Chapter 10: Bacillus Cereus

Updated: 8/1/00

• Potential Food Safety Hazard
• Control Measures
• Growth
• Heat Resistance
• Analytical Procedures
• Food Sampling and Preparation of Sample Homogenate
• Bacillus cereus
• Bacillus cereus Diarrheal Enterotoxin
• Most Probable Number from Serial Dilutions
• Other Analytical Procedures
• Commercial Test Products
• References

Contents

Potential Food Safety Hazard

Food poisoning caused by Bacillus cereus may occur when foods are prepared and held without adequate refrigeration for several h before serving. B. cereus is an aerobic sporeforming bacterium that is commonly found in soil, on vegetables, and in many raw and processed foods. Consumption of foods that contain >10⁶B. cereus/g may result in food poisoning. Foods incriminated in past outbreaks include cooked meat and vegetables, boiled or fried rice, vanilla sauce, custards, soups, and raw vegetable sprouts. Two types of illness have been attributed to the consumption of foods contaminated with B. cereus. The first and better known is characterized by abdominal pain and diarrhea; it has an incubation period of 4-16 h and symptoms that last for 12-24 h. The second, which is characterized by an acute attack of nausea and vomiting, occurs within 1-5 h after consumption of contaminated food; diarrhea is not a common feature in this type of illness (Rhodehamel and Harmon, 1998).

Contents

Control Measures

B. cereus is a common food contaminant. Effective control measures depend on destruction by a heat process and temperature control to prevent spore germination and multiplication of vegetative cells in cooked, ready-to-eat foods. Measures to reduce or eliminate the threat of food poisoning by B. cereus include: 1) Avoid preparing food too far in advance of planned service, 2) Avoid holding cooked foods at room temperature, 3) Use quick chill methods to cool foods below 7.2 ºC (45ºF) within 4 h of preparation; store in shallow pans/small quantities with the food less than 4 inches (10.2 cm) deep; if food is especially thick (e.g., refried beans), store no more than 3 inches [7.6 cm] deep). Hold/store hot foods above 60ºC (140ºF) until served, and 5) Reheat foods rapidly to 74ºC (165ºF) or above (Kramer and Gilbert, 1989; Reed, 1994).

Contents

FDA Guideline

FDA to assess situations on a case by case basis.

Contents

Growth

Table 10-1. Limiting conditions for B. cereus growth.

Contents

Heat Resistance

Table 10-2. B. cereus spore heat resistance.

Contents

Analytical Procedures

Contents

Food Sampling and Preparation of Sample Homogenate

Food Sampling and Preparation of Sample Homogenate (Andrews and June, 1998)

The adequacy and condition of the sample or specimen received for examination are of primary importance. If samples are improperly collected and mishandled or are not representative of the sampled lot, the laboratory results will be meaningless. Because interpretations about a large consignment of food are based on a relatively small sample of the lot, established sampling procedures must be applied uniformly. A representative sample is essential when pathogens or toxins are sparsely distributed within the food or when disposal of a food shipment depends on the demonstrated bacterial content in relation to a legal standard.

The number of units that comprise a representative sample from a designated lot of a food product must be statistically significant. The composition and nature of each lot affects the homogeneity and uniformity of the total sample mass. The proper statistical sampling procedure, according to whether the food is solid, semisolid, viscous, or liquid, must be determined by the collector at the time of sampling by using the Investigations Operation Manual (FDA, 1993). Sampling and sample plans are discussed in detail in ICMSF (1986).

Whenever possible, submit samples to the laboratory in the original unopened containers. If products are in bulk or in containers too large for submission to the laboratory, transfer representative portions to sterile containers under aseptic conditions. There can be no compromise in the use of sterile sampling equipment and the use of aseptic technique. Sterilize one-piece stainless steel spoons, forceps, spatulas, and scissors in an autoclave or dry-heat oven. Use of a propane torch or dipping the instrument in alcohol and igniting is dangerous and may be inadequate for sterilizing equipment.

Use containers that are clean, dry, leak-proof, wide-mouthed, sterile, and of a size suitable for samples of the product. Containers such as plastic jars or metal cans that are leak-proof may be hermetically sealed. Whenever possible, avoid glass containers, which may break and contaminate the food product. For dry materials, use sterile metal boxes, cans, bags, or packets with suitable closures. Sterile plastic bags (for dry, unfrozen materials only) or plastic bottles are useful containers for line samples. Take care not to overfill bags or permit puncture by wire closure. Identify each sample unit (defined later) with a properly marked strip of masking tape. Do not use a felt pen on plastic because the ink might penetrate the container. Whenever possible, obtain at least 100 g for each sample unit. Submit open and closed controls of sterile containers with the sample.

Deliver samples to the laboratory promptly with the original storage conditions maintained as nearly as possible. When collecting liquid samples, take an additional sample as a temperature control. Check the temperature of the control sample at the time of collection and on receipt at the laboratory. Make a record for all samples of the times and dates of collection and of arrival at the laboratory. Dry or canned foods that are not perishable and are collected at ambient temperatures need not be refrigerated. Transport frozen or refrigerated products in approved insulated containers of rigid construction so that they will arrive at the laboratory unchanged. Collect frozen samples in pre-chilled containers.

Place containers in a freezer long enough to chill them thoroughly. Keep frozen samples solidly frozen at all times. Cool refrigerated samples, except shellfish and shell stock, in ice at 0-4ºC and transport them in a sample chest with suitable refrigerant capable of maintaining the sample at 0-4ºC until arrival at the laboratory. Do not freeze refrigerated products. Unless otherwise specified, refrigerated samples should not be analyzed more than 36 h after collection. Special conditions apply to the collection and storage of shucked, unfrozen shellfish and shell stock (APHA, 1985). Pack samples of shucked shellfish immediately in crushed ice (no temperature specified) until analyzed; keep shell stock above freezing but below 10ºC. Examine refrigerated shellfish and shell stock within 6 h of collection but in no case more than 24 h after collection. Further details on sample handling and shipment may be found in the Investigations Operation Manual (FDA, 1993) and the Laboratory Procedures Manual (FDA, 1989). The Investigations Operation Manual (FDA, 1993) contains sampling plans for various microorganisms. Some of those commonly used are presented here.

1. Sampling plans
1. Aerobic plate counts, total coliforms, fecal coliforms, Escherichia coli (including enteropathogenic strains), Staphylococcus spp., Vibrio spp., Shigella spp., Campylobacter spp., Yersinia spp., Bacillus cereus, and Clostridium perfringens
1. Sample collection. From any lot of food, collect ten 8 ounce (227 g) subsamples (or retail packages) at random. Do not break or cut larger retail packages to obtain an 8 ounce (227 g) subsample. Collect the intact retail unit as the subsample even if it is larger than 8 ounce (227 g).
2. Sample analysis. Analyze samples as indicated in current compliance programs.
2. Equipment and materials
1. Mechanical blender. Several types are available. Use blender that has several operating speeds or rheostat. The term "high-speed blender" designates mixer with 4 canted, sharp-edge, stainless steel blades rotating at bottom of 4 lobe jar at 10,000-12,000 rpm or with equivalent shearing action. Suspended solids are reduced to fine pulp by action of blades and by lobular container, which swirls suspended solids into blades. Waring blender, or equivalent, meets these requirements.
2. Sterile glass or metal high-speed blender jar. 1000 ml, with cover, resistant to autoclaving for 60 min at 121ºC.
3. Balance, with weights. 2000 g capacity, sensitivity of 0.1 g.
4. Sterile beakers. 250 ml, low-form, covered with aluminum foil.
5. Sterile graduated pipets. 1.0 and 10.0 ml.
6. Butterfield's phosphate-buffered dilution water (R11). Sterilized in bottles to yield final volume of 90 ± 1 ml.
7. Sterile knives, forks, spatulas, forceps, scissors, tablespoons, and tongue depressors. For sample handling.
3. Receipt of samples
1. The official food sample is collected by the FDA inspector or investigator. As soon as the sample arrives at the laboratory, the analyst should note its general physical condition. If the sample cannot be analyzed immediately, it should be stored as described later. Whether the sample is to be analyzed for regulatory purposes, for investigation of a foodborne illness outbreak, or for a bacteriological survey, strict adherence to the recommendations described here is essential.
2. Condition of sampling container. Check sampling containers for gross physical defects. Carefully inspect plastic bags and bottles for tears, pinholes, and puncture marks. If sample units were collected in plastic bottles, check bottles for fractures and loose lids. If plastic bags were used for sampling, be certain that twist wires did not puncture surrounding bags. Any cross-contamination resulting from one or more of above defects would invalidate the sample, and the collecting district should be notified (see 3-e, below).
3. Labeling and records. Be certain that each sample is accompanied by a completed copy of the Collection Report (Form FD-464) and officially sealed with tape (FD-415a) bearing the sample number, collecting official's name, and date. Assign each sample unit an individual unit number and analyze as a discrete unit unless the sample is composited as described previously in this chapter.
4. Adherence to sampling plan. Most foods are collected under a specifically designed sampling plan in one of several ongoing compliance programs. Foods to be examined for Salmonella, however, are sampled according to a statistically based sampling plan designed exclusively for use with this pathogen. Depending on the food and the type of analysis to be performed, determine whether the food has been sampled according to the most appropriate sampling plan.
5. Storage. If possible, examine samples immediately upon receipt. If analysis must be postponed, however, store frozen samples at -20°C until examination. Refrigerate unfrozen perishable samples at 0-4°C not longer than 36 h. Store nonperishable, canned, or low-moisture foods at room temperature until analysis.
6. Notification of collecting district. If a sample fails to meet the above criteria and is therefore not analyzed, notify the collecting district so that a valid sample can be obtained and the possibility of a recurrence reduced.
4. Thawing

Use aseptic technique when handling product. Before handling or analysis of sample, clean immediate and surrounding work areas. In addition, swab immediate work area with commercial germicidal agent. Preferably, do not thaw frozen samples before analysis. If necessary to temper a frozen sample to obtain an analytical portion, thaw it in the original container or in the container in which it was received in the laboratory. Whenever possible, avoid transferring the sample to a second container for thawing. Normally, a sample can be thawed at 2-5ºC within 18 h. If rapid thawing is desired, thaw the sample at less than 45ºC for not more than 15 min. When thawing a sample at elevated temperatures, agitate the sample continuously in thermostatically controlled water bath.
5. Mixing

Various degrees of non-uniform distribution of microorganisms are to be expected in any food sample. To ensure more even distribution, shake liquid samples thoroughly and, if practical, mix dried samples with sterile spoons or other utensils before withdrawing the analytical unit from a sample of 100 g or greater. Use a 50 g analytical unit of liquid or dry food to determine aerobic plate count value and most probable number of coliforms. Other analytical unit sizes (e.g., 25 g for Salmonella) may be recommended, depending on specific analysis to be performed. Use analytical unit size and diluent volume recommended for appropriate Bacteriological Analytical Manual method being used. If contents of package are obviously not homogeneous (e.g., a frozen dinner), macerate entire contents of package and withdraw the analytical unit, or, preferably, analyze each different food portion separately, depending on purpose of test.
6. Weighing

Tare high-speed blender jar; then aseptically and accurately (± 0.1 g) weigh unthawed food (if frozen) into jar. If entire sample weighs less than the required amount, weigh portion equivalent to one-half of sample and adjust amount of diluent or broth accordingly. Total volume in blender must completely cover blades.
7. Blending and diluting of samples requiring enumeration of microorganisms
1. All foods other than nut meat halves and larger pieces, and nut meal. Add 450 ml Butterfield's phosphate-buffered dilution water to blender jar containing 50 g analytical unit and blend 2 min. This results in a dilution of 10^-1. Make dilutions of original homogenate promptly, using pipets that deliver required volume accurately. Do not deliver less than 10% of total volume of pipet. For example, do not use pipet with capacity greater than 10 ml to deliver 1 ml volumes; for delivering 0.1 ml volumes, do not use pipet with capacity greater than 1.0 ml. Prepare all decimal dilutions with 90 ml of sterile diluent plus 10 ml of previous dilution, unless otherwise specified. Shake all dilutions vigorously 25 times in 30 cm (1 foot) arc in 7 s. Not more than 15 min should elapse from the time sample is blended until all dilutions are in appropriate media.
2. Nut meat halves and larger pieces. Aseptically weigh 50 g analytical unit into sterile screw-cap jar. Add 50 ml diluent (7-a, above) and shake vigorously 50 times through 30 cm arc to obtain 10⁰ dilution. Let stand 3-5 min and shake 5 times through 30 cm arc to resuspend just before making serial dilutions and inoculations.
3. Nut meal. Aseptically weigh 10 g analytical unit into sterile screw-cap jar. Add 90 ml of diluent (7-a, above) and shake vigorously 50 times through 30 cm arc to obtain 10^-1 dilution. Let stand 3-5 min and shake 5 times through 30 cm arc to resuspend just before making serial dilutions and inoculations.

Contents

Bacillus cereus (Rhodehamel and Harmon, 1998)

Examination of Foods for B. cereus

1. Sampling

If the quantity of food to be examined is large, take representative samples of 50 g each from different parts of the suspect food because contamination may be unevenly distributed.

2. Transporting and storage of samples

Transport and examine samples promptly without freezing, if possible. If samples must be shipped to the laboratory, pack them in insulated shipping containers with enough gel-type refrigerant to maintain them at 6ºC or below. Upon receipt in the laboratory, store the samples at 4ºC and analyze as soon as possible. If analysis cannot be started within 4 d after collection, freeze samples rapidly and store at -20ºC until examined. Thaw at room temperature and proceed with analysis as usual. Dehydrated foods may be stored at room temperature and shipped without refrigeration.

Enumeration and Confirmation of B. cereus in Foods

1. Equipment and materials
1. Pipets, 1, 5, and 10 ml, graduated in 0.1 ml units
2. Glass spreading rods (e.g., hockey stick) 3-4 mm diameter with 45-55 mm spreading area
3. Incubators, 30 ± 2ºC and 35 ± 2ºC
4. Colony counter
5. Marking pen, black felt type
6. Large and small Bunsen burners
7. Wire loops, No. 24 nichrome or platinum wire, 2 mm and 3 mm id
8. Vortex mixer
9. Microscope, microscope slides, and cover slips
10. Culture tubes, 13 x 100 mm, sterile
11. Test tubes, 16 x 125 mm, or spot plate
12. Bottles, 3 ounce (89 ml), sterile
13. Anaerobic jar, BBL GasPak, with H₂+ CO₂ generator envelopes and catalyst
14. Water bath, 48-50ºC
15. Culture tube racks
16. Staining rack
17. Petri dishes, sterile, 15 x 100 mm
2. Media and reagents
1. Mannitol-egg yolk-polymyxin (MYP) agar plates (M95)
2. Egg yolk emulsion, 50% (M51)
3. Trypticase soy-polymyxin broth (M158)
4. Polymyxin B solutions for MYP agar (0.1%) and trypticase soy-polymyxin broth (0.15%) (M95 and M158)
5. Phenol red glucose broth (M122)
6. Tyrosine agar (M170)
7. Lysozyme broth (M90)
8. Voges-Proskauer medium (M177)
9. Nitrate broth (M108)
10. Nutrient agar for B. cereus (M113)
11. Motility medium (B. cereus) (M100)
12. Trypticase soy-sheep blood agar (M159)
13. Nitrite detection reagents (R48)
14. Butterfield's phosphate-buffered dilution water (R11) sterilized in bottles to yield final volumes of 450 ± 5 ml and 90 ± 2 ml
15. Voges-Proskauer test reagents (R89)
16. Creatine crystals
17. Gram stain reagents (R32)
18. Basic fuchsin staining solution (R3)
19. Methanol
3. Sample preparation

Using aseptic technique, weigh 50 g of sample into sterile blender jar. Add 450 ml Butterfield's phosphate-buffered dilution water (1:10 dilution) and blend for 2 min at high speed (18,000-21,000 rpm). Using the 1:10 dilution, make serial dilutions of sample for enumeration of B. cereus as described in D or E, below.
 

4. Plate count of B. cereus

Prepare serial dilutions from 10^-2 to 10^-6 by transferring 10 ml homogenized sample (1:10 dilution) to 90 ml dilution blank, mixing well with vigorous shaking, and continuing until 10^-6 dilution is reached. Inoculate duplicate MYP agar plates with each dilution of sample (including 1:10) by spreading 0.1 ml evenly onto surface of each plate with sterile glass spreading rod. Incubate plates 24 h at 30ºC and observe for colonies surrounded by precipitate zone, which indicates that lecithinase is produced. B. cereus colonies are usually a pink color which becomes more intense after additional incubation.

If reactions are not clear, incubate plates for additional 24 h before counting colonies. Select plates that contain an estimated 15-150 eosin pink, lecithinase-producing colonies. Mark bottom of plates into zones with black felt pen to facilitate counting and count colonies that are typical of B. cereus. This is the presumptive plate count of B. cereus. Pick 5 or more presumptive positive colonies from the MYP agar plates and transfer to nutrient agar slants for confirmation as B. cereus. Confirm isolates as B. cereus as described in F and G, below. Calculate number of B. cereus cells/g of sample, based on percentage of colonies tested that are confirmed as B. cereus. For example, if average count obtained with 10^-4 dilution of sample was 65 and 4 of 5 colonies tested were confirmed as B. cereus, the number of B. cereus cells/g of food is 65 x 4/5 x 10,000 x 10 = 5,200,000. (NOTE: Dilution factor is tenfold higher than sample dilution because only 0.1 ml was tested.)

5. Most probable number (MPN) of B. cereus

The MPN technique is recommended for enumerating B. cereus in foods that are expected to contain fewer than 10 B. cereus organisms/g. It may also be preferred for examining certain dehydrated starchy foods for which the plate count technique is inappropriate.

Inoculate 3-tube MPN series in trypticase soy-polymyxin broth, using 1 ml inoculum of 10^-1, 10^-2, and 10^-3 dilutions of sample with 3 tubes at each dilution. (Additional dilutions should also be tested if B. cereus population is expected to exceed 10³/g.) Incubate tubes 48 ± 2 h at 30ºC and observe for dense growth, which is typical of B. cereus. Streak cultures from positive tubes onto separate MYP agar plates and incubate plates 24-48 h at 30ºC. Pick one or more eosin pink, lecithinase-positive colonies from each MYP agar plate and transfer to nutrient agar slants for confirmation as B. cereus. Confirm isolates as B. cereus as described in F and G, below, and calculate MPN of B. cereus cells/g of sample based on the number of tubes at each dilution in which the presence of B. cereus was confirmed.

6. Confirmation of B. cereus

Pick 5 or more eosin pink, lecithinase-positive colonies from MYP agar plates and transfer to nutrient agar slants. Incubate slants 24 h at 30ºC. Prepare Gram-stained smears from slants and examine microscopically. B. cereus will appear as large Gram-positive bacilli in short-to-long chains; spores are ellipsoidal, central to subterminal, and do not swell the sporangium. Transfer 3 mm loopful of culture from each slant to 13 x 100 mm tube containing 0.5 ml of sterile phosphate-buffered dilution water and suspend culture in diluent with Vortex mixer. Use suspended cultures to inoculate the following confirmatory media:

1. Phenol red glucose broth. Inoculate 3 ml broth with 2 mm loopful of culture. Incubate tubes anaerobically 24 h at 35ºC in GasPak anaerobic jar. Shake tubes vigorously and observe for growth as indicated by increased turbidity and color change from red to yellow, which indicates that acid has been produced anaerobically from glucose. A partial color change from red to orange/yellow may occur, even in uninoculated control tubes, due to a pH reduction upon exposure of media to CO² formed in GasPak anaerobic jars. Be sure to use appropriate positive and negative controls so that a distinction can be made between positive and "false-positive" reactions.
2. Nitrate broth. Inoculate 5 ml broth with 3 mm loopful of culture. Incubate tubes 24 h at 35ºC. To test for nitrite, add 0.25 ml each of nitrite test reagents A and C to each culture. An orange color, which develops within 10 min, indicates that nitrate has been reduced to nitrite.
3. Modified VP medium. Inoculate 5 ml medium with 3 mm loopful of culture and incubate tubes 48 ± 2 h at 35ºC. Test for production of acetylmethyl-carbinol by pipetting 1 ml culture into 16 x 125 mm test tube and adding 0.6 ml alpha-naphthol solution (R89) and 0.2 ml 40% potassium hydroxide (R89). Shake, and add a few crystals of creatine. Observe results after holding for 1 h at room temperature. Test is positive if pink or violet color develops.
4. Tyrosine agar. Inoculate entire surface of tyrosine agar slant with 3 mm loopful of culture. Incubate slants 48 h at 35ºC. Observe for clearing of medium near growth, which indicates that tyrosine has been decomposed. Examine negative slants for obvious signs of growth, and incubate for a total of 7 d before considering as negative.
5. Lysozyme broth. Inoculate 2.5 ml of nutrient broth containing 0.001% lysozyme with 2 mm loopful of culture. Also inoculate 2.5 ml of plain nutrient broth as positive control. Incubate tubes 24 h at 35ºC. Examine for growth in lysozyme broth and in nutrient broth control. Incubate negative tubes for additional 24 h before discarding.
6. MYP agar. This test may be omitted if test results were clear-cut with original MYP agar plates and there was no interference from other microorganisms which were present. Mark bottom of a plate into 6-8 equal sections with felt marking pen, and label each section. Inoculate premarked 4 cm² area of MYP agar plate by gently touching surface of agar with 2 mm loopful of culture. (Six or more cultures can be tested in this manner on one plate.) Allow inoculum to be fully absorbed before incubating for 24 h at 35ºC. Check plates for lecithinase production as indicated by zone of precipitation surrounding growth. Mannitol is not fermented by isolate if growth and surrounding medium are eosin pink. (Yellow color indicates that acid is produced from mannitol.) B. cereus colonies are usually lecithinase-positive and mannitol-negative on MYP agar.
7. Record results obtained with the different confirmatory tests. Tentatively identify as B. cereus those isolates which 1) produce large Gram-positive rods with spores that do not swell the sporangium; 2) produce lecithinase and do not ferment mannitol on MYP agar; 3) grow and produce acid from glucose anaerobically; 4) reduce nitrate to nitrite (a few strains may be negative); 5) produce acetylmethylcarbinol (VP-positive); 6) decompose L-tyrosine; and 7) grow in the presence of 0.001% lysozyme.

These basic characteristics are shared with other members of the B. cereus group, including the rhizoid strains B. mycoides, the crystalliferous insect pathogen B. thuringiensis, and the mammalian pathogen B. anthracis. However, these species can usually be differentiated from B. cereus by determining specific characteristics typical of each species or variety. The tests described in G, below, are useful for this purpose and can easily be performed in most laboratories. Strains that produce atypical results from these tests require additional analysis before they can be classified as B. cereus.

7. Tests for differentiating members of the B. cereus group (Table 10-3)

Table 10-3. Differential characteristics of large-celled Group I Bacillus species.

Feature	B. cereus	B. thuringiensis	B. cereus var. mycoides	B. anthracis	B. megaterium
Gram reaction	+a	+	+	+	+
Catalase	+	+	+	+	+
Motility	+/-b	+/-	-c	-	+/-
Reduction of nitrate	+	+/-	+	+	-d
Tyrosine decomposed	+	+	+/-	-d	+/-
Lysozyme-resistant	+	+	+	+	-
Egg yolk reduction	+	+	+	+	-
Anaerobic utilization of glucose	+	+	+	+	-
VP reaction	+	+	+	+	-
Acid produced from mannitol	-	-	-	-	+
Hemolysis (sheep RBC)	+	+	+	-d	-
Known pathogenicityc	Produces enterotoxins	Endotoxin crystals pathogenic to insects	Rhizoidal growth	Pathogenic to animals and humans

a+, 90-100% of strains are positive
^b+/-, 50-50% of strains are positive
^c-, 90-100% of strains are negative
^d-, most strains are negative
^eSee "Section H, Limitations of methods for B. cereus"

The following tests are useful for differentiating typical strains of B. cereus from other members of the B. cereus group, including B. mycoides, B. thuringiensis, and B. anthracis.

1. Motility test. Inoculate BC motility medium by stabbing down the center with 3 mm loopful of 24 h culture suspension. Incubate tubes 18-24 h at 30ºC and examine for type of growth along stab line. Motile organisms produce diffuse growth out into the medium away from the stab. Nonmotile organisms produce growth only in and along stab. Alternatively, add 0.2 ml sterile distilled water to surface of nutrient agar slant and inoculate slant with 3 mm loopful of culture suspension. Incubate slant 6-8 h at 30ºC and suspend 3 mm loopful of liquid culture from base of slant in a drop of sterile water on microscope slide. Apply cover glass and examine immediately with microscope for motility. Report whether or not isolates tested were motile. Most strains of B. cereus and B. thuringiensis are motile by means of peritrichous flagella. B. anthracis and all except a few strains of B. mycoides are nonmotile. A few B. cereus strains are also nonmotile.
2. Rhizoid growth. Pour 18-20 ml nutrient agar into sterile 15 x 100 mm petri dishes and allow agar to dry at room temperature for 1-2 d. Inoculate by gently touching surface of medium near center of each plate with 2 mm loopful of 24 h culture suspension. Allow inoculum to be absorbed and incubate plates 48-72 h at 30ºC. Examine for development of rhizoid growth, which is characterized by production of colonies with long hair or root-like structures that may extend several centimeters from site of inoculation. Rough galaxy-shaped colonies are often produced by B. cereus strains and should not be confused with typical rhizoid growth, which is the definitive characteristic of B. mycoides. Most strains of this species are also nonmotile.
3. Test for hemolytic activity. Mark bottom of a plate into 6-8 equal sections with felt marking pen, and label each section. Inoculate a premarked 4 cm² area of trypticase soy-sheep blood agar plate by gently touching medium surface with 2 mm loopful of 24 h culture suspension. (Six or more cultures can be tested simultaneously on each plate.) Incubate plates 24 h at 35ºC. Examine plates for hemolytic activity. B. cereus cultures usually are strongly hemolytic and produce 2-4 mm zone of complete (b) hemolysis surrounding growth. Most B. thuringiensis and B. mycoides strains are also b-hemolytic. B. anthracis strains are usually nonhemolytic after 24 h incubation.
4. Test for protein toxin crystals. Inoculate nutrient agar slants with 3 mm loopfuls of 24 h culture suspensions. Incubate slants 24 h at 30ºC and then at room temperature 2-3 d. Prepare smears with sterile distilled water on microscope slides. Air-dry and lightly heat-fix by passing slide through flame of Bunsen burner. Place slide on staining rack and flood with methanol. Let stand 30 s, pour off methanol, and allow slide to air-dry. Return slide to staining rack and flood completely with 0.5% basic fuchsin or TB carbolfuchsin ZN stain (Difco). Heat slide gently from below with small Bunsen burner until steam is seen.

Wait 1-2 min and repeat this step. Let stand 30 s, pour off stain, and rinse slide thoroughly with clean tap water. Dry slide without blotting and examine under oil immersion for presence of free spores and darkly stained tetragonal (diamond-shaped) toxin crystals. Crystals are usually somewhat smaller than spores. Toxin crystals are usually abundant in a 3-4 d old culture of B. thuringiensis but cannot be detected by the staining technique until lysis of the sporangium has occurred. Therefore, unless free spores can be seen, cultures should be held at room temperature for a few more d and re-examined for toxin crystals. B. thuringiensis usually produces protein toxin crystals that can be detected by the staining technique either as free crystals or parasporal inclusion bodies within the exosporium. B. cereus and other members of the B. cereus group do not produce protein toxin crystals.

5. Interpreting test results. On the basis of the test results, identify as B. cereus those isolates which are actively motile and strongly hemolytic and do not produce rhizoid colonies or protein toxin crystals. Nonmotile B. cereus strains are also fairly common and a few strains are weakly hemolytic. These nonpathogenic strains of B. cereus can be differentiated from B. anthracis by their resistance to penicillin and g bacteriophage. CAUTION: Nonmotile, nonhemolytic isolates that are suspected to be B. anthracis should be submitted to a pathology laboratory such as the Centers for Disease Control and Prevention, Atlanta, GA, for identification or destroyed by autoclaving. Acrystalliferous variants of B. thuringiensis and nonrhizoid strains derived from B. mycoides cannot be distinguished from B. cereus by the cultural tests.
8. Limitations of method for B. cereus

The method described is intended primarily for use in the routine examination of foods. As noted in F, above, and in Table 10-3, the confirmatory tests recommended may in some instances be inadequate for distinguishing B. cereus from culturally similar organisms that could occasionally be encountered in foods. These organisms include 1) the insect pathogen B. thuringiensis, which produces protein toxin crystals; 2) B. mycoides, which characteristically produces rhizoid colonies on agar media; and 3) B. anthracis, which exhibits marked animal pathogenicity and is nonmotile. With the exception of B. thuringiensis, which is currently being used for insect control on food and forage crops, these organisms are seldom encountered in the routine examination of foods. The tests described in G, above, are usually adequate for distinguishing the typical strains of B. cereus from other members of the B. cereus group. However, results with atypical strains of B. cereus are quite variable, and further testing may be necessary to identify the isolates. Although a few diarrheal toxin detection kits are commercially available, none are recommended, pending further evaluation. At present, no practical tests for detecting the emetic toxin are available. Until reliable tests are available, cultural tests such as those described in this method must be relied upon for confirming isolates from foods as B. cereus.

Contents

Bacillus cereus Diarrheal Enterotoxin (Bennett, 1998)

Bacillus cereus is an aerobic sporeformer that is commonly found in soil, on vegetables, and in many raw and processed foods. Consumption of foods that contain large numbers of B. cereus (10⁶or more/g) may result in food poisoning, especially when foods are prepared and held for several h without adequate refrigeration before serving. Cooked meat and vegetables, boiled or fried rice, vanilla sauce, custards, soups, and raw vegetable sprouts have been incriminated in past outbreaks (Bennett and Harmon, 1988). Two types of illness are attributed to the consumption of foods contaminated with B. cereus. The first and better known is characterized by abdominal pain and diarrhea; it has an incubation period of 4-16 h and symptoms that last for 12-24 h (Lancette and Harmon, 1980; McFarland, 1907). The second, which is characterized by an acute attack of nausea and vomiting that occurs within 1-5 h after a meal; diarrhea is not a common feature in this type of illness.

Although certain physiological and cultural characteristics are necessary for identifying B. cereus (Lancette and Harmond, 1980), its enterotoxigenicity indicates whether a suspect strain may be a public health hazard. Evidence shows that diarrheal toxin is a distinct serological entity; in vitro methods that use specific antibodies have been developed to detect the toxin in culture fluids. The evidence for the emetic toxin, however, is still incomplete. This chapter presents a method for the routine culturing of suspect Bacillus spp., using a semisolid agar medium and a serological procedure (the microslide gel double diffusion test) to identify the enterotoxin.

1. Equipment and materials
1. Test tubes, 25 x 100 and 20 x 150 nm
2. Petri dishes, 15 x 100 and 20 x 150 mm, sterile
3. Bottles, prescription, 4 oz
4. Microscope slides, glass, pre-cleaned, 3 x 1 inch (7.62 x 2.54 cm)
5. Pipets, sterile, 1, 5, and 10 ml, graduated
6. Centrifuge tubes, 50 ml
7. Sterile bent glass spreaders
8. Electrical tape, 0.25 mm thick, 19.1 mm wide, available from Scotch Brand, 3M Co., Electro-Products Division, St. Paul, MN 55011.
9. Templates, plastic (Figure 10-1)
10. Silicone grease, high vacuum, available from Dow Corning Corp., Midland, MI 48640
11. Sponges, synthetic
12. Wooden applicator sticks
13. Glass tubing, 7 mm, for capillary pipets and de-bubblers
14. Pasteur pipets or disposable 30 or 40 µ1 pipets, available from Kensington Scientific Corp., 1165-67th St., Oakland, CA 94601, if capillary pipets are not available
15. Staining jars (Coplin or Wheaton)
16. Desk lamp
17. Incubator, 35 ± 1ºC
18. Hot plate, electric
19. Sterilizer (Arnold), flowing steam
20. Blender and sterile blender jars (see Chapter 1)
21. Centrifuge, high speed
22. Timer, interval
2. Media and reagents
1. Brain heart infusion (BHI) broth (M24)
2. Glucose, dextrose anhydrous
3. Gel diffusion agar, 1.2% (R28)
4. Nutrient agar slants (M112)
5. Distilled water, sterile
6. Phosphate-buffered dilution water (Butterfield's buffer) (R11)
7. Normal (physiological) saline, sterile (antisera diluent) (R63)
8. Thiazine Red R stain (R79)
9. Slide preserving solution for stained slides, 1% acetic acid and 1% glycerol (R69)
10. No. 1 McFarland standard (R42)
11. Antisera and reference enterotoxins
3. Preparation of materials and media
1. BHIG, 0.1%. Adjust BHI broth containing 0.1% glucose to pH 7.4 and dissolve by stirring. Distribute medium in 30 ml portions in 125 ml flasks and autoclave at 121ºC for 10 min.
2. No. 1 McFarland standard. Prepare turbidity standard No. 1 of McFarland nephelometer scale (5). Mix 1 part 1% BaCl₂ with 99 parts 1% H₂SO₄ in distilled water.
3. 1.2% Gel diffusion agar for gel diffusion slides. Prepare fluid base for agar in distilled water as follows: NaCl 0.85%; sodium barbital 0.8%; merthiolate 1:10,000 (crystalline), available from Eli Lilly and Co., Terre Haute, IN. Adjust pH to 7.4. Prepare agar by adding 1.2% Noble special agar (Difco). Melt agar mixture in Arnold sterilizer (steamer) and filter while hot, in steamer, through 2 layers of filter paper; dispense in small portions (15-25 ml) in 4 oz prescription bottles. (Remelting more than twice may break down purified agar.)
4. Thiazine Red R stain. Prepare 0.1% solution of Thiazine Red R stain in 1.0% acetic acid.
5. Preparation of slides. Wrap double layer of electrician's plastic insulating tape around both sides of glass slide, leaving 2.0 cm space in center. Apply tape as follows: Start a piece of tape 9.5-10 cm long about 0.5 cm from edge of undersurface of slide and wrap tightly around slide twice. Wipe area between tapes with cheesecloth soaked with 95% ethanol, and dry with dry cheesecloth. Coat upper surface area between tapes with 0.2% agar in distilled water as follows: Melt 0.2% bacteriological grade agar, and maintain at 55ºC or higher in screw-cap bottle. Hold slide over beaker placed on hot plate adjusted to 65-85ºC and pour or brush 0.2% agar over slide between 2 pieces of tape. Let excess agar drain into beaker. Return agar collected in beaker to original container for reuse. Wipe undersurface of slide. Place slide on tray and dry in dust-free atmosphere (e.g., incubator). NOTE: If slides are not clean, agar will roll off slide without coating it uniformly.
6. Preparation of slide assembly. Prepare plastic templates as described by Casman et al. (1969) (Figure 10-1). Spread thin film of silicone grease on side of template that will be placed next to agar, i.e., the side with the smaller holes. Place about 0.4 ml melted and cooled (55-60ºC) 1.2% diffusion agar between tapes.

Immediately lay silicone-coated template on melted agar and edges of bordering tapes. Place one edge of template on one of the tapes and bring opposite edge to rest gently on the other tape. Place slide in prepared petri dish (see 3-g, below) soon after agar solidifies and label slide with number, date, or other information.

7. Preparation of petri dishes for slide assemblies. Maintain necessary high humidity by saturating 2 strips of synthetic sponge (about ½ inch [1.3 cm] wide x ½ inch [1.3 cm] deep x 2½ inches [6.4 cm] long) with distilled water and placing them in each 20 x 150 mm petri dish. From 2 to 4 slide assemblies can be placed in each dish.
8. Recovery of used slides and templates. Clean slides without removing tape; rinse with tap water, brush to remove agar gel, boil in detergent solution for 15-20 min, rinse about 5 min in hot running water, and boil in distilled water. Place slides on end, using test tube rack or equivalent, and place in incubator to dry. If slides cannot be uniformly coated with hot 0.2% agar, they are not clean enough and must be washed again. Avoid exposure to excessive heat or plastic solvents when cleaning plastic templates. Place templates in a pan and pour hot detergent solution over them; let them soak 10-15 min. Use soft nylon brush to remove residual silicone grease. Rinse sequentially with tap water, distilled water, and 95% ethanol. Spread templates on towel to dry.
9. Directions for dissolving reagents used in slide gel. The reagents are supplied as lyophilized preparations of enterotoxins and their antisera. Rehydrate antisera in physiological saline. Rehydrate reference enterotoxins in physiological saline containing 0.3% proteose peptone, pH 7.0, or physiological saline containing 0.37% dehydrated BHI broth, pH 7.0. These preparations should produce faint but distinct reference lines in the slide gel diffusion test. The lines may be enhanced (see 5-c, below).
4. Procedure for enumeration and selection of B. cereus colonies

For examining food products, use procedures described for detecting B. cereus. Test isolates for enterotoxigenicity as described in 5, below.

Production of enterotoxin. Of the methods described for the production of enterotoxin, cultivation of B. cereus in BHIG (0.1% glucose, pH 7.4) is simple and requires no special apparatus other than a shaker. Add loopful of growth from nutrient agar slants to 3-5 ml sterile distilled water or saline. Inoculate BHIG with 0.5 ml of this aqueous suspension, which should contain about 300 million organisms/ml. Turbidity of suspension should be equivalent to No. 1 on McFarland nephelometer scale. Deliver suspension with sterile 1.0 ml pipet. Shake flasks at 3 ± 2ºC at 84-125 cycles/ml for 12 h. Good surface growth is obtained after 12 h of incubation. Transfer contents of flasks to 50 ml centrifuge tube. Remove organisms by high speed centrifugation (10 min at 32,800 x g). Examine supernatant for presence of enterotoxin by filling depots in slide gel diffusion assembly, as directed in 5, below.

5. Slide gel diffusion test

To prepare record sheet, draw hole pattern of template on record sheet, indicate contents of each well, and give each pattern on record sheet a number to correspond with number on slide.

1. Addition of reagents (Figure 10-2). Place suitable dilution of anti-enterotoxin (antiserum) in central well and place homologous reference enterotoxin in upper peripheral well (if diamond pattern is used); place material(s) under examination in well adjacent to well containing reference enterotoxin(s). Use reference toxins and antitoxins (antiserum), previously balanced, in concentrations that give line of precipitation about halfway between their respective wells. Adjust dilutions of reagents to give distinct but faint lines of precipitation for maximum sensitivity. (See 3-i for directions for dissolving reagents.) Prepare control slide with only reference toxin and antitoxin.

Fill wells to convexity with reagents, using Pasteur pipet (prepared by drawing out glass tubing of about 7 mm od) or disposable 30 or 40 µl pipet. Remove bubbles from all wells by probing with fine glass rod. Make rods by pulling glass tubing very fine, as in making capillary pipets, breaking it into about 2-1/2 inch (6.4 cm) lengths, and melting ends in flame. It is best to fill wells and remove bubbles against a dark background. Insert rods into all wells to remove trapped air bubbles that may not be visible. Let slides remain at room temperature in covered petri dishes containing moist sponge strips for 48-72 h before examination or for 24 h at 37ºC.

2. Reading the slide. Remove template by sliding it to one side. If necessary, clean slide by dipping momentarily in water and wiping bottom of slide; then stain as described below. Examine slide by holding over source of light and against dark background. Identify lines of precipitation through their coalescence with reference line of precipitation (Figure 10-3). If concentration of enterotoxin in test material is excessive, formation of reference line will be inhibited; test material must then be diluted and retested. Figure 10-4, diagram A, shows typical precipitate line inhibition caused by enterotoxin excess in test preparation reactant arrangement in Figure 10-2. Figure 10-5 shows typical line formation. Figure 10-6 shows a diluted preparation. Occasionally, atypical precipitate patterns that form may be difficult for inexperienced analysts to interpret. One of the most common atypical reactions is formation of lines not related to toxin but caused by other antigens in test material (Figure 10-7).
3. Staining of slides. Enhance lines of precipitation by immersing slide in Thiazine Red R strain for 5-10 min, and then examine. Such enhancement is necessary when reagents have been adjusted to give lines of precipitation that are only faintly visible. Use staining procedure described by Crowle (1958), modified slightly, when slide is to be preserved. Rinse away any reactant liquid remaining on slide by dipping slide momentarily in water and immersing it for 10 min in each of the following baths: 0.1% Thiazine Red R in 1% acetic acid; 1% acetic acid; and 1% acetic acid containing 1% glycerol. Drain excess fluid from slide and dry in 35ºC incubator for storage as permanent record. After prolonged storage, lines of precipitation may not be visible until slide is immersed in water.

Contents

Most Probable Number from Serial Dilutions (Garthright, 1998)

Background

The most probable number (MPN) is particularly useful for low concentrations of organisms (<100/g), especially in milk and water, and for those foods whose particulate matter may interfere with accurate colony counts. The following background observations are adapted and extended from the article on MPN by James T. Peeler and Foster D. McClure in the Bacteriological Analytical Manual (BAM), 7th edition.

Only viable organisms are enumerated by the MPN determination. If, in the microbiologist's experience, the bacteria in the prepared sample in question can be found attached in chains that are not separated by the preparation and dilution, the MPN should be judged as an estimate of growth units (GUs) or colony-forming units (CFUs) instead of individual bacteria. For simplicity, however, this appendix will speak of these GUs or CFUs as individual bacteria.

The following assumptions are necessary to support the MPN method. The sample is prepared in such a way that the bacteria are distributed randomly within it. The bacteria are separate, not clustered together, and they do not repel each other. The growth medium and conditions of incubation have been chosen so that every inoculum that contains even one viable organism will produce detectable growth.

The essence of the MPN method is the dilution of a sample to such a degree that inocula will sometimes but not always contain viable organisms. The "outcome", i.e., the numbers of inocula producing growth at each dilution, will imply an estimate of the original, undiluted concentration of bacteria in the sample. In order to obtain estimates over a broad range of possible concentrations, microbiologists use serial dilutions, incubating several tubes (or plates, etc.) at each dilution.

The first accurate estimation of the number of viable bacteria by the MPN method was published by McCrady (1915). Halvorson and Ziegler (1933), Eisenhart and Wilson (1943), and Cochran (1950) published articles on the statistical foundations of the MPN. Woodward (1957) recommended that MPN tables should omit those combinations of positive tubes (high for low concentrations and low for high concentrations) that are so improbable that they raise concerns about laboratory error or contamination. De Man (1983) published a confidence interval method that was modified to make the tables for this appendix.

Confidence intervals

The 95% confidence intervals in the tables have the following meaning.

Before the tubes are inoculated, the chance is at least 95% that the confidence interval associated with the eventual result will enclose the actual concentration.

It is possible to construct many different sets of intervals that satisfy this criterion. This manual uses a modification of the method of de Man (1983). De Man calculated his confidence limits iteratively from the smallest concentrations upward. Because this manual estimates concentrations of pathogens, the intervals have been shifted slightly upward by iterating from the largest concentrations downward.

Improbable outcomes

When excluding improbable outcomes, de Man's (1983) preferred degree of improbability was adopted. The included combinations of positive tubes are those that would be among the 99.985% most likely to result if their own MPNs were the actual bacterial concentrations. Therefore the entire set of results on any 10 different samples will be found in these tables at least 99% of the time.

Precision, bias, and extreme results

The MPNs and confidence limits have been expressed to 2 significant digits. For example, the entry "400" has been rounded from a number between 395 and 405.

Numerous articles have noted a bias toward over-estimation of microbial concentrations by the MPN. Garthright (1993) has shown, however, that there is no appreciable bias when the concentrations are expressed as logarithms, the customary units used for regressions and for combining results. Therefore, these MPNs have not been adjusted for bias.

Prior to this revision, the 8th edition tables showed the MPN for the (0,0,0) outcomes as less than the MPN of the (1,0,0) outcome. This made good numerical sense, but made for unacceptable complexity in trying to write acceptance standards for raw materials in terms of the BAM. This revision returns to the prior practice of recording the MPN for the (0,0,0) outcomes as less than the MPN for the (0,0,1) outcome, so that standards can once again be written in a simple manner in terms of all-negative outcomes.

Since no particular density is indicated for an outcome of (0.0.0), a density must be assigned arbitrarily (and stated explicitly in the report) in order to calculate statistics. For the logarithm of the density, log[0.5*MPN(1,0,0)] is a reasonable choice. For statistics using the (non-logarithmic) density itself, calculate once with a density of 0.0 and once with a density of 0.5*MPN(1,0,0). Either report both statistics or report one statistic accompanied by a comment on the difference between that statistic and the other one.

Selecting three dilutions for table reference

An MPN can be computed for any numbers of tubes at any numbers of dilutions. MPN values based on 3 decimal dilutions, however, are very close approximations to those based on 4 or more dilutions. When more than three dilutions are used in a decimal series of dilutions, refer to the 3 dilution table according to the following two cases, illustrated by Table 10-6 (with 5 tubes at each dilution).

Case 1. One or more dilutions show all tubes positive. Select the highest dilution that gives positive results in all tubes (even if a lower dilution gives negative results) and the next two higher dilutions (examples a and b); if positive results occur in higher unselected dilutions, shift each selection to the next higher dilution (example c). If there are still positive results in higher unselected dilutions, add those higher-dilution positive results to the results for the highest selected dilution (example d). If there were not enough higher dilutions tested to select three dilutions, then select the next lower dilutions (example e).

Case 2. No dilutions show all tubes positive. Select the 3 lowest dilutions (example f). If there are positive results in higher unselected dilutions, add those higher-dilution positive results to the results for the highest selected dilution (example g).

Other compendia of methods require that no excluded lower dilutions may have any negative tubes. This manual differs when the highest dilution that makes all tubes positive follows a lower dilution that has one or more negative tubes. Example b above would be read according to other compendia as (4, 5, 1, 0, 0) with MPN 4.8/g. The BAM reading, 33/g, is 7 times larger. The BAM selection method is based on FDA experience that for some organisms in some food matrices such outcomes as (2, 5, 1, 0, 0) and (0, 3, 1, 0, 0) occur too often to be random occurrences. In these cases, it appears that some factor (a competing organism or adverse set of compounds) is present at the lowest dilutions in such concentrations that it can reduce the detection of the target microbes.

Table 10-4. Table of examples.

Example	1.0 g	0.1 g	0.01 g	0.001 g	0.0001 g	Combination of positives	MPN/g
a	5	5	1	0	0	5-1-0	33
b	4	5	1	0	0	5-1-0	33
c	5	4	4	1	0	4-4-1	40
d	5	4	4	0	1	4-4-1	40
e	5	5	5	5	2	5-5-2	5400
f	0	0	1	0	0	0-0-1	0.20
g	4	4	1	1	0	4-4-2	4.7

Until further research clarifies this situation, analysts should continue to exclude dilutions lower than the highest dilution with all tubes positive. The findings should, however, report the extent to which such lower, partially negative dilutions have been excluded. Analysts working with materials with known limited complexity in research settings will want to use their professional judgement to read outcomes such as (4, 5, 1, 0, 0) as (4, 5, 1, 0, 0). They may also read outcomes such as (3, 5, 1, 0, 0) as too improbable to record, because they are not included in the tables.

Conversion of units

Tables 10-5 to 10-7 apply directly to inocula 0.1, 0.01, and 0.001 g. When different inocula are selected for table reference, multiply the MPN/g and confidence limits by whatever number is required to make the inocula match the table inocula. For example, if the inocula were 0.01, 0.001, and 0.0001 for 3 tubes each, multiplying by 10 would make these inocula match the table inocula. If the positive results from this 3 tube series were (3, 1, 0), one would multiply the Table 1 MPN/g estimate, 43/g, by 10 to arrive at 430/g.

Approximations for an unusual series of dilutions

The MPNs for a series of dilutions not addressed by any tables (e.g., resulting from accidental loss of some tubes) may be computed by iteration or may be estimated as follows. First, select the lowest dilution that doesn't have all positive results. Second, select the highest dilution with at least one positive result. Finally, select all the dilutions between them. Use only the selected dilutions in the following formula of Thomas (1942):



where ( )^1/2 means square root, P is the number of positive results, T is the total g in the selected dilutions, and N is the g of sample in the selected negative tubes.

The following examples will illustrate the application of Thomas’s formula. We assume that the dilutions are 1.0, 0.1, 0.01, 0.001, and 0.0001 g.

Example (1). Dilution results are (5/5, 10/10, 4/10, 2/10, 0/5). We use only (--,--, 4/10, 2/10,--); so T= 10*0.01 + 10*0.001 = 0.11. There are 6 negative tubes at 0.01 and 8 negative tubes at 0.001, so N = 6*0.01 + 8*0.001 = 0.068. There are 6 positive tubes, so



Example (2). Dilution results are (5/5, 10/10, 10/10, 0/10, 0/5). We use only(--,--, 10/10, 0/10,--), so by Thomas’s formula,



These two approximated MPNs compare well with the MPNs for (10, 4, 2) and (10,10,0) (i.e., 70/g and 240/g, respectively).

Example (2) above is a special case for which an exact solution for the two selected dilutions can be calculated directly, as follows. When all the results at the highest dilutions are negative, all the results at the remaining dilutions are positive, and when V is an individual inoculum at the highest dilution with all positive tubes, then



where T and N are defined as for Thomas's formula. For the second example above, the third dilution is the highest with positive portions, so V = 0.01. The MPN for the third and fourth dilution would be exactly



Approximate confidence limits can be calculated for the MPN by using the method of Cochran as follows. Let "a" denote the dilution ratio (e.g., a = 10 for the tables in this appendix). If there are n tubes per dilution, the standard error of log (MPN) is



where c = 0.58 for dilution ratios of 10 or more and c = 0.55 for dilution ratios less than 10. If the dilutions used for the MPN determination are not all the same, Peeler et al. (1992) recommend using for n the harmonic mean number of tubes in the k dilutions:



Special requirements and tables included

Requests for special computations and different designs will be honored as resources permit. Designs may be requested with more or less than 3 dilutions, uneven numbers of tubes, different confidence levels, etc. (Telephone or write the Division of Mathematics, FDA/CFSAN, 200 C St., SW, Washington, DC 20204.) The most-published designs, three 10-fold dilutions with 3, 5, or 10 tubes at each dilution, are presented here.

Table 10-5. For 3 tubes each at 0.1, 0.01, and 0.001 g inocula, the MPNs/g and 95% confidence intervals.

Positive tubes			MPN/g	Confidence limit		Positive tubes			MPN/g	Confidence limit
0.10	0.01	0.001		Low	High	0.10	0.01	0.001		Low	High
0	0	0	<3.0	--	9.5	2	2	0	21	4.5	42
0	0	1	3.0	0.15	9.6	2	2	1	28	8.7	94
0	1	0	3.0	0.15	11	2	2	2	35	8.7	94
0	1	1	6.1	1.2	18	2	3	0	29	8.7	94
0	2	0	6.2	1.2	18	2	3	1	36	8.7	94
0	3	0	9.4	3.6	38	3	0	0	23	4.6	94
1	0	0	3.6	0.17	18	3	0	1	38	8.7	110
1	0	1	7.2	1.3	18	3	0	2	64	17	180
1	0	2	11	3.6	38	3	1	0	43	9	180
1	1	0	7.4	1.3	20	3	1	1	75	17	200
1	1	1	11	3.6	38	3	1	2	120	37	420
1	2	0	11	3.6	42	3	1	3	160	40	420
1	2	1	15	4.5	42	3	2	0	93	18	420
1	3	0	16	4.5	42	3	2	1	150	37	420
2	0	0	9.2	1.4	38	3	2	2	210	40	430
2	0	1	14	3.6	42	3	2	3	290	90	1,000
2	0	2	20	4.5	42	3	3	0	240	42	1,000
2	1	0	15	3.7	42	3	3	1	460	90	2,000
2	1	1	20	4.5	42	3	3	2	1100	180	4,100
2	1	2	27	8.7	94	3	3	3	>1100	420	--

Table 10-6. For 5 tubes each at 0.1, 0.01, and 0.001 g inocula, the MPNs and 95% confidence intervals.

Positive tubes			MPN/g	Confidence limit		Positive tubes			MPN/g	Confidence limit
0.10	0.01	0.001		Low	High	0.10	0.01	0.001		Low	High
0	0	0	<1.8	--	6.8	3	3	2	24	9.8	70
0	0	1	1.8	0.09	6.8	3	4	0	21	6.8	40
0	1	0	1.8	0.09	6.9	3	4	1	24	9.8	70
0	1	1	3.69	0.7	10	3	5	0	25	9.8	70
0	2	0	3.7	0.7	10	4	0	0	13	4.1	35
0	2	1	5.5	1.8	15	4	0	1	17	5.9	36
0	3	0	5.6	1.8	15	4	0	2	21	6.8	40
1	0	0	2.0	0.1	10	4	0	3	25	9.8	70
1	0	1	4.0	0.7	10	4	1	0	17	6.0	40
1	0	2	6.0	1.8	15	4	1	1	21	6.8	42
1	1	0	4.0	0.7	12	4	1	2	26	9.8	70
1	1	1	6.1	1.8	15	4	1	3	31	10	70
1	1	2	8.1	3.4	22	4	2	0	22	6.8	50
1	2	0	6.1	1.8	15	4	2	1	26	9.8	70
1	2	1	8.2	3.4	22	4	2	2	32	10	70
1	3	0	8.3	3.4	22	4	2	3	38	14	100
1	3	1	10	3.5	22	4	3	0	27	9.9	70
1	4	0	11	3.5	22	4	3	1	33	10	70
2	0	0	4.5	0.79	15	4	3	2	39	14	100
2	0	1	6.8	1.8	15	4	4	0	34	14	100
2	0	2	9.1	3.4	22	4	4	1	40	14	100
2	1	0	6.8	1.8	17	4	4	2	47	15	120
2	1	1	9.2	3.4	22	4	5	0	41	14	100
2	1	2	12	4.1	26	4	5	1	48	15	120
2	2	0	9.3	3.4	22	5	0	0	23	6.8	70
2	2	1	12	4.1	26	5	0	1	31	10	70
2	2	2	14	5.9	36	5	0	2	43	14	100
2	3	0	12	4.1	26	5	0	3	58	22	150
2	3	1	14	5.9	36	5	1	0	33	10	100
2	4	0	15	5.9	36	5	1	1	46	14	120
3	0	0	7.8	2.1	22	5	1	2	63	22	150
3	0	1	11	3.5	23	5	1	3	84	34	220
3	0	2	13	5.6	35	5	2	0	49	15	150
3	1	0	11	3.5	26	5	2	1	70	22	170
3	1	1	14	5.6	36	5	2	2	94	34	230
3	1	2	17	6.0	36	5	2	3	120	36	250
3	2	0	14	5.7	36	5	2	4	150	58	400
3	2	1	17	6.8	40	5	3	0	79	22	220
3	2	2	20	6.8	40	5	3	1	110	34	250
3	3	0	17	6.8	40	5	3	2	140	52	400
3	3	1	21	6.8	40	5	3	3	180	70	400
5	3	4	210	70	400	5	5	0	240	70	710
5	4	0	130	36	400	5	5	1	350	100	1100
5	4	1	170	58	400	5	5	2	540	150	1700
5	4	2	220	70	440	5	5	3	920	220	2600
5	4	3	280	100	710	5	5	4	1600	400	4600
5	4	4	350	100	710	5	5	5	>1600	700	--
5	4	5	430	150	1100

Table 10-7. For 10 tubes at each of 0.1, 0.01, and 0.001 g inocula, the MPNs and 95% confidence intervals.

Positive tubes			MPN/g	Confidence limit		Positive tubes			MPN/g	Confidence limit
0.10	0.01	0.001		Low	High	0.10	0.01	0.001		Low	High
0	0	0	<0.90	--	3.1	4	0	2	6.8	3.0	14
0	0	1	.9	.040	3.1	4	1	0	5.6	2.2	12
0	0	2	1.8	.33	5.1	4	1	1	6.8	3.0	14
0	1	0	.90	.04	3.6	4	1	22	8.0	3.6	17
0	1	1	1.8	.33	5.1	4	2	0	6.8	3.0	15
0	2	0	1.8	.33	5.1	4	2	1	8.0	3.6	17
0	2	1	2.7	.80	7.2	4	2	2	9.2	3.7	17
0	3	0	2.7	.80	7.2	4	3	0	8.1	3.6	17
1	0	0	.94	.05	5.1	4	3	1	9.3	4.5	18
1	0	1	1.9	.33	5.1	4	3	2	10	5.0	20
1	0	2	2.8	.80	7.2	4	4	0	9.3	4.5	18
1	1	0	1.9	.33	5.7	4	4	1	11	5.0	20
1	1	1	2.9	.80	7.2	4	5	0	11	5.0	20
1	1	2	3.8	1.4	9.0	4	5	1	12	5.6	22
1	2	0	2.9	.80	7.2	4	6	0	12	5.6	22
1	2	1	3.8	1.4	9.0	5	0	0	6.0	2.5	14
1	3	0	3.8	1.4	9.0	5	0	1	7.2	3.1	15
1	3	1	4.8	2.1	11	5	0	2	8.5	3.6	17
1	4	0	4.8	2.1	11	5	0	3	9.8	4.5	18
2	0	0	2.0	.37	7.2	5	1	0	7.3	3.1	15
2	0	1	3.0	.81	7.3	5	1	1	8.5	3.6	17
2	0	2	4.0	1.4	9.0	5	1	2	9.8	4.5	18
2	1	0	3.0	.82	7.8	5	1	3	11	5.0	21
2	1	1	4.0	1.4	9.0	5	2	0	8.6	3.6	17
2	1	2	5.0	2.1	11	5	2	1	9.9	4.5	18
2	2	0	4.0	1.4	9.1	5	2	2	11	5.0	21
2	2	1	5.0	2.1	11	5	3	0	10	4.5	18
2	2	2	6.1	3.0	14	5	3	1	11	5.0	21
2	3	0	5.1	2.1	11	5	3	2	13	5.6	23
2	3	1	6.1	3.0	14	5	4	0	11	5.0	21
2	4	0	6.1	3.0	14	5	4	1	13	5.6	23
2	4	1	7.2	3.1	15	5	4	2	14	7.0	26
2	5	0	7.2	3.1	15	5	5	0	13	6.3	25
3	0	0	3.2	.90	9.0	5	5	1	14	7.0	26
3	0	1	4.2	1.4	9.1	5	6	0	14	7.0	26
3	0	2	5.3	2.1	11	6	0	0	7.8	3.1	17
3	1	0	4.2	1.4	10	6	0	1	9.2	3.6	17
3	1	1	5.3	2.1	11	6	0	2	11	5.0	20
3	1	2	6.4	3.0	14	6	0	3	12	5.6	22
3	2	0	5.3	2.1	12	6	1	0	9.2	3.7	18
3	2	1	6.4	3.0	14	6	1	1	11	5.0	21
3	2	2	7.5	3.1	15	6	1	2	12	5.6	22
3	3	0	6.5	3.0	14	6	1	3	14	7.0	26
3	3	1	7.6	3.1	15	6	2	0	11	5.0	21
3	3	2	8.7	3.6	17	6	2	1	12	5.6	22
3	4	0	8.7	3.6	17	6	2	2	14	7.0	26
3	4	1	8.7	3.6	17	6	2	3	15	7.4	30
3	5	0	8.8	3.6	17	6	3	0	12	5.6	23
4	0	0	4.5	1.6	11	6	3	1	14	7.0	26
4	0	1	5.6	2.2	12	6	3	2	15	7.4	30

Table 10-7 (cont.). For 10 tubes at each of 0.1, 0.01, and 0.001 g inocula, the MPNs and 95% confidence intervals.

Positive tubes			MPN/g	Confidence limit		Positive tubes			MPN/g	Confidence limit
0.10	0.01	0.001		Low	High	0.10	0.01	0.001		Low	High
6	4	0	14	7.0	26	8	3	1	21	10	39
6	4	1	15	7.4	30	8	3	2	24	11	44
6	4	2	17	9.0	34	8	3	3	26	12	50
6	5	0	16	7.4	30	8	4	0	22	10	39
6	5	1	17	9.0	34	8	4	1	24	11	44
6	5	2	19	9.0	34	8	4	2	26	12	50
6	6	0	17	9.0	34	8	4	3	29	14	58
6	6	1	19	9.0	34	8	5	0	24	11	44
6	7	0	19	9.0	34	8	5	1	27	12	50
7	0	0	10	4.5	20	8	5	2	29	14	58
7	0	1	12	5.0	21	8	5	3	32	15	62
7	0	2	13	6.3	25	8	6	0	27	12	50
7	0	3	15	7.2	28	8	6	1	30	14	58
7	1	0	12	5.0	22	8	6	2	33	15	62
7	1	1	13	6.3	25	8	7	0	30	14	58
7	1	2	15	7.2	28	8	7	1	33	17	73
7	1	3	17	7.7	31	8	7	2	36	17	74
7	2	0	13	6.4	26	8	8	0	34	17	73
7	2	1	15	7.2	28	8	8	1	37	17	74
7	2	2	17	7.7	31	9	0	0	17	7.5	31
7	2	3	19	9.0	34	9	0	1	19	9.0	34
7	3	0	15	7.2	30	9	0	2	22	10	39
7	3	1	17	9.0	34	9	0	3	24	11	44
7	3	2	19	9.0	34	9	1	0	19	9.0	39
7	3	3	21	10	39	9	1	1	22	10	40
7	4	0	17	9.0	34	9	1	2	25	11	44
7	4	1	19	9.0	34	9	1	3	28	14	58
7	4	2	21	10	39	9	1	4	31	14	58
7	4	3	23	11	44	9	2	0	22	10	44
7	5	0	19	9.0	34	9	2	1	25	11	46
7	5	1	21	10	39	9	2	2	28	14	58
7	5	2	23	11	44	9	2	3	32	14	58
7	6	0	21	10	39	9	2	4	35	17	73
7	6	1	23	11	44	9	3	0	25	12	50
7	6	2	25	12	46	9	3	1	29	14	58
7	7	0	23	11	44	9	3	2	32	15	62
7	7	1	26	12	50	9	3	3	36	17	74
8	0	0	13	5.6	25	9	3	4	40	20	91
8	0	1	15	7.0	26	9	4	0	29	14	58
8	0	2	17	7.5	30	9	4	1	33	15	62
8	0	3	19	9.0	34	9	4	2	37	17	74
8	1	0	15	7.1	28	9	4	3	41	20	91
8	1	1	17	7.7	31	9	4	4	45	20	91
8	1	2	19	9.0	34	9	5	0	33	17	73
8	1	3	21	10	39	9	5	1	37	17	74
8	2	0	17	7.7	34	9	5	2	42	20	91
8	2	1	19	9.0	34	9	5	3	46	20	91
8	2	2	21	10	39	9	5	4	51	25	120
8	2	3	23	11	44	9	6	0	38	17	74
8	3	0	19	9.0	34	9	6	1	43	20	91

Table 10-7 (cont.). For 10 tubes at each of 0.1, 0.01, and 0.001 g inocula, the MPNs and 95% confidence intervals.

Positive tubes			MPN/g	Confidence limit		Positive tubes			MPN/g	Confidence limit
0.10	0.01	0.001		Low	High	0.10	0.01	0.001		Low	High
9	6	2	47	21	100	10	6	0	79	34	180
9	6	3	53	25	120	10	6	1	94	39	180
9	7	0	44	20	91	10	6	2	110	50	210
9	7	1	49	21	100	10	6	3	120	57	220
9	7	2	54	25	120	10	6	4	140	70	280
9	7	3	60	26	120	10	6	5	160	74	280
9	8	0	50	25	120	10	6	6	180	91	350
9	8	1	55	25	120	10	7	0	100	44	210
9	8	2	61	26	120	10	7	1	120	50	220
9	8	3	68	30	140	10	7	2	140	61	280
9	9	0	57	25	120	10	7	3	150	73	280
9	9	1	63	30	140	10	7	4	170	91	350
9	9	2	70	30	140	10	7	5	190	91	350
10	0	0	23	11	44	10	7	6	220	100	380
10	0	1	27	12	50	10	7	7	240	110	480
10	0	2	31	14	58	10	8	0	130	60	250
10	0	3	37	17	73	10	8	1	150	70	280
10	1	0	27	12	57	10	8	2	170	80	350
10	1	1	32	14	61	10	8	3	200	90	350
10	1	2	38	17	74	10	8	4	220	100	380
10	1	3	44	20	91	10	8	5	250	120	480
10	1	4	52	25	120	10	8	6	280	120	480
10	2	0	33	15	73	10	8	7	310	150	620
10	2	1	39	17	79	10	8	8	350	150	620
10	2	2	46	20	91	10	9	0	170	74	310
10	2	3	54	25	120	10	9	1	200	91	380
10	2	4	63	30	140	10	9	2	230	100	480
10	3	0	40	17	91	10	9	3	260	120	480
10	3	1	47	20	100	10	9	4	300	140	620
10	3	2	56	25	120	10	9	5	350	150	630
10	3	3	66	30	140	10	9	6	400	180	820
10	3	4	77	34	150	10	9	7	460	210	970
10	3	5	89	39	180	10	9	8	530	210	970
10	4	0	49	21	120	10	9	9	610	280	1300
10	4	1	59	25	120	10	10	0	240	110	480
10	4	2	70	30	150	10	10	1	290	120	620
10	4	3	82	38	180	10	10	2	350	150	820
10	4	4	94	44	180	10	10	3	430	180	970
10	4	5	110	50	210	10	10	4	540	210	1300
10	5	0	62	26	140	10	10	5	700	280	1500
10	5	1	74	30	150	10	10	6	920	350	1900
10	5	2	87	38	180	10	10	7	1200	480	2400
10	5	3	100	44	180	10	10	8	1600	620	3400
10	5	4	110	50	210	10	10	9	2300	810	5300
10	5	5	130	57	220	10	10	10	>2300	1300	--
10	5	6	140	70	280

Contents

Other Analytical Procedures

• Bacillus cereus in foods: Enumeration and confirmation microbiological methods (AOAC, 1995a).
• Differentiation of members of Bacillus cereus group: Microbiological method (AOAC, 1995b).

Contents

Commercial Test Products

Table 10-8. Commercial test products for B. cereus.

Test Kit	Analytical Technique	Approx. Total Test Time1	Supplier
BCET-RPLA TD950  [Used to identify B. cereus diarrheal enterotoxin]	Reversed passive latex agglutination	24 h (food)  48 h (bacterial culture)	Oxoid, Inc.  Contact: Jim Bell  217 Colonnade Rd.  Nepean, Ontario K2E 7K3  Canada  Phone: 613/226-1318  E-mail: jbell@oxoid.ca
TECRA Bacillus diarrheal Enterotoxin VIA  [Used to detect GDE toxin and Bacillus spp. capable of producing enterotoxin]	ELISA	4-24 h	InternationalBioProducts  Contact: Mike Yeager 14780 NE 95th St. Redmond, WA 98052  Phone: 425/861-4918  E-mail: myeager@intlbioproducts.com Web: intlbioproducts.com

¹Includes enrichment

Contents

References

Andrews, W.H., and June, G.A. 1998. Food sampling and preparation of sample homogenate, Ch. 1. In Food and Drug Administration Bacteriological Analytical Manual, 8th ed. (revision A), (CD-ROM version). R.L. Merker (Ed.). AOAC International, Gaithersburg, MD.

AOAC. 1995a. Bacillus cereus in foods: Enumeration and confirmation microbiological methods. Sec. 17.8.01, Method 980.31. In Official Methods of Analysis of AOAC International, 16th ed., P.A. Cunniff (Ed.), 52-54.AOAC International, Gaithersburg, MD.

AOAC. 1995b. Differentiation of members of Bacillus cereus group: Microbiological method. Sec. 17.8.02, Method 983.26. In Official Methods of Analysis of AOAC International, 16th ed., P.A. Cunniff (Ed.), 54-55.AOAC International, Gaithersburg, MD.

Bennett, R.W. 1998. Bacillus cereus diarrheal enterotoxin. Ch. 15. In Food and Drug Administration Bacteriological Analytical Manual, 8th ed. (revision A), (CD-ROM version). R.L. Merker (Ed.). AOAC International, Gaithersburg, MD.

Bennett, R. W., and S. M. Harmon. 1988. Bacillus cereus food poisoning. In Laboratory Diagnosis of Infectious Diseases: Principles and Practice, Vol. I, pp. 83-93. A. Balows, W. J. Hausler, Jr., M. Ohashi, and A. Turano (Eds). Springer-Verlag, New York.

Casman, E. P., R. W. Bennett, A. E. Dorsey, and J. E. Stone. 1969. The micro-slide gel double diffusion test for the detection and assay of staphylococcal enterotoxins. Health Lab. Sci. 6:185-198.

Cochran, W. G. 1950. Estimation of bacterial densities by means of the "most probable number." Biometrics 6:105-116.

Crowle, A. J. 1958. A simplified micro double-diffusion agar precipitin technique. J. Lab. Clin. Med. 52:784.

de Man, J. C. 1983. MPN tables, corrected. Eur. J. Appl. Biotechnol. 17:301-305.

Eisenhart, C., and P. W. Wilson. 1943. Statistical methods and control in bacteriology. Bacteriol. Rev. 7:57-137.

FDA. 1998. Bacterial pathogen growth. Appendix 4. In Fish and Fishery Products Hazards and Controls Guide, 2^nd ed., p. 241-244. Department of Health and Human Services, Public Health Service, Food and Drug Administration, Center for Food Safety and Applied Nutrition, Office of Seafood, Washington, DC.

Fluer, F.S. and Ezepchuk, Y.V. 1970. Microbiologiya 39:396-401. Cited in Rhodehammel, E.J. and Harmon, S.M. (1995) Bacillus cereus. Ch.14. In Food and Drug Administration Bacteriological Analytical Manual, 8^th ed. L.A. TomLinson (Ed.), 14.01-14.08. AOAC International, Gaithersburg, MD.

Garthright, W. E. 1993. Bias in the logarithm of microbial density estimates from serial dilutions. Biom. J. 35: 3, 299-314.

Garthright, W.E. 1998. Most probable number determination from serial dilutions. Appendix 2. In Food and Drug Administration Bacteriological Analytical Manual, 8th ed. (revision A), App. 2, (CD-ROM version). R.L. Merker (Ed.). AOAC International, Gaithersburg, MD.

Gilbert, R.J., Stringer, M.F., and Pearce, J.M. 1974. The survival and growth of Bacillus cereus in boiled and fried rice in relation to outbreaks of food poisoning. J. Hyg. 73:433-444.

Halvorson, H. O., and N. R. Ziegler. 1933. Application of statistics to problems in bacteriology. J. Bacteriol. 25:101-121; 26:331-339; 26:559-567.

Kramer, J.M. and Gilbert, J.M. 1989. Bacillus cereus and other Bacillus species. Ch. 2, In Foodborne Bacterial Pathogens, M.P.Doyle (Ed.), p. 21-70. Marcel Dekker, Inc., New York.

Lancette, G. A., and S. M. Harmon. 1980. Enumeration and confirmation of Bacillus cereus in foods: Collaborative study. J. Assoc. Off. Anal. Chem. 63:581-586.

McCrady, M. H. 1915. The numerical interpretation of fermentation-tube results. J. Infect. Dis. 17:183-212.

McFarland, J. 1907. The nephelometer: An instrument for estimating the number of bacteria in suspensions used for calculating the opsonic index and for vaccines. J. Am. Med. Assoc. 49:1176.

Merker, R.L. (Ed.). 1998. Media and reagents, Appendix 3. In Food and Drug Administration Bacteriological Analytical Manual, 8th ed. (revision A), (CD-ROM version). AOAC International, Gaithersburg, MD.

Peeler, J. T., G. A. Houghtby, and A. P. Rainosek. 1992. The most probable number technique. In Compendium of Methods for the Microbiological Examination of Foods, 3rd ed. C. Vanderzant and D. F. Splittstoesser (Ed.), p. 105-120. American Public Health Association, Washington, DC.

Pradhan, L., Kanekar, P., and Godbole, S.H. 1985. Microbiology of spoiled mango pickles. Tolerance to salt, acidity and oil of the microbes isolated from spoiled mango pickles. J. Food Sci. Technol India, 22(5):339-341.

Reed, G.H. 1994. Foodborne illness (Part 4): Bacillus cereus gastroenteritis. Dairy, Food, and Environmental San. 14(2):87.

Rhodehamel, E.J., and Harmon, S.M. 1998. Bacillus cereus. Ch.14. In Food and Drug Administration Bacteriological Analytical Manual, 8th ed. (revision A), (CD-ROM version. R.L. Merker (Ed.). AOAC International, Gaithersburg, MD.

Thomas, H.A. 1942. Bacterial densities from fermentation tube tests. J. Am. Water Works Assoc. 34:572-576.

Woodward, R. L. 1957. How probable is the most probable number? J. Am. Water Works Assoc. 49:1060-1068.