Source: AUBURN UNIVERSITY submitted to
SUBUNIT VACCINE AGAINST BOVINE CHLAMYDIA (CHLAMYDOPHILA) PERCORUM INFECTION
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
TERMINATED
Funding Source
Reporting Frequency
Annual
Accession No.
0200543
Grant No.
2004-35204-14689
Project No.
ALAV-KALTENBOECK
Proposal No.
2004-01594
Multistate No.
(N/A)
Program Code
44.0
Project Start Date
Sep 1, 2004
Project End Date
Aug 31, 2008
Grant Year
2004
Project Director
KALTENBOECK, B.
Recipient Organization
AUBURN UNIVERSITY
108 M. WHITE SMITH HALL
AUBURN,AL 36849
Performing Department
PATHOBIOLOGY
Non Technical Summary
The prevailing view of bovine infection with intracellular bacteria of the genus Chlamydia is that of a curiosity, significant as rare zoonosis, but not for animal health and production. Modern ELISA and PCR techniques show a different picture, with very high seroprevalence, approaching 100%, and DNA prevalence as high as 50-60%, of clinically inapparent infection with Chlamydia psittaci (reclassified Chlamydophila abortus) and Chlamydia pecorum (Chlamydophila pecorum). We previously screened the C. abortus genome by expression library immunization for immunoprotective antigens in a mouse pneumonia model. Five genes protected as effectively as previous low-level infection with C. abortus. In a heifer C. abortus challenge model, vaccination with these protective genes or their corresponding proteins improved 6-week pregnancy rates after single artificial insemination to 83% from 50% in mock-vaccinated heifers. A future subunit vaccine will require additional C. pecorum protective antigens. We intend to identify in a mouse model protective C. pecorum vaccine candidate genes. We will create vaccine constructs of linear expression elements of each of approximately 1,000 open reading frames of C. pecorum. The C. pecorum genome whole genome sequence is already available to us. We will use genetic immunization with pools and single constructs to identify and confirm protective genes in three rounds of mouse respiratory challenge infection with C. pecorum.
Animal Health Component
100%
Research Effort Categories
Basic
50%
Applied
(N/A)
Developmental
50%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
3113310109025%
3113310110025%
3113410109025%
3113410110025%
Goals / Objectives
We will conduct a Chlamydophila pecorum genome-wide search for the best antigens mediating prophylactic immunity. To achieve this goal we propose the following specific objectives: 1) Create vaccine constructs of linear expression elements of each of approximately 1,000 PCR-amplified open reading frames of the Chlamydophila pecorum genome, and use genetic immunization with pools and single constructs to identify protective genes in two rounds of mouse respiratory challenge infection with C. pecorum. 2) Confirm protective C. pecorum vaccine candidates as whole-gene ubiquitin-fusion expression plasmids, reconstructed for mammalian codon usage, in the mouse respiratory challenge model. Background Classical investigations on diseases induced in cattle by intracellular bacteria of the order Chlamydiales have revealed severe, but rare, disease manifestations such as pneumonia, enteritis, polyarthritis, sporadic encephalomyelitis, and abortion and fertility disorders. Investigations using modern ELISA and PCR techniques show ubiquitous chlamydial infections in cattle. High seroprevalence, approaching 100%, and chlamydial genomic DNA prevalence as high as 50-60% is found of clinically inapparent bovine infection with Chlamydia psittaci (recently classified Chlamydophila abortus) and Chlamydia pecorum (recently classified Chlamydophila pecorum). In a challenge experiment of virgin heifer with C. abortus, previous infection and established immunity against C. abortus did not protect against depression of fertility by direct uterine challenge with C. abortus or by challenge through contact with shedding herd mates (cohorts). The results also indicated that pregnancy results at low uterine challenge were strongly influenced by concomitant cohort challenge, as well as by the level of pre-challenge immunity against C. abortus, expressed as anti-C. abortus IgM serum antibodies. Recently, we screened the C. abortus genome by ELI for immunoprotective antigens in a mouse pneumonia model. Five genes protected as effectively as previous low-level infection with C. abortus. In a heifer C. abortus cohort challenge model, vaccination with these protective genes or their corresponding proteins improved 6-week pregnancy rates after single artificial insemination to 83% from 50% in mock-vaccinated heifers. The high prevalence of C. pecorum suggests that a future vaccine against bovine chlamydial infection and/or disease will require C. pecorum protective antigens in addition to the already identified C. abortus antigens. Our long-term goal is to develop a subunit vaccine for prevention and/or therapy of bovine chlamydial diseases. Three developments provide us a unique opportunity to identify C. pecorum protective candidate antigens as components of such a subunit vaccine. First, we have recently validated the expression library immunization (ELI) approach for protective gene identification in C. abortus. Second, Dr. Stephen Johnston, the inventor of ELI, has developed methods that allow improved and accelerated deconvolution of the genomes of pathogens into the best vaccine candidates. Third, the complete genome sequence of C. pecorum is now available to us.
Project Methods
Objective 1 An essential component for directed ELI is access to the sequence of a pathogen genome. The PD laboratory has cultivated and highly purified a large quantity of C. pecorum type strain E58 (ATCC VR-628); isolated by McNutt in Iowa in 1940 from a case of sporadic bovine encephalomyelitis. C. pecorum E58 DNA was sequenced by Dr. Garry Myers at The Institute for Genome Research in Rockville, Maryland, and the PD has now access to the raw sequence. The availability of the C. pecorum genome sequence and purified DNA allows us to utilize LEE technology for genetic vaccination, co-developed by Dr. Sykes in the Johnston laboratory. The only requirement for protein expression using genetic immunization is a DNA fragment with a mammalian promoter, the gene of interest, and a transcriptional terminator. Linear dsDNA fragments like PCR products that encode these three components can be non-covalently linked together and directly injected into target animals. Since the gene gun requires only microgram quantities of DNA to induce protective immune responses, amplification of the DNA in E. coli is not required. By removing the bacterial requirements in substrate generation we eliminate any cloning biases as well as the time required for these extra manipulations. By combining the LEE technology with ELI technology, we can easily screen even pathogens with large genomes, once the sequence is known. The Sykes laboratory will use the sequence for Bioinformatics Analysis, identification of ORFs, and design of primers for ORF amplification which will be linked with promoter and terminator elements to produce LEEs. Using a murine prophylactic lung model, we will screen the LEE-ELI library for protective genes. Eight week-old female A/J mice will be used as low-dose (104 C. pecorum genomes) live-vaccinated protection controls, as naive disease controls, and for 3x vaccination with pools of the LEE constructs. All mice will be challenged with 106 C. pecorum genomes, and sacrificed 10 days later by carbon dioxide inhalation. Lungs will be weighed, snap-frozen, and DNA will be extracted. C. pecorum lung burdens will be determined by C. pecorum 23S rRNA FRET real-time PCR, and protection will be scored as reduction in the logarithm of chlamydial lung genomes as compared to naive controls. Objective 2 An important factor affecting protection in genetic immunization is the level of protein produced in the mammalian cell. In objective 2, we will improve the selected genes by changing the C. pecorum codons to the corresponding optimal codons for mammalian expression. This is likely to have a pronounced effect on protein production as the C. pecorum codon usage preference is likely far from that of mammals, similar to all other chlamydiae. Each of the vaccine candidate genes selected in objective 1 will be chemically resynthesized to specifications. The recoded constructs will be used in groups of 10-20 mice, as ubiquitin fusion constructs, to confirm their protective efficacy in the mouse model.

Progress 09/01/04 to 08/31/08

Outputs
OUTPUTS: 1) 1,685 C. pecorum ORF genetic immunization constructs have been designed, oligonucleotides synthesized, and high-quality vaccine constructs produced with 100% success rate were completed in November 2006. 2) An optimized biolistic (gene gun) vaccination modality was developed for vaccination with pools of ORF vaccine constructs. To enhance Chlamydia-protective Th1 immunity, both subunit genes encoding E. coli lethal toxin (LT) were included as genetic adjuvant in all vaccinations used for expression library screening in the A/J mouse model. 3) In the first round of ELI, a library of linear expression elements (LEE) of 1,685 PCR-amplified fragments of all putative ORFs of the C. pecorum genome was constructed. Pools of 8 LEEs each were arranged in a 18x12 rectangular matrix, and were combined into 18 vertical and 12 horizontal groups. Per vaccine group, 10 female, 6-week-old A/J mice were immunized by gene gun on days 0, 3, 6, 20, and 34. All mice were intranasally challenged with 4x108 C. pecorum strain E58 organisms 4 weeks later and sacrificed after 10 days. C. pecorum lung loads were quantified by 23S rRNA FRET PCR, and a protection score was calculated relative to unprotected naive mice (0 percent protection) and optimally protected mice that had received a low-dose live C. pecorum inoculum 4 weeks prior to the high-dose challenge infection (live-vaccinated, 100 percent protection). This round identified several protective pools and allowed reduction of the original pool of 1,053 C. pecorum vaccine candidate genes (a total of 1,685 partial or full gene constructs for vaccination) to 150 genes (a total of 270 partial or full gene constructs). 8) The second round of C. pecorum ELI with 150 candidate genes identified in August 2007 a pool of 55 C. pecorum vaccine candidate genes (90 ORF fragments) among the 270 gene constructs. 4) In the third round, completed in March 2008, these 55 genes were tested in a 8x7 matrix arrangement of pools of 7 or genes. Disappointingly, none of these pools mediated full protection, while several pools mediated partial protection with low variance of the data. Further analyses suggested that this low protection was associated with a Th2 shift of the immune response. This type of immunity is non-protective against chlamydial infection. We attribute this immune shift to an inadequate immunization approach, with potential over-vaccination and insufficient antigen removal that would result in persistent antigenic stimulus leading to Th2-biased immunity. 10) In a series of experiments that are in progress, we use a proven protective gene of C. pneumoniae in a C. pneumoniae - A/J and C57BL/6J mouse challenge model to identify an optimal vaccination modality for low-complexity or single antigen genetic immunization. The preliminary data suggest that either very low genetic antigen doses or a mixed modality genetic vaccine priming followed by protein antigen boost will yield highly protective vaccination outcomes. Once these optimization experiments are completed (anticipated for April 2009), we will use the optimal approach to identify the definitive C. pecorum vaccine antigens within the pool of 55 candidates. PARTICIPANTS: Not relevant to this project. TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: 1) Analyses of experimental results suggested that low protection observed during vaccination with low-complexity pools or single genes was associated with a Th2 shift of the immune response. This type of immunity is non-protective against chlamydial infection. We attribute this immune shift to an inadequate immunization approach, with potential over-vaccination and insufficient antigen removal that would result in persistent antigenic stimulus leading to Th2-biased immunity. 2) In a series of experiments that are in progress, we use a proven protective gene of C. pneumoniae in a C. pneumoniae - A/J and C57BL/6J mouse challenge model to identify an optimal vaccination modality for low-complexity or single antigen genetic immunization. The preliminary data suggest that either very low genetic antigen doses or a mixed modality genetic vaccine priming followed by protein antigen boost will yield highly protective vaccination outcomes. Once these optimization experiments are completed (anticipated for April 2009), we will use the optimal approach to identify the definitive C. pecorum vaccine antigens within the pool of 55 candidates.

Impacts
1) Out of the total of 1,053 C. pecorum genes (a total of 1,685 partial or full gene constructs for vaccination)a pool of 55 C. pecorum vaccine candidate genes (90 ORF fragments) were identified. 2) Findings during this investigation indicate the need for re-evaluation of genetic immunization in general, and specifically biolistic immunization by the gene gun. It appears that improved technology and stability of genetic vaccine constructs has resulted in long persistence of expression plasmids at the anatomical vaccination site, resulting in a shift towards Th2 immunity that is not protectiuve against chlamydiae and other intracellular bacteria. Ongoing investigations funded from non-CSREES sources allow continuation of the project to optimize genetic immunization against chlamydiae and other intracellular pathogens. Conclusive results from this investigation are anticipated for mid-2009. This will allow identification of the definitive C. pecorum vaccine candidate genes among the present 55 candidates. 3) It is anticipated that identification of 2-3 optimal C. pecorum vaccine candidate proteins will allow combination with the already known 3 best protective proteins for C. abortus to formulate and test a highly effective vaccine aganist chlamydial infections in cattle. 4) It is further anticipated that the knowledge about mechanisms that lead to Th2 immunity after genetic immunization will allow effective manipulation of teh immune response obtained by gteentic immunization. Thsi will have a major impact on the overall field of genetic immunization.

Publications

  • 1) Li, Y., Borovkov, A., A. Loskutov, S. Ahluwalia, C. Wang, D. Gao, K. F. Sykes, and B. Kaltenboeck. Identification of Chlamydophila pecorum vaccine candidate genes by expression library immunization. Proceedings of the 88th Annual Meeting of the Conference of Research Workers in Animal Disease. Chicago, Illinois, December 2007: Abstract 17.


Progress 01/01/07 to 12/31/07

Outputs
OUTPUTS: Preliminary results were reported at the Conference of Research Workers in Animal Disease in Chicago, December 2007. PARTICIPANTS: Yihang Li, Graduate Research Assistant, Auburn University, Auburn, AL Dongya Gao, Research Assistant III, Auburn University, Auburn, AL Bernhard Kaltenboeck, Professor, Principal Investigator, Auburn University, Auburn, AL Kathryn F. Sykes, Assistant Professor, Biodesign Institute, Arizona State University, Tempe, AZ Alexandre Borovkov, Research Assistant Professor, Biodesign Institute, Arizona State University, Tempe, AZ Andrey Loskutov, Research Assistant IV, Biodesign Institute, Arizona State University, Tempe, AZ TARGET AUDIENCES: Researchers in Veterinary Medicine and Animal Protection. PROJECT MODIFICATIONS: The modifications in the screening approach are driven by the need to search for stochastic combinations of 2 or more partially protective genes, since single genes do not mediate sufficeint protection.

Impacts
Beginning in December 2006, we initiated expression library immunization (ELI) in a murine respiratory disease model to identify candidates for a subunit vaccine against C. pecorum. A library of linear expression elements (LEE) of 1,685 PCR-amplified fragments of all putative ORFs of the C. pecorum genome was constructed. For selection of vaccine candidates, pools of 8 LEEs each were arranged in a 18x12 rectangular matrix, and the pools were combined into 18 vertical and 12 horizontal groups. Per such vaccine group, 10 female, 6-week-old A/J mice were immunized by gene gun vaccination on days 0, 3, 6, 20, and 34. All mice were intranasally challenged with 3.6x108 C. pecorum strain E58 organisms 4 weeks later and sacrificed after 10 days. C. pecorum lung loads were quantified by 23S rRNA FRET PCR, and a protection score was calculated relative to unprotected naive mice (0 percent protection) and optimally protected mice that had received a low-dose live C. pecorum inoculum 4 weeks prior to the high-dose challenge infection (live-vaccinated, 100 percent protection). Intersections between several highly protective vertical and horizontal vaccine groups were further dissected in a second round of ELI in an 18x9 rectangular matrix of single genes. Only weak protection was found in round 2. This suggests that single individual C. pecorum genes are not highly protective, instead only the combination of partially protective genes mediates subunit vaccine protection comparable to that observed in live-vaccinated mice. An experiment to identify the best combination of such partially protective genes is currently in progress.

Publications

  • An abstract of the report of preliminary results is available in the Proceeding of the 2007 Conference of Research Workers in Animal Disease in Chicago.


Progress 01/01/06 to 12/31/06

Outputs
The starting date of the project is October 1, 2006, with the actual account established in January 2007. Basic organizational procedures for the project have been done until December 31, 2006. The measures include detail planning of sampling methodology on the animal and in the laboratory. Further evaluation of sampling will be performed between 1/2007-4/2007, and actual routine sampling in bi-weekly intervals will be started in May 2007. At that time cows will become exposed to heat stress, and conditions for stress-induced immunosupprossion and chlamydial mastitis will be optimal.

Impacts
Identification of C. pecorum protective antigens is required for a vaccine that will generate immunity against both species of Chlamydia routinely found in ruminants, i.e. C. pecorum and C. abortus. C. abortus protective antigens have already been identified. Once protective antigens of C. pecorum are known, proteins of both species will be produced and incorporated into an experimental vaccine that elicits Thi-biased immunity using standard methodology. Preliminary data with a whole chlamydia organism vaccine indicate that Th1-biased vaccination against chlamydiae will reduce milk somatic cell counts in dairy cows, and will potentially improve fertility.

Publications

  • No publications reported this period


Progress 01/01/05 to 12/31/05

Outputs
A large stock of highly pure C. pecorum elementary bodies, sufficient for all challenge infections, was generated. The mouse lung disease model was retested for identification of an optimal inbred mouse strain. In collaboration on an NIH-funded project on Chlamydia taxogenomics, DNA of C. pecorum type strain E58 had been purified earlier, and the genome was sequenced from this DNA by G. Myers at The Institute for Genomic Research, Rockville, MD. After the annotated genome became available in January 2005, our primer design programs designed PCR oligonucleotides to amplify all genomic coding regions as 1,637 overlapping ORFs of 1,000 bp each. A C. pecorum mammalian expression library for vaccine screening of all ORFs is underway. Purified C. pecorum DNA serves as template for PCRs, and 15-base single stranded overhangs are created at both 5-prime and 3-prime ends to accommodate the annealing of promoter and terminator DNA products. These linear expression elements will be strategically pooled into 90 inoculating groups, each comprised of 40 ORFs, for biolistic delivery into mice. To favor stimulation of Th1-biased immunity, which is known to be important in the protective type of anti-chlamdial immune response, two plasmids encoding E. coli heat-labile enterotoxin subunits A (LTA) and B (LTB) will be included in the antigen inocula. In preliminary experiments, gene gun delivery with mutant, non-toxic LTA and LTB drove predominantly an antigen-specific strongly Th1-biased antibody response. We are currently confirming this effect with Chlamydia gene immunization in A/J mice and optimizing the C. pecorum intranasal challenge dose in an A/J challenge model. We anticipate starting the ELI screen of the whole C. pecorum genome in early 2006.

Impacts
Identification of C. pecorum protective antigens is required for a vaccine that will generate immunity against both species of Chlamydia routinely found in ruminants, i.e. C. pecorum and C. abortus. C. abortus protective antigens have already been identified. Once protective antigens of C. pecorum are known, proteins of both species will be produced and incorporated into an experimental vaccine that elicits Thi-biased immunity using standard methodology. Preliminary data with a whole chlamydia organism vaccine indicate that Th1-biased vaccination against chlamydiae will reduce milk somatic cell counts in dairy cows, and will potentially improve fertility.

Publications

  • No publications reported this period