Few things in life, most people will agree, can be worse than a hole inthe head. The exceptions are patients with certain neurologic diseases,for whom an aperture drilled in the skull may offer the opening to abetter life.Through such artificial orifices, neurosurgeons at Sweden's KarolinskaInstitute have administered recombinant nerve growth factorexperimentally, through the skull, to treat Alzheimer's disease. U.S.scientists are tooling up for similar trials, but fear technicalcomplications, because such invasive therapy is likely to be chronic.Compared to the futuristic nature of such recombinant molecules, thebrain surgeon's rotary cutting tool for perforating the skull isconceptually on a par with the stone hand-ax prehistoric man used torelieve headaches by chopping a hole in the head.Both are circumventing a basic law of nature, namely: `Large-moleculecompounds shall not pass from the bloodstream into the brain.' A goodthing, too; otherwise, every blood-borne pathogen would quickly infectthe vulnerable cells of the cerebrum.The locked gates to this blood-brain barrier consist of a seamless layerof endothelial cells with impermeable tight junctions.There must be a better way than a hole in the head, neuroscientists havemused, to invade the cerebral cavity with valid therapeutic intent. Oneroute, said biochemist Philip Friden, would be to hijack the vehicle bywhich iron compounds are smuggled past the barrier into the brain.Antibody Bypasses The BarrierFriden, associate scientific director of Alkermes Inc., is the principalauthor of a paper, dated Sept. 13, in the current Proceedings of theNational Academy of Sciences (PNAS). Its many-faceted title tells itsstory: "Intravenous administration of a transferrin receptor antibody-nerve growth factor conjugate prevents the degeneration of cholinergicstriatal neurons in a model of Huntington's disease."Friden told BioWorld Today what Alkermes, and his co-authors inChicago have achieved so far:Transferrin, as its name implies, is a natural blood protein by which thebody ferries iron across the barrier into the brain. "Iron is a requiredco-factor for enzymes mediating various metabolic processes in thebrain. Normally, the iron-transferrin complex binds to the transferrinreceptor on the endothelial cells, and are then transported to the brain,by some not-yet-defined mechanism," he said.To an antibody targeting that transferrin receptor, Alkermes hitchednerve growth factor (NGF). The antibody, like transferrin itself,coupled to the receptor, and took NGF along for the ride across thebarrier, via infusion into the jugular veins of rats.Not just any rats, but rodents made to model the nerve degenerationpeculiar to victims of Huntington's disease (HD). It takes humansmany years to develop the symptoms of HD, and as a rule brainautopsies to confirm that their affliction was indeed Huntington'srather than some other neurodegenerative malady.Anesthetized rats succumbed to neurochemical HD more quickly,following a trans-cerebral (hole-in-the-head) injection of quinolinicacid, aimed at the brain's striatal region. "Quinolinic acid," Friden said,"is an excitotoxin. It binds to certain nerve cells, and over-stimulatesthem to death."He added, "This chemical lesion may not be the same mechanism as inHD, but the population of neurons, including cells secreting cholineacetyltransferase (ChAT), that we saw die is very similar to those samecells lost in Huntington's."HD-model animals protected by the conjugated antibody-NGF infusionlost only 24 percent of their ChAT neurons, on average. Bycomparison, rats treated with an unconjugated mixture lost 43 percent,and untreated controls, 39 percent.This result, Friden emphasized, does not mean that a potentialtreatment for Huntington's is at hand: "In HD," he said, "a number ofdifferent cell types are lost. NGF is specific only for the cholinergic-neuron class. There's a whole host of other neurotrophic factors, forexample, brain-derived nerve factor (BDNF), glial-cell-derived nervefactor (GDNF)."The rodent experiments were carried out by Friden's collaborator,Jeffrey Kardower (the PNAS paper's first author) at Rush-PresbyterianMedical Center in Chicago.Alkermes' present purpose, Friden said, "is to show, using the generalneurotrophic factor NGF as a model, that we can deliver a factorperipherally to lesions with a specific loss of cholinergic factors, suchas Alzheimer's disease.""Primates," said Friden, "are the next species to look at before we'dever want to do anything in humans." Since early last year, Kardowerhas been attempting to go beyond the rat results in a score ofcynomolgus monkeys. One salient protocol change from the rat modelis using a human rather than a rodent transferrin-receptor antibody,which cross-reacts with primates, "and will eventually go intohumans.""We're getting some pretty interesting results in the monkeys," Fridenallowed. "If everything holds up, we'll be writing up our analysis in thenext few months."Alkermes' spokeswoman, Donna LaVoie, observed, "Obviously, weare looking to partner this program, for which we already have twoissued patents, with another pharmaceutical or biotech companyworking with neurotrophic factors." She said that the company "hassome research collaboration ongoing right now with a couple of[unnamed] companies."Friden mentioned that Amgen Inc. and Regeneron Pharmaceuticals Inc.are developing BDNF; Genentech Inc., NGF; Synergen Inc., GDNF.He noted that Synergen's glial-cell-derived nerve factor "seems to bespecific for the neurons that are lost in Parkinson's disease."He added, "They all want to get their factors into the brain by a non-invasive method. Our approach is the way to go, because we can give itby a shot in the arm, without drilling a hole in the head."Neurobiologist Fred Gage of the Scripps Cinic and Research Institute,La Jolla, Calif., welcomes Alkermes' peripheral administrationtechnique, but cautions that the rat model of Huntington's disease,which involved injecting an excitotoxin through the skull, raises aproblem."The concern there," Gage told BioWorld Today, "is that once yourmodel compromises the blood-brain barrier, and then you administersomething peripherally, that protein is going to have easier access. Itwouldn't be analogous to the normal situation, where the brain isactually intact."But, he also observed, the finding that the antibody-conjugated NGFworked better than non-conjugated "basically sounds really exciting." n

-- David N. Leff Science Editor

(c) 1997 American Health Consultants. All rights reserved.