Two-Twenty-Two

 BY DEREK LOWE

There's some interesting work being done on a disinfection technique that could have some good public health effects on the current coronavirus, on influenza spread, and on future outbreaks of both bacterial and viral diseases. As a small-molecule drug discovery guy, my thoughts naturally turn to fighting off these things through enzyme inhibitors and the like, but of course that's really hard to do. Effective compounds are not easy to come by in these areas, and the bacteria and viruses are of course constantly mutating (especially under the selection pressure of your new drugs!) So while those are valuable, an even better preventative is vaccination, when you can come up with an effective vaccine. But that's not always so easy, either - witness influenza, whose coat-protein-swapping ways keep it changing from season to season in ways that are very hard to keep up with. And of course witness the various infectious diseases for which no effective vaccine has ever been produced at all. An good vaccine is hard to beat, but good vaccines don't come easy.

What's even better than that are public health measures that are taken even more broadly and don't require individual actions. In the industrialized world, we took many of these a long time ago, things like providing clean water without pathogens floating around in it. It comes as a shock to us when that layer of defense breaks down, because we're so used to it that we take it as part of the natural order of things (which of course it's not - it takes money and effort and equipment to keep potable water coming out of the taps). Cleaner air is in the same category, and as someone who grew up in the 1960s and 1970s, I can tell you that the air in the US today is very noticeably better than what we were breathing back then, particularly in more heavily populated areas.

But air purification of viral and bacterial pathogens is not so easy. There are HEPA filters which will catch some of the pathogens, but getting the air turnover through them that's needed isn't so easy (and naturally, the filters have to be replaced at intervals), and there's ultraviolet light. The germicidal effects of short-wavelength UV have been known for a long time - it doesn't penetrate very far, but it's very effective. This technology is used for disinfection purposes in liquids (water purification and in pasteurization of things like milk and fruit juices), and for surface disinfection of fruits and vegetables, and there have been several studies of it for air purification in hospitals and other settings. These have been equivocal - those two links, for example, report significant results, but other studies have not, and fewer of the published articles have gone on to show actual reductions in infection, as opposed to reductions in the air counts of bacteria, fungi, and so on. You'd hope that that latter would connect with the former, but you have to prove these things, since a lot of seemingly straightforward ideas don't work out in practice.

The mechanism of all these disinfections is pretty straightforward: ultraviolet light is not good for living cells, and it's especially damaging to DNA and RNA as well as to unsaturated lipids (and thus ultimately to cell membranes). Some of that is direct photochemical damage, as in pyrimidine dimer formation with nucleic acids, and some of it (such as peroxidation of lipids) is downstream of the production of reactive oxygen species which do other kinds of damage as well if they overwhelm the cellular defenses against such things. All of these of course operate on pathogens, too, so the question has always been how to zap them without zapping ourselves. For air purification, there are are sorts of subtleties around the total light flux, the methods of exposing room air to it, and the various wavelengths of ultraviolet used, which can make things difficult to compare. All of those papers linked in the above paragraph use "UV-C", which is broadly defined as light of roughly 100 to 280 nm, but that's a lot of territory, and as anyone who's done photochemistry reactions can attest, different wavelength can give you different results. But it became apparent early on during our SARS-CoV-2 era that the current coronavirus was indeed inactivated by UV-C in general at several wavelengths and more work has gone into quantifying this effect.

This all leads to thoughts of large-scale disinfection of offices, meeting rooms, churches, restaurants and other public spaces, as described in this new article at The Atlantic. You'll note that it focuses on light at 222 nm, as do several of the links above. I was interested to see this described as a safe for human exposure, since as a bench scientist I've very much avoided exposure to such short wavelengths. But the reasoning makes sense: down in that range, the penetration of such light is only a few microns, and it doesn't even get past the dead cells on the surface of the skin and the cornea of the eye! Now if it did, it would certainly wreak all kinds of ultraviolet havoc, and that's what happens to bacteria and to viruses: they are small enough that they get thoroughly irradiated, and it does severe damage to them. Some of the papers above are from the Brenner group at Colombia, and they have been proposing for several years now that such far-UVC light (200-222 nm) could be deployed as an all-around germicidal method (down below 200 is impractical, because you get into the "vacuum UV" range where the light doesn't even get very far through the air before being absorbed by oxygen molecules themselves). In Japan, the Nakane group at Hirosaki University has been making the same case, and has demonstrated that indeed, 222 nm light does not induce DNA damage in mice, even at much higher flux than would be used for disinfection. That's very much as opposed to (say) 254 nm light, which starts doing damage immediately due to greater penetration into animal tissue.

The Brenner lab's experiments with other coronavirus strains in an UV-irradiated aerosol droplet chamber led them to conclude that at the current OSHA regulatory standard for continuous 222 nm light exposure that you would see 90% viral inactivation at 8 minutes and 99% inactivation in about 16 minutes. That seems promising, and I'm glad to report that there's a randomized trial underway in Nova Scotia in some long-term care facilities. They're running over two rounds of flu season using ordinary fluorescent lights along with the (invisible to the eye) UV ones as a control and looking to see if the number of respiratory infections goes down. We need more such studies, I'd say. The expectation would be that getting rid of the need to push the room air past a concealed hazardous UV fixture (as with those links in the third paragraph above) will improve things, but that has to be proven in the real world, too. We'll also need to collect more safety data, naturally, before we start bathing indoor spaces in invisible germicidal rays. But it's not at all a crazy idea, and it really deserves to be tried out thoroughly.

https://www.science.org/content/blog-post/two-twenty-two