On September 14, a Friday, Albuquerque officials presented Paul Torres and colleagues at the New Mexico Scientific Laboratory Division with two bags of spinach—one fresh, one frozen—that had been retrieved from the kitchens of patients suffering E. coli O157:H7 illness. It was an extraordinary turn of good fortune.
Recovering a viable form of E. coli from the spinach, however, remained a formidable task given that bacterial pathogens are generally present in low numbers in food samples and mixed in amongst millions of other types of microbes.
The frozen sample was ruled out immediately since water inside the bacteria expands upon freezing, lysing the organisms. So Torres, a supervisor in the lab’s environmental microbiology section, and his team of scientists set up to recover E. coli the fresh spinach using the standard method outlined in the US Food and Drug Administration’s Bacteriological Analytical Manual (BAM).
Because, said Torres, “we wanted to get it right,” the scientists also called Steve Weagant, an E. coli O157 expert at the FDA’s Pacific Regional Laboratory Northwest.
Weagant suggested that they try a second approach, the Food Emergency Response Network (FERN) method. This method was still being validated for use with spinach, but said Torres, “(Weagant) thought it was superior.”
The team proceeded with both methods, using a similar two-step process. First, scientists set up pre-enrichment broths and incubated spinach samples to revive any injured or dying microbial cells and get them reproducing. Then, they kicked up the temperature to the extremes preferred by E. coli and added selective antimicrobial agents to kill everything except E. coli—a “two-punch on the selection process.”
By Saturday, it was clear that the BAM method had failed to yield viable E. coli O157:H7. But by Sunday it appeared that the FERN method had worked. Torres’ team had grown magenta-stained colonies of something that could be E. coli O157:H7.
To verify, they characterized the organisms using four different test methods:
• Antigenic analysis—to confirm the presence of the O157 and H7 glycoprotein markers for which the organism is named.
• Biochemical analysis—to test for characteristic reactions with various enzyme-carbohydrate mixes.
• Polymerase chain reaction (a DNA amplification method)—to identify serotype.
• Pulsed-field gel electrophoresis—to generate a DNA “fingerprint” identifying the organism’s subtype.
Every test was consistent with the outbreak strain of E. coli O157:H7.
The value of the finding lay in the food packaging. Immediately, CDC, FDA and state investigators began to trace the origins of the spinach.
On September 21, the day after the New Mexico state public health laboratory announced its finding, the FDA narrowed the source to three counties in California’s Salinas Valley agricultural corridor and released a statement saying that spinach grown in “non-implicated areas” was safe to eat. On September 29, the agency announced that all spinach associated with the outbreak traced back to one grower. And on October 12, the California State PHL announced that it had identified the outbreak strain of E. coli O157:H7 in samples of cow manure collected near the spinach fields.
Public health laboratory data had broken the investigation wide open.