By David N. Leff

Editor's note: Science Scan is a roundup of recently published biotechnology-relevant research.

So far this month, three papers - two in Nature, one in PNAS - add new insights to the marathon quest for the molecular causes of Alzheimer's disease (AD) - and thence leads to drug therapy.

At this stage of knowledge, four human chromosomes get into the AD act, harboring genes that predispose to the disease:

The amyloid precursor protein (APP), some 700 amino acids long, resides on human chromosome 21. From this motherlode sequence, two proteolytic enzymes - beta-secretase and gamma-secretase, - cleave off short sequence fragments, which encode the amyloid neuritic plaques that are the hallmarks of AD neurodegeneration.

Presenilins 1 and 2 perch on chromosomes 14 and 1, respectively. These two proteins, plus APP, trigger early-onset AD - before the age of 60. The apolipoprotein-E gene (APOE), on chromosome 19, is a major risk factor for the more common late-onset AD - after the age of 70. Missense mutations in the gene encoding presenilin-1 is the commonest cause of familial, early-onset AD.

Last autumn, the beta-secretase enzyme was identified, providing one molecular target for AD drug-discovery efforts. Now, scientists at the Merck Research Laboratories in the U.S. and UK announce that presenilin-1 contains the active site of gamma-secretase. They present their evidence in the current issue of Nature, dated June 8, 2000, which reports: "Photoactivated beta-secretase inhibitors directed to the active site covalently label presenilin 1."

Two years ago, when other researchers inactivated presenilin-1 genes in mouse neurons, beta-secretase and presenilin expression were both abolished. A commentary on the Nature paper points out: "The development of clinically useful beta-secretase inhibitors is likely to be long and tedious. For instance, the inhibitors used by [the Merck co-authors] do not cross cell membranes efficiently, and will probably be of little use in vivo. The good news is that pharmaceutical companies now have another defined target for their drug-discovery machineries."

In the June 1, 2000, issue of Nature, neuropathologists at Case Western Reserve University, in Cleveland, Ohio, showed that a peculiar form of b-amyloid that lacks its first 10 amino acids, accumulates in the brains of AD patients carrying presenilin-1 mutations. They develop a form of the disease with an earlier onset and shorter duration, with unusual clinical symptoms. In a paper titled "Presenilin-1 mutations in Alzheimer's disease," they propose that "mutations of the presenilin-1 gene influence both beta-and gamma-secretase activities.

For 'Most Comprehensive Model' Of Alzheimer's, Mice Got Anti-Nerve-Growth Antibody Transgenes

While many animal models of Alzheimer's disease exist, none mimic the full range of symptoms typically seen in human AD patients. Now neuroscientists at the International School of Advanced Studies in Trieste, Italy, have tackled this deficit from offside. Reasoning that nerve growth factor (NGF) is presumed to have a hand in age-related neurodegenerative disease, they raised a strain of transgenic mice expressing a high-dose, anti-NGF recombinant antibody.

In the Proceedings of the National Academy of Sciences (PNAS) dated June 6, 2000, the group reports: "Alzheimer-like neurodegeneration in aged antinerve growth factor transgenic mice."

Their paper "demonstrates that these mice acquire an age-dependent neuropathology including amyloid plaques, tau protein [an AD hallmark] and neurofibrillary tangles in cortical and hippocampal neurons. These mice, they conclude, "represent, to our knowledge, the most comprehensive animal model for this severe neurodegenerative disease."

They point out that earlier efforts to achieve this result failed because "disrupting the NGF gene by homologous recombination led to the animals' early postnatal death." The co-authors obtained their anti-NGF mice by manipulating the genes of monoclonal antibodies directed at this factor. Besides tracking the neuronal and cerebral deficits of advancing murine old age, the transgenic animals also displayed deficiencies of cognition, memory and other behavioral attributes of AD.

Knocking Out Key Viral Virulence Factor In Mice Elucidates Roles of Interferon, PKR In Host Defense

Viral immunologists at Washington University in St. Louis reveal new mechanisms in the seesaw skirmishing between viral virulence and immune-system responses. Their paper in the Proceedings of the National Academy of Sciences (PNAS), dated May 23, 2000, bears the title "Specific phenotypic restoration of an attenuated virus by knockout of a host resistance gene." An accompanying commentary carries a somewhat trendier heading: " Maneuvering the internetworks of viral neuropathogenesis and evasion of host defense."

When a viral pathogen such as herpes simplex virus (HSV) invades a human host, the frontline defense is mounted by interferon. This antiviral cytokine gets its marching orders from a signaling protein kinase called PKR. The virulent virus deploys a gene product, ICP (infected cell protein) 34.5, which targets that central PKR molecule. Without that 34.5 virulence protein, HSV can't replicate in the mouse central nervous system. Knockout mice, lacking 34.5, attenuate that viral virulence, the paper reports, and identify the mechanism by which that viral protein acts in vivo.

In their KO animal model, ICP34.5 mutant viruses could discriminate in their lethal infectivity between normal and malignant cells - with implications for drafting HSV as a potential tumoricidal agent.

All Bets Are On In 'How Many Genes In Human Genome?' Guessing Game

As public and private genomicists round the track in their race to sequence the entire human genome, guesstimates are heating up as to just how many protein-coding genes there are in the DNA complement of Homo sapiens. The June issue of Nature Genetics offers a line-up of three qualified handicappers, whose studied predictions range from under 30,000 to 120,000 - to wit:

¿ 28,000 to 34,000 is the range reached by France's national genomics facility, Genoscope. Its paper is titled: "Estimate of human gene number provided by genome-wide analysis using Tetraodon nigroviridis [puffer fish] DNA sequence."

¿ Molecular biotechnologists at the University of Washington, Seattle, weigh in with "Analysis of expressed sequence tags [EST] indicates 35,000 human genes." Their actual prediction is 34,700.

¿ Employing the same EST analysis approach, The Institute for Genomic Research (TIGR), Rockville, Maryland, reports "Gene Index analysis of the human genome estimates approximately 120,000 genes."

In an editorial headed "The nature of the number," Nature Genetics observes that "Recent estimates from Incyte, DoubleTwist and Human Genome Sciences [assert] that the human genome contains over 100,000 genes. . . . potentially mute insinuation that commercial interests factor into the equation."

The journal concludes that, ". . . even with a complete, completely annotated genome in hand, we will still face the task of fathoming how it works."