Stability of bacterial DNA in
relation to microbial detection
in teeth
MALIN BRUNDIN
Department of Odontology
Umeå University, 2013
This work is protected by the Swedish Copyright Legislation (Act 1960:729) ISBN: 978-91-7459-743-1
ISSN: 0345-7532
Electronic version available at http://umu.diva-portal.org/ Set in Adobe Caslon Pro 10,5/12,6 and Myriad Pro Design and Layout in InDesign CS6 8.0 by the author. Printed by: Print & Media
Contents
Abstract vii
Preface ix
Introduction 1
The culture technique — a short history 2
Microbial infection of the pulp space 2
A new era 3
Molecular techniques used in detection of DNA 5 Polymerase chain reaction (PCR) 5 Quantitative real-time PCR (qPCR) 6 Molecular targets for assessment of viability 6 Stability of DNA 7
DNA decomposition 8
Reliability of molecular techniques in analyses of endodontic samples 8
Aims of the study 11
Overview of the methodology 13
Persistence of DNA from non-viable bacteria 13
Persistence of DNA from non-viable E. faecalis in ex vivo root canals 13 Persistence of DNA from non-viable F. nucleatum and P. anaerobius in vitro 13
Decay of DNA 14
Stabilisation of DNA 14
Hydroxyapatite 14
DNA preservation 17
Cell-bound—free DNA 18
Cell-bound DNA 19
Free DNA 19
Stabilisation of free DNA by binding to minerals 20
Hydroxyapatite 20
Human dentine 21
General discussion 23
Acknowledgements 27
Abstract
T
he fate of DNA from dead cells is an important issue when interpreting results from root canal infections analysed by the PCR technique. DNA from dead bacterial cells is known to be detectable long time after cell death and its stability is dependent on many different factors. This work investigated factors found in the root canal that could affect the recovery of microbial DNA.In an ex vivo experiment, DNA from non-viable gram-positive Enterococcus faecalis was inoculated in instrumented root canals and recovery of DNA was assessed by PCR over a two-year period. DNA was still recoverable two years after cell death in 21/25 teeth. The fate of DNA from the gram-negative bacteria Fusobacterium nucleatum and the gram-positive Peptostreptococcus anaerobius was assessed in vitro. DNA from dead F. nucleatum and P. anaerobius could be detected by PCR six months post cell death even though it was clear that the DNA was released from the cells due to lost of cell wall integrity during the experimental period. The decomposition rate of extracellular DNA was compared to cell-bound and it was evident that DNA still located inside the bacte-rium was much less prone to decay than extracellular DNA.
Free (extracellular) DNA is very prone to decay in a naked form. Binding to minerals is known to protect DNA from degradation. The fate of extracellular DNA was assessed after binding to ceramic hydroxyapatite and dentine. The data showed that free DNA, bound to these materials, was protected from spontaneous decay and from enzymatic decomposition by nucleases.
The main conclusions from this thesis were: i) DNA from dead bacteria can be detected by PCR years after cell death ex vivo and in vitro. ii) Cell-bound DNA is less prone to decomposition than extracellular DNA. iii) DNA is released from the bacterium some time after cell death. iv) Extracellular DNA bound to hydroxyapatite or dentine is pro-tected from spontaneous decomposition and enzymatic degradation.
Keywords:
Cell-bound DNA, cell-death, dentine, DNA binding affinity, DNA decomposition, DNA pres-ervation, extracellular DNA, hydroxyapatite, PCR, polymerase chain reaction.
Preface
The thesis is based on the following publications, which are referred to into the text by their Roman numerals.
I Brundin M, Figdor D, Roth C, Davies JK, Sundqvist G, Sjögren U. Persistence of dead-cell bacterial DNA in ex vivo root canals and influ-ence of nucleases on DNA decay in vitro.
Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2010; 110:789-94. II Brundin M, Figdor D, Sundqvist G, Sjögren U. Preservation of
Fuso-bacterium nucleatum and Peptostreptococcus anaerobius DNA after loss of cell viability.
Manuscript Submitted
III Brundin M, Figdor D, Sundqvist G, Sjögren U. DNA binding to hydroxyapatite: a potential mechanism for preservation of microbial DNA.
J Endod 2013; 39:211-6.
IV Brundin M, Figdor D, Johansson A, Sjögren U. Preservation of bacte-rial DNA by human dentine.
Article in press, J Endod
Introduction
M
odern studies in endodonticmicrobiology took a great leap forward in 1966 when Möller completed a seminal work on microbio-logical examination of root canals (Möller 1966). Prior to that time, numerous bac-teriological studies had been conducted on root canal samples, but almost all were impaired to a greater or lesser degree by shortcomings in their methodology. How were the samples taken? What steps were applied to isolate the field from the rest of the oral cavity, which was teeming with resident and transient microbiota? How was the field cleaned and disinfected to reduce and eliminate contaminating microbiota? What controls were estab-lished to check the operative field for con-tamination before sampling? Once taken, how were the samples handled and pro-cessed from the clinic to the laboratory? These essential questions had not been raised, let alone adequately addressed, yet were elemental for the proper scien-tific microbiological study of the infected root canal. Möller asked these questions and then sought the answers by careful, methodical work, which laid the
ground-work for the study of endodontic microbi-ology for the next 45 years, and well into the future.
Möller began his extensive methodo-logical studies by thoroughly cleaning the subject tooth, isolating it with rubber dam and disinfecting it. The disinfectant had to be inactivated to prevent carry-over of the antimicrobial into the bacterial growth media. At each stage, samples were taken and processed by culture to establish pre-cisely which were the most effective meth-ods. In this systematic way, he was able to determine a reliable protocol. Möller also developed and tested media for sampling, transfer and storage of isolates recovered from root canals to ensure that there were clinical and laboratory methods suita-ble for the optimal detection of bacteria (Möller 1966). Even today, Möller’s work provides the basic principles and rules for sampling and recovery of the root canal microbiota.
The culture technique — a short history It has long been known that bacteria are the predominant cause of pulp diseases (Miller 1894) and it is the culture tech-nique that has been used for microbial identification in all of the classic studies. Several classical culture studies in human and animal teeth have demonstrated that bacterial infection of the pulp space is essential for the development of apical periodontitis (Kakehashi et al. 1965; Sun-dqvist 1976). Using advanced techniques to isolate and culture bacteria from the root canal, much information has been provided about the composition of the flora of untreated root canals (Kantz and Henry 1974; Wittgow and Sabiston 1975; Sundqvist 1976), microbial associations amongst the microbiota (Sundqvist 1992), and about bacteria that are associated with acute symptomatic presentation of periapical disease (Sundqvist et al. 1989; Siqueira and Rôças 2009). Bacteriological studies using advanced culture techniques have been applied to the study of persis-tent infection of root canals in teeth with previous endodontic treatment to iso-late and identify the microbial aetiology (Molander et al. 1998; Sundqvist et al. 1998; Hancock et al. 2001; Siqueira and Rôças 2004).
Cultivation has long been considered to be the golden standard when analysing bacterial samples from root canals. How-ever, the method has limitations. Species present in very low numbers (10–102) in the bacterial sample may be lost in the cul-tivating and analysis procedure or during the transport to the laboratory. Cells from biofilm firmly attach to each other and their viability can be jeopardised during separation (Joyce et al. 2003; Kobayashi et al. 2007). These factors may contribute to misleading results and underestimations of bacterial growth.
The inability for some microbes (e.g. spirochetes) to grow in artificial environ-ments is a weakness of the method when surveying the root canal flora. Another limitation of the culture technique is the difficulty to identify and characterise bacteria growing on agar plates. Correct identification relies heavily on the oper-ator’s technical skills. Identification and characterisation is done by visual exam-ination of the growing colonies followed by further biochemical and physiological tests such as sugar fermentation patterns, enzyme synthesis and acidic fermentation products. By comparing the differences and similarities in these tests, bacteria can be classified into groups (taxa) and spe-cies. Classification of many oral bacteria remains unsolved as simple criteria for identification is not available. The use of molecular techniques, targeting the DNA of the bacterial cell, has lead to reclassi-fication and renaming of several species. The bacterial taxonomy is a dynamic pro-cess constantly updated with new species and renamed species due to more specific and stringent tests.
Microbial infection of the pulp space In the oral cavity, a diverse number of bacterial species are present (Aas et al. 2005; Paster et al. 2006; Faveri et al. 2008). All microorganisms in the oral cavity have the opportunity to invade the necrotic pulp space, but due to the spe-cific milieu in the root canal only a few species are able to survive and establish. Bacteria from the infected root canal are predominately strict anaerobic and the composition consists of several dif-ferent species in a number of 103 to 108 cells (Sundqvist 1976). The polymicrobial infection in the root canal has great sim-ilarities with the flora of the pathologic gingival pocket. In both cases the
com-In the next phase, proteolytic bacteria like Fusobacterium nucleatum and Bacte-roides species were predominant. Finally bacteria that can degrade serum proteins and use amino acids in their fermentation (Peptostreptococcus micros and Eubacterium brachy) were abundant in the infection (ter Steeg and van der Hoeven 1989).
In another study primate teeth were devitalized and infected by exposure of the pulps to the mouth flora. The pulp chambers were then sealed and samples were collected from the canals after var-ious observation periods (Fabricius et al. 1982). The proportion of obligate anaer-obic strains increased with time as shown in Figure 1.
In conclusion, the composition of the root canal infection is unique for every tooth, but will vary over time with changes in the environmental conditions.
A new era
During the past two decades, the intro-duction of molecular techniques have significantly enhanced the capacity for detection and identification of microbes from diverse environments. A key dif-ference between the molecular approach and culture techniques is that the former detects DNA, whereas the latter assesses viable cells. For clinical work, detection of bacterial DNA in a sample generally implies that the bacterium is present and a likely contributor to the infection. An important advantage of this technology is that rapid answers are available compared to days or weeks of incubation and bio-chemical tests. Moreover, the technique allows great accuracy for distinguishing between closely related species, which can be difficult with the culture-based meth-ods (Jervøe-Storm et al. 2005; Downes and Wade 2006).
Figure 1. Distribution of obligate and
faculta-tive bacteria detected after different times of closure (7–1080 days) of infected root canals in monkeys. Adapted from Fabricius et al. 1982. position depends on nutrition and oxygen factors but also interactions between the bacteria (Socransky and Haffajee 2002; Socransky and Haffajee 2005). As the conditions changes over time it will be reflected on composition of the flora (Fabricius et al. 1982; ter Steeg and van der Hoeven 1989). In vitro experiments show that rapidly growing saccarolytic bacteria (Streptococcus sp, Bifidobacte-rium adolescentis, EubacteBifidobacte-rium sabbureum) consuming carbohydrates (free glucose and the carbohydrate side-chains of gly-coproteins) dominated in the infection.
50
50
0
100
PER CENT OF TO TAL FL OR A OBLIGA TE ANAER OBIC BA C TERIA F ACUL TA TIVEL Y ANAER OBIC BA C TERIA 7d 90d 180d 1060dDENATURA-TION OF DNA PRIMERS ANNEALING EXTENSION OF DNA 2 COPIES OF TARGET DNA DENATURA-TION OF DNA 4 COPIES OF TARGET DNA
Figure 2. Conventional PCR requires two short DNA fragments (primers) that can bind to the
target DNA. The reaction starts with separating the target DNA strand by increasing the tem-perature (denaturation). When the suspension with the separated DNA strands cools down the primers can bind (anneal) to the specific part of the target DNA that is to be copied (amplified). With the primer as a starting point a new DNA strand is built (extension) in the presence of the heat stable enzyme DNA-polymerase and single units of DNA. The procedure is repeated 15–40 times (cycles).
Molecular techniques used in detection of DNA
The molecular approach targets specific DNA sequences, and the specificity of the molecular techniques allows species to be distinguished and characterised with greater resolution than by phenotypic and biochemical analysis used in the classical cultivation method.
Polymerase chain reaction (PCR)
The polymerase chain reaction is a well-es-tablished method based on detection of species-specific DNA sequences from the target microorganism. Kary B. Mullis invented the technology in the 1980’s and the discovery was rewarded with a Nobel prize in 1993 (Mullis and Faloona 1987). Using PCR, minute amounts of a specific DNA sequence can be amplified into mil-lions of copies (Fig. 2) and, after gel elec-trophoresis and staining with ethidium bromide, can be visualised by exposure to ultraviolet light. This means that very small amounts of DNA can be amplified and detected by this technique.
The PCR technique has revolutionised both the biological sciences and many clinical disciplines. The field of micro-biology has benefited in particular from the ability to rapidly detect and identify specific DNA sequences. Together with the application of the sequence encoding the 16S rRNA gene (Woese 1987), it has brought about a re-structure of microbial taxonomy that allows identification and classification with great precision.
The application of the PCR technique in endodontic microbiology was first described in 1997, using primers to suc-cessfully target species that are known to be difficult to culture or differentiate by biochemical methods (Conrads et al. 1997). In subsequent studies, the applica-tion of PCR in endodontic research has
led to identification of difficult-to-culture species such as Treponema denticola and Bacteroides forsythus being more frequently reported (Gonçalves and Mouton 1999; Siqueira et al. 2000), description of pre-viously undetected species (Munson et al. 2002) and not-yet-cultivable species (Aas et al. 2005). Application of this technol-ogy has greatly increased in recent years, for detecting the root canal microbiota. Studies using molecular techniques have generally reported a more diverse micro-bial flora associated with endodontic infections than that revealed by culture methods (Siqueira et al. 2000; Munson et al. 2002; Siqueira et al. 2007).
The PCR technique is a highly effective method that allows great specificity and incredible precision to microbial identifi-cation. However, it should be recognised that the conventional PCR method has some limitations. The reaction is sensitive to levels of divalent cations (Mg2+) (Cha and Thilly 1993) and the conditions must be optimised for each application. The primer (Fig. 3) design is very important. Primers have to be specific for the target DNA to be amplified, but should not be capable of annealing to each other.
3’ 5’ TGCCATAGACAAT Primer CGATAGGATCAACGGTATCTGTTTA Template 5’ 3’ 3’ 5’ TGCCATAGACAAAT : : : : : : : : : : : : : Annealing CGATAGGATCAACGGTATCTGTTTA 5’ 3’
Figure 3. The primer is a DNA sequence
reversed to the region of target on the DNA template and determine the starting point for PCR-replication.
A major consideration with PCR is the risk of contamination (Corless et al. 2000; Pääbo et al. 1989). As the technique is very sensitive, even the smallest contam-ination of target DNA can result in false positive results. In the oral cavity, all areas of the hard and soft tissues are covered with large numbers of microorganisms, so it is absolutely critical in endodontic microbiological research that contamina-tion is excluded and checked by sterility controls.
The results from PCR provide data about the presence of DNA from targeted microbes. But a limitation with conven-tional PCR is that no information is avail-able about the total amount of bacteria, nor the proportions of the different species in a mixed bacterial infection. Also, because DNA alone is the target, conventional PCR will amplify target oligonucleotides regardless of whether the DNA is derived from living or dead microbes. With con-ventional PCR, there is no distinction between DNA from living and dead bac-teria (Josephson et al. 1993; Kobayashi et al. 2009). The implications for this in endodontic research is that there is the potential for recovery of microbial DNA from root canal samples that contain both living and dead microbes. Thus, there is the possibility that microbes may be mistak-enly identified as active inhabitant in the root canal infection.
As these shortcomings of conventional PCR were identified over the years, refinements and modifications of the basic PCR technique have been progressively introduced to overcome or minimise limi-tations. Two methods are outlined below. Quantitative real-time PCR (qPCR) One innovation for assessing the amount of bacteria in a sample is the introduction of quantitative real-time PCR (qPCR).
Instead of analysing the PCR products by gel electrophoresis, the amplified DNA product is optically detected in real-time by fluorescent molecules bound to the amplified DNA. The fluorescent mole-cules are present either in free form or bound to a probe and are added to the PCR mixture. The resulting fluorescent signal is proportional to the amount of amplified DNA. In this way, quantifica-tion of content of a specific nucleic acid sequence is possible in the start sample. Molecular targets for assessment of viability Using culture methods, determining the viability of readily cultivable cells is straightforward, i.e. observation of growth on media means by direct impli-cation that there are viable, multiplying cells. However, with molecular methods, a notable challenge is how to determine whether cells are living or dead. One way to differentiate viable and dead cells, is to detect mRNA in a sample. The mRNA molecule transfers genetic information that directs synthesis of specific proteins. It is an unstable molecule with a rapid turnover and short half-life (Rauhut and Klug 1999), making it a suitable target for PCR. Detection of mRNA minimises the risk that the oligonucleotides derive from dead cells, since the presence of mRNA implies that the cells are viable or at least were recently alive (Donovan and Kush-ner 1986; Sheridan et al. 1998; Kobayashi et al. 2009). Single-stranded RNA can be amplified by PCR after an additional Reverse Transcription step (Lynas et al. 1989). The process of extracting and stor-ing RNA is sensitive due to rapid turn-over (Holland et al. 2003; Pérez-Novo et al. 2005) and low quality RNA may compromise the results (Fleige and Pfaffl 2006).
determining cell viability is assessment of the intactness of bacterial cell walls. This is based on the notion that cells with damaged cell walls are dead and that cells with intact walls are presumed viable. One commonly used method is the use of dyes to alternately stain intact cell walls or penetrate damaged cells and stain their contents. The commercially available LIVE/DEAD BacLight kit (Invitrogen) is a popular method that uses two stains, propidium iodide (PI) and SYTO9, which both stain nucleic acids. The green fluo-rescing SYTO9 can enter all cells, which is useful for assessing total cell counts; the red fluorescing PI enters only cells with damaged cytoplasmic membranes. Anal-ysis of the results from the LIVE/DEAD BacLight method can be done either by fluorescence microscopy, or flow cytome-try for quantitating cell numbers.
Assessment using molecular methods has been an elusive goal, but recently the possibility of exclusively detecting DNA from viable microorganisms has been shown by the use of the dye Propidium Monoazide (PMA) (Nocker et al. 2006). PMA enters only bacteria with compro-mised cell walls. Inside the cell, PMA can be linked by covalent binding to the DNA after exposure to strong visible light. Amplification of DNA bound to PMA has been shown to be inhibited, result-ing in a weak or no PCR signal (Nocker et al. 2006). Viable cells, with intact cell membranes, are not influenced by PMA and can be amplified in the PCR reac-tion (Nocker et al. 2006; Sánchez et al. 2013). This method has just recently been applied in endodontic microbiology (Kim et al. 2013).
Stability of DNA
Molecular methods have been applied to study ancient DNA in the emerging field
of paleomicrobiology. The first report was based on samples from ancient Egyptian mummies that were analysed for ancient DNA (Pääbo 1985). The technique has been used to obtain information about the presence of tuberculosis and other infectious diseases in ancient material (Nerlich et al. 1997; Donoghue et al. 2005; Papagrigorakis et al. 2006). It has also been possible to determine species of extinct animals and decide how these are related to today’s species through DNA analysis (Higuchi et al. 1984).
In paleomicrobiology, samples from bone and teeth are the most fruitful target as they have been shown to produce a greater yield of DNA after extraction compared to soft tissue (Pruvost et al. 2007). The minerals (in tooth and bone) may stabilise the DNA, which makes these tissues more suitable for the analysis of ancient DNA. Soft tissue will decom-pose faster and DNA may no longer be recoverable. A major problem in paleom-icrobiology is the risk of sample contam-ination with DNA from other sources, as the molecular techniques are very sensi-tive.
In endodontics, despite an increasing number of studies using molecular tech-niques, there has been limited discussion about whether the tissues found in bone and in particular teeth can influence the decay of bacterial DNA after cell death. With widely diverging results on the microbiota in infected root canals ana-lysed by culture versus molecular tech-niques, there is a question as to whether there may be factors in the root canal that affect the stability of the bacterial DNA. Bacteria that enter into the root canal but do not survive can be a source of error in the analysis of the present infection, if their DNA is able to persist for a long time. Another important issue, which is
not addressed in the endodontic litera-ture, is the risk of contamination from the surface of the tooth during sampling. DNA decomposition
Nucleic acids undergo spontaneous decomposition in solution (Lindahl 1993). The nucleotides in DNA and RNA are bound together by phosphodiester bonds. These bondings are weaker in RNA because of the presence of 2’-hydroxyl groups of ribose resulting in that RNA is very susceptible to hydrolysis. The phos-phodiester bonds of DNA are stronger, but instead the bondings between the sugar and the bases in DNA (N-glucosyl bonds) are labile. Once a base has been cleaved from the DNA strand, the DNA chain becomes weakened and will be cleaved within days (Lindahl 1993).
The rate of decomposition is affected by humidity, temperature and pH (Lin-dahl and Nyberg 1972; Burger et al. 1999; Götherström et al. 2002; Edwards et al. 2004). Access to water provides the abil-ity to hydrolysis, a high pH decreases the depurination rate and elevated tempera-ture accelerates the decomposition rate. Single-stranded DNA is 4-times more susceptible to decomposition than dou-ble-stranded DNA (Lindahl 1993).
DNA can be degraded by a specific enzyme, DNase. DNases are present in the cell cytoplasm to protect the cell from for-eign DNA, for example viruses. DNases also degrade DNA from dead cells into fragments to prevent free DNA circulat-ing in the blood system. There are some bacterial strains, predominately gram- negative species , that have the capacity to produce DNases (Porschen and Sonntag 1974). This is a likely virulence factor for these bacteria as the DNase can degrade the host chromosomal DNA (Minion et al. 1993).
Even though DNA is susceptible to decay, under favourable conditions it can be preserved for extended time periods, as demonstrated by the recovery of DNA from ancient human remains (Pääbo 1985; Tuross 1994). Using PCR methods, Mycobacterium tuberculosis DNA has been detected and amplified from 1000 year-olds mummies (Salo et al. 1994; Nerlich et al. 1997; Donoghue et al. 2005). It is well known that DNA can bind to min-erals and thereby become less prone to decay (Aardema et al. 1983; Lorenz and Wackernagel 1987; Romanowski et al. 1991; Demanéche et al. 2001). In 1956, Tiselius demonstrated that proteins have a specific binding affinity with hydroxyapa-tite and he created a model for separating proteins of different sizes (Tiselius et al. 1956). DNA has also been found to have a specific affinity to bind to hydroxyapa-tite (Bernardi 1965; Martinson 1973), this property has been used during DNA purification where DNA binds to the hydroxyapatite and can be separated from salts and unwanted proteins in a sample (Meinke et al. 1974; Colman et al. 1978). Salts are known to stabilise DNA (Mar-guet and Forterre 1998) but if present in too high concentration they inhibit the PCR reaction (Demeke and Jenkins 2010).
Reliability of molecular techniques in analyses of endodontic samples
There are disparate bacteriological results between studies using culture vs. molecu-lar techniques for analysis of the root canal microbiota. For example, why do molecu-lar techniques apparently pick up readily cultivated bacteria in greater prevalence than conventional culture? Is the reason a higher sensitivity, or are there factors in the root canal that preserve DNA from dead bacteria? Persistence of DNA after
cell-death might lead to false positive results and, in turn, give false information about the current flora. Thus, the kinetics of DNA decomposition from dead micro-bial cells is an important but unexplored process with implications for studies on endodontic infections.
In soft tissues, bacteria and bacterial DNA, will likely be destroyed by phago-cytosis or DNase activity. In hard tissues, like tooth and bone, there is a different and possibly unique situation. In the pulp spaces of teeth with necrotic and infected pulps, phagocytosis cannot occur as there is no blood supply, and the fate of any bac-terial DNA may be influenced by protec-tive factors. It has long been known that hydroxyapatite can bind DNA forming a
stable molecular complex. DNA binding to hydroxyapatite may protect DNA from degradation, which might account for longevity of DNA that has been observed in paleomicrobiological investigations.
When Möller developed his careful methods for sampling and cultivating bac-teria, he sought to answer many problems that had not been previously addressed in a satisfactory way. Today, with molecular analysis of root canal samples, we face many similar questions about reliabil-ity and authenticreliabil-ity of studies conducted using these increasingly popular methods. The studies contained in this thesis exam-ine some interesting questions, in particu-lar regarding the fate of bacterial DNA in the unique environment of the pulp space.
M
olecular methods have revo-lutionised microbial isolation and identification in many clinical disciplines and biology, yet there are some unique conditions applicable to analysis of the root canal microbiota that have attracted limited attention. The fate of DNA from microbes that have been in contact with the pulp space, or deadbac-teria in the root canal is of great relevance for studies in endodontic microbiology and has broader implications for paleom-icrobiology. The overall aim of this thesis was to systematically analyse a series of factors that influence decomposition and preservation of bacterial DNA relating to the root canal environment.
The specific aims of this thesis were:
• To investigate the recovery of amplifiable DNA when dead bacterial cells and free DNA are stored over time.
• To identify factors that affect the decomposition and preservation of DNA. • To examine DNA binding to hydroxyapatite and how hydroxyapatite binding
influ-ences DNA decomposition and preservation.
• To investigate DNA adsorption and release from human dentine and how dentine binding influences DNA decomposition in an in vitro bacterial community.
Overview of the
methodology
I
n this section, the methodology usedin studies I-IV is briefly outlined. Detailed descriptions can be found in the respective papers.
Persistence of DNA from non-viable bacteria
Persistence of DNA from non-viable E. fae-calis in ex vivo root canals
The fate of DNA from bacteria that do not survive in the root canal is uncertain, yet DNA longevity may confound recovery of authentic aetiologic participants in the disease process. The first study assessed the persistence of dead-cell bacterial DNA in ex vivo teeth.
Permanent intact teeth with mature root apices and no cracks were selected from collections of anonymous extracted teeth. After cleaning, teeth were embedded to the cemento-enamel junction in plaster and the root canals were instrumented under strict aseptic conditions. A sus-pension of heat-killed E. faecalis cells was inoculated into the instrumented root canals. After inoculation, the teeth were sealed and stored for up to 2 years. Sam-ples were taken at predetermined intervals during the observation period, processed
by PCR and analysed with gel electropho-resis (I).
Persistence of DNA from non-viable F. nucleatum and P. anaerobius in vitro In the ex vivo study (I), DNA from E. faecalis could be amplified at the end of the observation period (2 years). E. faecalis is a gram-positive facultative, known for its tough cell wall. It is possible that the stable bacterial cell wall retains the DNA intracellularly, protecting it from decom-position. Whether the results seen with E. faecalis are representative and applicable for other bacteria with a different method of cell killing is also of great interest. Therefore, a gram-negative anaerobe (F. nucleatum) and a gram-positive obligate anaerobe (P. anaerobius), both prevalent in the infected root canal were evaluated using an alternative killing method.
In in vitro experiments (II), bacterial cells from Fusobacterium nucleatum and Peptostreptococcus anaerobius were killed by exposure to air, re-suspended and inoculated in tubes and stored in phos-phate buffered saline (PBS) at 37°C for 6 months. Every month, samples were with-drawn and analysed for presence of ampli-fiable DNA by conventional PCR and measured by quantitative PCR (qPCR).
To assess the condition of the cell wall of the bacteria after cell death, scanning electron microscopy (SEM) was per-formed on viable cells, newly killed and cells that had been dead for 4 months (II). To further examine if there are differ-ences in stability between the gram-neg-ative and gram-positive cell wall after cell death, permeability of the different cell walls were assessed by treatment of the non-viable bacteria with the DNA bind-ing dye Propidium Monoazide (PMA) and also by vital staining (LIVE/DEAD) measured by flow cytometry (II).
Decay of DNA
DNA from dead bacteria can be detected by PCR a long time after cell death (I, II). Even though DNA remained cell-as-sociated after months of incubation, DNA gradually leaked out into the surrounding environment. In the root canal, dead cells and free DNA released from these cells may be exposed to various environmen-tal factors that might influence the DNA decomposition.
The impact of some factors on DNA degradation was further assessed to deter-mine if there was a difference in decay rate between cell-bound and extracellular DNA. In an in vitro assay (I), heat-killed E. faecalis cells (cell-bound DNA) and extracted E. faecalis DNA (free DNA) were inoculated into water, human sera, DNase, Elution solution and a culture of a DNase-producing species (Prevotella intermedia). At predetermined intervals (up to 3 weeks) recovery of DNA was assessed by gel electrophoresis.
Stabilisation of DNA Hydroxyapatite
After cell death, DNA was released into the surrounding environment (II) and
this extracellular DNA was prone to decomposition. It has previously been reported that DNA can bind to min-erals and in that complex the DNA is more resistant to decay. Hydroxyapa-tite is a major component of dentine, so it is relevant to confirm DNA binding to this mineral and whether this bind-ing stabilises naked DNA from degra-dation. F. nucleatum DNA was bound to hydroxyapatite by incubating the DNA with beads of ceramic hydroxyapatite. The DNA-coated hydroxyapatite was washed thoroughly and inoculated in dif-ferent media (water, sera). DNA bound to hydroxyapatite was also exposed to DNase I. At predetermined intervals (up to 3 months), recovery of amplifiable DNA was assessed by releasing the DNA from the hydroxyapatite beads using EDTA and evaluating the presence of residual DNA by PCR amplification and gel electropho-resis (III).
Human dentine
Hydroxyapatite was shown to have a pro-tective effect on DNA bound to it (III). The next logical step was to assess DNA binding to dentine and whether this influ-ences DNA preservation. Dentine con-tains many other components that may influence the protective effect on DNA.
Dentine shavings were collected from recently extracted teeth under aseptic conditions. Free F. nucleatum DNA was bound to the dentine shavings and the dentine-bound DNA was exposed to the same factors as in the hydroxyapatite study (III). Recovery of detectable DNA was assessed by releasing the DNA from the dentine by addition of EDTA and amplification of recovered DNA with PCR. Samples were analysed by gel elec-trophoresis (IV). As collagen type I is a notable component of dentine, DNA was
bound to collagen type I and treated in the same way as the dentine-bound DNA. DNA was released from collagen by addi-tion of Proteinase K. Recovery of detect-able DNA was assessed by PCR and gel electrophoresis (IV).
Stabilisation of bacterial DNA bound to human dentine was also assessed in an
in vitro bacterial community. Extracellu-lar F. nucleatum DNA and F. nucleatum DNA bound to dentine was exposed (3 months) to a mix of bacteria where one of the strains was DNase producing (P. intermedia). At predetermined intervals DNA recovery was assessed by PCR and gel electrophoresis (IV).
DNA preservation
Heat killed E. faecalis was inoculated in instrumented root canals of teeth ex vivo. The teeth were sealed and stored for up to two years. At 3, 6, 12 and 24 months root canal samples were collected and analysed by PCR. In 21/25 teeth DNA was still amplifiable two years after cell-death (I) (Table 1).
Young et al. found similar results when they in an in vitro experiment showed that DNA from heat-killed E. faeca-lis was detectable 1 year post cell-death when stored in phosphate buffered saline (Young et al. 2007).
Results and Discussion
In the indigenous root canal infection, the composition of the flora usually con-sists of equal proportions of gram-positive and gram-negative cells. In endodonti-cally treated teeth refractory to treatment, culture-based studies have shown that the residual infection is composed mainly of gram-positive bacteria. The most fre-quently found is E. faecalis (Molander et al. 1998; Sundqvist et al. 1998; Hancock et al. 2001; Siqueira and Rôças 2004). One explanation could be the robust cell characteristics that enable E. faecalis to withstand the endodontic treatment.
Whether this protects prokaryotic DNA Post-inoculation time (mo) Experimental teeth* Control teeth *
3 9/9 0/3
6 10/10 0/3
12 14/15 0/3
24 21/25 0/5
Table 1. Recovery of amplifiable Enterococcus faecalis DNA from ex vivo human teeth.
from degradative factors after cell death is an interesting question. The cell walls of gram-negative species are composed dif-ferently and are generally not considered to have the same stability.
Whether DNA from gram-negative and other gram-positive bacteria have comparable patterns to E. faecalis and per-sist long after cell-death was investigated in vitro (II). One of the most commonly found gram-negative species in untreated root canal infections, the strict anaerobic Fusobacterium nucleatum (Sundqvist 1994; Chugal et al. 2011), was killed by exposure to air. The bacterial cells were then stored in PBS for up to 6 months. Every month conventional PCR and qPCR was used to assess persistent amplifiable DNA. DNA from F. nucleatum was clearly detectable 6 months post cell death even though a decline of DNA content was noted over time. A gram-positive species (Peptos-treptococcus anaerobius) was also investi-gated and similar results were obtained. This is in line with a study using E. fae-calis where a 1000-fold decline of DNA
content was reported during the 1-year experimental period (Young et al. 2007). Cell bound – free DNA
When bacteria die, the cell wall degrades resulting in leakage of cell contents. Scan-ning electron microscopic examination of killed F. nucleatum cells (II) revealed that after exposure to air there was some shriv-elling, which increased over 4 months of incubation in PBS (Fig. 4). PCR analysis showed recoverable DNA in the cell pellet after 6 months incubation, but it was clear that the amount of cell-associated DNA had decreased over time. The concentra-tion of extracellular DNA in the super-natant rose progressively, indicating that the DNA had been released into the sur-rounding media from the damaged cells. The corresponding results for P. anaero-bius showed less pronounced deformation of the cell after 4 months of incubation (Fig. 4), however, DNA was found leak-ing out in the surroundleak-ing media (II).
DNA can thus appear in two forms. Cell-bound, i.e. the DNA is still located
Figure 4. SEM image of F. nucleatum
and P. anaerobius cells A. Viable F.
nucleatum B. Dead F. nucleatum
incu-bated in PBS for 4 months C. Viable
P. anaerobius D. Dead P. anaerobius
incubated in PBS for 4 months. Original magnification x 50000.
inside the bacteria, and free or extracellu-lar, i.e. released into the external environ-ment after cell lysis. Exactly how bacteria die in the root canal and what is the fate of DNA is not known, yet it is likely that both forms occur in infections in vivo.
The ability for DNA to persist in the root canal after cell death will depend on the form of the DNA, the root canal milieu, and whether the DNA will be exposed to factors that enhance or retard decay.
Cell-bound DNA
Heat-killed E. faecalis was inoculated in water, sera and elution solution, a media for storage of extracted DNA, and the rate of decomposition was compared to free DNA extracted from the cell. In vitro, DNA from heat killed E. faecalis was pre-served compared with extracellular DNA when exposed to different media (I). Cell-bound DNA was not degraded during the experimental period of three weeks (Table 2).
Cell-bound and free DNA was also exposed to DNase. Free DNA was degraded within hours both when exposed to a high concentration of DNase I and when exposed to a culture of the DNase producing Prevotella intermedia.
Cell-bound DNA was not notably affected by the DNase treatments (Table 2). With heat-killed cells (I) (60°C; 36-72h), the cell wall is most likely still intact as heat-treated bacteria lose their ability to repro-duce but still have intact cell walls. Kort et al. shocked gram-positive Bacillus sub-tilis at different temperatures and labelled the cells with propidium iodide to iden-tify cells with compromised membranes. Above 59°C cells were unable to grow but the cell membrane was impermeable for propidium iodide (Kort et al. 2008). Free DNA
The DNA molecule is subject to decaying factors. The in vitro experiments showed that extracellular DNA was prone to decomposition (I). Suspended in water, free DNA decomposed within days and sera and DNases enhanced the decom-position rate (Table 2). Free DNA was only stable in elution solution. DNA in free form was also prone to decay when exposed to DNases (I, III, IV)(Table 2, 3). Free DNA exposed to cultures including a DNase producing strain, was degraded within hours (I, IV) (Table 2, Fig. 5).
The DNA molecule is released from the bacterial cell at various times after cell DNase I Sera Sera Water Elution P. intermedia
(10%) (50%) solution
Extracted DNA < 30 min < 24 h < 24 h 7-13 d > 21 d < 3 Cell-bound DNA No decay in > 21 d > 21 d > 21 d > 21 d > 7 d*
30 min
* Assessed by polymerase chain reaction (PCR).
Table 2 E. faecalis DNA decay in vitro in various media.
death (II). The ability for the naked DNA molecule to persist over time is depend-ent on factors in the environmdepend-ent. An acidic pH, UV radiation, availability of oxygen, and high temperature are some factors known to affect DNA decom-position (Aoki et al. 1966; Clark et al. 1966; Lindahl and Nyberg 1972; Lindahl 1993; Höss et al. 1996; Burger et al. 1999; Edwards et al. 2004). DNases in the envi-ronment can efficiently cleave the exposed DNA molecule into fragments (Hell et al. 1972). Despite this, prokaryotic DNA from ancient specimens has been recov-ered long time after cell death (Salo et al. 1994; Nerlich et al. 1997; Drancourt et al. 1998).
As soft tissue decomposes quite rapidly, DNA from mineralized tissues are more suitable as targets in detecting DNA from ancient specimens (Lassen et al. 1994). The protection of the DNA molecule might be due to physical protection in bone and tooth from decomposing factors, but there is also the possibility of binding to minerals resulting in resistance to decay. The binding of DNA to minerals in clay is a well-known phenomenon (Goring and
Bartholomew 1952; Greaves and Wilson 1969; Lorenz and Wackernagel 1987; Cai et al. 2008). It has been shown that DNA bound to minerals is much more resistant to decay (Aardema et al. 1983; Paget et al. 1992; Cai et al. 2008). Hydroxyapa-tite, the predominant mineral in teeth and bone, can bind DNA (Bernardi 1965; Martinson 1973) and might be a factor of importance in preserving DNA over time. Stabilisation of free DNA by binding to minerals
Hydroxyapatite
The possibility for bacterial DNA to bind to ceramic hydroxyapatite (HA) and whether DNA bound to HA was more resistant to decay than free DNA in solu-tion was investigated (III). It was evident that HA had an affinity to bind DNA.
The uptake of DNA to HA occurred rapidly and within 90 minutes, the DNA was no longer detectable in the surround-ing media. The bindsurround-ing of DNA to min-erals is likely by ion attraction and the binding capacity is favoured of the pres-ence of cations (Lorenz and Wackernagel
HA bound DNA + + + Dentine bound DNA + + + Collagen bound DNA + + + Extracellular DNA – – +
Effect of Spontaneous Influence nucleases degradation of sera
(3 months) (3 months)
Table 3. Recovery of free and HA-, dentine-, or collagen-bound DNA after incubation with
1987; Paget et al. 1992; Poly et al. 2000) that act like a bridge between the nega-tively charged ions on the mineral and the backbone of DNA (Pietramellara et al. 2001). It is also suggested that enhanced adherence capacity in a low pH (Goring and Bartholomew 1952; Greaves and Wilson 1969; Lorenz and Wackernagel 1987; Khanna and Stotzky 1992) is due to protonation of the negatively charged DNA. This will result in an inhibition of the repelling forces to the negatively charged mineral surface (Khanna and Stotzky 1992).
Once bound to the mineral the DNA is less prone to decay. The binding to HA had a stabilising effect on spontaneous decomposition (Table 3) but also, DNA bound to HA was not decomposed when exposed to DNase I (III). It has previously been shown that DNA bound to clay is protected against DNase activity (Khanna and Stotzky 1992; Paget et al. 1992; Demanéche et al. 2001). The exact mech-anism of protection of adsorbed DNA against enzymatic degradation is not clar-ified. It is suggested the DNA molecule adsorbs with segments in contact with the mineral surface, alternating with seg-ments not in contact with the mineral and thereby available for enzymes (Paget et al. 1992). As part of the DNA chain is bound to the mineral surface, this cannot be cut by the DNase. After releasing the DNA from the mineral, fragments of DNA will be available for PCR amplification.
It has also been reported that minerals can adsorb enzymes, including DNase I (Paget et al. 1992; Cai et al. 2008). It seems reasonable to conclude that hydroxyapa-tite had the capacity to bind DNase I and render it unable to digest DNA (III). The protection of hydroxyapatite bound DNA against DNase activity is probably a com-bination of stabilisation of DNA due to
binding to HA and the capacity of HA to bind and inactivate DNase I.
DNases can be present in root canals and may originate from sera (Kurnick 1953; Miyauchi et al. 1989; Aitken et al. 1992) that perfuses through the apical foramen, or from bacterial species able to produce DNases (Porschen and Sonn-tag 1974; Leduc et al. 1995). Free DNA exposed to sera (10% and 50%; 24h) was decayed and could not be detected by gel electrophoresis (I) but when using PCR amplifiable DNA was still recoverable after three months incubation (III, IV). DNA bound to hydroxyapatite was not noticeably influenced by exposure to sera (III) (Table 3).
Human dentine
The stabilising effect by HA is interesting as hydroxyapatite is a major component of dentine (Bowes and Murray 1935; Ruse and Smith 1991). DNA had a binding affinity also to dentine (IV) and this made the DNA molecule less prone to decay by DNase I (Table 3), comparable with DNA bound to pure HA crystals (III).
Approximately 40–60 % of the dentine is hydroxyapatite (Bowes and Murray 1935; Ruse and Smith 1991) but there is also a notable quantity of proteins where collagen type I is common (Garant 2003; Jagr et al. 2012). Collagen also has the affinity to bind DNA (Izui et al. 1976). It was observed that DNA had a bind-ing affinity to collagen (IV). Spontane-ous decomposition of DNA was inhibited when bound to collagen and the colla-gen-DNA complex showed a stabilis-ing effect against enzymatic degradation (Table 3). Apart from collagen, other den-tine proteins might influence the binding strength between DNA and dentine, but additional factors were not investigated.
that dentine-bound DNA was also pro-tected against degradation when exposed to DNase I in high concentration (IV). The possibility of dentine to protect DNA from degradation was assessed in an in vitro bacterial community including a strain known to produce DNases (IV).
The species Pseudoramibacter alacto-lyticus, Parvimonas micra and the DNase producing Prevotella intermedia were sus-pended in 10% sera in a concentration of 106 to 108 cfu/mL. Free F. nucleatum DNA bound to dentine shavings were inoculated in the suspension for up to three months. At predetermined intervals, recovery of DNA was assessed by PCR and gel electrophoresis. After three months of incubation the dentine-bound
DNA was still detectable by PCR (Fig. 5). In controls, where free DNA not bound to dentine was exposed to the same bacterial mix, DNA decomposed to below detection limit within hours (Fig. 5). In this experiment it was noted that P. intermedia did not survive the whole experimental period, which may have caused a failure of continuous DNase production. To overcome this, fresh P. intermedia was added to the bacterial mix every other week. DNA bound to dentine was still detectable throughout the experiment but weaker bands on the gel were evident over time, which could mean that there is a dose-response relationship between the DNase produced by P. intermedia and the degradation of DNA (Fig 5).
Figure 5. Recovery of dentine-bound DNA after exposure to an in vitro bacterial community. A. DNA bound to dentine exposed to bacterial mix containing P. alactolyticus, P. micra and P.
intermedia in 10% sera. B. DNA bound to dentine exposed to bacterial mix as above with addition
General Discussion
S
tudies of the endodontic microbi-ota using molecular techniques have become increasingly popular over the last 15 years and the reasons are easy to understand—the method is straightfor-ward and the results are rapid and highly specific for target DNA. Nevertheless, the rigorous approach established almost half a century ago by Möller for the cul-tivation technique has not been applied to the molecular methods of today. The results shown herein demonstrate that the pulp space has some unique properties that make microbial sampling from a root canal infection quite different to samples isolated from other clinical sites.Findings from molecular studies of the root canal microbiota have shown a much more diverse flora compared to studies using advanced culture techniques. There may be numerous explanations for the variation in results, and two commonly given reasons for the difference are that the DNA-based techniques are more sen-sitive than the cultivation method and that the former detects bacteria that are not cultivable. However, these reasons cannot explain some observations, such as higher prevalence of species that are read-ily cultivable. There are other factors that
influence the results. One is the limita-tion of molecular methods to distinguish between viable and dead cells. Bacteria die in the root canal and detecting DNA from these non-viable organisms does not necessarily mean that this organism is alive and involved in the disease.
The studies in this thesis have shown that a number of factors that relate to the pulp space may have the capacity to delay DNA decay. Thus, in some situations the results from PCR-based analysis of root canal samples might be a historical reflection of bacteria that have entered the pulp space but not been able to survive the infection.
Both cultivation and PCR methods have limitations that may confound acquisition of reliable data on the root canal micro-biota. Whilst a principal benefit of the cultivation technique is that detected bac-teria can undoubtedly be claimed viable, a drawback is that some bacteria are diffi-cult to diffi-cultivate, or are not-yet-diffi-cultivable, which can produce false negative results. The PCR-based approach is highly spe-cific, but a drawback of conventional PCR is its inability to distinguish between DNA from viable and dead cells. This limitation may lead to recovery of species
assumed to be involved, but are actually dead or uninvolved in the infection.
There are several non-cultivation meth-ods claimed to distinguish viable from dead cells. Some methods are based on demonstration of cellular metabolic activ-ity (Lleo et al. 2000), others on cell wall integrity by dyes that only can enter a compromised cell wall and bind to the nucleic acid (Caron et al. 1998; Stocks 2004). Viability staining can be done separately and in combination, and be examined by flow cytometry, confocal laser scanning microscopy or fluorescence microscopy. The advantages that certain dyes only can enter a compromised cell wall has also been used in DNA-based detection methods. The photo-reactive DNA-binding dye Propidium Monoazide (PMA) has been used to separate viable from damaged cells in the PCR process. The dye has the ability to bind free DNA and DNA from cells with compromised cell walls, but cannot penetrate the intact wall of viable cells. After exposure to light energy, the DNA is not amplifiable (Nocker et al. 2006).
When bacteria die, the cell wall may remain intact and impermeable to stains for some time. Cells killed by exposure to isopropyl alcohol have been shown to lose their cell wall integrity within minutes (Sánchez et al. 2013) as alcohol will affect the solubility of lipids in the wall. Escher-ichia coli treated with lethal temperatures ≤72° C results in dead cells not permeable to nucleic staining dyes (Yang et al. 2011). The finding that a relatively mild, lethal heat treatment resulted in cell-death but not a permeable cell wall has been reported for Listeria innocua (Løvdal et al. 2011) and Bacillus subtilis (Kort et al. 2008). In experiments here, for air-killed F. nucleatum and P. anaerobius, the cell walls were impermeable for nucleic
stain-ing dyes for 3–7 weeks, respectively, after loss of cell viability. Thus, using the PMA technique, DNA from dead bacteria with intact cell walls will not be stained and may be a source of fault.
Metabolic activity in bacteria, shown by the presence of mRNA, an unstable mole-cule with short half-life is another molec-ular indicator of cell viability (Donovan and Kushner 1986; Sheridan et al. 1998). The RNA molecule is extracted from the bacterial cell and by using RT-PCR, com-plementary DNA (cDNA) is produced and can be used in the PCR reaction (Lynas et al. 1989). Amplifiable DNA is a marker that the cell produces RNA, and is a sign that the bacteria are viable (Vait-ilingom et al. 1998; Zhao et al. 2006). By this method, viable but not cultivable cells can be differentiated from dead cells (Pai et al. 2000). The technique is delicate due to the instability of the mRNA molecule (Deutscher 2006) as the RNA is sensitive to degradation in the sample process and can be lost by inadequate handling and storage of samples (Holland 2003; Pérez-Novo et al. 2005). These limitations may be the reason that only a few reports have been presented using this method when analysing root canal flora (Williams et al. 2006; Rôças and Siqueira 2010).
The results from the investigations con-tained in this thesis indicate that DNA from non-viable bacteria can be preserved for long time periods. A binding affinity of DNA for dentine and specifically the hydroxyapatite component will preserve DNA from decomposition and protect it from degradation by nucleases.
The key elements for establishing a rig-orous basis for sampling of the root canal microbiota using a molecular approach involves defining factors that influence DNA preservation and decomposition in this unique environment. This work has
provided basic information, which can be applied to develop refined methods and strategies. By clarifying these factors, it should lay a suitable framework to develop
improved methods that are specifically customised to sampling of the root canal microbiota.
Acknowledgements
I
wish to express my sincere gratitude tothose who have helped and supported me on this journey. Without friends and colleagues this project would not have been possible.
To my main supervisor and friend Ulf Sjögren. I am so happy that you intro-duced me to this interesting field of research! Thank you for your encourage-ment when things were tough and for restraining me when my ideas were too many and scattered. Thank you for keep-ing me on track for all of these years!! I am incredibly grateful for everything you have done for me.
To my co-supervisor David Figdor. Thank you David for all your help and support in my projects. Our fruitful dis-cussions over Skype, your great knowl-edge of the field and your excellent advice have been invaluable for me. Additionally, your English is excellent.
To my co-supervisor Anders Johansson. Thank you for sharing your scientific expertise, for all your advice and for your help in interpreting results. Thanks for supporting me when the stress broke the meter… and for explaining that it is okay to correct your supervisor when you know you know better.
To my dear friend, co-worker and super-visor at the lab Chrissie Roth. We have had so much fun at the lab! I have tried to pick up your laboratory tech skills but, of course, it is hard to keep up with someone with as much experience and knowledge as you have. Thank you for your friendly explanations when things went wrong… What would I have done without you!? To my mentor Göran Sundqvist. Thank you for helping design experiments and answering my strange questions. Thank you for being my co-author and teaching me how to conduct research the Umeå Way.
Rolf Claesson. Thank you for listening to my theories, filling in the blanks and for always being available when strange bacteria needed to be characterized. Also, thanks for all the words of wisdom and great laughs.
To my friend and helping hand Kerstin Lundgren. Thank you Kerstin for great assistance at the clinic and keeping track of the administrative details of my patients. But most of all, thank you for all the small talk and updates on clinic gossip.
Many thanks to my colleagues Majid Ebrahimi, Hans Ingridsson, Jon Lind-gren and Richard Johansson for covering for me at the clinic when I was not there. A special thank you to Jon, who helped me with some experiments in Paper IV.
Pernilla and Karin. Thank you for help-ing me with materials! But most of all, thank you for being my friends. It is nice to have somebody who knows what it is like, who can share my enthusiasm when things are at their best, but also who knows how to cheer up a tired, sad, frus-trated woman when she second-guesses herself. Girls, it’s time for a meeting…. Lillemor Hägglund and Anita Sevä. Thanks for help with incomprehensible paperwork.
Many thanks to Margareta Molin Thorén, Eva Utterberg and Susanna Marklund for their goodwill and support in this project.
My mother, Birgitta. Thank you for believing in me and always supporting me in my escapades! If mothers were flowers, I would have picked you!
To my family, David, Maja and Lucas. Thanks for your patience with the con-fused, tired, asocial wife/mother who you have lived with you for the past years. I love you!
Aardema BW, Lorenz MG, Krumbein WE (1983) Protection of Sediment-Adsorbed Transforming DNA Against Enzymatic Inactivation. Applied and Environmental
Microbiology 46, 417-20.
Aas JA, Paster BJ, Stokes LN, Olsen I, Dewhirst FE (2005) Defining the normal bacterial flora of the oral cavity. Journal of
Clinical Microbiology 43, 5721-32.
Aitken ML, Burke W, McDonald G, Shak S, Montgomery AB, Smith A (1992) Recom-binant human DNase inhalation in normal subjects and patients with cystic fibrosis. A phase 1 study. The Journal of the American
Medical Association 267, 1947-51.
Aoki S, Boyce RP, Howard-Flanders P (1966) Sensitization of Escherichia coli to radiation by bromouracil: excessive post-irradia-tion breakdown of deoxyribonucleic acid without concomitant synthesis. Nature
209, 686-8.
Bernardi G (1965) Chromatography of nucleic acids on hydroxyapatite. Nature
206, 779-83.
Bowes JH, Murray MM (1935) The chemical composition of teeth: The composition of human enamel and dentine. Biochemical
Journal 29, 2721-7.
Burger J, Hummel S, Herrmann B, Henke W (1999) DNA preservation: a microsatel-lite-DNA study on ancient skeletal remains.
Electrophoresis 20, 1722-8.
Cai P, Huang Q, Li M, Liang W (2008) Binding and degradation of DNA on montmo-rillonite coated by hydroxyl aluminum species. Colloids Surfaces B: Biointerfaces 62, 299-306.
Caron GN, Stephens P, Badley RA (1998) Assessment of bacterial viability status by flow cytometry and single cell sorting.
Journal of Applied Microbiology 84, 988-98.
Cha RS, Thilly WG (1993) Specificity, effi-ciency, and fidelity of PCR. PCR Methods
and Applications 3, S18-29.
Chugal N, Wang JK, Wang R et al. (2011) Molecular characterization of the micro-biota residing at the apical portion of infected root canals of human teeth.
Jour-nal of Endodontics 37, 1359-64.
Clark AJ, Chamberlin M, Boyce RP, How-ard-Flanders P (1966) Abnormal metabolic responce to ultraviolet light of a recombi-nation deficient mutant of Escherichia Coli K12. Journal of Molecular Biology 19, 442-54.
Colman A, Byers MJ, Primrose SB, Lyons A (1978) Rapid purification of plasmid DNAs by hydroxyapatite chromatography.
Euro-pean Journal of Biochemistry 91, 303-10.
Conrads G, Gharbia SE, Gulabivala K, Lampert F, Shah HN (1997) The use of a 16s rDNA directed PCR for the detection of endodontopathogenic bacteria. Journal of
Endodontics 23, 433-8.
Corless CE, Guiver M, Borrow R, Edwards-Jones V, Kaczmarski EB, Fox AJ (2000) Contamination and sensitivity issues with a real-time universal 16S rRNA PCR. Journal
of Clinical Microbiology 38, 1747-52.
Demanéche S, Jocteur-Monrozier L, Quiqua-mpoix H, Simonet P (2001) Evaluation of biological and physical protection against nuclease degradation of clay-bound plasmid DNA. Applied and Environmental
Microbiology 67, 293-9.
Demeke T, Jenkins GR (2010) Influence of DNA extraction methods, PCR inhibitors and quantification methods on real-time PCR assay of biotechnology-derived traits.
Analytical and Bioanalytical Chemistry 396,
1977-90.
Deutscher MP (2006) Degradation of RNA in bacteria: comparison of mRNA and stable RNA. Nucleic Acids Research 34, 659-66.
Donoghue HD, Marcsik A, Matheson C et al. (2005) Co-infection of Mycobacterium
tuberculosis and Mycobacterium leprae in
human archaeological samples: a possible explanation for the historical decline of leprosy. Proceedings of the Royal Society B
272, 389-94.
Donovan WP, Kushner SR (1986) Polynucle-otide phosphorylase and ribonuclease II are required for cell viability and mRNA turnover in Escherichia coli K-12. Proceedings
of the National Academy of Sciences of the United States of America 83, 120-4.
Downes J, Wade WG (2006)
Peptostrepto-coccus stomatis sp. nov., isolated from the
human oral cavity. International Journal of
Systematic and Evolutionary Microbiology
56, 751-4.
Drancourt M, Aboudharam G, Signoli M, Dutour O, Raoult D (1998) Detection of 400-year-old Yersinia pestis DNA in human dental pulp: an approach to the diagnosis of ancient septicemia. Proceedings of the
National Academy of Sciences of the United States of America 95, 12637-40.
Edwards CJ, MacHugh DE, Dobney KM et al. (2004) Ancient DNA analysis of 101 cattle remains: limits and prospects. Journal of
Archaeological Science 31, 695-710.
Fabricius L, Dahlén G, Öhman AE, Möller ÅJ (1982) Predominant indigenous oral bacteria isolated from infected root canals after varied times of closure. Scandinavian
Journal of Dental Research 90, 134-44.
Faveri M, Mayer MP, Feres M, de Figueiredo LC, Dewhirst FE, Paster BJ (2008) Microbio-logical diversity of generalized aggressive periodontitis by 16S rRNA clonal analysis.
Oral Microbiology and Immunology 23,
112-8.
Fleige S, Pfaffl MW (2006) RNA integrity and the effect on the real-time qRT-PCR performance. Molecular Aspects of Medicine
27, 126-39.
Garant P (2003) Dentin. In: Oral cells and
tissues, pp. 25-52: Quintessence Publishing
Co Lt.
Gonçalves RB, Mouton C (1999) Molecu-lar detection of Bacteroides forsythus in infected root canals. Journal of Endodontics
25, 336-40.
Goring CAI, Bartholomew WV (1952) Adsorp-tion of mononucleotides, nucleic acids, and nucleoproteins by clays. Soil Science
Greaves MP, Wilson MJ (1969) The adsorp-tion of nucleic acids by montmorillonite.
Soil Biology and Biochemistry 1, 317–23.
Götherström A, Collins M, Angerbjörn A, Lidén K (2002) Bone preservation and DNA amplification. Archaeometry 44, 395-404. Hancock HH, III, Sigurdsson A, Trope M,
Moiseiwitsch J (2001) Bacteria isolated after unsuccessful endodontic treatment in a North American population. Oral Surgery,
Oral Medicine, Oral Pathology, Oral Radiol-ogy and Endodontics 91, 579-86.
Hell A, Birnie GD, Slimming TK, Paul J (1972) Controlled fragmentation of DNA by DNase I. Analytical Biochemistry 48, 369-77. Higuchi R, Bowman B, Freiberger M, Ryder
OA, Wilson AC (1984) DNA sequences from the quagga, an extinct member of the horse family. Nature 312, 282-4.
Holland NT, Smith MT, Eskenazi B, Bastaki M (2003) Biological sample collection and processing for molecular epidemiological studies. Mutation Research 543, 217-34. Höss M, Jaruga P, Zastawny TH, Dizdaroglu
M, Pääbo S (1996) DNA damage and DNA sequence retrieval from ancient tissues.
Nucleic Acids Research 24, 1304-7.
Izui S, Lambert PH, Miescher PA (1976) In
vitro demonstration of a particular affinity
of glomerular basement membrane and collagen for DNA. A possible basis for a local formation of DNA-anti-DNA com-plexes in systemic lupus erythematosus.
The Journal of Experimental Medicine 144,
428-43.
Jágr M, Eckhardt A, Pataridis S, Mikšík I (2012) Comprehensive proteomic analysis of human dentin. European Journal of Oral
Sciences 120, 259-68.
Jervøe-Storm PM, Koltzscher M, Falk W, Dörfler A, Jepsen S (2005) Comparison of
culture and real-time PCR for detection and quantification of five putative periodonto-pathogenic bacteria in subgingival plaque samples. Journal of Clinical Periodontology
32, 778-83.
Josephson KL, Gerba CP, Pepper IL (1993) Polymerase chain reaction detection of nonviable bacterial pathogens. Applied and
Environmental Microbiology 59, 3513-5.
Joyce E, Phull SS, Lorimer JP, Mason TJ (2003) The development and evaluation of ultrasound for the treatment of bacterial suspensions. A study of frequency, power and sonication time on cultured Bacillus species. Ultrasonics Sonochemistry 10, 315-8.
Kakehashi S, Stanley HR, Fitzgerald RJ (1965) The Effects of Surgical Exposures of Dental Pulps in Germ-Free and Conventional Lab-oratory Rats. Oral Surgery, Oral Medicine,
Oral Pathology 20, 340-9.
Kantz WE, Henry CA (1974) Isolation and clas-sification of anaerobic bacteria from intact pulp chambers of non-vital teeth in man.
Archives of Oral Biology 19, 91-6.
Khanna M, Stotzky G (1992) Transformation of Bacillus subtilis by DNA bound on mont-morillonite and effect of DNase on the transforming ability of bound DNA. Applied
and Environmental Microbiology 58, 1930-9.
Kim SY, Shin Y, Lee CY, Jung IY (2013) In Vivo quantitative evaluation of live and dead bacteria in root canal infection by using propidium monoazide with real-time PCR.
Journal of Endodontics 39, 1359-63.
Kobayashi H, Oethinger M, Tuohy MJ, Procop GW, Hall GS, Bauer TW (2009) Limiting false-positive polymerase chain reaction results: detection of DNA and mRNA to differentiate viable from dead bacteria.
Diagnostic Microbiology and Infectious Disease 64, 445-7.
