Maladie du Hamburger (et Fosse Septique)

Abstract: Laboratory report from AP Environmental Systems class. Discusses new strains of E-coli bacteria, testing for e-coli bacteria, and resulting high levels of bacteria found in a backyard with a leaky septic system.

Introduction: In 1982, a news strain of Escherichia coli bacteria was found which is more dangerous and than many other strains. Since then, the strain (E. coli 0157:H7) has been identified with outbreaks of illness in such areas as hamburger, salad bars, apple cider, and swimming beaches. While most strains are harmless to humans, some (e.g., E. coli 0157:H7) can lead to death or injury. The symptoms of the infection include bloody diarrhea, hemorrhaging, cramps, and hemolytic uremic syndrome. The bacterium, when not in areas where it is supposed to be, produces a toxin detectable through Shiga toxin testing or pulsed-field gel electrophoresis.

Sources of E. coli bacteria are many. The average human excretes around 2 billion E. coli bacteria each day. Thus, leaking septic systems are one potential source of e. coli bacteria in water systems. Steve Smet of the Illinois Department of Public Health stated, "One bad (septic) system could lead to the results they saw," in reference to faulty septic systems leaking into Fyre Lake, Illinois. At another Illinois lake, toilet facilities at a beach, potable water supplies for showers and drinking, and cattle and sewage outlets were suspected, but eliminated as authorities suspected people of excreting directly into the water at the beach.

E. coli bacteria are not so much harmful in and of themselves, but rather an indicator organism for fecal contamination of water.

Although regulations vary by each state's health department, the Illinois Health Department suggests levels of coliform bacteria (i.e. fecal bacteria, E. coli) below 500 colonies per 100 mL. If readings of greater than 500 col/100mL are found on two consecutive days at a beach, the Health Department will close the beach.

Infection rates are not always high. In one Illinois case, levels of 660 and 900 col/10mL were found, but the infection rate was 0.3%, all of whom were children under 12.

Because the water was taken from an area surrounding a malfunctioning septic system, we expected to find significant e-coli infestations. Water from the attached house would enter the system, but bypass the filled tank to go directly into the field, where it would seep into surrounding areas without being broken down.

Experiment: Water for Trial I was collected from an area surrounding an odorous septic system in unincorporated Jackson Twp, Porter County, Indiana less than 24 hours before the test. Trial II water was procured from a small private pond in Valparaiso. Because my group forgot to bring water, we must rely on John Baumann's observations of the pond.

We filled inoculating needles with 1.0 mL of the sample water for Trials I and II, reading from the tip of the plunger in the inoculating needle, which was sterile for the first test. We were supplied with test dishes filled with prepared agar from a commercially available solution.

The agar (also known as agar-agar) was made from ME B. Generally, agars are gelatinous extracts of sea-weed used for bacterial cultures. This particular agar will cause E. coli bacteria to turn a metallic green.

The dishes were opened as slightly as possible and the water was dripped in quickly, within the time of 15 seconds. The dishes were then closed and swished from side to side to scatter the water sample over the complete surface area of the agar. Finally, the dishes were placed for 231/2 in an incubator to grow the bacteria.

Following Trial I, the needle was cleaned by first drawing in soapy water several times, then rinsing with clean water. The hottest tap water possible was used, and Procedure was repeated with Baumann's water (Trial II).

After the incubation period, the dishes were removed and the bacteria colonies counted. Only colonies with a metallic green color and a darker "nucleus"-type area were counted. Colonies were counted twice.

Following the count, the agar was scraped from the dish and disposed of.

Formulae: Area of Dish = pr²

Area of a Sector = r²Q

Area of inscribed triangle = 1/2(length of cord)(midpoint of cord to center of dish)

Standard Deviation was calculated using a TI-81 calculator pre-programmed to figure standard deviation.

Calculations: Calculations are shown with Figure 1.

Data: Data were scaled to the area of the dish, around 65 sq cm. For many groups, we scaled it by doubling the area if the area was around half of 65 (32.5).

Results for the supplied water varied tremendously, as the standard deviation of 405 indicates. Indeed, the standard deviation is greater than the average. Results varied from a low of 69 to a high of 1323.

For my group, we found one problem to be settling to one side of the dish. As Figure 1 shows, both dishes had settling to one side. (Figure 1 is shown as the dishes sat in the incubator.) The area in which E. coli actually had to grow was only about 22 sq cm. Large colonies were found further from the center; smaller colonies were towards the outer boundary.

Many different color colonies were found, but only the metallic green colored ones were counted.

Dan's toilet water was low because of his toilet bowl cleaner which probably killed his freshly-excreted E. coli (in the form of a bowel movement).

Several groups tested the same pond water, with results of 51, 236, 614, and 0 colonies/mL. This indicates a great range in results and that the tests we performed (or the testers themselves) are utterly unreliable, possibly because of inexperience.

Well water was negative, as it should be.

Water from a puddle on an asphalt driveway at a local pizzeria tested for 0 and 8 colonies/mL, indicating that rainwater has little E. coli, if any.

Error Analysis: Hoteliers who must send weekly readings to state Health Departments know well how to beat e-coli tests by killing the bacteria first. One former hotelier reported how he beat e-coli tests. He first took his sample on Thursday early in the morning so that chlorine would kill most bacteria and so that the pool would have sat as long as possible prior to the test. He then refrigerated the vial and sent it to Indianapolis on the last mail truck out of town. If the vial arrived on Friday, it would not be tested until afternoon, or if it arrived Saturday, the test would be Monday. Either way, the organisms would have been killed by a combination of cold temperatures, time, or darkness.

Because E. coli bacteria are sensitive to conditions, our results may have been skewed by the conditions of the test. Because we do not know exactly how long and under what conditions the water supplied sat, we can not know if any of these factors intervened. Portable kits can eliminate delays.

Another source of error was the small sample volume. Some commercially available kits use up to 100mL samples. Our 1mL volume may be too small to measure accurately. Also, the small volume did not completely cover the agar, even after significant agitation.

Next, the inoculating needle may have been contaminated during cleaning. The soap, despite at least 5 injections of clean water, may have remained in the needle and killed bacteria in the subsequent test. Although we used the same water as John Baumann's group, John recorded 614 colonies; whereas, we found no colonies. In short, the needle may have been too sterile. One source recommends that an autoclave be used to kill microörganisms, and a bunsen burner between trials. Bunsen burner heating would have killed bacteria without contaminating with soap.

Finally, significant water vapor amounts were suspended atop the dish. Some appears to have fallen onto the agar and diluted the sample.

Group data may be in error for several possible reasons:

• Groups counted bacteria that weren't really E. coli colonies.

• Groups errored in measuring areas of the dish.

• Groups reported data onto the board in improper form without notifying the class.

Conclusions: Some tests indicate that our water is unsafe by Illinois Department of Health standards. However, with ranges of 69 to 1323 col/mL, closing beaches on account of our tests would certainly be controversial (assuming our water was to come from a beach). The data proved to be highly unreliable, and the topic should be subjected to further study under more ideal conditions to be reliable.

In any event, the septic system near where the water was procured should be cleaned or pumped. If any children live in the area, they could suffer the consequences of coming into contact with water in the area.

We have recalled millions of pounds of hamburger because of E. coli, but some people are obviously oblivious to the threat right outside their back door.