1

1 2 3 4 5

Lantibiotic Production by Streptococcus mutans: Their Uses in Replacement Therapy for 6

the Prevention of Dental Caries and as Antibiotics for the Treatment 7

of Various Infectious Diseases 8

9 10 Authors: 11 James L. Smith 12 Ravi S. Orugunty 13 Jeffrey D. Hillman 14 15 Affiliation: 16 Oragenics, Inc. 17 13700 Progress Blvd. 18 Alachua, FL 32615 19 20 Corresponding Author: 21 J.D. Hillman 22 jhillman@oragenics.com 23 Telephone (386) 418-4018 24 Fax (386) 418-1660 25 26 27 28 Running Title: Replacement Therapy and Bacteriocins 29 30 31 32 33 34 35 36 37 38 39 40 41

2

Abstract 1

Five billion people worldwide suffer from dental caries, making it the most common, 2

chronic infectious disease of humankind. Mutans streptococci (MS) are the primary etiological 3

agent of dental caries. Streptococcus mutans strain JH1000 was originally isolated based on its 4

superior ability to colonize teeth, which was due to its production of a bacteriocin capable of 5

killing all other strains of S. mutans. Its cariogenic potential was essentially eliminated by 6

genetically modifying its ability to produce lactic acid, which is directly responsible for eroding 7

tooth mineral. The resulting strain is intended for use in a novel method for prevention of caries 8

called “replacement therapy”. A single application of the replacement strain is expected to 9

provide lifelong protection from most human decay by displacing disease-causing strains that are 10

naturally present on tooth surfaces. Once established in the MS niche, it will also prevent 11

colonization by disease-causing strains whenever the host comes in contact with them. The 12

bacteriocin produced by the replacement strain belongs to the lantibiotic family. These are an 13

important group of antimicrobials compounds. In addition to discussing our progress with 14

replacement therapy, we detail the current state of the art regarding the potential use of these 15

antibiotics as therapeutic agents. 16

3

Lantibiotic Production by Streptococcus mutans: Their Uses in Replacement Therapy for 1 the Prevention of Dental Caries and as Antibiotics for the Treatment of Various Infectious 2 Diseases 3 4 Dental Caries and Mutans Streptococci (MS) 5 6 Dental caries is the most common, chronic infectious disease of mankind. It affects 7

approximately 5 billion people worldwide. Despite the general perception that the incidence of 8

tooth decay has declined dramatically due to fluoride, sealants, and other preventative measures, 9

recent evidence indicates that the decline is really quite modest. During the period between 1999 10

and 2002, 41% of children aged 2-11 years had dental caries in their primary teeth, which is 11

unchanged since the 1988-94 reporting period. During the same period, 42% of children and 12

adolescents aged 6-19 years and approximately 90% of adults had dental caries in their 13

permanent teeth, representing a modest 10% and 6% reduction, respectively, since 1988-94 14

(Beltran-Aguilar et al., 2005). A recent Report by the Surgeon General on Oral Health in 15

America (2000) called dental caries ‘the silent epidemic’ and stated that those who suffer the 16

worst oral health are found among the poor of all ages, with poor children and poor older 17

Americans being particularly vulnerable. The caries incidence in Western Europe is comparable 18

to that in the U.S., while Japan reports a caries rate almost double that in the U.S. Certainly, 19

many developing countries are facing an explosion in caries rates as diets and other socio-20

economic factors change rapidly. 21

Dental caries is considered a multifactorial disease, meaning that diet, host factors and 22

bacteria all influence the incidence and severity of the disease. It is well established that 23

frequent consumption of large amounts of dietary sugar predisposes an individual to decay. 24

Similarly, fluoride ingestion during tooth formation and topically after tooth eruption promotes 25

enamel and dentine that are more resistant to solubilization by acids produced by bacteria. 26

4

Germfree animals that completely lack a bacterial flora do not develop dental caries regardless of 1

the amount of sugar in their diet, indicating an absolute role for bacteria in the etiology of this 2

disease. This provides the basis for most conventional approaches to preventing tooth decay, 3

such as brushing and flossing, where the goal is to reduce the number of cariogenic bacteria 4

below the level necessary for them to initiate clinically manifested disease (the so-called 5

minimum pathogenic dose). 6

In 1924, Clarke isolated a streptococcus species that he associated with human dental 7

caries lesions (Clarke, 1924; Loesche, 1986). He called these isolates Streptococcus mutans to 8

indicate their unusual oval shape when Gram stained. More than 3 decades passed before a 9

causal relationship was established. Keyes (Keyes, 1960; Loesche, 1986) observed that hamsters 10

harboring a non-cariogenic oral flora would develop caries when they were placed in cages with 11

caries active hamsters. He also observed that this did not occur if the caries active hamsters were 12

first given penicillin and erythromycin. Since the caries free hamsters were previously known to 13

contain a complex bacterial oral flora and did not show any of the signs of dental caries, Keyes 14

concluded that the caries active hamsters must have transmissible bacteria that cause the disease. 15

This finding was the basis for extensive research into the identification of the etiologic agent(s) 16

responsible for dental caries. Pioneering work in several laboratories (Fitzgerald, 1968; 17

Fitzgerald and Keyes, 1960; Keyes, 1968) demonstrated that S. mutans was highly cariogenic in 18

animal models, and it was subsequently demonstrated that it could be routinely associated with 19

human decay (Loesche, 1982). 20

Initially, S. mutans was subdivided into a number of subspecies based on differences in 21

surface antigenicity (Bratthall, 1970). More recently, S. mutans has been subdivided into seven 22

distinct species (S. mutans, S. sobrinus, S. rattus, S. cricetus, S. ferus, S. macacae and S. downei) 23

5

(Chikindas et al., 1997; Loesche, 1986; Whiley et al., 1988) based on their genetic and 1

phenotypic differences (Coykendall, 1989). However, since all of these species are very similar 2

and have the potential to cause dental caries, they are collectively referred to as mutans 3

streptococci (MS). S. mutans and S. sobrinus are the most common MS found in humans, and 4

are therefore likely to be responsible for most human tooth decay. They typically have been 5

observed in high numbers at the site of incipient caries lesions called white spot lesions. Other 6

bacteria have also been implicated in the etiology of dental caries, including other oral viridans 7

streptococci that colonize the tooth surface such as S. sanguis, S. oralis, and S. mitis. 8

Lactobacilli have also been implicated in the development of caries (Boyar and Bowden, 1985). 9

However, neither the viridans streptococci species mentioned nor lactobacilli were found in 10

significant numbers in white spot lesions (Balakrishnan et al., 2000; van Houte, 1980), 11

suggesting that these species are more likely to be involved in the progression of the disease 12

rather than in its initiation. The involvement of particular bacterial species in caries initiation is 13

referred to as the ‘specific plaque hypothesis’ (Loesche, 1982). A wealth of laboratory, animal 14

model, and epidemiological data strongly indicate that MS are the principal etiological agents of 15

dental caries. 16

MS have several important virulence factors that promote their ability to cause decay 17

(Kuramitsu, 2003). First, MS have surface molecules called adhesins that promote their ability 18

to attach to specific molecules that help to comprise the salivary pellicle covering the surfaces of 19

teeth. The initial, reversible attachment of MS to hard surfaces is highly specific, and provided 20

the model for the original demonstration of specific bacterial attachment (Hillman et al., 1970). 21

Second, once attached, MS will replicate and a produce a dense, extracellular, crosslinked 22

polyglucose holdfast called ‘dextran’ from sucrose (Hamada et al., 1989; Hamada et al., 1984; 23

6

Hosoya et al., 1993). The resulting microcolonies typically constitute approximately 1% of the 1

human dental plaque biofilm. Third, MS are homolactic fermenters when high sugar 2

concentrations are present in the environment. Most sugars commonly found in the human diet, 3

such as sucrose, fructose, glucose and galactose, are readily fermented by these species. The 4

lactic acid is the direct etiologic agent of caries (Hillman, 1978). It reduces the pH in the 5

microenvironment of the tooth and has the ability to chelate calcium, thereby promoting 6

dissolution of the hydroxyapatite that forms the hard, mineral portion of enamel and dentin. 7

Although viridans streptococci in general are homolactic fermenters in the presence of excess 8

sugar, and could therefore be cariogenic, it is believed that the particularly dense dextran 9

characteristic of MS helps to trap the acid against the tooth surface, thereby prolonging its 10

deleterious effect and making these species particularly virulent. This is the reason that the 11

amount of sugar in the diet and the frequency of its ingestion are considered to be extremely 12

important criteria in an individual’s susceptibility to dental caries (Koga et al., 1986; Loesche, 13

1986; Toi et al., 2005). In particular instances, high and frequent sugar consumption was found 14

to result in levels of MS equal to 30% of total cultivatable plaque flora, leading to rapid 15

deterioration of the dentition (Berkowitz, 2003). 16

17 The Role of Bacteriocins in the Natural History of MS Infections 18

19 Kelstrupp and Gibbons (Kelstrupp and Gibbons, 1970) were the first to report bacteriocin 20

production by MS. Since then a number of laboratories (Chikindas et al., 1997; Kelstrup and 21

Gibbons, 1969) have examined this property in several important regards. Bacteriocins produced 22

by MS have largely been referred to as ‘mutacins’ (Hamada and Ooshima, 1975a; Hamada and 23

Ooshima, 1975b), and more recently in certain cases as bacteriocin-like inhibitory substances 24

(BLIS). The literature suggests that 50 to 80% of all MS produced an antimicrobial compound 25

7

(Asensio et al., 1976; Fabio et al., 1987; Hamada and Ooshima, 1975b; Hillman, 1978; Parrot et 1

al., 1990a; Parrot et al., 1990b; Rogers, 1976; van Houte, 1980) that was effective in inhibiting a 2

limited spectrum of target strains, essentially all Gram positive bacteria. Based on the observed 3

differences in spectra of activity and on limited characterization of some of these molecules, it 4

appears that there is a great variety of MS mutacins. 5

The differences in mutacins produced by different strains of MS suggested their potential 6

usefulness as strain markers. Thus, one of the first practical applications of mutacin production 7

by MS was its use in epidemiological studies to determine the mode of transmission of these 8

species to the teeth of young children (Caufield and Walker, 1989). Like other oral viridans 9

streptococci that attach specifically to hard (e.g., tooth) surfaces, persistent colonization by MS 10

almost certainly does not occur in children until after the eruption of teeth (Berkowitz et al., 11

1981; Fujiwara et al., 1991; Mohan et al., 1998). However, some studies have isolated MS in 12

predentate children, although the cross-sectional nature of these studies cannot rule out the 13

possibility that the observed MS isolates are transients (Edwardsson and Mejare, 1978; Tanner et 14

al., 2002; Wan et al., 2003). This is likely, since MS has not been shown to persistently colonize 15

soft (mucosal) oral surfaces. 16

It was long presumed that contaminated saliva served as the vector for transmission, 17

which suggested that the mother (or the principal caretaker) of a child would serve as the likeliest 18

source of the infecting strain. This hypothesis was verified by a number of different laboratories 19

that demonstrated the spectrum of activity of the mutacin from MS strains isolated from mother 20

and child are almost always identical (Berkowitz and Jones, 1985; Emanuelsson and Wang, 21

1998; Emanuelsson et al., 1998; Li and Caufield, 1995). Mothers having a high number of MS 22

in their saliva transmit the infection to their children sooner and more efficiently than mothers 23

8

who have a low salivary concentration of MS (Berkowitz et al., 1981; Brambilla et al., 1998; 1

Kohler and Andreen, 1994). The period in which mothers are most likely to transmit MS to their 2

children, termed the window of infectivity, was established to be between 19 and 31 months of 3

age (Caufield et al., 1993). After this period of high infectivity, individuals are less prone to 4

acquire MS, presumably because other bacterial species at least partially fill their niche and 5

habitat. Evidence is sketchy, but indicates that once colonized by a particular strain of MS, it 6

becomes very difficult for superinfection by another MS strain to occur (Jordan et al., 1972; 7

Krasse et al., 1967; Ruangsri and Orstavik, 1977; Tanzer et al., 1982). Some experiments have 8

suggested that after primary acquisition of MS, the strain remain permanent in the oral flora 9

(Kozai et al., 1991), while other studies suggest that permanency of the oral flora is reached after 10

five years of age (Alaluusua et al., 1994). Genotypes of MS in children ages 3-8 have been 11

studied and show extended stability, indicating prolonged persistence of the MS strains 12

(Alaluusua et al., 1994; Emanuelsson and Thornqvist, 2000; Klein et al., 2004). This was also 13

seen in 3 year-long studies showing the persistence of common strains in mother-child pairs 14

(Caufield and Walker, 1989). Similarly, the persistence of MS has also been reported over a 16-15

year period (Kohler et al., 2003). There are reports of children harboring multiple strains of MS, 16

but it is not clear if these represent a single persistent strain plus transients (Alaluusua et al., 17

1994; Alaluusua et al., 1996; Caufield and Walker, 1989; Emanuelsson et al., 1998; 18

Emanuelsson and Thornqvist, 2000; Gronroos et al., 1998; Kohler et al., 2003; Kozai et al., 19

1999; Kreulen et al., 1997; Kulkarni et al., 1989; Mattos-Graner et al., 2001). In general, like 20

most other members of the indigenous bacterial flora, most MS transmission appears to be 21

vertical rather than horizontal. 22

9

There are several reports indicating that production of mutacin promotes transmission of 1

MS in animal models and from mother to child (Beighton et al., 1982; Vanderhoeven and 2

Rogers, 1979). Presumably, these compounds help MS to colonize and protect its habitat from 3

the same or similar species of bacteria. To fully appreciate this phenomenon, it must be 4

understood that dental plaque is typical of biofilms in general in having true architecture. 5

Contrary to our longstanding belief that the 500-700 different species of oral bacteria are 6

randomly intermixed on the tooth surface, investigators using confocal laser microscopy have 7

clearly demonstrated that dental plaque consists of microcolonies of species, alone or in 8

combination, separated by saliva-filled channels (Marsh 2005). The anatomy of the dental 9

plaque biofilm certainly must affect the ability of one species to positively or negatively interact 10

with another species. The importance of this phenomenon will be made apparent below. 11

12 Replacement Therapy for the Prevention of Dental Caries 13 14

As reviewed in his 1946 publication, Florey (Florey, 1946) pointed out that the use of 15

beneficial bacteria to fight pathogenic bacteria has been attempted by many laboratories for 16

many decades. This idea, traditionally known as replacement therapy but more recently as a type 17

of probiotics, is very appealing from both practical and theoretical standpoints. The 18

development of antibiotics revolutionized the practice of medicine in the second half of the 20th 19

century. Mortality due to infectious diseases decreased markedly during this period (Armstrong 20

et al., 1999). Since 1982, however, deaths stemming from infectious diseases have steadily 21

climbed in parallel with the rise of antibiotic resistant pathogens. A wide variety of medically 22

important bacteria are becoming increasingly resistant to commonly used antibiotics. 23

Vancomycin is considered to be the last line of defense against many serious bacterial infections. 24

The finding of vancomycin resistance strains of pathogenic bacteria is alarming; it portends the 25

10

rise of multidrug resistant pathogens that would be untreatable with currently available drugs. 1

The fear is that we will, in effect, return to the pre-antibiotic era unless new antibiotics or new 2

approaches such as replacement therapy are developed soon. From a theoretical standpoint, our 3

indigenous flora is composed almost exclusively of climax microorganisms. They are, in 4

evolutionary terms, “K strategists” in that they rely on being well adapted to their environment 5

(the human host) rather than on high rates of proliferation (r strategists) in order to assure their 6

survival. The gene-for-gene hypothesis (Person, 1967) states that host-parasite relationships are 7

inherently unstable: when a microorganism becomes a member of our indigenous flora, a 8

coevolutionary process occurs wherein mutations in the microorganism reduce its virulence and 9

mutations in the host increase its resistance. Eventually, host and microbe are brought into a 10

climax state where they coexist without significantly harming each other. It is certainly 11

reasonable that any microorganisms dependent on us for its daily survival would not want to 12

cause us harm. Empirically, most members of the approximately 700 species that comprise 13

human dental plaque are climax organisms. However, there are certain species that have not 14

completed this evolutionary process and retain some pathogenic potential. Also, in other 15

instances such as MS, microorganisms may have been forced out of their climax state by the 16

influences of civilization, as seen when increased reliance on dietary carbohydrates resulted in 17

increased dental caries. 18

In the case of dental caries, we can confidently assume that evolution will continue to act 19

on the principle etiologic agents, MS, to bring them into a new climax state wherein it no longer 20

expresses a pathogenic potential. However, for natural selection to act via spontaneous 21

mutations to eliminate this organisms’ residual virulence may require many thousands of years to 22

complete. The intention of modern day replacement therapy, as it applies to diseases caused by 23

11

indigenous microorganisms, is to greatly speed this natural evolutionary process. By careful 1

study of the pathogen, it may be possible to identify and modify certain of its genes to create, in 2

a relatively short span of time, a so-called effector strain that presages the climax organism that 3

evolution would eventually select. The effector strain is used to intentionally colonize the niche 4

in susceptible host tissues that is normally colonized by the pathogen. By being better adapted, a 5

properly constructed effector strain will prevent colonization of the pathogen by blocking 6

attachment sites, competing for essential nutrients, or other mechanisms. In this fashion, the 7

host is protected for as long as the effector strain persists as a member of the indigenous flora 8

which, ideally, is for the lifetime of the host. 9

Construction of an effector strain for the prevention of dental caries took into account the 10

following logical prerequisites: 11

12 1. It must persistently and preemptively colonize the MS niche, thereby preventing colonization 13

by wild-type (disease-causing) strains whenever the host comes into contact with them. 14

Ideally, it should be able to aggressively displace indigenous strains of MS, thereby allowing 15

even previously infected subjects to be treated with replacement therapy; 16

2. It must have a significantly reduced pathogenic potential; 17

3. It must be generally safe and not predispose the host to other diseases. 18

19 Colonization 20

Numerous studies (Jordan et al., 1972; Krasse et al., 1967; Ruangsri and Orstavik, 1977) 21

have documented the difficulty of persistently introducing laboratory strains of MS into the 22

mouths of humans, particularly if they already harbored an indigenous strain of this organism. 23

From the standpoint of replacement therapy of caries, these results suggest that implantation of 24

12

an effector strain would best be accomplished in children immediately after the onset of tooth 1

eruption and before their acquisition of a wild-type strain. In order to prevent supercolonization 2

by wild-type strains when the host comes in contact with them, an effector strain should have 3

some significant selective advantage to colonization. This would also enable subjects who have 4

already been infected with wild-type MS to be treated by replacement therapy. The ability of an 5

effector strain to preemptively colonize the human oral cavity and aggressively displace 6

indigenous wild-type strains was initially thought to be a complex phenomenon dependent on a 7

large number of phenotypic properties. However, it was discovered that a single phenotypic 8

property could provide the necessary selective advantage. A naturally occurring strain of S. 9

mutans was isolated from a human subject that produces a bacteriocin called mutacin 1140 that 10

is capable of killing virtually all other strains of MS against which it was tested (Hillman et al., 11

1984). Mutants were isolated that produced no detectable mutacin 1140 or that produced 12

approximately three-fold elevated amounts. The mutants were used in a rat model to correlate 13

mutacin production to preemptive colonization and aggressive displacement. While it has long 14

been generally assumed that bacteriocin production must serve a role in colonizing or protecting 15

a microbe’s habitat, to the best of our knowledge this work provided the most definitive evidence 16

to date demonstrating a role for a bacteriocin in the natural history of infection by a 17

microorganism. 18

A correlation was also made between mutacin 1140 production and the ability of S. 19

mutans to persistently colonize the oral cavities of human subjects and aggressively displace 20

indigenous mutans streptococci (Hillman et al., 1987; Hillman et al., 1985). Three years 21

following a single, 3 minute infection regimen involving brushing and flossing of a concentrated 22

cell suspension onto and between the teeth, all of the subjects remained colonized by the mutant 23

13

strain producing 3-fold elevated amounts of mutacin 1140. No other strains of MS were 1

observed in saliva and plaque samples of these colonized volunteers. The same results were 2

found recently, 14 years after colonization, for at least two of three subjects who are still 3

available for testing. These results indicate that this strain of S. mutans succeeded in satisfying 4

the first prerequisite for use as an effector strain in replacement therapy: It persistently and 5

preemptively colonized the S. mutans niche in the human oral cavity and it aggressively 6

displaced indigenous strains of this organism. Consequently, S. mutans strain JH1140, which 7

has a spontaneous mutation resulting in approximately 3-fold elevated production of mutacin 8

1140, served as the starting strain for construction of the effector strain for caries prevention. 9

Mutacin 1140 has been identified as a type A lantibiotic that resembles gallidermin. As a 10

naturally occurring product of S. mutans strain JH1140, it has been deemed a suitable agent for 11

promoting colonization by the effector strain in replacement therapy for prevention of dental 12

caries. Details of the chemistry and nature of mutacin 1140 are presented below. 13

14 Pathogenicity 15

In accord with the acidogenic theory of dental caries, lactic acid production by S. mutans 16

has long been considered to be the main pathogenic mechanism for production of caries lesions. 17

Consequently, lactate dehydrogenase (LDH) deficiency was chosen as the approach for reducing 18

acidogenicity in construction of the effector strain for caries prevention. Earlier work with a 19

closely related S. rattus strain had provided convincing evidence for the effectiveness of this 20

approach. LDH-deficient mutants were virtually non-cariogenic in gnotobiotic rats and did not 21

contribute significantly to the cariogenic potential of the indigenous flora in conventional 22

pathogen-free rats. Attempts to transfer these findings directly to S. mutans proved to be 23

difficult. LDH-deficient mutants of various strains of S. mutans, particularly including JH1140, 24

14

were not found using the same screening methods used to isolate mutants of S. rattus. Cloning 1

the structural gene encoding the S. mutans LDH provided the basis for solving this puzzle. 2

Standard insertional mutagenesis methods failed to yield LDH-deficient clones (Duncan and 3

Hillman, 1991; Hillman et al., 1994; Hillman et al., 1990), suggesting that LDH-deficiency was a 4

lethal mutation in most S. mutans strains. This hypothesis was definitively proven by creation of 5

a temperature sensitive LDH mutant (Chen et al., 1994). This isolate grew well at 30ºC but did 6

not grow at 42ºC under a variety of cultivation conditions. Chemostat studies indicated that 7

some aspect of glucose metabolism was toxic during growth under the non-permissive condition 8

(Hillman et al., 1996). The toxic effect could be overcome by limiting the amount of 9

environmental glucose. This and other data accorded with studies of S. mutans central 10

intermediary metabolism indicating that this organism has enzymatic activities, including 11

pyruvate formate-lyase (Takahashi et al., 1982) and pyruvate dehydrogenase (Carlsson et al., 12

1985; Hillman et al., 1987), for pyruvate dissimilation. However, at high sugar concentrations, 13

the levels of activity of these enzymes are apparently insufficient to compensate for the absence 14

of LDH. It was found (Hillman et al., 1996) that a supplemental alcohol dehydrogenase (ADH) 15

activity, when expressed in the temperature sensitive LDH mutant, could complement LDH 16

deficiency. 17

With this background of information, the effector strain construction started with the ldh 18

gene cloned into an appropriate suicide vector for S. mutans. Essentially the entire gene except 19

for transcription and translation signal sequences was deleted and replaced with the Zymomonas 20

mobilis open reading frame for alcohol dehydrogenase (ADH) II. Transformation of the 21

recombinant molecule into the S. mutans starting strain, JH1140, and allelic exchange resulted in 22

the isogenic mutant, BCS3-L1. This effector strain had no measurable LDH activity and 23

15

approximately 10-fold elevated levels of ADH activity relative to its parent. Fermentation end-1

product analysis showed that BCS3-L1 made no detectable lactic acid. As predicted from earlier 2

work with S. rattus, much of the metabolized carbon was converted to the neutral end products, 3

ethanol and acetoin. Under various cultivation conditions, including growth on a variety of 4

sugars and polyols, such as sucrose, fructose, lactose, mannitol, and sorbitol, BCS3-L1 yielded 5

final pH values that were 0.4 to 1.2 pH units higher than those of its parent, JH1140. The reduced 6

acidogenic potential of BCS3-L1 was reflected in its dramatically decreased cariogenic potential, 7

as shown in several animal models (Hillman et al., 2000). Introduction of the ldh mutant allele 8

into JH1140 had no measurable effects on mutacin 1140 production and other phenotypic 9

properties known to be important in the natural history of infection by S. mutans. This includes 10

intracellular polysaccharide storage, dextran production, specific bacterial attachment and 11

aciduricity. 12

The results of these studies provided strong evidence that BCS3-L1, an LDH-deficient 13

derivative of the mutacin 1140-producing strain, had significantly reduced pathogenic potential, 14

and thus satisfies the second prerequisite for use as an effector strain in replacement therapy for 15

dental caries. 16

17 Safety and stability 18

To serve as an effector strain in replacement therapy of dental caries, BCS3-L1 must be 19

safe in several important regards. First, it must be genetically stable. In the case of BCS3-L1, 20

reacquisition of an acidogenic phenotype by spontaneous reversion is extremely unlikely because 21

construction of BCS3-L1 involved deletion of essentially the entire ldh open reading frame. 22

Although horizontal transmission of an ldh gene is possible, S. mutans is not known to serve as a 23

donor in conjugation. Also, rare bacteriophages that infect MS have failed to demonstrate either 24

16

general or specific transducing ability (A. Delisle, personal communication). In addition, 1

transformability is known to be very strain dependent (Perry and Kuramitsu, 1981), and although 2

it has been shown (Li et al., 2001) that BCS3-L1 has a low transformation frequency due to a 3

natural mutation in its gene for competence stimulating peptide (comC), allelic exchange has 4

been used to delete the comE gene to provide added assurance that exogenous DNA will not 5

transform this strain. Laboratory and animal model testing of the comE derivative demonstrated 6

a transformation frequency < 10-9. The BCS3-L1 derivative defective in comE is, thus, the 7

prototype effector strain for use in replacement therapy for prevention of dental caries. It has 8

been given the trade name ‘SMaRT©’ for Streptococcus mutans associated Replacement 9

Therapy. 10

BCS3-L1 has been extensively tested for safety and efficacy both in the laboratory and in 11

animal models. Both acute and chronic toxicity studies revealed no adverse effects. It was 12

shown (Hillman et al., 2000) that after colonization for 6 months, the mean weights of 13

conventional Sprague-Dawley rats colonized with BCS3-L1 did not differ significantly from 14

animals colonized with JH1140 or S. mutans-free control animals. Histopathological 15

examination revealed no treatment-related lesions in any of the 44 organs and tissues examined. 16

Mutacin production by BCS3-L1 and the change in fermentation products resulting from LDH 17

deficiency could conceivably upset plaque ecology and lead to the bloom of another 18

microorganism with pathogenic potential. As mentioned above, studies (Costerton et al., 1994) 19

have provided an appreciation for the complicated structural architecture of biofilms, including 20

dental plaque. Following specific initial attachment to a surface, the growth of cells leads to the 21

formation of differentiated mushroom- and pillar-like structures consisting of cells embedded in 22

their extracellular polysaccharide matrix. Between these cellular structures are water-filled 23

17

spaces that serve as channels for the introduction of nutrients and the elimination of waste 1

products. In biofilms consisting of two or more species of bacteria, each cellular structure may 2

be pure or mixtures of cells depending on the pressures imposed by positive and negative 3

bacterial interactions. These interactions may also extend over a finite area to affect the general 4

composition of plaque in a particular habitat. Clearly, however, there is a physical limit to this 5

sphere of influence. The mutacin 1140 up-producing strain of S. mutans eliminated mutacin-6

sensitive indigenous strains of MS but had no effect on indigenous S. oralis strains that were 7

equally sensitive to mutacin killing in vitro (Hillman, 1978). These results indicate that S. 8

mutans has a physically distinct habitat that is separated from the S. oralis habitat by a distance 9

sufficient for dilution to reduce the concentration of mutacin below its minimal inhibitory 10

concentration. A similar explanation could account for the failure to observe qualitative or 11

quantitative changes in the plaque of rats following long-term infection with an LDH deficient 12

mutant, even though the mutant’s metabolic end-products are different from those of the wild-13

type strain (Stashenko and Hillman, 1989). 14

A final aspect of replacement therapy safety is the requirement for controlled spread of 15

the effector strain within the population. Mutacin 1140 clearly provides a selective advantage to 16

BCS3-L1 colonization, but the minimal infectious dose has not been determined for this or any 17

other S. mutans strain in humans. As described above, horizontal transmission of natural strains 18

appears to be a rare event, but mutacin 1140 production may promote its occurrence. Wives and 19

children of the two subjects infected with the mutacin up-producing S. mutans strain were not 20

colonized when tested 14 years after the initial infection regimen (J.D. Hillman, unpublished 21

results). Clearly, additional studies with larger populations will have to be performed to properly 22

18

measure the potential for horizontal transmission. It is expected that, like wild-type strains of S. 1

mutans, vertical transmission of BCS3-L1 from mother to child will occur at a high frequency. 2

Previous attempts to implement replacement therapy for various diseases have used 3

naturally occurring effector strains that were fortuitously isolated and found to be significantly 4

less virulent than other biotypes of the same species. No attention was given to the need for 5

genetic stability, with the result that these attempts invariably failed because of reversion or 6

acquisition of virulence factors. In other instances, effector strains did not have the necessary 7

phenotypic advantage to ensure persistent colonization and the means to prevent the colonization 8

or bloom of the pathogen. The results of the work presented here provide a sense of optimism 9

that replacement therapy will be a fruitful approach for the prevention of dental caries in humans. 10

We have taken advantage of the fact that MS are a very well studied group of organisms that 11

have a relatively small array of virulence factors. The major achievements over the 25 years that 12

this work has been ongoing were the isolation of a S. mutans strain which produces a bacteriocin 13

that gives it a significant advantage over other MS strains in colonization and the development 14

and use of recombinant DNA techniques that enabled us to provide the greatest possible 15

assurance of genetic stability. SMaRT is currently undergoing clinical trials to test its safety and 16

effectiveness in the prevention of dental caries. 17

The identification of mutacin 1140 as an essential component in construction of an 18

effector strain for replacement therapy led to an appreciation for this molecule’s chemistry and 19

biology. In particular, its broad spectrum of activity suggested the possibility that the purified 20

molecule, produced in commercially viable amounts, could serve in the treatment of a variety of 21

infectious diseases. The following section describes our work devoted to this area of study. 22

23 Production and Purification of Mutacin 1140 24

19

A genetic approach was used (Gutierrez et al., 1996) to determine that mutacin 1140 1

belongs to the family of antibiotics called lantibiotics. Lantibiotics are ribosomally synthesized 2

antimicrobial peptides produced by several species of Gram-positive bacteria. They are a group 3

of small peptide antibiotics that range from 22 to 34 amino acids, and they have extensive 4

posttranslational modifications. Serine and threonine residues are enzymatically dehydrated to 5

yield dehydroalanine (dha) and dehydrobuterine (dhb), respectively; the sulfhydryl groups of 6

nearby cysteine residues frequently undergo enzymatically catalyzed Michael addition to the 7

vinyl groups of dha to yield an alanine residue linked by a thioether bridge to another alanine 8

residue, so-called lanthionine (lan); similarly, the addition of a cysteine sulfhydryl group to dhb 9

yields α-butyric acid (abu) linked by a thioether to alanine, so-called β-methyllanthionine 10

(MeLan). In addition to the presence of lanthionine rings, other post-translational modifications 11

such as oxidative decarboxylation of the carboxyl terminus, as in the case of mutacin 1140, gives 12

an S-amino vinyl group. The occurrence of the unusual amino acids lanthionine (Lan) and β-13

methyllanthionine (MeLan) define lantibiotics and give them their name (Chatterjee et al., 2005). 14

These molecules have been found generally to have a broad spectrum of activity and they are 15

very heat stable. The lantibiotic nisin has found widespread use as a food preservative. The 16

primary mode of bactericidal activity of type A lantibiotics is believed to be disruption of the 17

cytoplasmic membrane causing the efflux of ions and metabolites and causing desynergization of 18

the target cell (Abee et al., 1995; Bruno et al., 1992; Garcera et al., 1993; Sahl et al., 1987). 19

However more recent evidence suggests that lantibiotics also act by inhibiting cell wall synthesis 20

by binding to and abducting lipid II. Lantibiotics are classified as Type A, Type B, Type IIa and 21

Type IIb depending on the peptide length, structure and heat stability. 22

20

As mentioned, the demand for new antibiotics and antibacterials is on the rise due to the 1

emergence of bacterial strains that are resistant to one or more groups of antibiotics. Lantibiotics 2

as a class are interesting antimicrobial agents that are being developed as potential drug 3

candidates because of their unusual structure and unique mode of action. This section will deal 4

with the various investigations that have been reported towards the successful manufacture of 5

these antimicrobial peptides. 6

One the most important criterion for the successful development of an antibiotic is the 7

ability to produce it in sufficient amounts. Usually several hundreds of grams of antibiotic are 8

required for pre-clinical development. The successful production of many antibiotics has 9

required optimization of fermentation conditions to achieve optimum yields. Nisin A (see Figure 10

1A) is the earliest of the lantibiotics to be discovered and has been used successfully in the food 11

industry as a preservative. Its biosynthetic pathway is well understood and the food groups that 12

use nisin as a preservative include meat and milk products (Gonzalez and Gryczka, 1988; 13

Gonzalez and Gryczka, 1987; Nauth, 2000a; Nauth, 2000b ; Nauth and Brooks, 2000 ; Nauth et 14

al., 2001; Pasch et al., 2004). The nisin used in these products is not purified. Inhibitory 15

concentrations are obtained by adding nisin-containing whey or milk solids derived from dairy 16

cultures inoculated with an appropriate Lactobacillus lactis strain. Thus, to date, a formal scale-17

up method for production of clinical grade nisin has not yet been reported. Recently, an 18

interesting approach to purifying nisin was reported in a patent by Coughlin et al. (Coughlin and 19

Crabb, 2004), who optimized conditions for proteolytic digestion to selectively degrade 20

approximately 40 contaminating peptides without degrading nisin. The authors concluded that 21

they achieved approximately 90% purity with 87.5% yield. Other approaches have yielded small 22

quantities of other lantibiotics in relatively pure form as well (Chatterjee et al., 2005). With the 23

21

exceptions noted below, large scale production and purification to clinical grade material has not 1

been achieved for the vast majority of the 40 plus lantibiotics currently known. This provides an 2

important challenge in light of the need for new antibiotics to replace ones that are losing their 3

effectiveness. 4

Gallidermin is a lantibiotic that is produced by Staphylococcus gallinarum and is a 5

structural analog of epidermin. It is active against a broad spectrum of Gram positive bacteria. 6

It has antibacterial activity similar to that of erythromycin and fusidin and is considered to have a 7

potential in the treatment of endocarditis and abscesses. Gallidermin was also found to be highly 8

active against Propionibacterium acnes that is known to cause skin infections. A natural 9

gallidermin variant, called staphylococcin T, was also identified and isolated from a strain of 10

Staphylococcus cohnii by (Furmanek et al., 1999). Gallidermin production through fermentation 11

has been extensively studied in order to achieve commercially viable yields. The amounts of this 12

antibiotic that are produced in optimized fermentations are 400-500 mg per liter, and methods 13

have been developed to enable accumulation of sufficient amounts of purified product for pre-14

clinical and clinical studies. 15

In order to maximize the production of gallidermin by fermentation, Kempf (Kempf et 16

al., 1997; Kempf et al., 1999a; Kempf et al., 1999b; Kempf et al., 2000) studied a variety of 17

parameters. The strain used for these studies was Staphylococcus gallinarum Tü 3928. The 18

concentration of dissolved oxygen was varied in the fermentation from 4 to 30% of saturation. It 19

was found that at high oxygen concentrations the biomass of the bacteria increased, but dropping 20

the concentration of oxygen increased the production of the antibiotic relative to biomass. 21

Therefore, to maximize the yield of the antibiotic with regards to oxygen concentrations it was 22

found that biphasic growth was best wherein a fully aerobic growth phase was followed by a 23

22

reduced oxygen growth phase. This led to an increase in the biomass during the aerobic phase 1

and a substantial increase in the antibiotic production in the oxygen reduced phase. Initially, the 2

inoculation and fermentation media consisted of meat extract, malt extract and calcium chloride. 3

The fermentations were carried out at 37°C and pH 6.5. The use of meat extract proved to be 4

undesirable due to both its high cost and the fact that it served as a possible source of prions. It 5

was found that yeast extract substituted very well for the meat extract. The optimized medium 6

afforded about 300 mg of gallidermin per liter. This process was scaled up successfully to 200 7

liters and the cost of the medium was reduced by 93%. 8

Yeast extract consists of several amino acids and peptides that are metabolized during the 9

course of fermentation. It would be logical to assume that at lease certain amino acids, if not all, 10

would have an effect on the production of gallidermin. Thus, the next set of experiments 11

designed and executed by Kempf (Kempf et al., 1999a) was to study the effect of adding 12

additional aliquots of amino acids such as glutamate, aspartate, asparagine, glycine, serine, 13

threonine, alanine and arginine into fermentation cultures containing yeast extract and maltose. 14

All of these amino acids were found to increase the product concentration to about 400 mg per 15

liter in a batch culture, the best amino acids being glutamate, glycine, serine, and threonine. 16

Glutamate was chosen to perform the optimizations on larger scales due to its lower cost 17

compared to other amino acids. It was reported that the amount of the antibiotic produced 18

increased to about 460 mg in a fed batch culture. Addition of simple ammonium salts did not 19

increase the amounts of antibiotic being produced, suggesting that the amino acids function as a 20

source of carbon rather than a source of nitrogen. 21

Large scale fermentation of the lantibiotic mutacin I/III, produced by S. mutans strain 22

CH43 and UA787, has been achieved in liquid medium as recently reported in a patent by 23

23

Caufield et al. (Caufield and He, 2004). The fermentation was carried out using a medium 1

comprised of yeast extract, peptone, and sucrose, which were inoculated with cells grown in 2

Todd-Hewitt broth. The fermentation was carried out at 37ºC and the pH of the broth was 3

maintained at 5.6 and stirred at 50-250 rpm. The fermentation was sampled every 8 hours to 4

monitor the production of antibiotic. The results seemed to suggest that peak production was 5

achieved at the end of the exponential growth phase of the antibiotics. It was also noted that the 6

production of the antibiotic required the formation of bio-film by the organism. The sucrose 7

levels were monitored periodically and after 24 hours the fermentation was halted. The cell free 8

broth was extracted with Amberlite-XAD resin to recover the antibiotic. The antibiotic was 9

recovered from the resin using a stepwise elution with 40%, 50%, 60%, 70%, 80% and 90% 10

ethanol. The bioactivity of various fractions was checked by a plate assay using S. sanguis 11

NY101 as an indicator strain. The bioactive fractions were lyophilized to dryness. The dried 12

powder was dissolved in 6M urea and the insoluble residue was removed by filtration. The 13

filtrate was first applied to a reverse-phase 30 SOURCE 15RPC custom column. Elution was 14

carried out with a fragmented gradient of water and acetonitrile containing 0.1% trifluoroacetic 15

acid using an AKTA Purifier (Amersham Pharmacia Biotech, Piscataway, N.J.). Active 16

fractions were collected and lyophilized to produce pure mutacin III powder. Then it was 17

purified again by performing another HPLC with a 15 cm 5RPC using similar procedures as 18

described above. Mutacin I/III lacks the robustness of many other lantibiotic antibiotics, but may 19

find practical application if its development of its production and purification proceed apace and 20

if its pre-clinical data indicate safety and efficacy in the treatment of select infectious diseases. 21

Mutacin B-Ny266 is a bacteriocin isolated from S. mutans B-Ny266 that is very similar 22

to mutacin 1140. Several sugars such as sucrose, fructose, galactose and cheese whey permeate 23

24

were used in fermentation studies to identify variables that affect lantibiotic production 1

(Guillaume et al., 2004). It was found that galactose was the best carbon source with regards to 2

production of the mutacin B-Ny266. Several nitrogen sources such as cheese whey permeate, 3

yeast extract, cottonseed hydrolysate, peptone and neopeptone were also investigated and yeast 4

extract was found to be optimal for the production of mutacin B-Ny266. The pH of the 5

fermentation medium was held at pH 6.0 by using calcium carbonate. It was found that calcium 6

is an important cation for the production of B-Ny266. The incorporation of other cations, such 7

as potassium and sodium, into the medium in place of calcium did not lead to the production of 8

significant levels of the mutacin. The exact function of calcium is not know but the authors 9

speculated that the divalent calcium cations bind to negatively charged molecules on the 10

membrane surface of the bacteria and give immunity to the lantibiotic producing bacteria by 11

preventing the positively charged mutacin molecules from binding. This effect may play a 12

protective role for the integrity of the lipid membrane of the producing microorganism. 13

Alternatively, calcium may stimulate the synthesis of the prepeptide or the activation of the 14

prepeptide maturation enzymes and the transport out of the cell (Matsusaki et al., 1996; Vuyst 15

and Vandamme, 1993). The inoculum for the fermentation was grown overnight in trypticase 16

soy broth yeast extract at 37ºC. The fermentation was carried out at 37ºC and the pH was 17

maintained at 7.0. The levels of mutacin B-Ny-266 produced were studied using a deferred 18

antagonism assay against Micrococcus luteus. Using these methods, mutacin concentrations up 19

to 500 mg/liter were achieved (Marc C. Lavoie, Personal Communication). Methods for 20

purification of the active molecule have not been reported. 21

Hillman’s group reported the initial isolation of a S. mutans strain producing mutacin 22

1140 (Hillman et al., 1984) and, subsequently, its genetic and biochemical analysis (Hillman et 23

25

al., 1998). Subsequently, mutacin 1140-producing MS have been independently isolated, and in 1

one study appeared to constitute approximately 1-2% of the total MS strains recovered from 2

human teeth (Hillman, unpublished), suggesting that there may be an evolutionary selection for 3

strains producing this lantibiotic. Mutacin 1140 has some very desirable properties from the 4

standpoint of its use as a therapeutic agent in the treatment of infectious diseases. It is a small, 5

stable molecule with a very broad spectrum of activity that includes most Gram-positive bacteria. 6

Mutacin 1140 is bacteriocidal and has rapid kill kinetics. It is very poorly immunogenic judging 7

from the inability to raise antibodies against it in rabbits and mice, with and without conjugation 8

to carrier molecules or treatment with gluteraldehyde. It has been suggested that the 3 9

dimensional structure of mutacin 1140 (see below), which has lanthionines covering the surfaces 10

of the molecule, is responsible for its poor immunogenicity: lantibiotics are relatively rare in 11

nature, and since they are the only known molecules to use lanthionine residues, it may be that 12

the immune repertoires of mammals do not include reactive lymphocytes. Certainly of particular 13

importance is the fact that mutacin 1140 has a novel mechanism of action in its ability to bind to 14

and abduct lipid II from the primary sites of active peptidoglycan biosynthesis. The binding of 15

mutacin 1140 to lipid II occurs at a site that is distinct from the binding site for vancomycin. 16

Spontaneous or chemically induced resistance to mutacin 1140 has not been found to date in 17

multidrug resistant clinical strains of Staphylococcus aureus or Enterococcus faecalis or in MS 18

(Hillman, unpublished). 19

The physical nature of mutacin 1140 has been extensively studied. The covalent structure 20

of mutacin 1140 was established by amino acid sequencing, mass spectrometry, and nuclear 21

magnetic resonance (see Figure 1B). The position of the various thioether bridges and 22

dehydrated amino acids were determined (Smith et al., 2000) by using sodium borohydride and 23

26

ethanethiol derivatization, an innovative double labeling method, followed by Edman 1

sequencing. In these studies the unsaturated amino acid residues in the molecule (dha-5 and dhb-2

14) were reduced using sodium borohydride and the reduced molecule was treated with 3

ethanethiol to open the thioether rings, which enabled the positions of dehydrated residues to be 4

identified during Edman sequencing. The relative positions of the various residues that 5

contained the lanthionine rings and dehydrated residues were easily distinguished following this 6

simple labeling procedure. The absolute positions of the thioether linkages for mutacin 1140 7

were obtained by Nuclear Magnetic Resonance. Purified mutacin 1140 was analyzed by means 8

of ESI-MS and MS/MS. Mutacin 1140 has a calculated mass of 2266 Da indicating the 9

occurrence of complete post-translational modifications. MS/MS of the doubly charged 10

molecular ion yielded a complex spectrum of daughter ions. The major ion was the doubly 11

charged molecular ion, indicating a noteworthy stability of the peptide. The MS/MS spectra 12

showed ions with the loss or addition of 32 or 33 mass units, suggesting the loss or addition of 13

sulfur or sulfhydryl groups. The interpretation of the b-ions was supported by an additional 14

experiment in which the N-terminal portion was labeled with 2-iminothiolane, and the doubly 15

charged molecular ion with two additions (to the first two N-terminal amino-acid residues, Phe 16

and Lys) was subjected again to collision-induced dissociation. The data obtained from these 17

experiments was consistent with two dehydrated amino acid residues and an S- [aminovinyl]-18

cysteine. Mutacin 1140 with one and two additions was also subjected to collision-induced 19

dissociation with good results. Daughter ions of the doubly charged molecule of mutacin 1140 20

with two additions of 2-mercaptoethanol were also analyzed by MS/MS. The 2-mercaptoethanol 21

added 78 and 156 mass units to all b-ions following positions 5 and 14, respectively, 22

demonstrating that these residues are Dha-5 and Dhb-14. The y-ion series gave similar results. 23

27

The absence of secondary daughter ions derived from b-ions or y-ions by loss or addition of S or 1

SH atoms in the Ala-12 to the Gly-15 region suggested that there are no thioether bridges over 2

this region. Therefore, mutacin 1140 could be cleaved by trypsin after the Arg-13 residue to 3

yield two fragments. Mutacin 1140 was subjected to tryptic digestion using both soluble and 4

immobilized sequencing-grade trypsin with identical results. The N-terminal fragment Phe-1 to 5

Arg-13 was observed as a 1401 Da peptide that was identified by tandem mass spectrometry. 6

The C-terminal fragment was not detected, probably due to the deamination of Dhb-14 after the 7

cleavage (similar observations pertaining to epidermin and gallidermin have been made). With 8

the available sequencing and Mass Spectrometry data, NMR techniques were used to assign the 9

positions of the lanthionine rings in mutacin 1140 (Smith et al., 2000), and NMR was also used 10

to definitively establish the correct thioether pairings. TOCSY, NOESY, HMQC, and HMBC 11

NMR data on mutacin 1140 were collected. A sequential Hαi to HNi+1 walk could be performed 12

for most of the amino acids for assignments, and long range NOEs were used to identify the 13

correct thioether bridging. Thus, using the NMR data, the complete 3 dimensional structure of 14

mutacin 1140 was deduced and as shown by Smith and coworkers (Smith et al., 2003). 15

A reporter gene construct in which the gene for ß-galactosidase is fused to the structural 16

gene for mutacin 1140 indicated that expression of the structural gene occurs throughout the 17

entire growth cycle in a very tightly regulated fashion (JD Hillman, unpublished). Although the 18

producer strain, JH1140, has a gene that resembles immunity genes in other lantibiotic gene 19

clusters, the minimal inhibitory concentration is very low (ca. 1 µg/ml). JH1140 prevents 20

autoinhibition of its growth by turning expression of the structural gene on and off in a fairly 21

regular, sawtoothed fashion, possibly in direct response to environmental mutacin 1140 22

concentrations. Maximum rates of production for mutacin 1140 occurred at mid- to late 23

28

logarithmic growth phase and for a short period at the beginning of the stationary phase. To 1

overcome the production limitation caused by regulation of expression, a ‘solid phase 2

fermentation’ method was developed (Hillman et al., 1998). In this method, medium was made 3

semi-solid by the addition of 0.5% agarose, which was then stab inoculated with JH1140. This 4

approach allowed mutacin to diffuse away from the producer cells for an extended period, 5

prolonging the period of active production. The typical yields for this protocol following 6

purification were only about 100-200 μg per liter, but this represented a substantial increase over 7

broth fermentation methods, and allowed us to perform the structural analyses described above. 8

Recently we have developed conditions for producing mutacin 1140 by growing JH1140 9

in liquid medium. A number of parameters were studied and optimized using a Sixfors™ 10

Laboratory Multifermentor (ATR Biotech, Laurel, MD). The Sixfors essentially consists of six 11

reactions vessels that can be individually controlled for temperature, oxygen and pH using PID 12

controllers. The data for all the parameters are recorded in real time throughout the course of the 13

fermentation. Following the lead of Kempf and coworkers (Kempf et al., 1997; Kempf et al., 14

1999a; Kempf et al., 1999b; Kempf et al., 2000) in their reported optimizations of gallidermin 15

production in liquid broth, we initially tested their optimized conditions in the production of 16

mutacin 1140 and noted marked improvement over yields obtained from standard broth cultures. 17

From that starting point we optimized the fermentation conditions by altering individual 18

variables such as pH, carbon source, nitrogen source, oxygen tension, divalent cations, 19

temperature, length of incubation, size of the inocula and agitation. The inocula used in these 20

experiments were grown following a standard operating procedure from a seed bank of JH 1140 21

in order to help minimize variability in this facet of the studies. The amount of mutacin 22

produced was measured using a deferred antagonism assay using an M. luteus target strain. The 23

29

optimized conditions resulted in the reproducible production of about 500-1000 mg per liter of 1

mutacin 1140. The method has also been scaled up from 250 ml as performed in the Sixfors to 2

30 liters without any measurable change in the level of production. 3

Purification of mutacin 1140 from the fermentation medium has also been extensively 4

investigated. Our current method, which achieves >90% purity and 60% yields, involves 5

centrifugation to remove the bacterial cells, adjustment of the culture liquor to 4.0 with 6

concentrated hydrochloric acid, and addition of HP-20 resin (Supelco) to extract the antibiotic. 7

The resin is then treated sequentially with 20% and 40% acetonitrile to remove bound impurities 8

and then with 80% aqueous acetonitrile to quantitatively elute the mutacin. The 80% acetonitrile 9

fraction is concentrated to dryness and subjected to reverse phase HPLC using a C18 column and 10

gradient elution with acetonitrile. The 50-60% acetonitrile fractions contain the bioactivity and 11

the presence of mutacin 1140 has been confirmed by MALDI-TOF, amino acid analysis, and 12

LC-MS. The bioactive fraction is subjected to a second polishing column to give the final, 13

purified mutacin 1140. Gram quantities of mutacin 1140 have been manufactured using these 14

methods and preclinical trials initiated. 15

16 Conclusions 17

The novel S. mutans bacteriocin, mutacin 1140, has proven itself to be of considerable 18

value from a number of different standpoints. Mutant analysis provided definitive evidence for 19

the role of this molecule in promoting the ability of the producer strain to colonize its natural 20

habitat, the enamel surface. This result led to its use as an integral factor in the development of 21

an effector strain for the replacement therapy of dental caries. Clinical trials have shown that a 22

single application of a prototype effector strain leads to lifelong colonization and displacement of 23

indigenous strains of MS. If ultimately successful in clinical trials, replacement therapy for the 24

30

prevention of dental caries should serve as a model for prevention of a variety of human and 1

non-human infectious diseases. 2

Although the class of antibiotics known as lantibiotics has been recognized for over half a 3

century, their usefulness has been limited to food preservation because of the difficulty in 4

obtaining their large scale production and purification. Multidrug resistant bacteria have taken 5

up residence in hospitals worldwide. According to the Center for Disease Control (CDC), nearly 6

two million patients in the United States currently acquire an infection in the hospital each year. 7

Of those patients, about 90,000 died in 2003 as a result of their infection, up from 13,300 patient 8

deaths in 1992. More than 70 percent of the bacteria that cause hospital-acquired infections are 9

resistant to at least one of the drugs most commonly used to treat them. Persons infected with 10

drug-resistant organisms are more likely to have longer hospital stays and require treatment with 11

second or third choice drugs that may be less effective, more toxic, and more expensive. There 12

is a need to teach more appropriate use of antibiotics, but this is not going to solve the current 13

dilemma. More importantly, we need new antibiotics. Clinical application of lantibiotics is a 14

viable prospect: mutacin 1140 is a small, stable molecule that has a broad spectrum of activity 15

and very low immunogenicity. Preliminary studies indicate that it is not cytotoxic in therapeutic 16

doses. Of the handful of new antibiotics developed during the past decade, very few have novel 17

mechanisms of action. This is an important criterion since it portends the ease with which 18

pathogens will be able to develop resistance to a new antibiotic. Mutacin 1140 was one of 19

several molecules that were used to demonstrate a bactericidal mechanism of action that involves 20

binding to and abducting lipid II, a membrane localized enzyme essential for peptidoglycan 21

synthesis. This is clearly a novel mechanism, and is reflected in the finding that isolates of S. 22

aureus and E. faecalis resistant to vancomycin and various other antibiotics have so far shown no 23

31

resistance or propensity to develop resistance to mutacin 1140. Preclinical and clinical trials will 1

ultimately determine the value of mutacin 1140 for clinical application in the treatment of 2

various infectious diseases. Since lantibiotics are ribosomally synthesized molecules, site 3

directed mutagenesis will enable tweaking of the existing structure to improve safety or efficacy, 4

if necessary. 5

6 7 8 Acknowledgements 9

The authors wish to acknowledge the grant support of the National Institutes of Health. 10

We also appreciate the help of Dr. Eric Chojnicki, Jacob Pollock and other Oragenics staff for 11

their contributions in the development of replacement therapy and mutacin 1140. Our thanks to 12

Stephanie C. Haas for her help in preparing this manuscript. 13

14

15 Selected Reading 16 17 Abee, T., Krockel, L., and Hill, C. (1995). Bacteriocins: Modes of action and potentials in food 18 preservation and control of food poisoning. International Journal of Food Microbiology 28, 169-19 185. 20 Alaluusua, S., Alaluusua, S.J., Karjalainen, J., Saarela, M., Holttinen, T., Kallio, M., Holtta, P., 21 Torkko, H., Relander, P., and Asikainen, S. (1994). The demonstration by ribotyping of the 22 stability of oral Streptococcus Mutans infection over 5 to 7 years in children. Archives of Oral 23 Biology 39, 467-471. 24 Alaluusua, S., Matto, J., Gronroos, L., Innila, S., Torkko, H., Asikainen, S., JousimiesSomer, H., 25 and Saarela, M. (1996). Oral colonization by more than one clonal type of mutans streptococcus 26 in children with nursing-bottle dental caries. Archives of Oral Biology 41, 167-173. 27 Armstrong, G.L., Conn, L.A., and Pinner, R.W. (1999). Trends in infectious disease mortality in 28 the United States during the 20th century. PAMA 281, 61-66. 29 Asensio, C., Perezdiaz, J.C., Martinez, M.C., and Baquero, F. (1976). New family of low-30 molecular weight antibiotics from enterobacteria. Biochemical and Biophysical Research 31 Communications 69, 7-14. 32 Balakrishnan, M., Simmonds, R.S., and Tagg, J.R. (2000). Dental caries is a preventable 33 infectious disease. Australian Dental Journal 45, 235-245. 34

32

Beighton, D., Hayday, H., and Walker, J. (1982). The acquisition of Streptococcus mutans by 1 infant monkeys (Macaca fascicularis) and its relationship to the initiation of dental caries. 2 Journal of General Microbiology 128, 1881-1892. 3 Beltran-Aguilar, E.D., Barker, L.K., Canto, M.T., Dye, B.A., Gooch, B.F., Griffin, S.O., Hyman, 4 J., Jaramillo, F., Kingman, A., Nowjack-Raymer, R., et al. (2005). Surveillance for dental caries, 5 dental sealants, tooth retention, edentulism, and enamel fluorosis-United States, 1988-1994 and 6 1999-2002. MMWR Surveill Summ 54, 1-43. 7 Berkowitz, R.J. (2003). Cause, treatment and prevention of early childhood caries: a 8 microbiological perspective. Journal of Canadian Dental Association 69, 304-307. 9 Berkowitz, R.J., and Jones, P. (1985). Mouth-to-mouth transmission of the bacterium 10 Streptococcus mutans between mother and child. Archives of Oral Biology 30, 377-379. 11 Berkowitz, R.J., Turner, J., and Green, P. (1981). Maternal salivary levels of Streptococcus 12 mutans and primary oral infection of infants. Archives of Oral Biology 26, 147-149. 13 Boyar, R.M., and Bowden, G.H. (1985). The Microflora associated with the progression of 14 incipient carious lesions in teeth of children living in a water-fluoridated Area. Caries Research 15 19, 298-306. 16 Brambilla, E., Felloni, A., Gagliani, M., Malerba, A., Garcia-Godoy, F., and Strohmenger, L. 17 (1998). Caries prevention during pregnancy: Results of a 30-month study. Journal of the 18 American Dental Association 129, 871-877. 19 Bratthall, D. (1970). Demonstration of five serological groups of streptococcal strains resembling 20 Streptococcus mutans. Odontol Revy 21, 143-152. 21 Bruno, M.E.C., Kaiser, A., and Montville, T.J. (1992). Depletion of proton motive force by nisin 22 in Listeria monocytogenes cells. Applied and Environmental Microbiology 58, 2255-2259. 23 Carlsson, J., Kujala, U., and Edlund, M.B.K. (1985). Pyruvate dehydrogenase activity in 24 Streptococcus mutans. Infection and Immunity 49, 674-678. 25 Caufield, P.W., Cutter, G.R., and Dasanayake, A.P. (1993). Initial acquisition of mutans 26 streptococci by infants - evidence for a discrete window of infectivity. Journal of Dental 27 Research 72, 37-45. 28 Caufield, P.W., and He, Z. (2004). Enhanced production of Streptococcus mutans mutacin I and 29 III. United States Patent Application 20,040,146,983 (July 29, 2004). 30 Caufield, P.W., and Walker, T.M. (1989). Genetic diversity within Streptococcus mutans evident 31 from chromosomal DNA restriction fragment polymorphisms. Journal of Clinical Microbiology 32 27, 274-278. 33 Chatterjee, C., Paul, M., Xie, L.L., and van der Donk, W.A. (2005). Biosynthesis and mode of 34 action of lantibiotics. Chemical Reviews 105, 633-683. 35 Chen, A.P., Hillman, J.D., and Duncan, M. (1994). L-(+)-Lactate dehydrogenase-deficiency is 36 lethal in Streptococcus mutans. Journal of Bacteriology 176, 1542-1545. 37 Chikindas, M.L., Novak, J., Caufield, P.W., Schilling, K., and Tagg, J.R. (1997). Microbially-38 produced peptides having potential application to the prevention of dental caries. International 39 Journal of Antimicrobial Agents 9, 95-105. 40 Clarke, J.K. (1924). On the bacterial factor in the aetiology of dental caries. British Journal of 41 Experimental Pathology 5, 141-147. 42 Costerton, J.W., Lewandowski, Z., Debeer, D., Caldwell, D., Korber, D., and James, G. (1994). 43 Biofilms, the Customized Microniche. Journal of Bacteriology 176, 2137-2142. 44 Coughlin, R.T., and Crabb, J.H. (2004). Method of purifying lantibiotics. United States Patent 45 Application 20,040,072,333 (April 15, 2004). 46

33

Coykendall, A.L. (1989). Classification and identification of the viridans streptococci. Clinical 1 Microbiology Reviews 2, 315-328. 2 Duncan, M.J., and Hillman, J.D. (1991). DNA-Sequence and In vitro Mutagenesis of the Gene 3 Encoding the Fructose-1,6-Diphosphate-Dependent L-(+)-Lactate Dehydrogenase of 4 Streptococcus Mutans. Infection and Immunity 59, 3930-3934. 5 Edwardsson, S., and Mejare, B. (1978). Streptococcus-Milleri (Guthof) and Streptococcus-6 mutans in Mouths of Infants before and after Tooth Eruption. Archives of Oral Biology 23, 811-7 814. 8 Emanuelsson, I.M.R., and Wang, X.M. (1998). Demonstration of identical strains of mutans 9 streptococci within Chinese families by genotyping. European Journal of Oral Sciences 106, 10 788-794. 11 Emanuelsson, I.R., Li, Y., and Bratthall, D. (1998). Genotyping shows different strains of 12 mutans streptococci between father and child and within parental pairs in Swedish families. Oral 13 Microbiology and Immunology 13, 271-277. 14 Emanuelsson, I.R., and Thornqvist, E. (2000). Genotypes of mutans streptococci tend to persist 15 in their host for several years. Caries Research 34, 133-139. 16 Fabio, U., Bondi, M., Manicardi, G., Messi, P., and Neglia, R. (1987). Production of Bacteriocin-17 Like Substances by Human Oral Streptococci. Microbiologica 10, 363-370. 18 Fitzgerald, R.J. (1968). Dental caries research in gnotobiotic animals. Caries Research 2, 139-19 146. 20 Fitzgerald, R.J., and Keyes, P.H. (1960). Demonstration of the etiologic role of streptococci in 21 experimental caries in the hamster. Journal of American Dental Association 61, 9-19. 22 Florey, H.W. (1946). The Use of Micro-organisms for Therapeutic Purposes. Yale Journal of 23 Biology and Medicine 19, 101-117. 24 Fujiwara, T., Sasada, E., Mima, N., and Ooshima, T. (1991). Caries Prevalence and Salivary 25 Mutans Streptococci in 0-2-Year-Old Children of Japan. Community Dentistry and Oral 26 Epidemiology 19, 151-154. 27 Furmanek, B., Kaczorowski, T., Bugalski, R., Bielawski, K., Bogdanowicz, J., and Podhajska, A. 28 J. (1999). Identification, characterization and purification of the lantibiotic staphylococcin T, a 29 natural gallidermin variant. Journal of Applied Microbiology 87, 856-866. 30 Garcera, M.J.G., Elferink, M.G.L., Driessen, A.J.M., and Konings, W.N. (1993). Invitro Pore-31 Forming Activity of the Lantibiotic Nisin - Role of Protonmotive Force and Lipid-Composition. 32 European Journal of Biochemistry 212, 417-422. 33 Gonzalez, C.F., and Gryczka, A.J. (1988). Derived nisin producing microorganisms, method of 34 production and use and products obtained thereby. United States Patent 4,740,593 (April 26, 35 1988). 36 Gonzalez, C.F., and Gryczka, A.J. (1987). Derived nisin producing microorganisms, method of 37 production and use and products obtained thereby. United States Patent 4,716,115 (December 38 29, 1987). 39 Gronroos, L., Saarela, M., Matto, J., Tanner-Salo, U., Vuorela, A., and Alaluusua, S. (1998). 40 Mutacin production by Streptococcus mutans may promote transmission of bacteria from mother 41 to child. Infection and Immunity 66, 2595-2600. 42 Guillaume, N., Augera, I., Beaudoina, M., Morencya, H., LaPointe, G., and Lavoiea, M.C. 43 (2004). Improved methods for mutacin detection and production. Journal of Microbiological 44 Methods 59, 351-361. 45

34

Gutierrez, J.A., Crowley, P.J., Brown, D.P., Hillman, J.D., Youngman, P., and Bleiweis, A.S. 1 (1996). Insertional mutagenesis and recovery of interrupted genes of Streptococcus mutans by 2 using transposon Tn917: Preliminary characterization of mutants displaying acid sensitivity and 3 nutritional requirements. Journal of Bacteriology 178, 4166-4175. 4 Hamada, S., Horikoshi, T., Minami, T., Okahashi, N., and Koga, T. (1989). Purification and 5 Characterization of Cell-Associated Glucosyltransferase Synthesizing Water-Insoluble Glucan 6 from Serotype-C Streptococcus-Mutans. Journal of General Microbiology 135, 335-344. 7 Hamada, S., Koga, T., and Ooshima, T. (1984). Virulence Factors of Streptococcus-Mutans and 8 Dental-Caries Prevention. Journal of Dental Research 63, 407-411. 9 Hamada, S., and Ooshima, T. (1975a). Inhibitory Spectrum of a Bacteriocin-Like Substance 10 (Mutacin) Produced by Some Strains of Streptococcus-Mutans. Journal of Dental Research 54, 11 140-145. 12 Hamada, S., and Ooshima, T. (1975b). Production and Properties of Bacteriocins (Mutacins) 13 from Streptococcus-Mutans. Archives of Oral Biology 20, 641 14 Hillman, J.D. (1978). Lactate-Dehydrogenase Mutants of Streptococcus-Mutans - Isolation and 15 Preliminary Characterization. Infection and Immunity 21, 206-212. 16 Hillman, J.D., Brooks, T.A., Michalek, S.M., Harmon, C.C., Snoep, J.L., and Van Der Weijden, 17 C.C. (2000). Construction and characterization of an effector strain of Streptococcus mutans for 18 replacement therapy of dental caries. Infection and Immunity 68, 543-549. 19 Hillman, J.D., Chen, A.P., Duncan, M., and Lee, S.W. (1994). Evidence That L-(+)-Lactate 20 Dehydrogenase-Deficiency Is Lethal in Streptococcus-Mutans. Infection and Immunity 62, 60-21 64. 22 Hillman, J.D., Chen, A.P., and Snoep, J.L. (1996). Genetic and physiological analysis of the 23 lethal effect of L-(+)-lactate dehydrogenase deficiency in Streptococcus mutans: 24 Complementation by alcohol dehydrogenase from Zymomonas mobilis. Infection and Immunity 25 64, 4319-4323. 26 Hillman, J.D., Duncan, M.J., and Stashenko, K.P. (1990). Cloning and expression of the gene 27 encoding the fructose-1,6-diphosphate-dependent L-(+)-lactate dehydrogenase of Streptococcus 28 mutans. Infection and Immunity 58, 1290-1295. 29 Hillman, J.D., Dzuback, A.L., and Andrews, S.W. (1987). Colonization of the human oral cavity 30 by a Streptococcus mutans mutant producing increased bacteriocin. Journal of Dental Research 31 66, 1092-1094. 32 Hillman, J.D., Johnson, K.P., and Yaphe, B.I. (1984). Isolation of a Streptococcus mutans strain 33 producing a novel bacteriocin. Infection and Immunity 44, 141-144. 34 Hillman, J.D., Novak, J., Sagura, E., Gutierrez, J.A., Brooks, T.A., Crowley, P.J., Hess, M., 35 Azizi, A., Leung, K.P., Cvitkovitch, D., and Bleiweis, A.S. (1998). Genetic and biochemical 36 analysis of mutacin 1140, a lantibiotic from Streptococcus mutans. Infection and Immunity 66, 37 2743-2749. 38 Hillman, J.D., Vanhoute, J., and Gibbons, R.J. (1970). Sorption of bacteria to human enamel 39 powder. Archives of Oral Biology 15, 899. 40 Hillman, J.D., Yaphe, B.I., and Johnson, K.P. (1985). Colonization of the human oral cavity by a 41 strain of Streptococcus mutans. Journal of Dental Research 64, 1272-1274. 42 Hosoya, K.I., Kubo, H., Natsume, H., Sugibayashi, K., Morimoto, Y., and Yamashita, S. (1993). 43 The structural barrier of absorptive mucosae - site difference of the permeability of fluorescein 44 isothiocyanate-labeled dextran in rabbits. Biopharmaceutics & Drug Disposition 14, 685-695. 45

35

Jordan, H.V., Englander, H.R., Engler, W.O., and Kulczyk, S. (1972). Observations on the 1 implantation and transmission of Streptococcus mutans in humans. Journal of Dental Research 2 51, 515-518. 3 Kelstrup, J., and Gibbons, R.J. (1969). Bacteriocins from human and rodent streptococci. 4 Archives of Oral Biology 14, 251. 5 Kelstrupp, J., and Gibbons, R.J. (1970). Fingerprinting human oral streptococci by bacteriocin 6 production and sensitivity. Arch Oral Biol 15, 1109-1116. 7 Kempf, M., Theobald, U., and Fiedler, H.P. (1997). Influence of dissolved O-2 on the 8 fermentative production of gallidermin by Staphylococcus gallinarum. Biotechnology Letters 19, 9 1063-1065. 10 Kempf, M., Theobald, U., and Fiedler, H.P. (1999a). Correlation between the consumption of 11 amino acids and the production of the antibiotic gallidermin by Staphylococcus gallinarum. 12 Biotechnology Letters 21, 959-963. 13 Kempf, M., Theobald, U., and Fiedler, H.P. (1999b). Economic improvement of the fermentative 14 production of gallidermin by Staphylococcus gallinarum. Biotechnology Letters 21, 663-667. 15 Kempf, M., Theobald, U., and Fiedler, H.P. (2000). Production of the antibiotic gallidermin by 16 Staphylococcus gallinarum - development of a scale-up procedure. Biotechnology Letters 22, 17 123-128. 18 Keyes, P.H. (1960). The infectious and transmissible nature of experimental dental caries. 19 Findings and implications. Archives of Oral Biology 1, 304-320. 20 Keyes, P.H. (1968). Research in dental caries. Journal of American Dental Association 76, 1356-21 1373. 22 Klein, M.N., Florio, F.M., Pereira, A.C., Hofling, J.F., and Goncalves, R.B. (2004). Longitudinal 23 study of transmission, diversity, and stability of Streptococcus mutans and Streptococcus 24 sobrinus genotypes in Brazilian nursery children. Journal of Clinical Microbiology 42, 4620-25 4626. 26 Koga, T., Asakawa, H., Okahashi, N., and Hamada, S. (1986). Sucrose-dependent cell adherence 27 and cariogenicity of serotype-C Streptococcus mutans. Journal of General Microbiology 132, 28 2873-2883. 29 Kohler, B., and Andreen, I. (1994). Influence of caries-preventive measures in mothers on 30 cariogenic bacteria and caries experience in their children. Archives of Oral Biology 39, 907-31 911. 32 Kohler, B., Lundberg, A.B., Birkhed, D., and Papapanou, P.N. (2003). Longitudinal study of 33 intrafamilial mutans streptococci ribotypes. European Journal of Oral Sciences 111, 383-389. 34 Kozai, K., Nakayama, R., Tedjosasongko, U., Kuwahara, S., Suzuki, J., Okada, M., and 35 Nagasaka, N. (1999). Intrafamilial distribution of mutans streptococci in Japanese families and 36 possibility of father-to-child transmission. Microbiology and Immunology 43, 99-106. 37 Kozai, K., Wang, D.S., Sandham, H.J., and Phillips, H.I. (1991). Changes in strains of mutans 38 streptococci induced by treatment with chlorhexidine varnish. Journal of Dental Research 70, 39 1252-1257. 40 Krasse, B., Edwardsson, S., Svensson, I., and Trell, L. (1967). Implantation of caries inducing 41 streptococci in the human oral cavity. Archive of Oral Biology 12, 231-236. 42 Kreulen, C.M., deSoet, H.J., Hogeveen, R., and Veerkamp, J.S.J. (1997). Streptococcus mutans 43 in children using nursing bottles. Journal of Dentistry for Children 64, 107. 44

36

Kulkarni, G.V., Chan, K.H., and Sandham, H.J. (1989). An investigation into the use of 1 restriction endonuclease analysis for the study of transmission of mutans streptococci. Journal of 2 Dental Research 68, 1155-1161. 3 Kuramitsu, H.K. (2003). Molecular genetic analysis of the virulence of oral bacterial pathogens: 4 An historical perspective. Critical Reviews in Oral Biology & Medicine 14, 331-344. 5 Li, Y., and Caufield, P.W. (1995). The fidelity of initial acquisition of mutans streptococci by 6 infants from their mothers. Journal of Dental Research 74, 681-685. 7 Li, Y.H., Lau, P.C.Y., Lee, J.H., Ellen, R.P., and Cvitkovitch, D.G. (2001). Natural genetic 8 transformation of Streptococcus mutans growing in biofilms. Journal of Bacteriology 183, 897-9 908. 10 Loesche, W.J. (1982). Dental caries: a treatable infection. 3rd edition, Springfield). 11 Loesche, W.J. (1986). Role of Streptococcus mutans in human dental decay. Microbial Reviews 12 50, 353-380 13 Marsh , P.D. (2005). Dental plaque: biological significance of a biofilm and community life-14 style. J Clinical Periodontology 32, 7-15. 15 Matsusaki, H., Endo, N., Sonomoto, K., and Ishizaki, A. (1996). Lantibiotic nisin Z fermentative 16 production by Lactococcus lactis IO-1: relationship between production of the lantibiotic and 17 lactate and cell growth. Appl Microbiology and Biotechnology 45, 36-40. 18 Mattos-Graner, R.O., Li, Y.H., Caufield, P.W., Duncan, M., and Smith, D.J. (2001). Genotypic 19 diversity of mutans streptococci in Brazilian nursery children suggests horizontal transmission. 20 Journal of Clinical Microbiology 39, 2313-2316. 21 Mohan, A., Morse, D.E., O'Sullivan, D.M., and Tinanoff, N. (1998). The relationship between 22 bottle usage/content, age, and number of teeth with mutans streptococci colonization in 6-24-23 month-old children. Community Dentistry and Oral Epidemiology 26, 12-20. 24 Nauth, K.R., and Lynum, M. (2000a). Stabilization of cream cheese compositions using nisin-25 producing cultures. United States Patent 6,110,509 (August 29, 2000). 26 Nauth, K.R., and Lynum, M. (2000b). Stabilization of mayonnaise spreads using whey from 27 nisin-producing cultures. United States Patent 6,113,954 (September 5, 2000). 28 Nauth, K.R., and Brooks, S. (2000). Stabilization of fermented dairy compositions using whey 29 from nisin-producing cultures. United States Patent 6,136,351 (October 24, 2000). 30 Nauth, K.R., Ruffie D.D., and Roman, M.G. (2001). Stabilization of cooked meat compositions 31 stabilized by nisin-containing whey and method of making. United States Patent 6,242,017 (June 32 5, 2001). 33 Parrot, M., Caufield, P.W., and Lavoie, M.C. (1990a). Preliminary characterization of 4 34 bacteriocins from Streptococcus mutans. Canadian Journal of Microbiology 36, 123-130. 35 Parrot, M., Drean, M.F., Trahan, L., and Lavoie, M.C. (1990b). Incidence of bacteriocinogeny 36 among fresh isolates of Streptococcus mutans. Canadian Journal of Microbiology 36, 507-509. 37 Pasch, J.H., Roman, M.G., Brooks, S., and Bell, J.L. (2004). Stabilization of cooked pasta 38 compositions using whey from nisin-producing cultures. United States Patent 6,797,308 39 (September 28, 2004). 40 Perry, D., and Kuramitsu, H.K. (1981). Genetic transformation of Streptococcus mutans. 41 Infection and Immunity 32, 1295-1297. 42 Person, C. (1967). Genetical adjustment of fungi to their environment. In The Fungi, G. A. A. 43 Sussman, ed. (New York, New York Academic Press), pp. 395-415. 44 Rogers, A.H. (1976). Bacteriocinogeny and properties of some bacteriocins of Streptococcus 45 mutans. Archives of Oral Biology 21, 99-104. 46

37

Ruangsri, P., and Orstavik, D. (1977). Effect of the acquired pellicle and of dental plaque on the 1 implantation of Streptococcus mutans on tooth surfaces in man. Caries Research 11, 204-210. 2 Sahl, H.G., Kordel, M., and Benz, R. (1987). Voltage-dependent depolarization of bacterial 3 membranes and artificial lipid bilayers by the peptide antibiotic nisin. Archives of Microbiology 4 149, 120-124. 5 Smith, L., Novak, J., Rocca, J., McClung, S., Hillman, J.D., and Edison, A.S. (2000). Covalent 6 structure of mutacin 1140 and a novel method for the rapid identification of lantibiotics. 7 European Journal of Biochemistry 267, 6810-6816. 8 Smith, L., Zachariah, C., Thirumoorthy, R., Rocca, J., Novak, J., Hillman, J.D., and Edison, A. 9 S. (2003). Structure and dynamics of the lantibiotic mutacin 1140. Biochemistry 42, 10372-10 10384. 11 Stashenko, K.P., and Hillman, J.D. (1989). Microflora of plaque in rats following infection with 12 an LDH-deficient mutant of Streptococcus rattus. Caries Research 23, 375-377. 13 Takahashi, S., Abbe, K., and Yamada, T. (1982). Purification of pyruvate formate-lyase from 14 Streptococcus mutans and its regulatory properties. Journal of Bacteriology 149, 1034-1040. 15 Tanner, A.C.R., Milgrom, P.M., Kent, R., Mokeem, S.A., Page, R.C., Riedy, C.A., Weinstein, P., 16 and Bruss, J. (2002). The microbiota of young children from tooth and tongue samples. Journal 17 of Dental Research 81, 53-57. 18 Tanzer, J.M., Krasse, B., and Svanberg, M. (1982). Conditions for implantation of Streptococcus 19 mutans mutant 805 in adult human mouths (abstract). Journal of Dental Research 61 (Spec. Iss), 20 334. 21 Toi, C.S., Cleaton-Jones, P., and Fatti, P. (2005). Characterization of Streptococcus mutans 22 diversity by determining restriction fragment-length polymorphisms of the gtfB gene of isolates 23 from 5-year-old children and their mothers. Antonie Van Leeuwenhoek International Journal of 24 General and Molecular Microbiology 88, 75-85. 25 van Houte, J. (1980). Bacterial specificity in the etiology of dental caries. International Dental 26 Journal 30, 305-326. 27 Vanderhoeven, J.S., and Rogers, A.H. (1979). Stability of the resident microflora and the 28 bacteriocinogeny of Streptococcus mutans as factors affecting its establishment in specific 29 pathogen-free rats. Infection and Immunity 23, 206-212. 30 Vuyst, D., and Vandamme, E.J. (1993). Influence of the phosphorus and nitrogen source on nisin 31 production in Lactococcus lactis subsp. lactis batch fermentations using a complex medium. 32 Appl Microbioly and Biotechnology 40, 17-22. 33 Wan, A.K.L., Seow, W.K., Purdie, D.M., Bird, P.S., Walsh, L.J., and Tudehope, D.I. (2003). A 34 longitudinal study of Streptococcus mutans colonization in infants after tooth eruption. Journal 35 of Dental Research 82, 504-508. 36 Whiley, R.A., Russell, R.R.B., Hardie, J.M., and Beighton, D. (1988). Streptococcus downei Sp. 37 Nov for strains previously described as Streptococcus mutans Serotype-H. International Journal 38 of Systematic Bacteriology 38, 25-29. 39 40 41 42

38

1

2 3 Figure 1. (A) Representative figure of the covalent structure of the lantibiotic nisin A showing 4 the thioether bridge linkages. (B) Representative figure of the covalent structure of the 5 lantibiotic mutacin 1140 showing the thioether bridge linkages. 6 7 8 9 10 11 12 13