Radiation Hormesis 2

Hormesis and Nuclear Power Plant Release Dose Rate Profiles

Hormesis is the theory that too little radiation is bad for you. Here's one theory on how it works. Normal metabolic processes produce about 11,600 potentially DNA damaging, Reactive Oxygen Species (ROS) per cell per second. This is essentially a steady state process. Radiation also produces ROS. 1 mGy/year produces on average 200 ROS per cell per year.\cite{feinendegen-2002} A trifle compared with the metabolic ROS.

But here's the difference. When you do the numbers, at 1 mGy/y, each cell gets hit with a radioactive particle only a couple of times a year. Each hit creates a microburst of about 100 ROS per cell. That microburst lasts less than a microsecond. During that microburst, the ROS production rate is 6000 times larger than the steady state metabolic rate. The theory is that evolution has used this sharp burst a handful of times a year to trigger our ROS scavenging, DNA repair, and damaged cell killing systems.

I find this unproven theory plausible. Evolution has a way of making lemonade out of lemons, taking a phenomenon which should be detrimental and turning it into a benefit. It is undisputed that ROS production is a double edge sword. On the one hand, the ROS damage DNA. On the other hand, they announce their presence and trigger an array of defense mechanisms. This scheme maximizes the signaling side while creating nearly negligible damage.

There's a lot to like about this idea. It explains why organisms do very poorly when they are placed in an artificially low radiation environment, for example, a lead box deep in a mine.\cite{luckey-2007} It could explain why humans flock to high radiation spas and beaches. Whatever the actual health benefits, something makes them feel better. Even Byers swore by Radithor before it killed him.

But how far can we push this? Metabolic processes produce 10 to 50 Double Strand Breaks(DSB) per day.\cite{vilenchik-2003, costes-2021, bouwman-2018} Radiation creates DSB's at the rate of about 0.04 DSB's/mGy.\cite{white-2016, bissell-2011} So at 1 mGy/year the radiation damage is less than 0.04 / (365.0 * 10) = 1.1e-5 the metabolic damage. Completely trivial. At 1 mGy/d, we have 0.04 /10 = 0.004 radiation DSB's per metabolic DSB's. Still negligible compared with metabolic damage. But at 1 mGy/d, we are triggering the repair response in each cell, twice a day. How much do we gain by triggering powerful chemical processes hundreds of times more than normal? At what point is this like holding the starter button down after the engine has started?

In a nuclear power plant release, just about all members of the public will see a nearly step jump in dose rate, followed by a rapid decline for about 50 days due to I-131 decay, and then a multi-decade long period of slightly elevated dose rates mainly due to Cs-137. Figure 1 shows the reconstructed air dose rate profile for one group after Fukushima assuming no evacuation. All nuclear plant release public dose profiles will have roughly the same shape.

Figure 1. Example individual NPP release dose rate profile

If we accept the microburst theory, the initial jump in dose rate will trigger a panoply of repair/defense mechanisms; but it seems unreasonable to assume that after that initial response repeated triggering will result in much net benefit. Soon or later the damage will overwhelm the signaling.

The Warsaw Model

In the last post, I claimed no hormetian has stepped forward with a well defined harm model, that recognizes the importance of repair time. This is not quite true. Professor Fornalski's group in Poland is working on the Warsaw Model. The Warsaw Model exposes a 3-D array of cells to a dose rate profile through time. It keeps track of each cell's status: healthy, damaged, repaired, mutated, cancerous, dead. It attempt to model metabolic damage, radiation damage, repair including hormetic response, and cell killing.

Currently, the Warsaw Model is limited in the type of dose rate profiles it can accept. But it can run a single spike and a constant dose rate. When the model is exercised on a single spike, with and without the hormetic response, it shows a strong but rather short lived reduction in unrepaired cells, Figure 2. But when the model is run on the constant dose rate profile, the difference nearly disappears, Figure 3. We reach a sort of equilibrium very quickly.

Figure 2. Warsaw Model: Single Spike. Ratio of Damaged Cells without/with Hormetic Response. If the blue dots are below 1.00, less damaged cells with hormetic response than without.

Figure 3. Warsaw Model: Flat Profile. Ratio of Damaged Cells without/with Hormetic Response. Vertical and horizontal scales different from Figure 2.

There are situations in which hormesis can be important. A nuclear power plant release is not one of them.

It's important to recognize that the Warsaw Model does not claim that radiation is good for you. Nor does it claim that there is a threshold dose rate below which there is absolutely zero harm. The adaptive response in the Warsaw Model reduces the number of unrepaired cells from what it would have been without the response. The effect is to shift the harm curve to the right; Figure 4.

Figure 4. Adaptive Response per Warsaw Model, reference \cite{fornalski-2016}

Note also the Warsaw Model ends up with an S-shaped curve. If the Warsaw Model claims the cancer incidence curve is sigmoid in dose rate, we can short circuit the process by fitting a sigmoid curve to the cancer incidence data we have, starting with the RERF data base. That is exactly what the SNT model does.