Copyright is owned by the Author of the thesis.  Permission is given for 
a copy to be downloaded by an individual for the purpose of research and 
private study only.  The thesis may not be reproduced elsewhere without 
the permission of the Author. 
 



i 
 

Presence of coagulase-negative 

Staphylococcus in dairy products 

 

 

 

 

 

 

 

 

A thesis presented in partial fulfilment of the requirements for the degree of 

Master of Science at Massey University, Palmerston North, New Zealand 

 

 

 

 

 

 

 

 

 

 

 

 

Jacinda Aplin 

2014 



ii 
 

Abstract 

Species within the genus Staphylococcus produce various virulence 

factors, including staphylococcal enterotoxins. Because of their 

production of enterotoxins, Staphylococcus is the third most common 

pathogen responsible for outbreaks of food poisoning worldwide. 

Whereas Staphylococcus aureus, a coagulase-positive Staphylococcus, 

is the leading cause of these outbreaks, coagulase-negative 

staphylococci (CNS) species are also present in food and are able to 

produce enterotoxins. More specifically, such CNS species have been 

linked with dairy-related food poisoning outbreaks. However, to date, no 

research investigating CNS species in New Zealand and their presence 

in food has been reported. This study therefore sets out to isolate, 

identify and characterise CNS from New Zealand milk and dairy 

products, and to evaluate their toxin-producing potential. The results from 

this study showed that MALDI−TOF MS is a rapid, reliable and accurate 

method for identifying CNS definitively to the genus level and, on most 

occasions, to the species level in dairy products and is therefore a 

potential alternative to the traditional phenotypic, such as commercial 

identification kits, and genotypic, such as sequencing, methods that are 

currently used. Of the 42 isolates analysed, none of the CNS isolates 

tested produced enterotoxin in vitro, however, 2 isolates were found to 

possess an enterotoxin gene. This shows a low propensity for CNS 

isolates in New Zealand dairy products to be a food safety risk.  

 

 

 

 

 

 

 

 

 



iii 
 

Acknowledgements 

I would firstly like to thank my supervisors, Dr Denise Lindsay and Dr Jon 
Palmer, who allowed me to really take charge of this project from the outset, all 
the while patiently giving me advice, feedback, encouragement, support  and 
their expertise every step of the way. 
 
Secondly, I would like to thank my manager, Dr Steve Holroyd, for supporting 
me in completing my PGDipSc and MSc over the last three years and allowing 
me the time to complete my laboratory work. Also to my subsequent manager, 
Dr Joanna Shepherd, for allowing time to complete my laboratory work as part 
of the PGDipSc project.  
 
I would also like to thank Fonterra for financially supporting me through my 
studies over the last 3 years. 
 
I would like to thank the Process Technical Microbiologists, Fonterra, located 
around New Zealand, for providing the dairy samples, and Analytical Services, 
Fonterra, for providing the Baird−Parker plate samples for this project. 
 
Finally, I would like to thank Dr Susan Paulin, Institute of Environmental Science 
and Research Ltd (ESR) for providing the enterotoxin producing S. aureus 
species for this project. 
 
 
 

 

 

 

 

 

 

 

 

 

 

 

 

 



iv 
 

Abbreviations 

BHI brain heart infusion broth 

bp base pair 

BP Baird−Parker agar 

CFU/mL colony forming units per millilitres 

CNS  coagulase-negative Staphylococcus/staphylococci 

CPS coagulase-positive Staphylococcus/staphylococci 

DNA  deoxyribonucleic acid 

DNase deoxyribonuclease 

kb kilobase 

kDa kilodalton 

MALDI−TOF MS matrix-assisted laser desorption ionisation−time of flight mass spectrometry 

MSP main spectrum  

m/z mass to charge ratio 

PCR polymerase chain reaction 

RAPD randomly amplified polymorphic DNA  

RNA ribonucleic acid 

rRNA ribosomal ribonucleic acid 

rpoB a gene that encodes the β-subunit of RNA polymerase 

SAgs superantigens 

SE(s) staphylocoocal enterotoxin(s) 

SEL(s) staphylococcal-enterotoxin-like superantigens 

sodA a gene that encodes the manganese-dependent superoxide dismutase 

subsp. subspecies 

spp. species 

TSA tryptone soy agar 

TSB tryptone soy broth 

TSST-1 toxic shock syndrome toxin 1  

tuf a gene that encodes the elongation factor Tu 

 

 

 

 

 

 

 

  



v 
 

Table of Contents 

Abstract                                                                                                                                                           ii 

Acknowledgements                                                                                                                                        iii          

Abbreviations                                                                                                                                                 iv          

List of Figures                                                                                                                                                vii        

List of Tables                                                                                                                                                  ix           

CHAPTER ONE: INTRODUCTION                                                                                                                1           

1.1 The Genus Staphylococcus                                                                                                       1           

1.2 Coagulase-negative Staphylococcus (CNS)                                                                              2           

1.3 Staphylococcal Enterotoxins (SEs)                                                                                            2             

1.4 Presence in Food                                                                                                                       3 

1.5 Enterotoxigenic Potential of CNS Species in Food                                                                    6           

1.6 New Zealand                                                                                                                              7             

1.7 Protein Profiling − A Novel and Rapid Method for Identifying CNS and their Enterotoxins       9          

1.8 Aims of this Investigation                                                                                                         12           

CHAPTER TWO: MATERIALS AND METHODS                                                                                         13            

2.1 General Requirements                                                                                                                           13          

2.2 General Methods                                                                                                                                    15          

2.2.1 Collection of samples and bacterial isolation                                                                        15           

2.2.2 Positive and negative controls                                                                                              18           

2.2.3 Phenotypic identification                                                                                                       18           

2.2.4 MALDI−TOF MS analysis                                                                                                     22            

2.2.5 Identification by 16S rDNA sequencing                                                                                26           

2.2.6 Immunoassay test                                                                                                                28            

2.2.7 SE PCR                                                                                                 30 

2.2.8 SE multiplex PCR              34          

2.2.9 Interpretation of resulting identities                                                                                      35             

CHAPTER THREE: RESULTS AND DISCUSSION                                                                                   36             

3.1 Product Analysis                                                                                                                                   36 

3.2 Identification of Coagulase-negative Species                                                                                       36            

3.2.1 Identification of CNS isolates using the BBL™ Crystal™ Gram Positive identification                     

         system, Remel RapID™ Staph Plus and MALDI−TOF MS                                                 36 

             3.2.2 Definitive identification of seven CNS isolates using 16S rDNA sequencing                        41              



vi 
 

             3.2.3 Comparison of MALDI−TOF MS identification and 16S rDNA sequencing                           42           

             3.2.4 Overall examination of identification results                                                                          45                                                                                                                                                                                                                                                              

3.3 Staphylococcal Enterotoxin                                                                                                                    48          

             3.3.1 Immunoassay test results                                                                                                       48           

             3.3.2 SE gene identification of seven isolates                                                                                 48   

             3.3.3 SE gene identification using multiplex PCR                                                                           48 

             3.3.4 Overall examination of enterotoxin results                                                                             53                                                                                                                                    

CHAPTER FOUR: SUMMARISING DISCUSSION AND CONCLUSION                                                    55    

 4.1.1 Identification of coagulase negative species                                                                         55 

 4.1.2 Staphylococcal enterotoxin potential of CNS species                                                           56 

4.2 Conclusions                                                                                                                                            58 

4.3 Limitations of the present study                                                                                                              59        

4.4 Future Work                                                                                                                                            59                                                                                            

CHAPTER FIVE: LIMITATIONS                                                                                                                   56 

REFERENCES                                                                                                                                             60            

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



vii 
 

List of Figures 

 
Fig. 1. Baird-Parker agar plate showing typical Staphylococcus aureus colonies (2A) with a 
visible clearing or halo around the black colony (2B, indicated with arrows)……………………..16  
 
Fig. 2. Baird-Parker agar plate showing non-typical black colonies (Staphylococcus epidermidis) 
with no visible clearing or halo…………………………………………………………………………17  
 
Fig. 3. Two Biochemical identification test kits used to identify the 42 Staphylococcus species.  
The BBL™ Crystal™ Gram Positive identification system (A, B), and the Remel RapID ™ Staph 
Plus identification system (C, D)………………………………………………………………………19 
 
Fig. 4. The BBL™ Crystal™ Gram positive identification system showing a typical result for a S. 
aureus species (A) and a coagulase negative species (S. saprophyticus, B)……………………20 
 
Fig. 5. Remel RapID ™ Staph Plus identification system showing a typical result for a S. aureus 
species (A) and a coagulase negative species (S. saprophyticus, B)…………………………….21 
 
Fig. 6.  The basic principle of Matrix assisted Laser Desorption Ionisation –time of flight mass 
spectrometry (MALDI-TOF MS)………………………………………………………………………23 
 
Fig. 7. The laboratory bench Microflex LT MALDI-TOF MS set-up (A) and the target plate (B) 
used for placing samples for analysis………………………………………………………………...25 
 
Fig. 8.  The Implen NanoPhotometer® P300 used to quantify DNA of the 7 isolates sequenced 
using 16S rDNA…………………………………………………………………………………………27 
 
Fig. 9. The complete Ridascreen® SET A, B, C, D, E immunoassay test kit..............................29 
 
Fig. 10. The Techne TC-512 thermal cycler used to generate PCR products for SE PCR, 
multiplex PCR and 16S rDNA sequencing…………………………………………………………...31 

 

Fig. 11. The Invitrogen E-Gel® system used for SE PCR analysis……………………………….32 

 
Fig. 12. The UVitec Essential V2 used to photograph the gels generated using both the E-Gel 
system for the SE PCR results and the traditional method for the multiplex PCR results………33 

 
Fig. 13. An MSP dendrogram generated by the Bruker Daltonics MALDI Biotyper 3.0 Offline 
Classification software for seven CNS isolates, for which their identification did not agree among 
the three methods BBL™ Crystal™ Gram Positive identification system, Remel RapID™ Staph 
Plus and MALDI−TOF MS. Isolate identifications based on 16S rDNA sequencing are as 
follows: isolates 1 and 2 – S. epidermidis, isolates 3, 4 and 7 – S. pasteuri, isolate 5 – S. 
saprophyticus, isolate 6 – S. scuiri subsp. sciuri…………………………………………………….43                                                                                                                                             
                                                                 
Fig. 14. A phylogenetic tree generated by MegAlign using Clustal V alignment of the seven CNS 
isolates for which the identification did not agree among the three methods (BBL™ Crystal™ 
Gram Positive identification system, Remel RapID™ Staph Plus and MALDI−TOF MS). 
Bootstrap values > 70% are shown. Isolate identifications based on 16S rDNA sequencing are 
as follows: isolates 1 and 2 – S. epidermidis, isolates 3, 4 and 7 – S. pasteuri, isolate 5 – S. 
saprophyticus, isolate 6 – S. scuiri subsp. sciuri…………………………………………………….44                                                                         
 
Fig. 15. Gel electrophoresis of CNS isolates 1-14 screened for SEA-SEE genes using multiplex 
PCR……………………………………………………………………………………………………….50      
 

Fig. 16. Gel electrophoresis of CNS isolates 15-28 screened for SEA-SEE genes using 

multiplex PCR……………………………………………………………………………………………51 

 



viii 
 

Fig. 17. Gel electrophoresis of CNS isolates 29-42 screened for SEA-SEE genes using 

multiplex PCR…………………………………………………………………………………………...52                                          

                      

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



ix 
 

List of Tables 

Table 1 Identification of 42 CNS isolates using three methods, i.e. the BBL™ Crystal™ Gram 

Positive identification system, Remel RapID™ Staph Plus and MALDI−TOF MS……. 38   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



1 
 

Chapter One: Introduction 

 

1.1 The Genus Staphylococcus 

Staphylococcus species are non-motile, non-sporeforming, encapsulated, Gram-

positive, catalase-positive, facultatively anaerobic, chemorganotrophic cocci with a 

respiratory and fermentative metabolism at an optimal temperature of 37°C (Le Loir 

et al. 2003). Staphylococcal cells form spherical single cells, paired cocci, short 

chains or grape-like clusters (Le Loir et al. 2003) and produce opaque white to cream 

coloured colonies that are 2–3 mm in diameter on uncrowded areas of blood agar. 

 

The Staphylococcus genus is widely distributed in nature, and is found living on the 

skin and mucous membranes of warm-blooded mammals and in environmental 

sources, i.e. soil, sand, air and water, and is also isolated from protein-based foods, 

i.e. meat, cheese and milk (Irlinger 2008; Even et al. 2010). 

 

Within the genus Staphylococcus, there are 42 known species and 24 subspecies 

(Vasconcelos & Cunha 2010). Various phenotypic and genotypic characteristics, 

including nuclease production, antibiotic resistance, i.e. methicillin or novobiocin, and 

DNA similarity, are often used to classify Staphylococcus species. However, the 

classification of the 42 known Staphylococcus species is predominantly based on the 

production of the enzyme, coagulase, into two distinct groups: coagulase-positive 

Staphylococcus (CPS) and coagulase-negative Staphylococcus (CNS). 

 

Staphylococcus species possess numerous virulence factors, including lipases, 

proteases, coagulase, haemolysins (α, β, γ and δ), deoxyribonuclease (DNase), 

leukocidin, exfoliative toxins A and B, toxic shock syndrome toxin 1 (TSST-1) and a 

family of emetic pyrogenic superantigens (SAgs) (Bartolomeoli et al. 2009). The 

ability to produce coagulase is typically associated with another virulence factor, the 

production of enterotoxins. However, CNS isolates, on occasion, have also been 

shown to produce enterotoxins (Zell et al. 2008). 

 

 



2 
 

1.2 Coagulase-negative Staphylococcus (CNS) 

 

Species of CNS are commonly associated with the cutaneous system (skin and 

mucosal membranes) of the human body (Vasconcelos & Cunha 2010). Therefore, 

CNS isolates are frequently encountered in human clinical samples (Gaillot et al. 

2000). However, these coagulase-negative species are regarded as either a 

contaminant or a natural part of the human microbiota (Kloos & Bannerman 1994; 

Çitak et al. 2003) and not the primary pathogen causing an illness. 

 

CNS strains can also be used in the food processing industry, contributing to flavour, 

aroma formation and pigmentation/colour functionalities, particularly in fermented 

foods (e.g. cheeses and sausages), and include the non-enterotoxin-producing 

Staphylococcus carnosus and Staphylococcus xylosus (Irlinger 2008; Coton et al. 

2010; Even et al. 2010; Podkowik et al.2013). The use of these strains of 

staphylococci in food processing is largely historical; for example, for over 60 years, 

S. carnosus has been used alone or in combination with other microorganisms 

(pediococci or lactobacilli) as a starter culture for the production of raw sausage (Zell 

et al. 2008). The true extent of the potential and actual virulence of CNS species has 

not been fully investigated. Therefore, CNS has largely been overlooked as a 

potential human pathogen, both clinically and in the food industry. 

 

1.3 Staphylococcal Enterotoxins (SEs) 

 

SEs are heat-stable low molecular weight (~ 22−28 kDa) extracellular proteins that 

belong to a large family of staphylococcal and streptococcal pyrogenic toxic SAgs 

(Argudín et al. 2010). SEs can cause toxic-shock-like syndromes and have been 

implicated in food poisoning incidences (Balaban & Rasooly, 2000). Therefore, SEs 

can function as both potent gastrointestinal toxins and SAgs (Balaban & Rasooly, 

2000). 

 

Enterotoxins are produced throughout the logarithmic growth phase or during the 

transition from the exponential phase to the stationary phase (Derzelle et al. 2009; 



3 
 

Argudín et al. 2010). These toxins are toxic in high nanogram to low microgram 

quantities (Larkin et al. 2009; Argudín et al. 2010) and are resistant to conditions that 

easily destroy the bacteria that produce them, i.e. heat treatment, low pH and 

proteolytic enzymes. As a result, they retain their activity in the digestive tract after 

ingestion (Argudín et al. 2010). SEs are produced when the water activity is > 0.86 

and the temperature is between 10 and 46°C (Colombari et al. 2007). The optimum 

temperature range for the production of the maximum level of toxin is 40–45°C 

(Colombari et al. 2007). 

 

SEs are named according to their emetic activities, i.e. those that induce emesis 

following oral administration in a primate model (Le Loir et al. 2003; Derzelle et al. 

2009; Argudín et al. 2010; Podkowik et al. 2013). These include SEA, SEB, SEC (C1, 

C2, C3, ovine and bovine variants), SED, SEE, SEG, SEH, SEI, SER, SES and SET 

(Balaban & Rasooly 2000; Derzelle et al. 2009; Vasconcelos & Cunha 2010). 

Enterotoxins that do not have emetic properties or have not yet been tested are 

designated as staphylococcal-enterotoxin-like superantigen (SEL) proteins and 

include SELJ, SELK, SELL, SELM, SELN,SELO, SELP, SELQ, SELU and SELV 

(Derzelle et al. 2009; Argudín et al. 2010; Vasconcelos & Cunha 2010). The toxic 

shock staphylococcal toxin, TSST-1, initially designated as SEF, lacks emetic activity 

and is therefore not classified as a true enterotoxin (Argudín et al. 2010). 

 

1.4 Presence in Food 

 

Staphylococcus is the third most common pathogen responsible for outbreaks of food 

poisoning worldwide (Atanassova et al. 2001; Veras et al. 2008; Podkowik et al. 

2013). The main sources of staphylococcal food contamination are: 

 

 improper storage and handling of food contaminated with staphylococci (Udo 

et al. 1999;  Atanassova et al. 2001; Le Loir et al. 2003; Veras et al. 2008; 

Bartolomeoli et al. 2009; Argudín et al. 2010);  

 the use of birds and mammals, i.e. cattle, sheep and goats, that either are 

non-symptomatic staphylococci carriers or are actively infected (i.e. mastitis) 



4 
 

and their products in the manufacture of food (Atanassova et al. 2001; Le Loir 

et al. 2003; Bartolomeoli et al. 2009; Hennekinne et al. 2009; Argudín et al. 

2010).  

 

As staphylococcal species are a common commensal of the skin and mucous 

membranes, i.e. the nose and hands of humans, food handlers can transmit 

staphylococci via manual contact or through respiratory secretions (Argudín et al. 

2010; Pinchuk et al. 2010). Transmission of staphylococci can occur during the 

manufacture or cooking of foods (Hennekinne et al. 2009). Therefore, this type of 

transmission is largely associated with cooked or processed foods (Argudín et al. 

2010).  

 

Le Loir et al. (2003) noted that, in all food poisoning cases related to staphylococci, 

the food or one of the ingredients was contaminated with an SE-producing strain and 

was exposed to temperatures that allow the bacteria to grow for an extended period 

of time. Exposure to these inappropriate temperatures is a result of the refrigeration 

process failing or because a growth-permissive temperature is required during 

processing, e.g. cheesemaking (Le Loir et al. 2003). 

 

Dairy animals with subclinical staphylococcal mastitis are a primary source of 

contamination, which can result in large numbers being shed into dairy-derived 

products or ingredients (Bartolomeoli et al. 2009; Hennekinne et al. 2009; Argudín et 

al. 2010). Staphylococci can also be transferred to foods via air, dust and food 

contact surfaces (Argudín et al. 2010). SEs can be produced by staphylococci in milk 

or milk products prior to pasteurisation and, although the bacterial cells will be 

destroyed by this process, the enterotoxins can retain their toxic activity (Bartolomeoli 

et al. 2009). SEs are regarded as heat resistant, with the temperature required for the 

thermal destruction curve to traverse one log cycle (z value) ranging from 25 to 33°C 

and the D121°C value (time in minutes at 121°C for 90% destruction) ranging from 8.3 

to 34 min (Erickson 2003). However, differences in heat resistance and stability do 

exist among enterotoxins, i.e. SEC > SEB > SEA and SEE > SEA > SEI, respectively 

(Erickson 2003). Therefore, heat stability depends on the medium in which the toxin 



5 
 

is present, the pH, the salt concentration and other environmental factors related to 

the level of toxin denaturation (Balaban & Rasooly 2000). In fact, SEs can show 

biphasic (faster first initial phase and slower second phase) inactivation at different 

pHs, and several constituents of foods (e.g. proteins, carbohydrates and chitin) can 

increase the thermal stability of SEs by up to fivefold, e.g. the time required to 

inactivate SEB in milk (30 μg/mL) at 100°C is 1.3 times greater than that in Veronal 

buffer (Erickson 2003). 

 

Staphylococcal food poisoning is the consumption of food containing sufficient 

amounts of one or more enterotoxins to result in intoxication (Wieneke et al. 1993; 

Argudín et al. 2010). The symptoms have a rapid onset (2–4 h) and include nausea, 

vomiting and abdominal cramping with or without diarrhoea (Atanassova et al. 2001; 

Le Loir et al. 2003; Argudín et al. 2010). This disease is usually self-limiting and the 

symptoms typically resolve within 24−48 h after onset (Le Loir et al. 2003; Argudín et 

al. 2010). Incidences that require hospitalisation are normally associated with infants 

or the elderly (Argudín et al. 2010). 

 

Foods often associated with staphylococcal contamination and intoxication include 

meat and meat products, poultry and egg products, milk and dairy products, salads, 

bakery products (particularly cream-filled pastries and cakes) and sandwich fillings 

(Veras et al. 2008; Argudín et al. 2010). Therefore, the foods at greatest risk are 

those in which the normal bacterial flora has been destroyed, e.g. cooked meats, or 

inhibited, e.g. cured, salted meats (Institute of Environmental Science and Research 

Limited 2001). Staphylococci therefore grow well in cooked foods that are high in 

protein, sugar or salt, foods that are low in acid or foods with moist fillings (Institute of 

Environmental Science and Research Limited 2001). However, the origins of 

staphylococcal food poisoning differ widely among countries. Le Loir et al. (2003) 

suggest that this is due to the differences in consumption and food habits in different 

countries.  

 

SEA alone or together with other SEs/SELs is the most commonly reported 

enterotoxin in food and is considered to be the main cause of Staphylococcus aureus 



6 
 

food poisoning (Vernozy-Rozand et al. 2004; Veras et al. 2008; Argudín et al. 2010). 

For example, in Japan in 2000, SEA was implicated in a food poisoning outbreak in 

low fat milk and drinking yoghurt produced from contaminated skim milk powder 

(Asao et al. 2003). When SEA, presumably produced before pasteurisation, was 

heated three times at 130°C for 2 or 4 s, it still retained immunological and biological 

activities (Asao et al. 2003). SEB, SEC and SED alone have also been implicated in 

staphylococcal food poisoning throughout the world (Argudín et al. 2010).  

 

1.5 Enterotoxigenic Potential of CNS Species in Food 

 

CNS are often associated with food. For example, all Staphylococcus species 

isolated from dried cured Iberian ham and from slightly fermented sausages (fuet, 

chorizo and salchichón) were CNS (Rodríguez et al. 1996; Martin et al. 2006), with S. 

xylosus being the main species in both these studies.   

 

Several recent studies have shown that some CNS strains do have the ability to 

produce enterotoxins. For example, a human isolate of Staphylococcus warneri 

(CCRC 12929) was shown to produce both SEA and SED (Chou & Chen 1997). 

Mhlambi et al. (2010) and Naidoo & Lindsay (2010) isolated CNS species from the 

dried meat biltong and showed that they produced enterotoxins. Furthermore, CNS 

isolated from various foods, including milk and cheese, were shown to possess not 

only the SEA gene but also the SEC1 gene (Cunha et al. 2006). Although CNS may 

carry SE genes, often enterotoxins either are not produced in vitro or their production 

is not detected using tests such as enzyme-linked immunosorbent assay and 

reversed passive latex agglutination (Rodríguez et al. 1996; Cunha et al. 2006; Veras 

et al. 2010). This could be due to the toxin being produced by CNS below the 

detection limit of these methods or the non-expression of these SE genes (Cunha et 

al. 2006).  

 

CNS strains that produce enterotoxins have also been found in dairy and related 

products. For example, CNS were isolated from goats’ milk, whey and cheese (lactic 

and Camembert) (Vernozy-Rozand et al. 1996). S. xylosus was predominant in 



7 
 

goats’ milk, Staphylococcus saprophyticus was predominant in whey, 

Staphylococcus simulans was predominant in lactic cheese and S. xylosus/equorum 

was predominant in Camembert-type cheese (Vernozy-Rozand et al. 1996). Eleven 

out of a total of 187 CNS isolated from goats’ milk, whey and cheese products were 

enterotoxigenic, producing SEE (Vernozy-Rozand et al. 1996). SEA, SED, SEE and 

SEH were produced by 35 CNS strains isolated from fermented fish, meat starter 

culture, sausage, soy, cured ham, cheese (hard and soft) and water buffalo raw milk 

(Zell et al. 2008). Amongst these CNS strains, SED and SEH were identified as the 

most common toxins and, of the 35 isolates tested, these enterotoxins were 

produced by five and eight isolates respectively (Zell et al. 2008). CNS strains 

isolated from Minas soft cheese included S. saprophyticus, S. warneri, 

Staphylococcus epidermidis, S. xylosus, Staphylococcus haemolyticus and 

Staphylococcus schleriferi subsp. schleiferi (Rall et al. 2010). The gene coding for 

SEA was the most frequently found (18.5%), with genes for SEB and SEC also being 

present (Rall et al. 2010). However, none of the CNS isolates isolated from Minas 

soft cheese produced any of the enterotoxins (SEA, SEB, SEC or SED) in vitro (Rall 

et al. 2010). CNS isolated from sheep milk have been shown to produce SEA, SEC, 

SED and combinations of SEs, i.e. SECD and SEBCD (Bautista et al. 1988). The 

predominant SE produced by these CNS was SED (Bautista et al. 1988).   

 

In addition, CNS have been isolated from foods typically associated with CPS food 

poisoning, e.g. bakery goods, milk, cheese, sandwiches and pork (Cunha et al. 

2006). In some instances, they have also been implicated as being responsible for 

food poisoning outbreaks. For example, CNS isolates were responsible for a food 

poisoning outbreak associated with unpasteurised milk in Brazil and involved SEC 

and SED (Do Carmo et al. 2004). CNS strains isolated from Minas soft cheese 

included S. saprophyticus, S. warneri, S. epidermidis, S. xylosus, S. haemolyticus 

and S. schleriferi subsp. schleiferi (Veras et al. 2008). The most frequently identified 

SE gene produced by these CNS was sea (18.5%) (Veras et al. 2008). Therefore, 

CNS could play a greater role in SE food poisoning outbreaks than presently thought. 

 

1.6 New Zealand 



8 
 

 

SE outbreaks have been reported in New Zealand, with yoghurt, hot ham 

sandwiches, Christmas hams and freeze-dried trampers’ meals being implicated 

(Institute of Environmental Science and Research Limited 2001). However, research 

on the presence of S. aureus in New Zealand has largely centred on strains from 

clinical and bovine origin (Jarvis & Lawrence 1970, 1971; Grinberg et al. 2004) and 

their enterotoxin production (Jarvis & Lawrence 1970, 1971). Jarvis & Lawrence 

found that 11 of 35 clinical Staphylococcus strains were enterotoxigenic and noted 

that this was similar to that reported in the United States. However, the predominant 

enterotoxin was SEB (seven of the 35 clinical strains (20%)), compared with SEA 

being reported as the leading cause of food poisoning in the United States at the time 

(Jarvis & Lawrence 1970). Although this same study found that no bovine strains 

produced SEA, SEB or SEC, it was limited in terms of the number of enterotoxins 

tested for, with only SEA, SEB and SEC being investigated. This may have been 

because these were the only SEs identifiable at the time. SED was mentioned but 

not tested for. As it also appears that strains were not identified to the species level, 

CNS species may have been included. 

 

Howard (2006) identified and enumerated mastitis pathogens in bulk tank milk from 

New Zealand dairy cows. Those identified included aesculin-positive streptococci, 

coliforms, S. aureus and CNS species. The CNS group was present in 51% of bulk 

tank milk samples at greater than 1000 CFU/mL compared with 49, 11 and 0% for 

streptococci, coliforms and S. aureus respectively. These results indicate that CNS 

species are found associated with dairy cows and these results suggest that the use 

of this raw milk is a potential source of CNS contamination and therefore that 

enterotoxin production could occur in the manufacturing plant or in the final product if 

good manufacturing practices are not adhered to. However, this study identified and 

differentiated S. aureus isolates from CNS only by their ability to produce coagulase, 

catalase and haemolytic colonies. As no biochemical tests or proteomic or molecular 

methods were used, it is possible that S. aureus isolates were misidentified as other 

coagulase-positive species or CNS and vice versa. 

 



9 
 

In 2010 Hudson & On reviewed the methods that are used to evaluate CPS and their 

production of enterotoxin in dairy products and that could be applied by small 

businesses producing raw milk cheeses in New Zealand. However, to date, no 

research investigating CNS species in New Zealand and their presence in food has 

been reported. Therefore, the question remains as to the presence of CNS in milk 

and dairy products in New Zealand and their potential to produce enterotoxins. 

Numerous phenotypic and genotypic methods are available to identify and 

characterise staphylococci and their enterotoxins. However, the rapid method, 

matrix-assisted laser desorption ionisation−time of flight mass spectrometry 

(MALDI−TOF MS), has been suggested as a novel bacterial identification 

methodology that could contribute greatly to improving the study of CNS and their 

enterotoxin ability. 

 

1.7 Protein Profiling − A Novel and Rapid Method for Identifying CNS and their 

Enterotoxins 

 

MALDI−TOF MS detects and examines the pattern of proteins, predominantly 

ribosomal proteins, directly from samples such as intact bacteria (Dupont et al. 2010) 

and, based on their mass-to-charge ratio (m/z), a reproducible spectrum consisting of 

a series of peaks is produced within minutes (Carbonnelle et al. 2007; Dupont et al. 

2010).  

 

Traditionally, bacterial identification has been based on phenotypic methods, i.e. 

Gram-staining and biochemical test patterns, and genotypic methods, i.e. rpoB and 

sodA gene sequencing.  Although some of these methods are rapid, complete 

identification by many traditional methods can take from hours to days (Seng et al. 

2009). MALDI−TOF MS was first proposed as a method for identifying bacteria over 

30 years ago (Anhalt & Fenselau 1975; Seng et al. 2009). However, it is only now 

gaining ground as a rapid, inexpensive, reliable and accurate method for bacterial 

identification (Rajakaruna et al. 2009; Seng et al. 2009; Dubois et al. 2010). This re-

emergence is due to the advances in the technology, improvements in the sample 

preparation and analysis of larger biomolecules (Croxatto et al. 2012; Wieser et al. 



10 
 

2012). MALDI−TOF MS could therefore replace these phenotypic methods and could 

rival genotypic methods as the gold standard for the identification of bacteria 

(Carbonnelle et al. 2007).  

 

The incidence of hospital-acquired infection linked to CNS is increasing and, with 

this, there is an increase in the need for an accurate identification of staphylococci at 

the species level (Dupont et al. 2010). Most commercial identification kits remain 

unable to differentiate between the CNS species (Dupont et al. 2010). Therefore, the 

identification of Staphylococcus species is a prime example of where MALDI−TOF 

MS is applicable. 

 

In the last decade, there has been an increase in research into the use of 

MALDI−TOF MS systems in the identification of bacteria, e.g. Keys et al. (2004), 

Ferreira et al. (2011) and Carbonnelle et al. (2012). A large number of studies have 

also used this method to identify Staphylococcus species (Bernardo et al. 2002; 

Carbonnelle et al. 2007; Rajakaruna et al. 2009; Dubois et al. 2010; Dupont et al. 

2010; Fox et al. 2010, 2011). These studies compared the ability of MALDI−TOF MS 

to identify staphylococci with the abilities of more traditional phenotypic methods 

(manual and automated) and/or genotypic methods (sodA gene and 16S ribosomal 

ribonucleic acid (rRNA) sequencing). The MALDI−TOF MS method achieved results 

that were comparable with those from more traditional methods (Rajakaruna et al. 

2009; Fox et al. 2010; Dubois et al. 2010; Carbonnelle et al. 2012). Dupont et al. 

(2010) suggested that the MALDI−TOF MS method out-performed the more 

traditional methods, and it was shown to be more sensitive in the identification of 

staphylococcal isolates to the species level (Carbonnelle et al. 2007; Fox et al. 2010; 

Dubois et al. 2010; Dupont et al. 2010; Carbonnelle et al. 2012). In one study by 

Dupont et al. (2010), MALDI−TOF MS was compared with Phoenix Microbiology, an 

automated nephelometry system, and Vitex-2, an automatic colorimetric system. The 

correct identification was obtained for 97.4% of clinical CNS isolates identified by 

MALDI−TOF MS compared with 79 and 78.6% for Phoenix Microbiology and Vitek-2 

respectively. In the study by Dubois et al. (2010), 99.3% of clinical and environmental 

(food and plant) staphylococcal isolates were correctly identified to the species level. 



11 
 

These studies suggest that MALDI−TOF MS is a highly sensitive method that could 

be used routinely to identify CNS to the species level. 

 

Fox et al. (2011) noted that the identification of CNS from environmental samples to 

species level is not adequately achievable with either traditional methods or 

MALDI−TOF MS. However, these authors described a method using MALDI−TOF 

MS and one-dimensional gel electrophoresis that improved the identification of 

environmental isolates, including Staphylococcus species.  

 

MALDI−TOF MS has already been shown to identify bacteria at the genus, species 

and subspecies level as part of protein profiling and bacterial proteomics. It could 

also be used to detect recombinant proteins, characterise targeted and unknown 

proteins, analyse bacterial RNA and DNA and detect virulence factors, including 

enterotoxins. Methods for use with MALDI−TOF MS to detect enterotoxins have 

already been studied (Brun et al. 2007; Dupuis et al. 2008) and could replace the 

immunological methods that are currently used to detect enterotoxins. This could 

improve investigations into staphylococcal food-borne poisoning outbreaks (Dupuis 

et al. 2008). This could also mean that MALDI−TOF MS could be used to identify as 

yet unidentified or previously unidentifiable SEs. 

 

 

 

 

 

 

 

 

 

 

 



12 
 

1.8 Aims of this Investigation 

The purpose of this study was to isolate, identify and characterise CNS from New 

Zealand milk and dairy products, and to evaluate their toxin-producing potential. 

 

1.8.1 Isolation of CNS from New Zealand milk and dairy products 

Objectives  

a) To isolate CNS bacteria from a variety of processed dairy products. 

b) To select and optimise a suitable method(s) for enumerating CNS from dairy 

products. 

 

1.8.2 Identification and characterisation of CNS from New Zealand milk and 

dairy products  

Objective  

a) To select and optimise suitable phenotypic and genotypic methods for identifying 

CNS at the species level: for example, biochemical tests, i.e. BBL™ Crystal™ 

Gram Positive identification system (BD Diagnostic Systems, Becton Dickinson, 

Franklin Lakes, NJ, USA) and Remel RapID™ Staph Plus (Oxoid, Basingstoke, 

UK), protein profiling, i.e. MALDI–TOF MS (Bruker Daltonik GmbH), and 

molecular methods, i.e. 16S rDNA seqeuncing. 

 

1.8.3 Evaluation of the toxin-producing potential of CNS isolated from dairy 

products 

Objectives  

a) To select and optimise a suitable method(s) for determining if CNS have the 

genes present to produce enterotoxins. 

b) To test the enterotoxin-producing ability of selected CNS strains in dairy 

products. 

 



13 
 

Chapter Two: Materials and Methods 

 

2.1 General Requirements 

 

The following materials and reagents were obtained from Oxoid, Basingstoke, UK:  

tryptone soy broth (TSB), tryptone soy agar (TSA) and Remel RapID™ Staph Plus 

biochemical test kits. 

 

Baird−Parker agar (BP) plates and peptone diluents were obtained from Fort Richard 

Laboratories, Auckland, New Zealand. 

 

BBL™ Crystal™ Gram Positive identification systems were obtained from BD 

Diagnostic Systems, Becton Dickinson, Franklin Lakes, NJ, USA. 

 

α-Cyano-4-hydroxy-cinnamic acid was obtained from Bruker Daltonik GmbH, 

Bremen, Germany. 

 

The standard solution (acetonitrile 50%, water 47.5% and trifluoroacetic acid 2.5%) 

for use in the MALDI−TOF method was obtained from Fluka Analytical, Buch, 

Switzerland. 

 

The following were obtained from Merck KGaA, Darmstadt, Germany: acetonitrile 

and ethanol. 

 

Formic acid was obtained from Fisher Scientific, Loughborough, Leicestershire, 

England. 

 

PCR grade water was obtained from Roche Diagnostics, Mannheim, Germany. 

 

The following were obtained from Zymo Research, Irvine, CA, USA: Quick-gDNA™ 

MiniPrep method and DNA Clean & Concentrator™-5 kit.  

 



14 
 

The following materials and methods were obtained from Invitrogen, Carlsbad, CA, 

USA: 1 Kb Plus DNA Ladder, Platinum® Multiplex PCR Master Mix and E- 

Gel® EX 2% gels. 

 

The primer sets for the 16S rDNA sequencing (U1392R and Bac27F), SE PCR 

(SEAfor, SEArev, SEBfor, SEBrev, SECfor, SECrev, SEDfor, SEDrev, SEHfor, 

SEHrev) and the multiplex SE PCR (SEA-F, SEA-R, SEB-F, SEB-R, SEC-F, SEC-R, 

SED-F, SED-R, SEE-F and SEE-R) were obtained from Life Technologies, Penrose, 

Auckland, NZ. 

 

The DNA loading buffer (EZ-Vision™ Two DNA Dye and Buffer 6x) was obtained 

from AMRESCO Incorporated, Solon, OH, USA. 

 

The Tris-Borate-EDTA (TBE) 10X Solution was obtained from Fischer Scientific, Fair 

Lawn, New Jersey, USA. 

 

Bio-Rad certified molecular biology agarose was obtained from Bio-Rad 

Laboratories, Hercules, CA, USA. 

 

The Ridascreen® Set A, B, C, D, E immunoassay kits were obtained from R-

Biopharm, Darmstadt, Germany. 

 

 

 

 

 

 

 

 

 

 

 



15 
 

2.2 General Methods  

2.2.1 Collection of samples and bacterial isolation 

As Staphylococcus species are naturally associated with milk, BP plates from a 

routine dairy testing laboratory and dairy products (milk powder, cheese and butter) 

from various dairy manufacturing plants from around New Zealand were collected. 

For the dairy products, samples were processed according to ISO 8261 / IDF 122: 

2001 (E) (International Organization for Standardization 2001) and then pre-

incubated at 37°C for 18 h according to the method described by Silva et al. (2005). 

Single colonies were then isolated by spread plating each sample on to BP plates 

(incubated at 37°C overnight). A proportion of colonies (at least 10% of the colonies 

present on each plate) were selected and grown in TSB overnight at 37°C and stock 

cultures were prepared (stored at −80°C). For colonies originating from BP plates, 

typical (Fig. 1) and non-typical (Fig. 2) Staphylococcus (black, with or without halos) 

were selected and cultured in TSB overnight at 37°C and then frozen stock cultures 

were prepared (stored at −80°C). In this way, 460 bacterial isolates were obtained for 

further study.  

 

 

 

 

 

 

 

 

 

 



16 
 

 

Figure 1 Baird-Parker agar plate showing typical Staphylococcus aureus colonies (1A) with a visible 
clearing or halo around the black colony (1B, indicated with arrows).  

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A B 



17 
 

 

Figure 2. Baird-Parker agar plate showing non-typical black colonies (Staphylococcus epidermidis) 

with no visible clearing or halo.  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



18 
 

 

2.2.2 Positive and negative controls  

The reference strain Staphylococcus aureus subsp. aureus NZRM 917 (New Zealand 

Reference Culture Collection Medical Section, ESR Kenepuru Science Centre, 

Porirua, New Zealand) was used as a known and/or negative control for the 

identification methods.  

 

The positive control isolates, S. aureus S27 (SEA producer), S. aureus S12 (SEB 

producer) and S. aureus S34 (SED producer) were obtained from the Food 

Assurance culture collection (Fonterra Research and Development Centre, 

Palmerston North, New Zealand) and were used as positive controls for the toxin 

tests and the multiplex toxin PCR method. Their toxin production status was 

determined by the Institute of Environmental Science and Research. 

 

The positive control isolates 3376 (SEC producer) and 3377 (SEE producer) were 

obtained from the Institute of Environmental Science and Research and were used 

as positive controls for the multiplex toxin PCR method. 

 

2.2.3 Phenotypic identification 

Bacterial isolates were identified using the BBL™ Crystal™ Gram Positive 

identification system (Fig. 3A & Fig. 4) and Remel RapID™ Staph Plus biochemical 

test kits (Fig. 3C & Fig. 5), according to the manufacturers’ instructions. 

 

 

 

 

 



19 
 

  

         

Figure 3. Two Biochemical identification test kits used to identify the 42 Staphylococcus species.  The 
BBL™ Crystal™ Gram Positive identification system (A, B), and the Remel RapID ™ Staph Plus 
identification system (C, D).  

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A

s

w

o

ll

e

n

 

r

o

u

n

d

 

li

g

a

m

e

n

t

, 

i

n

g

u

i

n

a

l 

h

e

r

n

i

a 

C

s

w

o

ll

e

n

 

r

o

u

n

d

 

li

g

a

m

e

n

t

, 

i

n

g

u

i

n

a

l 

h

e

r

n

i

a 

B

s

w

o

ll

e

n

 

r

o

u

n

d

 

li

g

a

m

e

n

t

, 

i

n

g

u

i

n

a

l 

h

e

r

n

i

a 

D

s

w

o

ll

e

n

 

r

o

u

n

d

 

li

g

a

m

e

n

t

, 

i

n

g

u

i

n

a

l 

h

e

r

n

i

a 



20 
 

 

 

 

Figure 4.  The BBL™ Crystal™ Gram Positive identification system showing a typical result for a S. 
aureus species (A) and a coagulase negative species (S. saprophyticus, B). 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A

s

w

o

ll

e

n

 

r

o

u

n

d

 

li

g

a

m

e

n

t

, 

i

n

g

u

i

n

a

l 

h

e

r

n

i

a 

B

s

w

o

ll

e

n

 

r

o

u

n

d

 

li

g

a

m

e

n

t

, 

i

n

g

u

i

n

a

l 

h

e

r

n

i

a 



21 
 

 

Figure 5.  Remel RapID ™ Staph Plus identification system showing a typical result for a S. aureus 
species (A) and a coagulase negative species (S. saprophyticus, B). 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A

s

w

o

ll

e

n

 

r

o

u

n

d

 

li

g

a

m

e

n

t

, 

i

n

g

u

i

n

a

l 

h

e

r

n

i

a 

B

s

w

o

ll

e

n

 

r

o

u

n

d

 

li

g

a

m

e

n

t

, 

i

n

g

u

i

n

a

l 

h

e

r

n

i

a 



22 
 

2.2.4 MALDI−TOF MS analysis 

 

The basic principles of MALDI-TOF MS analysis are shown in Fig. 6. In general 

MALDI-TOF MS analysis involves a soft ionisation mechanism of proteins (Clark et 

al. 2013). Soft ionisation allows analysis of large biomolecules, including ribosomal 

proteins of up to 100kDa (Clark et al. 2013).  In order for this mechanism to work the 

direct smear or purified protein of a sample is saturated with the matrix, a low-mass 

organic compound (Clark et al. 2013). The matrix is essential for this mechanism as it 

acts as a scaffold, allowing ionisation to occur and supplies protons for the ionisation 

of the sample (Clark et al. 2013). The drying of the sample and matrix enables them 

to co-crystallise so that the sample is embedded into the matrix (Clark et al. 2013).  

 

The crystallised sample-matrix is irradiated using a UV laser beam, with the intensity 

determined by a beam attenuator that is standardised according to the 

manufacturer’s settings (Clark et al. 2013). The photons from the laser and matrix 

molecules take up the energy from the beam, and this interaction triggers a 

sublimation of the matrix into a gas phase (Clark et al. 2013). This forms a plume 

which then results in the ionisation of the sample (Clark et al. 2013). 

 

Following ionisation the proteins within the sample are analysed by the mass 

analyser. This characterises the composition of the sample as a spectrum of mass-

to-charge (m/z) ratios (Clark et al. 2013).These ratios are electrodynamic 

measurements of how quickly ions from the sample move through the time of flight 

(TOF) tube and reach the detector (Clark et al. 2013).  These spectra are then 

compared to a database of defined reference spectra, allowing the identification of 

the micro-organism (Clark et al. 2013). 

 

 

 

 

 



23 
 

 

Figure 6.  The basic principle of Matrix assisted Laser Desorption Ionisation –time of flight mass 

spectrometry (MALDI-TOF MS). 

(http://www.mayomedicallaboratories.com/articles/communique/2013/01-maldi-tof-mass-

spectrometry/index.html) 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



24 
 

 

For the MALDI-TOF MS analysis using the Bruker system, cells from a single colony 

from an overnight streak plate on TSA were used for each isolate, according to either 

the direct method or the extraction method, as outlined in the manufacturer’s 

instructions. The direct method was used as an initial screen for each colony. Briefly, 

the colony was added to the MALDI target plate (Fig. 7B), air dried, overlaid with 1 µL 

of matrix, air dried again and then processed in the Microflex LT (Bruker Daltonik 

GmbH, Bremen, Germany) mass spectrometer (Fig. 7A) using the Flex Control 

program.  

 

For the extraction method, a loopful of colony was suspended in sterile demineralised 

water and 100% ethanol. The proteins were extracted using an equal volume of 70% 

formic acid and 100% acetonitrile, and the resulting supernatant was added to the 

MALDI target plate, air dried, overlaid with 1 µL of matrix, air dried again and then 

processed as described above.  

 

The resulting spectra were analysed using the Bruker Daltonics MALDI Biotyper 3.0 

Real Time Classification program and a main spectrum (MSP) dendrogram was 

generated using the Bruker Daltonics MALDI Biotyper 3.0 Offline Classification 

software. 

 

 

 

 

 



25 
 

 

Figure 7. The laboratory bench Microflex LT MALDI-TOF MS set-up (A) and the target plate (B) 
used for placing samples for analysis. 

 

 

 

 

 

 

 

 

 

A

s

w

o

ll

e

n

 

r

o

u

n

d

 

li

g

a

m

e

n

t

, 

i

n

g

u

i

n

a

l 

h

e

r

n

i

a 

B

s

w

o

ll

e

n

 

r

o

u

n

d

 

li

g

a

m

e

n

t

, 

i

n

g

u

i

n

a

l 

h

e

r

n

i

a 



26 
 

2.2.5 Identification by 16S rDNA sequencing 

Isolates were streak plated on to TSA and incubated overnight at 37°C. A heavy 

inoculum of each isolate was suspended in 200 µL of peptone diluent and then 

processed according to the Quick-gDNA™ MiniPrep method (Zymo Research). The 

universal bacterial primer sets, U1392R (5ʹ-ACG GGCGGT GTG TRC-3ʹ) and 

Bac27F (5ʹ-AGA GTT TGA TCM TGG CTC AG-3ʹ), were used for the amplification of 

16S rRNA in combination with PCR Master Mix (Fermentas Life Sciences, 

Burlington, Ontario, Canada; www.fermentas.com), according to the manufacturer’s 

instructions, and yielded a product of approximately 1300 bp. Polymerase chain 

reaction (PCR) amplifications were performed under the following conditions: an 

initial denaturation of template DNA at 94°C for 3 min; a series of 35 cycles 

consisting of denaturation (94°C, 30 s), annealing (60°C, 45 s) and extension (1 min 

30 s, 72°C); a final extension at 72°C for 7 min. The generated PCR products were 

further purified using the DNA Clean & Concentrator™-5 kit (Zymo Research) and 

the resulting DNA concentration was quantified using the NanoPhotometer® P300 

(Implen GmbH, Munchen, Germany) (Fig 8). The purified PCR product was 

sequenced, and the resulting sequences were analysed by BLAST 

(http://www.ncbi.nlm.nih.gov/BLAST/) against 16S rRNA sequences from GenBank 

(GenBank database of the National Center for Biotechnology Information, 

http://www.ncbi.nlm.nih.gov/GenBank/). A phylogenetic tree highlighting the genetic 

similarities of the isolates was constructed using MegAlign (DNASTAR, 

www.dnastar.com) after alignment with Clustal V. Accession numbers for the isolates 

were: JX519584, JX519585, JX519586, JX519587, JX519588, JX519590 and 

JX519591. 

 

 

 

 

 

 

 

 

http://www.fermentas.com/
http://www.ncbi.nlm.nih.gov/BLAST/
http://www.ncbi.nlm.nih.gov/GenBank/
http://www.dnastar.com/


27 
 

 

Figure 8.  The Implen NanoPhotometer® P300 used to quantify DNA of the 7 isolates sequenced 
using 16S rDNA. 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



28 
 

2.2.6 Immunoassay test 

Isolates were grown in TSB overnight at 37°C. At least 10 mL of culture was 

centrifuged for 5 min at 3500 x g. The supernatant was then sterile filtered using a 

0.22 µm syringe filter (Merck Millipore, Darmstadt, Germany) and 100 µL of this 

filtrate was used per well in the assay. The immunoassay test kit, Ridascreen® SET 

A, B, C, D, E (R-Biopharm, Darmstadt, Germany), and associated reagents were 

used as outlined in the manufacturer’s instructions (Fig. 9).  

 

The absorbance was measured at 450 nm, using a Multiskan GO spectrophotometer 

(Fisher Scientific, Loughborough, Leicestershire, England). The threshold value was 

calculated by adding 0.15 to the negative control value. The sample was deemed to 

contain no SEs or SEs at a concentration below the detection limit if the absorbance 

test value was less than the threshold value. The isolate was considered to produce 

SEs if the absorbance test value was greater than or equal to the threshold value. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



29 
 

 

Figure 9.  The complete Ridascreen® SET A, B, C, D, E immunoassay test kit. (http://www.r-
biopharm.com/wp-content/uploads/de/news/lebens-futtermittelanalytik/neues-produkt-ridascreen-set-total-
10885/ridascreen-SET-total-de.jpg) 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

http://www.r-biopharm.com/wp-content/uploads/de/news/lebens-futtermittelanalytik/neues-produkt-ridascreen-set-total-10885/ridascreen-SET-total-de.jpg
http://www.r-biopharm.com/wp-content/uploads/de/news/lebens-futtermittelanalytik/neues-produkt-ridascreen-set-total-10885/ridascreen-SET-total-de.jpg
http://www.r-biopharm.com/wp-content/uploads/de/news/lebens-futtermittelanalytik/neues-produkt-ridascreen-set-total-10885/ridascreen-SET-total-de.jpg


30 
 

 

2.2.7 SE PCR 

Enterotoxin SEA, SEB and SED are the most frequently implicated enterotoxins in 

food poisoning outbreaks (Balaban & Rasooly 2000; Bartolomeoli et al. 2009). 

Therefore isolates were screened for enterotoxin genes SEA, SEB, SED and SEH. 

Isolates were streak plated on to TSA and incubated overnight at 37°C. A heavy 

inoculum of each isolate was suspended in 200 µL of peptone diluent and then 

processed according to the Quick-gDNA™ MiniPrep method. The bacterial primer 

sets SEAfor (5ʹ-CCTTTGGAAACGGTTAAAAC-3ʹ), SEArev (5ʹ-

CTCTGMACCTTYCCATCAAA-3ʹ), SEBfor (5ʹ-GGG TATTTGAAGA TGGTAAAAATT-

3ʹ), SEBrev (5ʹ-AGGCGAGTTGTTAAATTCATAGAGTT-3ʹ) and SEDfor (5ʹ-

TCAATTTGTGGATAAATGGTGTAC-3ʹ), SEDrev (5ʹ-TTTCC TCCGAGAGTATCA 

TTAT-3ʹ) were used for the amplification of 16S rRNA in combination with PCR 

Master Mix, according to the manufacturer’s instructions, and yielded products of 

approximately 128, 140 and 154 bp respectively (Derzelle et al. 2009). PCR 

amplifications were performed using the Techne TC-512 thermal cycler (Bio-Techne, 

Minneapolis, USA) (Fig. 10), under the following conditions: an initial denaturation of 

template DNA at 95°C for 8 min; 45 amplification cycles (95°C for 10 s, 58°C for 10 s, 

72°C for 10 s); a cooling step to 4°C. The E-Gel® system (Invitrogen, Fig. 11) using 

the 2% E-Gel was set up according to the manufacturer’s instructions, using program 

7 for 12 min. The seven isolates, S. aureus NZRM 917, positive controls (S27, S12 

and S34) and a negative control (water) were run on the gel and an image was 

generated using the Essential V2 (UVitec Limited, Cambridge, UK) gel 

documentation system (Fig. 12). 

 

 

 

 

 

 

 

 



31 
 

 

Figure 10. The Techne TC-512 thermal cycler used to generate PCR products for SE PCR, multiplex 
PCR and 16S rDNA sequencing.  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



32 
 

 

 

Figure 11. The Invitrogen E-Gel® system used for SE PCR analysis. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



33 
 

 

Figure 12. The UVitec Essential V2 used to photograph the gels generated using both the E-Gel 
system for the SE PCR results and the traditional method for the multiplex PCR results. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



34 
 

2.2.8 SE Multiplex PCR 

As the primers used in the PCR method described in 2.2.8 were specific for S. 

aureus sp, the method and primers employed by Park et al. (2011) was used. The 42 

isolates were streak plated on to TSA and incubated overnight at 37°C. A heavy 

inoculum of each isolate was suspended in 200 µL of peptone diluent and then 

processed according to the Quick-gDNA™ MiniPrep method. The bacterial primer 

sets SEA-F (5’-CAGCATACTATATTGTTTAAAGGC-3’), SEA-R (5’- 

CCTCTGAACCTTCCCATC-3’), SEB-F (5’- GTATGGTGGTGTAACTGAGCA-3’), 

SEB-R (5’- TCAATCTTCACATCTTTAGAATCA-3’), SEC-F (5’- 

CTCAAGAACTAGACATAAAAGCTAGG-3’), SEC-R (5’- 

TCAAAATCGGATTAACATTATCC-3’), SED-F (5’- 

CTAGTTTGGTAATATCTCCTTTAAACG-3’), SED-R (5’- 

TTAATGCTATATCTTATAGGGTAAACATC-3’) SEE-F (5’- 

CAGTACCTATAGATAAAGTTAAAACAAGC-3’) and SEE-R (5’- 

TAACTTACCGTGGACCCTTC-3’) were used. These primer sets were used in 

combination with Platinum® Multiplex PCR Master Mix (Invitrogen, Carlsbad, CA, 

USA), according to the manufacturer’s instructions, and yielded products of 

approximately 400, 351, 271, 319 and 178 bp respectively. PCR amplifications were 

performed under the following conditions: an initial denaturation of DNA at 95 °C for 

10 min was followed by 35 cycles of amplification (95 °C for 30 s, 53 °C for 45 s and 

72 °C for 90 s), ending with a final extension at 72 °C for 10 min. A 4% gel was 

prepared using 4g Bio-Rad agarose and 100mL 1X TBE buffer, heated in a 

microwave until agarose melted, cooled and poured into a 100mL Bio-Rad casting 

tray. EZ-Vision loading buffer was prepared according to the manufacturer’s 

instructions and 1µL aliquots were pre-mixed with 5µL each of the 42 isolates, S. 

aureus NZRM 917, positive controls (S27, S12, S34, 3376 and 3377) and a negative 

control (water). The Bio-Rad gel electrophoresis system was set up according to the 

manufacturer’s instructions and an image was generated using the Essential V2 

(UVitec Limited, Cambridge, UK) gel documentation system (Fig. 12). 

 
 

 

 



35 
 

2.2.9 Interpretation of resulting identities 

For this study, we defined a “correct identity” as one for which two or more of the 

methods (MALDI−TOF and the two biochemical kits) tested gave the same result. 

When the identities of the isolates were different for all three methods, 16S rDNA 

sequence analysis was then considered to determine the “correct identity”.  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



36 
 

Chapter Three: Results and Discussion 

 

3.1 Product Analysis 

 

Dairy product samples from around New Zealand were provided for this study and 

included cheese, butter, butter milk powder, skim milk powder, cream, food additives, 

food bars, curd, yoghurt, dairy spread and UHT milk. Of these product types, cheese 

produced the greatest number of CNS isolates for further analysis (48 out of 126 

CNS isolates, 38%). This was followed by butter and butter milk powder with 33.3% 

and 11.1% of the total number of CNS isolates, respectively. 

 

Of the product types included in this study cheese also provided the greatest 

diversity of CNS species, with S. saprophyticus, S. epidermidis, S. warneri, S. 

pasteuri and S. capitis all being isolated from this dairy product. While yoghurt only 

produced 7 CNS isolates, 4 different CNS species were isolated from it and included 

S. epidermidis, S. warneri, S. hominis and S. capitis. For all other product types 

either one or two species predominated.  

 

 

3.2 Identification of Coagulase-negative Species 

 

3.2.1 Identification of CNS isolates using the BBL™ Crystal™ Gram Positive 

identification system, Remel RapID™ Staph Plus and MALDI−TOF MS  

Of the 427 isolates screened using MALDI−TOF MS, 301 were identified as 

Enterococcus species including E. casseliflavis, E. faecium, E. faecalis, E. 

faecalis/gallinarum and E. gallinarum (136), Macrococcus caseolyticus (70), Bacillus 

species including B. cereus, B. cereus/mycoides, B. lichenformis, B. mycoides, B. 

oleronius, B. pumilus, B. sonorensis, B. subtilis and B. weihenstephanensis/mycoides 

(44), Lysinibacillus fusiformis (29), Staphylococcus aureus (17), Paenibacillus 

thiaminlyticus (2), Lactococcus lactis (2) and Bifidobacterium animalis (1). The 

remaining 126 were identified as CNS species. BP agar was originally developed as 

a selective medium for S. aureus (Baird−Parker, 1962) and is the gold standard 



37 
 

medium that is used in food microbiology for the enumeration of staphylococcal 

species. It was interesting to note, in this current study, that colonies that are 

traditionally thought of as Staphylococcus (a typical black, shiny convex colony 

surrounded by a clear zone) were isolated 4% of the time on this medium. This 

emphasised the importance of a confirmation step as routine practice for these 

bacteria.  

 

Of the CNS isolates identified in this study, 42 were selected, representing isolates 

from a variety of products, and were further analysed using the BBL™ Crystal™ 

Gram Positive identification system, the Remel RapID™ Staph Plus method and 

MALDI−TOF MS applying the full extraction protocol (Table 1). A reference culture, 

S. aureus NZRM 917, was used as a positive control.   

 

 



38 
 

Table 1. Identification of 42 CNS isolates using 3 methods including, the BBL™ Crystal™ Gram Positive identification system, Remel 

RapID™ Staph Plus, and MALDI-TOF MS. 

 

 Isolate Number MALDI-TOF Identity BBL™ Crystal™ Gram Positive 
identification system 

Remel RapID™ Staph Plus 
method 

Agreement between 3 
methods 

1 S. saprophyticus (2.206)* S. saprophyticus (97%)** S. saprophyticus (58%) 

3 S. saprophyticus (2.048) S. saprophyticus (99.6%) S. saprophyticus (58%) 

5 S. epidermidis (2.094) S. epidermidis (>99.99%) S. epidermidis (>99.9%) 

6 S. warneri (1.998) S. warneri (99%) S. warneri (>99.9%) 

7 S. epidermidis (2.181) S. epidermidis (99.98%) S. epidermidis (99.8%) 

9 S. capitis (2.049) S. capitis (99.7%) S. capitis ss capitis (>99.9%) 

11 S. warneri (1.967) S. warneri (99.99%) S. warneri (>99.9%) 

12 S. warneri (2.099) S. warneri (96%) S. warneri (>99.9%) 

16 S. saprophyticus (1.996) S. saprophyticus (99.7%) S. saprophyticus (68%) 

17 S. saprophyticus (1.888) S. saprophyticus (99.9%) S. saprophyticus (68%) 

18 S. saprophyticus (1.8) S. saprophyticus (99.5%) S. saprophyticus (>99.9%) 

19 S. saprophyticus (2.11) S. saprophyticus (99.99%) S. saprophyticus (>99.9%) 

20 S. epidermidis (2.243) S. epidermidis (99.7%) S. epidermidis (>99.9%) 

22 S. saprophyticus (2.018) S. saprophyticus (99.9%) S. saprophyticus (>99.9%) 

23 S. saprophyticus (1.891) S. saprophyticus (>99.99%) S. saprophyticus (>99.9%) 

24 S. saprophyticus (1.947) S. saprophyticus (99.9%) S. saprophyticus (>99.9%) 

25 S. saprophyticus (2.028) S. saprophyticus (65%) S. saprophyticus (>99.9%) 

26 S. warneri (1.796) S. warneri (99%) S. warneri (>99.9%) 

29 S. capitis (1.812) S. capitis (99.95%) S. capitis ss capitis (>99.9%) 

31 S. epidermidis (2.060) S. epidermidis (99.7%) S. epidermidis (99.9%) 

32 S. warneri (1.915) S. warneri (99.9%) S. warneri (>99.9%) 

33 S. epidermidis (2.17) S. epidermidis (97%) S. epidermidis (>99.9%) 

36 S. capitis (2.31) S. capitis (95%) S. capitis ss capitis (99.8%) 

38 S. capitis (1.723) S. capitis (99.7) S. capitis ss capitis (51%) 

40 S. warneri (2.009) S. warneri (65%) S. warneri (>99.9%) 

41 S. epidermidis (2.122) S. epidermidis (99%) S. epidermidis (>99.9%) 

42 S. capitis (2.203) S. capitis (98%) S. capitis ss capitis (99.8%) 

Reference culture S. aureus (2.331) S. aureus (99.7%) S. aureus (>99.9%) 

Agreement between 2 
methods 

2 S. saprophyticus (2.013) S. saprophyticus (52%) S. warneri (99.8%) 
4 S. epidermidis (2.151) S. epidermidis (>99.99%) S. aureus (90%) 

10 S. epidermidis (2.03) S. epidermidis (98.7%) S. saprophyticus (99%) 
21 S. sciuri (1.884) S. sciuri (>99.99%) S. aureus (59%) 
27 S. capitis (2.153) S. capitis (99.7%) S. caprae (>99.9%) 
34 S. capitis (2.009) S. capitis (74%) S. hominis ss hominis (58%) 
35 S. capitis (2.118) S. capitis (95%) S. hominis ss hominis (89%) 
8 S. epidermidis (2.166) S. intermedius (99.91%) S. epidermidis (>99.9%) 



39 
 

30 S. hominis (2.211) S. saprophyticus (99.6%) S. hominis ss hominis (>99.9%) 
39 S. hominis (2.246) S. saprophyticus (>99.99%) S. hominis ss hominis (99%) 

No agreement 13 S. epidermidis (2.107) S. intermedius (77%) S. saprophyticus (92%) 
14 S. pasteuri (2.244) S. haemolyticus (67%) S. xylosus (96%) 
15 S. pasteuri (2.151) S. warneri (71%) S. warneri (>99.9%) 
28 S. warneri (2.27) Micrococcus sp. (58%) S. haemolyticus (>99.9%) 
37 S. warneri (2.45) S. aureus (99%) Kocuria rosea (>99.9%) 

*A score value of >1.7 = genus level identification, >2 = species level identification (as per manufacturer’s instruction – Bruker).  **Grading based on the API Identification system 

 

 

 

 



40 
 

The 3 methods used to identify the CNS isolates in this study agreed on 27 

occasions (64%). On 10 occasions (24%), corresponding identities between 

MALDI−TOF analysis and either of the 2 biochemical test kits was obtained. For 

these isolates, the BBL™ Crystal™ Gram Positive identification system 

corresponded on more occasions with MALDI−TOF analysis (7 out of 10 times) than 

the Remel RapID™ Staph Plus method (3 out of 10 times). For 1 isolate, the 2 

biochemical kits agreed, but the identity was different when based on MALDI−TOF 

analysis. For 4 of the isolates, no agreement between the identities was obtained 

using the 3 methods (Table 1).   

 

Biochemical test methods have been widely used to identify staphylococcal species 

in both clinical and food-related studies. The results of this study suggest that Remel 

RapID™ Staph Plus provides an identification of CNS species that is comparable 

with that of the BBL™ Crystal™ Gram Positive identification system. A previous 

study compared the BBL™ Crystal™ Gram Positive identification system with the 

API Staph (bioMérieux Diagnostic, Germany) and API Strep (bioMérieux Diagnostic) 

methods in identifying Gram-positive cocci, including 77 staphylococcal strains (von 

Baum et al., 1998). The BBL™ Crystal™ Gram Positive identification system overall 

gave a 90% rate of correct species identification, similar to that of the established API 

systems (von Baum et al., 1998).  Another study compared API 32 Staph ID 

(bioMérieux Diagnostic), the BBL™ Crystal™ Gram Positive identification system 

and the molecular method, restriction fragment length polymorphism of the amplified 

tuf gene (Alexopoulou et al., 2006). However, unlike the previous study, this study 

found that the API system had a higher overall agreement with the molecular method 

than the BBL™ Crystal™ Gram Positive identification system (Alexopoulou et al., 

2006).  

 

In comparing the ease of use of the 2 biochemical methods, Remel RapID™ Staph 

Plus provides results more rapidly (4–6 h) than the BBL™ Crystal™ Gram Positive 

identification system (18−24 h). Whereas the BBL™ Crystal™ Gram Positive 

identification system requires additional equipment (BBL™ Crystal™ Panel Viewer), 

which is simple to use, Remel RapID™ Staph Plus is more labour intense because it 



41 
 

requires the addition of reagents to some of the wells. In contrast, MALDI−TOF MS 

provides results within minutes and, although the principle behind the machine is 

complex, the laboratory method itself is relatively simple. This method also includes 

variations that can be applied to the method to aid and improve identification.  

  

Of these 42 isolates, 5 (12%) had an identification that did not agree between the 3 

methods or for which the MALDI−TOF identity did not agree. For example, isolate 28 

was identified as a Micrococcus species by the BBL™ Crystal™ Gram Positive 

identification system, and isolate 37 was identified as Kocuria rosea by Remel 

RapID™ Staph Plus. To obtain a definitive identification, these 5 isolates were further 

analysed using 16S rDNA sequencing. Along with these 5 isolates, 2 additional 

isolates were selected. One in which all 3 methods agreed on the identification 

(isolate 25) and one where only 2 methods agreed (isolate 4). 

 

3.2.2 Definitive identification of seven CNS isolates using 16S rDNA 

sequencing 

In comparison with 16S rDNA sequencing, the BBL™ Crystal™ Gram Positive 

identification system correctly identified two out of the seven isolates to the species 

level and four out of the seven isolates to the genus level. However, it was unable to 

identify one of the seven isolates that were confirmed as staphylococcal species by 

16S rDNA sequencing.  

 

The Remel RapID™ Staph Plus system correctly identified one out of the seven 

isolates to the species level and five out of the seven isolates to the genus level. Only 

one out of the seven isolates, confirmed as a staphylococcal species by 16S rDNA 

sequencing, was not identified by this method. 

 

For the MALDI−TOF MS method, six out of the seven isolates were correctly 

identified to the species level and one out of the seven isolates was correctly 

identified to the genus level. This method was therefore able to correctly identify all 

staphylococcal isolates to at least the genus level when compared with 16S rDNA 

sequencing. 



42 
 

 

3.2.3 Comparison of MALDI−TOF MS identification and 16S rDNA sequencing 

To further compare the identification for the isolates obtained using MALDI−TOF MS 

with that obtained using 16S rDNA sequencing, both an MSP dendrogram (Fig. 13) 

and a phylogenetic tree (Fig. 14) were generated. Identities and clustering of isolates 

1 (S. epidermidis), 3 (S. pasteuri), 4 (S. pasteuri), 5 (S. saprophyticus) and 7 (S. 

pasteuri) (99−100% similarities using 16S rDNA sequencing) were similar for both 

the MSP dendrogram (Fig. 13) and the phylogenetic tree (Fig. 14). The only 

exceptions were isolates 2 and 6. Isolate 2 was identified by MALDI−TOF analysis 

and 16S rDNA sequencing as S. epidermidis. However, it clustered with strains of S. 

capitis on the phylogenetic tree (Fig. 14). S. epidermidis and S. capitis do form part of 

the same phylogenetic cluster group based on 16S rDNA sequencing, which may 

account for this anomaly (Takahashi et al., 1999). Isolate 6 was identified as 

Staphylococcus sciuri (99% sequence similarity) using 16S rDNA sequencing and as 

S. warneri by MALDI−TOF analysis. This isolate clustered with an S. sciuri reference 

strain on the phylogenetic tree (Fig. 14), but with other CNS reference strains on the 

MSP dendrogram (Fig. 13).  



43 
 

 

 

Figure 13. An MSP dendrogram generated by the Bruker Daltonics MALDI Biotyper 3.0 Offline Classification software for seven CNS isolates, for which 

their identification did not agree among the three methods BBL™ Crystal™ Gram Positive identification system, Remel RapID™ Staph Plus and 

MALDI−TOF MS. Isolate identifications based on 16S rDNA sequencing are as follows: isolates 1 and 2 – S. epidermidis, isolates 3, 4 and 7 – S. 

pasteuri, isolate 5 – S. saprophyticus, isolate 6 – S. scuiri subsp. sciuri. 



44 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 14. A phylogenetic tree generated by MegAlign using Clustal V alignment of the seven CNS 

isolates for which the identification did not agree among the three methods (BBL™ Crystal™ Gram 

Positive identification system, Remel RapID™ Staph Plus and MALDI−TOF MS). Bootstrap values > 

70% are shown. Isolate identifications based on 16S rDNA sequencing are as follows: isolates 1 and 2 

– S. epidermidis, isolates 3, 4 and 7 – S. pasteuri, isolate 5 – S. saprophyticus, isolate 6 – S. scuiri 

subsp. sciuri.  

 

 

 

 

 

 

 

 

 
 

0 

4.3 

2 4 

S. epidermidis  P114 
Isolate 1  

88 

S. epidermidis MGB 83-2 

73 

S. epidermidis ATCC 14990 

 

S. epidermidis 100911 
S. capitis DSM 6180 
S. capitis DSM 20325 

88 

S. capitis DSM 20326 

91 

S. capitis ATCC 27840 

92 

Isolate 2 

84 

S. aureus ATCC 14458 
S. aureus NZRM 917 

78 

S. warneri DSM 20036  
S. warneri DSM 20316  

98 

S. warneri DSM 20036 
Isolate 7 
Isolate 4 
S. pasteuri DSM 10656 
S. pasteuri ATCC 51129  

77 

Isolate 3 
S. saprophyticus DSM 20038 
S. saprophyticus DSM 20229 

99.9 

S. saprophyticus ATCC 15305 

99.9 

Isolate 5 
S. sciuri DSM 20345  
S. sciuri DSM 6671 

99.9 

Isolate 6 



45 
 

3.2.4 Overall Examination of Identification Results 

Biochemical test methods have been used widely to identify staphylococcal species 

in both clinical and food-related studies. The results from this study suggest that the 

Remel RapID™ Staph Plus identification system provides an identification of CNS 

species that is comparable with that of the BBL™ Crystal™ Gram Positive 

identification system. A previous study compared the BBL™ Crystal™ Gram Positive 

identification system with the API Staph and API Strep methods in identifying Gram-

positive cocci, including 77 staphylococcal strains (von Baum et al. 1998). The BBL™ 

Crystal™ Gram Positive identification system overall gave a 90% rate of correct 

species identification, similar to that of the established API systems (von Baum et al. 

1998).  Another study compared API 32 Staph ID, BBL™ Crystal™ Gram Positive 

identification system and the molecular method, restriction fragment length 

polymorphism of the amplified tuf gene (Alexopoulou et al. 2006). However, unlike 

the previous study, this study found that the API system had a higher overall 

agreement with the molecular method than the BBL™ Crystal™ Gram Positive 

identification system (Alexopoulou et al. 2006).  

 

16S rDNA sequencing was the reference standard for the identification of 

microorganisms in this study. Overall, the identification of CNS by MALDI−TOF MS 

did compare well with 16S rDNA sequencing. Although MALDI−TOF MS results, like 

those of the biochemical methods, are limited by the number and types of 

microorganisms in the database, reference strains and new strains can be added to a 

separate database linked to the Bruker database. This is important for the food 

industry, where many of the strains seen are from the environment and are not likely 

to be in the existing Bruker database (which is largely based on isolates of clinical 

origin).   

 

The findings in this study are comparable with those from the study by Loonen et al. 

(2012), in which the phenotypic methods, Vitek-2 and ID 32 Staph, the genotypic 

methods, partial 16S and tuf gene sequencing, and the proteomic method 

MALDI−TOF MS were compared in differentiating 117 clinical CNS isolates and 25 

reference strains. MALDI−TOF MS correctly identified 99.3% of the CNS isolates 



46 
 

compared with 93, 70.4, 85.9 and 92.3% for tuf gene sequencing, 16S sequencing, 

ID 32 Staph and Vitek-2 respectively (Loonen et al. 2012). The 16S rDNA 

sequencing was the least successful method for identifying CNS species in the study 

by Loonen et al. (2012), with 29.6% of isolates incorrectly identified and nine out of 

the 25 reference strains (36%) misidentified to the species level. Using this method, 

both S. lugdunesis and S. saprophyticus were misidentified. This differs from this 

present study, in which the similarities for the seven isolates sequenced using the 

16S gene were between 99 and 100% and this method was able to identify one S. 

saprophyticus isolate. Loonen et al. (2012) do note that whole 16S sequencing would 

improve the differentiation of Staphylococcus species using this specific gene.  

 

Matsuda et al. (2012) developed a variation on the full extraction method used with 

MALDI−TOF MS, referred to as the on-plate extraction method, and compared this 

new method with both the direct method and the full extraction method for identifying 

273 clinical staphylococcal isolates and 14 type and reference strains. This new 

method showed no statistically significant differences from the full extraction method 

at the genus level (P = 0.1545) but significantly higher identification rates than the full 

extraction method at the species level (P < 0.0001), particularly in relation to the 

species S. caprae and S. saprophyticus (Matsuda et al. 2012). This study found 

slightly lower identification rates than that of Loonen et al. (2012) and this present 

study, with only 89.5, 80.8 and 60.2% of isolates correctly identified to the species 

level for the on-plate extraction, full extraction and direct methods respectively 

(Matsuda et al. 2012). Similarly Romero-Gómez et al. (2012) reported that 

MALDI−TOF MS correctly identified only 63.3% of CNS isolates in their study using 

the full extraction method. However, whereas the sample was spotted on to the 

target and dried at room temperature in the on-plate extraction method of Matsuda et 

al. (2012), the target was dried at 55°C for the full extraction method. These are not 

the recommended drying conditions of the manufacturers and could have had an 

effect on protein quality and the spectrum produced (Matsuda et al. 2012). Therefore, 

this could account for the low score and the incorrect identification obtained by these 

authors. 

 



47 
 

MALDI−TOF scores below 2.0 affected the correct identification of staphylococci in 

the studies by Matsuda et al. (2012) and Romero-Gómez et al. (2012). However, a 

clinical study by El Bouri et al. (2012) assessed the agreement between MALDI−TOF 

MS and the phenotypic methods BD Phoenix and API for 928 isolates and found that 

including scores between 1.7 and 2.0 for 45 staphylococcal isolates aided the 

MALDI−TOF MS method in achieving a 100% agreement in the identification and 

differentiation between S. aureus and CNS species. This was also observed in the 

study by Carpaij et al. (2011). This study evaluated the use of MALDI−TOF MS for 

identifying 62 clinical CNS isolates in comparison with tuf gene sequencing. Along 

with using scores above 2.0, these authors also considered scores between 1.8 and 

2.0 for duplicate samples as reliable results (Carpaij et al. 2011). The species 

assignment by MALDI−TOF MS was in agreement with tuf gene sequencing for all 62 

isolates (Carpaij et al. 2011). 

 

An emerging view of CNS is that they are essentially opportunistic microorganisms 

that are important nosocomial pathogens and pose a serious risk for the immuno-

compromised, the elderly, neonates and patients with an orthopaedic or cardiac 

prosthesis or prolonged catheter use (d’Azevedo et al. 2010). Therefore, because of 

the increase in the number of nosocomial infections associated with CNS 

(Alexopoulou et al. 2006) and the identification of CNS strains that produce 

enterotoxins in food, the application of MALDI−TOF MS could be as well suited to the 

clinical laboratory as to the food microbiology laboratory.  

 

The results from this study showed that MALDI−TOF MS is a rapid, reliable and 

accurate method for identifying CNS definitively to the genus level and, on most 

occasions, to the species level in dairy products and is therefore a potential 

alternative to the traditional phenotypic, such as commercial identification kits, and 

genotypic, such as sequencing, methods that are currently used (Dupuis et al. 2008).  

 

 

 



48 
 

3.3 Staphylococcal Enterotoxin 

 

3.3.1 Immunoassay test results 

To determine whether enterotoxins were produced by CNS species, 42 isolates were 

tested for SEA, SEB, SEC, SED and SEE production using the Ridascreen® SET A, 

B, C, D, E (R-Biopharm) immunoassay test kit. The three S. aureus positive controls, 

S27, S12 and S34, produced their respective enterotoxins, SEA, SEB and SED. 

However, none of the 42 CNS isolates produced SEA, SEB, SEC, SED or SEE in 

vitro broth culture. 

 

3.3.2 SE gene identification of seven isolates 

As no detectable enterotoxin was produced by any of the 42 CNS isolates, seven of 

these isolated from different products and of different staphylococcal species were 

selected to determine whether they possessed the genes for SEA, SEB, SED and 

SEH using the method of Derzelle et al. (2009). None of the seven selected CNS 

isolates possessed sea, seb, sed or seh genes using this method. 

 

3.3.3 SE gene identification using multiplex PCR 

As no CNS isolates were identified to possess any of the SE genes (sea, seb, sed or 

seh) tested for the 42 CNS isolates were rescreened using a multiplex PCR method 

developed by Park et al. (2011). In this current study set one from the Park et al. 

(2011) method was used to determine the presence of sea-see genes. The S. aureus 

positive control species (S27, S12 and S34) and the S. aureus reference strain 

NZRM 917 were also included.  

 

This method identified that 2 of the 42 CNS isolates possessed a SE gene (Figure 15 

and Figure 16). Isolate 12, a Staphylococcus warneri (Figure 15, lane 13) gave two 

distinct bands of equivalent size to those observed for SEB producing control strain 

(Figure 15, lane 17) and SED (Figure 15, lane 19). Isolate 12 also gave a faint band 



49 
 

of approximately 400bp, equivalent to that observed for the SEA producing control 

strain (Figure 15, lane 16). Isolate 17, a Staphylococcus saprophyticus, gave a band 

of approximately 319bp (Figure 16, lane 4) that is equivalent to the band observed for 

the SED producing control strain (Figure 16, lane 19).  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



50 
 

 

Figure 15. Gel electrophoresis of CNS isolates 1-14 screened for SEA-SEE genes using multiplex PCR. (Lane 1 1Kb ladder, Lane 2 Isolate 1, Lane 3 

Isolate 2, Lane 4 Isolate 3, Lane 5 Isolate 4, Lane 6 Isolate 5, Lane 7 Isolate 6, Lane 8 Isolate 7, Lane 9 Isolate 8, Lane 10 Isolate 9, Lane 11 Isolate 10, 

Lane 12 Isolate 11, Lane 13 Isolate 12, Lane 14 Isolate 13, Lane 15 Isolate 14. Lane 16 – 20 SE producing controls; Lane 16 SEA producing S. aureus, 

Lane 17 SEB producing S. aureus, Lane 18 SEC producing  S. aureus, Lane 19 SED producing S. aureus, Lane 20 SEE producing S. aureus.

1         2         3         4        5        6         7         8           9       10       11      12      13       14      15       16      17       18      19       20        

20202020       

SEB gene 

SEA gene 

SED gene 

SEC gene 

SEB gene 

SEE 

gene 

400bp 

300bp 



51 
 

 

 

 

Figure 16. Gel electrophoresis of CNS isolates 15-28 screened for SEA-SEE genes using multiplex 

PCR. (Lane 1 1Kb ladder, Lane 2 Isolate 15, Lane 3 Isolate 16, Lane 4 Isolate 17, Lane 5 Isolate 18, 

Lane 6 Isolate 19, Lane 7 Isolate 20, Lane 8 Isolate 21, Lane 9 Isolate 22, Lane 10 Isolate 23, Lane 11 

Isolate 24, Lane 12 Isolate 25, Lane 13 Isolate 26, Lane 14 Isolate 27, Lane 15 Isolate 28.  Lane 16 – 

20 SE producing controls; Lane 16  SEA producing S. aureus, Lane 17 SEB producing S. aureus, 

Lane 18 SEC producing  S. aureus, Lane 19 SED producing S. aureus, Lane 20 SEE producing S. 

aureus). 

 

 

 

 

 

 

 

 

 

 

 

SED gene 

1         2         3         4        5        6         7         8           9       10       11      12      13       14      15       16      17       18      19       20        

20202020       

300bp 

400bp 

500bp 



52 
 

 

 

 

Figure 17. Gel electrophoresis of CNS isolates 29-42 screened for SEA-SEE genes using multiplex 

PCR. (Lane 1 1Kb ladder, Lane 2 Isolate 29, Lane 3 Isolate 30, Lane 4 Isolate 31, Lane 5 Isolate 32, 

Lane 6 Isolate 33, Lane 7 Isolate 34, Lane 8 Isolate 35, Lane 9 Isolate 36, Lane 10 Isolate 37, Lane 11 

Isolate 38, Lane 12 Isolate 39, Lane 13 Isolate 40, Lane 14 Isolate 41, Lane 15 Isolate 42. Lane 16 – 

20 SE producing controls; Lane 16 SEA producing S. aureus, Lane 17 SEB producing S. aureus, Lane 

18 SEC producing  S. aureus, Lane 19 SED producing S. aureus, Lane 20 SEE producing S. aureus). 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

1         2         3         4        5        6         7         8           9       10       11      12      13       14      15       16      17       18      19       20        

20202020       

400bp 

300bp 



53 
 

 
 

3.3.4 Overall Examination of Enterotoxin Results 
 
Although the Ridascreen® SET A, B, C, D, E (R-Biopharm) immunoassay test kit 

method describes the use of brain heart infusion broth (BHI) broth for culturing 

isolates, TSB was used in this study. To confirm that this did not inhibit the 

production of enterotoxins by the CNS isolates, the S. aureus positive controls were 

grown in both BHI and TSB and put through the immunoassay. Each of the positive 

controls, S27, S12 and S34, produced detectable levels of their respective 

enterotoxins, after growth in both BHI and TSB. This confirms that growth in TSB 

does not affect enterotoxin production. 

 

The Ridascreen® SET A, B, C, D, E (R-Biopharm) immunoassay test kit is prone to 

cross-reactivity between SEA and SEE, and between SEB and SEC. In this study, 

S27, a SEA-producing isolate, gave a value above the threshold for the SEA 

antibody but also reacted with the SEE antibody, when grown in BHI. S12, a SEB-

producing isolate, gave a value above the threshold for the SEB antibody and also 

the SEC antibody when grown in both BHI and TSB. Although the additional 

reactivity of S27 and S12 to SEE and SEC respectively can be attributed to cross-

reactivity, S12 also produced a value above the threshold for SEE. This method 

therefore identified an additional enterotoxin produced by S12 that was not previously 

identified using the 3M™ TECRA™ Staph Enterotoxin immunoassay. 

 

The production of SEs by CNS species was not detected using the immunassay; 

however, these isolates were tested only in pure culture. The production of SEs by 

these species may occur when tested for directly from food products or when food is 

artificially contaminated with a CNS isolate at the enterotoxin-producing level (103 

CFU/mL). The conditions used to manufacture a food product along with the 

physicochemical conditions of a food may also play a major role in CNS producing 

enterotoxins.  

 

200 200 



54 
 

Although no SE production was detected by the immunoassay, the multiplex PCR 

sequencing did show that 2 of the 42 CNS isolates analysed possessed SE genes. 

Other studies have also shown a small proportion of isolates possess SE genes For 

example, in the study by Even et al. (2010) only one out of 129 CNS isolates was 

found to possess an enterotoxin gene and in the study by Ruaro et al. (2013) none of 

the 77 CNS isolates from raw milk and cheese possessed a enterotoxin gene. 

 

It is also worth noting that isolate 17 (S. saprophyticus) produced a band equivalent 

to that of SEA gene, the most commonly reported enterotoxin in food and therefore 

the main cause of Staphylococcus aureus food poisoning (Vernozy-Rozand et al. 

2004; Veras et al. 2008; Argudín et al. 2010) and isolate 12 (S. warneri) produced 

bands equivalent to both SEB and SED, both of which have also been implicated in 

staphylococcal food poisoning throughout the world (Argudín et al. 2010). Although 

neither of these isolates produced enterotoxin(s) in the immunoassay if these 

staphylococci where present in large enough numbers, or exposed to favourable 

growth temperatures in food ingredients or products, this could result in them 

producing enterotoxins and could lead to food poisoning of an individual or 

individuals.   

 

 

 

 

 

 

 

 

 

 

 

 

 

 



55 
 

 

 

Chapter Four: Summarising Discussion and 

Conclusion 

 

The third most frequent cause of food poisoning worldwide is due to Staphylococcus, 

with the coagulase positive species, S. aureus being identified as the leading cause 

(Atanassova et al. 2001; Veras et al. 2008; Podkowik et al. 2013). The ability of S. 

aureus to cause food poisoning is due to the production of enterotoxins under 

favourable conditions. However there is increasing evidence that the coagulase 

negative staphylococci (CNS) species also possess and are capable of producing 

these enterotoxins and therefore have been or could potentially be linked to food 

poisoning outbreaks. While research has been carried out in various countries there 

is a lack of research into the potential of CNS to cause food poisoning as generally 

related to food industry and more specifically as linked to dairy in New Zealand.  

 

4.1.1 Identification of coagulase-negative species 

A number of dairy products including cheese, butter, milk powder (butter and skim 

milk), cream, food additives, food bars, curd, yoghurt, dairy spread and UHT milk, 

were screened in order to isolate staphylococci species. To determine whether these 

isolates were S. aureus, CNS or other species three methods were used, matrix-

assisted laser desorption ionisation – time of flight mass spectrometry (MALDI-TOF 

MS), BBL™ Crystal™ Gram Positive identification system and the Remel RapID ™ 

Staph Plus identification system. 

 

While various studies have used the BBL™ Crystal™ Gram Positive identification 

system to identify Gram positive species, including staphylococci, the use of the 

Remel RapID ™ Staph Plus identification system to identify and differentiate 

staphylococci species has not been extensively investigated.  However this newer 

system provides a result within 28 hours (with an incubation time of 4 hours) and 

includes 40 species of staphylococci compared to the BBL™ Crystal™ Gram Positive 



56 
 

identification system which requires a minimum incubation time of 18 hours therefore 

results within 48 hours and only includes 25 staphylococci species. The Remel 

biochemical kit therefore is a fast biochemical alternative to the BBL method and 

could be a useful method within a microbiology laboratory, particularly clinical. 

It can be difficult to differentiate between coagulase positive and negative 

staphylococci species using traditional methods. The use of MALDI-TOF MS could 

improve rapidity of identification even further providing a fast turnaround time and 

accurate identity of staphylococci. This could be of significant benefit to both the 

clinical and food industries, not only in relation to the identification of staphylococci 

but all bacterial, yeast and mould species in general. However MALDI-TOF MS is not 

an approved validated method for the food industry, although it is extensively used in 

clinical settings. Therefore working towards gaining validation of this method for use 

with food is an important next step in its use in the food industry. This method could 

be a valuable tool in locating the source of contamination of a product as part of a 

trace back and identifying staphylococci species responsible for food poisoning 

outbreaks. 

 
 
4.1.2 Staphylococcal enterotoxin potential of CNS species 

This study used two different PCR methods to determine whether any of the 42 CNS 

isolates possess an enterotoxin gene. The SE PCR method by Derzelle et al. (2009) 

did not indicate that any of the 42 CNS species analysed possessed enterotoxins 

SEA, SEB, SED or SEH. However the multiplex PCR method by Park et al. (2011) 

identified 1 species that possesses SED and a second that possesses SEB and 

SED. The number of isolates identified as possessing one or more of the enterotoxin 

genes, SEA-SEE, in this study is low, it is still comparative to other studies for 

example Even et al. (2010) and Ruaro et al. (2013).  

 

Although 2 CNS isolates were shown to possess enterotoxin genes this was not 

associated with the production of the enterotoxin in vitro (broth culture). However 

these CNS may be capable of producing their respective SE under the right 

conditions in vivo and while this was not undertaken in this present study it is well 

worth investigating. 



57 
 

 

With the increasing evidence of CNS species possessing the enterotoxin genes 

(Chou & Chen 1997; Cunha et al. 2006; Veras et al. 2008; Mhlambi et al. 2010; 

Naidoo & Lindsay 2010) and a CNS species being linked to a food poisoning 

outbreak (Do Carmo et al. 2004) this indicates that staphylococci species should not 

be excluded when testing for SE production in the manufacture of food products. 

CNS species could therefore be a potential source of enterotoxin production in dairy 

products; however the risk is likely to be low. 

 

While the risk of food poisoning linked to CNS species is low these staphylococci 

species have been shown to be a major source of mastitis in dairy herds globally 

(Zhang & Maddox 2000; Pyorala & Taponen 2009; Tenhagen et al. 2009; Sampimon 

et al. 2010) and in New Zealand (Howard 2006). A delay to identify and separate 

those within the herd suffering from CNS associated mastitis and or even those with 

an increased CNS population is a potential source of contamination into the milk 

supply, which under favourable conditions could result in the production of 

enterotoxins. This could therefore in the future lead to an increase in the risk of CNS 

to cause food poisoning.  

 

In addition to the CNS mastitis causing species being a potential source of 

enterotoxin in dairy the consumption of raw unpasteurised milk is increasingly being 

advocated for in relation to perceived health benefits and improved taste over 

pasteurised milk. This has resulted in it becoming an increasing consumer trend 

globally and in New Zealand. With the consumption of raw milk comes the increased 

risk of exposure to disease causing bacteria, including staphylococci species and this 

risk is increased further when there is poor processing, hygiene and storage of the 

milk on farm and then transport, storage and handling by the consumer. This could 

also serve as another potential source of enterotoxin and therefore increase the risk 

of food poisoning associated with unprocessed dairy products.  

 

The main methods to detect enterotoxin production available commercially are 

immunoassay kits and these kits are currently only able to detect enterotoxins SEA-



58 
 

SEE. MALDI-TOF MS is a potential alternative to these methods and while the use of 

this method to detect enterotoxins has been studied (Brun et al. 2007; Dupuis et al. 

2008) further research is still required. Therefore MALDI-TOF MS could replace the 

immunological methods and improve investigations into staphylococcal food 

poisoning outbreaks breaks (Dupuis et al. 2008) by extending to include all currently 

known SEs and SELs and also as yet unidentified or previously unidentifiable SEs. 

 

4.2 Conclusions 

 The Remel RapID™ Staph Plus identification system provides an identification 

of CNS species that is comparable with that of the BBL™ Crystal™ Gram 

Positive identification system. 

 

 MALDI−TOF MS is a rapid, reliable and accurate method for identifying CNS 

definitively to the genus level and, on most occasions, to the species level in 

dairy products and is therefore a potential alternative to the traditional 

phenotypic, such as commercial identification kits, and genotypic, such as 

sequencing, methods that are currently used. 

 

 SE production in vitro and in TSB broth by CNS was not detected using the 

immunoassay method, Ridascreen® SET A, B, C, D, E. 

 

 None of the CNS isolates were found to possess an enterotoxin gene when 

using the single SE PCR method. However using the multiplex PCR method 

one isolate was found to possess both the SEB and SED genes, and another 

the SEA gene. 

 

 The possession of enterotoxin genes by two CNS isolates indicates that 

coagulase producing Staphylococcus aureus species are not the only potential 

source of enterotoxin production in dairy products, although the risk is likely 

low.  

 

 



59 
 

4.3 Limitations of the Present Study  

 This study only screened dairy products from one dairy season, therefore only 

gives a snapshot of CNS isolates present within the New Zealand dairy 

industry. 

 

4.4 Future Work  

 The enterotoxin immunoassay could be repeated for all 42 CNS in vivo i.e. 

within various dairy products. 

 16S rDNA sequencing could be completed for all 42 CNS isolates. 

 The multiplex PCR screening for SEA-SEE genes could be extended to 

include the remaining 101 CNS isolated from the dairy products in this study. 

 Multiplex PCR analysis could be extended to include SEG-SEI and SEJ-SELU 

genes using the method by Park et al. (2011).  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



60 
 

References  

Alexopoulou, K., Foka, A., Petinaki, E., Jelastopulu, E., Dimitracopoulos, G., and 

Spiliopoulou, I. (2006) Comparison of two commercial methods with PCR restriction fragment 

length polymorphism of the tuf gene in the identification of coagulase-negative staphylococci. 

Letters in Applied Microbiology 43: 450–454. 

 

Anhalt, J.P., and Fenselau, C. (1975) Identification of bacteria using mass spectrometry. 

Analytical Chemistry 47: 219–225. 

 

Argudín, M.Á., Mendoza, M.C., and Rodicio, M.R. (2010) Food poisoning and 

Staphylococcus aureus enterotoxins. Toxins 2: 1751–1773. 

 

Asao, T., Kumeda, Y., Kawai, T., Shibata, T., Oda, H., Haruki, K., Nakazawa, H., and Kozaki, 

S. (2003) An extensive outbreak of staphylococcal food poisoning due to low-fat milk in 

Japan: estimation of enterotoxin A in the incriminated milk and powdered skim milk. 

Epidemiology and Infection 133: 33–40. 

 

Atanassova, V., Meindl, A., and Ring, C. (2001) Prevalence of Staphylococcus aureus and 

staphylococcal enterotoxins in raw pork and uncooked smoked