Fast-Acting Toxins

I have a question that’s been bugging me for a long time, and today’s column on fast-acting poisons ( ) seems the appropriate reason to air it.
In the original Edmund O’Brian movie D.O.A., O’Brien’s character is seliberately dosed with “Luminous Toxin”, which actually does glow, apparently. I would’ve thought it was something made up fo the film, but IIRC, there’s a coda at the end claiming it’s a for-real substance that will kill you within hours, as in the movie (If you haven’t seen it, do. One of the great B-movies. O’Brien tracks down his own killer. Ignore the awful remake, which uses only the basic idea and telegraphs its solution.)
I’ve been trying to learn what “luminous toxin” is ever since. Nobody else seems to call anything by that name. From the fact that it glows, I imagined it was something radioactive, like Polonium, or maybe radium salt, or something. Does anyone know what this was supposed to be?

D.O.A. is in the public domain, so you can watch it online. The end credits say “The medical facts in this motion picture are authentic. Luminous toxin is a descriptive term for an actual poison. Technical Adviser, Edward F. Dunne. M.D.” The use of the phrase “descriptive term” suggests to me that this isn’t the real name of the poison.

Speaking of luminous poisons reminds me of the glass of poisoned milk(?) that Cary Grant served Joan Fontaine in Suspicion. I don’t remember if they ever said what kind of poison that was supposed to be.

Please help me with that last paragraph.

  1. D2O interferes with cell processes and causes cell death. Doubt it. The extra neutron does not interfere with any chemical processes. I think another thread is necessary here.
  2. The idea of it costing $500 a quart. Am i missing something?? Do you think a terrorist can afford such a steep cost? [SIZE=1]sarcasm[/SIZE]

Ordinarily, it’s a pretty good approximation to say that different isotopes of the same element behave identically chemically. But in this specific case, the extra neutron produces a significant change in the mass of the atom, so the change in its chemical properties will be more significant than for other isotopes. And biological processes depend on some pretty subtle reactions, such that even such a small change can be significant. It’s still almost good enough, though, which is why the lethal dose is so large: You need to replace most of the hydrogen in your victim’s body with deuterium.

I would dispute the claim that it’s hard to detect, though… Most labs might not think to check for it, but the same is probably true for many other obscure poisons. And if they do think to check for it, it’s unmistakable, and the evidence will last forever, unlike toxic compounds which will break down with time.

1.) I figured it wasn’t the real name for the poison. But that just helps obscure the real nature of it.

2.) I haven’t seem suspicion, but I read or heard an interview with Hitchcock in which he said they used a concealed bulb to make the milk “glow”, and they did that because it made it seem more evil, not because it was supposed to be luminous. As I haven’t seen it, I have no idea what it’s supposed to be.

I am suprised that Cecil didn’t mention HCN, aka prussic acid, probably one of quickest poisons known. I was also under the impression that someone had developed an antidote for ricin recently.


The key there is if they think to check for it. Typically, when searching for toxins, they will use a mass spectrometer that is not geared towards detecting low molecular weight molecules. Instead they will use probably liquid chromatagraphy with electrospray ionization or some similar “high-pressure” type set up. (High-pressure meaning 10[sup]-6[/sup] torr rather than 10[sup]-9[/sup] torr.) This means that the low molecular weight region is saturated with atmospheric noise that would block out any direct detection of D[sub]2[/sub]O. Minor variations in the isotopic ratios of compounds derived from the D[sub]2[/sub]O are unlikely to be noticed.

My numbers are a little off the top of my head but the principle is the same. If you want I have a text on MS to get the real numbers.

Isn’t it wise that they have set dimethyl mercury as the standard for Mercury NMR? Any of those organo mercury compounds are freaking frightening.

Everyone is missing the key point WRT D2O and detection. D2O is toxic because the deuteriums will exchange with many of the protons in your cells. In some cases – most notably in catalytic proteins (enzymes) – the exchange can lead to decreased or altered function of the enzyme. At sufficiently high levels of deuterium exchange, your body will begin to have difficulty maintaining its normal physiological state and toxic effects will be observed. Studies in dogs have shown that toxic effects are observed in the 30-35% range (

Under these conditions, virtually any biological sample analysed by HPLC-MS will show unmistakable alterations in the isotopic abundances of the analyte. There would be no need to look at the ratio of H2O/D2O in the low mass range of the spectrum – the analyte signals would scream with the evidence.

BTW – high pressure in this case refers to the pressure of the liquid chromatograph: 2000-3000 PSI vs. a few PSI for a gravity or flash column. The pressure in the vacuum chamber is unrelated to the ionization type – witness the fact that atmospheric pressure ionization methods like electrospray can be interfaced with mass specs like TOFs that need very good vacuums (<10e-8 torr). And the “noise” at low mass is not the result of atmospheric signals. With the exception of water, none of the abundant components of ambient air – oxygen, nitrogen, carbon dioxide, argon – are observed in electrospray ionization MS.

How many of these protons on an enzyme would actually exchange? Over the course of time, more and more would exchange, but most of these protons on enzymes do not have water as their source as far as I know. I would guess that D[sub]2[/sub]O would get toxic before most of the protons were exchanged on the higher molecular weight molecules. Your site does not give %protons exchanged on higher weight molecules, it gives %D[sub]2[/sub]O in the system. Cholesterol for example, would only exchange one proton. Then given the use of water as an eluent, most of the exchangable protons would exchange back before they ever get near the analyzer. So you are suggesting that a difference of a few Daltons is going to be noticed out of a thousand? I don’t think your average lab tech will pick that out.

ESI-TOF may exist but it is far from standard equipment unless there have been recent advancements that I’m not aware of. ESI is an atmospheric pressure ionization technique. Getting the sample from atmospheric pressure down to 10[sup]-8[/sup] torr is no small task. I guess your statement could read “The pressure of a vacuum chamber is unrelated to vacuum type if you have an infinite amount of money.”

This could be true. Water I assume would be because it is used as an eluent.

Non-geek warning – this is going to get technical.

I should have been more specific. There is, of course, a wide variation observed in the rates of H/D exchange in biological molecules. For all practical purposes, aliphatic C-H bonds don’t exchange. OTOH, very acidic protons exchange very rapidly.

I was thinking primarily of the behavior of proteins, where the H/D exchange of the amide hydrogens is widely used as a technique for characterizing solution dynamics (see, for example, The H/D exchange rate depends – amongst other things – on pH, and it happens that at physiological pH the intrinsic exchange rate is relatively high. Thus, at 30% mole fraction D2O, one would expect that a large proportion of the amide sites in the proteome would reflect the solvent deuteration state. (Different sites exchange at different rates – giving rise to the utility of the technique for protein dynamics – so not all will exchange, but many will at pH 7.4).

For a 50 kDa protein containing 450-500 amino acids, if 50% of the sites exchanged to the compostition of the solvent, the protein would gain about 75 Da. Since intact proteins are commonly analyzed by MS to determine phosphorylation state (a shift of 80 Da), it is not unreasonable to think that a 75 Da shift is noticable.

More to the point, proteins are normally analyzed in MS by tryptic digestion followed by LC-MS/MS of the peptides, where the mass shifts would be unmistakable.

Back exchange is not an necessarily an impediment either, since LC-MS is usually performed with 0.1% formic acid in the mobile phase, resulting in a pH near the minimum H/D exchange rate. In a practical sense, I often analyze samples subjected to H/D exchange using a 20 minute gradient LC run with good results. I will concede, though, that sample cleanup performed in aqueous solution by an unsuspecting analyst might very likely result in back exchange that would remove the deuterium before it is ever observed.

I think there have been advances you are unaware of, although they are not really recent :wink: ESI-TOF has been in common use for more than ten years, and the first commercial version I can remember – the Perseptive Mariner – was introduced in 1997. At this point, several of the major MS vendors have ESI-TOF instruments (google “bruker microtof”, “agilent 6210”) in addition to the nearly ubiquitous QqTOF-style hybrid instruments like the Waters QTOF and Applied Biosystems QSTAR. While all of these instruments cost more than your car, they are not expensive in mass spec terms. It isn’t really all that hard to go from 760 torr to 10[sup]-8[/sup] torr – a couple of skimmer cones, a rotary pump and a decent sized turbomolecular pump on the front end, a small turbo at the back end, and you are good to go.

Water because it is is the only abundant component of ambient air with any reasonable gas phase basicity. Gak – there’s a phrase that takes be back a lot of years; we worked with atmospheric pressure ionization (APCI on Sciex instruments) for years before John Fenn thought of applying electrospray to mass spec. Before protein mass spec, we used the all the same technology (except the ESI emitter) to dectect odor compounds in ambient air.


You win.

Actually my understanding of mass spec is about 7 years old, but that was textbook stuff so who knows how old it is before that. I would have thought my prof. for that class would have known, since aerosol MS was his specialty but perhapse he stuck to the book. Anyway, I still bet those ESI-TOF’s are significantly more expensive than the ESI-quads are and with the ESI-quads I wouldn’t expect one to know the difference at such high-MW.

Welcome to the straightdope message board. I don’t think anyone flinched.

Somewhat of a hijack, but since the OP mentioned “glowing toxins”: If you were radioactive, would you glow?


I don’t know if you are still hanging around, but this back exchange thing has me bothered a bit. You obviously know your stuff here, so perhaps you can give me the information that I am looking for.

The question is what protons are you talking about? I made a list of the types of exchangable protons in protiens. What we need to find is a proton that is significantly less acidic than an alcohol, but more acidic than an alkyl proton. I know this from experience. I can easily make any alcohol proton dissappear by shaking it in D[sub]2[/sub]O. (easily meaning in less than a second.)

Under the conditions you mentioned for testing (0.1% formic acid) the amines are all going to be protonated. That makes amine protons so acidic that they are out of the question. (It made me a little curious that you picked 0.1% formic acid as being less susceptible to exchange.) Thiol protons are also way out being more acidic than alcohols. The guanidine derivatives are way out for the same reason as the amines.

That leaves possibly alpha carbonyl protons. I did not see any malonic ester type protons in standard amino acids, otherwise I would have thought them perfect. The trouble is at human physiology pH (~7.4), I would not expect exchange to take place quickly on these protons.

I can see where H/D exchange would tell you alot about an enzyme/protien but that is a different case because it is there and they know it is there. They can use the right ratio of deuterated solvents to detect these things.

If you don’t know that it is there you can only use H[sub]2[/sub]O as the eluent. In my experience, this type of exchange takes seconds. That means that by the time it gets to the analyzer any real increased deuterium content is gone.

Clearly you know a lot about this subject. So perhaps you can tell me what protons I’m looking for so I don’t waste my life looking up pKa/pKb values for everything. Otherwise I’m still a bit skeptical.

Yes I have been spinning this in my brain for days.

I didn’t think the remake (which starred Dennis Quaid, Meg Ryan and Daniel Stern) was awful. It wasn’t the best movie ever, I’ll grant you, but I like it better than the original.

Anyway, ISTR in the remake the poison was established to be “radium chloride.” If that’s a real compound, I assume it’s the same thing as the “radium salt” you refer to. And once Quaid’s character learned he’d been killed via spiked coffee, he looked in the bottom of his mug and there were a few drops of glowing residue.

Hope that helps.

(raises hand sheepishly) I did.

I didn’t recall, which is why I asked. My saying “radium compound” was a guess.
Radium chloride exists, but according to one source it isn’t very soluble. If so, it would make a rotten poison to put in a drink. (The CRC handbook lists it as “s” though – soluble). Nobody I’ver checked in the Internet will say anything definite about its short-range toxicity, although one site suggests it ought to be pretty toxic chemically (besides its radioactivity) by analogy with other chemicals above it in the periodic table. So it’s arguably possible that this is what they had i n mind.

Well, would it need to be soluble? If it lay undissolved in his coffee and he just didn’t notice when he drank it, wouldn’t it still be deadly?

Well, in that case:

1.) He might not actually drink it – he might not finish the glass/cup (especially if there’s gritty stuff in the bottom)

2.) If it doesn’t dissolve in his stomache acids, but just passes through his body, it won’t do him any harm.