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

« Well, Okay: The Ugliest Biopharma Sites? | Main | Conference Thoughts »

October 13, 2010

A Cautionary Tale

Email This Entry

Posted by Derek

For those who haven't seen it, I heard Adam Renslo of UCSF present this work yesterday. His group was looking for inhibitors of cruzain, a target for Chagas' disease, which is certainly a worthy cause (and a tough target). They found a series of oxadiazoles, which are, to be sure, rather ugly (but no uglier than a lot of chemical matter you see in the kinase field, among others). They had affinity, they had reasonable SAR, and the team drove the potencies down against the target. . .only to find, late in the game, that it was all an illusion.

These compounds are aggregators. (That link takes you to a post about a follow-up paper from the UCSF folks, covering this debacle and others). What's striking is that this artifact (compound aggregation under the assay conditions) mimicked a plausible SAR - it wasn't just some random thing that made the numbers hard to interpret. No, it looks like Renslo and his team ended up optimizing for aggregation. As he put it in his presentation, "You'll find what you're looking for".

His other quote at the end of the talk was "Small molecules are much stranger than we've been led to believe", and I can't argue with that one either. Before anyone makes a comment about how his group should have checked their assay more thoroughly, or how they shouldn't have been trying to push such an unpleasant-looking series of compounds anyway - in general, about how this wouldn't have happened to you - pause for a moment, and be honest. Renslo was in this paper, and I thank him for it.

Comments (28) + TrackBacks (0) | Category: Drug Assays


1. Simon on October 13, 2010 9:40 AM writes...

Interesting and thought-provoking.

This one was 'rectified' --- but how many false cases do you think are out there? - Can someone come up with other examples from the past?

Permalink to Comment

2. RB Woodweird on October 13, 2010 9:43 AM writes...

So they found a clump of some compound which showed desirable binding to their target? Why can they not characterize the clump and prepare a structral/electronic twin so the clump - now a transient lump held together by electrostatic forces - is a permanent structure held together covalently?

Permalink to Comment

3. imatter on October 13, 2010 10:09 AM writes...

Unfortunately, "You'll find what you're looking for" is a tough lesson to learn so late in a scientific career.

Permalink to Comment

4. Fries with That? on October 13, 2010 10:30 AM writes...

I'm pretty sure the article also mentions they used a lower detergent concentration than they should have, meaning if they would have used the proper detergent concentration in the first place they likely would not have obtained aggregators.
They are beating the aggregator dead horse to death, by milking every artifact they get by publishing a paper, and also, insidiously, linking HTS in general to such artifacts.
Ok, I get it, HTS IS BAD BAD BAD because aggregators happen. I get it , I get it. Ok, now please stop publishing this nonsense, the word is out now and has been for quite some time. YAWN. BTH, use the correct detergents in your assays if you do decide to run a high throughput screen.

Permalink to Comment

5. gyges on October 13, 2010 10:37 AM writes...

"Before anyone makes a comment about how his group should have checked their assay more thoroughly, or how they shouldn't have been trying to push such an unpleasant-looking series of compounds anyway - in general, about how this wouldn't have happened to you - pause for a moment, and be honest."

Ever met a chemist, Derek?

Permalink to Comment

6. Cialisize Me on October 13, 2010 11:56 AM writes...

Is there an issue with the oxadiazole itself, and if so, can someone post opinions on what is the inherent problem? Or... is the issue with the whole molecule with its array of H-bond donors and acceptors that can promote self-association? I haven't read the papers yet.
Thanks, C.M.

Permalink to Comment

7. ronathan richardson on October 13, 2010 12:03 PM writes...

I'm wondering if someone could answer--is aggregation supposed to be a problem only for binding/inhibition assays, or are whole-cell assays susceptible? Because you can't run your whole cell assays in .01% triton...

Permalink to Comment

8. weirdo on October 13, 2010 12:12 PM writes...

Yeah, well, they had high micromolar inhibitors and I don't see any evidence that they attempted to validate the "activity" in whole-cell assays (re: #7).

They were asking for trouble. Begging for it. A hard lesson for these students, I am sure.

Permalink to Comment

9. Fries with That? on October 13, 2010 12:40 PM writes...

I'm not an expert but I believe it is only an in vitro artifact, I'm not sure an aggregated compound could even get into a cell.
I also doubt aggregation would happen inside a cell either.
The Shoichet lab was looking into any possibility of cell based aggregation a few years back, but I don't believe they found any artifacts to publish.

Permalink to Comment

10. Wavefunction on October 13, 2010 12:58 PM writes...

Aggregators are now well-established caveats to watch out for in HTS, thanks to the efforts of the Shoichet group. I had blogged about a paper in which they ran a screen against beta-lactamase and found no less than 95% of the molecules to be aggregators. And I wouldn't be surprised if using higher detergent concentrations may have other adverse effects.

Permalink to Comment

11. Moody Blue on October 13, 2010 1:10 PM writes...

@7&9 Well if the aggregation happens in the cell-based assays and gives the desired outcome, so be it. Would that be a problem, though? Then wouldn't the same phenomenon occur in vivo and lead to efficacy? I thought the whole issue of aggregation was brought to light by the Schoichet group as there discrepancies in the enzyme assays and cell-based ones.

Permalink to Comment

12. RTK on October 13, 2010 3:38 PM writes...

The way that aggregators disrupt in vitro assays is not by specific binding but by trapping the target in an aggregated blob, effectively removing it from the assay. In the presence of aggregating compounds the protein target can be precipitated from the reaction mix. This should not be an issue in cell based assays at all.

Permalink to Comment

13. Fries with That? on October 13, 2010 6:52 PM writes...

Also, several drugs which may work fine in cells and in the body may display aggregation properties at high micromolar concentrations in a non-cell based screen. The aggregation properties in these cases have nothing to do with the efficacy of the drug, since in the cell the drug is acting in a normal stoichiometric fashion. The use of Triton-100 detergent at a particular concentration has been used in the past to diminish the formation of these blobs in vitro, I believe the Renslo group freely admits they mistakenly thought they were using the "correct" detergent concentration when in fact they were using a much lower amount.
No one said drug discovery is easy, but this group at UCSF has a snobbish attitude towards HTS and is constantly harping on the negatives of HTS while singing the praises of tethering. Need I say more?

Permalink to Comment

14. Aspirin on October 13, 2010 8:42 PM writes...

Fries: The two groups you are talking about are different ones. The tethering group is Jim Wells's group, nothing to do with Shoichet. As far as I know Shoichet has not pushed tethering.

Permalink to Comment

15. Fries with That? on October 13, 2010 8:51 PM writes...

They are in the same department on the same floor in the same building and Wells is the Dept. Chair of Shoichet's Dept. Both Shoichet and Wells groups have snobbish attitudes towards HTS. Renslo is affiliated with Wells. Perhaps Shoichet also has snobbish attitudes towards tethering, if that is what you imply?

Permalink to Comment

16. Cellbiio on October 13, 2010 11:21 PM writes...

Cell based activity can be seen with compounds that form micelles or lamellae (maybe these are the lay terms for what chemists call blobs? Just kidding) . Certain receptor systems are more sensitive to membrane organization, so some compounds at CMC, usually around the micromolar or higher range in my experience, can appear to have specificity, but only if activity is evaluated at the magical dose, and not too much biology is examined. Dose response curves will reveal the telltale sign of no activity then abrupt 100% inhibition over a very small concentration change. Of course, the curve can look better if you space out the concentrations and specify a sigmoidal curve with a nice Hill slope. If you can never catch a data point in the half maximal zone, you've probably got a problem. Wouldn't this also be true for the in vitro stacking blobs that precipitate an assay component? They see data points in the middle of the curve. Maybe the amount of target or substrate removed is proportional to the size of the blob?

Screened 100s of thousands compounds in cell assays. Pilot work showed the 10 uM hit rate was huge. Settled on 1 uM as concentration to find compounds working by reasonable mechanisms, could go to 3 uM, but that pushed it. I never trust data in cell based assays above uM unless there is a lot of other data that confirms mechanism and cell integrity.

Permalink to Comment

17. Eka-silicon on October 14, 2010 2:28 AM writes...

By and large, academic medhcem is a quagmire, but I don't know anything at all about the Renslo lab, so this is a general question. Would this have go on for so long in industry?

I think the multi-department meeting at biotech/pharma are a key component of sorting messes like this out, and maybe there are insufficient chances in academia for that kind of interaction to occur?

Permalink to Comment

18. weirdo on October 14, 2010 8:48 AM writes...

Fries: Snobbishness.

Have you actually met the so-called "snob"?

I have. We've collaborated. He's a great guy. In person, no snobbishness at all.

Permalink to Comment

19. Fries with That? on October 14, 2010 9:53 AM writes...

Which one are you talking about Weirdo?

Permalink to Comment

20. Aspirin on October 14, 2010 10:06 AM writes...

I don't know who you are talking about but I have met both Shoichet and Wells and none of them came across as snobs to me (although Wells is more amiable than Shoichet). In fact I met Wells a few days after he had been elected to the NAS so there could have been plenty of potential gloating, but I didn't see any.

Permalink to Comment

21. weirdo on October 14, 2010 10:57 AM writes...

Wells. Don't know Shoichet.

But I haven't see anything in Shoichet's papers or talks that come across as snobbish. In fact, he clearly uses HTS in his work. Maybe he's an advocate for scientists doing things properly and not wasting time & resources (or publishing crap), but what's wrong with that?

If that's snobbish, we need a lot more snobs.

Permalink to Comment

22. Fries with That? on October 14, 2010 11:37 AM writes...

Aspirin and Weirdo, I am talking about the attitudes displayed by those labs specifically towards HTS screening, even though each lab has taken advantage of HTS in their own way. Stay on topic with the use of the word Snob, as directed towards HTS. Both groups continuously push fragment screening, and in Wells lab, tethering, as superior techniques to plain old fashioned screening.

Permalink to Comment

23. JC on October 14, 2010 1:44 PM writes...

There's an empty building the next parking lot over used by a company that was big into tethering.

Permalink to Comment

24. Sirtuin skeptic on October 14, 2010 1:45 PM writes...

Hmmm, anybody check the Sirtis compounds for aggregation?

Permalink to Comment

25. RubyD on October 15, 2010 3:21 PM writes...

Hey, Fries:

You seem just a bit bitter about the UCSF HTS. You seem to have intimate insider knowledge. Did you have a bad experience with them, or did you work for them at one time and got canned or harassed?

Permalink to Comment

26. Adam Renslo on October 16, 2010 5:03 PM writes...

Perhaps it is worth clarifying a couple points for those without access to the original paper.

The original leads (aminopyridine esters) were validated as competitive inhibitors (lineweaver-burk analysis) using an appropriate assay condition that controls for aggregation (0.01% triton). Early SAR studies were also conducted under this condition, including for example replacement of the ester function with an oxadiazole or isoxazole ring (both tolerated).

It was at this point the error occurred - in transferring the assay between labs, a decimal point got moved one position. The interesting thing is that those early leads were equally active under the two conditions and so it was not obvious from the data that anything had changed. Of course something fundamental had changed and the assay was now selecting for aggregation rather than for stoichiometric inhibition.

The second surprise was that two logs of additional potency could be realized by optimizing under the low triton condition (in fact this SAR felt as 'real' as any I have experienced in a decade of work in medchem). The lack of cell-based activity had not been a major concern when we were in uM land but once we got nM inhibitors that were still inactive in cells we got suspicious and discovered the error.

Finally, everyone can rest assured that no graduate students were harmed in the making of this story. The whole of the chemistry effort was completed in a couple month's time by a staff chemist in the SMDC. As for Rafaela, she learned something interesting and important about aggregation phenomena and this story fit nicely into the larger theme of her thesis.

Permalink to Comment

27. Another Anon on October 19, 2010 11:47 AM writes...

Aggregation (and other false positives) isn't just a problem for HTS. Academic labs seem to happily publish their "Finding Micromolar Hits Against Target X Using Virtual Screening" papers and never bother to show that their hits are stoichiometric and reversible. Of course, they usually see activity in cells, but lots of compounds do strange things to cells, and most of them are just cytotoxic. I blame the referees myself...

Permalink to Comment

28. Andrea on June 25, 2013 1:39 PM writes...

Congress's health and safety, and you won t be tempted to reach for a quick view by any person trying to handle and operate the facility. For a RIC or whatever your departement calls it to be walked on under a carpet. Donaghue pleaded guilty at Bolton crown court to three breaches of the regulations. Teaching re-enforces learning Let your children share their books with the younger students and lead a discussion afterwords. These are just a glimpse of the entire construction jobs that is usually required.

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