By Mary Welch

Irori Inc., a privately held company based in La Jolla, Calif., and Bristol-Myers Squibb Co. (BMS) will develop a miniaturized high-throughput combinatorial chemistry system that could eventually net Irori more than $10 million.

As part of the two-year deal, expected to be disclosed today, BMS made a $4.5 million equity investment in Irori and will pay milestone payments and licensing fees.

The system, which will use Irori's NanoReactor technology, allows BMS, of New York, to synthesize thousands of compounds of known structure per week, thus speeding its lead drug discovery process. The NanoReactor technology also will reduce the cost of reagents (the chemical ingredients needed to make compounds) since the compounds will be produced in micro-quantities nearly 10 times smaller than existing systems. The technology should be up and running within two years.

"We've worked with Irori for quite some time and I am very impressed," said David Floyd, vice president of discovery chemistry at BMS. "The NanoReactor is the most logical next step to screen very large numbers of single compounds in the most efficient way."

The collaboration has several benefits for Irori, a company that develops and markets systems, products and associated consumables for drug discovery. In 1997, it had sales of $3 million, the first full year it sold products.

"On the financial side, we will be profitable in the fourth quarter of 1998," said Richard Brown, vice president of business development for Irori. "This alliance, and a second one to be announced in a few weeks, should be sufficient to meet our financing requirements without additional venture capital funding. An [initial public offering], however, is a definite possibility."

The deal also brings the three-year-old company to the lead discovery side of combinatorial chemistry, in addition to its work in lead optimization.

"With lead optimization, once you have a hit (an identified compound with desirable biological activity), you then make hundreds to thousands of chemical variants," said Brown.

These closely related synthesized compounds are tested and examined to understand how modifications to the lead compound's molecular structure affects its biological activity.

The two standard combinatorial chemistry methods are the parallel synthesis and the split/pool method. Parallel synthesis has the advantage of providing individually isolated compounds ready for biological testing. The split/pool method is 10 to 1,000 times more productive than parallel synthesis but produces complex mixtures of compounds rather than individual ones.

Method Uses Radio Frequency Tags

In 1996, Irori introduced its AccuTag-100 Combinatorial Chemistry System with a directed sorting technique. The system was the first commercially available method to let scientists simultaneously gain the advantages of both techniques — namely generating large libraries — but also identifying and tagging each compound. Using a radio frequency (Rf) tag, scientists can label and track each compound, thus allowing combinatorial libraries of thousands of discrete compounds to be rapidly synthesized and then tracked through the whole combinatorial chemistry process.

Each compound is synthesized in a Rf-labeled microreactor, called a MicroKan Reactor. Hundreds of these microreactors are used together in each chemical reaction vessel — which is the size of a whiskey flask. By splitting and pooling these microreactors, large numbers of discrete compounds are synthesized from a small number of chemical reactions.

The typical way of producing 1,000 compounds would be to perform up to 3,000 chemical reactions, which could take several weeks, Using the Irori's microreactor, those same 1,000 compounds could be produced from 30 chemical reactions in 1 to 2 days.

However, the cost and physical size of the microreactor, while ideal for hundreds to thousands of compounds, would be "cumbersome for tens of thousands of discrete (separate) compounds," Brown said, adding, "The microreactor is about the size of a cigarette filter and you put several hundred of these into each chemical reaction flask." Each reactor costs between $2 to $3 and there must be a reactor for each compound, he explained.

"What we've been wanting to do is expand this technology to lead discovery, where tens of thousands of those compounds are searched looking for that initial hit," he said.

Hockey-Puck-Sized Microreactor Coming Soon

Enter BMS. The company already was using Irori's AccuTag system but wanted to know if Irori could develop a miniature reactor. Together the two companies will develop a new NanoReactor, which will be one-fifth to one-tenth the size of the existing microreactor and will likely look like an "itty bitty hockey puck," Brown said.

The NanoReactor will use Irori's "directed sorting process but, in order to enhance throughput and dramatically reduce the cost per compound, the miniaturized NanoReactors will use a 2D bar code technology," he added.

The 2D bar code looks like a miniaturized checker board with its unique pattern of black and white spots becoming the tag associated with each compound.

Irori currently has several collaborations with pharmaceutical companies and last year opened Irori Europe Ltd., a sales, marketing and service office in Tarpoley, U.K., (near Manchester).

The company, now with 50 employees, was founded by Michael Nova, former vice president of business development with Prizm Pharmaceuticals Inc., of San Diego, Calif., and now a consultant, and K.C. Nicolaou, chairman of the chemistry department of Scripps Research Institute, in La Jolla, Calif.

BMS's stock (NYSE:BMY) closed Tuesday at $109, up $1. *