Anyone who thinks scientists lead a glamorous, exciting life should talk to Jennifer Snyder, a graduate student at the University of Maryland’s School of Medicine. During July of 2003, while all her friends were out enjoying themselves, Snyder spent 10 days trying to get 40 laboratory mice to urinate into little plastic tubes.
Snyder’s goal was to extract enough genetic material from mouse urine to determine which of the 5,611 genes in a pathogenic strain of E.coli bacteria were turned on, or expressed, in mice with urinary tract infections. The results of her research, conducted in collaboration with scientists from the University of Michigan Medical School and the University of Wisconsin-Madison, will be published in the November 2004 issue of Infection and Immunity.
Results from the study could lead to new, more effective treatments for urinary tract infections, which are the cause of an estimated eight million physician visits in the United States each year. Nearly half of all American women have had at least one urinary tract infection, and know all too well how incapacitating and difficult to treat they can be.
“Even though Escherichia coli’s genome was sequenced in 2001, we didn’t know which genes were active during infection of an animal host or what proteins were present during the infection stage,” says Harry L.T. Mobley, Ph.D., a professor and chair of microbiology and immunology at the University of Michigan Medical School. Mobley directed the study while he was a faculty member at the University of Maryland.
“Now we can quantify the activity of every gene in a representative strain of uropathogenic E. coli, called CFT073, during infection of the mammalian urinary tract,” adds Mobley. “Knowing its gene expression profile essentially gives us a snapshot of what’s happening inside the urinary tract, from the pathogen’s point of view.”
The study was designed to compare gene expression levels for the CFT073 strain of E. coli as the bacteria grew in three different environments. The first was a container of nutrient-rich Luria broth. The second was a container of pooled urine from healthy female human volunteers. The third was inside the urinary tracts of 40 female mice whose bladders were inoculated with CFT073 E. coli.
According to Mobley, mice are often used in research on uropathogenic bacteria, because the urinary system in mice is similar to humans, and mice respond to urinary tract infections in the same way.
To identify the specific E. coli genes that were “turned on” during infection, the scientists first had to collect messenger molecules called RNA. RNA copies genetic information from active genes and carries it to the ribosome, which produces proteins in response to each gene’s coded instructions.
It’s one thing to extract RNA from bacteria in a test tube. It’s not so easy when the bacteria are inside a mouse.
“Collecting the urine was the most labor-intensive part of the study,” says Snyder. “Because the urinary tract is naturally sterile, we knew that the E.coli bacteria we put into the bladder was what would come out in the urine. But it was very important to collect every drop of urine, so we’d have enough RNA to make the experiment work.”
Catheterizing the mice was not an option, because it could introduce foreign bacteria into their bladders. So Snyder developed her own urine collection protocol. Positioning each mouse so its urethra was directly over the open end of a collection tube, Snyder pressed on the back of the mouse’s neck or gently massaged its bladder to stimulate urination.
It took about 45 minutes for Snyder and Virginia Lockatell, a technician and co-author on the study, to extract a precious drop of urine from each of the 40 mice in the study. Eight hours of mouse massage per day produced about 15 milliliters (one-half ounce) of urine, which Snyder processed immediately to stabilize the RNA before it could degrade.
After 10 days, Snyder processed the collected RNA to create DNA copies of the RNA molecules. Working with co-author Rodney A. Welch at the University of Wisconsin Medical School, Snyder used a DNA microarray customized for the CFT073 strain of E. coli to determine which genes were active in each of the three growth media.
The results showed that E.coli had different gene expression profiles when it grew inside the urinary tract, in pooled human urine or in nutrient broth. The researchers used the bacteria growing in Luria broth as the study’s control. They compared the DNA microarray signal intensity from bacteria growing inside the urinary tract of an infected mouse, and in pooled human urine, to the signal intensity of the control bacteria. Only genes with signals at least two times more or less intense than corresponding levels of signal intensity in control bacteria were included in the study.
The DNA microarray data identified 313 E. coli genes that were up-regulated and 207 genes that were down-regulated in mouse urine as compared to Luria broth. Many of the most highly up-regulated genes were known virulence genes in E. coli, but the study also found 13 new candidate virulence genes, which are not found in other strains of non-pathogenic E. coli.
“The most highly expressed genes in mouse urine were translational, indicating that the bacteria were growing rapidly – much more rapidly than in Luria broth,” Mobley says. “We were surprised to see high levels of an adhesion gene called type 1 fimbriae, but hardly any of 11 other fimbrial gene clusters expressed during the infection. And we found unusual sugars taken up by E. coli in the mouse, which were not utilized by the bacteria in other growth media.”
In addition to learning what the bacteria were doing inside the urinary tract, the DNA microarray data also provided valuable information about growing conditions E. coli encountered there, and how they differed from conditions outside a living animal.
“Now that we know the active virulence factors in the animal host and the specific proteins expressed on E. coli’s outer membrane during infection, researchers will have specific targets for the development of vaccines or therapeutic drugs designed to kill invading E. coli bacteria,” Mobley adds.
The study was funded by the National Institutes of Health. Additional collaborators on the study include Brian J. Haugen from the University of Wisconsin-Madison, and Eric L. Buckles, David E. Johnson and Michael S. Donnenberg from the University of Maryland School of Medicine.