Res. MicrobioL

© INSTITUT PASTEUR/ELSEVIER Paris 1992

1992, 143, 605-613

Genetic approaches to cell biology and metabolism of spirochetes C . W . P e n n (1) (*), p . j .

B a s s f o r d (2) (*) ( , ) , D . B . Y e l t o n (3), j . D u n n (4), D . R , N e l s o n (5), M. F u k u n a g a (6) a n d G. S t a n e k (7)

eli School of Biological Sciences, University of Birmingham, Birmingham BI5 2TT (UK), ~2)Department of Microbiology and Immuaology, University of North Carolina, Chapel Hill, NC 27514 (USA). 13) 2095 Health Sciences Center, Department of Microbiology and Immunology, West Virginia University, Morgantown, WV26506 (USA), r4j Department of Biology, Brookhaven Laboratory, Upt3n, N Y 11973 (USA), rs) Department of Microbiology, U~iversity of Rhode Island, 318 Morrill Science Building, Kingston, RI028810812 (USA), 16) Faculty of Pharmacy and Pharmaceutical Sciences, University of Fukuyama, 985 Higashimura-Cho, Fukuyama, Hiroshima 729-02 (Japan), and 17) Hygiene Institut der Univer~itat Wien, Kinderspitalgasse 15, Vienna (Austria)

SUMMARY

Genetic analysis and methodology have only comparatively recently been applied to the study of spirochetes. Although genetic transfer procedures for spirochetes are not widely available, there are several examples of progress in genetic analysis of spirochetes by other approaches. Some examples of these approaches are the following. 1 ) Genes for ~ynthetic pathways in Treponema and Leptospira have been cloned by oomplementatlo,~ of Escherichia coil serving as plasmid hosts. 2) The OspA protein of Borrelia burgdorferi has been overexpressed in E. coil without the signal peptide; the recombinant product has been suitable for circular dichroism as well as other biochemical analyses. 3) The heat shock proteins of B. burgdorferi are homologous to heat shock proteins of E. coIL 4) Enzyme activity profiles of B. burgdorferi and other spirochetes show strain heterogeneity and also indicate which biosynthetic and enzymatic activities are conserved within different spirochetes. 5) The gene organization of rRNA genes have revealed differences between spirochetes and other types of bacteria.

Key-words: Spirochete, Genetics, Biosynthesis; Pathways, Metabolism, Heat shock proteins, Enzymatic profiles, RNA.

(*) Chairpersonsand authors of the present summary of session, in collaboration with prescI:tingparticipantsat the "Molecular Biology of Spirochetes'" meeting. October 1991, Annecy, France. ('1"~Re~:ently deceased.

C. IV. PENN E T AL.

606

other hand differences between species, for example, in comparisons with T. denticola, are very significant. Bassford's group was, however, able to find a very marked difference between strains by radiolabelling o f proteins in intact organisms in the presence of erythromycin. The incorporation of 35Smethionine into proteins was drastically inhibited by erythromycin at 10-100 ~.g per ml in both Nichols and Gauthier strains, whereas label!~ng in Street Strain i4 was not inhibited in the presence of up to 400 izg per ml (Stamm et ol., 1988). This result was very significant ; erythromycin is the second line drug of choice recommended for syphilis tleatment where penicillin is not appropriate, and this strain was isolated by J. Clark of tile Centres for Disease Control in 1977 from a patient in w h o m erythromycin therapy had failed.

Introduction Genetic analysis of spirochetes, as for other bacteria, can be a powerful tool for understanding the basic and metabolic functions of these organisms. For the Enterobacteriaceae, Pseudomonads, and other easily cultured, metabolically versatile organisms, the analysis of mutants such as auxotrophs has been a central and highly versatile tool. In contrast, we have not been able to isolate or manipulate such mutants of spirochetes to this extent. Nevertheless, it has recently been possible to isolate cloned genes frc,n spirochetes and to learn from the recombinants aonut the metabolic capabilities of these organisms. Furthermore, by enhancing expression of spirochetal genes in more pliable and "user-friendly" host organisms, it h i s been possible to ~ , r y out ~iglfii'icant studies on the structure and function of spirochetal protein which would not otherwise have been available ia sufficient quantity for this type of analysis. Cloniag techniques and new methods for genomic analysis have also made it possible to m a p the location oY important genes, even of organisms that are poorly or not at all cultivable. At the same time, methods for the enzymological study of such organisms have also been refined to make it possible to derive metabolic profiles, again even for the most fastidious spirochetes. In this section, the application of these new technologies to the exploration of metabolic and cell biological attributes of spirochetes is reviewed.

We c.oald also make detailed comparisons between strains in their D N A sequences. Bassford and colleagues compared the tmpC sequence determined in van E m b d e n ' s laboratory with that of a homologous clone from Street Strain 14 (Hubbard et aL, 1991), and found that differences were very minor indeed - - just two conservative base substitutions. This suggests that little mutation is occurring in the organism (the Nichols strain was isolated about 65 years earlier), and the exceUeat conservation in this sequence suggests it could be a good target for P C R amplification from diverse isolates for use in diagnosis. Another approach in which geaetic methods could be exploited was gene cloning, which has been done in many laboratories. For most of the genes which have been cloned, we do not know the functions o f the proteins encoded ; one exception is the flageUins. We can, however, enhance the value of the approach by looking for function, for example in the complementation of auxotrophies. In a limited search for such complementation, Bassford's group identified T. pallidum genes for proline biosynthesis. In E. colt, the synthesis of proline requires three enzymes encoded by the proA, proB and proC genes. Frank Gherardini, a postdoctoral associate in Bassford's laboratory, identified cosmid clones complementing null mutations in the E. colt proA and proC genes, but not proB (Gherardini et al., 1990). The complementing activity for proC was localized to a 2.3-kb fragment that encoded proteins of 41 and 27 kDa. The 27-kDa protein was responsible for the proC c o m p l e m e n t i n g activity. T h e T. pallidum 1-pyrroline-5-carboxylate (P5C) reductase had considerable homology at the C-terminal end with that

The genetics of Treponema pallidum Philip Bassford described his first involvement with genetic approaches to the study of Treponema pallidum, as an Escherichia colt geneticist. His first reaction was " D o e s this organism really have genetics?" We could not culture it, there were no clonal populations, and no mutants could be obtained. (The obvious approach to clone and express genes o f T. pallidum in E. colt might be referred to as " p u n k genetics" I) However, there were some other things which could be done and which have illuminated some aspects of this organism. First, we could compare strains. Few new strains are available, but comparisons of protein profiles by one-dimensional SDS-PAGE between the Nichols strain, Street Strain 14, and the Gauthier yaws isolate of subspecies pertenue showed that they were very similar with only minor differences. The significance of these small differences is elusive; on the

Hsp ORF

= heat shock protein. = open reading frame.

] Ic

PBS

= phosphate-buffered saline.

GENETIC APPROACH~,S TO CELL B I O L O G Y A N D M E T A B O L I S M OF SPIROCHETES

6O7

of the P5C reductase cloned from a limited nt~mber of other organisms including E. coil

The leuB gene of L. interrogans

Since the genome o f T. pallidum is small, about one million base pairs (Walker et aL, 1991), one w~uld expect to find limited biosynthetic capabilities. Bassford questioned whether it was surprising that the organism appears capable of synthesizing proline. H u m a n blood contains no free proline, and thus he speculated that treponemes m~y need to synthesize proline and other amino acids during infection. In further exploitation of the auxotropb complementation approach, Bassfc rd's group identified serB, but could not complement E. co!i mutations in three additional biosynthetic pathways. Some other genes were identified using a related approach. By screening ,alidixic-acid-re.qstant E. coli containing T. pallidum DNA sequences, sensitive clones were identified that contained the gyrB gene which was adjacent t4~ gyrA. Degradative enzyme activities were also sought. While protease or lipase activities were not obtained, DNase activity was identified. A number o f antigens in addition to those so far reported were also cloned.

David Yelton reported cloning of the leuB gene from L. intcrrogans serovar pomona strain kennewicki into E. coil The gene was isolated from a clone bank following partial EcoR1 digestion of Leptospira DNA and ligation into a pUCI3 vector. The information needed to complement an E. coil leuB defect was carried on a 6.8 kDa fragment of lepto.;piral DNA, This fragment was unable to complem : n t defects in other leucine biosynthetic genes of E. coll. Subcloning and Tn5 mutagenesis showed that the information needed to complement several independent leuB defects in E. coil was contained within a 1,2-kb region of the cloned leptospiral DNA. Southern blot ana!ysi~ of ,V.enne,;.icki DNA, using this cloned fragment as a probe, revealed only a single DNA fragment having homology with the leuBcomplementing region. These results indicated thin the activity which was cloned represented the leuB gene of L. interrogans and not some pseudogene which was able to complement the leuB defect found in E. coll.

Another interesting activity cloned from Leptospira biflexa, by functional complementation in E. coil, was RecA (Stamm et al., 1991a). Antiserum to the RecA protein of E. coil also precipitated the Leptospira protein from both saprophytic and pathogenic strains; its expression was induced by agents known to activate the SOS response. No homologue o f this protein was found in T. pallidum, either by antibody reactivity or by gene cloning. While, as a negative result, this is of limited value, nevertheless it leads to interesting speculation; it seems rather odd, and suggests a very limited capability o f the organism for genetic recombination, for example in the acquisition o f genes from other sources.

A 3.l-kb EcoR! fragment containing the leuB complementing region was subcloned from the original plasm,.'d. Using maxicell analysis (Sancar et al., 1979), Yelton and colleagues found that three proteins were encoded by this DNA fragment. Both strands of this entire region have been sequenced (Sanger et al., 1977). Three open reading frames (ORF) were identified. These ORF encode proteins of 40.9, 38.8 and 15.0 kDa. Subclones, containing deletions of various sizes, were examined for t_heir ability to coral:dement a leuB mutation in E. coil This analysis showed that the 38.8-kDa protein was required for complementation and that the other two proteins were not required for complementation.

Loin Stamm had also looked at the heat shock response in various spirochetes (Stamm et al., 1991b). With the exception of 7". pallidum, all those examined which encompassed L. biflexa, L. interrogans, T. phagedenis, T. denticola and B. burgdorferi showed heat shock responses, defined as incorporation of 35S-methionine into novel proteins following elevation of temperature from 30°C to 37°C ~r 42°C. Temperatures ab.we 36°C soon stopped protein synthesis in T. pallidum. Homologues of two major heat shock proteins, GroEL and DnaK, were identified by radioimmunoprecipitiation in the various spirochetes, including f. paJlidum. Finally, a student in Bassford's group, John Hardham, had done some random sequencing of clones from T. pallidum using vector primers, to see what could be identified as sequences homologous with known sequences in the databases. While several homologues were found, for example, a DNA polymerase E, an enolase, and a tetrahydrofolate pathway enzyme, no great surprises emerged.

A search of the databases revealed that the 38.8 kDa protein showed significant amino acid homology with the 3-isopropylmalate dehydrogenases cloned from six other microorganisms. Homologies ranged from 69 to 79 percent. Enzyme assays showed that the cloned gene was a beta-isop~opylmalate dehydrogenase. The 15.0-kDa protein had a very basic isoelectric point and showed limited homology to a variety of DNA-binding proteins. The 40.9-kDa protein was found to show approximately 60 % homology with the ornithine-oxo-acid aminotransferases isolated from both h u m a n and S. cerevisiae. The genes ei~coding these proteins were organized as follows: ORF-I (40.9 kDa protein, putative o r n i t h i n e - o r o - a c i d a m i n o t r a n s f e r a s e ) ; ORF-2 (38.8 k D a protein, b e t a - i s o p r o p y l m a l a t e dehydrogenase, leuB gene); ORF-3 (l 5.0-kDa protein). All of the ORF were transcribed in the same direction starting with ORF-1 and proceeding towards ORF-3. Thus the organization of the leptospir.q! gcne~ involved in leucine biosynthesis is dramatically differ-

6O8

C. W. PENN E T AL.

ent from that found i~aE. co~i (Somers et al., 1973) or Bacdlus subtilis (Ward and Zahler, 1973), where these genes are grouped together in an operon having the following organization : regulatory elements leuA -leuB-leuC-leuD.

Collins, CO, they are testing the purified OspA-257 to see how well it can protect hamsters against a natural challenge of B. burgdorferi spirochetes, i.e. allowing B. burgdorferi-infected ticks (nymphal !xodes damminO to feed on immunized animals.

Regulation of the leuB gene of L. interrogans was investigated. Cultures were grown in either EMJH medium or EMJH medium to which 100 i~g/ml of either L-leucine, L-isoleucine or L-valine had been added. Leptospira cells were grown to mid-log phase and total cellular RNA was extracted. The RNA was used for Northern blot analysis using a probe derived from the cloned leuB gene. All of the cultures contained leuB.specific RNA in approximately equal amounts. To confirm these findings, enzyme assays were performed on cellular extracts prepared by sonication of mid-log phase cells grown in the various media. Debris was removed by centrifugation and the st~pe~ znt fluids were assayed for betaisopropy~malate dehydrogenase activity. Aii of the extracts were found to contain enzyme. The specific ac,ivities of the extracts were all approximately the same (about 0.05 Izmoles of product/min/~.g protein). These data suggest that the leuB gene of L. interrogans is not subject to regulation at either the transcriptional or the translational level.

Dunn and his coworkers are currently attempting to crystallize OspA in order to determine its X-ray structure. In the interim, they have used circular dichroism to characterize the secondary structure and study conformational changes in OspA protein. The results showed that OspA contains mostly beta-sheet (32 070 anti-parallel, 10 o70 parallel) configurations, beta-turn (18 %) and random coil (30 %), but little alpha-helix (11 o70).These values are quite different from those obtained using purely predictive methods based solely on the amino acid sequence of OspA, which predict a much higher percentage of alpha-helix.

High level expression and analysis of B. burgdorferi surface proteins John Dunn described how the ospA gene from a North American strain ofB. burgdorferi, B31, had been cloned under the control of transcription and translation signals from bacteriophage T7. The full length, 273 amino acid (31 t,Da) OspA lipoprotein was expressed poorly in E. coli and was associated with the insoluble membrane fraction. The low level of expression presumably is due to the accumulated toxic effects of OspA localizing at the E. coil cell membrane during expression and, as expected, the fu~l-length protein remained associated with the E. coil membrane fraction unless solubilized by detergents. In contrast, a truncated form of OspA lacking the amino-terminal signal sequence which normally would direct localization of the protein to the membrane was expressed at very high levels (at least 100 mg/litre or greater than 50 070 of the total protein), was highly soluble in aqueous solution (at least 50 mg/ml) and eluted from a gel f~Itt'ation column as a monomer under non-denaturing conditions. Dunn and colleagues developed an efficient procedure for purifying large amounts of recombinant OspA to homogeneity and tile molar extinction coefficient at 280 nm of purified OspA was determined, simplifying quantitation in immunological testing (Dunn et al., 1990). Together with B. Johnson at the Center3 for Disease Control in Fo~

Dunn's group has also used steady-state and timeresolved fluorescence to investigate the static and dynamic aspects of OspA conformation. OspA contains a single tryptophan residue which can be selectively excited at wavelengths > 295 nm and the fluorescence emission is particularly sensitive to the local environment. The fluorescence signal provides a probe for changes in protein conformation, since such changes are expected to cause changes in the environment of tryptophan. The emission peak (330 nm) of OspA and its response to various ionic and anionic quenchers indicated that OspA's tryptophan is buried within the native protein in a relatively hydrophobic environment; however, it is completely exposed to the solvent, as expected for a random coil, when OspA is denatured by high temperature (at least 80°C) or guanidine-HCI. Both CD and fluorescence measurements revealed that the native conformation of OspA was highly stable; no significant changes were seen from pH 3-11, and the protein was stable at low and high salt concentrations. The protein returned to its native state, as reflected by CD and fluorescent measurements, after being held at 100°C for up to 10 min. Interestingly, the truncated OspA protein was exceptionally resistant to digestion by trypsin and human plasmin, even though it is rich in lysine residues (41 residues, 16 mol 070) which are distributed rather uniformly throughout the protein. OspA has only two arginines, so the number of residues with positive charge at physiological pH is 44 (including the N-terminus). The recombinant protein contains 23 glutamic and 18 aspartic acids; so the total number of residues carrying a negative charge (42 at physiological pH including the C-terminus) nearly balances the number of positive charges. A substantial number (28) of these charged residues are located immediately adjacent to, or one residue away from, a residue of opposite charge. The overall pattern of positive and negative charges suggests that these positive and

GENETIC A P P R O A C H E S TO CELL B I O L O G Y A N D M E T A B O L I S M OF SPIROCHETES negative charges might form "salt bridges" that play a substantial role in stabilizing the native protein (France et al., 1992). Studies underway with OspB and another surface protein from B. burgdorferi, as well as the Vmp7 protein from the related spirochete, B. hermsii, led Dunn and colleagues to believe that a cloning strategy similar to that used to express the truncated Osl~A can provide a facile method for producing variations of Borrelia lipoproteins which are highly expressed in E. coil and solnble without exposure to detergents. Initial CD studies with purified Vmp7 protein indicated that it had a spectrum which is characteristic of proteins rich in alpha-helices. It will be interesting to determine whether this observed difference in secondary structure between OspA and Vmp7 plays a critical role in the immunogenicity of these proteins. Heat shock response and antigens of B. burgdorferi

David Nelson commenced by describing how, when cells experience an elevation in temperature, they respond by ceasing growth and in a transient manner, preferentially synthesize a small number of proteins termed heat shock (or stress) proteins (Hsp). This response to heat stress is termed the heat shock response (Lindquist, 1986). Other stress response proteins may be induced by a variety of environmental stresses such as exposure to alcohols, oxidants, abnormal proteins, and the inflammatory response. Stress proteins help the stressed cell to survive and recover from stress (Lindquist, 1986). Some Hsp, particularly members of the Hsp70 (or DnaK) and Hsp60 (or GroEL) families are highly conserved at the protein and DNA sequence levels. Members of these two families of proteins function as protein chaperones, aiding in protein folding and unfolding, as well as protein transport and secretion. Recent studies in a number of organisms have demonstrated that some Hsp are excellent antigens and may be immunodominant (Shinnick, 1991). This appears to be particularly true for Hsp60 and Hsp70. For example, two of the five or six immunodominant antigens of Mycobacterium leprae and M. tuberculosis are homologs of Hsp60 and Hsp70. Additionally, studies by several different investigators have linked some autoimmune diseases, such as rheumatoid arthritis, Behcet's syndrome, and insulindependent diabetes, with autoimmune recognition of Hsp60 (Shinnick, 1991 ; Winfield and Jarjour, 1991). These observations have suggested that many autoimmune disease symptoms may be associated with chronic bacterial infections and autoimmr~ne recognition of Hsp60 homologs (Winfield and Jarjour, 1991). Nelson hypothesized that, since Lyme disease is a chronic bacterial infection with autoimmune-like

609

pathologies, which include arthritis and various neuropathies, studies characterizing the heat shock response and Hsp of B. burgdorferi may help to elucidate the nature of Lyme disease and its relationship to autoimmunity. His approach was to examine the heat shock response of B. burgdorferi in order to identify and characterize the major Hsp. He then determined whether any Hsp was an antigen in infected humans and, if so, whether any antigenic Hsp could serve as autoimmunogens. The heat shock ~esponne of B. butgdu~feii

To test the hypothesis that some B. burgdorferi Hsp are antigens recognized by the infected host, Hsp produced by the bacterium had to be identified. Five to seven Hsp were detected by SDS-PAGE and fluorography of proteins from cells labelled with 35Smethionine after shifts from 33°C to 37°C or 40°C or from 20°C to 33°C, 37°C or 40°C. Ana!ysis of 35S-methionine-labelled Hsp by two-dimensional electronhoresis and autoradiography revealed 12 Hsp. Western immunoblot analysis with antisera against the highly conserved Hsp60 (GroEL) and Hsp70 (DnaK) proteins revealed a single heat-shockinducible 72 K protein band that reacted with antibodies to E. coli DnaK a::a with antibodies to the M. tuberculosis 71 kDa Hsp homoiog of DnaK (Carreiro et al., 1990). Two heat-shock-inducible proteins at 66 kDa and at 60 kDa reacted with antibodies against the M. tuberculosis 65 kDa Hsp homolog of E. coli GroEL (Carreiro et al., 1990). These data indicate that B. burgdorferi cells do produce Hsp homologous to the highly conserved DnaK and GroEL. Immune recognition of B. burgdorferi and human Hsp60 Determinations of whether any B. burgdorferi Hsp are antigens in infected Lyme disease patients were carried out by a combination of Western blot analysis and autoradiography. Human immune sera collected from six Lyme disease patients reacted with both the 60- and 66-kDa Hsp, but failed to react with the 72-kDa Hsp (Carreiro et al., 1990). These data indicated that the two GroEL-like Hsp are immunodominant antigens. These findings suggested the possibility that either of these Hsp60 homologs could trigger an autoimmune recognition of the human Hsp60 homolog. In order to detect autoimmune recognition of human Hsp60 by Lyme disease patient sera, six hightitre patient serum samples were pooled and used to probe Western b~ots of proteins from normal and heat-shocked human (THP-1) cells (Girouard et al., 1992). hlent:.cal protein blots were also probed with

C. W. PENN E T AL.

610

pooled sera from healthy controls and with antiGroEL antibody. The pooled patient sera reacted most intensely with a 65-kDa band that also reacted with the anti-OroEL antibody. In both cases, staining was most intense in the heat-shocked cell lysate. Control sera failed to react with the "t'HP-I cell lysate. To determine how widespread was the presence of autoreactive antibodies against human Hsp60 in Lyme disease patients, sera from 22 high-titre patients were tested for their ability to react with a recombinant form of human Hsp60 and with B. burgdorferi cell lysates (Girouard et aL, 1992). Ali Lyme disease patient sera reacted strongly with B. burgdorferi proteins. The two Hsp60 homologs of B. burgdorferi were recognized by > 90 % of the serum samples. Flagellin reacted with almost 70 % of the sera, while OspA and OspB we~'erecognized by 50 % and 63 °70of the patient sera, respectively. When the same patient sera were used to probe blots of human Hsp60, over 80 070of the sera reacted with this protein. In the control group, 50 % of the sera showed reactivity against some component of B. burgdorferi and only 28 % of the sera reacted with human Hsp60. These data show a significant correlation between the recognition of B. burgdorferi Hsp60 proteins and the autorecognition of human Hsp60 by Lyme disease patient sera. Further, this association of autoreactive antibodies to the presence of Lyme disease was significant when compared to control sera from a group of healthy volunteers. To investigate the antigenic relationship of B. burgdorferi Hsp60 proteins to the human Hsp60, Western blots of B. burgdorferi proteins and of human Hsp60 were probed with a panel of polyclonal and monoclonal antibodies raised against hsp60 homologs. These studies indicated that the B. burgdorferi Hsp60 homologs and the human Hsp60 do possess shared epitapes (Girouard et al., 1992). That is, several mAb raised against bacterial Hsp60 pro-

teins react with both B. burgdorferi Hsp60 and human Hsp60. Conversely, a mAb raised against human Hsp60 also recognized the B. burgdorferi Hsp60 homologs. Taken together, the data described by Nelson demonstrated a correlation between the presence of autoreactive antibodies to human Hsp60 and Lyme disease and also suggested the possibility that eutoimmune reactions play a role in the development of Lyme disease. Further, the presence of' shared epitopes on human and B. burgdorferi Hsp60 proteins indicated the possible role of molecular mimicry in the development of autoreactive antibodies. Finally, Nelson pointed out th~.t while he and colleagues observed autoreactivity to human Hsp60, they did not know the nature of the reactive epitopes and it was not clear wh~'ther autoreactive antibodies play a role in th." pathology of Lyme disease.

Enzyme activities of Borrelia strains Gerold Stanek described how B. burgdorferi strains have been studied by various methods including analysis with monoelonal antibodies, plasmid analysis and comparison of the genome. Little was known, however, about enzyme activities of spirochetes, particularly of B. burgdorferi. Stanek and colleagues (Mammad Manafi, Cornelia Koch, unpublished) used the standard biochemical and enzymatic procedures (Amato et al., 1991 ; Manafi et aL, 1991) which have been developed by API systems (SA, Montalieu-Vercieu, France) for the rapid identification and characterization of microorganisms, and summarized the results of an extensive enzymatic characterization of 10 Borrelia strains using API research kits. Table I shows the designation and origin of the ten Borrelia strains used, including nine strains of B. burgdorferi and one strain of B. hermsii. SDS-

Table !, Designation and origin of the 10 Borrelia strains used. Strain

B. burgdorferi B. burgdorferi B. burgdorferi B. burgdorferi B. burgdorferi B. burgdorferi B. burgdorferi B. burgdorferi B. burgdorferi B. hermsfi

Designation B31 VL3 VT1 HT2 HBL1 VHI VSI VT2 W BH

Origin Tick (L dammini) Cerebrospinal fluid (human) Tick (1. ricinus) Skin (mouse) Bladder wall (mouse) Myocardial biopsy (human) Skin (ACA) Tick (1. ricinus) Cerebrc~spinal fluid (human)

Country USA Austria Austria Austria Austria Austria Austria CFSR Austria US~,

GENETIC A P P R O A C H E S TO CELL B I O L O G Y A N D METABOLISJ~4 OF SPIROCHETES

KD 94

:~ B31 VL3 VT1 HT'2 HBL1VH1 VS1 v-r2w

- ~

611

BH.

~

30-

2014-

Fig. !. SDS-PAGE of whole-cen proteins of 9 B. burgdorJeri (B31, VL3, VTI, HT2, HBL i, VH 1, VSI, VT2, W) strains, and one B. hermsii (Bh) strain. Polyacrylamide gel (12.5 %) stained with Coomassie blue. MWS = molecular weight standard; KD = kilodaltons.

PAGE profiles of the 10 Borrelia strains tested were heterogeneous (fig. 1). Borreliae were cultivated in Barbour-Stoenner-Kelly medium (Barbour, 1984) for 7 days at 32°C. Cells were harvested by centrifugation and washed with sterile phosphate-buffered saline. Each strain was resuspended in PBS and used in the enzyme profile tests. The commercially available API ZYM and the experimental API galleries, aminopeptidases AP1-AP6, API glycosidases and API esterases were used. The enzyme kits are semiquantitative micromethods using 97 chromogenic enzyme substrates, including 04 for aminopeptidases, 20 for glycosidases, 10 for esterases and one each for alkaline phosphatase, acid phosphatase and phosphoamidase. The tests were performed according to the instructions of the producer. Enzymatic activities by API ZYM were graded according to the intensity of colour with the API ZYM colour reaction chart. The other test strips for aminopeptidases 1 to 6 and esterases display the positive reaction by appearance of orange colour for amir~opeptid,~ses, and

violet colour for esterases. All tests were repeated once in the same way. The results were reproducible and gave clear reactions. The specific enzyme activities of B. burgdorferi strains tested were widely similar, and also resembled those of the B. hermsii strain in tests for 43 peptidase, 20 glycosidase and 7 esterase activities. However, they differed from those of B. hermsii in 11 peptidase activities. Three esterase and 4 peptidase activities which were present in the B. hermsii strain were variable in B. burgdorferi isolates, and 3 esterase and 15 peptidase activities absent in B. hermsii were variable in B. burgdorferi. Peptidases are important for the utilization of peptides as sources of extraneous individual amino acids for protein synthesis, or as sources of carbon and latrogen. All Borrelia strains displayed a uniform glycosidase zymogram. The esterase pattern was almost identical up to a chain length of 8, from butyrate (C4) to nonano-

612

C. IV. PENN E T AL.

ate (C9). Two strains (W and VSI) had lipase activity. They hydrolysed long chain carboxylic acids up to a chain length of 18 carbon atoms. It is interesting that strain W with the highest lipase activity was isolated from the cerebrospinal fluid of a patient wittl lymphocytic meningitis. All strains had a uniform phosphatase and phosphoamidase activity. B. hermsii, however, differed markedly in its zymogram by lack of I l peptidases for utilization of mono- and dipeptides, which were present in all B. burgdorferi strains tested.

rRNA gene organization in spirochetes Masahito Fukunaga restated the definition of spirochetes based on their spiral shape and axially coiled fibrils. Despite these morphological similarities, there are some differences among them regarding habitat and physiological and biochemical metabolism. Molecular characterization methods, such as rRNA gene sequence co~nparison and DNADNA hybridization, have provided insight into the taxonomy of spirochetes. Fukunaga and colleagues were interested in studying the organization and regu. lation of the genes coding for the rRNA of spirochetes. Their interest in the genes in spirochetes was aroused because the organization of these rRNA genes in leptospires is unique (Fukunaga and Mifuchi, 1989), and organization and transcription of genes for rRNA is an essential process in the growth of any organism. Fukunaga and coworkers cloned DNA fragments containing the rRNA genes of B. burgdorferi and T. phagedenis. A restriction map of the fragment was constructed and organization of the rRNA genes was determined by hybridization. The number of rRNA genes in some other spirochetes was also determined by genomic hybridization. The genomic DNA fragment isolated from B. burgdorferi strain B31 contained a single copy of the 16S rRNA sequence and another cloned fragment contained both 23S and 5S sequences. These findings suggest that these rRNA genes are expressed independently in the organism. The location, linkage and number of rRNA genes in T. phagedenis strain Reiter and T. pallidum strain Nichols was determined. These organisms have two rRNA gene sets in their genome and in the rRNA gene sets for strain Reiter, the rRNA genes are closely linked and arranged in the order i6S-23S-5S. This arrangement is identical to that found io other common eubacteria. The results reported at this meeting and in earlier work of Fukunaga's group (Fukunaga et aL, 1990a,b, 1992a,b) have shown that the organization of rRNA genes among the spirochetes is divergent

and all types of rRNA gene organization are found among these organisms. The rRNA gene organization in lep~.ospiresis quite unique and that in treponemes is identical to the organization in common eubacteria. The organization of rRNA genes in B. burgdorferi is different from that in other spirochetes and similar to that of the thermophilic bacterium Thermus thermophilus and the planctomycete Pirellula marina. The coincidence of the organization of rRNA genes of these organisms is surprising, because they are members of different eubacterial families. Further studies on the expression and regulation of rRNA genes and nucleotide sequence determinations in comparison with those of other organisms should provide insight into the evolution of eubacteria.

Conclusion Despite the fact that all spirochetes are fastidious, and some, notably T. pallidum and closely related pathogenic treponemes, remain essentially noncultivable in serial subculture, the work described in this section illustrates that progress is being made in the elucidation of metabolic and cellular machinery in these organisms. Some paradoxes and puzzles remain. These seems to be a discrepancy between the observed small genome sizes of some of the organ-. isms and the apparent presence of some synthetic genes which might be expected to be "dispensable". The apparent absence from some species of cellular functions which we might expect to be almost universally present in bacteria, ~uch as RecA and the heat shock response, is curious and thought-provoking. Another example is the widely divergent organization of rRNA genes. These findings lead to the question: do we really have a clear idea of the phylogeny and relatedness of spirochetes? Hopefully, the further application of experimental approaches like those described above, and their extension into new areas, will provide more and firmer answers to this and other questions in the future.

Approche g~n~tique de la biologie cellulaire et du m~tabolisme des spirochetes Ce n'est que r6cemment que les spirochetes ont b6n6fici~: de la m6thodologie g6n6tique. Si les m6canismes habituels de transfert g6n6tique demeurent encore largement indisponibles pour ces bact&ies, de substanciels progr6s dans leur analyse g6n~tique ont cependant 6t6 r6alis6s dont on peut citer les exempies suivants. 1) Des g~nes impliqu6s dans les voles de biosynth6se des acides amin6s chez Treponema et Leptos-

GENETIC A P P R O A C H E S TO CELL B I O L O G Y A N D M E T A B O L I S M OF SPIROCHETES pira ont 6t6 clon~s par compl6mentation de mutants de Escherichia coli utilis6s comme h6te pour les plasmides recombinants. 2) La prot6ine OspA de Borrelia burgdorferi a ~t6 surexprim6e chez E. coli sans peptide signal ; le produit recombinant a 6t6 utilis6 Iors de diverses analyses biochimiques comme, en particulier, le dichroisme circulaire. 3) Les prot6ines exprim6es en r6ponse au choc thermique sont homologues aux prot6ines correspondantes de E. coli. 4) Les profils d'activit~ enzymatique de B. burgdorferi et autres spiroch&es montrent l'existence d'une h.,5.'t6rog6n6it6 des souches, lls indiquent aussi quelles activit6s biosynth6tiques et enzymatiques sont eonserv6es parmi les diff6rents spiroeh&es. 5) L'organisation des g6nes sp6eifiant les ARNr des spiroeh&es est diff6rente de celle des autres baet6ries. Mots-clds: Spirochete, G6n6tique, Biosynth~se; Voies, M&abolisme, ARN, Prot6ines de choc thermique, Profils enzymatiques.

References Amato, R.F., Bottone, E.J. & Amsterdam, D. (1991), Substrate profile systems ~'or,he identification of bacteria and yeasts by rapid and automated approaches, in "Manual of clinical microbiology". Fifth edition, chapter 18 (Balows, A., Hausler, W.J., Herrmann, K.L., Isenberg, H.D. & Shadomy, H.J.). American Society for Microbiology, Washington, DC. Barbour, A.G. (1984), Isolation and cultivation of Lyme disease spiroehaetes. Yale J. Biol. Med., 57, 521-525. Carriero, M.M., Laux, D.C. & Nelson, D.R. 0990), Characterization zf :he heat shock response and identification of heat shock protein antigens of Borrelia burgdorferi. Infect. Immun., 58, 2186-2191. Dunn, J.J., Lade, B.A. & Barbour, A.G. (1990), Outer surface protein A (OspA) from the Lyme disease spirochete, Borrelia burgdorferi: high level expression and purification of a soluble recombinant form of OspA. Protein exp. Purif., 1, 159-168. France, L.L., Kieleczawa, J., Dunn, J.J., Hind, G. & Sutherland, J.C. (1992), Structural analysis of an outer surface protein from the Lyme disease spirochete, Borrelia burgdorferi, using circular dichroism and fluorescence spectroscopy. Biochem. biophys. Acta (Amst.) (in press). Fukunaga, M., Horie, I. & Mifuchi, I. (1990a), Isolation and characterization of the 5S rRNA gene of Leptospira interrogans. J. Bact., 172, 3264-3268. Fukunaga, M., Masuzawa, T., Okuzako, N., Mifuchi, I. & Yanagihara, Y. (1990b), Linkage of ribosomal

613

RNA genes in Leptospira. Microbiol. ImmunoL, 34, 565-673. Fukunaga, M. & Mifuchi, i. 0989), Unique organization of Leptospira interrogans rRNA 3enes. J. Bact., 171, 5763-5767. Fukunaga, M., Mifuchi, l., Yanagihara, Y. & Okuzako, N. (1992a), Comparison of flanking regions of the 5S ribosomal ribonucleic acid (rRNA) genes in Leptospira :~iflexa and Leptospira interrogans, Chem. pharm. Bull., 40, 544-546, Fukunaga, M., Okuzako, N., Mifuchi, l., Arimitsu, Y. & Seki, M. (1992b), Organization of the ribosomal RNA genes in Treponema phagedenis and Treponema pallidum. Microbiol. Immunol., 36, 161-167. Gherardini, F.C., Hobbs, M.M., Stamm, L.V. & Bassford, P.J. (1990), Complementation of an Escherichia coli proC mutation by a gene cloned from Treponemapallidum. J. Bact., 172, 2996-3002. Girouard, L.A., Laux, D.C., Gupta, R.S., Jindal, S., Young, R.A. & Nelson, D.R. (1992), Immune recognition of human Hsp60 by Lyme disease patient sera. Infect. Immun. (submitted). Hubbard, C.L., Gherardini, F.C., Bassford, P.J. & Stature, L.V. (1991), Molecular cloning and characterization of a 35.5-Kilodalton lipoprotein of Treponema pallidum. Infect. lmmun., 59, 1521-i528. Lindquist, S. (1986), The heat shock response. Ann. Rev. Biochem., 55, 1151-1191. Manafi, M., Kneifel, W. & Bascomb, S. (1991), Fluorogenic and chromogenic substrates used in bacterial diagnosis. MicrobioL Rev., 53, 335-348. Sancar, A., Hack, A.M. & Rupp, W.D. (1979), Simple method for identification of plasmid-encoded proteins. J. Bact., 137, 692-693. Sanger, F.~ Nicklen, S. & Coulson, A.R. (1977), DNA sequencing with chain-terminating inhibitors. Proc. nat. Acad. Sci. (Wash.), 74, 5463-5467. Shinnick, T.M. (1991), Heat shock proteins as antigens of bacterial and parasitic pathogens. Curr. Top. Microbiol. Immunol.. 167, 145-160. Somers, J.M., Amzallag, A. & Middleton, R.B. (1973), Genetic fine structure of the leucine operon ol Escherichia coil K-12. J. Bact., 113, 1268-1272. Stature, L.V., Gherardini, F.C., Parrish, E.A. & Moomaw, C.R. (1991b), Heat shock response in spirochetes. Infect. Immun., 59, 1572-1575. Stamm, L.V., Parrish, E.A. & Gherardini, F.C. (1991a), Cloning of the recA gene from a free-living ieptospire and distribution of RecA-like protein among spirochetes. Appl. Environ. MicrobioL, 57, 183-189. Stamm, L.V., Stapleton, J.T. & Bassford, P.J. (1988L In vitro assay to demonstrate high-level erythromycin resistance of a clinical isolate of Treponema pallidum. Antimicrob. Agents a. Chemother., 32, 164-169. Walker, E.M., Arnett, J.K., Heath, J.D. & Norris, S.J. ( 1991), Treponema pallidum subsp, pallidum has a single, circular chromosome with a size of ---900kilobase pairs. Infect. lmmun.. 59, 2476-2479. Ward, J.B. & Zahler, S.A. (1973), Regulation of leucine biosynthesisin Bacillus subtilis. J. Bact., 116, 727-735. Winfield, J.B. & Jarjour, W.N. (1991), Stress proteins, autoimmunity, and autoimmune disease. Curr. Top. A¢icrobioL lmmunol., 167, 160-189.

Genetic approaches to cell biology and metabolism of spirochetes.

Genetic analysis and methodology have only comparatively recently been applied to the study of spirochetes. Although genetic transfer procedures for s...
725KB Sizes 0 Downloads 0 Views