New antibiotic

[arcticpenguin]

http://www.sciencedaily.com/releases/2003/10/031024064511.htm

According to the study, CBR703 inhibitors hindered the ability of RNA polymerase in Escherichia coli to perform crucial catalytic functions, such as building molecules of RNA. Compounds in the CBR703 series – all are synthetic chemicals – render RNA polymerase useless by binding to a specific place on the enzyme – a necessary step in the process.
...
While the CBR703 inhibitors seemed to stop gene expression in E. coli, the researchers found that the compounds wouldn't inhibit RNA polymerase in human cells.
...
Compounds in the CBR703 series – all are synthetic chemicals – render RNA polymerase useless by binding to a specific place on the enzyme – a necessary step in the process.

This is unusual. Most existing antibiotics are from bacteria or fungus, or are slight modifications of natural compounds.

Since this compound works by a novel method - it blocks a different site on RANpol than other drugs, it could potentially be used in combination with other drugs.

 

[American]

I'd like to know how they determined the molecules might be good candidates in the first place. My hunches are they either created a trillion random compounds and screened them (using some kinda high throughput method), or less likely they engineered it by predicting the chemistry at the specific site it binds to...

I'm happy-- NIH recently made a strategic decision to focus their billions on anti-viral research over antibiotics. Don't ask me how they made that value judgement, but if dollar figures matter, then it's nice to see this "second priority" making some progress as well.


edit-- that last point means nothing. It's probably a random hit by a dedicated team that has little to do with the larger scope of national funding.

 

[Rolfe]

Antibiotic = derived from a mould or whatever
Antibacterial = not so derived.

Antibacterials include sulphonamide-type drugs, trimethoprim, metronidazole and stuff like that. Pure pedantry and semantics.

Rolfe.

 

[Hellbound]

I'd like to know how they determined the molecules might be good candidates in the first place. My hunches are they either created a trillion random compounds and screened them (using some kinda high throughput method), or less likely they engineered it by predicting the chemistry at the specific site it binds to...

My understanding is that in modern labs they rely on computer modeling of various compounds. They can model the particular virus or bacteria (or a specific part of it, such as the RNA strain), and then look for various chemical combinations that create molecular shapes likely to bind there.

It's sort of a combination of the two ideas you have. The computer modelling gets it down to a set of possible compunds, and testing reveals whether these compunds work.

Not an expert, by any means, but that's my understanding.

 

[arcticpenguin]

My understanding is that in modern labs they rely on computer modeling of various compounds. They can model the particular virus or bacteria (or a specific part of it, such as the RNA strain), and then look for various chemical combinations that create molecular shapes likely to bind there.

It's sort of a combination of the two ideas you have. The computer modelling gets it down to a set of possible compunds, and testing reveals whether these compunds work.

Not an expert, by any means, but that's my understanding.

Yes, both techniques are tried in concert with each other.

A "lead compound" can be identifed in any way; random assays, modeling, whatever.

'Molecular modeling' and 'docking' make use of computers to interpret the 3-dimensional structure of a target molecule (elucidated with either NMR or X-ray crystallography) and predict ligands to fit the target. 'Combinatorial chemistry' is the generation of large numbers of related compounds to see if a better one can be found. The big pharma companies will have crystallographers, modelers, chemists and physiologists all working together.

Just because they found a molecule that inhibits bacterial growth in a test tube is no guarantee that it will yield an effective drug. There are other things to be checked. Is it safe for humans? Can it be absorbed by the body? Will it be broken doen by the body before it can do its job? Which types of bacteria is it effective against?