CHAPTER ONE

INTRODUCTION

1.1 Background of the study

Yogurt is one of the most popular termented dairy product widerly
consumed all over the world. It is obtained by lactic acid termentation of
milk by the action of a starter culture containing streptococcus thermophilus
and Lactobacillus delbruckii spp. bulgaricus. The role of these two general
in yogurt manufacture can be summarized as milk acidification and
synthesis of aromatic compound (Serra 2009). In Nigeria, It is a popular
drink due to its nutritional probiotic and organoleptic characteristics but
yogurt can be contaminated by fungi able to modify organoleptic
characteristics and be a risk of human health yogurt can be easily
subjected to microbial contamination as especially by fungi which grow and
reproduce and environment with feasible oxygen. Some of the molds can
produce secondary toxic metabolites resisting to common reclamation
treatments. Aflatoxins are carcinogenic and toxic, which is a secondary
metabolic product of some Aspergillums spp. (Issazadeh 2012) worked on
60 yogurt samples and out of these 60 sample, 60 sample, 59 sample were
found to be contaminated with AFML. (Ahmad 2013) worked on quality
assessment of yogurt produced at industrial and evaluation of yogurt
products on Qena city, Egypt. They showed that out of 100 random
samples purchased from various dairy shops, street vendors and
supermarkets located in 72% and 35% of small and large scale yogurt
samples though bottled yogurt is a popular drink in Kaduna metropolis,
Kaduna, Nigeria, very little research has done on micro-biological.
The other ten samples were collected without any registration no.
From HAFDAC and the labeling on these sample did not have any
manufacturing date and expiration date. For non registered samples were
collected and labeled as a1-a5. For each brand, 2 sample were collected
and labeled accordingly. After collection, the samples were put into a cooler
packaged with ice and then were small scale and were commercially
available in Faisalabad, Pakistan. They observed that the conform count of
1
yogurt produced at small scale was much higher than that of branded
samples produce at large scale (Montagna 1998) studied on
microbiological quality of 166 samples and observed that 7.2% of the
contaminated samples were observed and positive for fungi. (El-mast
2012) worked on microbiological quality of yogurt of different brand
approved by NAFDAC and also of some yogurt not approved by NAFDAC
is a governmental organization which is in charge of maintaining the quality
of food products made in transported to the Microbiological Laboratory of
Kaduna State University within 3h for analysis purpose.

Yogurt has been considered to be one of the most consumed

effective probiotics throughout the world. Considerable number of live
microbes given through food which enhances human health are said to be
probiotic. These beneficial microbes often improvise lactose digestion and
inhibit lactose intolerance. The study done by (Francisco 2015) infers that
mostly the strain of lactobacillus bulgaricus in yogurt production. Also, the
consumption of yogurt can be because and fungi other than the species
lactobacillus bulgaricus and streptococcus thermophilus. Contamination of
yogurt occurs in various stages which include contamination during the
production process or change in weather from cold to warm temperatures
or duration from manufacturing to selling and insufficiently refrigeration.
The contamination level in both locally and industrially made yogurt was
studied using various protocols. Most studies suggest that contamination
occurs due to unhygienic practices during the production level. Inter of
contaminated yogurts not only causes food poisoning but also services as
the most suitable medium for almost a wide range of microbes like yeasts
moulds gram negative psychrophiles conform lactic acid bacteria and so
on. Since the yogurt is eaten as such without heating or boiling the
changes get of getting ill are at high rate.

The market for sustainable and healthy plant-based animal protein

alternatives is growing rapidly and is expected to increase further.
Consumers’ motivation for choosing vegetarian and vegan products can
include health-related reasons (such as animal protein allergies and lactose
intolerance) and positive impacts on the environment and ani- mal welfare
2
(Aschemann-Witzel 2020; De Boer 2017; Nevalainen 2023). As a result,
this leads to the emergence of new dairy and meat alternatives. The plant-
based formulations of dairy and meat substitutes can include a wide array
of different ingredients with variations in chemical, nutritional, and
microbiological composition. Next to soy, which is still extensively used to
replace animal protein, the selection and availability of more sustainable
plant-protein sources are also increasing (EUVEPRO, 2019; Lima and
Mellinger, 2022). This in- cludes pulses (pea, faba bean, chickpea, mung
bean), cereals and pseu- docereals (rice, oat, wheat, quinoa), nuts
(almond, cashew), and oilseeds (canola and sunflower) (Eeig, 2018).
These plant sources are not new to the food industry, but the way in which
they are processed and applied can be. For example, almonds, cashews,
pea and oats are now widely. applied in plant-based analogues of milk,
yogurt, and cheese (Cardello 2022, Moss 2022, Tangyu 2019). However,
the reformulation of traditional products with new plant ingredients or
alterations of processing conditions for the known ingredients requires
careful consideration with respect to the microbiological safety and stability
of the developed recipes.

Similar to milk-based products, the manufacturing of plant-based

dairy alternatives often includes high-temperature treatment (at least
pasteurization) intended to eliminate foodborne pathogens and spoilage
microorganisms. Spore-forming bacteria such as Bacillus and Clostridium
are particularly difficult to remove from ingredients and food manufacturing
facilities. The ability to form robust endospores allows Bacillus and
Clostridium species to survive a variety of hostile conditions and occupy
many diverse environments (Wells-Bennik 2016). Moreover, many Bacillus
species are ubiquitous in soil and can serve as plant growth-promoting
bacteria, contributing to soil functionality and fertility (Bashan and Holguin,
1998, Khan 2018, Saxena 2020). As a result, Bacillus species are often
detected in many types of foods, especially those of plant origin. This
includes members of the Bacillus subtilis group and the Bacillus cereus
group, some of which are known to produce toxins that are harmful to
humans.

3
 B. subtilis group members are typically non-pathogenic and include
Bacillus licheniformis, Bacillus pumilus, Bacillus amyloliquefaciens, Bacillus
mojavensis, Bacillus vallismortis, Bacillus sonorensis, Bacillus atrophaeus
and B. subtilis itself with subspecies subtilis and spizizenii (Fritze, 2004;
Logan and De Von, 2009). These species typically have optimum growth
temperatures between 30 "C and 50 °C, with B. licheniformis reported to
grow at temperatures as high as 58 °C (Warth, 1978). Species such as B.
subtilis, B. amyloliquefaciens, and B. licheniformis are commonly found in a
variety of food ingredients and foods, including cocoa, herbs, spices, bread,
soup, milk, and milk powder (Lima 2011, Lucking 2013, Miller 2015, Oomes
2007). The bacteria from this group are often associated with food
spoilage, however, cases of B. licheniformis producing lichenysin toxin at
cytotoxic levels were reported by several authors (Madslien 2013, Mikkola
2000; Yeak 2022).

The B. cereus group (also known as B. cereus sensu lato group) is a

species complex that contains numerous closely related species, including
Bacillus weihenstephanensis, Bacillus cytotoxicus, Bacillus mycoides,
Bacillus pseudomycoides, Bacillus wiedmannil, Bacillus toyo nensis,
Bacillus anthracis, Bacillus thuringiensis, and Bacillus cereus sensu stricto
(Fritze, 2004, Liu 2015). In addition, nine new species were added to the B.
cereus group, which can only be distinguished by the Bacillus proteolyticus,
Bacillus nitratireducens and Bacillus parumycoides whole-genome
sequencing. These are Bacillus paranthracis, Bacillus pacificus, Bacillus
tropicus, Bacillus albus, Bacillus mobilis, Bacillus luti, (Liu 2017). The
nomenclature of species in the group has been subject to various changes
over the years (reviewed by Carrull 2022).The species in this group are
known to be pathogenic and vary in their ability to cause illness in humans
and animals. Dozens of foodborne outbreaks are reported every year due
to B. cereus and its toxins (EFSA, 2021). Depending on the toxin gene(s)
carried by a B. cereus strain, it can cause emetic or diarrheal disease
(Dietrich 2021). Strains with the ces gene cluster can produce the heat-
stable toxin cereulide in food, causing the emetic syndrome when the
product is ingested. If strains contain genes encoding enterotoxins, such as

4
hemolysin BL. (hbl), non-hemolytic enterotoxin (nhe), and cytotoxin K
(cytK), these toxins can be produced in the small intestine causing
diarrhea. Different combi- nations of toxin genes may be present in different
strains (Ehling-Schulz 2006, Guinebretière 2010, Marques Rossi 2018),
Anaerobic spore-forming Clostridium species can spoil a wide range of
foods, including dairy products, meat and poultry products, fresh and
canned fruits and vegetables, and usually produce gas and/or putrid odours
(De Jong, 1989). Although most species associated with food spoilage do
not cause illness, Clostridium perfringens and Clostridium botulinum are
known causes of severe foodborne disease (EFSA, 2021; García 2021;
Peck and van Vliet, 2016).

The presence of spore-forming microorganisms is highly undesirable

in raw food ingredients, as this can compromise both safety and quality of
the final product. To avoid such problems, food manufacturers use various
inactivation and food preservation strategies. However, heat treatment to
remove spores or fermentation to prevent growth of spoilers (as commonly
used in dairy products), may not be suitable for plant-based dairy
alternatives. This is due to the differences in chemical and physical
properties of plant-based ingredients compared to milk (for instance,
differences in protein solubility/precipitation or substrate availability).
Furthermore, there are still many knowledge gaps regarding the microbial
composition of different sources of plant pro- teins, where the initial levels
of microbial contaminants are not consistent and can be influenced by a
wide range of factors, such as geographical location, climate, cultivation
and harvesting conditions of the crops. In addition, plant ingredients can
come from various sources and in different product formats, such as
concentrated liquids/pastes, powders (concentrates and isolates), flours
based on whole seeds, or nuts and grains, each with varying levels of
purity and protein content. A better understanding of microorganisms
present in various ingredients is essential for conducting microbial risk
assessments and managing microbial safety and spoilage risks throughout
the plant-based food production chain.

5
 In this study microbial contaminants were investigated in 88 samples
of different plant-based ingredients with a focus on total viable bacteria and
spore-forming microorganisms, namely, aerobic mesophilic and
thermophilic spore formers, anaerobic mesophilic spore formers, and B.
cereus group species. The identities of the most abundant isolates per
group were determined. Toxin profiles were assessed for the isolated
strains belonging to the B. cereus group. In C. sporogenes isolates
absence of genes encoding the botulinum toxin (BoNT) A and B was
verified, and for isolates of B. licheniformis the presence of the IchAA gene
that encodes lichenysin production was investigated. The results are
discussed in view of the need for a holistic approach across the food chain
to manage animal protein alternatives and prevent microbiological spoilage
and/or safety issues in plant-based animal protein alternative.

Man has been consuming fermented milk since the beginning of

civilization (Mckinley, 2005). Yoghurt, a fermented dairy product popularly
consumed around the world, can be achieved by the fermentation lactic
acid in milk by the activity of a starter culture of Lactobacillus delbruckii
sub-specie bulgaris and Streptococcus thermophiles. The two genera play
important roles in the manufacture of yogurt from milk product through
acidification and synthesis of aromatic compounds (Serra 2009). The lactic
acid fermentation of milk is also a means of prolonging the shelf life of the
nutrients in milk (Hui, 1992; Oyeleke, 2009).

Yogurts are regarded as ready to drink foods commonly taken for

energy production and for health, all over the world, especially in Nigeria
(Alli 2010). It can also be consumed as a drink beverage to quench thirst
(Alfa- Lawal, 1984). This product blend is rich in protein and improves
healthy living (Cueva and Aryana, 2008). It is a balanced food which
contains virtually almost all the nutrients present in milk and in more
absolvable form. There is a wide range of flavors available to spice it
(Anther, 1986, Oyeleke, 2009). Due to its nutritional, organoleptic and
probiotic qualities, it is a popular drink that is in high demand (De 2014).

6
 Industrial processes have improved the quality, storage,
transportation and commercialization of the product. Microbiological
parameters especially, general Coliforms, Escherichia coli and
Enterococcus bacteria counts are commonly used to ascertain these
conditions (Tamine and Robison, 2007). The qualification of acidity and
temperature are also observed in order to evaluate the preservation of
yogurt status; statistics shows a variation from 0.6 to 1.5g of lactic acid per
100g of product and the temperature of preservation in dairy industries and
markets must not be higher than 10°C (Rodrigues 2010).

Yogurt being a product of fermented milk from fresh milk can be

easily contaminated. Yeast and mould are considered as primarily
contaminants in Nigeria (Suriyarachchi and Fleet, 1981, Oyeleke, 2009).
Fungi are capable of growing and reproducing in acidic environment with
appreciable oxygen (De 2014). Some species of the genus Aspergillus
have been implicated in the production of secondary metabolites known as
aflatoxins which are toxic and carcinogenic (Issazadeh 2012).
Contamination by fungi or moulds can result in changing or modifying the
taste, colour, texture and organoleptic characteristics of the yogurt.
Thereby, humans can be at risk taking or drinking it.

Ota is characterized by low level of environmental sanitation, poor

waste disposal and lack of bacteriologically free water (potable). The aim of
this study is to ascertain the microbial load or quality of some commercially
vended yogurts retailed in Ota metropolis.

Yogurts is the fermentation of dairy product that is obtained by lactic

acid fermentation of milk through the activity of a starter culture of two
bacteria, Streptococcus thermophilus and Lactobacillus delbruckii sub-
specie bulgaris. These microorganisms play important roles in the
production of yogurt, from milk product through acidification and synthesis
of aromatic compounds (Serra 2009). This lactic acid fermentation of milk
also serves as a means of prolonging the shelf of the nutrients in milk
(Oyeleke 2009, Hui, 1992).

7
 Yogurts are regarded as ready to drink foods usually consumed as
drink to quench thirst (Alfa- Lawal, 1984). This product is rich in protein and
improves healthy living (Cueva and Aryana, 2008). Yogurt is a balanced
food drink which contains virtually almost all the nutrients present in milk
and in more absolvable form and can be produced from whole or skimmed
milk and there is a wide range of flavors available to spice it (Anther, 1986;
Oyeleke 2009). Due to the increasing demand for yogurt drinks, automated
equipment is required to facilitate industrial production, in other to meet
with both quality and safety quality products (Salinas, 1986).

The improved industrial process has improved the storage,

transportation and commercialization of the product (Tamine and Robinson,
2007). Yogurts like any other milk product are liable to contamination of
bacteria, yeast and moulds, which are primarily source of contaminants in
commercially produced yogurt in Nigeria (Oyeleke, 2009; Suriyarachchi
and Fleet, 1981). Ota is characterized by low level of environmental
sanitation, poor waste disposal and lack of bacteriologically free water
(potable).

In 2000, World Health Organization (WHO) recognized food and drink

safety as an essential public health function. Food safety can be defined as
the opposite of food risk that is "the probabilities of not being adversely
affected from ingesting a particular food or drink" (Henson and Traill, 1993).
Lack of adequate awareness on the origin and severity of foodborne
diseases that made food handling practices less motivated to change. Un-
hygienic food drinks can affect the health of individuals and this can be
prevented by practicing food safety hygiene (Schafer 1993). It's important
to have clear knowledge about prevention of foodborne diseases, though
gain of such knowledge might not lead to total change in people's behavior
and that is why approach is very essential (Henson and Traill, 1993:
Gotsch, 2012). Attitude, knowledge and practice formed by manner with
impressions that comes as regards economic, social, cultural aspect of life
(Scharff, 2010).

8
 Most persons lack knowledge of basic rules pertaining to food
hygiene and consequently, report shows that approach and ability of
consumers generally demographic and socio-economic differs in their
background, such as age, gender and academic status (Sockett, 1995;
Wilcock 2004). Men show more risky approach than women and the
prevalence of risky behavior also increased with increasing soci-economic
status (Altekruse 1999). Research revealed in the United States (U.S.) by
(Unklesbay 1998) showed that students entered into a curriculum which
consist food safety data possesses greater food safety ability contrast to
others. Lifestyle changes has shown considerable influence on user's in the
U.S.; results show that consumers lack adequate knowledge on the
microorganisms causing food poisoning, type of foods associated with
these microorganisms, and the need of avoidance of cross contamination
(Williamson 1992).

There are three important aspects to be considered in the incidence

of food drink pathogens as regards the personnel handling food drink:
Approach, practice and attitude (Scharif and Al-Malki, 2009). "Persons that
prepare and sell food drink are known as Food drink handlers (WHO,
1989). Ability and approach are essential; these indicate differences within
food safety ability and self-reported methods (Woodburn and Raab, 1997).
Results show that most of the outbreaks resulted from improper food drink
handling practices (Ehiri and Morris, 1996). Different studies showed that
considerable number of illnesses is caused by improper handling practices
(Flint 2005). Observation shows that most peddler exhibit poor handling
method, and so food drinks are exposed to hazardous situations such as
cross contaminations and undesirable temperature (Ekanem, 1998).

The pathogens can be ingested through the nose, skin and faeces
especially through hands of the handlers (WHO, 1989). Research revealed
that Escherichia coli, Salmonella, Campylobacter and other bacteria remain
viable under finger tips and other surfaces for some days (Howes 1996).

Hawking of food drinks in the street is a source of diet of most people

towns (Suneetha 2011). About 2.5 billion people worldwide rely on vended
9
food drinks every day, and as such most nations have observed changes in
their socio-economic level in about ten years ago; the changes ledto
significant growth in the majority of vended street food drinks (WHO, 1996).
With increase in population and urbanization in developing countries, it is
expected that vended food drinks will be on the increase (WHO, 1996).

1.2 Street-hawked food drink has many benefits and these include:

1. It provides source of income for majority of people.

2. It is not expensive, convenient, and nutritive drink for both rural and
urban area.

3. It also provides business low capital and allows for self-employment

(WHO, 1996).

Beside the fact that, hawking of these drinks can result to problem of
waste disposal in the city, the major concern is how safe they are.
These are possible cause of severe drink poisoning outbreaks in most
nations around the world, via microbiological threat (WHO, 1996).

1.3 Statement of the problem

The way and manner vendors handle and transport the finished
product from the manufacturers or production site to their various outlets for
sales to the consumers poses a challenge and can result in post-production
contamination. They do not put into consideration temperature and the way
it's been handled if the package is dumped in an un-hygienic truck or
vehicle, it gets contaminated. These result in risk in public health.

Generally, the challenge of yogurt production in Nigeria is

characterized by inadequate housing facility, lack of bacteriologically free
water (potable); poor waste management, lack of adequate environmental
sanitation in areas of high population density and people with low income.
The Agencies in charge of enforcing good environmental sanitation laws
are not usually available to caution or prosecute offenders. This regulatory

10
negligence results in the nonchalant attitude of these industries in
disposing of their waste adequately.

1.4 Aims and objectives

The purpose of this research is to assess the quality of selected

yogurt commercially produced and sold in Ota Metropolis, Ogun State.
Specific objectives are:

1. To isolate and characterize microorganisms (fungi and bacteria) that

may be present in the yogurt samples.

2. To ascertain whether microbial isolates are pathogenic.

3. To ascertain the presence of starter cultures in the products

11
 CHAPTER TWO
LITERATURE OVERVIEW

2.1 CONTAMINATION DEPENDING ON WEATHER

The manufacturing process of yoghurt is quite simple and easy which

infers that any peon could do it. This paves way for higher contamination
rates since yoghurt can be manufactured at any time, at any place, with
any equipment by anyone. Thus, safe consumption of yoghurt is still an
unanswered question. Research done by clearly shows that high hygiene
practices should be undertaken in the production process of both small and
large-scale industries. Random yoghurt samples obtained from both large
and small industries in Qena district of Egypt showed maximum
contamination rates. In their study the microbiological quality was analysed
from ready to sell yoghurt containers, 50 samples from small scale
industries and 50 samples from large scale industries were put to test for
the estimation of psychrotrophs, enterococcus, Coliforms, yeasts and
moulds.

Psychrotrophic bacteria that grow in refrigeration temperatures are

known to cause food poisoning due to the products like lipase and protease
after enzyme conversion by assimilating nutrients from dairy products. The
presence of enterococci may be as a result of poor sanitation of the
industry since the organism is highly resistant to detergents, drying and
freezing conditions. Coliforms are often used as indicator organisms for
faecal pollution. Their presence in dairy based products also indicates that
other enteropathogenic microbes may be present. The presence of
Staphylococcus aureus in milk and milk-based products may be because of
its initial contamination through personal sharing and in handling and
production procedures. The samples examined from small scale industries
had been contaminated with about 92% of psychrotrophs with count range
of 1.7 * 10 ^ 3 * t 3 * 10 ^ 5 with average count 3.9 x 10/g. The samples
examined from large scale industries had been contaminated with about
70% of psychrotrophs with count range of 4 * 10 ^ 2 to 6 * 10 ^ 4 with
average count 6.8 * 10 ^ 3 / 1. The high count was due to inadequate heat
12
treatment, improper processing, and careless handling or because of the
presence of high psychrophiles in raw milk that was used for production.
The presence of enterococci in small scale industries ranged from 2.5 * 10
^ 2 to 1.6 * 10 ^ 5 with an average count 1.7 x 10 /g with 58 percent in total.
In large scale industries the average count was 2 * 10 ^ 3 / g ranging from 1
* 10 ^ 2 to 1.5 x 10 with 40% in Contamination by enterococci in yoghurt
often implies the negligence of hygienic control measures during handling
and production.

Contamination by Staphylococcus aureus in small scale level was

72% with 8.5 * 10 ^ 3/g as an average count. Whereas in large scale
industries Staphylococcus aureus contamination was 36% with 9.4*10^2/g
as average count. Presence of Staphylococcus aureus in yoghurt is
because of humans who act as mediatory channel for infection by way of
coughing, sneezing, or handling equipment after touching lesions infected
with Staphylococcus aureus, Yeast and mould count also estimated gave
average count of 1.4 * 10 ^ 4 / g ranging from 2.5 * 10 ^ 2 to 1.4 * 10 ^ 5 in
small scale and 3.9 * 10 ^ 2 / g as average count in large scale with range
of 1 * 10 ^ I to 1.4 * 10 ^ 3 Poor sanitary practices are the major cause for
deposition of yeasts and moulds in high acid product like yoghurt.

2.2 CONTAMINATION BY ADULTERANTS

A study conducted by (Rahman 2020) in Bangladesh cities revealed

the presence of few adulterants in addition to high protein levels. The study
showed local, unbranded yoghurt sample contained about 29% urea, 19%
ammonium sulphate, 33% starch and 19% hydrogen peroxide. Likewise,
branded, commercial yoghurt samples contained 29% starch, 14% urea,
36% ammonium sulphate and 21% hydrogen peroxide. Commercial urea is
mostly added to increase the level of non-protein nitrogen content and
hydrogen peroxide is used for long term storage that acted as a
preservative. In this study adulterants like starch, hydrogen peroxide,
ammonium sulphate and urea were found which is an indication of health
hazard. Because high level of starch could be a reason for diarrhoea
because of its effects like indigestion of starch in colon. Ammonium
13
sulphate and hydrogen peroxide can cause intestine to get inflamed. From
this study it can be seen that both branded and unbranded yoghurt were
adulterated with chemical substances like urea, ammonium sulphate,
starch and hydrogen peroxide.

2.3 SIGNS OF CONTAMINATED YOGHURT

A detailed report on inward collapse and swelling of yoghurt packs

was done by. Before buying yoghurt, it is essential to see the
manufacturing and expiry date and certain signs of contamination that
could be observed with our naked eye. Typically, the research shows that
samples analysed within expiry date had contaminants responsible for
inward collapsing of the lid and swelling seal due to anaerobic condition by
release of CO2 by certain microflora. Three different brands studied showed
comparative yield of bacteria, yeasts and moulds. From the samples
collected, one brand had bacterial count that reached a level of 10 ^ 7 *
cfu / g the second brand had no bacterial contaminants but the main
contaminants were moulds with colony count range of (10^2) to 10^5 *cfu/g
) while the third brand had both yeasts and moulds ranging between 10^6
and 10^5 *cfu/g.

Contaminating strain of brand one was found to be Acetobacter aceti

and samples from brand two had organisms like Mucor hiemalis Wehmer.
Mucor racemosus Fres and PenicilliumverrucosumDierckx var cyclopium
were identified. In brand three samples, moulds like Mucorhiemalis and
Mucor racemosus and yeasts like Debaryomyceslansenii were isolated.
The inward collapse of the yoghurt pack was mostly because of
Acetobacter aceti and swelling was because of yeasts and moulds.
However, after swelling inward collapse was seen after five to seven days
of incubation at refrigeration temperature. From their study it is evident that
certain alterations in the yoghurt packets can be taken as a sign of
contamination, but then again if the manufacturing practices were good and
in strict correct conditions of refrigeration there can always be a way to
slow down such kind of alterations.

14
Aerobic mesophilic and heat resistant thermophilic spore formers by
(Eijlander 2020) with a slight modification. For TMS, 1 ml. of 10-fold serial
dilutions of heated (80 °C, 10 min) and homog enized samples were pour-
plated using tryptic soy agar (TSA). A layer of 2% non-nutrient agar
(instead of 1.5% used by (Eijlander 2020) was poured on top of the
solidified TSA agar to prevent swarming of spore formers. Plates were
incubated for 48-72 h at 37 ± 2 °C and counted. For HRTS, the procedure
was identical with the difference that the initial heating step was 100 °C for
30 min and incubation of the plates took place at 55 ± 2 °C.

2.4 ANAEROBIC SULFITE REDUCING CLOSTRIDIUM SPORES

(SRCS)

To enumerate SRCS, homogenized samples were heated at 80 °C

for 10 min, cooled, 10-fold diluted in PPS and then 1 ml. of the required
dilution was pour-plated using iron sulfite agar containing 0.5 g/L sul- fite
(ISA-0.5) according to the ISO 15213:2003 (International Organi- zation for
Standardization, 2003). Solidified plates were incubated at 37. 2°C for 48-
72 h inside the anaerobic chamber (SHEL LAB Bactron, BACTRON900-2;
Cornelius, USA; gas composition: 90% N2, 5% CO2, 5 % H₂).

2.5 HEAT RESISTANT MOULDS (HRM)

For the detection of HRM, samples were heated at 80 "C for 30 min,
cooled and then pour-plated (1 ml. of the sample) onto 90 mm Petri dishes
with 18 to 20 mL of malt extract agar (MEA) supplemented with 0.1 mg/ml.
chloramphenicol (SR0078E, Thermo Scientific). Solidified plates were
incubated at 25 ± 2 °C and colonies were counted after 5. days up to 14
days of incubation.

2.6 ENUMERATION OF BACTERIA AND SPORES

Bacterial colonies were counted using a lightbox and recorded as the

number of colony forming units (CFU) per gram of the ingredient. For
samples with counts between 0 and 10 CFU/g results were registered as
10 CFU/g (The United States Food and Drug Administration (FDA), 2001)

15
which is the limit of detection (LOD) of the applied microbio- logical
methods. The limit of quantification (LOQ) was 100 CFU/g. In case no
colonies were detected on plates, this was considered as <10 CFU/g and
reported as "Not detected" (ND).

2.7 BACTERIA ISOLATION AND IDENTIFICATION

Bacterial colonies with different morphologies were transferred from

enumeration plates to fresh agar plates and incubated aerobically (TVC,
TMS, HRTS, BCES) or anaerobically (SRCS). A maximum of five separate
colonies from the best countable dilution plate per sample were picked. If
the colony morphologies were highly diverse, in those cases up to 10
colonies were selected. The culture media and incubation conditions
applied were identical to those used in the corresponding enumeration
methods. After incubation and before identification all isolated strains were
stored at 80 °C in Micro Bank tubes (PRO-LAB Diagnostics, Richmond Hill,
CA) according to the manufacturer's instructions. Isolates that were
identified as species belonging to the Bacillus and Clostridium genera were
placed in the NIZO culture collection.

2.8 IDENTIFICATION USING MALDIT OF MS

The identification of bacterial isolates was performed using Matrix-

Assisted Laser Desorption/Ionization Time-Of-Flight Mass Spectrom etry
(MALDI-TOF MS) with the MALDI Biotyper Sirius System (Bruker
Daltonics, Bruker Scientific Instruments, Billerica, MA) (see details about
consumables and software used in Васterial isolates from-80 °C stocks
were transferred to fresh agar plates and incubated aerobically or
anaerobically as listed. The recovered and purified colonies were then
smeared onto a position on the MBT Biotarget 96 plate, overlaid with 1 pl.
of a-Cyano-4-hydrox- ycinnamic acid (HCCA) matrix and left at room
temperature until completely dry (30 min). The manufacturer's bacterial test
standard (BTS) was used for the instrument calibration. For each strain,
two preparations of sample material were analysed. The identification was
2,3 performed using the BDAL library.

16
2.9 IDENTIFICATION USING 16S rRNA GENE SEQUENCING

To perform identification based on the 16S rRNA analysis, genomic

DNA of microbial isolates was extracted using InstaGene matrix (Bio- Rad
Laboratories) according to the manufacturer's instructions. The primers
(200 nM) listed were used for amplification of the 165 rRNA gene, using the
Promega #M750C Taq PCR master mix. The PCR reactions were
performed in the GeneAmp™ PCR System 9700 (Applied Biosystems)
thermal cycler under following conditions: a polymerase activation step of 1
min at 95 °C, followed by 35 cycles of 30 s at 95 °C (denaturation), 45 s at
50 °C (annealing) and an extension step at 72 C for 45 s. The PCR
amplicons were checked using the E-Gel electrophoresis system (REF
G521802, Invitrogen, Thermo Fisher Scientific) and puri fied using Gene
JET PCR purification columns (REF K0702, Thermo Scientific) according to
the manufacturer's instructions. Purified PCR product (7.5 µL) was mixed
with 2.5 µL forward (27F) or reverse primer (1492R) and sent for the
LightRun Tube sequencing analysis at Eurofins Genomics (Ebersberg,
Germany). The sequencing result files in within the Bacillus cereus group,
B. anthracis, B. cereus, B. mycoides, B. pseudomycoides, B. thuringiensis
and B. weihenstephanensis are closely related and cannot be well
distinguished only based on the MALDI-TOF MS analysis. In particular, B.
cereus and B. anthracis MALDI-TOF MS spectra are very similar. B.
anthracis is not included in the Maldi Biotyper commercial database.

Clostridium sporogenes and C. botulinum Group I are closely related

and show highly similar MALDI-TOF MS spectra and/or 165 rRNA
sequencing. C. botulinum is not included in the commercial database.

17
Targeted Primer Amplified Sequence (5-3) Sequence Reference
gene fragment reference or DNA
size (dp) of the strain

hbl HD3F 1091 GTA AAT TAI GAT GAI B. cereus

HA4R CAA TTTC ATCC 14579
AGA ATA GGC ATT
CAT AGA TT

nhe NA2F 766 AAF CTG CTC TTC B. cereus (Ebling-Schulz et

NB1R GTA TTC NVH1230/88 al., 2006)
TTT FTT GAA ATA AGC
TGT GG

Cytk CKF2 421 ACA GAT ATC GGI B. cereus

CKR5 GAA AAT GC NVH0391/98
CAA GTT ACT TGA
CGT GTT GC

Ces CesF1 1271 GGT GAC ACA TTA B. cereus

CesR2 TCA TAT AAG GTG F4810/72
GTA AGC GAA CCT
GTC TGT AAC A

ichAA CY21 472 CGA TTC CGG GTT B. lichemiformis (Yeak et al.,
CY22 ATT GAC TG ATCG14580 2022)
CGC TTC ATA TTG
TGC GTT CC

boNT-A AF 282 TGC AGG ACA AAT C. botulinum

AR GGA ACC AGT NCTC 57
TCC ACC CCA AAA
TGG TAT TCC

boNT-B BF 311 CCT CCA TTT GCG C. botulinum (Takeshi et al.,

BR AGA GGT ACG NCTC 59 1996)
CTC TTG GAG TGG C. botulinum
AAC GGT ACG NCTC 94

16S rRNA 27F 1500 AGA GTT TGA TCC

1492R TGG CTC AG
GGT TAC CTT GTT
ACG ACT T

Table 1: Oligonucleotide primers used in this study

18
2.10 DNA EXTRACTION AND DETECTION OF TOXIN-ENCODING
GENES

Bacterial isolates confirmed as B. licheniformis, C. sporogenes/tepidum

and B. cereus group species were analysed for the presence of toxin-
encoding genes using a PCR method. All primers used in this analysis .

For B. cereus group isolates, the presence of the following genes was
investigated: hbl-encoding hemolysin BL, nhe encoding non-hemolytic
enterotoxin, cytk encoding cytotoxin K, and ces gene cluster that is
responsible for production of cereulide. B. licheniformis isolates were
screened for the presence of the lichenysin synthetase gene IchAA. As C.
sporogenes and proteolytic C. botulinum are genetically closely related and
cannot be well distinguished based on the f MALDI-TOF MS and 165 rRNA
gene sequencing (Brunt 2020), all isolated C. sporogenes/ tepidum strains
were checked for the absence of bonNT/A and boNT/B genes which
encode C. botulinum neurotoxins A and B, respectively. Details about PCR
analyses performed are presented in the following sections.

For the DNA extraction, selected bacterial stocks were plated onto
Brain Hearth Infusion (BHI) agar and incubated at 30 deg 0C for 24 h.
Single colonies were inoculated in 15 mL of BHI broth and incubated
overnight at 30 deg 0C Cell pellets were collected by centrifugation
(Eppendorf 5804R, Eppendorf AG, Hamburg, Germany) at 3000 g for 5-10
min, resus- pended and lysed in the S.T.A.R. buffer (03335208001, Roche
Molecular Systems Inc., Branchburg, USA) through beating with 0.1 mm
diameter zirconia beads at 95 deg 0C for 10 min. Bacterial DNA was
subsequently. extracted from the supernatant using the Maxwell 16
Instrument (Promega, Madison, WI, USA) and Maxwell DNA isolation kit
(Prom- ega), according to the manufacturer's instructions. During the PCR
analysis, nuclease-free water was always used as a negative control and
16S rRNA gene amplification was applied as a positive control for the PCR
reaction.

19
2.10.1 BACILLUS CEREUS TOXIN PROFILING

For 147 bacterial colonies isolated from selective (MYP, PEMBA,

RAPID 'B. cereus) and non-selective (PCA, TSA) media and confirmed as
B. cereus group strains, a PCR method published by (Ehling-Schulz 2006)
was used with the following modifications: a duplex PCR was performed to
detect cytk and nhe genes, whereas hbl and ces genes were checked by
the monoplex PCR. The PCR mixture (25 µL.) consisted of 0.2mM of each
dNTP, 25 units/mL of Taq DNA polymerase and buffer (pH 8.5) (M750C
PCR Master Mix, Promega), 3 mM MgCl2, oligonucleotide primers (0.4 µM
for the ces gene, 1 µM for the hbl gene, and 0.5 µM for the the and cytk
genes), 2 ul. Template DNA and 4 µl. nuclease-free water. Each
amplification started with a polymerase activation step of 3 min at 95 deg 0C
followed by 35 cycles (40 for hbl) of 30 s at 95 deg 0C (denaturation), 30 s at
50 deg0C (annealing), 1 min at 72 deg0C extension), and ended with a final
elongation step at 72deg0C for 2 min. The amplification reactions were
performed in Biometra Tone Thermal Cycler (Analytik Jena AG, Jena,
Germany). B. cereus strain ATCC 14579 was used as a positive control for
the presence of the nhe, cytk and hbl genes, and B. cereus F4810/72 for
the presence of the ces gene. Oligonucleotide primers for cytk were
directed at highly conserved regions of the toxin gene to detect both forms
of cytk (cytK-1 and cytK-2) (Ehling-Schule 2006, Fagerlund 2004).

2.10.2 BACILLUS LICHENIFORMIS ICHAA GENE DETECTION

Detection of the IchAA gene (encoding lichenysin toxin) was per-

formed by the PCR analysis as described in section 2.5.1 with a few
modifications. Each amplification started with a polymerase activation step
of 2 min at 95 deg 0C followed by 35 cycles of 20 s at 95 °C (denatur-
ation), 10 s at 55 deg 0C annealing) and an extension step at 70 deg 0C for
10 s. The KOD Hot Start DNA polymerase kit (Novagen 71086-3, Toyobo
Co., Osaka, Japan) was used in all reactions. The final reaction mixture (25
µL) contained a 1:10 dilution of 10x Buffer for KOD Hot Start DNA
Polymerase, 0.02U / mu * L KOD Hot Start DNA Polymerase, 1.5 mM
MgSO4. dNTPs 0.2 mM (each), oligonucleotide primers at 0.3 uM 2 μμ.
20
Template DNA and PCR-grade water added to 25 µL. B. licheniformis
NIZO 4090 strain was used as a positive control for the IchAA gene.

2.10.3 CLOSTRIDIUM BOTULINUM TOXIN A AND B GENES

DETECTION

The PCR mixture was prepared with nuclease-free water, Dream Taq
Hot Start DNA Polymerase with associated buffer, dNTPs (2 mM), primer
sets at 10 µM and 2 µL of DNA template (25ng / mL) to 48 µL. Master Mix,
for a total reaction volume of 50 µl. Each amplifi- cation started with a
polymerase activation step of 1 min at 95 deg 0C fol- lowed by 30 cycles of
30 s at 95 deg 0C (denaturation), 45 s at 55 deg 0C (annealing) and
extension step at 72 deg 0C for 45 s. Final elongation was performed at 72
deg 0C for 10 min. DNA of strains C. botulinum NCTC 57, NCTC 59 and
NCTC 94 were used as positive controls.

2.11 IDENTIFICATIONS OF MICROBIAL CONTAMINANTS IN PLANT-

BASED INGREDIENTS

Altogether, 845 individual bacterial colonies were isolated from the

highest dilutions of the plates used to enumerate TVC, TMS, HRTS, SRCS
and BCES in plant-based ingredient samples. The highest number of
isolates was obtained from the TVC and TMS plates and comprised 310
and 261 bacterial colonies, respectively. Fewer isolates were picked from
the plates on which BCES, SRSC, and HTRS were counted, largely
because of the lower levels at which these microorganisms were present.
In total, 160 microbial colonies were isolated from the BCES analysis, 76
from the SRCS plates and the lowest number of isolates (38) was ob tained
from the HRTS plates. The iden- tifications of isolated colonies revealed
representatives of 33 different bacterial genera. An overview of the isolates
at genus and species level per microbiological analysis is presented and
species per analysis and per ingredient category. At the highest dilutions of
the TVC plate assay, spore-forming and non-spore-forming
microorganisms were detected. The isolated spore formers belonged to the
genera Bacillus, Brevibacillus, Fictibacillus, Lysi- nibacillus, Paenibacillus
and Streptomyces, with Bacillus species accounting for 60% of all TVC
21
identifications. Overall, B. cereus group members, B. licheniformis and B.
amyloliquefaciens comprised 27%, 25% and 9% of Bacillus identifications,
and 16%, 15% and 6 % of all TVC identifications, respectively. Isolates of
non-spore-forming species were represented by 18 different general of
various Gram negative and Gram positive bacteria. The diversity of non-
spore-forming species encoun- tered in the main groups of ingredients was
higher in pulses than in (pseudo)cereals and drupes. Isolates obtained as
part of the TMS analysis mainly belonged to four spore-forming genera:
Bacillus, Geobacillus, Lysinibacillus and Paeniba- cillus, and occasionally
some non-spore-forming species were detected. The majority of the TMS
isolates (87 %) were identified as species belonging to the Bacillus genus,
where 42 % of isolates were identified as B. licheniformis, 21% as B.
subtilis and 11 % as B. cereus group species. In total, 31 different Bacillus
species were isolated from various plant-based in- gredients in TMS and
TVS analyses, with the highest variety of Bacillus species detected in pulse
samples, then in (pseudo)cereals, and drupes. Isolates belonging to the B.
subtilis group and B. cereus group occurred the most frequently, B. subtilis
itself was detected in 33 % of samples, while B. licheniformis and B. cereus
group species were found in 47% of all ingredients. Most of the confirmed
B. subtilis and B. cereus group isolates originated from pea and oat
samples.

Presumptive B. cereus (BCES) isolates that were picked from selective

media were mostly confirmed as B. cereus group members with 22% of
isolates identified as B. cytotoxicus. Occasionally, species were identified
belonging to the genus Lysinibacillus (one time in pea concentrate, mung
bean flour and chickpea isolate), Paenibacillus (one time in mung bean
powder), and to the B. subtilis group (9% of all BCES identifications).
Moreover, MYP agar sometimes allowed the growth of mannitol-fermenting
bacteria (resulting in yellow colouration of the medium), and this interfered
with the counting and isolation of typical B. cereus colonies. A higher
selectivity was observed for the RAPID'B. cereus agar than for MYP and
PEMBA: all colonies picked from this medium were identified as B. cereus
group members (in total 55 microbial isolates). Of the selective media used

22
in this study, this me dium was therefore preferred for enumeration of the
B. cereus group species in plant-based ingredients.

For the HRTS isolates (surviving 30 min at 100°C and growing at

0
55 C), approximately half were identified as strict thermophiles (Geo-
bacillus stearothermophilus, Geobacillus subterraneus and
Aneurinibacillus) and around 39% belonged to the B. subtilis group. On two
occasions colonies of vegetative bacteria were identified, namely,
Staphylococcus haemolyticus and Micrococcus luteus, which we attribute
to the cross-contamination during the sample handling or plating.

Most of the collected isolates of anaerobic mesophilic spore formers

(96%) (SRCS) were identified as Clostridium spp., including C. sporogenes
and C. tepidum (41%), Clostridium diolis and Clostridium beijerinckii (12%),
Clostridium amylolyticum (9%) and Clostridium argentinense (8%).
Additionally, Clostridium cochlearium, Clostridium butyricum, and
Clostridium tertium were detected in some sample. The largest variety of
the SRCS species was observed in pulses, lower in drupes, and no isolates
were recovered from samples of (pseudo) cereals.

2.12 DETECTION OF TOXIN GENES

The potential of isolated bacterial strains to produce toxins associ

ated with foodborne infections or intoxications was investigated using the
PCR-based methods. An overview of the toxin genes detected in selected
bacterial isolates is presented. This analysis revealed that from 147 B.
cereus group isolates, approximately 9% of the strains contained the ces
gene which encodes the heat-stable emetic toxin cereulide. For
enterotoxin-encoding genes, 28% of all strains carried the hbl gene, 42%
were positive for cytk, and the non-hemolytic enterotoxin the gene was
detected in 69% of the strains. Moreover, around 9% of the B. cereus
group isolates contained genes for both emetic toxin and one of the
enterotoxins (nhe, hbl or cytk). In 4% of the isolates none of the targeted
toxin genes was detected. A detailed overview of the toxin profiles for
selected B. cereus group iso- lates is presented

23
 For B. licheniformis, a total of 98 bacterial isolates were analysed,
and 98% of the strains rendered the IchAA PCR product, indicating that
nearly all of these isolates have potential to produce lichenysin.

Additionally, because C. sporogenes and C. botulinum Group I

strains are genetically closely related and their MALDI-TOF MS spectra
cannot be distinguished using the applied database, all 34C. tepidum/spor
ogenes isolates were analysed for the presence of the boNT-A/B genes,
but these genes were not detected in any of the strains

Species Number of % of positive PCR results

isolates
IchAA cytk hbl nhe ces boNT-A/B

B. licheniformis 98 98% - - - - -

B. cereus group 147 - 42% 28% 69% 9% -

C. sporpgens/tepidum 34

Table 3: Toxin gene presence in bacterial isolates from plant-based ingredients. Selected isolates
were analyzed for the presence of toxin encoding genes (lichenysin (lchAA), hemolytic enterotoxin
HBL (hbl), non-hemolytic enterotoxin NHE (nhe), cytotoxin K (cytK), cereulide (ces), and botulunum
toxin BoNT-A/B (boNT-A/B) using PCR.

Ingredients can occur at different points in the production chain, for

instance, during cultivation of the crop (influenced by geographical location,
weather/climate or the use of biological pesticides such as B. thuringiensis
(Johler 2018), harvesting (inevitably leading to contamination with soil and
dust and microorganisms therein), ship- ment and storage of harvested
crops. The processes used to obtain different forms of ingredients can also
influence levels and types of microorganisms encountered. As an example,
concentrates of pulses are obtained via dry fractionation, which does not
involve wetting or heating steps. On the other hand, the production of pulse
isolates is a wet process involving dissolving, heating and evaporation
steps, during which thermophilic spore-forming bacteria (which may be
24
present in ingredients and equipment) can survive, grow and produce
spores that contaminate the product. Depending on the ingredient, a large
spectrum of microorganisms was initially present in plant-based samples as
part of the TVC isolates, with the highest diversity of non-spore formers in
pulses. In conclusion, differences in starting concentrations and types of
microbial contaminants present in plant-based ingredients, even of the
same type, can be influenced by many factors, most of which are difficult to
control.

In this study, sample preparation of plant-based ingredients pre-

sented some challenges, particularly for the spore counting methods that
involve a heating step prior to plating. Many of the ingredients were poorly
dispersible and sometimes could not be homogeneously sus- pended in the
BPW or PPS solutions. In addition, the applied heat treatment caused
gelation in many samples, probably due to starch gelatinization and/or
protein coagulation upon heating. This can lead to uneven distribution of
microorganisms in samples, and particles of the coagulated ingredient in
the plating agar can make counting difficult, especially at low dilutions. To
reduce gel- ling upon heating and to obtain more homogeneous samples,
less concentrated solutions (> 1:10) of the ingredient can be heated but this
will affect the limit of detection unless a higher volume of the diluted sample
is plated. In addition, the use of starch-degrading enzymes such as
amylase can be considered to improve sample homogeneity prior to
performing serial dilutions. Another challenge posed by plant-based
ingredients is counting colonies on plates when swarming Bacillus spe cies
(such as B. licheniformis) are present in relatively high concentra- tions, as
was the case for many samples. Adding the non-nutrient overlay, as
applied in this study (and as described by Eijlander 2020), reduced the
spread of swarming bacteria across the plate and facilitated counting.

Moreover, their virulence and toxigenic potential can differ. In this

study, 9% of the B. cereus group strains isolated from various ingredients
contained the gene required for cereulide production. Cereulide- producing
strains represent a potential food safety hazard even if no viable organisms
are detected, for example, if this heat-stable toxin is produced prior to heat
25
inactivation of the cells. It was recently reported that psychrotolerant B.
cereus isolates can grow and produce cereulide in cereal-based substrate
at temperatures as low as 12 °C, with higher levels being produced at 20
°C (Ellouz 2021). Cereulide pro- duction can even take place at
temperatures below 10°C, as reported for a psychrotrophic strain by
Thorsen et al. (2006). It is remarkable that in our study PCR toxin typing
showed a low prevalence (69%) of nhe- positive B. cereus group isolates.
This appears to be a contradiction to the assumption that nearly all B.
cereus group strains carry the nhe genes (Guinebretière 2010). A total of
76% (35 out of 46) of the the negative strains were identified as B.
cytotoxicus (belonging to PanC group VII) based on MALDI-TOF MS and
16S rRNA identifications (Supplementary Tables 4C, 5A). Thus, the low the
prevalence is thought to result from the high frequency of B. cytotoxicus
which carries the nhe gene variant that is not detected with the primers
used in this study (Fagerlund 2007,Guinebretière 2010). B. cytotoxicus was
frequently encountered in pea, coconut and almond ingredients and all
analysed isolates carried the cytk gene as determined by the PCR anal-
ysis. This species is distinguished from other species in the B. cereus
group by its ability to grow at 50 °C (Guinebretiere, 2008, Le Mare 2021)
and produce cytotoxin K-1 (CytK1) (Fagerlund,2004, Guinebretiere 2006).
Cases of B. cytotoxicus food poisoning have been also associated with
plant-based foods, for instance, vegetable puree (Laind 2000), cooked
semolina and potato purees.

(Guinebretière 2013). B. cytotoxicus thermotolerance and its ability to

survive moderate heat treatments are important properties to consider in
food manufacturing processes that are operated at elevated temperatures.

The results from our study on the prevalence of toxin genes that are
present in spore formers isolated from plant-based ingredients provide a
first indication of potential risks when these ingredients are used in foods. A
more thorough evaluation is needed to accurately predict and minimize the
actual food safety risks associated with the presence of spore formers in
plant-based ingredients. This includes defining the conditions under which
spores of potentially pathogenic species can survive, germinate and
26
multiply in foods, and when and to what extent toxins are produced. This is
part of our ongoing studies. Overall, the findings of this work present
valuable data regarding the types and levels of microorganisms in various
plant-protein sources. Combined with knowledge of heat inactivation
kinetics and growth characteristics (pH, aw, temperature) of various
species, microbiological risk assessments for novel plant-based
formulations can be performed. This will support the development of
effective strategies to control microbial contamination within the production
chain of plant-based meat and dairy alternatives. Supplementary data to
this article can be found online.

This work was funded by the European Union's Horizon 2020

Research and Innovation Program under the Marie Sklodowska-Curie grant
agreement No. 721456 and No. 956126 and by the Topsector Agri & Food
(grant No. LWV21011). Within the Topsector, industry, knowledge institutes
and the Dutch government work together in in- novations for safe and
healthy food for nine billion people in a resilient world.

27