About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Don't Like It? Well, Just Don't Cite It! | Main | A Quick Quiz (Re: Antibacterials) »

March 11, 2009

Bacteria: Respect Must Be Paid

Email This Entry

Posted by Derek

I’ve had the opportunity to learn more about antibacterial drug discovery in the last year or so – that was one of the few therapeutic areas I hadn’t worked in, actually. And although I already knew that it was no picnic in the park (I’d heard the complaints), doing it myself has given me a new respect for the nasty resilience of bacteria.

I’ve been used to having my compounds go into cell assays, where a good number of them fail. That’s expected – every medicinal chemist knows that some of the potent compounds in the primary assay (against the purified protein target) are going to wipe out when they go up against cells. Cells have membranes, for one thing, and they have bailing pumps built into them to spit out molecules that they don’t recognize. I’ve seen compounds, as has everyone who does drug discovery, that bounce right off the cell assay while closely related analogs work just fine. That’s why you run the assay, to weed those guys out – you may not every understand what specifically went wrong, but you at least get a chance to try to avoid it as you go on.

But bacteria are different beasts. Their independent, free-living nature makes them nastier than even tumor cell lines. Cancer cells, aggressive creatures though they are, still expect to get their food delivered (and their garbage hauled) by the bloodstream. (That’s what makes angiogenesis a drug target in oncology). But bacteria have to search out their own meals, fighting it out with every other bacterium in the area while doing so. Their membranes are like armor plate compared to a lot of higher-organism cell lines, with the gram-negative organisms taking the trophy. Or is it the mycobacteria? (They're both awful, and the proportion of compounds that fail when you move past the pure-protein stage is thus far higher). They react to threats, communicate with each other, and reproduce like crazy. It’s like dealing with a swarm of tiny, self-replicating attack submarines.

So yes, finding an effective new antibacterial drug is a real triumph, and it’s not a triumph that’s been happening very often in recent years. This gets mentioned a lot in the popular press, when they feel like running a Coming of the Superbugs piece, and one of the usual explanations is that drug companies got out of the area years ago because they thought the problem wasn’t big enough to worry about. That’s part of the explanation – or was, quite a while ago. And the finances are different in this space, true. You’re never going to have a multibillion dollar blockbuster, because a new agent is going to be reserved just for infections that are unresponsive to the older drugs. But it’ll still sell.

No, there are plenty of companies working in the area now, and many that never left. And the need for new agents is clear, and has been for quite a while now. The real reason that we don’t have lots of new antibacterial drugs is that it’s really hard to find them, for one thing, and the the bacterial are more than capable of fighting back when we try.

Update: for more on the topic, see here.

Comments (25) + TrackBacks (0) | Category: Infectious Diseases


1. The Pharmacoepidemiologist on March 11, 2009 8:02 AM writes...

Does anyone know anything about a second round of layoffs at Sanofi later this year (as in after June)? Also, what about the major consolidation taking place at JNJ? I'm hearing that pink slips are being assembled. Again, anyone?

Permalink to Comment

2. Anonymous on March 11, 2009 8:26 AM writes...

GSK has a long history in anti-bacterials, and got retapamulin (new class of antibacterial, the pleuromutilins) onto the market a couple of years ago. They 'turbo-charged' infectious diseases research in 2007 by slashing headcount and making most of the remaining unfortunates go onto 3 year temporary contracts rather than the permanent contracts they were on before. How motivating.

Permalink to Comment

3. milkshake on March 11, 2009 8:30 AM writes...

In my first job in US, a colleague asked me to give her all my compounds and intermediates for antibacterial screening in growth inhibition assays. As the results started pouring in, my impression was that it was quite easy to come up with hits in the single digit microgram/mL range but pretty hard to get anything useful. As I remember the stuff that came out were either extremely greasy prenoid-like alcohols (and also some ring-chlorinated benzylic alcohols similar to Triclosan), there were unpleasant-looking compounds like nitrofurfural (and its thiophene analog) - turns out there is an antibiotic nitrofurantoin on the market. Greasy amines with hemolytic activities also showed up. My colleague was very excited but I tried to explain to her how hard it would be to develop crap hits like these into reasonable leads, even if these compounds were non-toxic...

Permalink to Comment

4. TW Andrews on March 11, 2009 10:11 AM writes...

My understanding is that a lot of anti-microbials are refinements/reformulations of natural products which bacteria use to kill each other. Is that the case, or are there other good sources of anti-bacterials?

Permalink to Comment

5. Lynn on March 11, 2009 10:13 AM writes...

Microbiologist, ex-Big Pharma, now consulting in antibacterial discovery. I agree that the lack of new antibacterials is largely due to the fact that it's hard to find them. First, because of the permeability problems you allude to. Second (re milkshake's post) to the poor fitness of the chemical libraries that are used to screen for antibacterials these days (They are generally alkylators or detergents. In the latter case, they kill gram positives by surfaction, the antibacterial activity generally not being due to interaction with a specific target). Third because an inhibitor of a single enzyme (if it does encounter its target) will in all probability select for resistance overnight. And it's hard to find novel natural product antibiotics. Big Pharma has been slashing antibacterial research and seems to think small pharma will come up with new entities to in-license. Good luck to them. And to the rest of us - who will be needing new antibacterials in the future. This is a problem that needs close collaboration between microbiologists and medicinal chemists in a research setting.

Permalink to Comment

6. The Inorganic Gardener on March 11, 2009 10:16 AM writes...

The Gram positive organisms (things like Bacillus spp., Mycobacterium spp. and all of that lot) are the ones with the big waxy walls. Gram negatives (Klebsiella spp., Pseudomonas spp., Escherichia spp. (yes, E. coli is really not the only Escherichia - there are plenty of others!) and Yersinia spp. spring to mind in terms of "of medical interest".

Bacteria are really very good (esp. the Gram positives) at hijacking DNA from one another. In comes a wave of naladixic acid and you're not resistant? No problem, one quick conjugation with a nearby cell and you can horizontally transfer the gene and gain resistance. Sod Darwinian evolution - horizontal transfer is much more fun.

The Bacteria are a fascinating kingdom. Their abilities far outstrip those of the Eukarya (plants, animals, fungi etc). Growth at 378K? No problem. Survival at 394K? Again, not a problem. Ditto growth on methanol, benzene, toluene, carbon disulfide, potassium cyanide and a variety of other "nasties". Use of uranium ions as an energy source whilst assimilating carbon from CO2 in the air. Resistance to thallium and other heavy metals.

You name it, I can probably find you a bacterium that can do it (within sensible reason of course - I don't know many which can drive a car etc!).


Permalink to Comment

7. fragment_boy on March 11, 2009 10:20 AM writes...

I agree that as Pharma moved onto the 'single target' approach it made anti-infective programs near impossible for the reasons outlined by Lynn in post 5.

For a successful anti-bacterial campaign you have to roll back to the good old days of phenotypic screening... at least then you know that your agent is A) getting in and B) having an effect.

Sadly thats not cool and trendy so pharma aint going to do it and a startup wont get funding for it.

We should pretend that we are Jannsen.

Permalink to Comment

8. qetzal on March 11, 2009 11:09 AM writes...


I'm interested by some of the points made by Lynn (#5), The Inorganic Gardner (#6), and fragment_boy (#7). Can you comment on how the typical strategies in antibacterial drug discovery compare to those for, say, anticancer drugs? Do most pharmas approach them in a similar way? (E.g., select a molecular target, look for hits in some cell-free assay, then move to cell-based assays, etc.) If so, do you agree with the previous commenters that this may be a mistake? Or are there significant differences, and if so, what are they?

Just curious. :-)

Permalink to Comment

9. JRad on March 11, 2009 11:55 AM writes...

So having been in both infectious diseases and cancer research (ex-big pharma), lack of new discoveries in AI is due to a combination of percieved lack of market size for AIs despite resistance being an issue for the last 15 years(exit big Pharma), the move away from natural products and toward small molecule synthetics (less molecular complexity, easier med chem, and lower COPS)and the magic bullet screening approach (one target, one kill). While perhaps an effective paradigm for other TAs (emphasize PERHAPS), a different paradigm is required for AI discovery and from an organizational level this is a problem in big companies (lack of R&D synergy with the rest of the organization). AI discovery looks to be uncompetitive from that POV. It looks totally different for someone when they are dying from methicillin or vancomycin resistant infections. A perfect niche for focused smaller company efforts with the right combination of expertise, infrastrucutre and approach.

Permalink to Comment

10. medicamenta vera on March 11, 2009 11:56 AM writes...

Just Curious:
Single molecular targetting has been the prevailing paradigm since I started working in at Big Pharma in the 80's. This approach facilitates:
1) Driving SAR
2) Structure based drug design (eg, crystallography, etc)
3) Explaining proposed MOA to the FDA (and upper management)

It is generally appreciated that clinical efficacy and differentiation are driven by a number of factors. For example, the exact mechanism by which amlodipine works is not well understood. Lipitor which did not have a stellar pre-clinical profile has been proposed to have "pleiotropic" effects. (I would be interested in non-cardiovascular examples.)

Sometimes a drug appears to work by having more than one mechanism of action. A recent example I'm interested in is istaroxime a Na+/K+ ATPase inhibitor in Phase II development which is also a SERCA 2a activator. The latter activity is proposed to make this molecule better tolerated that digoxin. (Having a shorter half-life helps too.) Based on the literature it appears that the initial goal was to develop a Na+/K+ ATPase inhibitor, and when istaroxime became a development compound the SERCA 2A activity became apparent, and was used to rationalize the greater therapeutic window vs. digoxin.

One final note, you might find The Artemisinin Project and the work of Jay Keasling of interest. It is a great story, and a peek into the future.

Permalink to Comment

11. karen on March 11, 2009 12:02 PM writes...

There can also be toxicity issues. If you find something potent enough to get past the bacteria's defences, it may be potent enough to do bad things to your liver, kidneys, etc. I worked on an antibiotic some years ago that was very potent, but had the unfortunate side effect of killing some of the test animals from liver toxicity. (And it was a natural product and very hard to make, so not much commercial potential.)

Permalink to Comment

12. emjeff on March 11, 2009 12:43 PM writes...

Sure, it's not easy finding viable compounds against a bacrterial target. But is anti-hypertensive research any easier, when you have to prove that your candidate makes people live longer?

The real reason antibiotics are given short shrift is of course economic . It is no cheaper to develop an anti-microbial than a drug in any other therapeutic area. But, you only give an antibiotic for 1-2 weeks as opposed to months/years for diabetic agents for example. The business model just does not work.

Permalink to Comment

13. Andrew Ryan on March 11, 2009 1:10 PM writes...

Another microbiologist here. An important consideration is that new antibiotics don't have to kill the bacteria per se--they could just disable it so the immune system can clear it out. Small molecules that specifically inhibit virulence factors like type three secretion--sort of hypodermic that punctures cells and injects toxins and other proteins to hijack the host cell--are promising candidates.

Permalink to Comment

14. Retread on March 11, 2009 3:07 PM writes...

I'm surprised that no one has mentioned quorum sensing as a line of attack. The most resistant infections (a la lung infections in cystic fibrosis, urinary tract infections in unfortunates with indwelling catheters) are those in which the organism has formed a biofilm, rendering them much more resistant to antibiotics (if they can even diffuse into the film to get to the bugs). Quorum sensing molecules are intimately involved in biofilm formation, and even better, the molecules must be extracellular to work. Receptors for them are on the outside of the organisms, to boot Are any of you folks out there working on antiquorum sensing molecules?

A good place to get an overview is [ Cell vol. 125 pp. 237 - 246 '06 ] by Losick and Bassler -- who did a lot of the early work. Perhaps once the film has formed the ballgame is over, quorum molecules no longer being required, but it seems like it's worth a look. Is anyone doing this?

Permalink to Comment

15. Lynn on March 11, 2009 3:53 PM writes...

to JRad at #9 - I agree that small pharma may be able to succeed in the hard job of NOVEL antibacterial discovery (as opposed to the relatively straightforward modification of an older class) if they do embrace new paradigms (as opposed to new "platforms") - which recognize and try to overcome the obstacles to antibacterial discovery (permeability and the need for multitargeting [or at least low resistance-potential]). But this will take, as you say, the right combination of expertise [available now as refugees from big pharma] and some freedom from the need for an immediate product candidate and monetary return (not so easy in small pharma). I still believe novel natural products are out there, too.

As to anti-virulence and quorum sensing - they have been worked on, mostly by small pharma and largely avoided by large pharma. So far, no leads I'm aware of. It is not clear to me that they would be sufficiently therapeutic (as opposed to prophylactic) and I believe that they are just as prone to rapid resistance selection as any single enzymatic target. Perhaps they'd be useful as adjunctive or combination therapy.

fragment_boy - I heartily agree with your phenotypic screening comment.

Permalink to Comment

16. Dave C on March 11, 2009 4:09 PM writes...

To add to the many challenges already discussed: if you develop a brand new magic molecule that works against drug resistant strains and you market your drug ethically (i.e., only to those who have resistant strains), you're targeting a very small market. Add this to emjeff's comment that patients will take the new drug for a short time. The drug would be sold into a slice of a comparatively small market. Unfortunately, market forces aren't conducive to new antibiotics.

Also, consider the challenge of clinical trials with the magic molecule: how do you prove effectiveness in people? You can't have placebo controls. Ethically, the primary subjects would be those for whom all other therapies failed. You have to identify & give them the new drug before they die from what could be a very fast-growing infection (i.e., death within hours or days). Then try to do statistics on this highly variable patient population to show that your drug actually worked.

Permalink to Comment

17. Dave_n on March 11, 2009 6:49 PM writes...

I loved the GSK has extensive antibiotic experience piece above. I was in the old SKF antibiotic discovery program prior to the "mol biol onslaught" and we had compounds in clinical use. However, head-count was needed so out the door went the a/b discovery team. Fast forward a few years and around $2B later. Excellent mol. biol papers, no drug....rethink...combichem approach...$1B+ expenditures, no drugs...resuscitate a very old natural product class and modify slightly. Do not point out that it is an old molecule in new clothes...SUCCESS. However the group running the program no longer exist and GSK is reorganizing yet again. Even more entertainingly, I still get requests from headhunters asking if I woudl like to get back into antibiotic discovery with a well-known Anglo-American corporation. Would appear that the HR department does not know their history.

Permalink to Comment

18. np on March 11, 2009 10:50 PM writes...

A great paper on GSK's antibiotic drug discovery program is Payne, D. J. Nature Drug Discovery Reviews 2007, 29.

Permalink to Comment

19. RB Woodweird on March 12, 2009 6:33 AM writes...

At Warner Bros. Pharmaceuticals, we like to dress our antibacterials up in tiny little blonde wigs and have them wear mascara and lipstick, then hang out waiting for the bacteria to come by and get all smitten over them. Then BOOM!

Permalink to Comment

20. retread on March 12, 2009 8:01 AM writes...

#15 Lynn

You're quite right that if you attack the receptor for the quorum molecule (antibodies, small molecule inhibitors) the bacteria will mutate around it. Look at what's happened in man with the kinase inhibitors.

I don't think the bugs would deal quite so easily with attacks on the messenger itself (the quorum molecule), some of which are quite wierd (one such called AI-2 even contains boron [ Nature vol. 415 p. 546 '02 ] ), and none of which (as far as I know) occur in eukaryotes.

Permalink to Comment

21. srp on March 12, 2009 10:31 PM writes...

Has anyone thought about trying to judo the evolutionary process of the bugs by encouraging the mild ones to reproduce like mad and outcompete the nasty ones?

Permalink to Comment

22. David Govett on March 13, 2009 2:07 PM writes...

It's a never-ending arms race, at least until the Singularity. Then it will be payback for all the misery bacteria have caused humanity.

Permalink to Comment

23. Chip on March 13, 2009 2:37 PM writes...

As a pharmacy student, I want to thank you all for pointing out why, exactly, I have to sit through all of the the medchem and micro.

Fascinating. I appreciate you making it interesting.

Permalink to Comment

24. Uniline on March 16, 2009 9:26 PM writes...

Bacteria are an important part of our body. As long as they are under control we benefit from them. I think to control the virulence is a better approach, so inhibition of quorum sensing is a good approach.

Permalink to Comment

25. Jane Yao on July 18, 2012 2:04 PM writes...

Screen our Newly Isolated compound library to generate new drug leads

Please take a look at our unique sample library containing low hanging fruits, and consider screening it in your next drug lead discovery.

We ( provide over 12,000 non-commercially available compounds and fractions obtained by column separation of worldwide chemically untapped natural products.


Health Resource Pharmaceuticals LLC

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry