Brucella Ovis Research Articles

CORRELATION STUDY BETWEEN POSITIVE SEROLOGIC TESTS AND BACTERIOLOGICAL AND MORPHOPATHOLOGICAL EXAMINATION FOR BRUCELLA OVIS INFECTION IN THE AREA OF CLUJ COUNTY

Research was conducted during 2000 - 2004 on a number of 60.986 blood samples from rammers, 530 pairs of testicles for bacteriological examination and 76 testicle samples processed for histopathology. Serologic and morphopathologic investigations were performed in the Veterinary State Laboratory Cluj (LSVS) and the bacteriology department in the Institute of Diagnosis and Animal Health (IDSA) Bucharest. Following the examination by serological RFC resulted a number of 4200 positive tests from 60.986 samples tested. Thus, the percentage of sickness caused by infectious epididimitis in rammers is 6.88%. Investigations by bacteriological examination, conducted at IDSA Bucharest on a number of 530 samples testicles (representing evidence and aries), revealed the presence of infectious epididimitis of rammers at 77.36%, 22.64% of the samples were negative. Having established a hierarchy of isolated germs involved in the aetiology of epididimitis we observed that 62.02% of cases are caused by Brucella ovis, 6.60% of cases by B. ovis in association with Arcanobacterium pyogenes and 7 ,74% by Arcanobacterium pyogenes only. Histological examination was conducted on a number of 76 testicles samples, of which 46 samples (60.52%) were with pathologic changes that can be attributed to infectious orchiepididimitis of rammers. Correlation is positive between histological examination and bacteriological examination, from a proportion of 100% testing positive for pathology, we managed to isolate Brucella ovis from 83.33%. Between necropsic exam results and serological examination by RFC, the correlation is negative, only 4.17% of the proportion of 100% diagnosed serological, presents pathological lesions of epididimitis. This type of relationship is valid also between serological examination by RFC and palpatory clinical examination. Considering RFC as a standard method to diagnose infectious epididimitis and comparing results with those of bacteriological examination carried out on the same samples that were diagnosed positive with RFC, we found that between these two methods there is a positive correlation. If RFC detected positive samples in the proportion of 100%, bacteriological examination detected the evidence of infected germs in a proportion of 77.36%, of which the Brucella ovis is 69.62%. Following bacteriological tests carried out on the same samples diagnosed as positive by serological examination (RFC), we find that the reaction of complement fixation although very sensitive, does not meet all the requirements, especially in regard of specificity, detecting also other bacterial antigens similar to B . ovis. For these reasons, in order to avoid false positive reactions from rammers with high economic and biologic value, testing is required to be made by at least two different methods (RFC and ELISA or RFC and bacteriological examination of semen).

Efficacy of bp26 and bp26/omp31 B. melitensis Rev.1 deletion mutants against Brucella ovis in rams

Brucella melitensis Rev.1 is the most effective vaccine against B. ovis infection in sheep but induces antibodies interfering with B. melitensis diagnosis. Brucella BP26 and Omp31 proteins are differential diagnostic antigens. Single or double bp26 and omp31 Rev.1 deletion mutants have been proven effective against B. melitensis in sheep. Here, the CGV26 (deleted in bp26 gene) and CGV2631 (deleted in both bp26 and omp31 genes) mutants have been tested for efficacy against B. ovis in rams. Either inoculated subcutaneously or conjunctivally, both mutants conferred significant protection against B. ovis. The protection induced by CGV26 was similar to that of Rev.1 but significantly higher than that conferred by CGV2631. In conclusion, the CGV26 mutant, in association with the adequate diagnostic strategy, could be a useful alternative to Rev.1 for sheep vaccination against B. ovis infections in those countries performing simultaneously B. melitensis and B. ovis eradication campaigns.

Differential expression of inflammatory and immune response genes in rams experimentally infected with a rough virulent strain of Brucella ovis

Infection of sheep with Brucella ovis results in ovine brucellosis, a disease characterized by infertility in rams, abortion in ewes and increased perinatal mortality in lambs. During the course of the infection both the ovine immune response and host cell gene expression are modified. The objective of this research was to conduct a preliminary characterization of differential gene expression in rams experimentally infected with B. ovis by microarray hybridization and real-time RT-PCR. Of the 600 ruminant inflammatory and immune response genes that were analyzed in the microarray, 20 and 14 genes displayed an expression fold change >1.75 with a P-value <0.05 at 15 and 60 days post-challenge (dpc), respectively. Of these genes, 16 were upregulated and 4 were downregulated in infected rams at 15 dpc. At 60 dpc, 11 and 3 genes were up- and down-regulated in infected rams, respectively. Only four genes, desmoglein, epithelial sodium channel, alpha subunit (ENaC-alpha), interleukin 18 binding protein (IL18BP) and macrophage migration inhibition factor (MIF) were found upregulated in infected rams at both 15 and 60 dpc. The analysis of differentially expressed genes demonstrated activation of inflammatory and innate immune pathways in infected animals. B. ovis infection also resulted in upregulation of genes involved in phagocytosis and downregulation of protective host defense mechanisms, both of which may contribute to the chronicity of B. ovis infection. The gene expression profiles differed between rams with severe and moderate B. ovis infection. This is the first analysis of differential gene expression in rough brucellae and particularly in B. ovis-infected rams. The characterization of the genes and their expression profiles in response to B. ovis infection further contributes to our understanding of the molecular mechanisms of infection and the pathogenesis of brucellosis.

Stability of Poly(ε-caprolactone) Microparticles Containing Brucella ovis Antigens as a Vaccine Delivery System Against Brucellosis

In previous works, our research group has successfully proved the use of subcellular vaccines based on poly(epsilon-caprolactone) (PEC) microparticles containing an antigenic extract of Brucella ovis (HS) against experimental brucellosis in both mice and rams. However, the successful exploitation of pharmaceutical products, and therefore of this product as veterinary vaccine, requires preservation of both biological activity and native structure in all steps of development from purification to storage. In this context, we have carried out an accelerated stability study to evaluate the relative stability of HS when loading in PEC microparticles. For this purpose, freeze-dried microparticles were stored at 40 +/- 1 degrees C and 75% RH as a preliminary analysis of a stability testing. The results showed that both physico-chemical (size, morphology, antigen content, release profile) and biological (integrity and antigenicity of the HS) properties were preserved after 6 months of storage. On the contrary, after 1 year of storage, the HS release profile was dramatically affected probably due to a progressive loss of the polymer microstructure. In addition, the degradation and loss of the antigenicity of the HS components was also evident by SDS-PAGE and immunoblotting analysis. In fact, after 12 months of storage, only the integrity and antigenicity of two of the major protective proteins of the HS antigenic complex were preserved.

Evaluation of three different antigens in an indirect enzyme-linked immunoassay for the detection of antibodies against Brucella abortus SRB51 in vaccinated heifers

The live attenuated Brucella abortus SRB51 (SRB51) is a partial O-chain-deprived mutant. The relative lack of the polysaccharide prevents it from inducing antibodies detectable by most of the serological tests used for the diagnosis of bovine brucellosis. The performance of three antigens used in an indirect enzyme-linked immunoassay test for detecting SRB51 antibodies were evaluated. A homogeneous group of twenty-five 10-month-old Hereford heifers was used. The animals were bled on day 0 and then subcutaneously vaccinated with 2 ml of a commercially available SRB51 vaccine (Schering-Plough™) containing 1 × 10 7 to 3.4 × 10 7 viable cells. Blood samples without anticoagulant for sera obtaining were then collected at days 30, 90, 210 and 360 post-vaccination. To detect the SRB51 antibodies, Brucella ovis hot saline extract, B. ovis RLPS (RLPS), and SRB51-RLPS were used. The buffered antigen plate agglutination test and an indirect enzyme-linked immunoassay (I-ELISA) using the smooth LPS (SLPS) antigen from B. abortus were used as control tests. All the sera samples were negative in the BPA test and in the standard I-ELISA using the SLPS. The SRB51-RLPS and the B. ovis RLPS antigens performed better than the B. ovis hot saline extract antigen.

Incidence of Brucella species in marine mammals of the German North Sea

In this study, organ samples from 426 common seals Phoca vitulina, 298 harbour porpoises Phocoena phocoena, 34 grey seals Halichoerus grypus and 10 other marine mammals were assessed for the presence of Brucella species. Forty-seven common seals, 2 harbour porpoises and 1 grey seal were found to be positive for these bacteria. A total of 91 Brucella strains were successfully isolated, due to the fact that Brucella spp. were found in more than one organ sample in 15 animals. The primary organ in which the bacteria were present was the lung. In addition, 2 strains were isolated from lungworms (Parafilaroides spp.). Forty-nine of the isolated strains were selected for further analysis using conventional phenotyping methods. Molecular characterisation was carried out by analysing the IS711 and omp2 loci. With respect to the distribution of the IS711 loci in the genome, the 49 field isolates differed strongly from the terrestrial Brucella species and marginally from the marine Brucella reference strain NCTC12890. Based on the results of the PCR restriction fragment length polymorphism (PCR-RFLP) investigation of the omp2 locus, the majority of the Brucella field isolates were classified as B. pinnipediae, recently proposed B. pinnipedialis, possessing 1 omp2a gene and 1 omp2b gene. Two field isolates revealed the presence of 2 omp2a genes, as has been described for Brucella ovis. To our knowledge, these results confirm for the first time the presence of Brucella species in the marine mammal population of the German North Sea. These findings highlight the need for additional research on the relevance of these Brucella species for marine hosts and their environment.

A bioinformatic approach to understanding antibiotic resistance in intracellular bacteria through whole genome analysis

Intracellular bacteria survive within eukaryotic host cells and are difficult to kill with certain antibiotics. As a result, antibiotic resistance in intracellular bacteria is becoming commonplace in healthcare institutions. Owing to the lack of methods available for transforming these bacteria, we evaluated the mechanisms of resistance using molecular methods and in silico genome analysis. The objective of this review was to understand the molecular mechanisms of antibiotic resistance through in silico comparisons of the genomes of obligate and facultative intracellular bacteria. The available data on in vitro mutants reported for intracellular bacteria were also reviewed. These genomic data were analysed to find natural mutations in known target genes involved in antibiotic resistance and to look for the presence or absence of different resistance determinants. Our analysis revealed the presence of tetracycline resistance protein (Tet) in Bartonella quintana, Francisella tularensis and Brucella ovis; moreover, most of the Francisella strains possessed the blaA gene, AmpG protein and metallo-β-lactamase family protein. The presence or absence of folP (dihydropteroate synthase) and folA (dihydrofolate reductase) genes in the genome could explain natural resistance to co-trimoxazole. Finally, multiple genes encoding different efflux pumps were studied. This in silico approach was an effective method for understanding the mechanisms of antibiotic resistance in intracellular bacteria. The whole genome sequence analysis will help to predict several important phenotypic characteristics, in particular resistance to different antibiotics. In the future, stable mutants should be obtained through transformation methods in order to demonstrate experimentally the determinants of resistance in intracellular bacteria.

Allergen immunotherapy with nanoparticles containing lipopolysaccharide from Brucella ovis

The adjuvant and protective capacity against anaphylactic shock of the association between rough lipopolysaccharide of Brucella ovis (LPS) coencapsulated with ovalbumin (OVA), as a model allergen, in Gantrez ® AN nanoparticles was investigated. Several strategies were performed in order to study the adjuvant effect of the LPS either encapsulated or coating the nanoparticles. OVA, as well as LPS, was incorporated either during the manufacturing process (OVA-encapsulated or LPS-encapsulated nanoparticles, respectively) or after the preparation (OVA-coated or LPS-coated nanoparticles, respectively). After the administration of 10 μg of OVA incorporated in the different formulations, all the nanoparticles, with or without LPS, were capable of amplifying the immune response (IgG 1 and IgG 2a). However, in a model of sensitized mice to OVA, the formulation with OVA and LPS-entrapped inside the nanoparticles administered intradermally in three doses of 3 μg of OVA each was the only treatment that totally protected the mice from death after a challenge with an intraperitoneal injection of OVA. In contrast, the control group administered with OVA adsorbed onto a commercial alhydrogel adjuvant showed 80% mortality. These results are highly suggestive for the valuable use of Gantrez ® nanoparticles combined with rough LPS of B. ovis in immunotherapy.

Brucella melitensisB115-based complement fixation test to detect antibodies induced byBrucellarough strains

To assess the efficiency of a Brucella melitensis B115 rough strain, naturally devoid of anticomplementary activity, used as antigen in a complement fixation test (CFT) to detect antibodies induced by Brucella strains with rough phenotype, such as Brucella abortus RB51, Brucella ovis and Brucella canis. Complement fixation testing was performed on sera from RB51-vaccinated cattle and buffaloes, B. ovis-infected sheep and B. canis-infected dogs using B115, RB51 and the hot saline extract (HSE) as antigens. The B115-based CFT proved highly sensitive and specific in detecting rough antibodies and its efficiency was comparable with that of RB51 and HSE-based CFT. Brucella melitensis B115 can be successfully used as an antigen in CFT to detect antibodies induced by Brucella rough strains. Brucella melitensis B115 antigen may represent an improvement over Brucella rough strains for Brucella antibody detection by CFT, thus enhancing the efficiency of brucellosis surveillance systems. Owing to the absence of anticomplementary activity, it does not require particular growth conditions or modifications and can be accurately standardized. The B115-based CFT may constitute a suitable supplementary test for the diagnosis of human infections owing to rough Brucellae.

Micro-organism-like nanoparticles for oral antigen delivery

The aim of this work was to design and evaluate microorganism-like nanoparticles able to either mimic adhesion and colonization patterns of bacteria or to enhance and modulate the host immune response. Thus poly(anhydride) nanoparticles were either coated with adhesive factors (either with flagellin from Salmonella enteritidis or mannosamine) or loaded with lipopolysaccharide of Brucella ovis. The mucosal affinity of the nanoparticles was investigated by bioadhesion and fluorescence microscopy studies. The immunoadjuvant properties were checked by the oral route using ovalbumin as an antigen model. The results indicated that both flagellin and mannosamine nanoparticles display a stronger mucosal affinity to normal mucosa and Peyer’s patches than control ones, whereas LPS entrapped ones did not. Surprisingly, LPS formulation was unable to enhance the immune response by the oral route. However, flagellin and mannosamine nanoparticles induced stronger and more balanced serum levels of both IgG1 (Th2) and IgG2a (Th1) than control ones. In addition, oral immunization with ligand formulations induced high levels of fecal IgA.

Infecção em cão por Brucella abortus: relato de caso

Brucella abortus infection is reported in a dog from a rural area that presented at clinical evaluation left testicular enlargement and right testicular decrease. Serum resulted negative to rapid agglutination test and agar gel immunodifusion with Brucella ovis antigen but positive to buffered plate agglutination test, tube agglutination test and 2- Mercapthoetanol with B. abortus antigen. Brucella isolation was negative in blood, testicular material, semen and urine. Brucella DNA was detected in PCR from urine and blood.

Detection of Ovine Antibody to Brucella ovis by Indirect Enzyme Immunoassay

Because some batch‐to‐batch variation in the preparation of rough lipopolysaccharide (RLPS) from Brucella ovis has been experienced, several protocols were tested to establish the most reliable method for detection of antibody in indirect enzyme immunoassay. An early version of the assay gave a performance index (PI=sum of optimum percent sensitivity and percent specificity, determined by receiver operator characteristic analysis) of 198.6. This assay used RLPS from B. ovis as the antigen and a monoclonal antibody specific for bovine IgG1 heavy chain‐enzyme conjugate for detection. This was not repeatable using other batches of antigen. Newer versions of the assay generally had decreased sensitivity values, giving PIs of 193. Use of a recombinant protein A/G‐enzyme conjugate did not improve the PI (PI=190), giving reduced specificity and higher sensitivity. The final version used B. abortus RB51 RLPS as the antigen and protein A/G‐enzyme conjugate for detection, giving a PI of 197. Because of the batch uniformity of the B. abortus RB51 RLPS and the versatility of the protein A/G‐enzyme conjugate, the latter version appears to be the most useful for diagnostic serology.

A DNA vaccine coding for the chimera BLSOmp31 induced a better degree of protection against B. ovis and a similar degree of protection against B. melitensis than Rev.1 vaccination

In the present study, we reported an attempt to improve the immunogenicity and protective capacity of the chimera BLSOmp31 using a different antigen delivery: DNA vaccination. Vaccination of BALB/c mice with the DNA vaccine coding for the chimera BLSOmp31 (pCIBLSOmp31) provided the best protection level against Brucella ovis, which was significantly higher than the given by the co-delivery of both plasmids coding for the whole proteins (pcDNABLS+pCIOmp31) and even higher than the control vaccine Rev.1. Moreover, pCIBLSOmp31 induced higher protection against Brucella melitensis than pcDNABLS+pCIOmp31 but similar protection than Rev.1. The chimera induced a strong humoral response against the inserted peptide. It also induced peptide- and BLS-specific cytotoxic T responses. The insertion of this peptide on BLS induced stronger T helper 1 responses specific for the carrier (BLS), thus our results represent a case of synergic strengthening between two Brucella antigens. Hitherto, this is the first indication that a recombinant subunit vaccine elicits greater protection than whole Brucella.

Gantrez® AN nanoparticles as an adjuvant for oral immunotherapy with allergens

The aim of this study was to investigate the adjuvant properties of oral-administered Gantrez AN nanoparticles with ovalbumin (as allergen model) and, in some cases, lipopolysaccharide of Brucella ovis as immunomodulator. For this purpose, BALB/c mice were administered by oral gavage with OVA nanoparticles and both Th1 and Th2 markers (IgG2a and IgG1, respectively) were enhanced. On the other hand, these carriers administered by oral route were able to protect a model of sensitized mice to ovalbumin from anaphylactic shock. These results are highly suggestive for the valuable use of Gantrez nanoparticles in oral immunotherapy with allergens.

A recombinant subunit vaccine based on the insertion of 27 amino acids from Omp31 to the N-terminus of BLS induced a similar degree of protection against B. ovis than Rev.1 vaccination

The development of an effective subunit vaccine against brucellosis is a research area of intense interest. The enzyme lumazine synthase from Brucella spp. (BLS) is highly immunogenic, presumably due to its decameric arrangement and remarkable stability. In this work we decided to develop a chimera with the scaffold protein BLS decorated with 10 copies of a known protective epitope derived from an outer membrane protein of 31 kDa (Omp31) from Brucella spp. Vaccination of BALB/c mice with the chimera as a recombinant protein (rBLSOmp31) provided the best protection level against Brucella ovis, which was higher than the given by the co-delivery of both recombinant proteins (rBLS + rOmp31) and similar than the control vaccine Brucella melitensis strain Rev.1. Moreover rBLSOmp31 induced protection against Brucella melitensis but to a lesser degree than Rev.1. The chimera induced a strong humoral response against the inserted peptide. It also induced peptide- and BLS-specific T helper 1 and cytotoxic T responses. In conclusion, our results indicate that BLSOmp31 could be a useful candidate for the development of subunit vaccines against brucellosis since it elicits humoral, T helper and cytotoxic immune responses and protection against smooth and rough species of Brucella.

Inquérito soro-epidemiológico e fatores de risco associados à infecção por Brucella ovis em carneiros deslanados do semi-árido da Paraíba

Foi realizado um levantamento soro-epidemiológico da brucelose por Brucella ovis em reprodutores ovinos deslanados na Paraíba com os objetivos de verificar a prevalência e distribuição da infecção por B. ovis em propriedades rurais e analisar os possíveis fatores de risco associados à infecção. Foram investigadas 283 propriedades criadoras de ovinos, das quais foram colhidas 498 amostras de soro sanguíneo de carneiros, a partir de oito meses de idade, nas mesorregiões do Sertão Paraibano e Borborema. Todos os soros foram examinados pela IDGA (teste de triagem) e RFC (teste confirmatório). De acordo com as análises, 8,59% (I.C.95% = 5,83%-12,48%) das propriedades apresentaram evidência sorológica de infecção por B. ovis, com uma prevalência de 5,57% (I.C.95% = 3,86%-7,97%) de reprodutores sororreagentes. Nas propriedades que higienizavam suas instalações com maior freqüência, a soropositividade foi estatisticamente inferior (p &lt; 0,05).

Multiplex PCR for the detection ofBrucella ovis,Actinobacillus seminisandHistophilus somniin ram semen

To develop a multiplex polymerase chain reaction (PCR) assay for the rapid detection of Brucella ovis, Actinobacillus seminis, Histophilus somni in fresh ram semen samples. The multiplex assay was based on the single PCR assays published for the detection of A seminis and B ovis, and the forward primer published for the detection of H somni; an alternative reverse primer for H somni was designed in this study. Culture and PCR of 295 fresh semen samples were carried out. The multiplex PCR was far more successful in the detection of H somni (45/295) than culture (23/295). A seminis was also detected in more semen samples by multiplex PCR (29/295) than culture (13/295) and B ovis was detected in three samples using both PCR and culture. No amplifications were detected with DNA from a range of bacterial isolates including species associated with epididymitis in rams. This PCR could be used as a complementary test, or alternative to culture of ram semen and other biological samples for the detection B ovis, H somni and A seminis.

Evaluation of pyrrolidonyl arylamidase for the identification of nonfermenting Gram-negative rods

To evaluate the activity of pyrrolidonyl arylamidase (PYR) for the differentiation and identification of nonfermenting gram negative rods (NFGNR), 293 isolates were tested. A 24 h culture of each test organism was prepared. From this a 108–109 cfu/mL suspension was added to 0.25 mL of sterile physiologic solution. A PYR disk was then added and the test was incubated for 30 minutes at 35–37°C, at environmental atmosphere. Reading was done by adding 1 drop of cinnamaldehyde reagent. Strains of Acinetobacter baumannii, Acinetobacter haemolyticus, Alcaligenes faecalis, Bergeyella zoohelcum, Bordetella bronchiseptica, Bordetella hinzii, Brevundimonas diminuta, Brevundimonas vesicularis, Brucella ovis, Brucella spp., Brucella suis, Burkholderia cepacia complex, Moraxella catarrhalis, Moraxella lacunata, Moraxella nonliquefaciens, Moraxella osloensis, Oligella ureolytica, Pseudomonas alcaligenes, Pseudomonas mendocina, Pseudomonas pseudoalcaligenes, Pseudomonas putida, Pseudomonas stutzeri, Pseudomonas Vb3, Psychrobacter phenylpyruvicus, and Stenotrophomonas maltophilia were PYR negative. On the other hand Achromobacter piechaudii, Achromobacter denitrificans, Achromobacter xylosoxidans, Burkholderia gladioli, Chryseobacterium gleum-indologenes, Comamonas testosroni, Cupriavidus pauculus, Delftia acidovorans, Elizabethkingia meningoseptica, Myroides spp., Ochrobactrum anthropi, Pseudomonas oryzihabitans, Ralstonia pickettii, Rhizobium radiobacter, Shewanella spp., Sphingobacterium multivorum, Sphingobacterium spiritivorum, and Weeksella virosa were PYR positive. Finally, Acinetobacter lwoffii, Pseudomonas aeruginosa, Pseudomonas fluorescens, Roseomonas spp., and Sphingomonas paucimobilis-parapaucimobilis were PYR variable. PYR testing should be considered as a useful tool to facilitate the identification of NFGNR.

Detection of antibodies to Brucella ovis in sheep milk using B. ovis and B. canis antigen

The diagnostic techniques most widely used for detecting brucellosis caused by Brucella ovis are serological tests such as complement fixation (CFT), agar gel immunodiffusion (AGID), and ELISAs. However, to our knowledge, milk tests, with the advantage that samples may be taken in a non invasive manner, have not been investigated as diagnostic tools. We studied 144 samples of milk and sera from lactating ewes, comparing bacteriological studies, serological and milk tests using Brucella canis and B. ovis antigens. A group of 75 ewes in an uninfected flock were serologically negative to rapid slide agglutination test (RSAT), indirect ELISA (IELISA)- B. canis, AGID and IELISA- B. ovis. The milk of these ewes had an IELISA- B. canis mean (% P) value of 16.18 (S.D. 5.63), while the IELISA- B. ovis had a mean (% P) value of 12.52 (S.D. 4.94). A cut-off value of (% P) 27.44 (+2 S.D.) or (% P) 33 (+3 S.D.) was determined by milk-ELISA- B. canis and (% P) 22.4 (+2 S.D.) and (% P) 27.34 (+3 S.D.) by milk-IELISA- B. ovis. These cut-off values were adjusted by receiver–operator characteristics (ROC) analysis using 23 positive samples from infected ewes, which indicated a milk-IELISA- B. canis cut-off value of (% P) 33 and milk-IELISA- B. ovis of (% P) 26 with 100% sensitivity and specificity. Based on our results, we propose that, following a study of a larger number of samples, the milk-IELISA- B. canis could be considered a suitable test for the diagnosis of B. ovis brucellosis in lactating ewes.

CGHScan: finding variable regions using high-density microarray comparative genomic hybridization data

BackgroundComparative genomic hybridization can rapidly identify chromosomal regions that vary between organisms and tissues. This technique has been applied to detecting differences between normal and cancerous tissues in eukaryotes as well as genomic variability in microbial strains and species. The density of oligonucleotide probes available on current microarray platforms is particularly well-suited for comparisons of organisms with smaller genomes like bacteria and yeast where an entire genome can be assayed on a single microarray with high resolution. Available methods for analyzing these experiments typically confine analyses to data from pre-defined annotated genome features, such as entire genes. Many of these methods are ill suited for datasets with the number of measurements typical of high-density microarrays.ResultsWe present an algorithm for analyzing microarray hybridization data to aid identification of regions that vary between an unsequenced genome and a sequenced reference genome. The program, CGHScan, uses an iterative random walk approach integrating multi-layered significance testing to detect these regions from comparative genomic hybridization data. The algorithm tolerates a high level of noise in measurements of individual probe intensities and is relatively insensitive to the choice of method for normalizing probe intensity values and identifying probes that differ between samples. When applied to comparative genomic hybridization data from a published experiment, CGHScan identified eight of nine known deletions in a Brucella ovis strain as compared to Brucella melitensis. The same result was obtained using two different normalization methods and two different scores to classify data for individual probes as representing conserved or variable genomic regions. The undetected region is a small (58 base pair) deletion that is below the resolution of CGHScan given the array design employed in the study.ConclusionCGHScan is an effective tool for analyzing comparative genomic hybridization data from high-density microarrays. The algorithm is capable of accurately identifying known variable regions and is tolerant of high noise and varying methods of data preprocessing. Statistical analysis is used to define each variable region providing a robust and reliable method for rapid identification of genomic differences independent of annotated gene boundaries.