borrelia channel-forming proteins: structure and function

Borrelia channel-forming proteins: structure and function

Ignas Bunikis

Department of Molecular Biology Laboratory for Molecular Infection Medicine Sweden (MIMS)

Umeå University Umeå 2010

III | P a g e

To my parents Skiriu tėvams

“May the Force be with you.”

Yoda "No problem! :)"

Iggy

IV | P a g e

V | P a g e

Table of Contents

Table of Contents .................................................................................................. V

Abstract .............................................................................................................. VIII

Papers in this thesis .............................................................................................. IX

Papers not included in this thesis .......................................................................... X

Introduction ........................................................................................................... 1

Lyme disease and relapsing fever ...................................................................... 2

History ........................................................................................................... 2

Clinical manifestations ................................................................................... 3

Detection and treatment ............................................................................... 4

Prevention ..................................................................................................... 6

Borrelia .............................................................................................................. 8

Classification .................................................................................................. 8

General characteristics .................................................................................. 9

Transmission and adaptation ...................................................................... 11

Genomes of Borrelia ........................................................................................ 14

Borrelia genomes ......................................................................................... 14

Plasmids and virulence ................................................................................ 15

Genomic relationships ................................................................................. 16

Metabolism of Borrelia .................................................................................... 18

Limited de novo biosynthetic capacity of Borrelia ...................................... 18

Carbon utilization ........................................................................................ 18

ATP synthesis ............................................................................................... 20

Cell envelope of Borrelia ................................................................................. 21

Borrelia membranes .................................................................................... 21

General characteristics of porins ................................................................. 22

VI | P a g e

Characterization of porins ........................................................................... 24

Porins in spirohetes ..................................................................................... 25

P66 ............................................................................................................... 26

Efflux through channel-tunnels ....................................................................... 28

Export systems ............................................................................................. 28

Efflux pumps ................................................................................................ 29

RND superfamily .......................................................................................... 30

Accessory proteins ....................................................................................... 30

Channel-tunnels........................................................................................... 31

Structure and characteristics of TolC .......................................................... 32

Aims of the thesis ................................................................................................ 35

Results and discussion ......................................................................................... 36

Paper I. ............................................................................................................. 36

Identification of efflux system in a Borrelia ................................................ 36

Involvement of BesC in antibiotic resistance and virulence ........................ 37

Channel-forming properties of purified BesC.............................................. 38

A structural model of BesC .......................................................................... 39

BesA is an atypical adaptor protein ............................................................. 40

Paper II. ............................................................................................................ 41

Purification and identification of a 38-kDa, pore-forming protein ............. 41

Analysis of the amino acid sequences of Oms38 ........................................ 42

Biophysical properties of Oms38 ................................................................. 42

Paper III. ........................................................................................................... 44

New porin in B. burgdorferi outer membrane ............................................ 44

Presence of DipA in other Borrelia species and homology to Oms38 ......... 45

DipA exhibits specificity for dicarboxylates ................................................. 45

Paper IV. .......................................................................................................... 48

VII | P a g e

The high interspecies homology of P66 ...................................................... 48

Purification and identification of the P66 homologues in Lyme disease and relapsing fever Borrelia spirohetes.............................................................. 49

P66 homologues exhibit similar biophysical properties .............................. 50

Paper V. ........................................................................................................... 51

P66 pore size estimation ............................................................................. 51

Channel interior ........................................................................................... 53

P66 is composed of several subunits ........................................................... 53

Conclusions .......................................................................................................... 55

Acknowledgements ............................................................................................. 56

References ........................................................................................................... 59

Publications I-V .................................................................................................... 77

VIII | P a g e

Abstract Borrelia is a Gram-negative, corkscrew-shaped bacterium transmitted by infected ticks or lice. Borreliae are subdivided into pathogens of two diseases: Lyme disease, caused mainly by B. burgdorferi, B. afzelii and B. garinii; and relapsing fever caused primarily by B. duttonii, B. hermsii, B. recurrentis or B. crocidurae. Both diseases differ in their manifestations, duration times and dissemination patterns. Antibiotics are the major therapeutics, although unfortunately antibiotic treatment is not always beneficial. To date, drug resistance mechanisms in B. burgdorferi are unknown. Transporters of the resistance-nodulation-division (RND) family appear to be involved in drug resistance, especially in Gram-negative bacteria. They consist of three components: a cytoplasmic membrane export system, a membrane fusion protein (MFP), and an outer membrane factor (OMP). The major antibiotic efflux activity of this type in Escherichia coli is mediated by the tripartite multidrug resistance pump AcrAB-TolC. Based on the sequence homology we conclude that the besA (bb0140), besB (bb0141) and besC (bb0142) genes code for a similar efflux system in B. burgdorferi. We created a deletion mutant of besC. The minimal inhibitory concentration (MIC) values of B. burgdorferi carrying an inactive besC gene were 4- to 8-fold lower than in the wild type strain. Animal experiments showed that the besC mutant was unable to infect mice. Black lipid bilayer experiments were carried out to determine the biophysical properties of purified BesC. This study showed the importance of BesC protein for B. burgdorferi pathogenicity and resistance to antibiotics, although its importance in clinical isolates is not known. Due to its small genome, Borrelia is metabolically and biosynthetically deficient, thereby making it highly dependent on nutrients provided by their hosts. The uptake of nutrients by Borrelia is not yet completely understood. We describe the purification and characterization of a 36-kDa protein that functions as a putative dicarboxylate-specific porin in the outer membrane of Borrelia. The protein was designated as DipA, for dicarboxylate-specific porin A. DipA was biophysically characterized using the black lipid bilayer assay. The permeation of KCl through the channel could be partly blocked by titrating the DipA-mediated membrane conductance with increasing concentrations of different organic dicarboxylic anions. The obtained results imply that DipA does not form a general diffusion pore, but a porin with a binding site specific for dicarboxylates which play important key roles in the deficient metabolic and biosynthetic pathways of Borrelia species. The presence of porin P66 has been shown in both Lyme disease and relapsing fever spirochetes. In our study, purified P66 homologues from Lyme disease species B. burgdorferi, B. afzelii and B. garinii and relapsing fever species B. duttonii, B. recurrentis and B. hermsii were compared and their biophysical properties were further characterized in black lipid bilayer assay. Subsequently, the channel diameter of B. burgdorferi P66 was investigated in more detail. For this study, different nonelectrolytes with known hydrodynamic radii were used. This allowed us to determine the effective diameter of the P66 channel lumen. Furthermore, the blockage of the channel after addition of nonelectrolytes revealed seven subconducting states and indicated a heptameric structure of the P66 channel. These results may give more insight into the functional properties of this important porin.

IX | P a g e

Papers in this thesis

This thesis is based on the following articles and manuscripts, which will be referred to by their Roman numerical (I-V).

I. An RND-type efflux system in Borrelia burgdorferi is involved in virulence and resistance to antimicrobial compounds. Bunikis I, Denker K, Östberg Y, Andersen C, Benz R, Bergström S. PLoS Pathog. 2008 Feb 29;4(2):e1000009.

II. Oms38 is the first identified pore-forming protein in the outer membrane of relapsing fever spirochetes. Thein M, Bunikis I, Denker K, Larsson C, Cutler S, Drancourt M, Schwan TG, Mentele R, Lottspeich F, Bergström S, Benz R. J Bacteriol. 2008 Nov;190(21):7035-42. Epub 2008 Aug 29.

III. DipA, a pore-forming protein in the outer membrane of Lyme disease Borrelia exhibits specificity for the permeation of dicarboxylates. Thein M, Bonde M*, Bunikis I*, Denker K, Sickmann A, Bergström S, Benz R. (Submitted)

IV. P66 porins are present in both Lyme disease and relapsing fever spirochetes: a comparison of the biophysical properties of P66 porins from six Borrelia species. Barcena-Uribarri I, Thein M, Sacher A, Bunikis I, Bonde M, Bergström S, Benz R. Biochim Biophys Acta. 2010 Feb 24.

V. The use of nonelectrolytes as molecular tools reveals the channel size and the oligomeric constitution of the Borrelia burgdorferi porin P66. Barcena-Uribarri I, Thein M, Maier E, Bonde M, Bunikis I, Bergström S, Benz R. (Manuscript)

* These authors contributed equally to this work.

X | P a g e

Papers not included in this thesis

Contribution has been made by the author to the following articles that are not included in this thesis:

The etiological agent of Lyme disease, Borrelia burgdorferi, appears to contain only a few small RNA molecules. Östberg Y, Bunikis I, Bergström S, Johansson J. J Bacteriol. 2004 Dec;186(24):8472-7. Tickborne relapsing fever diagnosis obscured by malaria, Togo. Nordstrand A, Bunikis I, Larsson C, Tsogbe K, Schwan TG, Nilsson M, Bergström S. Emerg Infect Dis. 2007 Jan;13(1):117-23.

1 | P a g e

Introduction

Borrelia is a fascinating organism! So is the research on this bacterium! It reminds a little bit of jigsaw puzzle. The only big difference is that this particular puzzle does not have a defined size or a known number of pieces... Since its discovery, scientists have been captivated by this pathogen. In addition to many other studies, throughout the years hard and extensive work has been done in studying and understanding epidemiology, entomology, spirochete morphology, disease progression, its complex life cycle, virulence mechanisms, phenomenal antigenic variation, metabolism, host-pathogen interactions, genome organization, molecular mechanisms underlying cell processes, structural and biophysical properties of smallest building blocks of Borrelia. Each research group, each scientist contributes in solving fragments of a big picture by piecing together knowledge, ideas, experiments and discoveries made every day. In this thesis I present a few small puzzle pieces with the hope that they can fill in some gaps in the knowledge of Borrelia outer membrane components, their role in the pathogenicity and virulence of this spirochete. Perhaps one day, the combined efforts of many researchers around the globe will reveal the complete picture of this intriguing and unique bacterial species when all the pieces of the puzzle fall into place. But till then, there are many unsolved mysteries to be dealt with, many unanswered or unasked question to be answered or asked, and many discoveries to be made. There's plenty of research to be done, no question about that!

2 | P a g e

Lyme disease and relapsing fever

History The first known case of acrodermatitis chronica atrophicans, a late skin manifestation of Lyme borreliosis was described in 1883 by German physician Buchwald (Buchwald 1883). Later, a case of erythema migrans, the characteristic skin rash at the initial stage of Lyme borreliosis, was described by the Swedish dermatologist Arvid Afzelius, who suggested that the reason for the rash was the bite of a tick or another insect (Afzelius 1921). The first description of Lyme arthritis was reported in 1977 when Steere and coworkers described an unusually high incidence of acute arthritis among children in Old Lyme, Connecticut (Steere, Malawista et al. 1977). Several patients reported the development of a skin rash before the onset of arthritis. Because of the geographical clustering of the cases in the wooded areas and peak occurrence in the summer, Steere et al. postulated that the symptoms arose from transmission of an agent by an arthropod vector. A few years later, in 1981, while looking for rickettsiae, William Burgdorfer encountered long, irregularly coiled spirochetes when examining the midguts of Ixodes scapularis ticks (Burgdorfer 1984). In 1982, Burgdorfer and coworkers reported the correlation between the spirochete and Lyme disease (Burgdorfer, Barbour et al. 1982). In honor of its discoverer, the Lyme disease spirochete was named Borrelia burgdorferi (Johnson, Schmid et al. 1984). To acknowledge early findings of the pioneer Afzelius, a common etiological agent of Lyme disease in Europe was named B. afzelii (Canica, Nato et al. 1993).

In contrast to Lyme disease, relapsing fever has been recorded and described since ancient times, even though the disease had not been given its contemporary name and the causative agent had not yet been determined. An epidemic fever was described by Hippocrates, called the "ardent fever", and during the following centuries preceding modern times, several outbreaks of epidemic fever had been suggested to be relapsing fever (Felsenfeld 1971). In middle of the 18th century, the first clinical features were well described in a documented outbreak in Ireland. This epidemic spread over the British Islands, and the disease was for the first time designated "relapsing fever" in 1843 (Southern and Sanford 1969; Felsenfeld 1971). From England, relapsing fever was introduced to the European continent and the United States, causing

3 | P a g e

outbreaks in populations living under poor conditions. The organism causing the epidemic disease was discovered in 1868 by the German physician Obermeier, and was named Spirocheta obermeieri, but later on the organism was designated Borrelia recurrentis (Felsenfeld 1971). The biggest outbreaks of epidemic relapsing fever occurred during World War I and II when hundreds of thousands soldiers and civilians were infected. A similar disease was described on the African continent for the first time by the famous explorer Dr. Livingstone in 1857, but it was not until the beginning of the 20th century that presence of Borrelia in blood samples was noticed. Relapsing fever has since then been reported in almost all parts of the world, with some exceptions (Felsenfeld 1971).

Clinical manifestations Clinically, Lyme disease is most similar to syphilis, both exhibiting multisystem involvement and mimicry of other diseases (Strle, Picken et al. 1997). The most common sign of Lyme borreliosis is a red skin rash called erythema migrans observed in 80-89% of all the clinical cases in Europe and North America (Steere 1989; Mehnert and Krause 2005). The erythema migrans expands from the site of tick bite with central clearing, and without treatment it can reach one meter in diameter (Stanek and Strle 2003). The skin lesion is often accompanied by influenza-like symptoms such as nausea, fatigue, fever and headache. This is followed by a disseminated infection within days or weeks affecting the nervous system, joints, skin and heart. Subsequently, a persistent infection can evolve within weeks or months (Steere 2001). European patients can develop a chronic skin manifestation called acrodermatitis chronica atrophicans. Live spirochetes have been isolated from such lesions more than 10 years after the onset of the disease (Asbrink and Hovmark 1985). In about 10% of the patients, no symptoms of infection can be seen (Steere 2001).

In contrast to Lyme borreliosis, relapsing fever is an acute disease with a high degree of bacteremia. The incubation period is 2-18 days after the tick bite. An infected patient develops a sudden high fever, accompanied by chills, severe headache and other undefined flu-like symptoms. Borrelia spirochetes are highly invasive and can be found in variety of tissues, although spleen and liver are affected the most during early infection. The symptomatic period of

4 | P a g e

relapsing fever lasts for 3-5 days. Thereafter, the patient experiences a pronounced temperature elevation, followed by a sudden temperature drop down to normal or even lower level than pre-infection. The febrile period is a result of massive bacteremia as the Borrelia multiplies in the blood, reaching as many as 108 bacteria per milliliter of blood. After the body temperature has gone down to normal, a period of well-being occurs when no bacteria are detected in the blood. An afebrile period lasts for 4-7 days whereafter the disease relapses with subsequent fever peaks and spirochetemia. The first attack is commonly more severe and prolonged than the following relapses (Southern and Sanford 1969). Borrelia can penetrate the blood-brain barrier and infect brain, which can be manifested by meningitis, facial palsy and neuritis (Cadavid and Barbour 1998). In addition, relapsing fever is associated with a high incidence of pregnancy complications, such as spontaneous abortions and preterm delivery (Jongen, van Roosmalen et al. 1997).

Detection and treatment Tick bite history and general awareness after staying in a Lyme borreliosis endemic area are crucial for diagnosis. Persons who developed a skin lesion or other illness within one month after removing an attached tick should consult a physician. The erythema migrans is still one of the most easily noticeable signs of an infection (Wormser, Nadelman et al. 2000). Detection of Borrelia by polymerase chain reaction (PCR) is useful, but the method has a major drawback since it cannot distinguish between living and dead bacteria. However, it can be the method of choice in later stages of disease (Steere 2001). Isolation of bacteria in culture is the ultimate proof, although it is time-consuming, has low sensitivity and depends on infectious species and the stage of disease (Nadelman and Wormser 1998; Wilske 2005). Antibody detection of surface localized antigens is a common, constantly improving method, but cannot be applied at the initial stages of the disease (Steere 2001). When used, it is recommended that positive Enzyme-Linked Immunosorbent Assay (ELISA) results are confirmed by Western blot analysis (Wilske 2005).

Due to diffuse first symptoms resembling influenza, relapsing fever can be difficult to diagnose. Recently, it has been shown that relapsing fever can be obscured by malaria infection (Nordstrand, Bunikis et al. 2007). Prevalence of

5 | P a g e

malaria in the same areas as relapsing fever and symptomatic similarities of both diseases can lead to misdiagnosis and mistreatment. In addition, very often a tick bite is left unnoticed because of the nocturnal lifestyle of the soft-bodied ticks transmitting Borrelia and their ability to complete the blood-meal in as fast as a few minutes. Together with the fact that relapsing fever does not cause erythema migrans, the start of the disease becomes difficult to detect.

The easiest way to detect ongoing relapsing fever infection is by examining a blood sample from the infected patient, since the Borrelia could be detected immediately by dark-field microscopy or by Wright-Giemsa staining. However, such detection requires a sufficient number of bacteria in the blood, such as during the first peak of spirochetemia, a time period that is very easily missed. The sensibility of this method could be improved by removing the red blood cells and concentrating the spirochetes in the plasma (Larsson and Bergström 2008). Serological tests are available, although such tests have limited sensitivity due to high antigenic variation and cross-reactivity with other Borrelia species. GlpQ, a protein that is conserved among all members of relapsing fever Borrelia, but is absent in Lyme disease spirochetes, can be used to discriminate between these two agents (Schwan, Schrumpf et al. 1996; Nordstrand, Bunikis et al. 2007). Even though serological diagnosis of relapsing fever is possible, the time point for effective treatment could be missed. The antibody response towards the bacterium develops after approximately 4-6 days (Yokota, Morshed et al. 1997), sufficient time for Borrelia to disseminate to tissues such as brain with limited access for antibiotics (Andersson, Nordstrand et al. 2007).

In most cases, use of antibiotics against Borrelia results in successful treatment (Bryceson, Parry et al. 1970). Commonly used antibiotics are penicillin, erythromycin and tetracycline. The clinical manifestations and age of the patient can influence type and length of the antibiotic treatment (Wormser, Nadelman et al. 2000). In the case of relapsing fever, it should be kept in mind that treatment with antibiotics can cause a life-threatening Jarish-Herxheimer reaction provoked by release of bacterial components when bacteria are lysed by antibiotics (Bryceson 1976).

6 | P a g e

Prevention The most effective approach to prevent and reduce the incidence of Lyme disease is to avoid contact with infected ticks. Limited outdoor activities in tick infested areas and properly tucked clothes not leaving any exposed skin provide effective protection. The risk of attracting ticks also can be reduced by application of mosquito or tick repellents. Still, removing any attached ticks within 24 hours of attachment reduces the chance of infection dramatically. It is therefore important to examine the body for ticks when staying in endemic areas (Piesman, Mather et al. 1987). Although difficult, reducing the vector population through habitat manipulation and insecticides is possible. Another approach is vaccination. Unfortunately, currently there is no vaccine available against Lyme borreliosis. However, a vaccine called LYMErix was commercially available in the USA from 1998 but withdrawn from the market in 2002 due to rare cases of vaccinated persons developing autoimmune arthritis, although this was never proven (Steere 2006). The vaccine was based on recombinant outer surface protein A (rOspA) and conferred protection in a unique way. Bactericidal antibodies generated against OspA eliminate B. burgdorferi within the tick during feeding, preventing infectious spirochetes from entering the host (Fikrig, Telford et al. 1992). LYMErix was only effective against B. burgdorferi infections, and therefore availability was restricted to the USA, where this genospecies alone is responsible for all clinical cases. In Europe and Asia, B. burgdorferi is less common. Instead, B. garinii and B. afzelii are the major causative agents for Lyme borreliosis. A vaccine for the European market therefore needs to provide protection against all three genospecies. Now LYMErix vaccine is available for dogs only (Ma, Hine et al. 1996). Different attempts to identify new, suitable vaccine candidates have been tried through the years, but this has not yet led to any commercially available vaccine.

The situation is similar for relapsing fever, currently there is no vaccine available. Therefore human infection can only be avoided by reducing or eliminating the vectors that transmit the disease. This can be difficult in areas where the vector lives in close association with humans. Relapsing fever transmitting ticks live in the cracks and crevices in over 80% of mud-huts (Barclay and Coulter 1990), a common type of house in sub-Saharan Africa (Sonenshine 1997). Different repellants and bed-nets have been used and proven to be successful in reducing exposure of humans to the ticks

7 | P a g e

(Sonenshine 1997). Fighting against louse-borne relapsing fever is even more difficult. Improvement of living conditions and hygienic standards would help eradicate the louse, although in areas affected by poverty this is a very hard task.

8 | P a g e

Borrelia

Classification Borrelia belongs to the Spirochete phylum in the class of Spirochaetes and the order of Spirochetales. Examples of other human disease-causing spirochetes are Treponema pallidum (causing syphilis), T. denticola (periodontal diseases) and Leptospira interorrgans (leptospirosis) (Paster and Dewhirst 2000). The genus Borrelia is comprised of many different genospecies, but the two groups causing disease in humans are Lyme Borrelia genospecies and relapsing fever Borrelia genospecies.

The Lyme disease Borrelia group of spirochetes includes 12 known genospecies. B. burgdorferi, B. garinii and B. afzelii are known to cause disease in humans (Stanek and Strle 2003). Lately, a few clinical cases due to B. bissetii, B. lusitaniae and B. spielmanii infections also have been reported (Maraspin, Cimperman et al. 2002; Collares-Pereira, Couceiro et al. 2004; Fingerle, Schulte-Spechtel et al. 2008). The tick vector, geographical distribution and reservoir animals differ between the genospecies. In North America, clinical cases of Lyme borreliosis are due to infection of B. burgdorferi. In Europe and Asia the situation is more complex due to three causative agents, namely B. garinii, B. afzelii and B. burgdorferi, although the latter is not very common in Asia (Kurtenbach, Hanincova et al. 2006). In addition, these genospecies are spread by two different vectors: Ixodes ricinus in Europe and I. persulcatus in Asia. The habitats of these two ticks overlap in Eastern Europe (Kurtenbach, Hanincova et al. 2006). Lyme disease Borrelia can be genotyped using a phenotypic method which includes monoclonal antibodies reacting against different classes of the outer surface protein A (OspA). Using this method, seven OspA serotypes have been identified (Wilske, Preac-Mursic et al. 1993). OspC has also been the basis of classification with monoclonal antibodies, separating Lyme Borrelia into additional thirteen classes (Wilske, Jauris-Heipke et al. 1995). In recent years, the phenotypic typing methods are being replaced in favor of genotypic methods based on DNA sequence comparison. Genes used for genotyping are p66 (porin), flaB (a structural protein of flagella), ospA and ospC (Assous, Postic et al. 1994; Pinne, Thein et al. 2007). The drawback using these genes is the high level of sequence similarity among different genospecies. In a comparative study the intergenic spacer (IGS) located on the chromosome between rrs (16S)

9 | P a g e

and rrlA (23S) genes was identified as the most suitable locus for genospecies identification and evolutionary studies since this intergenic sequence is not under selective pressure and therefore is highly polymorphic (Bunikis, Garpmo et al. 2004).

The relapsing fever group is further divided into "new world" and "old world" Borrelia species based on the continent in which their vectors exist. Old world species are present on the European, African and Asian continents, whereas new world species are defined as those present in South and North America. Many known relapsing fever species are classified according to their association with their arthropod vector. The ticks are adapted to live in their specific microhabitat, thereby making the Borrelia also specific for a certain geographic location (Sonenshine 1997). The tick-Borrelia relationship is in many cases so specific that one Borrelia species cannot be transmitted by the vector specific for other Borrelia species (Barbour and Hayes 1986). There has also been a division of species based on the range of mammals that they infect. For example, louse-borne B. recurrentis is considered to be a strictly human-specific pathogen. New technologies and genotyping based on DNA sequencing lead to a revision of the phylogeny. The 16S rRNA gene has been used to compare and distinguish different species, although due to a high degree of conservation and only minor sequence differences, it is hard to separate closely-related organisms (Fox, Wisotzkey et al. 1992; Ras, Lascola et al. 1996). Therefore, the above described genotyping based on intergenic spacer sequence is a valuable alternative.

General characteristics The Borrelia spirochetes have an obligate parasitic lifestyle whereby they are only able to survive in a tick vector or a vertebrate host. They are coiled, up to 30 µm long and less than 1 µm thick (Figure+1) (Goldstein, Buttle et al. 1996). Variation can, however, occur between genospecies. The generation time is determined to be on average 5 to 12 hours in vitro during the logarithmic growth phase at the optimal temperature of about 35°C, which is extremely long compared to other bacteria (Barbour 1984). The bacterium consists of an outer membrane with numerous lipoproteins on the surface, a peptidoglycan layer in the periplasmic space, and an inner membrane that surrounds the

10 | P a g e

cytoplasm (Barbour and Hayes 1986). All Borrelia species are highly motile and move by way of a spinning/wave-like motion by rotating their flagella clockwise or counter-clockwise (Goldstein, Charon et al. 1994). The flagella are inserted at the ends of the bacterium and, like the cytoskeleton of an eukaryotic cell, they render the spirochetes wave-like in appearance (Wolgemuth, Charon et al. 2006). The number of flagella differs between different species and can be as many as 30 (Charon, Greenberg et al. 1992). The lack of flagella results in a long rod-shaped non-motile bacterium (Motaleb, Corum et al. 2000). Although Borrelia have both inner and outer membranes and are generally considered to be Gram-negative bacteria, they have many features similar to Gram-positive bacteria such as the absence of lipopolysaccharide (LPS) and presence of membrane glycolipids (Takayama, Rothenberg et al. 1987; Hossain, Wellensiek et al. 2001). Borrelia has a very segmented genome with a linear chromosome and high number of circular and linear plasmids, which carry genes that are necessary for the bacteria to fulfil the infection cycle. The total size of the genome does not exceed 2M base pairs and encodes limited biosynthetic genes, suggesting that many nutritional components are taken up from the host (Fraser, Casjens et al. 1997; Casjens 2000). This makes the bacterium difficult to culture and a complex medium, Barbour-Stoenner-Kelly II (BSKII) (Barbour 1984), supplemented with rabbit serum is needed.

Figure 1. Borrelia burgdorferi. Photo taken by Ignas Bunikis in the year 2004, half a year before starting his PhD studies.

11 | P a g e

Transmission and adaptation Lyme disease and relapsing fever are zoonotic diseases, transmitted by the arthropod vector to different vertebrate hosts (Barbour and Hayes 1986). The bacteria are strictly dependent on the arthropod and the reservoir host to be maintained in nature, and have therefore a biphasic life cycle where they alternate between infecting the vector and the vertebrate. The success of this alternating lifestyle is dependent on adaptation to different environments. Several features contribute to the success of ticks as vectors for pathogen transmission: a long lifespan and durability, as they can resist starving for years, and an ability to infect a majority of vertebrates (Sonenshine 1997). During its life cycle, a tick goes through three developmental stages: the larva, the nymph and the adult stage. All three developmental forms require blood-meals before transformation to the next stage (Sonenshine 1997). The transmission of Borrelia occurs when a naϊve larva, nymph or tick feeds on a Borrelia-infected animal. Borrelia replicates in the midgut and when their number increases, they migrate towards the tick internal organs, such as nervous tissue, the salivary glands and the coxal organs (Barbour and Hayes 1986). The bacteria can persist in the tick vector for several years before they are transmitted to a vertebrate host with the blood-meal (Sonenshine 1997).

Ticks transmitting Borrelia are divided into two major families: the hard-bodied tick (Ixodidae) and the soft-bodied tick (Argasidae). The hard-bodied ticks are vectors for Borrelia species causing Lyme disease. Ixodes tick feeding is a slow process and can last from 3–5 days for a larva, up to 4–7 days for a nymph and 7–11 days for an adult tick (Sonenshine 1997). To be able to infect a new host, the spirochetes need to penetrate the epithelium of the tick midgut and migrate through the hemolymph to the salivary glands. In the case of the nymph, the migration process takes 36-48 hours and starts when the tick begins to feed and blood fills up the midgut (Ribeiro, Mather et al. 1987; De Silva and Fikrig 1995). During the first 24 hours of nymphal attachment, virtually no transmission of spirochetes to the host takes place. After 48 hours transmission, occurs but particularly efficient transmission takes place only after 72 hours (Piesman, Mather et al. 1987). Removing a feeding tick within 24 hours after attachment is therefore an efficient way of preventing spread of spirochetes to a new host. Tick feeding triggers a shift in expression of major outer surface proteins of Borrelia. In the unfed tick midgut, Lyme Borrelia spirochetes express OspA, a

12 | P a g e

lipoprotein anchored to the bacterial outer membrane (Schwan, Piesman et al. 1995). OspA has been shown to mediate adherence to the tick midgut epithelium (Pal, Li et al. 2004). This interaction is essential for colonization of the tick gut. When the tick starts to feed on a host, the incoming blood changes the environment of the midgut. The OspA expression is ceased and, instead, up-regulated expression of another lipoprotein OspC occurs (Schwan, Piesman et al. 1995). The reciprocal expression of OspA and OspC correlates with exit of spirochetes from midgut, migration through the hemolymph to the salivary glands and entry into the new host. OspC is essential for the initial infection of the mammalian host (Grimm, Tilly et al. 2004).

The largest group of Argasidae family is the Ornithodoros, which is the main vector for relapsing fever. As all ticks, they need blood-meals for their development. Soft-bodied ticks live in close proximity to their vertebrate hosts and are the most active during the night. Ornithodoros ticks are fast feeders — they do not consume large volumes of blood at once, instead they can feed several times and on multiple hosts. This increases the spreading frequency of bacteria (Sonenshine 1997). When the infected tick feeds on a naϊve animal, the bacteria are deposited into the vertebrate skin, rapidly migrate into circulation and multiply there, reaching high numbers. The bacteria subsequently leave the circulation and penetrate various tissues such as the liver, spleen, kidneys or brain. B. recurrentis, causing epidemic relapsing fever, is transmitted by the human body louse Pediculus humanus humanus. Lice are relatively short-lived and molt three times before reaching adult stage. They typically feed five times per day (Raoult and Roux 1999). Louse-ingested Borrelia takes the route from the midgut to the hemolymph where it multiplies. Later it disseminates to all cavities of the louse body without entering the tissues (Southern and Sanford 1969). The transmission of B. recurrentis to humans does not occur through the bite, instead the bacteria are transferred when the louse is crushed and the spirochete-containing hemolymph is rubbed into the bite site (Barbour and Hayes 1986).

Early researchers noticed that every relapse of the disease was caused by a new variant of the original infecting strain. Eventually, it was evident that the relapsing feature was a result of antigenic variation of the dominant surface protein, the Vmp's (Barbour, Tessier et al. 1982; Stoenner, Dodd et al. 1982). During a fever peak, the majority of the spirochetes express one type of Vmp,

13 | P a g e

eliciting the host to mount an antibody response towards this serotype, subsequently clearing the organisms from the circulation. However, a second wave of bacteria, expressing a different Vmp, will multiply and cause the disease to relapse. Antigenic variation contributes to maintenance of Borrelia in nature since it prolongs bacterial survival in the circulation, thereby increasing the chance of successful transmission to a naϊve host (Barbour and Hayes 1986). In addition, Vmp's are involved in adaptation to various surroundings and determinants of tissue tropism (Cadavid, Thomas et al. 1994; Cadavid, Pachner et al. 2001).

14 | P a g e

Genomes of Borrelia

Borrelia genomes Borrelia is genomically unique and not closely related to any other bacteria, including the other spirochetes. B. burgdorferi and B. hermsii represent the two major branches within the Borrelia genus, wherein B. burgdorferi typifies the ‘Lyme disease agent-like’ branch and B. hermsii typifies the ‘relapsing fever agent-like’ branch (Paster, Dewhirst et al. 1991). Each Borrelia species carries a linear chromosome about 900 kbp in length and multiple circular and linear low copy number plasmids that are usually, but not always, in the 9–62 kbp range. The chromosomes of these bacteria all have quite similar gene contents, whereas the plasmids are more variable. Individual Borrelia cells harbor a more diverse plasmid complement than that of any other bacterium — the sequenced genome of B. burgdorferi type strain B31 has 21 plasmids, and several additional plasmids appear to have been lost between isolation of the strain in 1982 and the completion of its genome sequence (Casjens, Palmer et al. 2000). Overall, the 900-kbp chromosomes, often called the ‘large’ or ‘main’ chromosomes, carry the great majority of the genes that encode metabolic enzymes, and the 400 to 650 kbp of plasmid DNAs carry the bulk of the surface lipoprotein encoding genes. The chromosome carries tightly packed genes, as is typical of bacteria, while many of the linear plasmids have substantially lower gene densities and many apparently decaying pseudogenes (Casjens 2000; Casjens, Palmer et al. 2000).

The complete 910,725-bp sequence of the B. burgdorferi strain B31 linear chromosome was published in 1997 by Fraser et al. (Fraser, Casjens et al. 1997), and it remains the only Borrelia strain whose genome sequence is known to be complete (some plasmid sequences remain unassembled in the others). Its nucleotide composition is 28.6% G+C and is predicted to contain 846 apparently intact protein encoding genes and ten or fewer genes that appear to be truncated or contain frame shift mutations; the fraction of mutational inactivated chromosomal genes is among the lowest of the analyzed bacterial genomes. Although smaller bacterial genomes are known, this chromosome size is near the small end of the spectrum (Casjens 1998). The protein-encoding genes occupy 93% of the chromosome, a typical value for a bacterial genome not undergoing current reduction in size (Lynch 2006). These protein-encoding

15 | P a g e

genes are quite highly clustered according to function, often suggesting transcription as operons; about 67% of the genes are oriented such that they are transcribed away from the center of the chromosome. About 60% of the strain B31 chromosome’s predicted genes have some similarity to a gene in another organism whose role or function is at least partly understood; about 10% are similar to known genes in other organisms whose roles are unknown; and about 30% are unique to Borrelia and have unknown functions (Fraser, Casjens et al. 1997). The chromosome carries what appears to be a rather minimal set of the genes that are required for cell maintenance and replication. The biosynthetic and intermediary metabolic capacity of Borrelia is very limited. Genes that encode enzymes that perform functions in respiration, amino acid synthesis, nucleotide synthesis, lipid synthesis and enzyme cofactor synthesis are almost completely lacking, in agreement with their many and fastidious requirements for growth in culture and their evolutionary adaptation to two highly specific host environments. Given the requirement that they scavenge essentially all amino acids, nucleosides, lipids and cofactors from their environment, it is perhaps surprising that ‘only’ 16 transporters were identified in the initial analysis, and only a few more have been identified since then. Some of these transporters probably have generic specificity and have been hypothesized to bring in multiple related small molecules (Fraser, Casjens et al. 1997).

Plasmids and virulence With the notable exception of cp26, the plasmids are not required for growth in culture (Sadziene, Wilske et al. 1993). A relationship between the plasmid content of B. burgdorferi and virulence has been well established. In B. burgdorferi strain B31, linear plasmids lp25 and lp28-1 are necessary for high infectivity in mice (Labandeira-Rey and Skare 2001; Purser, Lawrenz et al. 2003), while lp25 and lp28-4 are important for maintenance in ticks (Grimm, Tilly et al. 2005; Strother and de Silva 2005). Circular plasmid cp26 is found in all isolates and encodes OspC, which is required for establishment of infection in mice (Pal, Yang et al. 2004). Recently, lack of lp36 was shown to correlate with reduced infectivity in mice, and the defect could be partially complemented by the lp36-encoded adenine deaminase (Jewett, Byram et al. 2007). However, B. burgdorferi isolates that contain all these plasmids may still be attenuated for

16 | P a g e

virulence in mice (Wang, Ojaimi et al. 2002), suggesting that additional virulence factors remain to be identified.

Genomic relationships Since the publication of the B. burgdorferi strain B31 genome sequence, the complete nucleotide sequences of the chromosomes of B. garinii strain PBi and B. afzelii strain PKo have been published along with the complete sequences of some of their circular and linear plasmids (Glockner, Lehmann et al. 2004; Glockner, Schulte-Spechtel et al. 2006). In 2008, Lescot et al. reported the sequences of B. duttonii Ly and B. recurrentis A1 and some of their plasmids (Lescot, Audic et al. 2008). In addition, chromosome sequences for relapsing fever isolates B. hermsii DAH and B. turicatae 91E135 are available through GenBank. Data suggest that chromosomes of all the Borrelia spp. have highly similar gene orders. The analysis of the sequence of the B. garinii PBi and B. afzelii PKo chromosomes showed that the two chromosomes are essentially completely co-linear with the B. burgdorferi chromosome except for the regions very near the telomeres, and that the common regions of PBi and PKo are about 93% identical in nucleotide sequence (Glockner, Lehmann et al. 2004; Glockner, Schulte-Spechtel et al. 2006). PKo and PBi are about 93% identical whereas those of B31 and PKo are 91% identical, indicating that the three species, B. burgdorferi, B. garinii and B. afzelii, are approximately equidistantly related. Analysis of B. hermsii strain DAH and B. turicatae strain 91E135 chromosomal draft sequences showed that these two strains contain three purine metabolism genes that B. burgdorferi lacks (Pettersson, Schrumpf et al. 2007). In addition, the relapsing fever Borrelia chromosomes have two genes that encode enzymes for glycerol-3-phosphate acquisition that are not present in the Lyme disease spirochetes (Schwan, Battisti et al. 2003). Lescot et al. found only 13 genes unique to the Lyme agents and seventeen unique to the relapsing fever agents when their B. duttonii and B. recurrentis chromosome sequences were compared to the B. burgdorferi B31 chromosome sequence (Lescot, Audic et al. 2008). Thus, the relapsing fever and Lyme disease agent chromosomes are quite similar overall. The B. garinii PBi and B. afzelii PKo genomes carry eight and six linear plasmids, respectively, as well as at least three and nine circular plasmids, respectively. The circular plasmids are largely homologous to their B. burgdorferi relatives (Maier, Bremer et al. 1988; Glockner, Lehmann et al. 2004;

17 | P a g e

Glockner, Schulte-Spechtel et al. 2006). Among the linear plasmids, both PBi and PKo also carry plasmids that are very similar to B. burgdorferi lp54. The relapsing fever Borrelia also harbor numerous linear and circular plasmids in the same size ranges as those of the Lyme agents, and at least some isolates have one or two much larger linear plasmids in the 100–220 kbp size range (Lescot, Audic et al. 2008). The 180-kbp linear plasmid of B. hermsii HS1 carries nucleotide metabolism genes that are not present in the Lyme agent Borrelia genomes (Zhong, Skouloubris et al. 2006). Lescot et al. reported the sequences of seventeen B. duttonii and eight B. recurrentis plasmid sequence contigs (Lescot, Audic et al. 2008). From this work, it is clear that although the genes are similar, the gene organization of plasmids in relapsing fever Borrelia is often quite different from that in B. burgdorferi.

18 | P a g e

Metabolism of Borrelia

Limited de novo biosynthetic capacity of Borrelia Borrelia burgdorferi is an obligate bacterial pathogen with no known ability to survive outside of either the arthropod vector or mammalian host. This apparent inability to survive outside of a host implies that B. burgdorferi is restricted in terms of its metabolic capabilities and is consistent with its fastidious nature. On the other hand, the ability to survive in such disparate hosts as ticks and warm-blooded mammals implies that B. burgdorferi is adept at incorporating metabolites found in one or both of the arthropod and mammalian species they infect. Identification of first specific Borrelia porin involved putative nutrient uptake is described in Paper II and Paper III of this thesis. The genome sequence of B. burgdorferi indicates that the primary source for energy production is by fermentation via the Embden-Meyerhof pathway (Fraser, Casjens et al. 1997; Das, Hegyi et al. 2000; von Lackum and Stevenson 2005). No homologues of TCA enzymes or proteins involved in electron transport and oxidative phosphorylation are encoded in the genome (Fraser, Casjens et al. 1997; Das, Hegyi et al. 2000; von Lackum and Stevenson 2005). Despite the limited battery of enzymes involved in energy production, B. burgdorferi is able to survive and persist for long periods of time inside the host they colonize.

Carbon utilization B. burgdorferi uses phosphotransferase system (PTS), a family of well characterized sugar transporters, for acquisition of carbohydrates across the cytoplasmic membrane. It is predicted that Borrelia possesses six distinct PTS transporters which mediate the accumulation of glucose, N-acetylglucosamine, chitobiose, and mannose (Fraser, Casjens et al. 1997; Saier 2000; von Lackum and Stevenson 2005). Three of the six PTS transporters are thought to be competent for glucose uptake, suggesting that glucose is the preferred carbon source of B. burgdorferi. In addition to the PTS transporters, it is predicted that B. burgdorferi utilizes an Mgl ABC system for glucose transport. Mannose and chitobiose are predicted to be transported via PTS as well and used as alternative carbon sources (Fraser, Casjens et al. 1997; von Lackum and

19 | P a g e

Stevenson 2005). Although the genome sequence suggested that additional carbon sources, including fructose, galactose, and ribose, could be utilized, studies have shown Borrelia growth limitation in modified BSK media using these carbohydrates. It was determined that B. burgdorferi growth is restricted to glucose, maltose, N-acetylglucosamine, chitobiose, mannose, and glycerol as carbon sources during in vitro growth (von Lackum and Stevenson 2005).

Most likely, Borrelia uses glucose to produce galactose, which is required for the synthesis of the major membrane glycolipid, α-monogalactosyl diacylglycerol (α-MgalDAG) (Östberg, Berg et al. 2007). This glycolipid has been proposed to be required for membrane stability and function (Mannock, Lewis et al. 2001). The homologues of all of the enzymes required for the oxidative phase of the pentose phosphate pathway are present within B. burgdorferi, while several enzymes are missing from the non-oxidative phase of the pentose phosphate pathway suggesting that ribose is probably required for RNA and riboflavin biosynthesis and not for generating energy. This would explain the presence in Borrelia of a putative ribose transport system and inability to use it as a carbon source during in vitro growth (Fraser, Casjens et al. 1997; von Lackum and Stevenson 2005).

The transport of chitobiose is of particular interest as it is a derivative of the chitin exoskeleton of the arthropod vector, specifically a β-1,4-linked disaccharide of N-acetyl glucosamine (NAG) that is not found in mammals. It was experimentally shown that B. burgdorferi can transport chitobiose, although the genetic inactivation of the chbC gene (chitobiose PTS transporter) did not affect growth in either the tick vector or mammalian hosts, indicating that other sugars present within the tick can compensate for the loss of this transport system (Tilly, Elias et al. 2001; Tilly, Grimm et al. 2004). Alternatively, a putative beta-glucosidase, encoded by bb0620, could function to cleave chitobiose to generate NAG molecules that can then be transported through PTS transporters (Fraser, Casjens et al. 1997). After NAG is phosphorylated, it can then either be incorporated into cell wall biosynthesis or deacetylated to generate glucosamine-6-phosphate that is then converted to fructose-6-phosphate, a critical intermediate substrate in the glycolytic pathway. In addition, it has been observed that NAG can serve as an alternative carbon source and support in vitro growth of B. burgdorferi (von Lackum and Stevenson 2005).

20 | P a g e

The B. burgdorferi genome is predicted to contain an operon (bb0240 through bb0243) encoding a glycerol transporter, enzymes that convert glycerol to glycerol-3-phosphate and isomerize glycerol-3-phosphate to dihydroxyacetonephosphate. Once dihydroxyacetone phosphate is formed, it can isomerize into glyceraldehyde-3-phosphate, which then proceeds to the glycolytic pathway to generate ATP. In addition to its entry into glycolysis, glycerol-3-phosphate can be shuttled into phospholipid, lipoprotein and glycolipid biosynthesis (Fraser, Casjens et al. 1997; Saier 2000; von Lackum and Stevenson 2005).

ATP synthesis B. burgdorferi lacks genes homologous to those encoding proteins involved in oxidative phosphorylation or the TCA cycle (Fraser, Casjens et al. 1997; Das, Hegyi et al. 2000). They also do not appear to make energy via the pentose phosphate pathway, but instead use it to generate building blocks for DNA and RNA (von Lackum and Stevenson 2005). However, they do have all of the genes required for the glycolytic pathway. The major metabolic pathway for ATP generation in B. burgdorferi appears to be substrate-level phosphorylation mediated by the phosphoglycerate kinase and pyruvate kinase enzymatic steps during fermentation (Fraser, Casjens et al. 1997).

Most bacteria generate a proton gradient across their cytoplasmic membrane due to electron transport systems that result in a greater concentration of protons outside the cytoplasm. The protons then cross the concentration gradient through an F-type ATPase in a process that generates ATP within the cytoplasm of the bacterium. However, B. burgdorferi does not encode an F-type ATPase. In contrast to the majority of bacteria, it possesses a V-type ATPase which functions in reverse compared to F-type ATPases, and thus utilizes energy instead of synthesizing ATP (Fraser, Casjens et al. 1997). Most likely, the V-type ATPase in B. burgdorferi functions primarily as a H+ pump that establishes a proton motive force across the inner membrane, thereby driving some proton-coupled transport systems and B. burgdorferi motility. The exact function of this important protein complex remains to be experimentally determined.

21 | P a g e

Cell envelope of Borrelia

Borrelia membranes The borrelial outer cell membrane is fluid and consists of 45–62% protein, 23–50% lipid and 3–4% carbohydrate (Barbour, Hayes et al. 1986). Its composition differs significantly from that of Gram-negative bacteria. A first, particularly distinguishing, feature is the absence of phosphatidylethanolamine (Belisle, Brandt et al. 1994) and lipopolysaccharide (LPS) (Takayama, Rothenberg et al. 1987) and the presence of non-LPS glycolipid antigens (Eiffert, Lotter et al. 1991; Wheeler, Garcia Monco et al. 1993; Belisle, Brandt et al. 1994). These glycolipids represent about 50% of the total lipids and contain only galactose as the monosaccharide constituent (Hossain, Wellensiek et al. 2001). Secondly, the B. burgdorferi outer membrane exhibits a relatively low density of transmembrane-spanning proteins as determined by freeze fracture EM studies (Walker, Borenstein et al. 1991; Radolf 1994; Jones, Bourell et al. 1995). This may explain why Borrelia is more susceptible than Gram-negative bacteria to detergents or to disruption by routine physical manipulations such as centrifugation and resuspension. Thirdly, the outer membrane contains an unusually large number of lipoproteins (Brandt, Riley et al. 1990), many of which are on the bacterial surface, the host–pathogen interface, where they act as adhesins, targets for bactericidal antibodies, or receptors for various molecules. The phospholipid content of borrelial membranes differs from those of other bacteria, including other spirochetes, such as Treponema pallidum. Mass spectroscopic analysis identified α-monogalactosyl diacylglycerol (α-MGalDAG, 36.1%), phosphatidylcholine (PC, 11.3%), and phosphatidylglycerol (PG, 10.5%) as the major lipids in B. burgdorferi (Hossain, Wellensiek et al. 2001). Also, a cholesteryl galactoside with an immunogenic motif has been described in B. burgdorferi (Ben-Menachem, Kubler-Kielb et al. 2003; Schroder, Eckert et al. 2008). Previous studies also showed that the membranes of the relapsing fever species B. hermsii contain MGalDAG, PC, PC, cholesteryl glucoside, and an acylated cholesteryl glucoside (Livermore, Bey et al. 1978). Diglyceride-based glycolipids, like the major MGalDAG of the B. burgdorferi envelope, are most widespread in Gram-positive bacteria and mycoplasmas, but essentially absent among Gram-negative bacteria, with the exception of many photosynthetic species (including plant chloroplasts) and some Pseudomonas species (Shaw

22 | P a g e

1970). Recently, the monogalactosyl-1,2-diacylglycerol synthase from B. burgdorferi (bbMGS) was identified, cloned, and functionally verified to show that this enzyme catalyzes synthesis of the major membrane lipid MGalDAG. Interestingly, the characterized enzyme is closely related to many Gram-positive analogues (Östberg, Berg et al. 2007). Borrelia spirochetes contain an unusual abundance of proteins covalently modified by lipids, i.e., lipoproteins (Haake 2000). A recent re-evaluation using a modified prediction algorithm identified up to 127 lipoproteins in B. burgdorferi genome, corresponding to 7.8% of open reading frames (Setubal, Reis et al. 2006). This proportion is significantly higher than in other complete bacterial genomes such as Treponema pallidum (2.1%) or Helicobacter pylori (1.3%) (Fraser, Casjens et al. 1997; Casjens, Palmer et al. 2000). Surface-exposed lipoproteins also play an important role in adaptive responses and pathogenicity of Borrelia spirochetes. The most abundant and best studied outer membrane proteins of Borrelia spirochetes are lipoproteins called outer surface proteins (Osps). Osps have very little sequence or structural homology to any other known proteins. Characterization of Osps in different B. burgdorferi sensu lato strains revealed considerable heterogeneity within and among different species. The most studied Osps are OspA, OspB, OspC, OspE/Erps/CRASPS, and the Dbps. The reciprocal expression of OspA/B and OspC during Borrelia transition from its tick vector to the mammalian host represents how stage-specific protein expression can contribute to pathogenesis during the natural cycle of spirochetal transmission. In unfed ticks, Borrelia expresses large amounts of OspA and OspB and almost no OspC. The production of OspA and OspB proteins decreases in the mammalian host and instead the spirochete promotes production of OspC (Schwan, Piesman et al. 1995; Ohnishi, Piesman et al. 2001). Vaccination with a rOspA protects against subsequent B. burgdorferi infection (Wheeler, Garcia Monco et al. 1993). Despite its efficacy, it is not used as a vaccine anymore, in part because of assumed side-effects (Lathrop, Ball et al. 2002). Immunization with OspC also confers protective immunity in animal studies (Steere, Sikand et al. 1998).

General characteristics of porins In contrast to E. coli, the fluid and fragile outer membrane of Borrelia contains very low amounts of integral membrane proteins (Brandt, Riley et al. 1990; Walker, Borenstein et al. 1991; Radolf, Bourell et al. 1994). The integral outer

23 | P a g e

membrane proteins partly maintain bacterial cell structure, bind to different substrates, are involved in solute or protein transport, signal transduction, interaction with cells and fulfill a number of other tasks important for bacteria (Achouak, Heulin et al. 2001). Secretion or uptake of nutrients through the membranes is accomplished by several pore-forming proteins (Nakae 1975; Benz 1994) that can be divided into three different classes: general diffusion pores, specific channels containing a binding site for a certain solute, and actively transporting, energy-consuming channels. In recent years, numerous structures of outer membrane proteins were established by crystallography, which lead to further subdivision of these proteins into five groups (Schulz 2004). These include general porins, the most abundant species of outer membrane proteins. A good example is E. coli OmpF (Cowan, Schirmer et al. 1992), which allows the unspecific diffusion of hydrophilic molecules and may show a slight selectivity towards ions. The specific porins are structurally closely related to the general porins although their β-barrels have two more strands, as in the case of E. coli LamB (Schirmer, Keller et al. 1995). A third, distinct group consists of composed pores, where the β-barrel itself is a homo-oligomer, such as a dimer in gramicidin A (Ketchem, Hu et al. 1993), a heptamer in hemolysin (Song, Hobaugh et al. 1996) and trimer in TolC (Koronakis, Sharff et al. 2000). The group of outer membrane transport proteins has a much larger β-barrel and additional internal domains. They do not form a pore, but open and close for the passage through the outer membrane of relatively large molecules, such as iron-containing siderophores, in case of FhuA (Locher, Rees et al. 1998) and cobalamins, in the case of BtuB (Chimento, Mohanty et al. 2003). The fifth group of outer membrane proteins (poreless Omps) contains small β-barrels that are not likely to form a pore, rather they function as membrane anchors, such is OmpA in E. coli (Pautsch and Schulz 1998).

The primary sequence of the vast majority of outer membrane porins does not show any indication of α-helical structures, although it represents the typical structure for membrane proteins (Kyte and Doolittle 1982). As a general rule, the outer membrane porins are trimers formed by monomers consisting 16- or 18-stranded β-barrels, whereas the inner membrane proteins consist of mostly α-helices. The cylindrical structure implies that on average every second amino acid in the outer membrane spanning β-sheets is hydrophobic because it faces the lipid membrane or is hydrophilic and points towards the channel interior

24 | P a g e

(Vogel and Jahnig 1986). Another structural feature of porins is the inward folded loop attached to the inner side of the barrel wall, which constricts the internal diameter and stabilizes the porin β-barrel. Residues located in this loop determine the channel properties, like the diameter, ion selectivity or substrate affinity (Bauer, Struyve et al. 1989; Jordy, Andersen et al. 1996). Interestingly, these loops show very high sequence variation among homologous porins that might aid in evasion of serum antibodies, bacteriophages or proteases. Although porins show very low sequence homology, the similarities in topology and charge distribution are striking (Schirmer 1998). For instance, the C-terminal transmembrane segment contains an aromatic amino acid that is absolutely conserved in classical porins of Enterobacteriaceae (Struyve, Moons et al. 1991). Yet, this is not the case for the spirochetal porins characterized to date (Nikaido 2003).

In addition to transport of solutes, some porins can have other functions. They can insert into eukaryotic cell membranes (Rudel, Schmid et al. 1996), activate complement (Alberti, Marques et al. 1996), act as receptors for bacteriophages (Yu, Ding et al. 1998) or be targets for bactericidal compounds (Sallmann, Baveye-Descamps et al. 1999). Moreover, surface-exposed regions of porins can adhere to cellular receptors, a feature observed for Shigella flexeneri OmpC (Bernardini, Sanna et al. 1993), Treponema denticola Msp (Fenno, Muller et al. 1996) and B. burgdorferi P66 (Coburn and Cugini 2003).

Characterization of porins There are several techniques used to study function and properties of porins in vitro. Among others, reconstitution systems such as the liposome-swelling assay (Nikaido and Rosenberg 1981) and measurements of exclusion limits (Nakae and Nikaido 1975) are used. Conductance recording across a planar lipid bilayer (also called a black lipid bilayer) is the technique used the most for an initial structural and functional study of membrane channels (Benz, Janko et al. 1978; Schindler and Rosenbusch 1978). Reconstitution experiments aim to reestablish the function of purified porins, demonstrating the integrity of the pore-forming complex after the isolation process. The bilayers can be built as planar membranes or vesicles. Black lipid membranes separating two aqueous phases are particularly stable for electrical measurements (Benz and Janko 1976). The

25 | P a g e

apparatus used for membrane formation consists of a Teflon chamber with a thin wall containing a hole with a diameter of 0.2–1 mm separating the two aqueous compartments. For membrane formation, lipid solution is painted over the hole to form a thick, colored lamella. The experiments can be started after the lamella thins out and turns optically black in reflected light, because the lipid bilayer is much thinner than the wavelength of visible light (Dilger and Benz 1985). The lipid bilayer technique allows very sensitive detection of current through the membrane, which gives the opportunity to measure membrane channel conductance with high accuracy. The bilayer itself has a very low conductance below 10 pS. The membrane current is measured with an electrode pair switched in series with the voltage source and a high sensitivity current amplifier. The addition of a channel-forming protein to the aqueous phase results in a stepwise increase of membrane conductance as a consequence of protein reconstituting into the lipid bilayer membrane and can be recorded. The single channel conductance for an individual porin is usually homogenous and correlates with the size of the channel formed (Trias and Benz 1994). Artificial lipid bilayer measurements in addition to rough estimates of the channel size can provide important data on the electrostatic properties of the pore, such as ion selectivity determined upon permeability ratio of cations over anions (Schirmer 1998).

Porins in spirohetes The first spirochetal protein reported to have pore-forming activity was a 36.5-kDa major protein of the outer membrane from Spirochaeta aurantia. In its native state, it forms trimers and exhibits 7.7-nS channel-forming activity in the planar lipid bilayer assay (Kropinski, Parr et al. 1987). Oligomeric forms have been observed for several spirochetal porins (Egli, Leung et al. 1993; Shang, Exner et al. 1995; Blanco, Champion et al. 1996). The 53-kDa major surface protein (Msp) from Treponema denticola has adhesin properties and porin activity. Interestingly, it was visible as a hexagonal array on the surface of T. denticola cells by electron microscopy (Egli, Leung et al. 1993; Mathers, Leung et al. 1996). Porin activity has also been demonstrated for native, rare outer membrane protein 1 (Tromp 1) of T. pallidum, exhibiting single channel conductance of 0.7 nS (Blanco, Champion et al. 1996). Purified recombinant OmpL1 of pathogenic Leptospira was reconstituted into planar lipid bilayers and

26 | P a g e

demonstrated 1.1 nS average conductance (Shang, Exner et al. 1995). Up-regulation of OmpL1 porin expression after excretion of virulent Leptospira from the host and following culture-attenuation suggests that OmpL1 plays an important role in the adaptive response (Haake, Champion et al. 1993; Shang, Exner et al. 1995). Borrelia, due to its limited metabolic and biosynthetic capacities, is highly dependent on nutrients provided by their hosts (Fraser, Casjens et al. 1997; Casjens, Palmer et al. 2000). Consequently, these parasites need to have efficient regulation of the nutrient uptake across the cell envelope. The first indication of porins in B. burgdorferi came through investigation of channel-forming activities in the planar lipid bilayer assay of outer membrane vesicles. Two porin activities of 0.6 nS and 12.6 nS were found (Skare, Shang et al. 1995). Subsequent work has so far characterized these possible porins in B. burgdorferi: P13 and its paralog A01 (Östberg, Pinne et al. 2002; Pinne, Östberg et al. 2004; Pinne, Denker et al. 2006) and P66 (Skare, Mirzabekov et al. 1997). In this thesis additional three Borrelia pore-forming activity exhibiting proteins Oms38, DipA and BesC are described (Paper I, II and III). Another protein, Oms28, exhibits channel-forming activity but has later been shown not to be an outer membrane pore. Both its localization in the membrane and its function has been challenged. This protein, now designated BBA74, was shown to be a periplasmic outer membrane associated protein that lacks typical porin properties (Mulay, Caimano et al. 2007). In contrast to Lyme disease Borrelia species, knowledge of porin activities in relapsing fever Borrelia is rather limited. There are indications of several pore-forming activities in outer membrane preparations of relapsing fever spirochetes (Shang, Skare et al. 1998; Thein, Bunikis et al. 2008) and genes with high homology to B. burgdorferi p13, oms28 and p66 can be found in the published genomes of the relapsing fever agents B. duttonii, B. recurrentis and B. hermsii. There are also indications of uncharacterized channel-forming activities present in B. burgdorferi (Östberg, Pinne et al. 2002).

P66 P66 was the second porin identified in the outer membrane of Lyme disease B. burgdorferi with single-channel conductance measuring 10 nS (Skare, Mirzabekov et al. 1997). In recent years P66 has been well studied and identified in both Lyme disease and relapsing fever spirochetes (Bunikis, Noppa

27 | P a g e

et al. 1995; Casjens, Palmer et al. 2000; Pinne, Thein et al. 2007; Lescot, Audic et al. 2008). Its atypical huge conductance would represent a large channel that could accommodate large solutes and facilitate their transport into the bacterial cell. Peptides, long chain fatty acids, deoxynucleotides and other large molecular weight solutes have no obvious outer membrane transport systems and are proposed to diffuse across the outer membrane of B. burgdorferi. Because of the large size pore formed by P66, it seems likely that this protein may play an important role in the passive transport of some of these solutes. Additionally, using a phage display library of B. Burgdorferi, Coburn et al. were able to identify P66 as an outer membrane adhesin that binds integrin (Coburn, Chege et al. 1999). Subsequent experiments with a P66 knock-out strain showed that the loss of this porin abolished integrin binding activity (Coburn and Cugini 2003). Therefore, in addition to pore-forming activity, P66 presumably functions as an integrin binding protein. This same dual function was demonstrated for the 64-kDa and 53-kDa (Msp) proteins from T. denticola. Both proteins have porin activity while the 64-kDa protein is also involved in binding of T. denticola cells to gingival fibroblasts in addition to the ability to bind various extracellular matrix proteins (Fenno, Muller et al. 1996; Fenno, Wong et al. 1997). P66 contains surface-exposed domains (Bunikis, Noppa et al. 1995; Bunikis, Noppa et al. 1996) which exhibit immunogenic potential (Barbour, Jasinskas et al. 2008). An additional function of P66 might be interaction with surface-localized lipoproteins. Osp proteins, in particular OspA, have been shown to limit the access of both antibodies and trypsin to the major surface loop of P66. Thus, the possible close contact between P66 and OspA, and other Osp proteins may hinder accessibility to the P66 protein (Bunikis and Barbour 1999). Further characterization of biophysical properties of P66 and its homologues from Lyme disease and relapsing fever borreliae is described in Paper IV. The atypical huge single-channel conductance of this interesting porin raised a question: Is the surprisingly high conductance due to a similarly outstanding huge channel diameter? The attempts to answer this question are described in Paper V. Even though knowledge about transport across the outer membrane of Borrelia is increasing, a great deal still remains to be determined.

28 | P a g e

Efflux through channel-tunnels

Export systems In addition to the inner membrane, exported substances like proteins or drugs have to pass the periplasmic space and have to overcome the second permeability barrier, the outer membrane. In the case of protein export out of the cell, one can distinguish roughly between sec-dependent and sec-independent export pathways. In the sec-dependent export pathways, proteins with a cleavable N-terminal secretion signal are directed to the sec system in the inner membrane, which transports them into the periplasmic space (Fekkes and Driessen 1999; Manting and Driessen 2000). This route is also known as the general secretion pathway (GSP). A database search showed that homologues of all essential components of the sec translocase complex are present in Borrelia (Fraser, Casjens et al. 1997). After or during translocation, the secretion signal is cleaved, resulting in processed periplasmic intermediates. For further export across the outer membrane there are six different mechanisms: the chaperone/usher pathway, the single accessory pathway, the autotransporter, the type II secretion system, the type IV pilus biogenesis system and the type IV secretion system (Henderson, Navarro-Garcia et al. 1998; Konninger, Hobbie et al. 1999; Nunn 1999; Cao and Saier 2001; Sandkvist 2001; Thanassi, Stathopoulos et al. 2002).

There are two distinct sec-independent secretion machineries. Both systems form assemblies which span the whole cell envelope and export proteins without a periplasmic intermediate. The type III secretion system forms the injection needle of pathogens and the flagellar export apparatus. These are complex assemblies formed by up to 20 proteins (Hueck 1998; Young, Schmiel et al. 1999). The other sec-independent export apparatus is the type I secretion system.

The export of drugs and cations is mediated by several distinct transporters in the inner membrane. They can be divided into five families (Paulsen, Chen et al. 2001; Saier and Paulsen 2001). One group belongs to the ancient superfamily of ATP-binding cassette (ABC) transporters using ATP as an energy source. The other four families are energized by proton motive force and act as proton antiporters. They belong to the major facilitator superfamily (MFS), the small

29 | P a g e

multidrug resistance (SMR) family, the multidrug and toxin extrusion (MATE) family and the resistance nodulation cell division (RND) family (Pao, Paulsen et al. 1998; Brown, Paulsen et al. 1999; Tseng, Gratwick et al. 1999; Chung, Krueger et al. 2001). Without any accessory proteins, these inner membrane transporters would not be able to export their substrates across the outer membrane. For three of the four transporter families, it is known that members interact with two other proteins and form so-called efflux pumps in order to expel drugs or cations out of the cell. A putative drug efflux system of B. burgdorferi is described in Paper I of this thesis.

Efflux pumps Efflux pumps, as well as the type I secretion system, engage an outer membrane protein of the channel-tunnel family to export substrates out of the cell. In addition to this outer membrane protein, the export machineries harbor a third protein called an accessory protein, also known as the membrane fusion protein (MFP).

Efflux pumps mediate resistance against diverse noxious substances, such as antibiotics, dyes, detergents, bile salts or heavy metals. They may, for example, raise antibiotic resistance by several orders of magnitude, rendering antibiotics clinically useless (Poole 2001). The involvement of an outer membrane channel-tunnel makes it possible to export the substance out of the cell into the medium. In order to enter the cell again, the expelled molecules must cross the outer membrane. Thus, the synergistic interplay between the efflux pump and the low permeability of the outer membrane is highly advantageous for Gram-negative bacteria. Like the type I secretion system, efflux pumps are tripartite export systems consisting of an inner membrane transporter, a periplasmic accessory protein and an outer membrane channel-tunnel.

Members of three distinct transporter families can be part of channel-tunnel-dependent efflux pumps. Inner membrane transporters belong either to the RND superfamily, to the MF superfamily or to the superfamily of ABC transporters. Genomic analysis revealed the existence of a fourth transporter family, the putative extrusion transporters (PET), which might also act in cooperation with an accessory and a channel-tunnel protein.

30 | P a g e

RND superfamily The RND transporter superfamily is a ubiquitous family with representation in all major kingdoms. It is grouped in seven families; two of them are involved in channel-tunnel-dependent efflux (Tseng, Gratwick et al. 1999). Dependent on the exported substrate, one distinguishes between the heavy metal efflux transporters (HME) and the hydrophobic and amphiphilic compounds efflux transporters (HAE).

The first solved structure of a transporter that is driven by proton motive force was AcrB of E. coli (Murakami, Nakashima et al. 2002). This transporter belongs to the HAE family and is part of the well-studied AcrAB/TolC multidrug efflux pump of E. coli. The amino acid sequence of AcrB can be divided into two halves with homologous regions implying that RND transporters arose as a result of a tandem, intragenic duplication event. The duplication is also reflected in the tertiary structure of the transporter, which is composed of two halves with similar architecture. The crystal structure revealed that AcrB functions as a homotrimer. The jellyfish-shaped structure is divided into a transmembrane domain, a pore domain and a TolC-docking domain. The subsequent crystallization of AcrB in a complex with four different ligands (Yu, McDermott et al. 2003) gave further insight into the mechanism of multidrug resistance efflux pumps.

Accessory proteins The accessory proteins involved in efflux pumps are grouped into several families (Saier 2000). Their classification correlates with the different transporter families with which they interact. Accessory proteins are characterized by two so-called lipoyl domains. The domains are approximately 30 residues long and form β-strands. These lipoyl domain motifs are found in all members of different accessory proteins (Dinh, Paulsen et al. 1994; Johnson and Church 1999). The structure of these domains is known from enzymes that transfer covalently attached lipoyl or biotinyl moieties between enzymatic components (Dardel, Davis et al. 1993; Athappilly and Hendrickson 1995; Green, Laue et al. 1995; Neuwald, Liu et al. 1997). The two motifs are connected by a three-residue-long linker in these proteins. In the case of the accessory proteins, this linker is much longer. The size of the linker is 41–65 (accessory proteins of

31 | P a g e

RND transporters), 79 (ABC transporters) and 111–121 residues (MF transporters), respectively. The secondary prediction for this linker region reveals that all different accessory protein families have two domains with a high coiled-coil probability separated by a gap of five to ten residues. It is predicted that that the coiled-coil domains form an α-helical hairpin. Accessory proteins possess a membrane anchor at the N-terminal end. This is either a transmembrane helix or a covalently bound fatty acid serving as a lipid anchor. This is the case for accessory proteins interacting with RND transporters. Studies revealed that this anchor is not necessary for the function of accessory proteins and can be removed experimentally without affecting functionality of the efflux pump (Zgurskaya and Nikaido 1999; Yoneyama, Maseda et al. 2000). For the RND transporter AcrB and the corresponding accessory protein AcrA, it could be shown by cross-linking that both proteins form a complex. The specificity of interaction between the accessory protein and the transporter lies in the periplasmic loops, as shown by chimeric RND transporters (Tikhonova, Wang et al. 2002). It has been experimentally proven that AcrA assembles into trimers as does its partner, AcrB.

Channel-tunnels According to the classification, the channel-tunnel family is named the outer membrane factor (OMF). Channel-tunnels are found in almost all Gram-negative bacteria (Sharff, Fanutti et al. 2001). This confirms the importance of this outer membrane protein for the cells. The channel-tunnel family is further divided into subfamilies based on the members’ different export functions: protein secretion, drug efflux and cation efflux. Prototypes of the three subfamilies are TolC of E. coli, OprM of P. aeruginosa and CzcC of Ralstonia eutrophus, respectively (Wandersman and Delepelaire 1990; Gotoh, Tsujimoto et al. 1995; Rensing, Pribyl et al. 1997). TolC is not only involved in protein secretion, but is also part of a diverse efflux pumps. This makes it an interesting and important protein due to its ability to interact with several distinct inner membrane complexes.

32 | P a g e

Structure and characteristics of TolC TolC forms homotrimers, which are cannon-shaped with a long axis measuring 100 Å (Koronakis, Sharff et al. 2000). One end of the cylinder is open, with an inner diameter of about 30 Å. Towards the other end, it becomes smaller, leading to an almost closed cylinder. The structure can be divided into three domains: the 40-Å-long channel domain, the 100-Å-long tunnel domain and the equatorial domain. The channel domain and the tunnel domain form a continuous tube, which is the origin of the name of this family: channel-tunnels. It is evident that TolC comprises a structural repeat, which was already recognized in the primary structure, before the protein was crystallized (Gross 1995; Johnson and Church 1999). It is most likely that the family of channel-tunnels evolved by gene duplication from a common ancestor. The overall length of TolC homologues varies, but mostly due to extensions at the N- and C-termini. Gaps or insertions exist only in the extracellular loops or equatorial domain. The length of the tunnel-forming helices is constant. It is clearly determined because residues that are important at the transitions between the different parts of the structure are highly conserved.

The channel domain anchors TolC in the outer membrane. It consists of 12 β-strands which are arranged in an antiparallel way to form a barrel. The β-barrel is a common feature of all outer membrane proteins crystallized so far (Buchanan 1999; Koebnik, Locher et al. 2000). Nevertheless, the architecture of β-barrel of channel-tunnels is different; β-barrels of all other known protein structures are formed by a single peptide chain. In the case of channel-tunnels, three protomers are necessary to form a single barrel, each contributing four β-strands. Despite similarities with porins, the interior of the TolC β-barrel is different from the interior of porins. TolC lacks the structural element of porins — an inward folded loop that constricts the internal diameter of the porin β-barrel. Residues located at this loop determine the channel properties, like diameter, ion selectivity or substrate affinity (Bauer, Struyve et al. 1989; Jordy, Andersen et al. 1996; Saint, Lou et al. 1996).

More than 80% of the TolC peptide chain is located in the periplasmic space forming the equatorial and tunnel domain. This 100-Å periplasmic extension is the most striking difference compared to all other outer membrane proteins known (Koronakis, Andersen et al. 2001). The tunnel domain consists entirely of α-helices. Above the equatorial domain the helices assemble into an α-barrel.

33 | P a g e

This is a structure that is not found in any other crystallized protein. The α-barrel is left-twisted, different from the right-twisted β-barrel of the channel domain. The opposite twist might mutually stabilize the barrels so that the structures do not collapse. Below the equatorial domain, the continuous helices pair together and form conventional coiled-coils. Seen from the periplasmic side, the coiled-coils arrange into a tight structure, almost sealing the tunnel. The diameter of the periplasmic entrance is 4 Å, which is too narrow for exported proteins or drugs to pass through. Therefore, the tunnel entrance has to open. An iris-like rearrangement of the inner coiled-coil is predicted to open the periplasmic entrance (Koronakis, Sharff et al. 2000). The circular network of residues at the tunnel entrance that keeps the tunnel in closed confirmation has been identified by Andersen and co-workers (Andersen, Koronakis et al. 2002). Furthermore, disruption of this network results in widening of the tunnel entrance, subsequently increasing the conductance 10-fold. This indicates most likely that the tunnel entrance of TolC is open, and it is estimated that the single-channel conductance corresponds to a diameter of around 16 Å, wide enough to allow export.

It is known that the tripartite export systems are not permanent assemblies. The transporter and the accessory protein form a complex, but interaction between the complex and the channel-tunnel is dynamic (Thanabalu, Koronakis et al. 1998; Zgurskaya and Nikaido 2000). Therefore, the predicted role of the accessory protein is recruitment and interaction with the channel-tunnel. This function is most likely mediated by the coiled-coil domains of the accessory proteins. The contact between the channel-tunnel and the inner membrane complex could function as a trigger for opening the tunnel entrance.

The ability of TolC to form pores was first described by Benz et al. (Benz, Maier et al. 1993). Stable pores in a planar lipid bilayer exhibited a single channel conductance of 85 pS in 1 M KCl. This is about 20-fold smaller than that of general diffusion pores like OmpF of E. coli (Bauer, Struyve et al. 1989). It is already known from research on porins that the residues lining the narrowest part of the channel have a strong influence on the electrophysiological properties (Bauer, Struyve et al. 1989; Jordy, Andersen et al. 1996; Saint, Lou et al. 1996). The narrowest part of TolC channel-tunnel is the periplasmic entrance. It is lined by six aspartate residues (two per monomer), establishing a highly electronegative area. The aspartate residues determine ion selectivity and

34 | P a g e

render TolC highly cation selective. In most channel-tunnel proteins, at least one of the two negatively charged residues is conserved. This underlines the importance of these residues. Cation selectivity is common for most pore-forming proteins of the outer membrane of E. coli (Benz 1994). It might be a general feature to protect cells from uptake of negatively charged, harmful substances like bile salts found in the environment of these bacteria. The conservation of these negative charges within the channel-tunnel family provides a possible target for new drugs aimed at the inhibition of efflux pumps.

35 | P a g e

Aims of the thesis

This thesis was initiated with the general aim to investigate novel, uncharacterized proteins in the outer membrane of Borrelia. Such proteins might be of great importance for pathogenicity and survival of spirochetes in different hosts during their complex life cycle. In addition, the aim was to improve knowledge about already established Borrelia porins. Emphasis was placed on studying the molecular, biochemical and structural properties of channel-forming proteins BesC, Oms38/DipA and P66.

Specific aims are summarized below.

• To purify BesC from outer membrane preparations of Borrelia burgdorferi

• To characterize the biophysical properties of BesC

• To determine the function of BesC

• To investigate role of BesABC efflux complex in vivo

• To isolate and characterize a novel porin in relapsing fever and Lyme disease Borrelia – Oms38/DipA

• To demonstrate the substrate specificity of DipA

• To determine the properties of P66 in different Borrelia species

• To investigate the channel size of P66 using nonelectrolytes

36 | P a g e

Results and discussion

Paper I.

In this study, using molecular, biophysical methods and bioinformatics, we characterize a putative multidrug efflux system in Borrelia burgdorferi, comprising the components BesA, BesB and BesC. Furthermore, we demonstrate that the BesC protein, a TolC homolog, is a novel virulence factor of B. burgdorferi.

Identification of efflux system in a Borrelia Comprehensive database search revealed that the B. burgdorferi B31 genome harbors a putative member of an outer membrane factor family hypothetical protein designated BB0142 (Yen, Peabody et al. 2002). The amino acid sequence of this protein revealed significant similarity (E value = 5-22) to proteins of the outer membrane efflux protein (OEP) family, including the E. coli outer membrane protein TolC (Marchler-Bauer and Bryant 2004). Investigation of the genes directly flanking bb0142 showed that bb0140 codes for an AcrB-like protein, an inner membrane transporter of the RND (resistance, nodulation, cell-division) family, and bb0141 codes for an AcrA-like protein, an adaptor protein also known as membrane fusion protein. Homologs of these proteins form tripartite multidrug efflux pumps in many Gram-negative bacteria. Therefore, we proposed renaming these genes besA (bb0141), besB (bb0140) and besC (bb0142) for Borrelia efflux system proteins A, B and C. The besABC system is the only multidrug efflux pump identified thus far in the B. burgdorferi genome. According to the genome sequence, the genes besB, besA and besC are oriented in the same direction, and separated by 18 and 8 bp, respectively (Fraser, Casjens et al. 1997). We analyzed RNA from low-passage strain B31 by RT-PCR and demonstrated that besB, besA and besC are transcribed as a single transcript.

37 | P a g e

Involvement of BesC in antibiotic resistance and virulence In order to investigate the potential involvement of BesC in antibiotic resistance and virulence of B. burgdorferi, we constructed a besC mutant. It has been shown previously that the outer membrane components are essential for functionality of multidrug efflux pumps (Sulavik, Houseweart et al. 2001). Therefore, susceptibility of besC::str mutant strain to different antimicrobials was tested in vitro. β-lactam antibiotics, macrolides, and tetracyclines are generally recommended for stage-dependent therapy of Lyme borreliosis (Mursic, Wilske et al. 1987). The mechanisms and actual proteins involved in Borrelia resistance to different antimicrobial agents have not been elucidated to date. Therefore, our finding of increased Borrelia susceptibility to several antibiotics due to inactivation of BesC, the homolog of E. coli TolC, is the first proof that Borrelia possesses an active efflux system that might be responsible for multidrug resistance. Strategies to identify efflux pump inhibitors are in progress (Lomovskaya and Watkins 2001). For these reasons, the identification and the characterization of such an efflux system in B. burgdorferi is very important for understanding the pathogenicity of this organism.

Several studies have shown that lack of efflux-pump expression by a Gram-negative bacterium has a deleterious effect on the ability of the bacterium to be pathogenic in animal models (Jerse, Sharma et al. 2003; Nishino, Latifi et al. 2006). Some bacterial efflux pumps export not only antibiotics and other substances, but also host-derived antimicrobial agents (Lee and Shafer 1999). This finding has led to the suggestion that the physiological role of these systems is evasion of such naturally produced molecules, thereby allowing the bacterium to survive in its ecological niche (Lacroix, Cloeckaert et al. 1996). To investigate the influence of besC inactivation on B. burgdorferi virulence, we performed animal infectivity experiments. Borrelia was cultured from organs of infected mice. In contrast to cultures from positive control mice containing vigorously growing bacteria, cultures from mice infected with the besC mutant exhibited no bacterial growth at all, suggesting that BesC is important for Borrelia to establish infection in mice.

38 | P a g e

Channel-forming properties of purified BesC We separated the outer membrane proteins from a ∆p66 knock-out strain by anionic exchange chromatography. We investigated channel-forming activities of different fractions in the black lipid bilayer assay. In one of the fractions containing uniform activity, we identified by mass spectrometry using peptide mass fingerprints (Sickmann, Mreyen et al. 2002) the protein BesC as the sole protein with a putative pore-forming function. To analyze the channels formed by BesC, we performed single-channel experiments with purified protein. Results showed that this protein forms obvious channels with an average single-channel conductance of 300 pS in 1 M KCl. In contrast, E. coli TolC forms a channel with a single-channel conductance of 80 pS in 1 M KCl, which is almost four-fold smaller than that for the channel of BesC (Benz, Maier et al. 1993; Koronakis, Sharff et al. 2000; Koronakis 2003; Andersen 2004). TolC of E. coli contains a circular network of inter- and intra-molecular connections between the three monomers that is responsible for the stability of the almost closed state of the channel-tunnel (Andersen, Koronakis et al. 2002). This network involves residues located at the inner and outer coiled coil of the tunnel domain, which are not conserved in BesC. This suggests that other interactions keep the helices of the tunnel entrance together; otherwise the single-channel conductance would be much higher. The modeled BesC structure shows that the two charged residues, Asp363 and Lys366, of different monomers are in close proximity, allowing formation of intermolecular salt bridges. Thus, most likely they are able to establish a similar circular network keeping the helices at the tunnel entrance in a closed conformation.

We performed zero-current membrane potential experiments to analyze the ion selectivity. After insertion of 100 to 1000 channels into the phosphatidylcholine membrane, the salt concentration on one side of the membrane was raised from 100 to 500 mM by addition of 3 M salt solution. The zero-current membrane potential was close to zero for KCl, which means that the ion permeability through the BesC pore follows the aqueous mobility of the ions and indicates that BesC forms a non-selective channel.

39 | P a g e

A structural model of BesC To further analyze the electrophysiological properties of BesC in comparison with the best characterized channel tunnel TolC of E. coli, we have modeled the structure of BesC. A special focus was put on residues lining the periplasmic entrance, which are known to have a major influence on the electrophysiological properties of channel-tunnels (Polleichtner and Andersen 2006). In comparison to TolC of E. coli, which forms pores of 80 pS in 1 M KCl with a high preference for cations due to six aspartate residues (Asp371 and Asp374) lining the tunnel entrance, the pores formed by BesC are almost non-selective. Looking at the modeled BesC tunnel entrance, it becomes apparent that the opening is slightly wider than that of TolC, explaining the higher single channel conductance. Furthermore, the charges lining the channel entrance are balanced. There are two oppositely charged residues per monomer, Asp363 and Lys366, which could explain BesC being non-selective for ions (Figure 2).

Figure 2. Comparison of channel entrance of E. coli TolC and B. burgdorferi BesC. Arrows indicate important amino acids, that define selectivity of both channels.

40 | P a g e

BesA is an atypical adaptor protein BesC needs to interact with the putative inner membrane complex formed by the AcrB homologue BesB, which serves as an inner membrane transporter, and the AcrA homologue BesA, a periplasmic adaptor protein. These three components are necessary to form a transport system spanning the two membranes for pumping out noxious compounds from the cytoplasm (Sulavik, Houseweart et al. 2001; Andersen 2003). A special characteristic of the Borrelia adaptor protein is the absence of the hairpin domain. Sequence alignment revealed that the loop, which connects the two halves of the highly conserved lipoyl domain, is just nine residues long instead of the 61 or 75 residues found in the proteins MexA of P. aeruginosa or AcrA of E. coli, respectively. This means that the long coiled-coil domain, which is suggested to stabilize the contact to the channel-tunnel in the assembled complex of AcrABTolC or MexABOprM, is missing (Stegmeier, Polleichtner et al. 2006; Lobedanz, Bokma et al. 2007). In the case of the Borrelia efflux pump, the modeled structures of the three Borrelia components suggest that the contact zone is restricted to the head-to-tail contact between the BesC channel-tunnel and the BesB/BesA complex. Recent biochemical data of the E. coli efflux pump provided evidence that loops of the upper rim of the inner membrane transporter AcrB are in contact with the channel tunnel and might trigger and stabilize the tunnel opening (Tamura, Murakami et al. 2005; Stegmeier, Polleichtner et al. 2006; Lobedanz, Bokma et al. 2007). We assume that in the case of the Borrelia efflux pump, this interaction is sufficient to trigger and stabilize opening of BesC. The already wider tunnel entrance of BesC requires less conformational changes and less interaction for the transition into the open state. From an evolutionary point of view, the Borrelia efflux pump presents an archetypical form of this apparatus because the adaptor protein lacking the coiled-coil domain resembles most the potential progenitor, the lipoyl-domain of acetyl transferase proteins (Johnson and Church 1999). This is consistent with the fact that Spirochetes present a very early branch in the evolution of bacteria (Gupta 2000).

41 | P a g e

Paper II.

This study describes the identification and characterization of the first relapsing fever porin, designated Oms38, which is present in the outer membranes of B. duttonii, B. hermsii, B. recurrentis, and B. turicatae.

Purification and identification of a 38-kDa, pore-forming protein Borreliae are limited in their metabolic and biosynthetic capacities and therefore are highly dependent on nutrients provided by their hosts (Fraser, Casjens et al. 1997). Relapsing fever Borrelia, in comparison with Lyme disease spirochetes, grows rapidly, reaching high cell densities in the blood in only a few days. To maintain this rapid growth, the bacterium needs to have efficient nutrient uptake from the plasma. The transport of nutrients and other molecules across the outer membrane is performed by pore-forming proteins, so-called porins. Intriguingly, although there are indications of their presence (Shang, Skare et al. 1998), no relapsing fever porin has been described in the

literature to date. The analysis of relapsing fever porins is of high importance not only for understanding the mechanisms of nutrient uptake, but because it could lead to identification of surface-exposed, immunogenic proteins, as putative candidates for vaccine development.

Outer membrane fractions of three relapsing fever Borrelia representatives, B. duttonii, B. hermsii, B. recurrentis were subjected to hydroxyapatite chromatography. Pore formation was found exclusively in fractions exhibiting a band that corresponded to a molecular mass of 38 kDa. To identify the gene encoding the 38-kDa protein, we performed partial amino acid sequencing of the protein fractions that contained only the corresponding protein band. N-terminal sequencing provided the first 19 N-terminal amino acids of the native 38-kDa protein of B. duttonii. Searching the genome sequences relapsing fever spirochetes B. duttonii revealed the gene coding for the 38-kDa protein and its complete amino acid sequence. According to its molecular mass, the protein was designated Oms38, for outer membrane spanning protein of 38 kDa.

42 | P a g e

Analysis of the amino acid sequences of Oms38 Partial sequencing of the purified Oms38 of B. duttonii allowed the identification of its gene within the genomes of the three other relapsing fever species, B. hermsii, B. recurrentis, and B. turicatae. The deduced amino acid sequences of Oms38 from the four relapsing fever species showed an overall identity of 58%. B. duttonii- and B. recurrentis-derived Oms38 shared the highest amino acid identity, 98%, whereas B. hermsii and B. turicatae Oms38 were 83% identical. The main differences were located in the N terminus, while the remaining parts of the sequences were highly similar. The N terminus of Oms38 of B. duttonii starts with three glutamic acids, as determined by N-terminal sequencing. This means that this protein contains a 19-amino-acid, N-terminal signal peptide with a putative recognition sequence for the leader peptidase, similar to those of other spirochetal outer membrane proteins (Cullen, Haake et al. 2004). Computational analyses (Bagos, Liakopoulos et al. 2004; Gromiha, Ahmad et al. 2004) revealed putative β-sheet stretches in the secondary structure of Oms38. This suggests that the protein may form a β-barrel in the outer membrane, like

most Gram-negative bacterial porins (Benz 1994; Saier 2000; Delcour 2002; Charbit 2003).

Biophysical properties of Oms38 The pores formed by Oms38 were studied in more detail by adding small amounts of Oms38 to black lipid bilayer membranes. The results revealed that all the Oms38 porins form channels with a low single-channel conductance of 80 pS in 1 M KCl. This is considerably lower than the conductance of all porins that were identified to date in Lyme disease-causing B. burgdorferi. The data demonstrate that the Oms38 porins from B. duttonii, B. hermsii, and B. recurrentis consistently exhibited the same average single-channel conductance of 80 pS in 1 M KCl, which is not surprising given the high sequence similarity of Oms38 proteins. A single-channel conductance of Oms38 pores is within the range of those of certain outer membrane porins of E. coli that have a lower conductance, such as the nucleotide-specific Tsx (10 pS) (Maier, Bremer et al. 1988) and the sugar-specific LamB (160 pS) (Benz, Schmid et al. 1986).

43 | P a g e

To obtain further information on the properties of the pores formed by Oms38, single-channel experiments were also performed with electrolytes other than KCl. Experiments demonstrated that the conductance of the Oms38 channels was dependent on the aqueous mobility of the ions. Replacement of chloride with the large and less mobile acetate resulted in a substantial influence on the single-channel conductance (30 pS for B. duttonii, 65 pS for B. hermsii, and 50 pS for B. recurrentis in 1 M KCH3COO, pH 7), whereas the replacement of potassium by lithium led to a lower influence on the conductance (50 pS, 75 pS, and 60 pS, respectively, in 1 M LiCl), suggesting selectivity for anions. We further investigated the selectivity of B. duttonii-derived Oms38 by multichannel

experiments under zero-current conditions in KCl, LiCl, and KCH3COO. Selectivity measurements allow quantification of the permeability of the Oms38 pore for anions relative to cations. For all three salts employed in this study, the more dilute side of the membrane became negative due to the preferential movement of anions over cations. This indicates preferential diffusion of anions through the pore.

Oms38 is the first identified, purified, and characterized putative porin in the outer membrane of the relapsing fever species B. duttonii, B. hermsii, and B. recurrentis. Although, detailed structural and functional analyses of Oms38 remain to be done in the future, this work contributes to the knowledge of proteins present in the outer membrane of relapsing fever Borrelia and their importance for physiology of these strictly host-dependent, pathogenic spirochetes.

44 | P a g e

Paper III.

This study describes identification and biophysical characterization of a dicarboxylate-specific porin in the outer membrane of B. burgdorferi, a homologue of the Oms38 porin of relapsing fever. It is the first identified solute-specific porin in Borrelia and is designated as DipA, for dicarboxylate-specific porin A.

New porin in B. burgdorferi outer membrane A few outer membrane-spanning proteins of B. burgdorferi have been characterized so far; four exhibit pore-forming activity. These are P13; Oms28, whose possible function as a porin has recently been questioned (Mulay, Caimano et al. 2007); P66 (Skare, Champion et al. 1996; Skare, Mirzabekov et al. 1997; Nilsson, Cooper et al. 2002; Östberg, Pinne et al. 2002); and BesC, part of a channel-tunnel efflux systems, identified in the study described in Paper I of this thesis (Bunikis, Denker et al. 2008). All these proteins form pores with a rather high conductance in black lipid bilayer. First identified porin of relapsing fever spirochetes, Oms38, described in Paper II of this thesis, has a single-channel conductance in the pS-range (Thein, Bunikis et al. 2008). Searching the genome of B. burgdorferi revealed a gene homologous to oms38 called “hypothetical protein bb0418” with unknown function. These findings indicated that the outer membrane fraction of B. burgdorferi could possibly contain further pore-forming components with lower conductance and have properties similar to those of Oms38 of relapsing fever Borrelia.

The outer membrane fraction of B. burgdorferi B31 ∆p66 was subjected to hydroxyapatite chromatography. One of the fractions showed high channel-forming activity of 50 pS in 1 M KCl, which differed clearly from the pore-forming activities of P13, Oms28, P66 and BesC. The 36-kDa protein present in the fraction was defined as the sole pore-forming component responsible for the formation of the 50 pS pores. To identify the gene coding for this 36-kDa protein, a silver-stained protein band was analyzed by mass spectrometry and identified by peptide mass fingerprinting. The gene of the protein band was

45 | P a g e

found to be a “hypothetical protein bb0418” of B. burgdorferi B31, now designated as dipA, for dicarboxylate-specific porin A.

Presence of DipA in other Borrelia species and homology to Oms38 Searching the published genomes of B. garinii PBi and B. afzelii PKo revealed proteins homologous to DipA in these closely related Lyme disease agents. The proteins share an amino acid sequence identity of 88%, demonstrating that the proteins are highly conserved. This is comparable to the high homology of other porins in these species (Bunikis, Noppa et al. 1995; Noppa, Östberg et al. 2001). The polyclonal antibodies raised against a recombinant C-terminal part of B. burgdorferi DipA identified the presence of homologues proteins in the Lyme disease species B. burgdorferi, B. afzelii and B. garinii and the relapsing fever species B. crocidurae, B. duttonii, B. hermsii, B. hispanica and B. recurrentis. We assume that the structures and functions of the DipA homologues are similar to DipA under in vivo conditions.

The prediction shows that the B. burgdorferi DipA signal peptide is 20 amino acids long and contains positive charges at the N-terminus, properties that are typical for borrelial signal peptides (Cullen, Haake et al. 2004). The predicted N-terminal cleavage sites agree with those of Oms38 derived from N-terminal amino acid sequencing (Thein, Bunikis et al. 2008). Further predictions indicated that B. garinii and B. afzelii DipA contain similar N-terminal extensions. Additional computational analyses predicted putative β-sheet regions in the primary sequences of B. burgdorferi, B. garinii and B. afzelii DipA. The secondary structure predictions supported the idea that the proteins may form a β-barrel cylinder consisting of about 14 β-sheets, similar to most typical Gram-negative bacterial porins, which contain 16 or 18 β-sheets (Benz 1994; Saier 2000; Delcour 2002; Charbit 2003).

DipA exhibits specificity for dicarboxylates The growth of Borrelia depends strictly on nutrients provided by their hosts. The organism has fastidious growth requirements of serum-supplemented mammalian tissue-culture medium for in vitro cultivation (Barbour 1984). In

46 | P a g e

addition, it is known that B. burgdorferi lacks genes coding for proteins of the tricarboxylic acid cycle or oxidative phosphorylation and for de novo synthesis of amino acids and nucleotides (Fraser, Casjens et al. 1997). This implicates that essential compounds or precursor of these compounds have to be imported into the cell. The small single-channel conductance of the DipA pore is comparable to the those of the substrate-specific E. coli channels Tsx (10 pS) (Maier, Bremer et al. 1988) and LamB (160 pS) (Benz, Schmid et al. 1986) under identical conditions. This led to hypothesis that DipA perhaps could be a channel specific for essential nutrients and could contain a binding site for them, similar to the carbohydrate-specific E. coli channel LamB (Ferenci, Schwentorat et al. 1980; Benz, Schmid et al. 1986).

This hypothesis was tested and the substrate-specificity could be demonstrated by multi-channel experiments. High affinity of DipA for dicarboxylates was revealed in titration experiments, performed with a variety of dicarboxylates and other related organic anions with high biological relevance. Substances blocked the ion current through DipA with a maximum block of channel conductance ranging from 20% for pyruvate to 31% for oxaloacetate. In analogy to other bacterial specific porins, it is likely that the DipA binding site, with its high affinity for dicarboxylic anions, increases the permeability of the channel for these metabolites. The presence of a binding site leads to accelerated transport of carbohydrate through LamB and of phosphate through OprP, especially at very low substrate concentrations (Benz, Schmid et al. 1987; Benz 1994). Thus, the permeability of a substrate-specific porin can surpass that of a general diffusion pore by orders of magnitude in spite of its smaller cross-section (Benz, Schmid et al. 1987).

Dicarboxylates, such as malate, succinate, oxaloacetate and 2-oxoglutarate, are major intermediates of the tricarboxylic acid cycle and mainly used for synthesis of amino acids. For example, oxaloacetate and 2-oxoglutarate are important substrates for the biosynthesis of asparagine, aspartic acid and glutamic acid, respectively, which are essential protein building blocks. In addition, C4-dicarboxylates other than succinate can be metabolized due to the lack of a functional tricarboxylic acid cycle in anaerobic energy metabolism of most bacteria (Janausch, Zientz et al. 2002). Taking these points into consideration, a potential dependence of the growth of Borrelia on this group of chemicals is likely. Consequently, DipA plays an important role in the uptake of

47 | P a g e

dicarboxylates across the outer membrane. It is noteworthy that DipA is not the first identified membrane channel specific for dicarboxylates. Previous studies revealed that the channel of spinach leaf peroxisomes is also specific for this class of chemicals (Reumann, Maier et al. 1998). Interestingly, in bacteria, the PorB porin of Chlamydia trachomatis is the first identified pore-forming outer membrane protein specific for dicarboxylates. It plays a crucial role in feeding the chlamydial tricarboxylic acid cycle and providing carbon and energy production intermediates (Kubo and Stephens 2001).

It is interesting to note that attempts to inactivate the dipA gene in B. burgdorferi so far have been unsuccessful (data not shown), which further suggests the importance of this protein for Borrelia. DipA is the first identified solute-specific porin in Borrelia. It contains at least one binding site with a high affinity for dicarboxylic anions and related compounds, and is therefore suggested to play a major role in the metabolic pathway for uptake of these nutrients.

48 | P a g e

Paper IV.

In this study, we investigated the presence of P66 homologues in both Lyme disease and relapsing fever Borrelia species and analyzed biophysical properties of these proteins in the black lipid bilayer assay.

The high interspecies homology of P66 In contrast to Lyme disease B. burgdorferi, knowledge about relapsing fever porins is very limited. To date, only one porin has been identified in relapsing fever spirochetes, described in Paper II of this thesis (Thein, Bunikis et al. 2008). Searching the available genome sequences of Lyme disease and relapsing fever Borrelia revealed the presence of P66 homologues in all species used in this study. P66 is a well-studied outer membrane protein present in Lyme disease Borrelia (Bunikis, Noppa et al. 1995; Skare, Mirzabekov et al. 1997; Coburn, Chege et al. 1999; Defoe and Coburn 2001; Coburn and Cugini 2003; Pinne, Thein et al. 2007). It was shown that B. burgdorferi P66 forms pores in artificial membranes with an extremely high single channel conductance of almost 10 nS in 1 M KCl (Skare, Mirzabekov et al. 1997). Furthermore, P66 acts as an adhesin that can bind to β3-integrin (Coburn, Chege et al. 1999; Defoe and Coburn 2001; Coburn and Cugini 2003), has surface-exposed domains (Bunikis, Noppa et al. 1995; Bunikis, Noppa et al. 1996) and exhibits immunogenic potential (Barbour, Jasinskas et al. 2008). This makes P66 of great interest in the search for vaccination candidates against Borrelia.

Comparison of the amino acid sequences of P66 from the Lyme disease B. burgdorferi, B. afzelii, B. garinii and relapsing fever B. duttonii, B. recurrentis, B. hermsii revealed 41% inter-species identity with highly conserved domains. Whereas P66 of the three Lyme disease species share a sequence identity of 90%, P66 proteins of the three relapsing fever strains were identical by 67% of their amino acid sequence. The P66 sequence identity of B. duttonii and B. recurrentis was highest, reaching an overall identity of 98%. The highly conserved domains present within all P66 species studied might be important structural determinants responsible for the channel structure and formation of 20 or more transmembrane β-sheets as secondary structure prediction

49 | P a g e

(Gromiha, Ahmad et al. 2004) suggested. This means that P66 proteins for all borrelial species could form β-barrel cylinders as many other outer membrane porins studied to date (Saier 2000; Delcour 2002; Charbit 2003).

Purification and identification of the P66 homologues in Lyme disease and relapsing fever Borrelia spirohetes From different Borrelia species, P66 homologues were isolated and purified using chromatography. Proteins exhibited a molecular mass of approximately 66 kDa for all Lyme disease and relapsing fever strains. Polyclonal antiserum raised against B. burgdorferi P66 reacted to purified proteins from all strains except to 66-kDa proteins purified from B. duttonii and of B. recurrentis. This could be due to variations within the primary sequence of B. burgdorferi P66 as compared with those of the two relapsing fever species. The 66-kDa proteins isolated from B. duttonii and from B. recurrentis in later experiments formed the typical 9-11 nS channels previously described for P66 (Skare, Mirzabekov et al. 1997). To compare biophysical properties of P66 from Lyme disease and relapsing fever Borrelia, we performed a detailed characterization of the P66 homologues in black lipid bilayer membranes, including B. burgdorferi P66 to employ the same conditions in all the measurements. Even though B. hermsii P66 was clearly identified by Western blot, it did not exhibit any activity in the black lipid bilayer assay despite the high pore-forming activity of the P66 homologues of all other tested species. This finding was in agreement with single-channel experiments using whole B. hermsii outer membrane fraction. There was no indication in these experiments that channels could be formed with a single-channel conductance of about 10 nS in 1 M KCl, which is similar to the conductance of the other P66 species (Shang, Skare et al. 1998; Thein, Bunikis et al. 2008). It seems to be possible that B. hermsii P66 is susceptible to treatment with detergents, which could destroy its pore-forming ability in a similar fashion as SDS destroys pore formation of all other P66 homologues. Other approaches using milder isolation conditions may be required to show the functionality of P66 in B. hermsii.

50 | P a g e

P66 homologues exhibit similar biophysical properties The single-channel conductance of the P66 homologues with pore-forming capability was studied. The conductance of the P66 pores from Lyme disease and relapsing fever Borrelia in 1 M KCl was within 9 and 11 nS. Experiments showed that all P66 homologues formed wide and water-filled channels without any obvious selectivity for anions or cations. In a previous study, it was demonstrated that P66 of B. burgdorferi showed voltage-dependent closure (Skare, Mirzabekov et al. 1997). We performed multi-channel experiments with all P66 homologues and determined that P66 from both Lyme disease and relapsing fever Borrelia species studied here exhibited voltage-gating with a slight species-dependent specificity.

Reconstitution experiments demonstrated that P66 from B. afzelii, B. garinii, B. duttonii and B. recurrentis formed consistently well-defined channels with a single-channel conductance in 1 M KCl that ranged between 9 and 11 nS, similar to that obtained here and previously for the pores formed by B. burgdorferi P66 (Skare, Mirzabekov et al. 1997). The small variations of single-channel conductance were probably due to minor species-specific differences of intrinsic pore properties which could be caused by amino acid exchanges in the sequences. However, it remains unclear whether the high single-channel conductance is caused by a likewise enormous diameter of the P66 pore as suggested previously (Skare, Mirzabekov et al. 1997). Attempts to answer this question are presented in Paper V of this thesis.

51 | P a g e

Paper V.

In this work, we further characterized the P66 porin from B. burgdorferi. The intriguing question of pore size and constitution of P66 was addressed.

P66 pore size estimation P66 is able to form pores in planar lipid bilayers with an enormously high single-channel conductance of 11 nS in 1 M KCl (Barcena-Uribarri, Thein et al. 2010). Porins from other spirochetes such as Spirocheta aurantia and Treponema denticola also exhibit extremely high single-channel conductance (Egli, Leung et al. 1993); this is atypical and rare for Gram-negative bacterial porins. To date, besides selectivity and estimated pore diameters, very little is understood about the apparent pore sizes and the structures of these outer membrane channels. The channel diameter of P66 has been calculated to be approximately 2.6 nm (Skare, Mirzabekov et al. 1997), which is rather large compared to other pore-forming outer membrane proteins (Benz 1985). This calculation of the P66 channel diameter was based on the assumption that the conductance of the channel is equal to the conductivity of a simple cylinder in an aqueous salt solution. The length of the cylinder was taken to be equal to the thickness of the membrane. This method does not take into account other important factors such as the form of the channel. To get better insight into the effective pore diameter, the conductance of the P66 channel reconstituted in planar lipid membranes was studied using nonelectrolytes as a function of their spherical size (Krasilnikov 2002). Large, nonpermeable nonelectrolytes with hydrodynamic radii between 0.94 and 2.50 nm (PEG 1000, PEG 3000 and PEG 6000) did not enter the P66 channel and showed no effect on its conductance. However, in the presence of small nonelectrolytes with hydrodynamic radii up to 0.60 nm, such as ethylene glycol, glycerol, arabinose, sorbitol, maltose and PEG 300, the P66 single-channel conductance decreased proportionally to that of the bulk solution conductivity. Surprisingly, the presence of PEG 400 and PEG 600 (hydrodynamic radii of 0.70 and 0.80 nm, respectively) in the bathing solution resulted in an exceptional low single-channel. This effect was most likely due to an interaction between the polymer and the channel interior, resulting in a blockage of the conductance.

52 | P a g e

PEG 1000 with the radius of 0.94 nm was the smallest nonelectrolyte that could not enter the P66 channel. This observation allowed us to determine the approximate radius of the P66 entrance. The possible constriction zone radius was estimated to be approximately 0.39 nm since sorbitol was the smallest nonelectrolyte that did not pass freely through the channel and did not fill it completely (Figure 3).

Figure 3. Schematic illustration of P66 monomer pore size estimation using nonelectrolytes with different hydrodynamic radii. Spherical particles represent PEG 1000 (red), sorbitol (yellow) and glycerol (green).

Our suggested P66 entrance diameter of ~1.9 nm is within the range of several other Gram-negative bacterial porins and other membrane channels that were characterized by the use of nonelectrolytes, such as Bacillus anthracis (PA63)7 (d ≈ 2 nm) (Nablo, Halverson et al. 2008) or Staphylococcus aureus α-toxin (d ≈ 1.35 nm) (Krasilnikov, Sabirov et al. 1992). These two channels exhibit single-channel conductance in 1 M KCl of ~180 pS (Orlik, Schiffler et al. 2005) and ~775 pS (Menestrina, Dalla Serra et al. 2003), respectively. Even though the P66 channel diameter is closer to that of Bacillus anthracis (PA63)7, its single-

53 | P a g e

channel conductance is about 60-fold higher than (PA63)7. The molecular organization of P66 as a complex in the outer membrane of Borrelia could be an explanation for this phenomenon.

Channel interior Membrane experiments in the presence of 20% PEG 400 or PEG 600 resulted in drastically reduced single-channel conductance. Interestingly, the kinetics of conductance decrease after addition of these compounds was very slow, which differs from that of substrate-binding porins (Andersen, Cseh et al. 1998; Orlik, Andersen et al. 2002; Denker, Orlik et al. 2005). This finding enforced the idea that specific interactions of PEG 400 and PEG 600 with the channel interior result in blocking the ion flux through the pore. Measurements of the current noise through open and nonelectrolyte-induced closed states of P66 channels suggest nonchemical (e.g., substrate binding) interaction of nonelectrolytes with the P66 channel. This observation has not been reported to date in any similar studies and requires further investigation.

P66 is composed of several subunits Additional measurements were performed to study PEG 400-induced blockage of the P66 channel on the single-channel level. After reconstitution of a single 11 nS P66 unit into the membrane, PEG 400 was added. The measurements

revealed a PEG-induced, stepwise closing of P66 in seven substrates with a conductance of about 1.5 nS in 1 M KCl. Together these results suggested that the P66 channel may be formed by a bundle of pores (Figure 4).

Figure 4. Schematic representation of putative heptameric oligomer of P66.

54 | P a g e

In addition, purified P66 was investigated on a Blue native PAGE gel. A band of approximately 460 KDa was observed and confirmed by immunoblot to be P66. Theoretically, the heptameric oligomer of P66 would have a molecular mass of 462 kDa. Furthermore, the six or seven substrates with a conductance of about 1.5 nS could match the 11 nS conductance of the oligomer. P66 could be the first known example of a porin constituted by a bundle of seven independent channels in a protein complex.

55 | P a g e

Conclusions

• Borrelia burgdorferi possesses an active RND type efflux system, which plays an important role in virulence of this pathogen

• BesC is a homologue of E. coli TolC and exhibits channel-forming activity in artificial membranes

• BesA is an atypical membrane fusion protein

• Oms38 is a novel porin in the outer membrane of relapsing fever Borrelia

• Lyme disease Borrelia harbors the Oms38 homologue DipA

• DipA exhibits specificity for dicarboxylic anions and is the first described solute-specific porin in Borrelia

• Both relapsing fever and Lyme disease spirochetes contain P66 homologues with similar biophysical characteristics

• The effective pore forming diameter of P66 is smaller than predicted

• P66 forms a high molecular mass oligomer with many single channel conductive units

56 | P a g e

Acknowledgements

Somebody once said that if there is a part of the thesis that has a significant social impact factor, then it must be the acknowledgements. So here it is!

The last seven years were quite a journey! A long one, but definitely worth it! I met a lot of interesting and charming people throughout the years. They helped me to grow, to expand my mind, to see things from another, sometimes totally unexpected, angle. A great number of them became my friends. Through them I learned about different cultures and customs, got introduced to a variety of cuisines, got dragged into one or two adventures and simply spent a lot of good time while elaborating on important (or perhaps not so important) matters of life over a glass of wine. Most importantly, now when I look at the world map, I see many places where I have friends! I send my gratitude to all of you!

A special word goes to my supervisor - Sven Bergström - I could've not had a better mentor than you! The positivity and enthusiasm you carry are simply infectious! Thank you for friendship, advice, help and motivation! We will keep on working and perhaps someday we will make you rich and famous!

GroupSvB. The word speaks for itself! A lot of good words go to former and present members of the group, and people in one or another way related to it. A big thanks goes to Yngve, my co-supervisor, for sharing your experience, tips and tricks; Betty, for being a friend and such a great host on both the East and West coasts; Marija, Latvian 'sesutė' and my snowboard coach; Pär and Anna, for the music, good words and "hemgjord björksnaps" (I think I still have some left), Christer and Jenny, for funny things happening "without" your involvement and always being ready to discuss any subject; Ingela, for being such a good safe keeper of our group! Twelve points goes to Elin, no doubt about that! Thanks to Mari, for help in the lab; Marie, for Star Trek and meeting with Quark! Patrik, for the right and positive attitude; Lelle, being course assistant with you was really fun; Lisette, for nice chats; Michaela and Stefan — last time with Play Station was just a warm up, next time we will take it seriously! Johan, for bringing a lot of fresh energy — I'm sure we can sequence whatever moves and has a spiral shape! May, still, the thought that you guys

57 | P a g e

have summer when we have winter, is disturbing... :) Jenny and Reine, our good neighbors, are you sure you are not from our group?

Have to send my regards to the rapidly expanding journal club. Still not as big as some competitors’, but full of the same valuable ideas and advice. Can it be different when we have Ulrich von Pawel-Rammingen and Constantin Urban sharing their thoughts and ideas! As you noticed it was a statement, not a question.

I have to say a separate thanks to Jörgen Johansson, mostly for my first co-authorship on a publication and cheering words! Since the last study on sRNAs in Borrelia, quite some time has passed, shall we go for an update? :)

Mazen and Margret, we had some good times enjoying smoke of narghile on one or two or more occasions! By the way, Mazen, I have another bottle of good whisky for you (see Mazen's thesis). Since I mentioned whiskey, I have to give tribute to the Whiskey Club I was once in and to its members who now are spread around the world — you guys introduced me to the Scottish finest!

Sakura and Hans, thanks for those nice dinners and sharing some good stories from the past (not all needs to be shared, Hans); Edmund, was always fun to go with you on different trips, have a beer and discuss all sorts of subjects, keep your good spirit and energy! Constance, Sara, Tiago, Stefanie, Barbro, Tianyan, Jonas, Stefan, Mikael, Maria, Lisandro, Olena, Gosia, Claudia, Annika, Stina, Linda and many more... You all make the Department a special place!

My gratitude goes to Bernt Eric Uhlin for help whenever it was needed; to Sun Nyunt Wai for lots of ideas and enthusiastic approaches to solving problems; to Johnny who can get you anything (he knows people in the right places); to Siv for all the sequences; to Marita, Ethel, Berit, Maria for taking care of all the important paper stuff; to Marek for smoothly running departmental Matrix; to personnel of "Media" and "Disken" for making our lives easier; to Mariella, for making the world a cleaner place!

I was privileged to collaborate with a guru of porins, Roland Benz. Thank you for your hospitality, introduction to the lipid bilayer world and memorable trips around Würzburg. My regards go to the whole "AG Benz", especially to Christian Andersen, for help with channel-tunnels and many, many great ideas;

58 | P a g e

to Katrin, Anna, Maite and Ivan — I'm glad I met you guys! And most of all I have to say big thanks to Marcus and Katja, for your friendship, laughs and fun times we had on those many trips exploring Bavaria!

I send my gratitude to Wolfram R. Zückert, who took great care of me in Kansas, introduced me to "tailgating" parties, baseball and American football! Regards to the members of his lab, especially to Ozan for the "real" Kansas beer and glimpse into college life; and to Min for a super Korean barbeque!

Acknowledgements would be incomplete if the Lithuanian gang would be not mentioned! When I was coming so far north, I thought I would be the only one from the Baltic states in Umeå. I was wrong. Totally! :) I met these amazing people who became my friends. Ernestas and Eglė, Jūratė and Darius, Ramunė and Arūnas, Karolis and Rima, Jaunius and others, I want to thank you all for your help and friendship throughout the years! For all the good times we had, for all the trips we took, for all the wine we drank, for all the fun we shared!

A special gratitude I would like to express to my parents for their endless support and encouragement! Ačių už nuolatinį palaikymą, motyvaciją, gerą ūpą ir buvimą šalia visais gyvenimo atvejais!

And finally I want say how thankful I am to my little northern family. Lobis, the four-legged companion, who is always ready to give a motivating tail wag and Sabrina, mein Herzblatt, meine zweite Hälfte, mein Schatz, meine ♥! I am most grateful to you, my dear! For your warmth, loving heart, ability to see beauty and joy in simple things of life and sharing that with me! I wouldn't be complete without you! I love you!

59 | P a g e

References

Achouak, W., T. Heulin, et al. (2001). "Multiple facets of bacterial porins." FEMS Microbiol Lett 199(1): 1-7.

Afzelius, A. (1921). "Erythema migrans." Acta Derm Venereol 2: 120-125. Alberti, S., G. Marques, et al. (1996). "Interaction between complement

subcomponent C1q and the Klebsiella pneumoniae porin OmpK36." Infect Immun 64(11): 4719-4725.

Andersen, C. (2003). "Channel-tunnels: outer membrane components of type I secretion systems and multidrug efflux pumps of Gram-negative bacteria." Rev Physiol Biochem Pharmacol 147: 122-165.

Andersen, C. (2004). Drug efflux and protein export through channel-tunnels. Weinheim/Germany, WILEY-VCH Verlag GmbH.

Andersen, C., R. Cseh, et al. (1998). "Study of sugar binding to the sucrose-specific ScrY channel of enteric bacteria using current noise analysis." J Membr Biol 164(3): 263-274.

Andersen, C., E. Koronakis, et al. (2002). "Transition to the open state of the TolC periplasmic tunnel entrance." Proc Natl Acad Sci U S A 99(17): 11103-11108.

Andersson, M., A. Nordstrand, et al. (2007). "In situ immune response in brain and kidney during early relapsing fever borreliosis." J Neuroimmunol 183(1-2): 26-32.

Asbrink, E. and A. Hovmark (1985). "Successful cultivation of spirochetes from skin lesions of patients with erythema chronicum migrans Afzelius and acrodermatitis chronica atrophicans." Acta Pathol Microbiol Immunol Scand B 93(2): 161-163.

Assous, M. V., D. Postic, et al. (1994). "Individualisation of two new genomic groups among American Borrelia burgdorferi sensu lato strains." FEMS Microbiol Lett 121(1): 93-98.

Athappilly, F. K. and W. A. Hendrickson (1995). "Structure of the biotinyl domain of acetyl-coenzyme A carboxylase determined by MAD phasing." Structure 3(12): 1407-1419.

Bagos, P. G., T. D. Liakopoulos, et al. (2004). "A Hidden Markov Model method, capable of predicting and discriminating beta-barrel outer membrane proteins." BMC Bioinformatics 5: 29.

Barbour, A. G. (1984). "Isolation and cultivation of Lyme disease spirochetes." Yale J Biol Med 57(4): 521-525.

Barbour, A. G. and S. F. Hayes (1986). "Biology of Borrelia species." Microbiol Rev 50(4): 381-400.

60 | P a g e

Barbour, A. G., S. F. Hayes, et al. (1986). "A Borrelia-specific monoclonal antibody binds to a flagellar epitope." Infect Immun 52(2): 549-554.

Barbour, A. G., A. Jasinskas, et al. (2008). "A genome-wide proteome array reveals a limited set of immunogens in natural infections of humans and white-footed mice with Borrelia burgdorferi." Infect Immun.

Barbour, A. G., A. Jasinskas, et al. (2008). "A genome-wide proteome array reveals a limited set of immunogens in natural infections of humans and white-footed mice with Borrelia burgdorferi." Infect Immun 76(8): 3374-3389.

Barbour, A. G., S. L. Tessier, et al. (1982). "Variable major proteins of Borrellia hermsii." J Exp Med 156(5): 1312-1324.

Barcena-Uribarri, I., M. Thein, et al. (2010). "P66 porins are present in both Lyme disease and relapsing fever spirochetes: A comparison of the biophysical properties of P66 porins from six Borrelia species." Biochim Biophys Acta.

Barclay, A. J. and J. B. Coulter (1990). "Tick-borne relapsing fever in central Tanzania." Trans R Soc Trop Med Hyg 84(6): 852-856.

Bauer, K., M. Struyve, et al. (1989). "One single lysine residue is responsible for the special interaction between polyphosphate and the outer membrane porin PhoE of Escherichia coli." J Biol Chem 264(28): 16393-16398.

Belisle, J. T., M. E. Brandt, et al. (1994). "Fatty acids of Treponema pallidum and Borrelia burgdorferi lipoproteins." J Bacteriol 176(8): 2151-2157.

Ben-Menachem, G., J. Kubler-Kielb, et al. (2003). "A newly discovered cholesteryl galactoside from Borrelia burgdorferi." Proc Natl Acad Sci U S A 100(13): 7913-7918.

Benz, R. (1985). "Porin from bacterial and mitochondrial outer membranes." CRC Crit Rev Biochem 19(2): 145-190.

Benz, R. (1994). "Bacterial Cell Wall." Elsevier, Amsterdam. Benz, R. (1994). Solute uptake through bacterial outer membranes. Bacterial

Cell Wall. G. J. M. Hackenbek R. Amsterdam, Elsevier Science B. V.: 397-423.

Benz, R. and K. Janko (1976). "Voltage-induce capacitance relaxation of lipid bilayer membranes. Effects of membrane composition." Biochim Biophys Acta 455(3): 721-738.

Benz, R., K. Janko, et al. (1978). "Formation of large, ion-permeable membrane channels by the matrix protein (porin) of Escherichia coli." Biochim Biophys Acta 511(3): 305-319.

Benz, R., E. Maier, et al. (1993). "TolC of Escherichia coli functions as an outer membrane channel." Zentralbl Bakteriol 278(2-3): 187-196.

61 | P a g e

Benz, R., A. Schmid, et al. (1986). "Pore formation by LamB of Escherichia coli in lipid bilayer membranes." J Bacteriol 165(3): 978-986.

Benz, R., A. Schmid, et al. (1987). "Mechanism of sugar transport through the sugar-specific LamB channel of Escherichia coli outer membrane." J Membr Biol 100(1): 21-29.

Bernardini, M. L., M. G. Sanna, et al. (1993). "OmpC is involved in invasion of epithelial cells by Shigella flexneri." Infect Immun 61(9): 3625-3635.

Blanco, D. R., C. I. Champion, et al. (1996). "Recombinant Treponema pallidum rare outer membrane protein 1 (Tromp1) expressed in Escherichia coli has porin activity and surface antigenic exposure." J Bacteriol 178(23): 6685-6692.

Brandt, M. E., B. S. Riley, et al. (1990). "Immunogenic integral membrane proteins of Borrelia burgdorferi are lipoproteins." Infect Immun 58(4): 983-991.

Brown, M. H., I. T. Paulsen, et al. (1999). "The multidrug efflux protein NorM is a prototype of a new family of transporters." Mol Microbiol 31(1): 394-395.

Bryceson, A. D. (1976). "Clinical pathology of the Jarisch-Herxheimer reaction." J Infect Dis 133(6): 696-704.

Bryceson, A. D., E. H. Parry, et al. (1970). "Louse-borne relapsing fever." Q J Med 39(153): 129-170.

Buchanan, S. K. (1999). "Beta-barrel proteins from bacterial outer membranes: structure, function and refolding." Curr Opin Struct Biol 9(4): 455-461.

Buchwald, A. (1883). "Ein Fall von diffuser idiopathischer Haut-Atrophie." Archives of Dermatological Research: 553-556.

Bunikis, I., K. Denker, et al. (2008). "An RND-type efflux system in Borrelia burgdorferi is involved in virulence and resistance to antimicrobial compounds." PLoS Pathog 4(2): e1000009.doi:1000010.1001371/journal.ppat.1000009.

Bunikis, J. and A. G. Barbour (1999). "Access of antibody or trypsin to an integral outer membrane protein (P66) of Borrelia burgdorferi is hindered by Osp lipoproteins." Infect Immun 67(6): 2874-2883.

Bunikis, J., U. Garpmo, et al. (2004). "Sequence typing reveals extensive strain diversity of the Lyme borreliosis agents Borrelia burgdorferi in North America and Borrelia afzelii in Europe." Microbiology 150(Pt 6): 1741-1755.

Bunikis, J., L. Noppa, et al. (1995). "Molecular analysis of a 66-kDa protein associated with the outer membrane of Lyme disease Borrelia." FEMS Microbiol Lett 131(2): 139-145.

Bunikis, J., L. Noppa, et al. (1996). "Surface exposure and species specificity of an immunoreactive domain of a 66-kilodalton outer membrane protein

62 | P a g e

(P66) of the Borrelia spp. that cause Lyme disease." Infect Immun 64(12): 5111-5116.

Bunikis, J., L. Noppa, et al. (1996). "Surface exposure and species specificity of an immunoreactive domain of a 66-kilodalton outer membrane protein (P66) of the Borrelia spp. that cause Lyme disease." Infect Immun 64(12): 5111-5116.

Burgdorfer, W. (1984). "Discovery of the Lyme disease spirochete and its relation to tick vectors." Yale J Biol Med 57(4): 515-520.

Burgdorfer, W., A. G. Barbour, et al. (1982). "Lyme disease-a tick-borne spirochetosis?" Science 216(4552): 1317-1319.

Cadavid, D. and A. G. Barbour (1998). "Neuroborreliosis during relapsing fever: review of the clinical manifestations, pathology, and treatment of infections in humans and experimental animals." Clin Infect Dis 26(1): 151-164.

Cadavid, D., A. R. Pachner, et al. (2001). "Isogenic serotypes of Borrelia turicatae show different localization in the brain and skin of mice." Infect Immun 69(5): 3389-3397.

Cadavid, D., D. D. Thomas, et al. (1994). "Variability of a bacterial surface protein and disease expression in a possible mouse model of systemic Lyme borreliosis." J Exp Med 179(2): 631-642.

Canica, M. M., F. Nato, et al. (1993). "Monoclonal antibodies for identification of Borrelia afzelii sp. nov. associated with late cutaneous manifestations of Lyme borreliosis." Scand J Infect Dis 25(4): 441-448.

Cao, T. B. and M. H. Saier, Jr. (2001). "Conjugal type IV macromolecular transfer systems of Gram-negative bacteria: organismal distribution, structural constraints and evolutionary conclusions." Microbiology 147(Pt 12): 3201-3214.

Casjens, S. (1998). "The diverse and dynamic structure of bacterial genomes." Annu Rev Genet 32: 339-377.

Casjens, S. (2000). "Borrelia genomes in the year 2000." J Mol Microbiol Biotechnol 2(4): 401-410.

Casjens, S., N. Palmer, et al. (2000). "A bacterial genome in flux: the twelve linear and nine circular extrachromosomal DNAs in an infectious isolate of the Lyme disease spirochete Borrelia burgdorferi." Mol Microbiol 35(3): 490-516.

Charbit, A. (2003). "Maltodextrin transport through lamb." Front Biosci 8: s265-274.

Charon, N. W., E. P. Greenberg, et al. (1992). "Spirochete chemotaxis, motility, and the structure of the spirochetal periplasmic flagella." Res Microbiol 143(6): 597-603.

63 | P a g e

Chimento, D. P., A. K. Mohanty, et al. (2003). "Substrate-induced transmembrane signaling in the cobalamin transporter BtuB." Nat Struct Biol 10(5): 394-401.

Chung, Y. J., C. Krueger, et al. (2001). "Size comparisons among integral membrane transport protein homologues in bacteria, Archaea, and Eucarya." J Bacteriol 183(3): 1012-1021.

Coburn, J., W. Chege, et al. (1999). "Characterization of a candidate Borrelia burgdorferi beta3-chain integrin ligand identified using a phage display library." Mol Microbiol 34(5): 926-940.

Coburn, J. and C. Cugini (2003). "Targeted mutation of the outer membrane protein P66 disrupts attachment of the Lyme disease agent, Borrelia burgdorferi, to integrin alphavbeta3." Proc Natl Acad Sci U S A 100(12): 7301-7306.

Collares-Pereira, M., S. Couceiro, et al. (2004). "First isolation of Borrelia lusitaniae from a human patient." J Clin Microbiol 42(3): 1316-1318.

Cowan, S. W., T. Schirmer, et al. (1992). "Crystal structures explain functional properties of two E. coli porins." Nature 358(6389): 727-733.

Cullen, P. A., D. A. Haake, et al. (2004). "Outer membrane proteins of pathogenic spirochetes." FEMS Microbiol Rev 28(3): 291-318.

Dardel, F., A. L. Davis, et al. (1993). "Three-dimensional structure of the lipoyl domain from Bacillus stearothermophilus pyruvate dehydrogenase multienzyme complex." J Mol Biol 229(4): 1037-1048.

Das, R., H. Hegyi, et al. (2000). "Genome analyses of spirochetes: a study of the protein structures, functions and metabolic pathways in Treponema pallidum and Borrelia burgdorferi." J Mol Microbiol Biotechnol 2(4): 387-392.

De Silva, A. M. and E. Fikrig (1995). "Growth and migration of Borrelia burgdorferi in Ixodes ticks during blood feeding." Am J Trop Med Hyg 53(4): 397-404.

Defoe, G. and J. Coburn (2001). "Delineation of Borrelia burgdorferi p66 sequences required for integrin alpha(IIb)beta(3) recognition." Infect Immun 69(5): 3455-3459.

Delcour, A. H. (2002). "Structure and function of pore-forming beta-barrels from bacteria." J Mol Microbiol Biotechnol 4(1): 1-10.

Denker, K., F. Orlik, et al. (2005). "Site-directed mutagenesis of the greasy slide aromatic residues within the LamB (maltoporin) channel of Escherichia coli: effect on ion and maltopentaose transport." J Mol Biol 352(3): 534-550.

Dilger, J. P. and R. Benz (1985). "Optical and electrical properties of thin monoolein lipid bilayers." J Membr Biol 85(2): 181-189.

64 | P a g e

Dinh, T., I. T. Paulsen, et al. (1994). "A family of extracytoplasmic proteins that allow transport of large molecules across the outer membranes of gram-negative bacteria." J Bacteriol 176(13): 3825-3831.

Egli, C., W. K. Leung, et al. (1993). "Pore-forming properties of the major 53-kilodalton surface antigen from the outer sheath of Treponema denticola." Infect Immun 61(5): 1694-1699.

Eiffert, H., H. Lotter, et al. (1991). "Identification of an immunoreactive non-proteinaleous component in Borrelia burgdorferi." Med Microbiol Immunol 180(5): 229-237.

Fekkes, P. and A. J. Driessen (1999). "Protein targeting to the bacterial cytoplasmic membrane." Microbiol Mol Biol Rev 63(1): 161-173.

Felsenfeld, O. (1971). "Borrelia: Strains, Vectors, Human and Animal Borreliosis." St Louis: Warren H Green inc.

Fenno, J. C., K. H. Muller, et al. (1996). "Sequence analysis, expression, and binding activity of recombinant major outer sheath protein (Msp) of Treponema denticola." J Bacteriol 178(9): 2489-2497.

Fenno, J. C., G. W. Wong, et al. (1997). "Conservation of msp, the gene encoding the major outer membrane protein of oral Treponema spp." J Bacteriol 179(4): 1082-1089.

Ferenci, T., M. Schwentorat, et al. (1980). "Lambda receptor in the outer membrane of Escherichia coli as a binding protein for maltodextrins and starch polysaccharides." J Bacteriol 142(2): 521-526.

Fikrig, E., S. R. Telford, 3rd, et al. (1992). "Elimination of Borrelia burgdorferi from vector ticks feeding on OspA-immunized mice." Proc Natl Acad Sci U S A 89(12): 5418-5421.

Fingerle, V., U. C. Schulte-Spechtel, et al. (2008). "Epidemiological aspects and molecular characterization of Borrelia burgdorferi s.l. from southern Germany with special respect to the new species Borrelia spielmanii sp. nov." Int J Med Microbiol 298(3-4): 279-290.

Fox, G. E., J. D. Wisotzkey, et al. (1992). "How close is close: 16S rRNA sequence identity may not be sufficient to guarantee species identity." Int J Syst Bacteriol 42(1): 166-170.

Fraser, C. M., S. Casjens, et al. (1997). "Genomic sequence of a Lyme disease spirochaete, Borrelia burgdorferi." Nature 390(6660): 580-586.

Glockner, G., R. Lehmann, et al. (2004). "Comparative analysis of the Borrelia garinii genome." Nucleic Acids Res 32(20): 6038-6046.

Glockner, G., U. Schulte-Spechtel, et al. (2006). "Comparative genome analysis: selection pressure on the Borrelia vls cassettes is essential for infectivity." BMC Genomics 7: 211.

65 | P a g e

Goldstein, S. F., K. F. Buttle, et al. (1996). "Structural analysis of the Leptospiraceae and Borrelia burgdorferi by high-voltage electron microscopy." J Bacteriol 178(22): 6539-6545.

Goldstein, S. F., N. W. Charon, et al. (1994). "Borrelia burgdorferi swims with a planar waveform similar to that of eukaryotic flagella." Proc Natl Acad Sci U S A 91(8): 3433-3437.

Gotoh, N., H. Tsujimoto, et al. (1995). "The outer membrane protein OprM of Pseudomonas aeruginosa is encoded by oprK of the mexA-mexB-oprK multidrug resistance operon." Antimicrob Agents Chemother 39(11): 2567-2569.

Green, J. D., E. D. Laue, et al. (1995). "Three-dimensional structure of a lipoyl domain from the dihydrolipoyl acetyltransferase component of the pyruvate dehydrogenase multienzyme complex of Escherichia coli." J Mol Biol 248(2): 328-343.

Grimm, D., K. Tilly, et al. (2005). "Defining plasmids required by Borrelia burgdorferi for colonization of tick vector Ixodes scapularis (Acari: Ixodidae)." J Med Entomol 42(4): 676-684.

Grimm, D., K. Tilly, et al. (2004). "Outer-surface protein C of the Lyme disease spirochete: a protein induced in ticks for infection of mammals." Proc Natl Acad Sci U S A 101(9): 3142-3147.

Gromiha, M. M., S. Ahmad, et al. (2004). "Neural network-based prediction of transmembrane beta-strand segments in outer membrane proteins." J Comput Chem 25(5): 762-767.

Gross, R. (1995). "Domain structure of the outer membrane transporter protein CyaE of Bordetella pertussis." Mol Microbiol 17(6): 1219-1220.

Gupta, R. S. (2000). "The natural evolutionary relationships among prokaryotes." Crit Rev Microbiol 26(2): 111-131.

Haake, D. A. (2000). "Spirochaetal lipoproteins and pathogenesis." Microbiology 146 ( Pt 7): 1491-1504.

Haake, D. A., C. I. Champion, et al. (1993). "Molecular cloning and sequence analysis of the gene encoding OmpL1, a transmembrane outer membrane protein of pathogenic Leptospira spp." J Bacteriol 175(13): 4225-4234.

Henderson, I. R., F. Navarro-Garcia, et al. (1998). "The great escape: structure and function of the autotransporter proteins." Trends Microbiol 6(9): 370-378.

Hossain, H., H. J. Wellensiek, et al. (2001). "Structural analysis of glycolipids from Borrelia burgdorferi." Biochimie 83(7): 683-692.

Hueck, C. J. (1998). "Type III protein secretion systems in bacterial pathogens of animals and plants." Microbiol Mol Biol Rev 62(2): 379-433.

66 | P a g e

Janausch, I. G., E. Zientz, et al. (2002). "C4-dicarboxylate carriers and sensors in bacteria." Biochim Biophys Acta 1553(1-2): 39-56.

Jerse, A. E., N. D. Sharma, et al. (2003). "A gonococcal efflux pump system enhances bacterial survival in a female mouse model of genital tract infection." Infect Immun 71(10): 5576-5582.

Jewett, M. W., R. Byram, et al. (2007). "Genetic basis for retention of a critical virulence plasmid of Borrelia burgdorferi." Mol Microbiol 66(4): 975-990.

Johnson, J. M. and G. M. Church (1999). "Alignment and structure prediction of divergent protein families: periplasmic and outer membrane proteins of bacterial efflux pumps." J Mol Biol 287(3): 695-715.

Johnson, R. C., G. P. Schmid, et al. (1984). "Borrelia burgdorferi sp. nov.: Etiologic Agent of Lyme Disease." Int. J. Syst. Bacteriol 34: 496-497.

Jones, J. D., K. W. Bourell, et al. (1995). "Membrane topology of Borrelia burgdorferi and Treponema pallidum lipoproteins." Infect Immun 63(7): 2424-2434.

Jongen, V. H., J. van Roosmalen, et al. (1997). "Tick-borne relapsing fever and pregnancy outcome in rural Tanzania." Acta Obstet Gynecol Scand 76(9): 834-838.

Jordy, M., C. Andersen, et al. (1996). "Rate constants of sugar transport through two LamB mutants of Escherichia coli: comparison with wild-type maltoporin and LamB of Salmonella typhimurium." J Mol Biol 259(4): 666-678.

Ketchem, R. R., W. Hu, et al. (1993). "High-resolution conformation of gramicidin A in a lipid bilayer by solid-state NMR." Science 261(5127): 1457-1460.

Koebnik, R., K. P. Locher, et al. (2000). "Structure and function of bacterial outer membrane proteins: barrels in a nutshell." Mol Microbiol 37(2): 239-253.

Konninger, U. W., S. Hobbie, et al. (1999). "The haemolysin-secreting ShlB protein of the outer membrane of Serratia marcescens: determination of surface-exposed residues and formation of ion-permeable pores by ShlB mutants in artificial lipid bilayer membranes." Mol Microbiol 32(6): 1212-1225.

Koronakis, V. (2003). "TolC--the bacterial exit duct for proteins and drugs." FEBS Lett 555(1): 66-71.

Koronakis, V., C. Andersen, et al. (2001). "Channel-tunnels." Curr Opin Struct Biol 11(4): 403-407.

Koronakis, V., A. Sharff, et al. (2000). "Crystal structure of the bacterial membrane protein TolC central to multidrug efflux and protein export." Nature 405(6789): 914-919.

67 | P a g e

Krasilnikov, O. V. (2002). Sizing channels with neutral polymers. Structure and Dynamics of Confined Polymers. K. Press. Dordrecht, Netherlands, Kasianowicz, J. J., Kellermayer M. S. Z., Deamer, D. W.: 97-115.

Krasilnikov, O. V., R. Z. Sabirov, et al. (1992). "A simple method for the determination of the pore radius of ion channels in planar lipid bilayer membranes." FEMS Microbiol Immunol 5(1-3): 93-100.

Kropinski, A. M., T. R. Parr, Jr., et al. (1987). "Isolation of the outer membrane and characterization of the major outer membrane protein from Spirochaeta aurantia." J Bacteriol 169(1): 172-179.

Kubo, A. and R. S. Stephens (2001). "Substrate-specific diffusion of select dicarboxylates through Chlamydia trachomatis PorB." Microbiology 147(Pt 11): 3135-3140.

Kurtenbach, K., K. Hanincova, et al. (2006). "Fundamental processes in the evolutionary ecology of Lyme borreliosis." Nat Rev Microbiol 4(9): 660-669.

Kyte, J. and R. F. Doolittle (1982). "A simple method for displaying the hydropathic character of a protein." J Mol Biol 157(1): 105-132.

Labandeira-Rey, M. and J. T. Skare (2001). "Decreased infectivity in Borrelia burgdorferi strain B31 is associated with loss of linear plasmid 25 or 28-1." Infect Immun 69(1): 446-455.

Lacroix, F. J., A. Cloeckaert, et al. (1996). "Salmonella typhimurium acrB-like gene: identification and role in resistance to biliary salts and detergents and in murine infection." FEMS Microbiol Lett 135(2-3): 161-167.

Larsson, C. and S. Bergström (2008). "A novel and simple method for laboratory diagnosis of relapsing Fever borreliosis." Open Microbiol J 2: 10-12.

Lathrop, S. L., R. Ball, et al. (2002). "Adverse event reports following vaccination for Lyme disease: December 1998-July 2000." Vaccine 20(11-12): 1603-1608.

Lee, E. H. and W. M. Shafer (1999). "The farAB-encoded efflux pump mediates resistance of gonococci to long-chained antibacterial fatty acids." Mol Microbiol 33(4): 839-845.

Lescot, M., S. Audic, et al. (2008). "The genome of Borrelia recurrentis, the agent of deadly louse-borne relapsing fever, is a degraded subset of tick-borne Borrelia duttonii." PLoS Genet 4(9): e1000185.

Livermore, B. P., R. F. Bey, et al. (1978). "Lipid metabolism of Borrelia hermsi." Infect Immun 20(1): 215-220.

Lobedanz, S., E. Bokma, et al. (2007). "A periplasmic coiled-coil interface underlying TolC recruitment and the assembly of bacterial drug efflux pumps." Proc Natl Acad Sci U S A 104(11): 4612-4617.

68 | P a g e

Locher, K. P., B. Rees, et al. (1998). "Transmembrane signaling across the ligand-gated FhuA receptor: crystal structures of free and ferrichrome-bound states reveal allosteric changes." Cell 95(6): 771-778.

Lomovskaya, O. and W. J. Watkins (2001). "Efflux pumps: their role in antibacterial drug discovery." Curr Med Chem 8(14): 1699-1711.

Lynch, M. (2006). "Streamlining and simplification of microbial genome architecture." Annu Rev Microbiol 60: 327-349.

Ma, J., P. M. Hine, et al. (1996). "Safety, efficacy, and immunogenicity of a recombinant Osp subunit canine Lyme disease vaccine." Vaccine 14(14): 1366-1374.

Maier, C., E. Bremer, et al. (1988). "Pore-forming activity of the Tsx protein from the outer membrane of Escherichia coli. Demonstration of a nucleoside-specific binding site." J Biol Chem 263(5): 2493-2499.

Mannock, D. A., R. N. Lewis, et al. (2001). "An analysis of the relationship between fatty acid composition and the lamellar gel to liquid-crystalline and the lamellar to inverted nonlamellar phase transition temperatures of phosphatidylethanolamines and diacyl-alpha-D-glucosyl glycerols." Eur Biophys J 30(7): 537-554.

Manting, E. H. and A. J. Driessen (2000). "Escherichia coli translocase: the unravelling of a molecular machine." Mol Microbiol 37(2): 226-238.

Maraspin, V., J. Cimperman, et al. (2002). "Solitary borrelial lymphocytoma in adult patients." Wien Klin Wochenschr 114(13-14): 515-523.

Marchler-Bauer, A. and S. H. Bryant (2004). "CD-Search: protein domain annotations on the fly." Nucleic Acids Res 32(Web Server issue): W327-331.

Mathers, D. A., W. K. Leung, et al. (1996). "The major surface protein complex of Treponema denticola depolarizes and induces ion channels in HeLa cell membranes." Infect Immun 64(8): 2904-2910.

Mehnert, W. H. and G. Krause (2005). "Surveillance of Lyme borreliosis in Germany, 2002 and 2003." Euro Surveill 10(4): 83-85.

Menestrina, G., M. Dalla Serra, et al. (2003). "Ion channels and bacterial infection: the case of beta-barrel pore-forming protein toxins of Staphylococcus aureus." FEBS Lett 552(1): 54-60.

Motaleb, M. A., L. Corum, et al. (2000). "Borrelia burgdorferi periplasmic flagella have both skeletal and motility functions." Proc Natl Acad Sci U S A 97(20): 10899-10904.

Mulay, V., M. J. Caimano, et al. (2007). "Borrelia burgdorferi BBA74, a periplasmic protein associated with the outer membrane, lacks porin-like properties." J Bacteriol 189(5): 2063-2068.

Murakami, S., R. Nakashima, et al. (2002). "Crystal structure of bacterial multidrug efflux transporter AcrB." Nature 419(6907): 587-593.

69 | P a g e

Mursic, V. P., B. Wilske, et al. (1987). "In vitro and in vivo susceptibility of Borrelia burgdorferi." Eur J Clin Microbiol 6(4): 424-426.

Nablo, B. J., K. M. Halverson, et al. (2008). "Sizing the Bacillus anthracis PA63 Channel with Nonelectrolyte Poly (Ethylene Glycols)." Biophysical Journal 95(3): 1157.

Nadelman, R. B. and G. P. Wormser (1998). "Lyme borreliosis." Lancet 352(9127): 557-565.

Nakae, T. (1975). "Outer membrane of Salmonella typhimurium: reconstitution of sucrose-permeable membrane vesicles." Biochem Biophys Res Commun 64(4): 1224-1230.

Nakae, T. and H. Nikaido (1975). "Outer membrane as a diffusion barrier in Salmonella typhimurium. Penetration of oligo- and polysaccharides into isolated outer membrane vesicles and cells with degraded peptidoglycan layer." J Biol Chem 250(18): 7359-7365.

Neuwald, A. F., J. S. Liu, et al. (1997). "Extracting protein alignment models from the sequence database." Nucleic Acids Res 25(9): 1665-1677.

Nikaido, H. (2003). "Molecular basis of bacterial outer membrane permeability revisited." Microbiol Mol Biol Rev 67(4): 593-656.

Nikaido, H. and E. Y. Rosenberg (1981). "Effect on solute size on diffusion rates through the transmembrane pores of the outer membrane of Escherichia coli." J Gen Physiol 77(2): 121-135.

Nilsson, C. L., H. J. Cooper, et al. (2002). "Characterization of the P13 membrane protein of Borrelia burgdorferi by mass spectrometry." J Am Soc Mass Spectrom 13(4): 295-299.

Nishino, K., T. Latifi, et al. (2006). "Virulence and drug resistance roles of multidrug efflux systems of Salmonella enterica serovar Typhimurium." Mol Microbiol 59(1): 126-141.

Noppa, L., Y. Östberg, et al. (2001). "P13, an integral membrane protein of Borrelia burgdorferi, is C-terminally processed and contains surface-exposed domains." Infect Immun 69(5): 3323-3334.

Nordstrand, A., I. Bunikis, et al. (2007). "Tickborne relapsing fever diagnosis obscured by malaria, Togo." Emerg Infect Dis 13(1): 117-123.

Nunn, D. (1999). "Bacterial type II protein export and pilus biogenesis: more than just homologies?" Trends Cell Biol 9(10): 402-408.

Ohnishi, J., J. Piesman, et al. (2001). "Antigenic and genetic heterogeneity of Borrelia burgdorferi populations transmitted by ticks." Proc Natl Acad Sci U S A 98(2): 670-675.

Orlik, F., C. Andersen, et al. (2002). "Site-directed mutagenesis of tyrosine 118 within the central constriction site of the LamB (maltoporin) channel of Escherichia coli. II. Effect on maltose and maltooligosaccharide binding kinetics." Biophys J 83(1): 309-321.

70 | P a g e

Orlik, F., B. Schiffler, et al. (2005). "Anthrax toxin protective antigen: inhibition of channel function by chloroquine and related compounds and study of binding kinetics using the current noise analysis." Biophys J 88(3): 1715-1724.

Östberg, Y., S. Berg, et al. (2007). "Functional analysis of a lipid galactosyltransferase synthesizing the major envelope lipid in the Lyme disease spirochete Borrelia burgdorferi." FEMS Microbiol Lett 272(1): 22-29.

Östberg, Y., M. Pinne, et al. (2002). "Elimination of channel-forming activity by insertional inactivation of the p13 gene in Borrelia burgdorferi." J Bacteriol 184(24): 6811-6819.

Pal, U., X. Li, et al. (2004). "TROSPA, an Ixodes scapularis receptor for Borrelia burgdorferi." Cell 119(4): 457-468.

Pal, U., X. Yang, et al. (2004). "OspC facilitates Borrelia burgdorferi invasion of Ixodes scapularis salivary glands." J Clin Invest 113(2): 220-230.

Pao, S. S., I. T. Paulsen, et al. (1998). "Major facilitator superfamily." Microbiol Mol Biol Rev 62(1): 1-34.

Paster, B. J. and F. E. Dewhirst (2000). "Phylogenetic foundation of spirochetes." J Mol Microbiol Biotechnol 2(4): 341-344.

Paster, B. J., F. E. Dewhirst, et al. (1991). "Phylogenetic analysis of the spirochetes." J Bacteriol 173(19): 6101-6109.

Paulsen, I. T., J. Chen, et al. (2001). "Comparative genomics of microbial drug efflux systems." J Mol Microbiol Biotechnol 3(2): 145-150.

Pautsch, A. and G. E. Schulz (1998). "Structure of the outer membrane protein A transmembrane domain." Nat Struct Biol 5(11): 1013-1017.

Pettersson, J., M. E. Schrumpf, et al. (2007). "Purine salvage pathways among Borrelia species." Infect Immun 75(8): 3877-3884.

Piesman, J., T. N. Mather, et al. (1987). "Duration of tick attachment and Borrelia burgdorferi transmission." J Clin Microbiol 25(3): 557-558.

Pinne, M., K. Denker, et al. (2006). "The BBA01 protein, a member of paralog family 48 from Borrelia burgdorferi, is potentially interchangeable with the channel-forming protein P13." J Bacteriol 188(12): 4207-4217.

Pinne, M., Y. Östberg, et al. (2004). "Molecular analysis of the channel-forming protein P13 and its paralogue family 48 from different Lyme disease Borrelia species." Microbiology 150(Pt 3): 549-559.

Pinne, M., M. Thein, et al. (2007). "Elimination of channel-forming activity by insertional inactivation of the p66 gene in Borrelia burgdorferi." FEMS Microbiol Lett 266(2): 241-249.

Pinne, M., M. Thein, et al. (2007). "Elimination of channel-forming activity by insertional inactivation of the p66 gene in Borrelia burgdorferi." FEMS Microbiol Lett 266(2): 241-249.

71 | P a g e

Polleichtner, G. and C. Andersen (2006). "The channel-tunnel HI1462 of Haemophilus influenzae reveals differences to Escherichia coli TolC." Microbiology 152(Pt 6): 1639-1647.

Poole, K. (2001). "Multidrug resistance in Gram-negative bacteria." Curr Opin Microbiol 4(5): 500-508.

Purser, J. E., M. B. Lawrenz, et al. (2003). "A plasmid-encoded nicotinamidase (PncA) is essential for infectivity of Borrelia burgdorferi in a mammalian host." Mol Microbiol 48(3): 753-764.

Radolf, J. D. (1994). "Role of outer membrane architecture in immune evasion by Treponema pallidum and Borrelia burgdorferi." Trends Microbiol 2(9): 307-311.

Radolf, J. D., K. W. Bourell, et al. (1994). "Analysis of Borrelia burgdorferi membrane architecture by freeze-fracture electron microscopy." J Bacteriol 176(1): 21-31.

Raoult, D. and V. Roux (1999). "The body louse as a vector of reemerging human diseases." Clin Infect Dis 29(4): 888-911.

Ras, N. M., B. Lascola, et al. (1996). "Phylogenesis of relapsing fever Borrelia spp." Int J Syst Bacteriol 46(4): 859-865.

Rensing, C., T. Pribyl, et al. (1997). "New functions for the three subunits of the CzcCBA cation-proton antiporter." J Bacteriol 179(22): 6871-6879.

Reumann, S., E. Maier, et al. (1998). "Permeability properties of the porin of spinach leaf peroxisomes." Eur J Biochem 251(1-2): 359-366.

Ribeiro, J. M., T. N. Mather, et al. (1987). "Dissemination and salivary delivery of Lyme disease spirochetes in vector ticks (Acari: Ixodidae)." J Med Entomol 24(2): 201-205.

Rudel, T., A. Schmid, et al. (1996). "Modulation of Neisseria porin (PorB) by cytosolic ATP/GTP of target cells: parallels between pathogen accommodation and mitochondrial endosymbiosis." Cell 85(3): 391-402.

Sadziene, A., B. Wilske, et al. (1993). "The cryptic ospC gene of Borrelia burgdorferi B31 is located on a circular plasmid." Infect Immun 61(5): 2192-2195.

Saier, M. H., Jr. (2000). "Families of proteins forming transmembrane channels." J Membr Biol 175(3): 165-180.

Saier, M. H., Jr. (2000). "A functional-phylogenetic classification system for transmembrane solute transporters." Microbiol Mol Biol Rev 64(2): 354-411.

Saier, M. H., Jr. and I. T. Paulsen (2001). "Phylogeny of multidrug transporters." Semin Cell Dev Biol 12(3): 205-213.

Saint, N., K. L. Lou, et al. (1996). "Structural and functional characterization of OmpF porin mutants selected for larger pore size. II. Functional characterization." J Biol Chem 271(34): 20676-20680.

72 | P a g e

Sallmann, F. R., S. Baveye-Descamps, et al. (1999). "Porins OmpC and PhoE of Escherichia coli as specific cell-surface targets of human lactoferrin. Binding characteristics and biological effects." J Biol Chem 274(23): 16107-16114.

Sandkvist, M. (2001). "Biology of type II secretion." Mol Microbiol 40(2): 271-283.

Schindler, H. and J. P. Rosenbusch (1978). "Matrix protein from Escherichia coli outer membranes forms voltage-controlled channels in lipid bilayers." Proc Natl Acad Sci U S A 75(8): 3751-3755.

Schirmer, T. (1998). "General and specific porins from bacterial outer membranes." J Struct Biol 121(2): 101-109.

Schirmer, T., T. A. Keller, et al. (1995). "Structural basis for sugar translocation through maltoporin channels at 3.1 A resolution." Science 267(5197): 512-514.

Schroder, N. W., J. Eckert, et al. (2008). "Immune responses induced by spirochetal outer membrane lipoproteins and glycolipids." Immunobiology 213(3-4): 329-340.

Schulz, G. E. (2004). "The structure of general porins." WILEY-VCH Verlag GmbH&Co. KGaA, Weinheim, Germany.

Schwan, T. G., J. M. Battisti, et al. (2003). "Glycerol-3-phosphate acquisition in spirochetes: distribution and biological activity of glycerophosphodiester phosphodiesterase (GlpQ) among Borrelia species." J Bacteriol 185(4): 1346-1356.

Schwan, T. G., J. Piesman, et al. (1995). "Induction of an outer surface protein on Borrelia burgdorferi during tick feeding." Proc Natl Acad Sci U S A 92(7): 2909-2913.

Schwan, T. G., M. E. Schrumpf, et al. (1996). "GlpQ: an antigen for serological discrimination between relapsing fever and Lyme borreliosis." J Clin Microbiol 34(10): 2483-2492.

Setubal, J. C., M. Reis, et al. (2006). "Lipoprotein computational prediction in spirochaetal genomes." Microbiology 152(Pt 1): 113-121.

Shang, E. S., M. M. Exner, et al. (1995). "The rare outer membrane protein, OmpL1, of pathogenic Leptospira species is a heat-modifiable porin." Infect Immun 63(8): 3174-3181.

Shang, E. S., J. T. Skare, et al. (1998). "Isolation and characterization of the outer membrane of Borrelia hermsii." Infect Immun 66(3): 1082-1091.

Sharff, A., C. Fanutti, et al. (2001). "The role of the TolC family in protein transport and multidrug efflux. From stereochemical certainty to mechanistic hypothesis." Eur J Biochem 268(19): 5011-5026.

Shaw, N. (1970). "Bacterial glycolipids." Bacteriol Rev 34(4): 365-377.

73 | P a g e

Sickmann, A., M. Mreyen, et al. (2002). "Identification of modified proteins by mass spectrometry." IUBMB Life 54(2): 51-57.

Skare, J. T., C. I. Champion, et al. (1996). "Porin activity of the native and recombinant outer membrane protein Oms28 of Borrelia burgdorferi." J Bacteriol 178(16): 4909-4918.

Skare, J. T., T. A. Mirzabekov, et al. (1997). "The Oms66 (p66) protein is a Borrelia burgdorferi porin." Infect Immun 65(9): 3654-3661.

Skare, J. T., E. S. Shang, et al. (1995). "Virulent strain associated outer membrane proteins of Borrelia burgdorferi." J Clin Invest 96(5): 2380-2392.

Sonenshine, D. E. (1997). "Biology of Ticks." Oxford University Press. Song, L., M. R. Hobaugh, et al. (1996). "Structure of staphylococcal alpha-

hemolysin, a heptameric transmembrane pore." Science 274(5294): 1859-1866.

Southern, P. M. J. and J. P. Sanford (1969). "Relapsing fever: A Clinical and Microbiological Review." Medicine 48(2): 129-150.

Stanek, G. and F. Strle (2003). "Lyme borreliosis." Lancet 362(9396): 1639-1647. Steere, A. C. (1989). "Lyme disease." N Engl J Med 321(9): 586-596. Steere, A. C. (2001). "Lyme disease." N Engl J Med 345(2): 115-125. Steere, A. C. (2006). "Lyme borreliosis in 2005, 30 years after initial observations

in Lyme Connecticut." Wien Klin Wochenschr 118(21-22): 625-633. Steere, A. C., S. E. Malawista, et al. (1977). "Lyme arthritis: an epidemic of

oligoarticular arthritis in children and adults in three connecticut communities." Arthritis Rheum 20(1): 7-17.

Steere, A. C., V. K. Sikand, et al. (1998). "Vaccination against Lyme disease with recombinant Borrelia burgdorferi outer-surface lipoprotein A with adjuvant. Lyme Disease Vaccine Study Group." N Engl J Med 339(4): 209-215.

Stegmeier, J. F., G. Polleichtner, et al. (2006). "Importance of the adaptor (membrane fusion) protein hairpin domain for the functionality of multidrug efflux pumps." Biochemistry 45(34): 10303-10312.

Stoenner, H. G., T. Dodd, et al. (1982). "Antigenic variation of Borrelia hermsii." J Exp Med 156(5): 1297-1311.

Strle, F., R. N. Picken, et al. (1997). "Clinical findings for patients with Lyme borreliosis caused by Borrelia burgdorferi sensu lato with genotypic and phenotypic similarities to strain 25015." Clin Infect Dis 25(2): 273-280.

Strother, K. O. and A. de Silva (2005). "Role of Borrelia burgdorferi linear plasmid 25 in infection of Ixodes scapularis ticks." J Bacteriol 187(16): 5776-5781.

74 | P a g e

Struyve, M., M. Moons, et al. (1991). "Carboxy-terminal phenylalanine is essential for the correct assembly of a bacterial outer membrane protein." J Mol Biol 218(1): 141-148.

Sulavik, M. C., C. Houseweart, et al. (2001). "Antibiotic susceptibility profiles of Escherichia coli strains lacking multidrug efflux pump genes." Antimicrob Agents Chemother 45(4): 1126-1136.

Takayama, K., R. J. Rothenberg, et al. (1987). "Absence of lipopolysaccharide in the Lyme disease spirochete, Borrelia burgdorferi." Infect Immun 55(9): 2311-2313.

Tamura, N., S. Murakami, et al. (2005). "Direct interaction of multidrug efflux transporter AcrB and outer membrane channel TolC detected via site-directed disulfide cross-linking." Biochemistry 44(33): 11115-11121.

Thanabalu, T., E. Koronakis, et al. (1998). "Substrate-induced assembly of a contiguous channel for protein export from E.coli: reversible bridging of an inner-membrane translocase to an outer membrane exit pore." EMBO J 17(22): 6487-6496.

Thanassi, D. G., C. Stathopoulos, et al. (2002). "Bacterial outer membrane ushers contain distinct targeting and assembly domains for pilus biogenesis." J Bacteriol 184(22): 6260-6269.

Thein, M., I. Bunikis, et al. (2008). "Oms38 is the first identified pore-forming protein in the outer membrane of relapsing fever spirochetes." J Bacteriol 190(21): 7035-7042.

Tikhonova, E. B., Q. Wang, et al. (2002). "Chimeric analysis of the multicomponent multidrug efflux transporters from gram-negative bacteria." J Bacteriol 184(23): 6499-6507.

Tilly, K., A. F. Elias, et al. (2001). "Genetics and regulation of chitobiose utilization in Borrelia burgdorferi." J Bacteriol 183(19): 5544-5553.

Tilly, K., D. Grimm, et al. (2004). "Infectious cycle analysis of a Borrelia burgdorferi mutant defective in transport of chitobiose, a tick cuticle component." Vector Borne Zoonotic Dis 4(2): 159-168.

Trias, J. and R. Benz (1994). "Permeability of the cell wall of Mycobacterium smegmatis." Mol Microbiol 14(2): 283-290.

Tseng, T. T., K. S. Gratwick, et al. (1999). "The RND permease superfamily: an ancient, ubiquitous and diverse family that includes human disease and development proteins." J Mol Microbiol Biotechnol 1(1): 107-125.

Vogel, H. and F. Jahnig (1986). "Models for the structure of outer-membrane proteins of Escherichia coli derived from raman spectroscopy and prediction methods." J Mol Biol 190(2): 191-199.

von Lackum, K. and B. Stevenson (2005). "Carbohydrate utilization by the Lyme borreliosis spirochete, Borrelia burgdorferi." FEMS Microbiol Lett 243(1): 173-179.

75 | P a g e

Walker, E. M., L. A. Borenstein, et al. (1991). "Analysis of outer membrane ultrastructure of pathogenic Treponema and Borrelia species by freeze-fracture electron microscopy." J Bacteriol 173(17): 5585-5588.

Wandersman, C. and P. Delepelaire (1990). "TolC, an Escherichia coli outer membrane protein required for hemolysin secretion." Proc Natl Acad Sci U S A 87(12): 4776-4780.

Wang, G., C. Ojaimi, et al. (2002). "Disease severity in a murine model of lyme borreliosis is associated with the genotype of the infecting Borrelia burgdorferi sensu stricto strain." J Infect Dis 186(6): 782-791.

Wheeler, C. M., J. C. Garcia Monco, et al. (1993). "Nonprotein antigens of Borrelia burgdorferi." J Infect Dis 167(3): 665-674.

Wilske, B. (2005). "Epidemiology and diagnosis of Lyme borreliosis." Ann Med 37(8): 568-579.

Wilske, B., S. Jauris-Heipke, et al. (1995). "Phenotypic analysis of outer surface protein C (OspC) of Borrelia burgdorferi sensu lato by monoclonal antibodies: relationship to genospecies and OspA serotype." J Clin Microbiol 33(1): 103-109.

Wilske, B., V. Preac-Mursic, et al. (1993). "An OspA serotyping system for Borrelia burgdorferi based on reactivity with monoclonal antibodies and OspA sequence analysis." J Clin Microbiol 31(2): 340-350.

Wolgemuth, C. W., N. W. Charon, et al. (2006). "The flagellar cytoskeleton of the spirochetes." J Mol Microbiol Biotechnol 11(3-5): 221-227.

Wormser, G. P., R. B. Nadelman, et al. (2000). "Practice guidelines for the treatment of Lyme disease. The Infectious Diseases Society of America." Clin Infect Dis 31 Suppl 1: 1-14.

Yen, M. R., C. R. Peabody, et al. (2002). "Protein-translocating outer membrane porins of Gram-negative bacteria." Biochim Biophys Acta 1562(1-2): 6-31.

Yokota, M., M. G. Morshed, et al. (1997). "Protective activity of Borrelia duttonii-specific immunoglobulin subclasses in mice." J Med Microbiol 46(8): 675-680.

Yoneyama, H., H. Maseda, et al. (2000). "Function of the membrane fusion protein, MexA, of the MexA, B-OprM efflux pump in Pseudomonas aeruginosa without an anchoring membrane." J Biol Chem 275(7): 4628-4634.

Young, G. M., D. H. Schmiel, et al. (1999). "A new pathway for the secretion of virulence factors by bacteria: the flagellar export apparatus functions as a protein-secretion system." Proc Natl Acad Sci U S A 96(11): 6456-6461.

Yu, E. W., G. McDermott, et al. (2003). "Structural basis of multiple drug-binding capacity of the AcrB multidrug efflux pump." Science 300(5621): 976-980.

76 | P a g e

Yu, S., H. Ding, et al. (1998). "Characterization of a phage specific to hemorrhagic Escherichia coli O157:H7 and disclosure of variations in host outer membrane protein ompC." J Biomed Sci 5(5): 370-382.

Zgurskaya, H. I. and H. Nikaido (1999). "AcrA is a highly asymmetric protein capable of spanning the periplasm." J Mol Biol 285(1): 409-420.

Zgurskaya, H. I. and H. Nikaido (2000). "Cross-linked complex between oligomeric periplasmic lipoprotein AcrA and the inner-membrane-associated multidrug efflux pump AcrB from Escherichia coli." J Bacteriol 182(15): 4264-4267.

Zhong, J., S. Skouloubris, et al. (2006). "Function and evolution of plasmid-borne genes for pyrimidine biosynthesis in Borrelia spp." J Bacteriol 188(3): 909-918.

77 | P a g e

Publications I-V

borrelia channel-forming proteins: structure and function

Documents