Cloning
of an immunodominant antigen gene of Bartonella bacilliformis
Kyle E. McFarley
The University of Montana IBS-CORE Program
Mentor- Dr. M.F. Minnick
INTRODUCTION
Bartonella bacilliformis is a facultative intracellular bacterial parasite of human erythrocytes and endothelial cells. Although other species of Bartonella are found worldwide B. bacilliformis is endemic to South America. Outbreaks of B. bacilliformis have been reported in the mountainous regions of Peru, Colombia, Ecuador, Bolivia, Chile, and Guatemala[1]. B. bacilliformis infections are not only a concern for people living in these areas[2] but also travelers who visit them[3]. In some areas of Peru up to 60% of the local population are seropositive for antibartonella antibodies[4] and as many as 15% are chronic asymptomatic carriers[5].
The phlebotamine sandfly, Lutzomyia verrucarum, is the vector for transmission of the bacterium to humans. The female sandfly transmits the pathogen during nocturnal blood feeding on humans. The bacterium, B. bacilliformis is not contagious between humans1. In a fascinating and tragic historical case a 26-year-old medical student, Daniel Alcides Carrion, inoculated himself in August 1885 with secretions from a red, bulging skin lesion on a bed-ridden patient with a condition termed “verruga peruana.” On October 5 of that same year Carrion died in Lima, Peru from human bartonellosis, also known as Carrion’s disease, Oroya fever, or verruga peruana[6].
Treatment of the disease has been fostered by the advent of antibiotics. Prior to antimicrobials, Bartonella infections were fatal nearly 40% of the time. B. bacilliformis is sensitive to most antimicrobials including penicillin, tetracycline, chloramphenicol, and various aminoglycosides.[7],8
RESEARCH METHODS
Sample collection.
The convalescent serum used in this experiment was obtained from Dr. Larry Laughlin at the Uniformed Health Services University. The serum was drawn on January 1997, from an eight-year-old boy in San Pedro, Peru. At the time of the blood draw the patient had verrugous lesions and was culture positive for B. bacilliformis.
Titering the l Zap
Express genomic library of B. bacilliformis (KC583)
The library titer was determined by plating a ten-fold serial dilution of the stock library. One microliter of phage suspended in S buffer from the serial dilutions was used to infect 200 ml of E. coli. XL1 Blue. Phage were incubated 25min. in a 37°C water bath to infect the host cell. Phage-infected host cells segregated by their serial dilution were then added to separate test tubes each containing 3ml of molten TB soft top agar, cooled to 55° C in a heating block. LB plates were prewarmed and labeled corresponding to the serial dilution of the l Zap Express library of B. bacilliformis. Each test tube containing a certain dilution of the l Zap Express library was suspended in 3 ml of the LB soft-top agar and poured onto a warmed LB plate and incubated overnight at 37°C.
It was determined from the titration of the library that a 1/10 dilution of the library achieved the desired plaque density. Plaque density was approx. 700 PFU per plate or 8700PFU/ml.
Probing the l Zap
Express library of B. bacilliformis by an Immunoblot protocol
It was determined through titration of l Zap Express library of B. bacilliformis that optimal plaque density was achieved with a 1/10 dilution of the stock library. 2ml of stock were added to 18ml of S buffer in an Eppendorf tube. This was repeated four times. Tubes were labeled. Four more Eppendorf tubes were prepared, each containing 200 ml of XLI Blue and also labeled A, B, C, & D. Four test tubes containing 3ml of LB soft-top agar were heated and then cooled in a 55°C heating block. 1ml of phage suspended in S buffer was taken from each of the four Eppendorf tubes labeled A, B, C, &D containing the 1/10 dilution of the library and added to corollary Eppendorf tubes containing 200 ml of XLI Blue. The infected cells were then incubated in a 37°C water bath for 25 minutes. LB plates were warmed as described in titration methods and corresponding labels of A, B; C, & D were given to each of the four plates. After the 25-minute bath infected host cells after the 25-minute bath were placed in a separate tube of preheated LB soft-top and then poured immediately onto a corollary LB plate. Plates were incubated for four hours at 37°C. Four Nitrocellulose filters were treated with IPTG and dried. These filters were layed over plates A, B, C, &D, and the plates were incubated overnight at 37°C to produce plaquing.
Before removal, nitrocellulose filters were labeled by pencil with their corresponding plate designation- (A, B, C, D). A needle tool was dipped in India ink and used to pierce the nitrocellulose filter and the LB agar in an asymmetric pattern to obtain the filter’s relationship to it’s corollary plate of origin. Filters were lifted using tweezers. Plates containing plaques were wrapped tightly in Parafilm and were stored at 4°C during development of filters.
The filters were immediately placed in a glass dish containing 50ml PBS (Ph 7.4) +0.3% Tween-20 +2% nonfat milk. The four suspended filters were allowed to rock and soak gently for 1 hour. After one hour the above solution was poured off and all the ingredients listed above were again added but, without the milk. In addition to these ingredients 50 ml of the primary antibody was added. The primary antibody was from patient “San Pedro” and standard safety protocol was followed when handling the human serum. Throughout the nitrocellulose development glass dishes were used. Probing of the nitrocellulose filters with the primary antibody was allowed to occur for 24 hours. The dish was wrapped in saran wrap to inhibit evaporation. After the 24 hour probing period the above solution was discarded and the filters were washed in the same dish 5 times in 50ml of PBS. After each wash the previous wash was discarded. After all five washes were completed the filters were soaked and rocked in 50ml PBS containing 25ml goat anti-human: HRP antibody. After one hour the above was discarded and the filters were washed again 5 times for ten minutes each wash in 50ml of PBS. After washes were accomplished the filters were developed with50ml PBS containing 75 ml of 30% H202,
0.04g of 4-Chloronapthol and 10 ml of MeOH. The developer was mixed and added to the glass dish containing the filters, immediately after the final wash. Development occurred for 10 min. on the rocker. Developed plaques were confirmed and the reaction was stopped with tap water. Positive plaques appeared dark blue in color. Since color reactions of this kind fade, the plaques were noted and enhanced with an indelible marking pen.
The plates containing the plaques that the nitrocellulose filters came from were removed from the 4°C storage and were taken together with the corresponding developed filters to a light table. The holes in the nitrocellulose filters made with the needle tool and stained with Indian ink made prior to the filters being lifted were matched to their corollary plate of origin. With the developed filter aligned perfectly below its corollary plate the positive plaque can be determined with a relative degree of certainty on the master plate. Using a borosilicate pipette, the plaque(s) were harvested and suspended in 90ml of S buffer in an Eppendorf tube. When the plaques are harvested in this way a plug is formed. The plug is as long as the agarose is deep. There may be several plugs per plaque. These tubes were labeled A, B, C, &D. If there were multiple positive plaques on a certain filter then they were labeled with a suffix 1,2,3, etc. Plate D had 4 positive plaques on it so the positive plugs suspended in S buffer in Eppendorf tubes were labeled D-1, D-2, D-3, & D-4. Were added to each Eppendorf tube containing the suspended plugs.
Two plugs were chosen to purify further. The two plugs were D-1 and D-2. These plugs were replated using the same methods as for the stock library. Nitrocellulose filters were laid over the plates as before and these filters were developed using all the same techniques as above. This was done to ensure that a very isolated plaque/ plug was obtained. We wanted an isolated positive plaque that would not be contaminated with phage expressing other proteins from B. bacilliformis that we were not interested in or rather that the antibodies from the patient sera did not recognize. After a new set of plugs was obtained and titered using the same protocol given as for the l Zap Express library, it was determined that the optimal density was achieved at a 1/100 dilution of the plug. Note that the phage elute into the buffer so it is not necessary to be concerned with the plug suspended in the buffer. The contents of the tubes were suspended and then I waited for the chloroform to settle to the bottom of the Eppendorf tube. It is important to allow the chloroform to settle out or the host cells that are used in further steps will be killed.
Purification of DNA from pKM-1 plasmid using DNA Perfect Prep Kit
DNA was purified using a DNA Perfect Prep Kit as per the manufacturer’s protocols. (5 Prime® 3 Prime, Inc. â Boulder, CO). Plasmids isolated with the Perfect Prep Kit were labeled pKM-1, pKM-2, pKM-3, and pKM-4. Each of the products was subjected to electrophoresis on a 0.8% agarose gel with l Hind III fragments and it was determined that pKM-1, pKM-2, pKM-3, and pKM-4 all contained the same insert. Bands were visualized by illumination with ultraviolet light. pKM-1 was used throughout the rest of the research and pKM-2, pKM-3, and pKM-4 were stored at -20° C.
In VIVO excision of the pKM-1 plasmid from the l Zap Express arms using the EXASSIST/ XLOLR system
In VIVO excision of the pKM-1 plasmid was executed using the manufacturer’s (Stratagenes) protocols.
Reintroducing pKM-1 into the XLOLR host cell.
Four seed cultures
of XLOLR were grown overnight in LB in a shaking incubator. The next day 1ml of
the culture was aspirated and placed into 20ml LB/Kanamcin 50. The bacterial
cultures were then grown for two hours at 37° C with shaking. 100ml (10 mM) of IPTG stock was added and grown 2
additional hours. The XLOLR culture was
spun at 6000 RPM for 5min and the supernatant decanted. The pellet was then
resupended in 1x PBS by vortexing. The suspended pellet was transferred to an
Eppendorf tube
Western Blot and SDS-PAGE gel comparing pKM-1 in XLOLR to pBK-CMV cloning vector in XLOLR.
A 12.5% acrylamide SDS-PAGE gel was prepared with a 0.4% stacking gel. Bacterial lycates from E. Coli containing PBK-CMV and PKM-1 were run next to a molecular weight color standard. Proteins were transferred to nitrocellulose per the Immunoblot protocol developed by M.F. Minnick, University of Montana. The gel was laid on Nitrocellulose paper cut to size. Another piece of nitrocellulose paper was laid over the gel and this sandwich was placed in the larger transfer sandwich running rig. Next, 4L of 20mM sodium phosphate buffer was made using 76 ml 0.2M NaH2HPO4 (Monobasic) and 324 ml 0.2M Na2HPO4 (dibasic). This mixture was brought to 4 liters with distilled water. 1.5 L was poured into the pan to make the transfer sandwich. The buffer in the tank was pre-cooled before introducing the sandwich. The sandwich was blotted in a water-cooled tank at 11 volts (0.2 Amp) overnight. Western Blot development was the same as development of nitrocellulose filters as described above.
Restriction Mapping of pKM-1
using restriction enzymes Hind III and Pst 1 (with l
DNA/Hind III fragments as the molecular weight markers)
Four Eppendorf tubes were prepared for the three different digests of pKM-1 and one for the molecular weight marker; the tubes were labeled 1-4. To tube one the following was added: one microliter of 10X react buffer, eight microliters of deionized water, and one microliter of l DNA/Hind III fragments. The l DNA/Hind III fragments were the DNA size standards in lane one of the gel that lanes 2-4 would be compared to. Into tube two the following was added: Three microliters of pKM-1 digested with one microliter of Hind III restriction enzyme, one microliter of 10x react buffer, and five microliters of deionized water. Tube three contained three microliters of pKM-1 digested with one microliter of Eco R1 restriction enzyme, and five microliters of deionized water. Tube four contained three microliters of pKM-1 digested with one microliter of Pst 1, one microliter of 10x react buffer, and five microliters deionized water. The contents of these four Eppendorf tubes were placed into four corresponding wells in a 0.8% agarose gel and subjected to electrophoresis.
An approximation of the size of each band that appeared in lanes 2-3 of the gel was determined by measuring with a millimeter ruler the distance between the molecular weight marker bands whose base pair size is known above and below the unknown fragment of interest. That measurement is then divided by the difference in size of the two known fragments. In this way a relationship is derived corresponding Xbp to One millimeter. Using the millimeter ruler again a measurement is taken between the closest known marker fragment and the unknown fragment of interest. The difference is then multiplied by the value for each derived millimeter unit as described above. This method was repeated for each of the fragments in lanes 2-4.
Taking the above data I compared the restricted fragments of pKM-1 with the already determined approximate size of pKM-1 and the known size of the pBK-CMV with the know restriction sites on pBK-CMV. The pBK-CMV phagemid vector contains 4518bp. The entire pBK-CMV plasmid has been mapped and common restriction enzyme cutting sites were determined. Using a map of pBK-CMV with all of these restriction sites labeled I was able to determine the size of the insert of DNA from the San Pedro serum recognized antigens that were excised and then cloned into the pBK-CMV phagemid which is now called pKM-1.
RESULTS
Probing of the l Zap Express library of B. Bacilliformis with the patient serum (San Pedro) resulted in twelve plaques that were recognized by human antibodies. From titration of the l Zap Express library of B. Bacilliformis, the patient serum antibodies recognized approx. 5 of every one hundred plaques from the current l Zap Express library of B. Bacilliformis Each plaque represented a clone of phage-infected E. Coli that were expressing proteins from B. Bacilliformis. A library of 12 serum recognized antigens from B. Bacilliformis was established.
One of the isolated plaques of that library was taken and the portion of B. Bacilliformis DNA cloned between the arms of the lambda phage was rescued into a double stranded circular plasmid in vitro using the EXASSIST/XLOLR system (Stratagene). The DNA from the resulting plasmid was purified and four samples of the DNA of that insert were produced. It was determined that each of the four DNA samples contained the same size insert.
Three samples of the pKM-1 plasmid were subjected to different restriction enzymes whose restriction sites are known on the pBK-CMV plasmid. For clarification pKM-1 is pBK-CMV with the insert. A restriction map of the insert was made. It was determined that there was an insert in the pBK-CMV plasmid based upon the known size of pBK-CMV. Also based upon the size of the inserts and the known restriction sites of the enzymes I was able to map the insert (Fig. 1). The length of the insert was determined to be approximately 2145 base pairs.
Fig. 1 Partial restriction endonuclease map of pKM-1.
An SDS-PAGE gel was run comparing proteins of pBK-CMV in XLOLR and pKM1 in XLOLR. At the same time a Western blot of these proteins was performed and was probed to determine the size of the antigen protein. A protein of approx 57,727 Da was observed on westerns containing pBK-CMV / XLOLR. This protein would require a gene of approximately 1443bp to code for it. This number of base pairs is within the size of the insert and this protein was not recognized in the pBK-CMV lane of the Western blot. In the SDS-Page gel comparing these two it is extremely difficult to distinguish this protein as is illustrated in Fig. 2.
. 
Fig. 2. Western blot analysis of genes expressed from pKM1. Lanes 1-Molecular weight standard, 2-E. coli XLOLR containing the vector, pBK-CMV, 3-XLOLR containing pKM-1.
DISSCUSSION
In summary, this project resulted in identification of a clone, expressing a gene of approx. 1443 bp. The gene encodes an antigenic protein of 57.727 Da that is recognized by human convalescent serum.
The next step in this research is to send the DNA from pKM-1 off for sequencing. Once the DNA sequence is obtained the predicted Amino acid sequence can be determined. In addition the secondary sequence can be predicted. The analysis of the DNA sequence will be made using the GenBank database to search for homologs using BLAST searches.
ACKNOWLEDGEMENTS
This work was funded by an IBS-CORE Undergraduate Research Fellowship to Kyle McFarley through a grant from the Howard Hughes Medical Institute to the University of Montana. I would like to thank all the members of the IBS-CORE program especially Carol Brewer, Carol Snetsinger, Anita Emborg, and Dr. Gannon. I would also like to thank members of the Minnick Laboratory, especially Mike Minnick for his patience, and dedication to the success of my research. Thanks also to Laura Smitherman, Sherry Coleman, and Leo Choquette for technical help and valuable discussions. I would finally like to thank my wife Dianne for her love and support of my interest in doing research.