KinasePro

Kinase Chemistry – Just a year and a half behind the times.

Archive for October, 2006

Couple-a-new-PDB’s

Posted by kinasepro on October 24, 2006

2BDF, and 2BDJ

These are the SRC / Abl phosphine oxide bearing purines from Ariad. Good stuff with a free bonus link to the published discussion: Chem Biol Drug Des 2006, 46.

OC1=CC=CC(CCN2C=NC3=C2N=C(C4CCCC4)N=C3NC5=CC=C(P(OC)(OC)=O)C=C5)=C1x-ray crystallography, SRC/ABL AP23464 AP23451 CCN2C=NC1=C(NC3=CC=C(P(O)(CP(O)(O)=O)=O)C=C3)N=C(C4CCC(N)CC4)N=C12 CP(C(C=C5)=CC=C5NC1=C(N=CN2CCC3=CC(O)=CC=C3)C2=NC(C4CCCC4)=N1)(C)=O

Posted in biotech, PDB, SRC | Leave a Comment »

AV-412

Posted by kinasepro on October 23, 2006

>> Update: Revise the structures to the acrylamide as in WO/2007/103233.

Hello Hello Aveo,EGFR Phase one recruitment initiated. Red Team, Blue Team, Go! Go! Go!

I see you’ve liscenced this AV-412 / MP-412 thingamajig, from Mitsubishi Pharma which had KinasePro a little worried at first, since he doesn’t like slogging through Japanese patents. Luckily for him though, the patents are written in english. More luckier: There’s only two real compounds in the claims.

Most of the kinasefans out here will already recognize this is bound to be a quinazoline patent, the only question is how did they get around the deluge of Tarceva / Iressa knock-offs. EP1690856 seems to be the only EGFR series coming from Mitsubishi:

Yep that’ll do. But only 2 compounds? Well so it is… In case your joining the party late. Tarceva bound to EGFR is: 1m17 and 1XKK also looks interesting when trying to interperet the above structures. So much for that region merely being accessible to sovlent, eh?

CC(N4CCN(C)CC4)(C)C#CC1=C(NC(CCOC5=CC=CC=C5)=O)C=C2C(N=CN=C2NC3=CC(Cl)=C(F)C=C3)=C1 CC(N4CCN(C)CC4)(C)C#CC1=C(NC(CCS(C5=CC=CC=C5)(=O)=O)=O)C=C2C(N=CN=C2NC3=CC(Cl)=C(F)C=C3)=C1

Posted in biotech, EGF | 3 Comments »

WO2006091450

Posted by kinasepro on October 22, 2006

Hi again Lexicon,

Kinasepro must admit that he missed your application from this passed August. Your press release bugged him though, so he wouldn’t let it go. If its not too late, I’d like to change the guess about LG293 from the last post.

LG293 = JAK3

Probably not my final answer, and of course I’ll be waiting with the rest of the peanut gallery to see what the story is over the next couple years. Heck, as far as Kinasepro knows LG293 isn’t even a kinase in which case it’s LG267 which is JAK3 (that actually kinda makes more sense) . So why does Kinasepro have this crazy idea? It’s because of WO2006091450 and the fact tha Rigel nominated R348 for JAK3 on the very same day.

While your application, like all your other body of matter patent applications, doesn’t list a single biochemical assay it does incorporate matter which bears a striking resemblance to Pfizers early JAK3 stuff. After a little digging it appears a number of companies have been poking around the azaindole / pyrrolopyrimidine motif, so today I’d say you’ve got some stiff competition.


There’s plenty of others. Pfizer hit this scene first back in ’99 (WO199965909) and Vertex and AZ have tons of applications around different chemotypes, There seems to be something special about the saturated ring – Neet – o. Presumably noone has been able to dial in pk / selectivity / etc. till recently.

Posted in biotech, JAK, Lexicon | Leave a Comment »

US20060234983

Posted by kinasepro on October 21, 2006

Hey Rigel, SYK prodrugs eh? I see you guys have been on this stuff like white on rice for a couple years. I suppose Kinasepro could wait for the Biorg. Med. Chem. Lett. to come out some time in 2010, but I’m gonna see if I couldn’t kick up a first draft to help you guys out. Here Goes:

So R112 sucks but since we’ve got this really cool assay and Pfizer’s had a bag of cash dangling out there for us, its not like we were just going to give up on the target. Yah in case you were wondering we all lost a ton of money back then.

Anyhow the early SAR for this project was carved around the arylaminopyrimidine scaffold based on a library hit, and and we put the kitchen sink both meta and para on the pyrimidine (WO2003063794 – 7/8/03; WO2004014382 – 2/19/04).

R112 had crummy solubility, so we thought the idea of shoving it up peoples noses sounded pretty good. You know, like an allergy indication – something sort of topical where you might get away without solubility? Put it together with some of the finest, and most advanced dispersal agents known to mankind and throw it into a few hundred sneezing folks and lets see what happens. Somehow it didn’t work out. Go figure, thats just how these things go sometimes I guess.But as I was saying, some of our best early compounds were bicycles(1), and since I had some luck with indoles on a previous project I really put a beat down on that indazole, and indole chemistry. C’mon they’re supposed to be phenol bioisosteres. It just wasn’t working out until we started making benzoxazines (2) and then we knew: these comopounds are for real!

r1

That brings me to these second generation analogs where we got a feel for the SAR. Along the road we learned a few things: Namely as we honed in on the benzoxazines we found that the saturated ring preferred a hindered center next to oxygen (4), that the other aryl was more or less solvent exposed, and that we could get away with putting tons of stuff out there (3). Oh yah – and that an azabenzoxazine improved our drug like properties (5). Sure they were harder to make, but you gotta do what you gotta do, eh? The pyrimidine wouldn’t budge, but oh yah, we did find some success with a few non-benzoxazines (WO2005013996). We had a handful of non-benzoxazines, but they just didn’t cut it. Insoluble as it is R406 was the best of the lot.

r406-We nominated the benzenesulfonic acid salt, and the phase 1 study was done around the same time as our Phase II for R112. This compound is oral and we proved we were hitting a biomarker but for some reason the stock didn’t go back up =<

r3To polish off this project we looked at some phosphonate prodrugs which eventually led to R788 (6), but pretty much everybody is on HCV now. US20060234983 published Thursday, and WO2006078846 came out back in January – they’re all a part of this work. The non-prodrug aza-analog of R406 had a solubility of like 1 ug / mL – while the phosphonate is like *POW* 5 mg/mL, it also gives us us a %f of about 30%. Oh, and R788 rox k thx.** The above is merely a lighthearted satirical dramatization derived from Kinasepro’s keen ability to filter the patent literature and press releases. Please feel free to complain in the comments or to kinasepro@gmail.com

>> Updated 10/23 >> Fixed a couple chemdraw errors, and would like to add a reader submission link to an article on R406 from August ’06 (free link). They got an xray but the picture… well the picture leaves something to be desired.

r4

Posted in Rigel, SYK | 4 Comments »

US7122552

Posted by kinasepro on October 19, 2006

Hi Vertex,

Nice JAK3 / CDK2 application. Kinasepro notices that you folks patent an aweful lot of compounds, and chemotypes. Keeps me busy, but I have to wonder how you’ll find time to develop all these things.

There have been innumerable CDK2 inhibitors enumerated, and Xrays described. JAK3 – hmm… not so much. Interesting to me if you guys actually found a scaffold that hits’m both, because of some key residue differences. (gatekeeper: met vs. phe, and a selectivity residue: cys909 vs asn322)

Anyhow cool fused-pyrazoles, I guess if you squint a little you could find some similarities with the Aurora kinase inhibitor: VX680

Posted in CDK2, JAK, Vertex | Leave a Comment »

WO2006106326

Posted by kinasepro on October 19, 2006

A big wuzzup to my peeps representin in the big 48,

Umm, thats country code 48. Umm, well you see I mean, ahh nevermind. Hi AstraZeneca nice CHK1, PDK1 AND PAK inhibitors. KP won’t lie, he hasnt seen a whole lot of PAK or PDK1 inhibitors, but he’s sure they’ve been implicated in everything from breast cancer to the tooth fairy. Chk1 PDK1 have been around the virtual-block for a while now, and there’s quite a few Xray’s in the PDB. Nothing quite like what you’re up to here though, so I’m quite happy to be poking through your latest application:

I don’t have a real good guess on how it’s binding, if pressed I’d say the amide is likely hitting the hinge, the aryl is headed off into H1, and the piperidines are in the ribose region (Chk1 acid residue in white below, is also in PDK1).

->> mid-post update, I wasn’t very clear on the binding mode until I cracked open PDK1 bound to staurosporine (1OKY) and overlayed it. I’d say I’m a little more confident now:

It strikes me that I’ve seen the indole-carboxamide as a kinase active scaffold before – something in the IKK vein. I’ll have to look it up and update tomorrow. Nice app.

Hej då
>> update >> indeed WO2006034317 is a Glaxo UK, IKK application from March, 06. which itself appears to be a continuation of SKB WO2005067923.   

 

Posted in Astra, Chk1 | Leave a Comment »

Some new X-Rays

Posted by kinasepro on October 18, 2006

Both published 10/17;An arylpyradizone complexed to P38: 2I0H:

If that 2.53Å to the hinge is real, thats awefully close, but what I really wanna know is who’s bright idea it was to leave on the protecting group?

And an inactive APO of DAP at 1.47Å: 2A2A

Posted in PDB | Leave a Comment »

WO20060105865

Posted by kinasepro on October 17, 2006

Hello there Merck KGaA,

Nice Chk1 application. This one’s a modification of a previous GSK application WO2006072354 [Thats glycogen synthase kinase, not glaxo (I know, kinasepro wishes one of them would change their name too)]

This functional group is not without precedent, occuring in BMY25368 a histamine H2-receptor antagonist and elsewhere in the medicinal chemistry patentscape. Notably Schering and Wyeth seem to be all over the functional group for the last 25 years. Not so much in Kinases, but somebody in Pearl River must like them because American Cyanimide, American Home Products, and Wyeth have all messed around with them quite a bit through their history. They describe them as alpha-amino acid isosteres: J. Med. Chem. 1998, 236.

I guess KP shouldn’t be surprised to find the group has been used before for kinase inhibitors. It doesn’t seem like anyone has taken them very far, but OSI has a series of derivatives that hit c-Kit WO2006034111, and Abbott tried to use them as a urea mimic in a series of PP1 analogs: US20040014756.

O=C(C(NC3=CC(C(NC(CN5CCOCC5)=O)=NN4)=C4C=C3)=C1N[C@H](C)C2=CC=CC(O)=C2)C1=O

Posted in Chk1, Merck DE | Leave a Comment »

WO2006108059

Posted by kinasepro on October 16, 2006

Hello Exelixis,

Kinasepro must lament that many of your claimed compounds he’s seen before. They appear in the recent Kirin application which published a mere 7 days before yours. WO2006104161 (published 10/5/06) Ooooh, 7 Days? I haven’t figured out the priority dates but you guys probably need a beer right about now, eh? Hopefully everybody signed their notebooks…

Sure there’s probably enough room room in the c-Met sandbox for both Excelixis and Kirin to play. Both groups appear to have found interesting heterocyclic tail pieces, and the Excelixis application relies less heavily on the acyl-ureas claiming also some things like phenethyl oxalamides, and a noteworthy benzothiazole-amide (scheme on p142-144).

Mostly this is just a continuation of: WO2005030140 where they were looking at the quinazoline analogs. Bit of a snoozer if you ask me, but KP does like some of your mesylate intermediates:

Somebody should write a review on morpholine isosteres.

O=C(C5(CC5)C(NCC4=CC=C(F)C=C4)=O)NC(C=C3)=CC=C3OC2=CC=NC1=CC(OC)=C(OC)C=C12 O=C(C(NCCC5=CC=CC=C5)=O)NC(C=C3F)=CC=C3OC2=CC=NC1=CC(OCCCN4CCOCC4)=C(OC)C=C12 O=C(OC(C)(C)C)N2C[C@]([C@]([H])(C2)CC(C1)OS(=O)(C)=O)1[H] O=C(OC(C)(C)C)N1C2CC(CC1CC2)CCOS(=O)(C)=O

Posted in c-Met, Exelixis | Leave a Comment »

WO2006104917

Posted by kinasepro on October 15, 2006

Hi again GSK,

Looks like Kinasepro was napping again while you guys nominated your latest p38 candidate. I like your process route, and thanks for reminding me why I’ll never make a good process chemist.

You see in step 2 KP would just chuck it in the microwave with some Ac2O, precipitate with ether, and carry the crude. Sure I’d end up with a 25% yield, but I’d be half way home by lunchtime! It appears as if maybe there’s a better way to do the cyclization. I’ve never heard of this activated Horner-Emmons reagent but I can’t imagine you’d stick it in the patent if it didn’t work, eh? (Yah, these process guys are just like that) Thanks for the tip.

Posted in GSK, p38 | Leave a Comment »