By Randall Osborne

NEW YORK ¿ Lexicon Genetics Inc. disclosed what the company said is a breakthrough in gene trapping technology, which will enable the company to create a database of significant human genes by this summer.

Arthur Sands, president and CEO of The Woodlands, Texas-based Lexicon, said scientists have refined and upgraded the company¿s mouse technology to trap human genes, with a method he said is far superior to traditional genomic sequencing and the expressed-sequence-tag (EST) method.

¿It¿s like refined ore vs. crude ore, and it¿s a way to detect where that ore is, more efficiently,¿ Sands told BioWorld Today, after presenting at the Biotechnology Industry Organization¿s CEO and Investors Conference here.

OmniBank, the company¿s library of more than 40,000 mouse embryonic stem cell clones based on gene trapping methods, is used by Lexicon and its subscribers to select drug-discovery targets. (See BioWorld Today, June 2, 1997, p. 1.)

The new LexGene Trap technology has trapped more than 20,000 human genes in the past few months, half of which are new, as determined by comparison with published gene-sequence databases, including the National Institutes of Health¿s Human Genome Project, Sands said.

More than 90 percent of Lexicon¿s trapped human DNA sequences have been verified as genes, and about half of those are novel, Sands said, adding that Lexicon is on track to finish its database of the entire ¿trappable¿ human genome this summer.

¿We know, conservatively, greater than 50,000 will be trappable, and 30 to 40 percent of these will be new,¿ Sands said.

¿We¿re going to apply a target discovery and validation model [with LexGene Trap], not a database model,¿ he added. ¿The trapped genes are going to be proprietary to Lexicon for some time, which allows us time to develop them as targets¿ for development with partners, Sands said.

In gene trapping, the company inserts thousands of retroviruses into the chromosomes of human cell lines.

¿When they land there, they activate ¿ scan, essentially ¿ for genes in the region,¿ Sands said. ¿They activate them because they carry their own, strong promoters. It creates what is called a transcriptionally normalized¿ library, because we now have our own strong promoter expressing all genes.¿

With EST technology, he said, researchers are ¿simply not finding novel genes anymore. ESTs rely on the endogenous promoter to express the gene. If there¿s a weak promoter, you¿re not going to find that gene.¿

Sands said LexGene beats genomic sequencing, because the latter requires working on all of the chromosomal DNA stretches, ¿97 percent of which do not contain genes. Even once you have sequenced all the chromosomes, it doesn¿t mean you¿re going to find a gene. They are segmented, and can be strewn over vast regions of genomic DNA. In trapping, we¿re focused on the coding regions of the chromosomes, so we don¿t have to sequence that 97 percent. And we bring our own promoter, so we normalize expression of all genes and get them all in an equal manner.¿

Genes found by trapping are likely to be more valuable, too, he said.

¿We know that these tend to be low-abundance genes,¿ Sands said. ¿They are probably the key regulatory molecules made by the body, and the body does not make a whole lot of them, unlike keratin in skin or collagen. The regulatory molecules are made in scarce amounts in key organs, like the pituitary gland and thyroid gland.¿

Although privately held Lexicon will create a database, the company does not intend to establish simple access arrangements, he said.

¿[Deals] will be done for target selection and done in a collaborative mode, but not in terms of raw information supply,¿ Sands said. ¿We would be looking to develop the target further with the partner, using mutant mouse technology, and we would be looking for larger royalties and up-front payments.¿

Progress already has been made, he said.

¿We¿ve made over 20 knockout lines of mice in-house, for looking at target validation,¿ Sands said. ¿We¿re categorizing these into well-known drug target categories. G-protein coupled receptors are prioritized and secreted proteins are categorized, and we¿re going to be making knockouts systematically in those categories to determine function, which will then take us to a disease category and lead us to the appropriate partnerships.¿

HGS¿ Effort Directed Toward Drug Discovery

Also at the conference, William Haseltine, chairman and CEO of Human Genome Sciences Inc. (HGS), detailed for the first time the functional genomics platform that has enabled his Rockville, Md.-based company to advance three genomics-based products into the clinic.

¿We constructed HGS from the beginning not primarily as a company to find genes, but as a new way to find drugs,¿ Haseltine said.

HGS¿ products are the cancer treatments Myeloid Progenitor Inhibitory Factor-1 and Keratinocyte Growth Factor-2; and Vascular Endothelial Growth Factor-2, which stimulates blood-vessel growth. (See BioWorld Today, Nov. 19, 1998, p. 1; Aug. 27, 1998, p. 1; and Nov. 11, 1997, p. 1.)

The company said it has used its functional genomics platform to isolate messenger RNAs from more than 95 percent of all human genes, of which 75 percent to 80 percent are fully functional, containing the ¿instructions¿ needed to produce the corresponding protein.

HGS also has identified about 12,000 newly discovered genes that encode for signaling proteins, and has completed the sequence analysis of about 9,000 genes and more than 11,000 proteins. About 500 of the newly discovered proteins are related to previously studied ones, including 35 new members of the interleukin family; 40 new members of the growth factor family; and more than 100 members of the G-protein coupled receptor family.

Rachel Leheny, an analyst with Warburg, Dillon, Read LLC, of New York, said genomics has come far since its advent in 1993, but investors are expecting still more.

¿We have seen no small molecule drugs that have come out of genomics enter the clinic yet, and only a few drugs that are therapeutic proteins have entered the clinic,¿ she told the conference.

¿This is going to be a critical issue for the next five or 10 years,¿ Leheny said. ¿People are doing a reality assessment. Both Wall Street and the pharmaceutical industry are saying, Show me the money.¿ People are saying, What can we do to make money [from genomics], now and in the future?¿¿

The conference, sponsored by the Biotechnology Industry Organization, investment firms and others, continues through Friday. n