The adult human body generates a gram a day of an antibiotic-like molecule called defensin.

"Defensins," explained molecular geneticist Charles Bevins, "are membrane-active peptides that can disrupt membrane function by poking holes in them. They are largely secreted by the immune system's scavenging cells, the neutrophils, which are the most abundant white blood cells in the circulation," he observed. "Five percent of a human neutrophil's weight consists of defensin peptides.

"These defensins could be regarded as a drug," Bevins continued, adding, "I refer to them as endogenous antibiotics.' That gram a day is a typical dose of the widely prescribed antibiotic, ampicillin. So in effect, we humans make a lot of defensin antibiotic."

Bevins, a staff scientist at the Cleveland Clinic Foundation, is senior author of an article in the current issue of Nature, dated March 19, 2003. Its title: "Protection against enteric salmonellosis in transgenic mice expressing a human intestinal defensin."

"I think one of the most important findings in this article," Bevins told BioWorld Today, "is that it serves as a proof of principle for the relative effectiveness of defensins in mucosal host defense. The field is expanding very rapidly; there's a lot of in vitro data supporting that these evolutionarily ancient molecules are important in host defense. But in vivo testing of their true effectiveness is minimal. So I think this Nature paper fills an important cog in that process.

"As we better understand the defensins," he went on, "we will have a clearer picture as to the mechanisms by which humans and other mammals - other animals - interact with the microbial environment in which we live. In the future, two things may spring out of it: One is that deficiencies or defects in the body's ability to make defensins could likely contribute to disease.

"And finally," Bevins suggested, "this is pie in the sky, but the neat thing about some of these host-defense molecules is that they are encoded by conventional genes, in our study, HD-5. So it's conceivable that future therapies could be generated from the transfer of genes into somatic cells in efforts to combat infectious diseases."

Neutrophils Meet Intestinal Paneth Cells

Another fruitful source of human defensins are the so-called Paneth cells, which stud the lining of the small intestine, where Salmonella causes food poisoning in people. Salmonella typhimurium is the infectious microbe to which the Cleveland co-authors exposed their experimental transgenic mice.

"Nita Salzman, the first author of this Nature article," Bevins noted, "has showed in a separate study that Salmonella - as part of the bug's virulence mechanism - actually tries to inhibit the Paneth cells' ability to make defensin peptides. So it's like a little war that goes on between the microbe and the host.

"In designing our mouse experiments," Bevins recounted, "we were very fortunate in determining that the human defensin was expressed in Paneth cells at about the right level. So we challenged those mice with S. typhimurium. We knew from early experiments that transgenic rodents seemed to survive better than the wild types [WT]. Many investigators study microbial infection in mice highly susceptible to Salmonella infection. So relatively low-challenge doses of the bacterium will cause a typhoid-like disease, to which the mice succumb.

"The rodent strain that we were using is not particularly susceptible to Salmonella," Bevins pointed out. "So we had to give very high doses of the germ in order to see any sort of disease manifestations. If we gave low doses, the WT mice wouldn't blink at the challenge. So when we administered particularly high doses of Salmonella, we saw a dramatic phenotype. WT types succumbed; they died from the high bacterial inoculum, whereas in the same circumstances the transgenics survived.

"As for the mechanism, we counted how many Salmonella numbers survived in the intestinal lumen. We found that far fewer bacteria remained alive in the lumens of the transgenics, compared to the WTs. So putting the results together we concluded that the HD-5 gene expression in the small intestine gave some extra killing power to the weapons that the mouse immune system used against the oral inoculation of bacteria. It reduced their number, so the transgenic mice were able to handle the inoculum because a few bugs out there in the lumen were trying to get across that mucosal membrane surface to cause disease. The WT mice had higher numbers of luminal bacteria, and were unable to successfully meet their lethal challenge."

Bevins addressed the question: "Viewed as a body-built antibiotic, do defensins risk attenuation or total decay by the bacterium's drug-resistance genes?"

Are Defensins At Risk Of Bacterial Drug Resistance?

"The answer," Bevins allowed, "is sort of.' Not conventional antibiotic resistance, the way we think of resistance, for instance, to ampicillin. The fuller answer is that it's more difficult for bacteria to devise means of resistance to antimicrobial peptides, such as defensins, than it is to some of the more conventional antibiotics. If you take one of the normal steps that people take for FDA approval of an antibiotic, where they achieve 90 percent killing of bacteria, then let them grow back up and do 14 such cycles, they do not generate bugs that are resistant to the antimicrobial peptides.

"Another fascinating way that bacteria seem to target defensin action," Bevins said, "is that certain bacteria can actually inject proteins into eukaryotic cells - into the host cells - and impair the ability of those cells to make defensin peptides.

"We showed last year," he recalled, "that the Paneth cell defensins in humans are expressed and stored as an inactive propeptide. They're made as kind of a masked precursor antibiotic ready to go. And they require proteolytic activation. In humans we found that activating enzyme was made in Paneth cells. It was a version of a common enzyme, trypsin, that's expressed in Paneth cells.

"Interestingly, the trypsin is also stored as an inactive proenzyme. So it, too, requires activation. Some of our studies are asking: What gets the whole process started?' After the Paneth cells release their contents, which contain inactive defensin and inactive trypsin, when do they get activated in humans? That's one of the issues," Bevins concluded, "we are investigating now."