A new article from The Lancet reports the synthesis of an antibiotic that not only kills all drug-resistant bacteria that have been tested (in vitro or in vivo in mice), but also seems impervious to being rendered ineffective by the evolution of bacteria. If this works out in humans, it would be a terrific advance in medicine: an antibiotic that can be used on people whose infections aren’t treatable because the bacteria are drug-resistant (this is common in TB, for instance), but also an antibiotic that seems to be impervious to the evolution of bacterial resistance. It’s also easy to synthesize using organic chemistry, and can be easily tweaked in its structure just in case some bacteria eventually do develop resistance.
Click on the screenshot below to read the original article, or download the pdf here. The researchers hail mostly from UC Santa Barbara, but also come from UC Davis, Singapore, and Australia.
I found the link from reader Jeannie, who sent me a short piece on kottke.org calling attention to the results. That linked to the original paper (above), but added this:
The discovery was serendipitous. The U.S. Army had a pressing need to charge cell phones while in the field — essential for soldier survival. Because bacteria are miniature power plants, compounds were designed by Bazan’s group to harness bacterial energy as a “‘microbial”’ battery. Later the idea arose to re-purpose these compounds as potential antibiotics.
“When asked to determine if the chemical compounds could serve as antibiotics, we thought they would be highly toxic to human cells similar to bleach,” said Mahan, the project lead investigator. “Most were toxic — but one was not — and it could kill every bacterial pathogen we tested.”
Such are the totally unexpected side effects of research, and although this was applied research, it’s also a justification for pure research. Remember: the whole apparatus for sequencing DNA, and then the CRISPR technique, gene editing, and so on, began with the simple observation that some bacteria live in hot springs near the boiling point, and some curious scientists who asked “I wonder how they do it.”
Back to the paper. My take will be short as it’s complicated and, to be sure, parts of it are beyond me. The compound they found was COE2-2hexyl, and below is the structure. It can be synthesized without much difficulty using standard methods of organic chemistry, so you don’t have to get it from massive quantities of fungi or other organisms. Captions are from the paper.
Asyou see, it consists of two aromatic central chains with four long carbon tails, each of which has a nitrogen atom in it:
It works, as implied in the caption above, by attacking the cell membranes of bacteria, disrupting essential functions of the membrane. These include the ability of the bacterium to absorb and emit cell contents, to help the bacterium metabolize, and also, critically, to divide. Here’s how COE2-2hexyl looks when it gets into the bacterial membrane and disrupts it. Other similar compounds, called COEs, have the same shape and do the same thing:
This compound was tested on 17 bacterial isolates taken from people with drug-resistant infections:
You’ll recognize some of these as bacteria that cause gonorrhea, tuberculosis, pneumonia, dysentery (Shigella flexneri) Acinetobacter baumannii, which causes bad infections associated with hospitals, and so on. All of these were isolates taken from humans who had shown antibiotic resistance. The drugs were tested in vitro, using mouse cell cultures that were infected with bacterial isolates from humans, and also in vivo, in mice that had been infected (there were of course controls that had been mock-treated). (I have to add that I feel sorry for the mice in the control group.)
As the authors note (my emphasis):
Expanded antibacterial activity analyses revealed that COE2-2hexyl exhibited broad antibacterial activity against all 17 clinical bacterial isolates tested (Table 1). Notably, methicillin-resistant S. aureus (MRSA, MT3302) and CRE K. pneumoniae (MT3325) were derived from sepsis patients with refractory bacteremia, whereby the CRE organism was resistant to 20/22 antibiotics determined by clinical VITEK testing (bioMerieux, Inc.) and 19/24 antibiotics determined by broth microdilution.
Note that it worked when nearly all antibiotics had failed. The authors also made two derivatives of this compound, adding one cabon atom to two of the four chains; these two compounds also showed antibacterial activity against 9 drug-resistant bacterial isolates tested.
Of course you’re wondering “well, this is great, but is it safe?” It was, even in higher doses, and in the doses that killed bacteria. But of course it was safe in mice but may not be in humans. Clinical testing will be in order, and that might take a long time before we see if this and its derivatives are truly “wonder drugs”.
Finally, testing the compound for relatively long periods against bacteria showed that the bacteria did NOT develop immunity to the drug (that’s via natural selection, of course), which is really good news, since few antibiotics have not been overcome by mutations that render bacteria immune to them. (I believe that the Streptococcus bacterium that causes “strep throat” has never evolved resistance to penicillin, nor has the polio virus evolved immunity to polio vaccines, but such cases are rare.) Now, as Orgel’s Second Rule states, “evolution is cleverer than you are,” and eventually, if COE2-2hexyl is used for long enough, bacteria might find a way around it. But right now, things look promising.
Finally why are bacteria unable to evolve resistance to COE2-2hexyl? The clue is in this sentence in the paper:
COE2-2hexyl had specific effects on multiple membrane-associated functions that may act together to disrupt bacterial cell viability and the evolution of drug-resistance through a mechanism of action distinct from most membrane disrupting antimicrobials or detergents which destabilize.
The compound, it seems, disrupts many different functions of the bacterial membrane, and while one disruption might be fixed by one or more mutations in the bacteria, something that screws up your system big time, and in multiple ways, may be impossible to repair, as bacterial mutations that overcome one disruption may make it harder to fix the other disruptions. The more ways a drug can screw up a bug, the less likely it is that the bug can evolve resistance. But remember—bacteria are clever.
I’ll end with the authors’ final paragraph about what’s good about this compound and what needs to be done (mostly efficacy and safety testint in humans). But if this thing works out, it will be a medical advance of almost unparalleled value (bolding is mine):
The ease of molecular design and modular nature of COEs offer many advantages over conventional antimicrobials due to their intermediate molecular size, sufficient aqueous solubility to achieve efficacy, and the absence of complex chemical structures/chiral centers, making synthesis simple, scalable and affordable. The COE refinement workflow potentially accelerates lead-compound optimization by more rapid screening of novel compounds for the iterative directed-design process. It also reduces the time and cost of subsequent biophysical characterization, medicinal chemistry and bioassays, ultimately facilitating the discovery of novel compounds with improved pharmacological properties. Additionally, COEs provide an approach to gain new insights into microbial physiology, including membrane structure/function and mechanism of drug action/resistance, while also generating a suite of tools that enable the modulation of bacterial and mammalian membranes for scientific or manufacturing uses. Notably, further COE safety and efficacy studies will need to be conducted on a larger scale to ensure adequate understanding of the clinical benefits and risks to assure clinical efficacy and toxicity before COEs can be added to the therapeutic armamentarium. Despite these limitations, the modular design of COEs enables the construction of a spectrum of compounds with the potential as a new versatile therapy for the emergence and rapid global spread of pathogens that are resistant to all, or nearly all, existing antimicrobial medicines.
34 thoughts on “A “miracle antibiotic” in the offing? New compound works against all multiple-drug-resistant bacterial strains, and also stymies the evolution of bacterial resistance”
Is this Gram negative and Gram positive bacteria then?
BTW I’d subscribe if the button works but it hasn’t for a long time.
Both. I’m working on the subscribe issue.
Since it works on gram-positive and gram-negative bacteria, they should call it AllGramCoe Universal antibiotic.
Ok I found it – hope nobody minds if I copy/paste the list from the paper – maybe one might find their favorites – I tried to improve the readability with carriage-returns :
Gram-negative bacterial isolates included:
A. baumannii ATCC 19606
A. baumanii ATCC 17978
E. coli DH5α
E. coli ATCC 25922
E. coli MG1655
E. coli BW25113
E. coli BW25113 ΔmutL::kan,
K. pneumoniae ATCC 13883 CRE
K. pneumoniae (MT3325), derived from a urinary/sepsis patient obtained from Santa Barbara Cottage Hospital (2017)
N. gonorrhoeae ATCC 700825
N. gonorrhoeae ATCC 49226
S. flexneri ATCC 29903
P. aeruginosa ATCC 10145
S. enterica serovar Typhimurium ATCC 14028
Y. pseudotuberculosis (YPIII).
Gram-positive clinical isolates included methicillin-resistant (MRSA) and -sensitive (MSSA) S. aureus: CA-MRSA USA300, MSSA Newman and 3 isolates derived from sepsis patients obtained from Santa Barbara Cottage Hospital (2016) termed MRSA Blood (MT3302); MRSA Wound (MT3315); MSSA Blood (MT3305).
24 S. pneumoniae clinical isolates included D39 (ser. 2) and Daw 1 (ser. 6).
Very interesting. If it pans out in clinical trials then it would compete with RNA based vaccines as the greatest biomedical revolution in recent history. Btw, the technology that goes into the Covid RNA vaccines is an absolute marvel of invention, much of which came from basic research. I actually got choked up reading about how clever those things are, as nicely explained here: https://berthub.eu/articles/posts/reverse-engineering-source-code-of-the-biontech-pfizer-vaccine/
I would argue it’s more important than mRNA vaccines. If it pans out, it seems like it is on the same sort of level as the original discovery on penicillin.
Fascinating and promising! Especially interesting is that it disrupts bacterial activity in multiple ways. This doesn’t preclude the evolution of resistance, but it might require that resistance arise in steps, extending the usefulness of the compound. I’m excited to learn more as the research proceeds.
The question is, what do multi-drug efflux pumps do with it. Perhaps nothing because it is in the …
Is this new compound in the inner membrane? I guess so…
there’s a great diagram of the outer membrane, inter membrane space, and inner membrane in this paper :
this figure :
ahem … periplasmic space.
I’ll try to cool it with commenting after this one :
Gram negative cell wall :
Gram positive cell wall :
… and a note that mycolic acids form a waxy layer on Mycobacterium tuberculosis – which needs acid fast staining because the Gram stain give inconclusive results ok I’ll stop.
The first of a novel class of antibiotics would be expected to destroy multi antibiotic-resistant bacteria with ease. When the the quinolone ciprofloxacin came out, it killed both gram-negative and gram-positive drug-resistant pathogens and worked on DNA gyrase, which seemed fortuitous. There was a case report of the drug “curing” the in vivo production of a resistance plasmid. For several years the only drug resistance seen was in individual patients who had taken the drug.
Nevertheless, that changed and quinolone resistance became wide-spread. It would be great if this new class of antibiotics does not allow resistance, but one has to be skeptical of the idea. Resistance has developed over time to every class of antibiotic – even to chemical sterilizers like phenol. Fingers crossed.
Here I am come to write much the same thing! Ever since antibiotic resistance became a problem (can you believe Staph. aureus used to be sensitive to simple peniciliin G? Not for long!) drug companies have been running out the canard that ‘this one will not cause resistance.’ All I can say is that if this comes to market I hope its use is restricted to those infections where it is needed and nothing else will do. And we had better keep it out of the hands of vets and the animal feed industry.
Does it kill good bacteria (assuming there are any) as well?
What is difference between bacterial cell membranes and other creature cell membranes (such as ours)?
Easy to make; but if it pans out, how much will it sell for?
I note rhat Treponema pallidum (the helical bacterium that causes Syphilis) never became really resistent to penicillin, although the doses used are much higher nowadays.
I’m sure that bacteria will sooner or later evolve to cope with these compounds, let us hope rather later than sooner. The observation that a mutation improves one angle at the same time makes it more vulnerable to other angles gives hope.
I wonder if these compounds will not wreak havoc on our mutualistic intestinal microbiote. Well classical antibiotics often do snd may lead to pseudomembranous cilitis (which can be deadly). Luckily we have probiotics now, and poop transplant.
If this works out in Human use they’ll deserve a Nobel Prize!
The large, “greasy” and multiply charged structure and the fact that they’re focusing on sepsis makes me wonder about the bioavailability of compounds like this (ie. can these be administered orally or IV and distribute through tissues to the point of infection). Just looking at the molecule, it doesn’t strike me as one that will be very bioavailable or useful for infections requiring antibiotics with wide distribution in the body, like skin and soft tissue infections etc. It looks like a detergent molecule.
I’d love to be wrong here though…
If this is useful as a new sepsis drug that can be directly administered IV safely, that will also be a great breakthrough.
Back in the day when I ran an antibacterial discovery group, the medicinal chemists would have laughed if I suggested working on this class of compounds. It will be interesting to see if this moves into clinical trials.
Many thanks, PCC(e), for informing us about this very important research in practical molecular biology. This research at U. Cal. Santa Barbara is supported both by NIH and
by the U.S. Army Research Office. The latter connection has been noticed, with shrill
disapproval, by “Progressive” activists: https://www.laprogressive.com/climate-change-2/ucsb-military-research. The screed at the link reads in part:
“Amid the towering Eucalyptus and the beautiful lagoon that frame UCSB’s Nobel Prize-winning campus, visionary social science professors plan conferences on climate justice, while others with doctorate degrees in engineering, chemistry, math, psychology, biology and computer science oversee teams of graduate students laboring for the Pentagon and its private military contractors who profit from an escalating arms race.
Yes, distinguished professors are collaborating with the U.S. Army to prepare for doomsday wars with Russia and China as the Pentagon rushes weapons to Ukraine and sends nuclear-powered submarines into the South China Sea.
Let’s organize and mobilize to demand UCSB sever ties with the war machine to invest, instead, in peace and climate sustainability. “
I am trying to get my head around this. This doesn’t affect the cell wall but the cell membranes. Now my knowledge of bacterial cell membranes is rusty but digging around I found this list of functions of bacterial membranes.
1. Osmotic or permeability barrier
2. Location of transport systems for specific solutes (nutrients and ions)
3. Energy generating functions, involving respiratory and photosynthetic electron transport systems, establishment of proton motive force, and transmembranous, ATP-synthesizing ATPase
4. Synthesis of membrane lipids (including lipopolysaccharide in Gram-negative cells)
5. Synthesis of murein (cell wall peptidoglycan)
6. Assembly and secretion of extracytoplasmic proteins
7. Coordination of DNA replication and segregation with septum formation and cell division
8. Chemotaxis (both motility per se and sensing functions)
9. Location of specialized enzyme system
Disrupting one of those would be bad enough but all of them would be catastrophic.
One big difference between prokaryotic and eukaryotic membranes is that the latter contain sterols whereas the former don’t – and that includes mitochondria.
I wonder ( wet finger in the air) if that prevents the same action in eukaryotes?
I work in antibiotic susceptibility. The claims in this paper are pretty outsized, and it’s unlikely these compounds will significantly impact antimicrobial susceptibility. Hate to be the killjoy. Our lead biologist had this to say (post if you want):
“This class of antibiotics has been around for a while. The claim of “does not evoke resistance” is pretty dubious in light of the ubiquitous efflux pump AcrB providing elevated resistance. They are similar to polymyxins, for which “full resistance” seems impossible, but they have limited solubility and must be administered via IV. The paper claims there is limited toxicity to mammalian cells relative to polymyxins, but there are no pharmacokinetic studies (which would tell us how long it takes the body to “clear” these compounds after dosing). Membrane-targeting antibiotics have a poor track record in the clinic, which is probably why these compounds have not received any industry adoption despite the initial reports 13 years ago.
The real promise of these antibiotics is the ability to produce new variations rapidly and cheaply in response to emerging resistance. The problem is that each variation would need to pass through the same extensive testing and regulatory processes.”
I found that there is already a paper out describing resistance to _similar_ compounds in Enterococcus faecalis.
“… in light of the ubiquitous efflux pump AcrB providing elevated resistance.”
My thought exactly – I was puzzled as to precisely where this compound liked to settle in the cell wall and how the pumps were disposed relative to that.
IOW it’s game over for that compound – it’s gettin’ pumped the hell out….?
These pumps are one of the main reasons Gram negative bacteria are so effective evolving resistance mechanisms.
Just a note for the other lay people on here who find this discussion difficult. I am actually able to follow the general thrust of the discussion thanks to a book I am just finishing: The Song of the Cell: An Exploration of Medicine and the New Human
Book by Siddhartha Mukherjee. Highly recommend it.
Mukherjee is an excellent communicator. I liked his “The Emperor of all Maladies,” about cancer.
None of the bacteria tested against this new antibiotic cause tuberculosis. Y. pseudotuberculosis is a species of Yersinia that causes lesions in lymph nodes that superficially resemble TB but the bacterium, and the disease, are not at all related to Mycobacterium tuberculosis. A single drug that could treat drug-resistant TB would be big news but this is not it, at least not according to this report.
A poster asked upthread if this antibiotic would kill off “good” bacteria if there are any. Indeed there are: trillions of them living in your gut. A problem we see in people with fatal diseases treated with a succession of ever more powerful antibiotics for infections that their immune systems can’t eradicate is superinfection with various fungi. This event often heralds death, albeit often due to the underlying fatal disease or various other complications of treatment, often not sepsis per se.
Women will be familiar with yeast infections from even short courses of “broad-spectrum” antibiotics that we have now. It’s not a “bug-drug” interaction. It’s a “drug-person-microbiome” interaction that leads to this.
I’ll read the article but KD33 @15 likely gets the prize.
Interesting mechanism of action but this compound would have a long way to go before it became a drug. I have some cold water for you all – there’s no way this class will be orally bioavailable so it would be given as repeated doses by IV administration. OK if you are hospitalized with a life-threatening infection but otherwise, a major hassle. This is belied by them dosing the compound in mice by the IV or IP route (injection into their peritoneal space). Some decent news is that they appear pretty potent and treating infected mice at 2mg/kg is a fairly low dose and just once a day implyong that mouse pharmacokinetics are good enough for it to work. Bad news is, that’s close to its solubility limit so much more work to do formulating it so, not good. The next step would be to model in other animals and cells what the human drug clearance will be. If this compound gets significantly metabolized by human liver enzymes, much work would be required to fix these showstopper issues and I’d argue that given it’s structure, that would be highly improbable.
Very early tests but that’s OK. You have to start somewhere. The drug is active in vitro against a wide variety of bacteria that cause infection in people including two highly resistant ones. (In a real test of spectrum, you would test it against dozens of individual isolates of each species, not just one or two of each that they did here.) It prevented death (in mice) from infection with the two highly resistant strains and it didn’t show obvious acute toxicity in cell lines or in mice. Whether this interesting molecule can be made into a drug that can be given safely to people is a long shot. We were so very very lucky with penicillin. The only challenge there was making enough of it.
The resistance question is interesting from a natural selection viewpoint. In many cases it can be shown that the genes for the resistance mechanism — there is a bewildering variety of mechanisms — were present in small proportions of stored cultures that were put into the freezers before the antibiotic in question was invented. Not always (or even commonly) does the strain need to wait to get lucky as a new mutation arises that happens to produce resistance. These low-frequency genes are selected for when the population is exposed to the relevant antibiotic, often during treatment of a single patient. (I’m eliding a lot here.) But when the selection pressure is removed, the population does not typically revert to wild type, suggesting that most of these mutations confer not just a relative advantage which is lost when the antibiotic is removed from the hospital, but an absolute advantage that maintains them at high prevalence in the strain. Bacteria do not seem to pay for the privilege of being resistant to antibiotics. Thus there is a ratchet effect.
This might be important for this new compound. Its appeal is that it might take several mutations instead of just one for full resistance to develop and the time to accumulate them all ought to slow down the inevitable acquisition of the full suite. But all these resistance genes from different individual cells could get recombined into one transposable element like a plasmid or a phage virus. Then if one lucky bacterial cell picks up the plasmid through conjugation or phage transfection, Bingo: instant resistance that will rapidly replace the wild sensitive type….provided it has not lost overall fitness in doing so.
Jerry notes that this is one of the prizes of pure research. You never know where it will lead and you can’t direct it to some end. It takes a prepared mind to notice something like this and then formulate a testable hypothesis.
This is great and hopeful news, and I don’t want to be seen as taking away from it, but . . .
I’m absolutely certain that bacteria can eventually evolve defenses. All we know is that they couldn’t within the time frame of the testing. Yes, the fact that the antibiotic kills in multiple ways will make resistance more difficult and slower to develop. But develop it will.
What I don’t much care for here are people who are absolutely certain they’re right. Did you read my post?
The bacteria that causes strep throat has not evolved resistance to penicillin (still the drug of choice) after decades and decades. Nor have the measles or polio virus evolved resistance to the vaccine in many years. See pp.131-132 of Why Evolution is True. In light of those observations, I am baffled by your “absolute certainty”. Can you predict the future?
Now, would you like to take back your “absolute certainty”?
To be fair, the streptococcus is the exception rather than the rule. And even the humble strep is slowly learning: I’m old enough to remember when I could treat lobar pneumonia or pneumococcal meningitis with simple benzyl penicillin (the original “penicillin G”) but you wouldn’t dare try that these days.
No one can have ‘absolute certainty’ about this, but all we know of bacterial evolution (and you probably know more than most of us) tells us it is likely that given enough generations some mutation will arise that gives some protection against these new-ish drugs. Polymyxins have been around for a long time, and you probably have some at home in the form of one of the constituents of Polysporin/Neosporin (formulation varies from country to country). Let us hope we are wrong (I would be delighted to be wrong!) to warn against believing all claims at face value, but we have seen these claims before and it did not end well.
“Can you predict the future?”
I’d like to point out a brilliant distinction I found from David Deutsch’s books – between prediction and prophecy. Was it in The Beginning of Infinity? Not sure…
I’m not arguing with anyone here – I’m only highlighting this important distinction that sometimes gets lost in the commotion – I also think it is very interesting!
Another curiosity is that Neisseria meningitidis which causes one form of meningitis that can occur in epidemics (but also in single cases) has never acquired resistance to penicillin either. (Because even with prompt treatment in healthy young people the case-fatality rate is many times higher than, say, Covid, prevention through vaccination is now the way to go. I should add, since this new antibiotic was developed through military-sponsored research, much of the success of meningitis vaccination comes from research sponsored by the United States Army in whose recruits this had been a serious problem for over a century.)
This is especially remarkable when you consider that N. gonorrhoea acquired a penicillin-resistance plasmid in the 1960s—numerous other resistance determinants soon followed—but has never transmitted any of them to the meningococcus even though other neisserias have picked it up. And it’s not because of anatomic separation. Gonorrhoea of the pharynx is common and the meningococcus is often found not breaching the peace in the nasopharynx of healthy children and young adults….one of many mysteries of how we get along with our bacteria.
This is not to say that strep and meningococcus never acquire resistance to any antibiotics, just that resistance to the drug of first choice has never been observed which is fortunate for us.
This is in contrast to plasmids carrying genes for antibiotic resistance in other ecological niches which are highly promiscuous with little care for genetic relatedness of the donor and recipient bacteria.
Of course nothing is certain about the future. But the dice are loaded against the antibiotics. Investors wisely look at the track record of antibiotics generally in making a bet that resistance won’t develop so quickly as to make their investment worthless, rather than taking too much confidence from the track record of strep and meningococcus (and syphilis, as someone noted above.)
Host factors in desperately ill patients make it exceedingly difficult to truly eradicate infection today in many cases, the ones which generate calls for newer antibiotics because the old ones don’t work anymore. This leads to prolonged treatment in an effort to do with drugs what the patient cannot do with his own defence mechanisms. This is the principal reason why antibiotic resistance evolves so widely and rapidly.