By David N. Leff

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

Escherichia coli is a bacterium that leads a double life - commensal and pathogenic. In its first guise, the bug colonizes the human gut, where among other useful functions it makes vitamin K, which helps stem hemorrhaging. In its sinister role, E. coli can infect foodstuffs, notably meat, and cause sometimes-fatal food poisoning - not to mention travelers' diarrhea.

Such contamination usually is transmitted by oral ingestion of fecal wastes - testimony to the bacterium's natural habitat in the intestines. But there's another route of transmission, which afflicts 1 in 3 American women before the age of 65. It's urinary tract infection (UTI). That tract - urine's conduit from bladder to micturition - is the urethra, nature's own built-in catheter, as it were.

In men, that tube follows a long, zigzag trajectory from bladder to urethral meatus at the tip of the penis. In women, the path is much shorter, ending with the urethral orifice in the vaginal wall.

From here to the nearby anus extends a short stretch of anatomy between the thighs, called the perineum, which urologists suggest is prone to fecal soiling. Ergo, women are more likely than men to suffer UTI. That suffering features among its hallmark symptoms intense pain, burning and frequent urination. Antibiotics can usually quell a bout of UTI, but the infection frequently stages a comeback - with a 25 percent chance of recurrence - often several attacks a year. And the strains of E. coli that specialize in UTI don't fight fair.

A paper in the Proceedings of the National Academy of Sciences (PNAS), dated Aug. 1, 2000, updates the bacterium's treacherous survivorship. Its title: "Bad bugs and beleaguered bladders: Interplay between uropathogenic Escherichia coli and innate host defenses."

Moving upstream into the bladder, the bacteria latch on to that organ's rugose inner surface by grappling hooks called pili. These hair-like projections deploy proteins at their tips that stick to the bladder lining like Velcro. That adherence saves the bacterium from being washed out by the downward flow of urine, but it also sends an immunological heads-up of a microbial invader to the bladder epithelial cells.

The PNAS paper reports that the bacteria can lurk dormant - or replicating - in the bladder lining, and serve as a reservoir for recurrent UTI. When physicians no longer find dead bacteria and sloughed-off bladder cells in a patient's urine sample, they conclude that their antibiotic treatment must have been successful. Then, if infection recurs, they presume that fresh E. coli from the gut, perineum or elsewhere have invaded the bladder again. In fact, it may have been colonized all along by latently hidden bugs - never hindered by their victim's immune defenses, or drug therapy.

"Future research," the paper concludes, "is aimed at identifying bacterial and host factors that can exacerbate or attenuate the severity of UTIs. It is hoped that such studies will eventually lead to more efficacious antibacterial therapeutics."

Microchips Capture Breast Tumor Gene Expression Patterns, Hint At Clinical Import

"Molecular portraits of human breast tumors" is the title of a paper in the Aug. 17, 2000, issue of Nature. Its authors, at Stanford University, suggest that tumors are as varied as the individuals they afflict - both in their natural history and their response to treatment.

They snapped their varied portraits by capturing gene expression patterns in 65 surgical specimens of breast tumors from 42 patients, using complementary DNA microchips representing 8,102 human genes. Twenty of the tumor specimens were sampled twice, before and after a 16-week course of doxorubicin chemotherapy.

"An important implication of this study," the paper pointed out, "is that the clinical designation of 'estrogen receptor negative' breast carcinoma encompasses at least two biologically distinct subtypes of tumors (basal-like and Erb-B2 positive), which may need to be treated as distinct diseases."

They made the point that their data "are far from having a complete picture of the diversity of breast tumors. When hundreds (instead of tens) of breast tumors have been characterized, a more defined tumor classification is likely, and statistically significant relationships with clinical parameters should be uncovered."

Hunt For Non-Familial Breast Cancer Genes Turns Up Linkage On Chromosome 13's Long Arm

It's now five years since discovery of BRCA1 and BRCA2, the two genes that account for familial mammary carcinoma. This inherited form of breast cancer afflicts 1 in 10 patients who have the disease, leaving 90 percent "sporadic" cases. Recent linkage analysis of 237 breast-ovarian cancer families showed 52 percent carrying BRCA1 and 32 percent BRCA2 - hence presumably at risk of acquiring the malignancy.

Now a search for such a risk factor in the somatic cells of sporadic patients canvassed 77 Scandinavian family members with no identified mutations in the two BRCA genes. It discovered a candidate susceptibility locus on the long arm of human chromosome 13. This finding is reported in the Aug. 15, 2000, issue of PNAS, under the heading: "Somatic deletions on hereditary breast cancers implicate [chromosome] 13q21 as a putative novel breast cancer susceptibility locus."

These studies, the paper pointed out, imply that breast cancer families attributable to those two genes "may be smaller than initially thought."