Methodologies for the characterization of microbes in industrial environments: a review.

Abstract

There is growing interest in research and development to develop novel tools to study, detect, and characterize microbes and their communities in industrial environments. However, knowledge about their validity in practical industrial use is still scarce. This review describes the advantages and limitations of traditional and molecular methods used for biofilm and/or planktonic cell studies, especially those performed with Listeria monocytogenes, Bacillus cereus, and/or Clostridium perfringens. In addition, the review addresses the importance of isolating the microorganisms from the industrial environment and the possibilities and future prospects for exploiting the described methods in the industrial environment.

Full text

Page 1

REVIEW PAPER
Methodologies for the characterization of microbes
in industrial environments: a review
Received: 1 November 2002/ Accepted: 2 April 2003/Published online: 23 May 2003
? Society for Industrial Microbiology 2003
Abstract There is growing interest in research and
development to develop novel tools to study, detect, and
characterize microbes and their communities in indus-
trial environments. However, knowledge about their
validity in practical industrial use is still scarce. This
review describes the advantages and limitations of tra-
ditional and molecular methods used for biofilm and/or
planktonic cell studies, especially those performed with
Listeria monocytogenes, Bacillus cereus, and/or Clos-
tridium perfringens. In addition, the review addresses the
importance of isolating the microorganisms from the
industrial environment and the possibilities and future
prospects for exploiting the described methods in the
industrial environment.
Keywords Biofilm Æ Culture Æ Molecular techniques Æ
Fingerprinting
Introduction
Microorganisms inhabiting the food and processing
industries are mostly benign, but some can be harmful to
the processing and safety of the product. Therefore, the
control of harmful microorganisms is essential. Indus-
trial processes that deal with any biological material
provide nutrients and conditions for microorganisms to
grow, either in the shelter of sessile biofilms on surfaces
or as planktonic cells in the circulating process waters.
Moreover, in most natural and industrial systems where
the supply of nutrients is sufficient, microorganisms
grow as spatially organized, matrix-enclosed, multispe-
cies communities in biofilms. Besides a solid surface, the
microbes need only water to initiate biofilm formation
[72, 73, 270]. Microbial biofilms and biofouling of sur-
faces and interfaces within the industrial environment
are major problems. In industry, the first step is identi-
fying the problem of biofilms and biofouling in a par-
ticular process or site. Subsequently, it is important to
determine the best possible methods for detection of
biofilms in situ, so that they can be characterized and
possibly further studied in the laboratory. Finally, this
information can be used to define strategies for con-
trolling biofilm formation in that specific environment
[208].
Microorganisms in food and in industrial environ-
ments are distributed unevenly; and there is a great
variation in the cell density and composition of micro-
bial population over space and time. Typically, the
microbial cells are located in the surfaces of the food
matrix and process equipment; and the cell density and
species distribution may vary in different parts of a food
product [77, 156, 158]. Changes in the ecosystem cause
continuous qualitative and quantitative variation in the
composition of the microbial community over time
[152]. Both intrinsic (e.g. chemical composition, natural
microbiota) and extrinsic factors (e.g. processing, stor-
age conditions) affect microbial growth [272] and
consequently the composition of the microbial commu-
nity. All these factors affect the actual sampling of the
industrial environment, making it a demanding task to
perform.
Published data on microbial detection and charac-
terization from industrial environments is abundant and
therefore the following review is restricted to the most
relevant and widely applied techniques and to just a few
specific bacterial species important in the food and
process industrial environments, namely Listeria mono-
cytogenes, Bacillus cereus, and Clostridium perfringens.
Both traditional and molecular identification and char-
acterization methods for bacteria and their communities
are discussed, in addition to typing methods for bacterial
isolates obtained from the industrial environment and/
or foodstuff. Future prospects for the exploitation of the
J Ind Microbiol Biotechnol (2003) 30: 327–356
DOI 10.1007/s10295-003-0056-y
Johanna Maukonen Æ Jaana Ma ¨ tto ¨ Æ Gun Wirtanen
Laura Raaska Æ Tiina Mattila-Sandholm Æ Maria Saarela
J. Maukonen Æ J. Ma ¨ tto ¨ Æ G. Wirtanen Æ L. Raaska
T. Mattila-Sandholm Æ M. Saarela (&)
VTT Biotechnology, P.O. Box 1500, 02044 VTT, Finland
E-mail: maria.saarela@vtt.fi
Tel.: +358-9-4564466
Fax: +358-9-4552103

Page 2

described methods in the industrial environment and
industrial samples are also addressed.
Hygiene and safety problems in the industrial
environment
Problems caused by bacterial biofilms
Biofilms protects microbes from hostile environments
and act as a trap for nutrient acquisition [270]. In mul-
tispeciesbiofilms,mixed-species
formed by a part of the sessile population when cells of
metabolically cooperative species are juxtaposed and are
thus in a position to benefit from interspecies substrate-
exchange and/or mutual end-product removal. This level
of structural organization and metabolic specialization
explains the remarkable metabolic efficiency of micro-
bial biofilms and their universal and inherent resistance
to antimicrobial agents [72]. In practice, a biofilm on
improperly cleaned surfaces is a barrier between mi-
crobes and disinfectants, antibiotics, or biocides [54,
228, 298, 458]. Biofilm components can also protect
microbes from the effects of steam: e.g. the bacterial
slime of Bacillus sp. improves the heat resistance of the
bacterium, extending the autoclaving time required for
efficient sterilization to several hours [270]. Besides
causing problems in cleaning and hygiene [184], biofilms
can cause energy losses and blockages in condenser
tubes, cooling fill materials, water and wastewater cir-
cuits, and heat exchangers [64]. Biofilms can occasion-
ally cause health risks by releasing pathogens into
drinking-water distribution systems [44, 453]. In food
processing water-supply systems, biofilms cause prob-
lems in granular activated carbon columns, reverse
osmosis membranes, ion exchange systems, degasifiers,
water storage tanks, and microporous membrane filters
[132, 287]. Commonly found microbes in the food
industry and on food contact surfaces are enterobacte-
ria, lactic acid bacteria, micrococci, streptococci, pseu-
domonads, and bacilli [458]. The formation of resistant
spores that can contaminate process equipment and
food products is a special concern for the food pro-
cessing industry and for the consumer [17].
The level of hygiene in the paper and board industry
is important, since the end-products are often in contact
with foodstuffs. The microbial isolates from the paper
and board industry are mainly bacilli, enterobacteria,
pseudomonads, or actinomycetes, but moulds, yeasts,
anaerobic sulfate-reducing bacteria, and clostridia may
also be detected [168, 333, 392]. The growth of B. cereus,
clostridia, coliforms, and staphylococci in the paper-
making process is detrimental to product hygiene [323,
379]. Aerobic and anaerobic spore-forming bacteria,
such as bacilli and clostridia, which are not killed during
the drying stage of paper-making, are the most impor-
tant microbes from the safety point of view [194, 324,
341, 407]. Slime build-up in paper-processing machines,
caused by microbial biofilms, may cause significant
microcoloniesare
economic losses, mainly due to machinery-running
problems in addition to spots, holes, and quality prob-
lems in the end-product. The machinery slime can also
contain polymers of microbial origin, fibers, and inor-
ganic precipitates. Common bacteria detected and
identifiedfrompaper-processing
include enterobacteria, bacilli, pseudomonads, and
Clavibacter spp. The total number of microbes in the
slime can reach 1012colony-forming units (cfu)/ml.
Pathogens, such as B. cereus, can also be found in these
machinery slimes. Anaerobic bacteria, such as sulfate-
reducing bacteria, can be involved in the initiation and
progress of corrosion [36, 168, 408]. Also, heat-stable
microbial metabolites, mainly enzymes and toxins, can
cause problems if migration occurs from a packaging
material into a foodstuff. Volatile metabolites, such as
the fatty acids produced by many Clostridium spp. and
the hydrogen sulfide produced by sulfate-reducing bac-
teria, can cause organoleptic problems in end-products
[107, 168, 341].
machinery slimes
Prevention of hygiene and safety problems
The elimination of biofilms is a very difficult and
demanding task, because many factors affect the
detachment, such as temperature, time, mechanical
forces, and chemical forces [453]. Harmful microorgan-
isms may enter the manufacturing process and reach the
end-product in several ways, e.g. through raw materials,
air in the manufacturing area, chemicals employed,
process surfaces, or factory personnel [193, 323].
The target of microbial control in a process line is
two-fold: to reduce or limit the number of microbes and
their activity and to prevent and control the formation
of deposits on process equipment. The present most
efficient means for limiting the growth of microbes are
good production hygiene, a rational running of the
process line, and a well designed use of biocides and
disinfectants. Novel means to control slime formation
are constantly sought, e.g. through the control of envi-
ronmental factors on the process line and the use of
surface-active agents, (bio)-dispersants, enzymes, and
new biocidal chemicals, in addition to non- or minimally
toxic chemicals [36, 108, 215]. The cleanliness of sur-
faces, the training of personnel and good manufacturing
and design practices are important in combating hygiene
problems in the food industry [179]. Disinfection after
the removal of biofilms, using suitable cleaning proce-
dures, is also required in food plants where wet surfaces
provide favorable conditions for microbial growth [116,
286].
In the food industry, equipment design and choice of
surface materials are important in fighting microbial
biofilm formation [179, 454]. The most practical material
in processing equipment is steel, which can be treated
with mechanical grinding, brushing, lapping, and elec-
trolytic or mechanical polishing. Dead ends, corners,
cracks,crevices, gaskets,valves,and jointsare
328

Page 3

vulnerable points for biofilm accumulation [67, 115,
322]. Poorly designed sampling valves can destroy an
entire process or give rise to incorrect information, due
to biofilm formation at measuring points. Valves are
vulnerable to microbial growth and thus constitute a
hygiene risk [67]. Also, hoses, tubes, filters, etc. con-
taining polyvinylchloride increase the risk of contami-
nation,duetothismaterial:
contaminated and it deteriorates more easily than steel
[331]. Problems with the accumulation of particulates
and cells occur whenever cleaning is inappropriate for
any reason [278]. Inadequate cleaning and sanitation of
surfaces coated with biofilms presents a source of con-
tamination within the process [458].
Achieving a clean food plant must be the aim of the
plant managers, who have to invest the necessary time
and money to accomplish it. Monitoring methods and
cleaning procedures, including the program, cleaning
agents, disinfectants, and cleaning equipment, must be
carefully planned [454]. Cleaning in the process industry
should be based on systematic planning. The knowledge
that microbes grow differently on surfaces, compared
with suspensions, is the first step in developing advanced
regimes in process hygiene [178]. Biofilm formation in
industrial systems reflects a disturbance in the process
[270]. Biofilms are less likely to accumulate in well
designed systems, which are effectively cleaned. Results
indicate that low-pressure cleaning in itself is not effective
enough to remove biofilms unless the cleaning agent is
effective[457].Theefficiencyofcleaningagentsisassessed
by their ability to remove biofilms from process surfaces
together with their ability to kill the bacteria present in
the biofilm [457]. The cleaning effect in open systems can
beenhancedusingdouble-foamingorthroughscrubbing.
In closed processes, the removal of biofilms from surfaces
can be performed using efficient flow conditions in com-
bination with effective cleaning agents [457]. Strong oxi-
dizing and/or disinfective agents are used to combat
microbial deposits on equipment surfaces in problem
areas. Satisfactory elimination of biofilms using only
disinfectant treatment cannot be achieved, even if the
agent is very effective against freely suspended cells [286].
Sources, problems, and control of microbial contami-
nants in industrial processes are presented in Fig. 1.
it ismore easily
Monitoring hygiene and safety problems
Monitoring practices based on sampling of the liquid
phase do not reflect the location or extent of microbes
growing in biofilms on surfaces [69]. The methods used
for monitoring process hygiene are often based on
conventional cultivation, using various types of agar
plates or adenosine triphosphate (ATP) measurement.
Conventional cultivation requires several days before
the result can be obtained and it enumerates cells able to
form colonies on the given agar [457]. The measurement
of ATP is an often used method for detecting biological
growth [4, 147], e.g. for the measurement of total
hygiene [457]. The detection limit of ATP measurement
for bacteria is 103–104cfu/ml [39].
The detection of deposit build-up on equipment sur-
faces at an early stage enables effective countermeasures
and thus results in an improvement in the process hy-
giene. Successful on-line monitoring of microbiological
deterioration in the process industry has great beneficial
impact, of both economic and environmental value. On-
line monitoring saves both expense and the environment
when gentle cleaning methods can be used and unnec-
essary procedures avoided.
A reliable identification of industrial microbial iso-
lates is often difficult to obtain. Over the past decade,
many improvements have been seen in both conventional
and modern methods for the detection and identification
of microorganisms from the industrial environment.
Phenotypic analyses (e.g. the fatty acid methyl ester test,
or sodium dodecyl sulfate–polyacrylamide gel electro-
phoresis) have traditionally played an important role in
microbial identification and classification. Genotypic
analyses (e.g. partial 16S rDNA sequencing, ribotyping)
have proved very useful and accurate in the identification
and classification of industrial microbial isolates, since
the physiological properties of the industrial microbes
may be different from those of the reference strains.
Industrialstrains are usually well adapted to their specific
environments and do not often possess the typical char-
acteristics of any species hitherto described. Thus, the
effective use of molecular methods requires the devel-
opment of extensive identification libraries. Further-
more, in many cases, the results of phenotypic and
genotypic tests are not in good agreement, which further
hampers identification [333, 413].
L. monocytogenes, B. cereus, and C. perfringens
in the industrial environment
L. monocytogenes, B. cereus, and C. perfringens are
importantpathogensinindustrial environments,
Fig. 1 Sources, problems, and control of microbial contaminants
in industrial processes
329

Page 4

especially due to their ability to endure adverse/harmful
process conditions.
L. monocytogenes is a significant food-borne patho-
gen and may cause epidemic and sporadic outbreaks.
The infective dose of L. monocytogenes is related both to
the level of contamination of the food product and to
the host susceptibility. L. monocytogenes is able to grow
across a wide range of temperatures (including very low
temperatures) and pHs; and it is extremely salt-tolerant.
Due to this good tolerance to environmental stress-fac-
tors, L. monocytogenes is difficult to remove from
the factory environment once it has become a part of
the house microbiota. The typical food vehicles for
L. monocytogenes are dairy, meat, and fish products [for
reviews, see 119, 251].
B. cereus is widely distributed in nature (soil contains
105–106spores/g) and is extremely tolerant to different
environmental stresses. B. cereus is a non-competitive
bacterium. However, food processes can select for it,
since pasteurization is insufficient to kill the spores and
many of the strains are psychrotrophic [18]. Spores of
B. cereus are very hydrophobic [196] and adhere tightly
to surfaces. Vegetative cells, especially cells in the late
stationary growth phase, are also hydrophobic [318].
B. cereus strains may produce emetics and/or entero-
toxins, which leads to food poisoning when a toxin-
producing strain is present at levels >105cells/g [200,
417]. But strains causing food poisoning at lower cell
levels (103–104cells/g) have also been found [18].
C. perfringens is mainly restricted to meat products,
since this bacterium is unable to synthesize several
amino acids. C. perfringens spores survive insufficient
heating of the food product, vegetative cells reproduce at
temperatures between 10 ?C and 47 ?C, and the gener-
ation time in optimal growth conditions is short [18].
Wild strains of C. perfringens are mainly enterotoxin-
negative. However, enterotoxin-positive C. perfringens
strains may cause food poisoning, especially through
cooked food [283]. Aerotolerant vegetative cells survive
for some time under aerobic conditions, but do not
multiply.
Characteristics of L. monocytogenes, B. cereus, and
C. perfringens are presented in Table 1.
Isolation of microorganisms from the industrial
environment
Sample collection and processing
Due to the spatial and temporal heterogeneity and
technical problems related to sampling in the industrial
environment, obtaining a representative sample from
certain foods and food-related industry is a demanding
task. Microbes are often tightly attached to surfaces and
the process equipment may contain parts that are hard
to access, such as dead-ends or bends in pipework.
Different methods, such as swabbing, rinsing, agar-
flooding, andcontactagar
employed for sampling in the industrial environment
[141, 333, 352, 456, 457]. The conditions during sample
transportation have a great impact on sample quality;
and the time between sampling and processing should be
limited to the minimum.
methods,havebeen
Table 1 Typical characteristics of Listeria monocytogenes, Bacillus cereus, and Clostridium perfringens. cfu Colony-forming units, ND not
detected
CharacteristicL. monocytogenes B. cereus C. perfringens
Minimum water activity
(aw; for growth)
Growth temperature
pH (for growth)
Heat resistance of spores
(at 100 ?C)
Number of bacteria per
gram in food
reported in infective cases
Incubation period
0.920.94 0.93
)1.5 ?C to 50 ?C
pH 4.1–9.4
–
4–55 ?C
pH 4.3–9.3
3–8 min
12–50 ?C
pH 5.5–9.0
0.3–13.0 min
<10–104cfu/g (high risk groups)
105–109cfu/g (normal population)
105–107cfu/g (diarrheal type)
105–108cfu/g (emetic type)
106–107cfu/g
18–20 h (diarrheal type), in other types
even 2 months
Diarrhea, fever (healthy adults)Sepsis,
meningitis, fecal infection (adult high
risk groups)Meningitis, sepsis,
pneumonia, fecal infections (new-borns)
Fever, preterm delivery, stillbirth
(during pregnancy)
Ready-to-eat foodstuffs with long
shelf-life (fish, meat products, soft
cheeses) and vacuum-packed products
8–16 h (diarrheal type)0.5–5.0 h
(emetic type)
Abdominal pain, nausea,
vomiting (emetic type)
Abdominal pain, diarrhea,
nausea (diarrheal type)
8–24 h
Symptoms
Abdominal pain,
nausea, acute diarrhea
Food vehiclesMeat products, soups, vegetables,
pudding, milk, and milk products
(diarrheal type) Rice, pasta
and noodles (emetic type)
Incompletely cooked
or slowly cooled food
products, meat, poultry,
shellfish, fish, and dairy
products
ND
ND
Occurrence in food biofilms
Occurrence in environmental
site biofilms
Yes (dairy, fish, meat)
Yes (plentiful)
ND (dairy)
Yes (non-food)
330

Page 5

Recovery of microbial cells from the sample matrix is
a critical step and may lead to biases in the qualitative
and quantitative estimation of the microbial community
[for a review, see 131]. In most cases, samples need to be
macerated and homogenized to liberate the microbial
cells from the sample matrix. The cells in natural sam-
ples, e.g. naturally contaminated food, can be tightly
attached to the matrix [62] and may need vigorous
processing. Maceration of a food matrix may change the
chemical environment of the sample and release sub-
stances that are toxic or inhibitory to some microbes.
Dilution is also a potential bias-causing step and needs
standardization [131]. Several methods have been
developed to concentrate the sample, either to decrease
the detection limit of the target microbes or to overcome
the inhibitory effect of the matrix on the detection
method, e.g. polymerase chain reaction (PCR). Selective
or non-selective enrichment cultures are widely used for
amplifying populations of food pathogens before
detection by cultivation or molecular techniques [62].
However, the enrichment step is not always optimal for
the recovery of the target species and may lead to false-
negative results, especially when a too-selective enrich-
ment medium is used for a sample containing injured
cells [131]. Moreover, enrichment precludes attempts to
quantitate results and, due to differences in growth rate
between different populations, enrichment may lead to a
bias in the recovery of different species [105]. Immuno-
magnetic separation (IMS) is a technique in which
magnetic particles coated with specific antibodies are
used to capture the target cells [312]. IMS has been used
to concentrate food pathogens from sample homogen-
ates or enrichment cultures, followed by detection with
cultivation, immunological, or molecular methods [62,
102, 188, 312]. In addition, centrifugation and filtration
techniques are commonly used in concentrating cells
from certain types of sample.
Microbial cells are exposed to several environmental
stresses during food processing and storage, which may
change the physiological status of the cells. In addition
to culturable and metabolically active cells and autoly-
sing dead cells, microbial cells in many other physio-
logical states can be found in samples [222]. In several
food matrices, the dominant cells are those in a sta-
tionary growth phase, which are still metabolically ac-
tive [131]. Adverse conditions, such as nutrient depletion
and low temperature, can lead to viable but non-cul-
turable cells (VBNC), which do not produce colonies on
media that normally support their growth. However,
VBNC cells remain metabolically active and infective
[222, 273]. Two different types of cells contribute to the
silent but active majority: (1) known species for which
the applied cultivation conditions are just not suitable or
which have entered a non-culturable state and (2) un-
known species that have never been cultured before, due
to a lack of suitable methods [16]. Sub-lethally injured
cells, which do not grow on selective media but grow on
non-selective media, may be present in processed sam-
ples and processed foods [152, 202]. In addition, cells in
certain microbial groups have the ability to form spores,
which are extremely resistant dormant states [222].
VBNC and injured cells may resuscitate or recover
under appropriate conditions [58, 391]. The recovery of
injured cells is highly dependent on the chemical com-
position of the enrichment medium and on the degree of
the injury and the presence of accompanying microbes
[244, 374, 391]. Besides culture, VBNC and injured cells
can be more easily detected by fluorescent staining or
molecular techniques [for reviews, see 152, 223, 274],
which do not rely on the viability of the target cells.
The sample-processing method is dependent on the
properties of the target microbial groups and on the
method used in the subsequent detection step. If detec-
tion is performed by culture-based methods, the target
strains must retain viability and culturability during
sample processing, whereas the inhibitory components
of the sample matrix play an important role in PCR-
based detection. The sample matrix studied plays an
important role when deciding which method to use for
microbial detection and identification.
Cultivation
An effective cultivation procedure for the detection of
food pathogens should suppress competitive microor-
ganisms to the extent that the diagnostic system allows
easy and reliable detection of the target genera/species.
Several selective culture-based techniques are available
for the detection and enumeration of L. monocytogenes,
B. cereus, and C. perfringens in environmental and food
samples; and the international standard methods for the
detection and enumeration of these food pathogens are
based on cultivation [77, 417]. The cultivation and sub-
sequent identification of isolates using conventional
techniques are time-consuming. It may take more than
one week to obtain the complete results [77, 338]. During
the past decade, much effort has been put into the
development of more rapid culture techniques, many of
which are based on the use of fluorogenic and/or chro-
mogenic culture media.
Microbial community analysis by cultivation
An ideal method for studying microbial communities
would detect and enumerate all microbial species present
in the samples with equal efficiency. It was speculated
that many microbial communities are too diverse to be
counted exhaustively, which led to the application of
statistical approaches for the estimation of diversity
[192]. In food microbiology, especially in food hygiene
surveys, cultivation has usually been aimed at the
detection of selected groups/species of microorganisms,
rather than the assessment of the complexity and
dynamics of the microbial community. Culture-inde-
pendent, DNA-based methods have also had limited
applications in theinvestigation ofmicrobial
331