Food Science & Technology · Food Science Core · also Food Microbiology, Microbiology of Foods, Food Microbiology and Safety, Molecular Food Microbiology
Which microorganisms grow in which foods and why, the pathogens and toxins behind food-borne disease, and the methods — conventional, rapid and molecular — used to find them. The PG units on predictive microbiology and genomic surveillance carry it beyond the hygiene-focused milk and meat microbiology in the Veterinary track (milk-meat-hygiene-food-safety).
In the exams: GATE XE-G and XL-U Section 2 (spoilage, toxins from microbes); ICAR AIEEA-PG Code 18 Unit IV; ICAR AICE 35.b Units 2–4
Food microbiology is the study of the microorganisms that live in food: the ones that spoil it, the ones that make people ill, and the ones that are used to make it. This subject deals with the first two. It asks which organisms grow in which foods and why, which pathogens and toxins cause food-borne disease, and how a laboratory finds them, from the classical plate count to whole-genome sequencing. Fermentation and food biotechnology are taught as a separate subject.
The subject matters because nearly every decision in a food plant is a microbiological decision. Formulation, chilling, shelf life, drain design, sampling plans and the response to a positive Listeria swab all depend on it. By World Health Organization estimates, food-borne disease causes hundreds of millions of illnesses a year, falling heaviest on young children in low- and middle-income countries.
The subject is examined in every food technology degree. In India it is Section 2 of the GATE XE-G and XL-U food technology papers, Unit IV of the ICAR AIEEA-PG Code 18 paper and part of the ICAR AICE-JRF/SRF 35.b Food Safety and Quality syllabus, and it is central to Food Safety Officer and Food Analyst recruitment. Examiners favour growth arithmetic, the characteristics that identify each pathogen, spoilage types and their causes, indicator organisms, the steps of standard detection methods, and two- and three-class sampling plans.
Microorganisms need available water, and the measure of availability is water activity (aw): the ratio of the vapour pressure of water over the food to that over pure water at the same temperature. Fresh meat, milk, fruit and vegetables have an aw of about 0.97 to 0.99. Water activity multiplied by 100 is the equilibrium relative humidity (ERH): the relative humidity of the air at which the food neither gains nor loses moisture.
Each group of organisms has a minimum aw for growth. Standard textbook values are about 0.91 for most spoilage bacteria, 0.88 for most yeasts and 0.80 for most moulds. Staphylococcus aureus can grow down to about 0.86 aerobically, which is why it is the pathogen of concern in cured meats and other salted foods. Proteolytic Clostridium botulinum needs about 0.94 and the non-proteolytic type about 0.97. Halophilic bacteria grow at about 0.75, xerophilic moulds at about 0.61 and osmophilic yeasts at about 0.60, which is why honey and dried fruit can still spoil. Below about 0.60 no microorganism grows, although many survive: Salmonella can persist for months in chocolate, peanut butter, spices and milk powder. Intermediate-moisture foods, with aw between about 0.60 and 0.90, are stable at room temperature only when aw is combined with other controls, because moulds and some yeasts still grow in that range.
Most bacteria grow best near neutrality and few pathogens grow below about pH 4.0 to 4.5, while many yeasts and moulds tolerate pH 2 to 3. The most important single number is pH 4.6: proteolytic C. botulinum does not grow or form toxin at or below it, so foods are divided into acid foods (pH 4.6 or below) and low-acid foods (above 4.6), a boundary written into canning law in many countries. The type of acid matters as well as the pH: weak organic acids such as acetic, lactic, sorbic and benzoic acid are more inhibitory than strong acids at the same pH, because the undissociated molecule crosses the cell membrane and releases protons inside the cell.
The oxidation-reduction (redox) potential, Eh, measures how readily the food gains or loses electrons. Aerobes such as Pseudomonas and moulds need a positive Eh; obligate anaerobes such as clostridia need a negative one; facultative organisms such as the Enterobacteriaceae and lactic acid bacteria grow across the range.
Some contain natural antimicrobials: the lysozyme and iron-binding ovotransferrin of egg white, the lactoperoxidase system and lactoferrin of milk. Biological structures act as barriers: the shell and membranes of an egg, the skin of fish and animals, and the peel of fruit. Once these are broken by bruising, cutting or mincing, spoilage accelerates, which is why minced meat spoils faster than an intact carcass.
Extrinsic factors are those of the storage environment. Temperature is the most important. Organisms are grouped by their cardinal temperatures: psychrophiles grow best below about 15 °C and do not grow above about 20 °C; psychrotrophs grow at refrigeration temperatures but have optima of 20 to 30 °C; mesophiles, which include most pathogens, have optima of about 30 to 40 °C; thermophiles have optima above about 45 °C. The relative humidity of the store affects surface aw: a dry food in a humid store picks up moisture and grows mould on its surface. The gas atmosphere matters: carbon dioxide inhibits many Gram-negative aerobes, which is the basis of modified atmosphere packaging, while removing oxygen favours lactic acid bacteria and anaerobes.
Hurdle technology, described by Leistner, combines several moderate hurdles, such as reduced aw, lowered pH, mild heat, preservatives and chilling, each too weak alone, to give a stable and safe food with less damage to quality.
Growth is by binary fission. During exponential growth the number N after time t is N = N0 × 2^n, where n = t/g is the number of generations and g the generation (doubling) time. The number of generations between two counts is n = (log N − log N0)/log 2, that is 3.32 × (log N − log N0). If a count rises from 10^3 to 10^6 CFU/g in 5 hours, n = 3.32 × 3 = 9.97 generations and g is about 30 minutes. The specific growth rate on a natural-log basis is µ = ln 2/g. Real growth curves add a lag phase before exponential growth and a stationary phase after it.
Spoilage is any change that makes a food unacceptable to the consumer: off-odours, off-flavours, slime, gas, discoloration or changes in texture. It is usually caused by organisms that reach high numbers, often above about 10^6 to 10^7 CFU per gram or per square centimetre, although the level differs between foods and between organisms.
At room temperature raw milk sours: mesophilic lactic acid bacteria, chiefly Lactococcus lactis, ferment lactose to lactic acid until casein coagulates. In refrigerated raw milk, Gram-negative psychrotrophs, especially Pseudomonas fluorescens, dominate. They are killed by pasteurisation and by UHT treatment, but many produce heat-stable proteases and lipases that survive even UHT. Those enzymes later cause bitterness, rancidity and age gelation in UHT milk, so the quality of the raw milk limits the shelf life of the heated product.
Pasteurised milk spoils in two ways. Post-pasteurisation contamination, from fillers and pipework, introduces Gram-negative psychrotrophs that grow quickly in the cold and give off-flavours and curdling. Heat-resistant spore-formers that survive pasteurisation, chiefly psychrotrophic Bacillus cereus and related species, germinate and grow slowly in the cold; their lecithinase breaks down the fat globule membrane, giving bitty cream, and their proteases cause sweet curdling, coagulation without acid. Coliforms in pasteurised milk indicate recontamination, because pasteurisation kills them.
The muscle of a healthy animal is nearly sterile; contamination comes from the hide, the gut contents, equipment and workers during slaughter and dressing.
Under chilled aerobic storage, Gram-negative psychrotrophs, especially Pseudomonas species, outgrow everything else. They first use glucose and, when it runs out at the surface, attack amino acids, releasing ammonia, sulphides and amines. Off-odours usually appear at around 10^7 CFU/cm² and surface slime at around 10^8. Meat with a high ultimate pH, such as dark, firm, dry meat from stressed animals, has little glucose and spoils sooner.
Vacuum packing or modified atmosphere packing with carbon dioxide suppresses pseudomonads and selects lactic acid bacteria, which grow slowly and give a mild sour odour, so the shelf life of vacuum-packed chilled beef is many weeks. Brochothrix thermosphacta, a facultative Gram-positive rod, gives cheesy or dairy odours where some oxygen remains. In high-pH meat, Shewanella putrefaciens produces hydrogen sulphide, which combines with myoglobin to give green sulphmyoglobin. Psychrotrophic clostridia, notably Clostridium estertheticum, cause blown pack spoilage of vacuum-packed chilled meat, with gross gas production.
Some lactic acid bacteria produce hydrogen peroxide that oxidises the cured meat pigment and causes greening of sausages and ham.
Fish spoil faster than meat. Their flesh has a higher post-mortem pH, little carbohydrate, a high content of non-protein nitrogen, and a natural flora adapted to cold water. Marine fish contain trimethylamine oxide (TMAO), which specific spoilage organisms, especially Shewanella putrefaciens, reduce to trimethylamine (TMA), the main source of the fishy odour; TMA and total volatile basic nitrogen are used as chemical indices of freshness. In fish of the tuna and mackerel families, bacteria such as Morganella morganii decarboxylate histidine to histamine if the fish are not chilled promptly; histamine is not destroyed by cooking or canning and causes scombroid poisoning.
A fresh egg is well defended. The cuticle and shell keep out most bacteria; the shell membranes filter; and the albumen is hostile, with an alkaline pH that rises to about 9 to 9.5 as carbon dioxide is lost, lysozyme that attacks Gram-positive cell walls and ovotransferrin that binds iron. The rots are named by appearance: green rots caused by Pseudomonas fluorescens, black rots caused by Proteus, Pseudomonas and Aeromonas species, red rots caused by Serratia, and colourless rots. Salmonella Enteritidis is a separate concern: it can infect the reproductive tract of the hen and contaminate the egg contents before the shell forms.
Grains carry a large and varied microflora from the field, but their low water activity at safe storage moisture keeps it dormant. Fungi are conventionally divided into field fungi and storage fungi. Field fungi, such as Fusarium, Alternaria and Cladosporium, invade the grain while it is still in the plant and need high moisture; they decline once the grain is dried. Storage fungi, chiefly species of Aspergillus and Penicillium, are xerophilic and grow in stored grain whose moisture has risen, from inadequate drying, condensation, leaks or insect activity, which itself produces water and heat. Their growth heats the grain, discolours it, reduces germination and can produce mycotoxins (chapter 7). The safe moisture content for storage differs between grains and climates and is set nationally; the principle is to keep the grain below the aw at which storage moulds grow, roughly 0.65 to 0.70.
Flour and meal carry spores of moulds and of Bacillus species, which matter when water is added.
Fruits are acidic, usually below pH 4.5, and rich in sugar, so bacteria are largely excluded and spoilage is caused by moulds and yeasts. Standard examples are Penicillium expansum, the cause of blue mould rot of apples and pears and a producer of the mycotoxin patulin; Penicillium digitatum and P. italicum, the green and blue moulds of citrus; Botrytis cinerea, grey mould of strawberries, grapes and many soft fruits, which grows at refrigeration temperatures; and Rhizopus stolonifer, a fast-growing soft rot of berries and stone fruit at warmer temperatures.
Post-harvest control combines careful handling to avoid wounds, prompt cooling, controlled atmosphere storage with reduced oxygen and raised carbon dioxide, sanitation of packing lines and wash water, and, where permitted, approved post-harvest fungicides.
Vegetables have a pH between about 5 and 7, so bacteria compete with moulds, and bacterial soft rot is the most important spoilage of fresh vegetables. Pectobacterium carotovorum (formerly Erwinia carotovora) and some Pseudomonas species produce pectinases that dissolve the middle lamella binding plant cells together, turning tissue into a watery, often foul-smelling mass.
Fresh-cut vegetables and salads are a special case. Cutting releases nutrients, removes the protective skin and increases respiration, so counts rise quickly unless the product is washed with a sanitiser and kept cold. Modified atmosphere packaging slows the vegetable's respiration and some spoilage organisms, but it does not stop psychrotrophic pathogens: Listeria monocytogenes can grow on refrigerated vegetables, and salads and sprouts have carried Salmonella and Shiga toxin-producing E. coli. Sprouted seeds are particularly risky because the warm, moist sprouting conditions let pathogens on the seed multiply greatly.
A commercially sterile can contains no microorganisms able to grow under normal storage conditions, but it is not absolutely sterile: thermophilic spores may survive and stay dormant unless the can is stored hot.
Underprocessing of low-acid foods allows mesophilic spore-formers to survive. Putrefactive anaerobes such as Clostridium sporogenes, whose spores are more heat resistant than those of C. botulinum and are used as a surrogate in process validation, swell the can and produce a putrid odour; survival of C. botulinum itself is the greatest hazard. Thermophilic spoilage follows slow cooling or storage at high temperatures, as in hot climates or vending machines. Flat sour spoilage, acid without gas so the can stays flat, is caused by Geobacillus stearothermophilus in low-acid foods and by Bacillus coagulans in acid foods such as tomato products. Thermophilic anaerobe (TA) spoilage, by Thermoanaerobacterium thermosaccharolyticum, produces hydrogen and carbon dioxide, swells the can and gives a sour, cheesy odour. Sulphide stinker spoilage, by Desulfotomaculum nigrificans, produces hydrogen sulphide that blackens the contents without swelling the can. Butyric acid anaerobes such as Clostridium pasteurianum spoil moderately acid foods.
Leaker spoilage is different: a mixed flora of non-spore-forming bacteria, often including cocci and rods, indicates contamination after processing through a faulty seam or by contaminated cooling water, not underprocessing. Chemical hydrogen swells, from corrosion of the can by acid foods, swell the can without any microbial cause. Swollen cans are described, in increasing severity, as flippers, springers, soft swells and hard swells.
Heat-resistant moulds, such as Byssochlamys fulva and B. nivea, have ascospores that survive the mild heat given to fruit products and spoil canned fruit and juices by softening the fruit. Alicyclobacillus acidoterrestris, an acid-tolerant spore-former, spoils fruit juices by producing guaiacol, a medicinal or smoky taint, without gas.
Baking kills vegetative cells and most mould spores, so bread leaves the oven nearly free of moulds; it is recontaminated by airborne spores during cooling, slicing and wrapping. Mould is the commonest spoilage of bread and cakes. Rhizopus stolonifer, the black bread mould, and species of Penicillium and Aspergillus are typical. Control relies on hygiene in cooling and slicing areas, filtered air, reduced aw, weak-acid preservatives such as propionate and sorbate where permitted, and modified atmosphere packaging.
Rope is a bacterial spoilage of bread caused by Bacillus subtilis and B. licheniformis, whose spores survive baking in the centre of the loaf. In warm weather they germinate and break down starch and protein, giving a soft, sticky, stringy crumb that can be pulled into threads, and a smell often compared to over-ripe melon. Acidifying the dough and using propionate help to control it.
Soft drinks, fruit juices, beer and wine are acidic, so their spoilage organisms are acid-tolerant. In soft drinks and juices, yeasts are the main problem. Zygosaccharomyces bailii is notorious because it tolerates high sugar, low pH and the weak-acid preservatives sorbate and benzoate at concentrations that stop most other yeasts. Acetic acid bacteria such as Acetobacter and Gluconobacter oxidise ethanol and sugars to acids in juices and in wine exposed to air.
In beer, hops inhibit most Gram-positive bacteria, but some lactic acid bacteria, notably Lactobacillus and Pediococcus species, carry hop-resistance genes and cause sourness, turbidity and the buttery off-flavour diacetyl.
Leuconostoc mesenteroides, which makes the slimy polysaccharide dextran from sucrose, is an important spoilage organism in sugar refining and in sucrose-containing beverages.
Salmonella is a genus of Gram-negative, facultatively anaerobic, usually motile rods in the Enterobacteriaceae. Serotyping by somatic (O) and flagellar (H) antigens under the White-Kauffmann-Le Minor scheme recognises more than 2,500 serovars. Typhoidal serovars (Typhi and Paratyphi) are restricted to humans and spread by faecal contamination of water and food. Non-typhoidal serovars, such as Enteritidis and Typhimurium, are zoonotic: animals, especially poultry and pigs, carry them in the gut, and they reach people in undercooked poultry, eggs, meat, raw milk, and produce or low-moisture foods contaminated by animal faeces. Infection causes gastroenteritis with diarrhoea, abdominal cramps and fever, usually 12 to 72 hours after eating. Salmonella grows between about 5 and 46 °C, does not grow at refrigeration temperatures, and is killed by normal cooking and pasteurisation, but it survives drying and freezing for long periods.
Shigella is closely related to E. coli. Its four species (S. dysenteriae, S. flexneri, S. boydii and S. sonnei) are non-motile, usually do not ferment lactose and produce no hydrogen sulphide. They are human-adapted, with no animal reservoir, so food-borne shigellosis is almost always due to an infected food handler or to faecally contaminated water used on produce. The infectious dose is very low, of the order of 10 to 100 cells, which explains easy person-to-person spread. Shigella invades the colonic epithelium and causes bacillary dysentery, with bloody, mucoid stools; S. dysenteriae type 1 produces Shiga toxin.
Listeria monocytogenes is a Gram-positive, non-spore-forming, facultatively anaerobic rod. It shows tumbling motility at 20 to 25 °C but little at 37 °C, is weakly beta-haemolytic, and gives a positive CAMP test with S. aureus. Its food-safety importance comes from its tolerance: it grows from close to 0 °C up to about 45 °C, tolerates salt concentrations around 10 per cent, and grows down to an aw of about 0.92. It is widespread in soil, water and the intestines of animals, and it establishes itself in food plants, in drains, on floors and in hard-to-clean equipment, from which it contaminates ready-to-eat foods after cooking. Invasive listeriosis affects mainly pregnant women, in whom it can cause abortion and stillbirth, newborns, the elderly and the immunocompromised. It has a long incubation period, up to several weeks, and a high case-fatality rate among hospitalised cases, often quoted as 20 to 30 per cent. Typical vehicles are soft cheeses, deli meats, smoked fish, pâtés, and refrigerated foods with long shelf lives. Serotypes 1/2a, 1/2b and 4b cause most human cases.
Shiga toxin-producing E. coli (STEC), including enterohaemorrhagic E. coli, produce Shiga toxins 1 and 2 and usually the intimin adhesin. Serotype O157:H7 is the best known, but non-O157 serogroups such as O26, O103, O111 and O145 cause a large share of cases. Ruminants, especially cattle, are the main reservoir. Vehicles include undercooked minced beef, raw milk, unpasteurised juice, leafy greens and sprouts. The infectious dose is low, of the order of 10 to 100 cells. Illness ranges from watery to bloody diarrhoea, and some patients, especially children, develop haemolytic uraemic syndrome with kidney failure. Most O157:H7 strains do not ferment sorbitol within 24 hours and are beta-glucuronidase negative, which distinguishes them from most other E. coli on selective media.
Campylobacter jejuni and C. coli are curved or spiral Gram-negative rods with darting motility. They are microaerophilic, growing best in about 5 per cent oxygen and 10 per cent carbon dioxide, and thermotolerant, growing at 42 °C but not below about 30 °C, so they do not multiply in food. They are nevertheless a leading cause of bacterial gastroenteritis in many high-income countries, because poultry carcasses are very commonly contaminated and cross-contamination in kitchens is easy. Illness follows after 2 to 5 days; a rare but serious sequel is Guillain-Barré syndrome, an acute paralytic neuropathy.
Vibrio species are curved Gram-negative rods of marine and estuarine water. V. parahaemolyticus is halophilic, needing salt, and causes gastroenteritis from raw or undercooked seafood; its virulence is linked to the thermostable direct haemolysin. V. vulnificus, from raw oysters or wounds exposed to seawater, causes septicaemia with high mortality in people with liver disease. V. cholerae serogroups O1 and O139 cause cholera through cholera toxin. On TCBS agar V. cholerae, which ferments sucrose, forms yellow colonies, while V. parahaemolyticus forms green ones.
Yersinia enterocolitica is a Gram-negative rod in the Enterobacteriaceae that grows at refrigeration temperatures. Pigs are the main reservoir, especially the tonsils, and undercooked pork is the classic vehicle. Infection causes enterocolitis and mesenteric lymphadenitis that can mimic appendicitis. Laboratories use cold enrichment and CIN agar, on which colonies have a red bull's-eye centre.
Botulism is a neuroparalytic illness caused by botulinum neurotoxin, the most potent biological toxin known. The toxin blocks release of acetylcholine at nerve endings, causing symmetrical descending flaccid paralysis that can progress to respiratory failure. Food-borne botulism is an intoxication: the toxin is formed in the food and eaten. Infant botulism and wound botulism are infections in which spores germinate in the body; honey is a recognised source of spores for infants.
C. botulinum is not one species but a group of strict anaerobes that share the ability to make the toxin. Two groups matter in food. Group I (proteolytic) strains, producing toxin types A, B or F, have very heat-resistant spores, a minimum growth temperature of about 10 °C, a minimum pH of 4.6 and a minimum aw of about 0.94. Their spores define the botulinum cook of low-acid canned food: a 12D process, using a D121 of about 0.21 minutes, gives a minimum F0 of about 2.5 minutes, and commercial processes are usually longer. Group II (non-proteolytic) strains, producing types B, E or F, grow and make toxin at temperatures as low as about 3 °C, need an aw of about 0.97 and a pH above about 5.0, and have spores of much lower heat resistance. They are the hazard in chilled, vacuum-packed and modified-atmosphere foods with long shelf lives, and type E is associated with fish. Because they do not putrefy the food, toxic food may look and smell normal.
The toxin itself is heat-labile and is inactivated by heating at about 85 °C for 5 minutes or by boiling, but spores need far more severe treatment. Control rests on the canning process, acidification to pH 4.6 or below, reduced aw, salt and nitrite in cured meats, refrigeration below 3 °C, and limits on shelf life of chilled foods.
Clostridium perfringens is an anaerobic, non-motile, Gram-positive spore-forming rod found in soil, dust and the intestines of people and animals. Its spores survive cooking. It grows extremely fast at warm temperatures, with generation times as short as about 10 minutes near 43 to 45 °C, and it grows up to about 50 °C. Outbreaks therefore follow the same pattern: a large quantity of meat, poultry, stew or gravy is cooked, then cooled slowly or held warm, spores germinate, cells multiply to high numbers, and the food is eaten without adequate reheating. In the intestine the cells sporulate and release C. perfringens enterotoxin, causing abdominal pain and diarrhoea, usually without vomiting or fever, 8 to 22 hours after eating. Enterotoxin-producing strains were classically called type A; a revised typing scheme published in 2018 places most of them in a new type F. Rapid cooling of cooked food through the range of about 50 to 15 °C, and reheating to a high internal temperature, prevent it.
Bacillus cereus is an aerobic to facultatively anaerobic, motile, Gram-positive spore-forming rod common in soil and on cereals, spices and dried foods. It causes two syndromes. The emetic syndrome is an intoxication by cereulide, a small cyclic peptide formed in the food, which is stable to heat, acid and proteolysis; nausea and vomiting begin 0.5 to 6 hours after eating, and cooked rice held warm or cooled slowly is the classic vehicle. The diarrhoeal syndrome is caused by heat-labile protein enterotoxins (haemolysin BL, non-haemolytic enterotoxin and cytotoxin K) produced by cells growing in the small intestine; diarrhoea and cramps begin 8 to 16 hours after eating meat dishes, vegetables, sauces or milk products. On mannitol egg yolk polymyxin (MYP) agar, B. cereus forms pink colonies, because it does not ferment mannitol, surrounded by a zone of egg-yolk precipitate from lecithinase.
Staphylococcus aureus is a Gram-positive, non-motile, facultatively anaerobic coccus that forms grape-like clusters. It is carried in the nose and on the skin of a large part of the human population and in infected wounds and boils, and it causes mastitis in dairy animals. Staphylococcal food poisoning is an intoxication by staphylococcal enterotoxins, a family of small proteins (SEA to SEE and many more) produced as the organism grows in food. The enterotoxins are heat-stable: they survive cooking, and even canning-level heat does not reliably inactivate them, so killing the cells does not make the food safe once toxin has formed. Symptoms are sudden nausea, vomiting and abdominal cramps, often with diarrhoea, 1 to 6 hours after eating, and recovery usually within a day or two. Enough toxin to cause illness generally forms only when counts reach about 10^5 to 10^6 CFU per gram.
S. aureus competes poorly with other bacteria, but it is tolerant of salt and low aw, growing down to about 0.86, so it is favoured in foods where competitors have been removed or inhibited: cooked meats, ham, poultry salads, cream-filled pastries, cheese and other foods handled after cooking. The typical outbreak is a cooked food handled by a carrier and then held at room temperature for hours. On Baird-Parker agar, with egg yolk and tellurite, colonies are black and shiny, surrounded by a clear zone; confirmation is by the coagulase test. Control depends on hand hygiene, keeping food handlers with skin infections away from food, and keeping cooked food hot or cold.
Mycotoxins are toxic secondary metabolites of moulds. Unlike bacterial food poisoning, mycotoxicosis is usually a chronic problem caused by repeated intake of contaminated staple foods, and the toxins are mostly stable to cooking and processing, so prevention must happen in the field and in storage. A mould can be present without toxin, and toxin can remain after the mould has died or been removed.
Aflatoxins are produced mainly by Aspergillus flavus and A. parasiticus in warm, humid conditions, in crops such as maize, groundnuts, tree nuts, cottonseed, chillies and spices. A. flavus usually produces aflatoxins B1 and B2, while A. parasiticus produces B1, B2, G1 and G2; the letters refer to blue or green fluorescence under ultraviolet light. Aflatoxin B1 is the most toxic and is classified by the International Agency for Research on Cancer as a Group 1 human carcinogen; it causes liver cancer, and the risk is greatly increased in people with chronic hepatitis B infection. The 1960 outbreak of Turkey X disease in Britain, traced to groundnut meal, led to the discovery of the aflatoxins. Dairy animals that eat feed containing B1 excrete the hydroxylated metabolite aflatoxin M1 in milk, which is why milk is monitored for M1.
Limits vary widely between jurisdictions. Codex sets a maximum level of 0.5 µg/kg for aflatoxin M1 in milk and 15 µg/kg for total aflatoxins in peanuts intended for further processing; the European Union is stricter for many products, for example setting 0.050 µg/kg for M1 in milk.
Ochratoxin A is a nephrotoxin and a possible human carcinogen. In temperate climates it is produced mainly by Penicillium verrucosum in stored cereals; in warmer climates by Aspergillus ochraceus and, in grapes, wine and dried vine fruit, by A. carbonarius. Patulin, produced by Penicillium expansum in rotting apples and pears, is monitored in apple juice; Codex and the EU set a maximum of 50 µg/kg in apple juice. Removing rotten fruit before pressing is the main control, and alcoholic fermentation destroys patulin.
Fusarium species are field fungi of cereals, and their toxins form mainly before harvest, especially in wet weather at flowering. Deoxynivalenol (DON, vomitoxin), a trichothecene produced by F. graminearum and F. culmorum in wheat, barley and maize, causes feed refusal and vomiting in pigs and is the most frequently detected mycotoxin in cereals in many surveys. T-2 toxin, another trichothecene, was associated with alimentary toxic aleukia in the former Soviet Union in the 1940s. Zearalenone, produced by F. graminearum, is oestrogenic and causes hyperoestrogenism and infertility in pigs. Fumonisins, produced by F. verticillioides and F. proliferatum in maize, cause leukoencephalomalacia in horses and pulmonary oedema in pigs, and are associated with oesophageal cancer and neural tube defects in human populations with high maize intake.
Mycotoxin control follows the stages of the food chain. Before harvest: resistant varieties, crop rotation and tillage to reduce inoculum, irrigation and insect control to reduce plant stress, and, for aflatoxin, biological control with non-toxigenic strains of A. flavus that compete with toxigenic ones. After harvest: rapid drying to a safe moisture content, clean and dry storage, control of insects and condensation, and monitoring of temperature. In processing: sorting, including optical sorting of nuts, cleaning and removal of damaged kernels, which concentrates toxin in a small fraction that can be discarded. For animal feed, permitted binders that adsorb some toxins in the gut are used.
Sampling is the largest source of error in mycotoxin testing, because toxin is distributed very unevenly: a few highly contaminated kernels can carry most of the toxin in a lot. Regulations therefore specify large aggregate samples, made from many incremental samples, which are ground and subsampled before analysis. Screening uses enzyme-linked immunosorbent assays and lateral flow test strips. Confirmatory methods use cleanup by immunoaffinity columns followed by high-performance liquid chromatography with fluorescence detection, or liquid chromatography with tandem mass spectrometry, which can measure many toxins at once.
Food-borne viruses differ from bacteria in ways that change how they are controlled. They are obligate intracellular parasites, so they do not multiply in food; food is only a vehicle. The infectious dose is very low. Most are non-enveloped, which makes them resistant to drying, freezing, acid and many disinfectants. Most cannot be grown easily in routine cell culture, so they are detected by molecular methods, and a positive result shows genome, not necessarily infectious virus.
Norovirus is the leading cause of food-borne gastroenteritis in many countries. It causes sudden vomiting and diarrhoea after 12 to 48 hours, lasting a day or two. The usual food-borne routes are infected food handlers who contaminate ready-to-eat foods, bivalve shellfish such as oysters that filter virus from sewage-polluted water, and berries and leafy produce irrigated or handled with contaminated water or hands. Hepatitis A virus causes liver inflammation after a long incubation period, averaging about four weeks; shellfish, frozen berries and food handlers are the main sources. Hepatitis E virus genotype 3 is zoonotic in pigs and wild boar and has been linked to undercooked pork liver and sausages. Control relies on excluding ill food handlers for a period after symptoms stop, hand washing with soap and water, harvesting shellfish only from classified clean waters, and thorough cooking. Standardised detection uses RT-qPCR methods such as ISO 15216 for hepatitis A virus and norovirus in food.
Protozoan parasites are transmitted mainly as resistant cysts or oocysts in faecally contaminated water and produce, or as tissue cysts in meat. Toxoplasma gondii has the cat as its definitive host, which sheds oocysts in faeces; people are infected by eating tissue cysts in undercooked meat, especially lamb, pork and game, or oocysts on unwashed produce or in water. Infection is usually mild, but infection acquired in pregnancy can damage the fetus, and it is serious in the immunocompromised. Cooking to a sufficient core temperature and deep freezing inactivate tissue cysts.
Cryptosporidium parvum and C. hominis cause watery diarrhoea that can be prolonged and severe in people with weakened immunity. Their oocysts are infectious when shed, highly resistant to chlorine, and small enough to pass some filters, so they cause waterborne outbreaks and outbreaks from unpasteurised juice, raw milk and salads. Cyclospora cayetanensis has caused outbreaks linked to imported fresh berries, herbs and leafy greens; its oocysts must mature in the environment before becoming infectious, so it does not spread directly from person to person.
Helminths reach people mainly through raw or undercooked meat and fish. Trichinella spiralis larvae encyst in the muscle of pigs, wild boar, horses and bears; eating undercooked meat causes trichinellosis, with muscle pain, fever and periorbital oedema. Meat inspection uses digestion of muscle samples to find the larvae. Freezing kills T. spiralis in pork under defined time-temperature conditions, but Arctic species such as T. nativa in game are freeze-tolerant. Taenia saginata, the beef tapeworm, and T. solium, the pork tapeworm, are acquired by eating cysticerci in undercooked meat. T. solium is more dangerous because a person who swallows its eggs, from a carrier's faeces, develops cysticercosis, and cysts in the brain cause epilepsy.
Anisakis simplex and related nematodes live in marine fish and squid. Eating raw, lightly cured or undercooked fish can cause acute abdominal pain when larvae penetrate the gut wall, and allergic reactions. Freezing kills the larvae, and fish to be eaten raw must be frozen first; regulations specify the conditions, for example at least −20 °C for 24 hours in the European Union, while US guidance specifies −20 °C for 7 days or shorter times at −35 °C.
Testing every lot of food for every pathogen is impossible, so laboratories test for organisms whose presence signals a problem. Following Mossel's distinction, an index organism signals the possible presence of an ecologically similar pathogen, as E. coli signals faecal contamination and so the possible presence of enteric pathogens. An indicator organism signals a failure of process or hygiene, as Enterobacteriaceae in a pasteurised product signal inadequate heating or recontamination.
The aerobic plate count, or total viable count, estimates the general microbial load. It reflects raw material quality, hygiene and storage history, but it is not a safety measure: a fermented food has a high count and is safe, while a food with a low count can carry a pathogen. Coliforms are Gram-negative, non-spore-forming, facultatively anaerobic rods that ferment lactose with acid and gas within 48 hours at 35 to 37 °C. The group includes environmental genera such as Enterobacter and Klebsiella, so coliforms in raw produce say little, but in pasteurised milk or a cooked product they show recontamination after heating, because heat kills them. Thermotolerant (faecal) coliforms ferment lactose at 44 to 44.5 °C and are closer to a faecal signal. E. coli itself is the best index of faecal contamination. Enterobacteriaceae, all glucose-fermenting and oxidase-negative, are a broader hygiene indicator favoured in Europe. Enterococci are more resistant to drying, freezing and heat and are used for water and some processed foods. In environmental monitoring, Listeria species are used as an indicator for L. monocytogenes, because conditions that let one species persist will let the pathogen persist.
Most counts begin with a homogenised sample, typically 10 g or 25 g in nine times its weight of diluent, giving a 10^−1 dilution, followed by ten-fold serial dilutions. In the pour plate method, 1 mL of a dilution is mixed with molten agar cooled to about 45 °C; in the spread plate method, 0.1 mL is spread on the surface of a set plate. Plates with between about 25 and 250 colonies, or 30 and 300 in some standards, are counted; each colony is assumed to come from one cell or clump, so results are reported as colony-forming units (CFU). The count is the number of colonies divided by the volume plated and by the dilution: 156 colonies from 0.1 mL of a 10^−4 dilution means 156 / (0.1 × 10^−4) = 1.56 × 10^7 CFU/g. Incubation conditions differ between standards; ISO 4833 incubates the aerobic count at 30 °C for 72 hours, while the US Bacteriological Analytical Manual uses 35 °C for 48 hours. Psychrotrophic counts use about 7 °C for 10 days.
The most probable number (MPN) method estimates numbers statistically from the pattern of positive tubes in a dilution series, commonly three or five tubes at each of three ten-fold dilutions. It suits low numbers and samples with particles that interfere with plating, and it is the classical method for coliforms and E. coli. In the classical coliform procedure, lauryl sulphate tryptose broth tubes showing gas are presumptive positives; transfer to brilliant green lactose bile broth with gas confirms coliforms; and transfer to EC broth at 44.5 °C with gas indicates faecal coliforms, followed by tests that confirm E. coli. The result is read from MPN tables for the combination of positive tubes and the sample amounts used.
Detection of a pathogen, rather than counting, usually asks whether it is present in a defined amount of food, often 25 g. Because the pathogen may be few in number and injured by processing, detection proceeds in stages. Non-selective pre-enrichment lets injured cells recover. Selective enrichment suppresses competitors. Plating on selective and differential agars produces colonies that can be recognised. Biochemical and serological tests confirm identity.
The standard Salmonella method illustrates the sequence. ISO 6579-1 uses pre-enrichment in buffered peptone water, selective enrichment in Rappaport-Vassiliadis soya broth and Muller-Kauffmann tetrathionate-novobiocin broth, and plating on xylose lysine deoxycholate (XLD) agar, where typical colonies are red with black centres from hydrogen sulphide, plus a second medium. Suspect colonies are confirmed with triple sugar iron, lysine and urease tests and serotyping with O and H antisera. For Listeria, ISO 11290-1 uses half Fraser then Fraser broth and plating on chromogenic agar such as ALOA, on which L. monocytogenes forms blue-green colonies with an opaque halo from phospholipase activity. E. coli O157 is concentrated by immunomagnetic separation and plated on cefixime-tellurite sorbitol MacConkey agar, where it forms colourless, sorbitol-negative colonies. Campylobacter needs microaerobic incubation at 41.5 °C on modified charcoal cefoperazone deoxycholate agar. Other standard media include Baird-Parker for S. aureus, MYP for B. cereus, tryptose sulphite cycloserine agar for C. perfringens (black colonies), TCBS for Vibrio, CIN for Yersinia, and dichloran rose bengal chloramphenicol agar or dichloran glycerol agar for yeasts and moulds.
ATP bioluminescence measures adenosine triphosphate on a surface or in a liquid. The firefly enzyme luciferase uses ATP to oxidise luciferin, producing light that a luminometer reads in relative light units within seconds. All living cells contain ATP, but so do food residues, so the test measures total organic soil, microbial and non-microbial together. That makes it a hygiene verification tool, used to check that a surface has been cleaned before production restarts, not a method for counting bacteria or detecting pathogens. Pass and fail limits are set by each plant for each site, and residues of some sanitisers can reduce the light signal.
Immunoassays use antibodies to detect antigens of a target organism or toxin. In a sandwich enzyme-linked immunosorbent assay (ELISA), a capture antibody bound to a surface holds the antigen, and a second, enzyme-labelled antibody produces a colour from a substrate. Lateral flow devices put the same chemistry on a strip: the sample flows along a membrane, labelled antibodies bind the antigen, and a visible line forms at a capture zone, with a control line to show the test ran. Immunoassays for bacteria typically need about 10^4 to 10^5 cells per mL, so they are used after an enrichment step that multiplies a few cells to detectable numbers. Their results are presumptive, and positives are confirmed by culture. Other rapid tools include chromogenic media, impedance and conductance methods that detect growth from changes in the medium, and MALDI-TOF mass spectrometry for identifying colonies from their protein profiles.
The polymerase chain reaction (PCR) amplifies a specific DNA sequence, such as a virulence or species-specific gene, through repeated cycles of denaturation, primer annealing and extension. Each cycle can double the target, so millions of copies are made in an hour or two. In food testing, PCR usually follows an enrichment, both to raise numbers above the detection limit and to dilute inhibitors, and to favour live cells.
Quantitative or real-time PCR (qPCR) measures fluorescence at each cycle. The threshold cycle (Ct) is the cycle at which fluorescence crosses a set threshold; the more starting target, the lower the Ct. A ten-fold dilution series gives a standard curve of Ct against log quantity, and the amplification efficiency is E = 10^(−1/slope) − 1, so a slope of −3.32 means 100 per cent efficiency, with each ten-fold change in target shifting Ct by about 3.32 cycles. Hydrolysis probes add specificity, and multiplex assays detect several targets in one tube, for instance the stx1, stx2 and eae genes of STEC. Reverse-transcription qPCR detects RNA viruses such as norovirus.
The main limitation is that PCR detects DNA from dead as well as live cells. A heat-treated food may give a positive result without any viable pathogen. Viability PCR adds a dye such as propidium monoazide that enters only cells with damaged membranes and blocks amplification of their DNA, reducing but not eliminating this problem. Every run needs negative, positive and internal amplification controls; a failed internal control means the result is invalid, not negative.
Whole-genome sequencing (WGS) reads the entire genome of an isolate. For food microbiology it has replaced several older typing methods at once, because a single sequence predicts species, serotype, virulence genes and antimicrobial resistance genes, and gives the highest possible resolution for comparing isolates. Public-health laboratories in many countries now use WGS for surveillance of Salmonella, Listeria, STEC and Campylobacter; in the United States the PulseNet network moved from pulsed-field gel electrophoresis to WGS.
Isolates are compared by single nucleotide polymorphism (SNP) analysis against a reference, or by core-genome multilocus sequence typing (cgMLST), which compares alleles at hundreds to thousands of genes. Closely related isolates, differing by only a few SNPs or alleles, suggest a common source, but no single cut-off defines an outbreak: interpretation must combine the genomic distance with epidemiological evidence, food traceback and the organism's known rate of diversification.
Predictive microbiology uses mathematical models to describe how microbial numbers change with time and conditions, so that shelf life and safety can be estimated without testing every combination of formulation and storage. Models are classed as primary, secondary and tertiary.
A primary model describes the change in numbers with time under constant conditions. For growth, it gives a lag time, a maximum specific growth rate (µmax) and a maximum population density. The modified Gompertz equation, the logistic equation, the Baranyi and Roberts model and the three-phase linear model are standard examples. For inactivation, the simplest primary model is log-linear, defined by the D value. Curved survivor curves are fitted by models such as the Weibull model, in which log(N/N0) = −(t/δ)^p; δ is the time to the first decimal reduction and p describes the shape, with p below 1 giving tailing and p above 1 giving a shoulder.
A secondary model describes how the parameters of the primary model change with environmental factors such as temperature, pH and aw. The square-root model of Ratkowsky and colleagues states that √µmax = b(T − Tmin) at suboptimal temperatures, where Tmin is a notional minimum temperature, and it has been extended to pH and aw. Cardinal parameter models use the minimum, optimum and maximum values of each factor, and the gamma concept multiplies the separate effects of each factor. Arrhenius-type models and polynomial response-surface models are also used. Probability or growth/no-growth models predict whether growth can occur at all under a given combination of factors. Tertiary models are software tools that combine the two.
Models are used to estimate shelf life, to judge the effect of a temperature abuse, to design formulations for hurdle-stable foods, to support HACCP critical limits, and as inputs to quantitative microbial risk assessment.
Every model is only as good as the data behind it and the range over which it was built. Most models are developed in laboratory broth, and real foods add structure, competing flora and variation between strains, so predictions in broth are often faster than growth in food. Models should not be extrapolated beyond the range of factors used to build them. Their performance is checked against independent data in the food: the bias factor, Bf = 10^(mean of log(predicted/observed)), shows whether the model is on average fail-safe (Bf above 1 for growth rate) or fail-dangerous; the accuracy factor, Af = 10^(mean of |log(predicted/observed)|), shows the average spread. Challenge tests, in which a food is inoculated with the pathogen and stored under realistic conditions, remain the reference when a model is uncertain.
A biofilm is a community of microorganisms attached to a surface and embedded in a matrix of extracellular polymeric substances that they produce: polysaccharides, proteins, lipids and extracellular DNA. Formation passes through conditioning of the surface by organic molecules, reversible then irreversible attachment, growth into microcolonies, maturation into a three-dimensional structure with water channels, and dispersal of cells that colonise new sites. Many bacteria coordinate this through quorum sensing, cell-density signalling by molecules such as acyl-homoserine lactones in Gram-negative bacteria and oligopeptides in Gram-positive bacteria.
Cells in a biofilm are far more resistant to sanitisers, heat and drying than the same cells in suspension, because the matrix limits penetration and binds active chemicals, because cells deep in the biofilm grow slowly, and because persister cells survive. Biofilms form in drains, floors, conveyor belts, gaskets, valves, dead legs, heat exchangers, filling heads and membranes. Important organisms include L. monocytogenes, which persists in plants for years; Salmonella in dry processing environments; and thermophilic bacilli such as Anoxybacillus and Geobacillus in milk powder evaporators, which shed spores into the product.
Control starts with hygienic design of equipment and buildings, following guidelines such as those of EHEDG and the 3-A standards: smooth, cleanable surfaces, no crevices or dead legs, and free drainage. Cleaning must remove the matrix before sanitising; the effect depends on time, mechanical action, chemistry and temperature. Clean-in-place cycles commonly run a pre-rinse, an alkaline wash to remove protein and fat, an intermediate rinse, an acid wash to remove mineral deposits, a final rinse and a sanitising step, with concentrations and temperatures validated for each plant. Sanitisers include chlorine compounds, peracetic acid and quaternary ammonium compounds. Environmental monitoring programmes sample zones graded by distance from the product, from zone 1, the food-contact surfaces, outwards, and hunt the source of each positive.
A microbiological criterion defines whether a food, a batch or a process is acceptable, based on the presence, absence or number of microorganisms, or the quantity of their toxins, per unit of food. The Codex principles and guidelines for microbiological criteria (CXG 21-1997) require a criterion to state its purpose, the food to which it applies, the microorganism or toxin, the analytical method, the sampling plan, the microbiological limits, the point in the food chain where it applies, and the action to be taken when it is not met.
Food safety criteria define acceptability of a product placed on the market; failure leads to withdrawal or recall. Process hygiene criteria indicate whether a production process is working as intended; failure calls for corrective action in the process. In the European Union, Regulation (EC) No 2073/2005 sets both kinds: for example, L. monocytogenes in ready-to-eat foods able to support its growth must not exceed 100 CFU/g during the shelf life, or must be absent in 25 g before the food leaves the producer's control if the producer cannot show that the limit will be met through the shelf life. The United States treats L. monocytogenes in ready-to-eat foods more strictly, generally as an adulterant when detected. In India, FSSAI publishes microbiological standards for food categories. Limits, food categories and methods differ between jurisdictions and are revised, so the current text must be checked.
Criteria are linked to public health through risk-based concepts. A food safety objective (FSO) is the maximum frequency or concentration of a hazard in a food at the time of consumption that provides the appropriate level of protection; a performance objective (PO) is the corresponding limit at an earlier step; and a performance criterion is the effect a control step must achieve, such as a 5-log reduction.
The International Commission on Microbiological Specifications for Foods (ICMSF) attributes plans are the standard framework. In a two-class plan, n sample units are taken from the lot and each is classed acceptable or defective against a limit m; the lot is accepted if no more than c units are defective. For pathogens such as Salmonella the plan is usually presence/absence with c = 0, so a single positive unit rejects the lot. In a three-class plan there are two limits: m separates good from marginal quality and M separates marginal from unacceptable. The lot is rejected if any unit exceeds M, or if more than c units fall between m and M. With n = 5, c = 2, m = 10^2 and M = 10^3 CFU/g, results of 50, 300, 800, 90 and 400 CFU/g give three marginal units and no unacceptable unit; three is more than c, so the lot is rejected.
The probability of accepting a lot depends on the true proportion of defective units, p, and on n and c. For a two-class plan with c = 0, it is Pa = (1 − p)^n. With n = 5 and 10 per cent of units defective, Pa = 0.9^5 = 0.59: such a lot will be accepted more than half the time. With n = 60, the same lot is accepted with probability 0.9^60, about 0.002. The plot of Pa against p is the operating characteristic curve. The arithmetic shows why sampling cannot prove a lot safe: low-level contamination needs very large numbers of samples to detect, and contamination is rarely spread evenly. ICMSF links the stringency of the plan to the hazard and to whether conditions after sampling will reduce, not change or increase it, through fifteen cases; for severe hazards in foods for vulnerable groups, it recommends large n with c = 0.
A student uses this subject to master the arithmetic and the identification facts that recur in GATE, ICAR and university papers: generations and generation time from two counts, CFU from a dilution plate, MPN read from a table, qPCR efficiency from a slope, and acceptance probability for a two-class plan. In the laboratory the same student prepares serial dilutions, pour and spread plates, runs the coliform MPN sequence and a Salmonella enrichment, and learns the colony appearance of each pathogen on its selective agar. Knowing which organism causes each named spoilage defect, and which toxins are heat-stable, answers most short questions.
Five years into a quality role in a dairy, meat, bakery or ready-meal plant, a technologist writes and runs the environmental monitoring programme, reads ATP results after cleaning, investigates a Listeria species positive in a drain, decides when to hold product, and interprets supplier certificates and laboratory reports. Shelf-life studies, challenge tests and predictive model outputs land on the same desk, as does the question of whether a spoiled return is a leaker, a temperature abuse or a formulation failure. Microbiology also drives the plant's HACCP hazard analysis and its corrective actions.
A senior technical manager or regulator with twenty-five years' experience sets microbiological specifications for products and suppliers, chooses sampling plans and the response to failures, signs off validation of new formulations and processes, and leads the response to a positive pathogen result or an outbreak link: trace, hold, recall, root-cause analysis and communication with authorities. Such a person judges when a rapid method is fit for release decisions, how to use WGS evidence from public-health agencies, and how to balance shelf life, safety and cost across a product range and across export markets with different criteria.
A researcher at this level works on the open problems: predictive models that hold in real food matrices and fluctuating temperatures, strain variability and stress responses of pathogens, persistence and biofilm ecology of Listeria in plants, culture-independent detection that reports viability, metagenomic surveillance of food chains, the significance of genomic clusters, quantitative microbial risk assessment that links criteria to public-health outcomes, and control of emerging hazards such as viruses on produce and spore-formers in new minimally processed foods. Such work is judged by whether its methods are validated and its uncertainty is stated.
This manual does not give MPN tables, full national microbiological standards (FSSAI, EU, US, Codex) by food category, or complete method protocols; questions supply table values where needed, and limits must be checked in the current regulation. Growth limits (minimum aw, pH and temperature) are standard textbook values that vary with strain, food and the other factors present. Prions, antimicrobial resistance in the food chain, probiotics, fermentation and starter cultures, and quantitative microbial risk assessment in detail are covered elsewhere or only mentioned. Recent taxonomic renaming of several spoilage organisms is noted only where the old name is still common in examinations.
Original practice questions written to the syllabus and pattern of the exams named in each question's examTags. They are not past papers and do not reproduce any board's questions.
These notes are free to read here, with no account.
What the Laisha app adds:
Free for ten days, then a subscription. We would rather write that here than have you find it later.