By David N. Leff

People line their closets with cedar veneer, because cedar trees secrete a volatile chemical that discourages moths.

A badly injured African clawed frog (Xenopus laevis) with an open wound will heal nicely in a pool highly contaminated with bacteria.

Even the lowly fungi produce potent antibiotics, notably penicillin, which humans enlist to fight bacterial infections.

And humans, of course, along with mammals in general, deploy an elaborate system of immune defenses, ranging from antibodies to T lymphocytes. Where these fail or fall short, we turn to commercial antibiotics, as a kind of prosthetic back-up.

If frogs and fungi manufacture their own self-defense antibiotics, why can't humans? Molecular biologist Michael Zasloff has found that they can, and do.

A decade ago, Zasloff, then chief of human genetics at the National Institute of Child Health and Human Development, performed that wound-healing frog experiment. In the amphibian's skin, he discovered a family of peptides with broad-spectrum antimicrobial activity, which he named "magainins" (after the Hebrew word for shield).

Zasloff is now executive vice president of Magainin Pharmaceuticals Inc., of Plymouth Meeting, Pa., and a principal author of an article in today's Cell, dated Feb. 21, 1997. Its title: "Human ß-defensin-1 is a salt-sensitive antibiotic in lung that is inactivated in cystic fibrosis."

"Prior to publication of this paper," Zasloff told BioWorld Today, "it was clear to me that the systems that are operative in the frog have to be operative in man, in one transformation or another."

Defensins are a counterpart of magainins. They occur, Zasloff explained, "on all of the wet surfaces in the human body, from airway to digestive tract. In a sense," he added, "what the body is trying to do is keep a microns-thin, viscous layer of mucus that keeps all our wet surfaces antibiotic-rich."

Unlike commercial antibiotics, Zasloff said, "defensins are peptides, about 30 amino acids long. "They attack the membranes of bacteria in such a way as to damage the pathogens' permeability barrier. This is a very difficult mechanism for a microorganism to overcome."

Defensins come in two main classes, alpha and beta, set apart by their differing molecular folding structures. Alpha defensins, Zasloff said, "represent part of the killing equipment of white blood cells, principally neutrophils."

Beta defensins, he added, "are produced on virtually all the epithelial wet surfaces, whether mouth, nose, prostate or urinary tract. Those were the peptides used in our experiments described in the Cell paper."

Those surfaces line the epithelial-cell inner walls of the respiratory and digestive tracts and shield them from infection.

In patients with cystic fibrosis (CF), this shield crumbles.

"The basis for mortality in CF," Zasloff said, "is the progressive destruction of the lungs by bacterial attack. This leads to respiratory failure, chronic infection and eventually death. It isn't that CF sufferers don't make defensins; they do. But the endogenous antibiotic is being secreted into a fluid mucus layer that is too salty."

Faulty regulation of sodium chloride trafficking in the body is a hallmark of cystic fibrosis. Some 30 years ago, Zasloff recalled, an Italian mother told her CF child's physician that the youngster's perspiration was extremely salty.

From this anecdotal observation came the sweat test, now widely used to differentially diagnose the genetic disease by measuring its sodium chloride concentration.

Zasloff noted, "If fluid produced by epithelial cells from a normal person's airway is incubated with Pseudomonas aeruginosa or Staphylococcus aureus * two ubiquitous bacteria not usually harmful to man, but dire in CF * those germs will die. But the bacteria shrug off the same fluid from airway secretions of a CF patient.

"Now," he continued, "if we dilute the salt content of the CF fluid to the normal concentration, the antibiotic activity returns, and kills the bacteria.

"That demonstration of why defensins don't defend the lungs of CF patients from destruction," Zasloff said, "was done not with fluid from a human being, but from xenograft mouse models, the work of James Wilson and his group at the University of Pennsylvania, in Philadelphia." Gene therapist Wilson is the Cell paper's other senior author.

To make their murine surrogate model of a human airway surface, the Penn team started with a series of rat tracheas, tiny tubes of cartilage devoid of epithelial cells. They populated those inert cylinders with airway cells from humans, with or without CF, and embedded these grafts under the skins of nude mice.

The implants grew into living pseudo-human tracheas, completely covered by intact layers of respiratory epithelial cells actively secreting the mucosal fluid. With that liquid, the Penn/Magainin teams were able to demonstrate:

* that the cells in those xenografts were making human defensin, and in abundance;

* that synthetic defensin had some activity at the normal concentration of airway fluid, but none at all at the salt conditions of CF airways;

* that normal and CF lungs alike express Magainin's complementary DNA human ß-defensin-1 gene, across the entire respiratory tract, from upper airway down to pulmonary alveoli.

"The very last experiment in the paper," Zasloff observed, "which was absolutely critical, reconstituted the rat trachea with normal human cells, in which Jim Wilson's antisense oligos specifically inhibited the defensin gene and its expression of the defensin peptide. So when we took the airway fluid out of that trachea, 99 percent of its antibiotic activity had been eliminated."

Multiple Medical Implications Arise

"Therapeutic intervention becomes very fascinating now," Zasloff said. "Wilson's group at Penn has the belief, or the hope, that it will be possible to correct the CFTR gene subtly. Complete 100-percent correction, apparently, is not needed to adequately regulate the salt environment of that airway mucus layer, as demonstrated in their xenografts. And they would like to introduce a modified defensin gene into that CF airway by gene therapy."

Magainin is working the other side of the CF street.

"We are still continuing on our mission to introduce an aerosolized peptide as a form of replacement. We have generated many peptides that exhibit the appropriate broad-spectrum activity and salt independence, against which bacteria will find it exceedingly difficult to develop resistance." *