The Microbrewery Laboratory Manual -
Part II: Bacteria Detection, Enumeration, and Identification
By Fal Allen
Republished from BrewingTechniques' September/October 1994.
Are your beer and brewery clean? The only way to be sure is to test, test, test. This article shows you how.
Bacteria are almost everywhere. They are in your floor drain (by the billions), on your floor, in the air, and all over your hands. Many kinds exist: aerobic, anaerobic, Gram positive, Gram negative, wort spoilers, beer spoilers, acid bacteria, and spore forming. It's enough to drive a brewer insane! But don't panic. Unless your brewing techniques are poor, you shouldn't have to worry about most of these bacteria. A few simple things can be done to detect and deal with the more problematic ones before they cause a major problem.
Laboratory work should be preventive, not reactive. A good brewer will detect a bacterial infection when it is low grade, before it has caused serious problems. The only way to do that is by regular, ongoing lab work. Once you can taste a bacterial infection, it is better to sewer the beer because the flavors will only get worse as the beer ages out in the market.
KNOW THY ENEMYTo eliminate the infecting bacteria, you need to know at what stage of your brewing process it was introduced. It will also be helpful to know what kind of bacteria it is, so you have a better idea of how to deal with it. Certain types of bacteria are more susceptible to specific types of sanitation.
Even the cleanest brewery plays host to bacteria, and the beer, at some point, can come into contact with some of them. This can happen at any stage of the brewing process, from the mash tun to the packaged product.
Part of our job as brewers is to minimize the number of bacteria that contact our beer. Periodic testing of various areas of the brewery will give you an idea of how good a job you are doing at keeping these bacteria at bay. You can test samples of beer at any stage of production and can swab areas that are supposed to be sanitary to see if your cleaning regime is doing its job.
Because bacteria can be introduced at any stage of production, it is important to test the beer at every stage. These tests can be very informative. If bacteria show up in a conditioning or bright beer tank but not in the fermentor, for example, you can be sure that the bacteria are being introduced somewhere downstream of the fermentor, probably in the hoses or the conditioning tank itself. This information gives you a better idea of how to attack the problem.
This installment of a four-part series on laboratory procedures for small-scale breweries discusses selective media testing of beer samples taken from fermentors and conditioning and bright beer tanks, membrane filtration of packaged product, forced wort testing, and the swabbing of clean areas to measure the effectiveness of the cleaning and sanitation techniques used. Last, a discussion of differential staining - used to help determine a bacterium's identity or, at minimum, its general classification - provides instruction for the last step to all of the testing methods discussed in this article.
SELECTIVE MEDIASelective media provide an easy way to check for bacteria. Many types of media are used, some for growing yeast and general bacteria, some with yeast inhibitors, some with Saccharomyces-specific inhibitors. (The next installment will discuss wild yeast detection.)
HLP medium: J.E. Siebel Sons' Co (Chicago) makes several good media. I have found that one of the best and easiest to use is the Siebel HLP medium. HLP stands for Hsu's Lactobacillus and Pediococcus medium, and it selects for, but is not specific to, those bacteria. Lactobacillus and Pediococcus are the most common types of beer-spoiling bacteria.
HLP is easy to prepare and use. It requires no autoclave and will detect both aerobic and anaerobic bacteria. It contains a yeast inhibitor to prevent the yeast from growing; any growth you see is bacteria.
Preparation of HLP medium: To prepare the HLP medium, suspend 7 g of powdered medium in 100 mL of distilled water (I have found that tap water works fine) in a flask (for an overview of the materials needed, see the equipment list in the accompanying box). Close the flask loosely with foil on top, a cotton plug, or other permeable closure. Bring this solution to a boil and boil for 2-3 min while swirling to avoid scorching. While the medium is still hot, place 9 mL in each of 11 sterile test tubes and cap. (Note: You may need to double up and prepare enough for 20 test tubes.) I like 16 X 150 mm screw-top test tubes. The prepared medium can be used that day, or you can store the test tubes for later use; store at 32-41 °F (0-5 °C) and reliquefy with hot water before use.
Sterile sampling: To use HLP medium, you first have to take sterile samples. I like to sample from a few random fermentors, all of the conditioning tanks, and the plate chiller. I also like to have one blank, or control test tube (no sample added) and one positive (inoculated with a sample from the floor drain).
You can put the samples into either sterile test tubes or flasks. I prefer to use test tubes with screw-top closures. Use one test tube for each sample. Never reuse a test tube until it has been cleaned and autoclaved. These samples will be added to the test tubes you previously prepared with HLP medium.
Label each of the HLP test tubes with the number of the corresponding tank to be sampled, and then label each sample collection test tube to match (see Figure 1).
Before taking samples, make sure all your equipment is available and ready, tie back your long hair, and remove your jewelry, rings, and wrist watch. Make sure the area is free of traffic. Close off the area if possible. Wash down your hands and forearms with 70% ethanol. Now you are ready.
First, sterilize the sample valve. Spray ethanol on and into the valve. With your hand torch, flame the valve inside and out. Do not overheat the valve, because overheating could damage it. Then start a slow stream of beer flowing. Count 1 second and then take your sample. You do not want the very first runnings, because they will be "cooked" by their contact with the hot valve. Close the test tube lid - remember to work as quickly as you can to minimize airborne contamination. Resanitize the sample valve and move on to the next one. Collect all of your samples before proceeding to the next step.
Adding sample to HLP medium: The HLP medium must be in a liquid state (75-85 °F; 24-30 °C). Starting with collection sample #1, pipette out 1 mL of sample and transfer it to the appropriately marked HLP test tube (Figure 1). Recap the tube containing both HLP and sample. Slowly invert it twice to distribute the sample evenly throughout the medium. Repeat for all samples using a clean, sterile pipette for each transfer.
Place the test tubes upright in a test tube rack, and place the rack in the incubator (or homemade temperature-controlled box) at 85 °F (30 °C). Examine for growth after 48 h and then again after 72-96 h. If you suspect that your test sample is heavily contaminated with acetic acid bacteria (vinegary off-flavors), you can overlay 2-4 mL of sterile paraffin wax on the surface of the medium after inoculation to suppress the growth of these bacteria and enable you to get clear HLP results.
Reading HLP tests: HLP medium is designed to grow Lactobacillus and Pediococcus bacteria. In addition to the yeast inhibitor that keeps yeast from growing, it forms an oxygen barrier on the surface so that anaerobic bacteria (such as Lactobacillus and Pediococcus) will be able to grow below that. Aerobic bacteria can grow on the surface of the media.
Look for the growth of colonies suspended within the medium and on the surface. They will be oval to round in shape and will continue to increase in size over time. This is one of the reasons for taking readings at 48 h and then again at 72 h. If you have any doubts, look at the sample you took from the floor drain; it will have grown one or more kinds of bacteria. If it shows no bacterial growth, either you have the cleanest floor drain on earth or something went wrong in the previous steps. Also, carefully examine the blank or control test tube. Any sign of bacterial growth indicates that the medium was somehow contaminated and all your results should be considered suspect. Retrace your steps and look for mistakes, then start again. I have never had either my positive or negative control test tubes show anything but the expected results.
This type of test works well in most cases, but you are only testing a 1-mL sample. In some areas of the brewery it is best to take a larger sample. In areas like the exit port of your wort chiller, for example, even a very small amount of bacteria can infect an entire batch of beer. Similarly, packaged product may not be consumed right away, and even a small infection will show itself over time.
MEMBRANE FILTRATIONThe accompanying box lists the equipment needed for membrane filtration testing. I use a Nalgene filter holder with receiver (product number #28199-440; Nalge Company, Rochester, New York). This filter holder will accept several sizes of filters. I use a 0.45-Ám filter from Millipore (Bedford, Massachusetts); 0.45 will catch all beer bacteria as well as any yeast in the beer. I also modified a picnic hand pump from Fox Equipment (Kansas City, Missouri) as a vacuum pump. I opened the pump and reversed the pump disc, transforming it from a pressure pump to a vacuum pump. It cost one-third of what you would pay in a scientific catalog, and it works just as well.
Membrane filtration examines areas where you expect to find low concentrations of bacteria; for example, brewery air, rinse water, or packaged product. By running a relatively large volume through a filter, you catch anything that is in that volume. Plating the filter sheet reveals what, if anything, grows, and in what quantity. The medium you use depends on what you are looking for. A sterile, filtered product should show no growth of yeast or bacteria; you might want to use a general growth media to look for either of these.
Next, open the top and add the 12-oz beer sample, pouring gently to minimize foaming. It should not be degassed. Cover promptly and stop up air entry ports with sterile cotton or put a slight amount of carbon dioxide pressure on to facilitate the flow (1-3 psi works well). Now use the vacuum pump to draw the liquid through the filter.
When the entire 12-oz sample has passed through the filter, remove the filter with sterile tweezers and place it on a petri dish with HLP medium (prepared as described above). Make sure that the filter is covered with medium. You may have to overlay the filter with a little extra medium after you place it in the petri dish. If you are looking for anaerobic bacteria, it may be necessary to incubate anaerobically.
Incubate for 2-4 days at 80 °F (27 °C). Look for signs of growth and record the number of colonies/mL of sample. The grid design on the filter provides a visual aid that makes location and enumeration easier. If colonies grow, you may want to do some further examination under a microscope. You may also want to Gram stain the bacteria for further classification (see "Differential Staining" section below). Knowing what you are dealing with will make it easier to devise corrective measures.
FORCED WORT TESTINGA forced wort test measures the cleanliness of the wort chiller and hard pipe or hosing leading up to the fermentor (before the yeast addition). Samples should be taken after passing through the wort chiller. You will need a way to aseptically sample wort in-line to the fermentor. Generally, this is done either through a rubber membrane that can be pierced by a syringe needle, such as the type that doctors use to take blood samples (Figure 2), or through a sample cock. We have a device we call "Alice," which is an end cap with a small sample valve in the center of it (Figure 3). Alice can be clamped onto a T fitting, and wort can then be sampled in-line.
The box on the next page lists the additional equipment needed for forced wort testing.
Procedure: First, sterilize your equipment. Wrap the entire sampling device in foil and autoclave it (15 min at 250 °F [121 °C] or 15 psi), and then let the apparatus cool. When you are ready to cast back (knock out), hook up your sampling device in-line to the fermentor. Start casting back, and as the first wort passes by, begin to allow the flask to fill up. Fill the flask one-third to one-half full and then close the sample valve (or remove the collection needle). Remove the hose from the sample valve and place the end in a small container of sanitizer to form an airlock. Store at 75-80 °F (~26 °C).
Check for signs of growth after 48 h. Nothing should be growing. Check again after 72-96 h. You may see something at 72 or 96 h, but it should be clear at 48 h. Bubbles coming from the airlock and haze in the middle and upper part of the liquid are signs of growth. The odor should smell clean and malty like fresh wort. If you see growth within 48 h, somewhere upstream of your sample point is not being cleaned and sanitized properly. You may want to plate out some of this wort sample and see what grows. Anytime you find growth in a medium, you can loop it out, streak it onto a slide, and look at it under a microscope. This may give you a better idea of what you're up against.
SWABBINGThis method is used to check for bacteria in areas that should be clean and sterile, such as cleaned-in-place fermentor hose ends, exit ports from hard piping, the plate cooler, or the filter.
Swab the area to be checked for growth. The area should be no larger than 10 cm2. Place the swab into a test tube of medium and seal. Remember to work as rapidly as possible to minimize contact time with the surrounding air. Label all the samples you take and place them upright in a test tube rack. Incubate as per the medium instructions (2-6 days at 75-80 °F (~26 °C). Check for growth on the ends of the swabs. Remember, you are swabbing areas that should be sterile. If you show any significant growth, you should rethink your cleaning and sterilization procedures.
DIFFERENTIAL STAINING (GRAM STAINING)Suppose any of the above tests show positive results for the presence of bacteria. What then? Differential staining will help you to identify - if only in general terms - the identity of the offending organism. Armed with this information, you will be better able to formulate an effective response.
The Gram stain technique was first developed by Christian Gram in 1884 and is used to classify bacteria. It is a method of general identification. Gram stain classification is based on the fact that Gram positive and Gram negative bacteria have differing cell wall composition.
The procedure involves a series of stains and rinses. Gram-positive bacteria cell walls retain stain. Gram- negative bacteria cell walls do not. Again, if you know what kind of bacteria you are dealing with, you will have a better idea of what steps you need to take to combat them. Different bacteria react differently to different sanitizers.
Procedure: Spread a drop or thin layer of bacteria suspension in the middle of a thoroughly clean slide. This drop should cover an area of about 10-20 mm2. Allow to air dry in a dust-free environment. Heat-fix the smear on the slide by quickly passing it through the flame of the Bunsen-type burner two or three times.
Examine under the microscope at your highest magnification (400-1000X). If you use an oil immersion lens, put a drop of immersion oil directly on the slide without a cover slip. Gram-positive bacteria and yeast cells will show as blue-black. Gram-negative bacteria will stain red. This technique is accurate only on young, viable cultures less than 72 h old, and the procedure must be strictly followed to obtain accurate results.
Lactobacillus and Pediococcus bacteria are Gram positive. Acetobacter, acetomonas, zymomonas, obesumbacterium, and coliform bacteria (escherichia and acetobacter) are Gram negative. These bacteria are the eight most common brewery bacterial contaminants (see Table I). For further discussion on each of these bacteria, consult reference 1 (pp. 208-214), reference 2 (pp. 741-759), or reference 3 (pp. 71-79).
ACKNOWLEDGEMENTSI would like to thank the following brewers who gave us hours of free consultation and helped us to develop our practical lab procedures: Jamie Emmerson (Full Sail Brewing, Portland, Oregon), Paul Saylor (Catamount Brewing, White River Junction, Vermont), John Battles (Indianapolis Brewing, Indianapolis, Indiana), Dave Logsden (Wyeast Laboratories, Hood River, Oregon). Thanks also to Toby Malina of ParaGraphics (Seattle, Washington) for the technical drawings.
REFERENCES(1) The Practical Brewer, M. Meilgaard (Master Brewers Association of the Americas, Madison, Wisconsin, 1977).
(2) Malting & Brewing Science, D.E. Briggs, J.S. Hough, R. Stevens, and T.W. Young (Chapman and Hall, London, 1981).
(3) Microbiological Methods (Siebel Institute, Chicago,1994).
FURTHER READING"Wort Composition," in The Practical Brewer, M. Meilgaard (Master Brewers Association of the Americas, Madison, Wisconsin, 1977).
ASBC Methods of Analysis, 8th ed. (American Society of Brewing Chemists, Egan, Minnesota, 1992).
Essential Quality Control for Brewers (Siebel Institute, Chicago, 1994).
|Issue 2.5 Table Of Contents|