µL for 50% of 15 N) aliquot of stock B was dispensed to each "sample" well on plate P B (marked with blue in Figure S6), resulting in standards (0.37% 15 N) was added into each "standard" well on plate P A and P B (marked with yellow in Figure S6) to prepare 32 reference standards on each plate. Finally, 270 µL (149 µL for 50% of 15 N) of the diluted E. coli culture was dispensed into each well except blank wells.

Chemicals and materials
In total, 32 replicates pairs of "sample" and "standard" wells were prepared on each plate. M9 minimal media depleted with 13 C (0.1% 13 C, 99.9% 12 C) were prepared by dissolving 500.00 mg of D-glucose (0.1% 13 C, 99.9% 12 C) in 100.0 g carbon-free M9 minimal media solution, followed by filtering with a 250 mL vacuum filtration system through a 0.2 µm PES membrane.

Preparation of E. coli sample on honeycomb well plate
Sample preparation workflow is shown in Figure S5. In each experiment, three stock solutions were used. Stock A for preparing sample S A , and stock B for preparing sample S B were obtained by mixing M9 minimal media at normal isotopic condition (1.1% of 13 C) with 13 C depleted minimal media (0.1% 13 C) or 13 C enriched M9 minimal media (99% of 13 C) at certain ratio (Table 2 and Table 3). M9 minimal media at normal isotopic condition (1.1% of 13 C) were used to prepare stock solution of standard A and standard B. The final solutions were dispensed into the honeycomb well plates using the Tecan robot or pipettes.
To test E. coli growth at 0.1-1.1% 13 C, stock solutions were prepared according to Table 2.
A 40 µL aliquot of the incubated E. coli culture (O.D. ≈ 0.2) was diluted in 10 mL M9 minimal media (normal) to prepare diluted E. coli culture. To minimize the cost of 13 C depleted media, pipettes were used for part of the sample preparation here. First, 400 µL M9 minimal media without bacteria were introduced into each of the border wells on both plates A and B to serve as blanks (no color code in Figure S6). Second, 360 µL aliquot of stock A was dispensed manually by pipette into each "sample" well on plate P A (marked with purple in Figure S6) to prepare 32 replicates of sample S A . Third, 360 µL aliquot of stock B was dispensed to each "sample" well on plate P B (marked with blue in Figure S6), resulting in 32 replicates of sample B. For the next step, 360 µL aliquot of stock solution of standards (1.1% 13 C) was added into each "standard" well on plate P A and P B (marked with yellow in Figure   S6) to prepare 32 reference standards on each plate. Finally, 40 µL of the diluted E. coli culture was dispensed into each well except blank wells with robot. In total, 32 replicates pairs of "sample" and "standard" wells were prepared on each plate.
To test E. coli growth at 3-13% 13 C, stock solutions were prepared according to  Figure S6). Second, 40 µL aliquot of stock A was dispensed into each "sample" well on plate P A (marked with purple in Figure S6) to prepare 32 replicates of sample S A . Third, 40 µL aliquot of stock B was dispensed to each "sample" well on plate P B (marked with blue in Figure S6), resulting in 32 replicates of sample B. For the next step, 40 µL aliquot of stock solution of standards (1.1% 13 C) was added into each "standard" well on plate P A and P B (marked with yellow in Figure S6) to prepare 32 reference standards on each plate. Finally, 260 µL of the diluted E. coli culture was dispensed into each well except blank wells. In total, 32 replicates pairs of "sample" and "standard" wells were prepared on each plate. µL M9 minimal media without bacteria were introduced with robot into each of the border wells ("edge cells") on both plates P A and P B (72 wells in total) to serve as blank samples (no color code in Figure S6) with robot. Second, 100 µL aliquot of stock A was dispensed into each "sample" well (marked with purple on plate P A in Figure S6) to prepare 32 replicates of sample S A manually by pipette. Third, 100 aliquot of stock solution of standard A was added manually by pipette into each "standard" well (marked with yellow on plate P A in Figure S6) to prepare 32 reference standards on plate. In the same way, wells were filled on plate P B .
Finally, 200 µL of the diluted E. coli culture was dispensed into each well except blank wells with robot. In total, 32 replicate pairs of "sample" and "standard" wells were prepared on each plate.  Figure S6) with Robot. Second, 100 µL aliquot of stock A was dispensed into each "sample" well (marked with purple on plate P A in Figure S6) to prepare 32 replicates of sample S A manually by pipette. Third, 100 aliquot of stock solution of standards was added manually by pipette into each "standard" well (marked with yellow on plate P A in Figure S6) to prepare 32 reference standards on plate. Finally, 200 µL of the diluted E. coli culture was dispensed into each well except blank wells with the robot. In total, 32 replicate pairs of "sample" and "standard" wells were prepared on the plate.

E. coli growth measurements
After sample preparation on the honeycomb well plate, E. coli concentration in each well was continuously monitored by measuring turbidity (with wide band filter 420-580 nm) using Bioscreen C instrument with continuous shaking at 39 o C. Turbidity was sampled every six minutes and was monitored for ca. 22 hours to obtain a raw growth curve.

Data analysis
Data analysis 33 was performed with Excel software as described in reference 33.
Using Microsoft Excel, the logarithm of turbidity was plotted against time. The slope for every 8-h interval was calculated, and the maximum value was determined. The extrapolation of the line with maximum slope to the background level of turbidity gave the lag time. The maximum turbidity for each replicate was taken as the maximum density. The obtained three values for each growth curve were treated in the same way as below.
For each "sample" A, the obtained value was normalized by that of the "standard" B.
To minimize the influence of nonstatistical outliers that could arise due to gross errors in sample preparation and handling (e.g. differences in the geometry of the honeycomb wells, position-dependent sensitivity of the BioScreen C detector, etc.), the 32 replicates were divided into 4 groups according to their positions on the honeycomb well plate (group 1: columns 1 and 2; …, group 4: columns 7 and 8). In each group, the median of the eight values was calculated and then the four medians were averaged to obtain the value for a given plate and its standard deviation.
Altogether, seven independent 32-replicate experiments were performed for each 15 Figure S1. Growth parameters of E. coli at 50% of 15 N in M9 minimal media, compared to normal isotopic conditions.