Patent Application
20230160888
The present invention refers to the medical or veterinary field. Particularly, the present invention is directed to an in vitro method for the diagnosis of subjects infected with Mycobacterium species, particularly subjects infected with Mycobacterium avium subsp. paratuberculosis (MAP) which is the causative agent of paratuberculosis (PTB).
MAP is a pathogenic bacterium which pertains to the genus Mycobacterium. MAP causes Johne's disease or PTB, which mainly affects ruminants such as cattle, and it has been associated with some human diseases such as Crohn's disease, ulcerative colitis or colorectal cancer [Pierce E S et al., 2018. Could Mycobacterium avium subspecies paratuberculosis cause Crohn's disease, ulcerative colitis . . . and colorectal cancer? Infect Agent Cancer. 2018 Jan. 4; 13:1. doi: 10.1186/s13027-017-0172-3. eCollection 2018] [Garvey M et al., 2018. Mycobacterium avium subspecies paratuberculosis: A possible causative agent in human morbidity and risk to public health safety. Open Vet J. 2018; 8(2):172-181. doi: 10.4314/ovj.v8i2.10. Epub 2018 May 19]. Moreover, MAP has been associated with an increasingly long list of inflammatory/autoimmune diseases: sarcoidosis, Blau syndrome, Hashimoto's thyroiditis, autoimmune diabetes (T1D), multiple sclerosis (MS), rheumatoid arthritis, lupus and Parkinson's disease [Dow C T et al., 2020. Proposing BCG Vaccination for Mycobacterium avium ss. paratuberculosis (MAP) Associated Autoimmune Diseases. Microorganisms. 2020 Feb. 5; 8(2). PMID: 32033287 PMCID: PMC7074941].
Transmission of MAP primarily occurs by the fecal-oral route through the ingestion of MAP contaminated feces, colostrum, or milk. Infection usually occurs within the first months of live of the animal but remains sub-clinical for an average of 2-5 years. After being ingested, MAP crosses the intestinal mucosa where it is phagocytosed by sub-epithelial macrophages establishing a chronic infection. MAP is able to survive and proliferate within phagosomes by inhibiting apoptosis and phagosome acidification and by preventing presentation of antigens to the immune system. As the infection progresses, the lesions in the intestine and mesenteric lymph nodes become more severe. Rather than focalized, the granulomatous infiltrate becomes diffuse disrupting the mucosal structure and function and affecting jejunum and ileum. According to their location and extension, the histological lesions present in the intestinal tissues of infected animals are classified into focal, multifocal, and diffuse categories [González J et al., 2005. Histopathological classification of lesions associated with natural paratuberculosis infection in cattle. J Comp Pathol. 2005 August-October; 133(2-3): 184-96. PMID: 16045917 DOI: 10.1016/j.jcpa.2005.04.007]. The focal lesions consist of small granulomas, mainly located in the jejunal and ileal lymph nodes. The multifocal lesions consist of well-demarcated granulomas in the intestinal lymphoid tissue and also in the intestinal lamina propia. The diffuse lesions were characterized by severe granulomatous enteritis and lymphadenitis. According to the inflammatory cell type present in the infiltrate and the amount of acid-fast bacilli (AFB), diffuse lesions were subdivided into diffuse lymphoplasmacytic or paucibacillary, intermediate and diffuse histiocytic or multibacillary lesions [Balseiro et al., 2018. Chronic regional intestinal inflammatory disease: A trans-species slow infection? Comp. Immunol., Microbiol., Infect. Dis. 2018. 62:88-100. DOI: https://doi.org/j.cimid.2018.12.001]. In the diffuse lymphoplasmacytic lesion type the cellular infiltrate is composed mainly of lymphocytes that extends through broad areas, contains rare AFBs and is clearly visible to the naked eye. The diffuse histiocytic lesion type consists of large numbers epithelioid cells, lymphocytes, macrophages, and numerous Langhans giant cells with huge amounts of AFBs, that affect large areas of the intestine and appears as a thickening of the intestinal wall readily visible upon opening the intestine. The diffuse intermediate type can also be seen under the naked eye affecting broad areas of the intestine as a thickening of the mucosa microscopically composed of lymphocytes, macrophages and scant Langhans giant cells with small numbers of AFB.
PTB is responsible for significant economic losses in dairy herds worldwide due to decreased milk production, increased management costs and premature culling from clinical disease. Several studies have demonstrated that more than 50% of the dairy cattle herds are positive for MAP antibodies in USA and in Europe and, therefore, bovine PTB can be considered endemic in these areas. Commercial inactivated vaccines against bovine PTB are very successful in reducing MAP presence in feces and tissues and in increasing both milk production and cattle productive life in infected farms when compared with unvaccinated farms. However, PTB vaccination with heat-killed inactivated vaccines is not allowed in most European countries due to its interference with Mycobacterium bovis detection tests. Therefore, PTB control is currently based on testing and culling plus preventing MAP transmission to susceptible animals using appropriate hygienic-sanitary strategies. However, the efficiency of the control programs based on the “test and cull” policy is strongly conditioned by the sensitivity of the tests used to identify early infections. Presently, fecal culture is considered “the gold standard” test for the ante-mortem diagnosis of MAP infection. However, individual fecal culture is expensive and time consuming and only detects advanced infections due to the relatively late onset of fecal shedding during the natural course of MAP infection. In fact, the sensitivity of the fecal culture is 70% in animals with PTB-associated clinical signs, but only 23-29% in animals with sub-clinical infection, which may shed MAP inteimittently and in lower numbers in feces and milk contaminating the environment and infecting other animals
Early stage diagnostics such as the IFN-γ release assay (IGRA) detects whether a T-cell mediated immune response has been elicited in response to mycobacterial antigens. The IGRA method uses a sandwich ELISA to detect the cellular response to MAP by measuring the difference in IFN-γ signals for whole blood samples activated by specific antigen (MAP extract) and non-specific antigen (M. phlei extract) (ID Vet). The IGRA only reflects MAP exposure, and thus cannot discriminate between individuals with controlled infection from those with sub-clinical disease.
The IDEXX ELISA, which detects specific antibodies produced against MAP, is nowadays widely used for the diagnosis of PTB. However, the IDEXX ELISA is associated with a high rate of false negative results (low sensitivity) [Vazquez et al., 2014. Latent infections are the most frequent form of paratuberculosis in slaughtered Friesian cattle. Spanish Journal of Agricultural Research 2014 12(4): 1049-1060. http://dx.doi.org/10.5424/sjar/2014124-5978]. This means that a high percentage of subjects, those latently infected, still test negative. False negative results are a concern because they hamper the implementation of control or eradication measures.
So, the detection of MAP infections, mainly sub-clinical infections before the bacterium is shed and transmitted to herd mates and possibly to humans, is still an unmet need. In other words, novel tools with high sensitivity and specificity are needed to detect MAP-infections, mainly at early stages of the infection and during the long choric phase that precedes the clinical stage of the disease.
The present invention is focused on solving the above cited problems and it is herein provided a new strategy for the diagnosis of subjects infected with MAP, which offers a high sensitivity and specificity, also at early stages or during the long chronic stage of the MAP infection.
Such as it has been explained above, the present invention is directed to an in vitro method for the diagnosis of subjects infected with Mycobacterium species, particularly subjects infected with MAP which is the causative agent of PTB.
Five of the many biomarkers identified by RNA-Seq were selected for the development of novel ELISAs for diagnosis purposes, according to their high expression levels in animals with focal or diffuse histological lesions with respect to the control animals without lesions.
Table 1 shows differential expression of host biomarker genes selected after RNA-Seq analysis of peripheral blood of Holstein cattle with focal and diffuse histological lesions versus control animals without detected lesions. The mean Log2 fold change is a measurement of differential gene expression between two compared groups.
Genes encoding for bovine proteins MMP8 (Matrix metallopeptidase 8) and ABCA13 (ATP binding cassette subfamily A member 13) showed significantly higher expression in animals with focal lesions, while genes encoding for bovine proteins FAM84A (family with sequence similarity 84-member A), SPARC (secreted protein acidic and cystein rich) and DES (desmin) were up-regulated in animals showing diffuse lesions. Biomarkers selection was based on high levels of expression, cellular location (extracellular proteins, present in sera, were the preferred choice for detection by ELISA) and commercial availability of ELISA kits for its specific detection.
The diagnostic potential of five ELISAs designed to detect MMP8, ABCA13, FAM84A, SPARC and DES was studied using serum samples from infected cows with distinct histological lesions in gut tissues representing various stages of the disease. The gold standard or reference technique used to classify the animals included in this study was the type of histological lesions present in their intestinal tissues. Significantly, such as it can be seen in the results provided below, it was not possible to validate the five biomarker-based ELISAs for the diagnosis of PTB. In fact, only three out of the five biomarker-based ELISAs were able to discriminate animals with focal lesions, two out of the five biomarker-based ELISAs biomarkers discriminated animals with multifocal lesions, two out of the five biomarker-based ELISAs discriminated animals with diffuse lesions and three out of the five biomarker-based ELISAs were able to discriminate all the infected animals from the control group. Indeed, the ELISAs for the detection of FAM84A and DES did not discriminate any group of infected animals (their AUC was always below 0.7).
Of note, the ELISAs for the detection of proteins ABCA13 and/or SPARC and /or MMP8 showed a high performance, specifically to discriminate the group of animals with focal lesions and all-lesions group (which comprises all the animals with focal, multifocal and diffuse histological lesions). Such as it is shown in the results provided below, the ELISA for the detection of the ABCA13 showed the most accurate diagnostic performance for both the detection of animals with focal lesions (AUC value of 0.866, p<0.001) and for the overall detection of animals with any type of histological lesions (AUC value of 0.825, p<0.001). Thus, the ELISA for the detection of ABCA13 is an accurate method for detection of animals with focal lesions and for overall detection of animals with any type of histological lesion.
Consequently, ABCA13 is herein presented as a reliable alternative for the methods currently used for the diagnosis of PTB. In fact, IDEXX ELISA is nowadays widely used for the diagnosis of PTB although this method is associated with a high rate of false negative results (low sensitivity). This means that a high percentage of infected subjects still test negative for the disease due to the low sensitivity of IDEXX ELISA. This problem is solved by the present invention because, such as it is herein explained, the ABCA13 showed a high sensitivity and specificity.
So, the first embodiment of the present invention refers to an in vitro method for diagnosing subjects infected with Mycobacterium species which comprises: a) Measuring the concentration of the protein ABCA13 in a biological sample obtained from the subject, and b) wherein if an increase of the concentration of the protein ABCA13 is identified in the step a) with respect to the concentration of the protein ABCA13 measured in uninfected control subjects, it is an indication that the subject is infected with Mycobacterium species.
In a preferred embodiment, the present invention refers to an in vitro method for diagnosing subjects infected with MAP which comprises: a) Measuring the concentration of the protein ABCA13 in a biological sample obtained from the subject, and b) wherein if an increase of the concentration of the protein ABCA13 is identified in the step a) with respect to the concentration of the protein ABCA13 measured in uninfected control subjects, it is an indication that the subject is infected with MAP.
In a preferred embodiment, the present invention refers to an in vitro method for the diagnosis of PTB in a subject which comprises: a) Measuring the concentration of the protein ABCA13 in a biological sample obtained from the subject, and b) wherein if an increase of the concentration of the protein ABCA13 is identified in the step a) with respect to the concentration of the protein ABCA13 measured in uninfected control subjects, it is an indication that the subject is suffering from PTB.
In a preferred embodiment, the above-described in vitro methods comprise: a) Measuring the concentration of the proteins [ABCA13 and SPARC] in a biological sample obtained from the subject, and b) wherein if an increase of the concentration of the proteins [ABCA13 and SPARC] is identified in the step a) with respect to the concentration of the proteins [ABCA13 and SPARC] measured in uninfected control subjects, it is an indication that the subject is suffering from PTB.
In a preferred embodiment, the above-described in vitro methods comprise: a) Measuring the concentration of the proteins [ABCA13 and SPARC and MMP8] in a biological sample obtained from the subject, and b) wherein if an increase of the concentration of the proteins [ABCA13 and SPARC and MMP8] is identified in the step a) with respect to the concentration of the proteins [ABCA13 and SPARC and MMP8] measured in uninfected control subjects, it is an indication that the subject is suffering from PTB.
In a preferred embodiment, the above-described in vitro methods comprise: a) Measuring the concentration of the proteins, in a biological sample obtained from the subject, wherein if the concentration obtained for any of the previously cited proteins is above N the established cut-off value, this is indicative that the subject is suffering from PTB.
In a preferred embodiment, the above-described in vitro methods comprise the diagnosis of latent or patent PTB, preferably latent PTB.
In a preferred embodiment, the biological sample is preferably a body fluid comprising serum, plasma or blood.
In a preferred embodiment, the subject is a mammal selected from the group consisting of: cattle, goat, sheep, horse, pig, deer, rabbit, wild boar, bison, llama, alpaca, opossum, badger, elephant or human being; preferably a farm-animal selected from the group consisting of: cow, goat, sheep, horse, pig, deer, llama or alpaca.
In a preferred embodiment, the above cited method is an immunoassay, preferably enzyme-linked immunosorbent assay (ELISA).
The second embodiment of the present invention refers to the in vitro use of the protein ABCA13, of the proteins [ABCA13 and SPARC], or of the proteins [ABCA1 3 and SPARC and MMP8], for diagnosing subjects infected with Mycobacterium species.
In a preferred embodiment, the proteins are used for diagnosing subjects infected with MAP.
In a preferred embodiment, the proteins are used for the diagnosis of PTB.
In a preferred embodiment, the proteins are used for the diagnosis of latent or patent PTB, preferably latent PTB.
The third embodiment of the present invention refers to a kit adapted for performing any of the above described methods which comprises tools or reagents for determining the concentration the protein ABCA13.
In a preferred embodiment, the kit comprises tools or reagents for determining the concentration of the proteins [ABCA13 and SPARC] or the proteins [ABCA13 and SPARC and MMP8].
In a preferred embodiment the present invention refers to the use of the above cited kit for diagnosing subjects infected with Mycobacterium species.
The fourth embodiment of the present invention refers to a method for treating subjects infected with Mycobacterium species, particularly with MAP, by applying a therapeutically effective dose or amount of a composition which brings about a positive therapeutic response in a subject infected with the bacteria. Significantly, this method comprises a first step wherein the subject is diagnosed by using any of the methods described above.
The fifth embodiment refers to a method for guaranteeing that a batch of milk comes from females free of MAP infection and therefore that the product is apt for human consumption without any significant risk of causing any of the above-mentioned human diseases.
For the purpose of the present invention the following terms are defined:
The following examples disclose a preferred way of carrying out the invention, without the intention of limiting its scope of protection.
Two groups of animals were included in this field study: Slaughtered group) Ninety-three Holstein Friesian cows (ranging from 0.81 to 12.66 years of age) came from 26 farms located in the Principality of Asturias (Northwest of Spain). In Asturias 25% of the herds and 1.9% of the animals were positive by serum ELISA in 2017 (data taken from the records of the Regional Government). Specifically, fifty-five animals came from a dairy farm with a mean herd size of 105 cows (2016-2019) and a mean prevalence of PTB of 6.30% in the sampling period, based on serum ELISA test (IDEXX laboratories, Hoofddorp, The Netherlands). Another thirty-four animals were chosen randomly from cows slaughtered in the local abattoir (coming from 24 different farms) and the remaining four cows came from SERIDA's Friesian cow farm (3% mean prevalence of PTB in the period 2018-2019). The PTB infection status of these 93 animals at the time of slaughter was determined by histopathology, serum ELISA test, and bacteriological culture and specific real-time PCR of tissues and feces. Information related to the presence of PTB-associated clinical signs such as gradual weight loss, diarrhea and decreased milk production was obtained from the farmers when possible; and PTB-free group) Sixty-one animals (ranging from 0.50 to 10.08 years of age) from a PTB-free faun in Asturias were used as the control group for the study. The PTB-free status of this control farm was verified yearly by IDEXX serum ELISA, the absence of clinical cases in the period 2016-2020 as well as by the absence of positive fecal bacteriological culture and PCR results in 2019.
Experimental procedures were approved by the SERIDA Animal Ethics Committee and authorised by the Regional Consejeria de Agroganaderia y Recursos Autoctonos del Principado de Asturias. Spain (authorization codes PROAE 29/2015 and PROAE 66/2019). All the procedures were carried out in accordance with the Directive 2012/63/EU of the European Parliament.
Isolation of genomic DNA from tissues and feces was performed using the MagMax Total Nucleic Acid Isolation kit according to the manufacturer's instructions (TeiinoFisher Scientifc, Lissieu, France). For detection of MAP DNA, the LSI VetMax Triplex real-time PCR was used according to the manufacturer's instructions (TermoFisher Scientifc, Lissieu, France). The kit enables real-time PCR detection of Map IS900 and F57 genes in DNA extracted from feces, liquid cultures, and tissues or colonies. Real-time PCR amplifications were performed using the MX3000P Real-Time PCR detection system (Stratagene, San Diego, USA) system with the following conditions: 1 cycle at 50° C. for 2 min, 1 cycle of 95° C. for 10 min, 45 cycles of denaturation at 95° C. for 15 s, and annealing/extension at 60° C. for 60 s.
For bacteriological culture, a pool (2 gr) of ileocecal lymph nodes, distal jejunal lymph node, ileocecal valve (ICV), and distal jejunum were decontaminated with 38 mL of hexa-decyl pyridinium chloride at a final concentration of 0.75% (Sigma, St. Louis, Mo.) and homogenized in a stomacher blender. After 30 min of incubation at room temperature, 15 mL of the suspension was transferred to a new tube and incubated overnight for decontamination and sedimentation. Approximately, 200 μl of the suspension was taken from the layer near the sediment and inoculated into two slants of Herrolds egg yolk medium (HEYM; Becton Dickinson, Sparks, Md.) and into two slants of Lowenstein-Jensen medium (LJ; Difco, Detroit, Mich.), both supplemented with 2 mg/L of Mycobactin J (ID.vet Innovative Diagnostics, Grabels, France). At the time of slaughter, feces were taken from the rectum of each animal, maintained at 4° C. and processed within 48 h after arrival at the laboratory. The fecal samples (2 g each) were decontaminated, blended in a stomacher, and cultured in HEYM and LJ, as previously described for tissue culture.
As mentioned above animals included in this study were classified according to the type of histological lesions present in their intestinal tissues. Tissue sections of distal jejunum, ICV and jejunum and ICV lymph nodes were collected from the 93 slaughtered cows, fixed in 10% neutral buffered formalin, sliced and embedded in paraffin wax using standard procedures. Afterwards, 4 μm sections were assessed by haematoxylin-eosin (HE) and Ziehl-Neelsen (ZN) staining for specific acid-fast bacteria detection. Slices were observed using an Olympus BH-2 light microscope and photographed using an Olympus DP-12 digital camera. The stained sections were examined by light microscopy for detection and classification of pathological lesions and for the presence of acid-fast bacteria (AFB).
Peripheral blood was collected from the tail vein of all the cows included in the study using BD Vacutainer Z serum clot activator Plastic tubes (Vacuette. Kremsmunster. Austria). After clotting, serum was separated by centrifugation for 20 min at 2,500 g at room temperature and stored at −20° C. until use. The concentrations of the selected biomarkers in the serum of each animal were measured using commercially available ELISAs according to the manufacturers' instructions. Quantitative sandwich ELISA kits Bovine Matrix Metalloproteinase 8 (MMP8) (Detection range 3.12-100 ng/mL); Bovine Protein FAM84A ELISA kit (Detection range 62.5-2000 pg/mL), Bovine SPARC ELISA kit (Detection range 0.78-50 ng/mL), and competitive Bovine ATP-binding cassette sub-family A member 13 (ABCA13) ELISA kit (Detection range 0-5000 pg/mL) and Bovine Desmin (DES) ELISA kit (Detection range 0-25 ng/mL) were used for specific detection of MMP8, FAM84A, ABCA13, SPARC and DES, respectively (MyBioSource, San Diego, Calif. USA).
The assay procedure for the specific detection of biomarker ABCA13 is described in more details as follows:
A standard curve was used to determine the concentration of biomarkers in the serum samples (average OD of each standard was plotted on the vertical axis against the concentration on the horizontal axis and the best fit drawn to generate a regression curve). Standards and samples were tested in duplicate. The mean value of the blank control was subtracted from mean raw OD values before result interpretation. For optimization various dilutions of the serum were tested (for instance: undiluted, 1:2, 1:4 and 1:8) and the dilution which showed a larger number of samples with measurement values included within the range of the standard curve was considered optimal. In most cases the optimal dilution was 1:2, however, samples with high concentrations of a specific biomarker had to be assayed at higher dilutions.
The AUC (area under the curve) and optimal cut-off value for each ELISA was determined by Receiver operator characteristic (ROC) curve analysis. The optimal cut off values for sensitivity and specificity were based on maximum Youden Index (J=Se+Sp−1).
The discriminatory power of each biomarker-based ELISA to differentiate between the different histological groups and the control group was determined according to [Muller, M. P., Tomlinson, G., Marrie, T. J., Tang, P., McGeer, A., Low, D. E., Detsky, A. S., Gold, W. L. (2005). Can routine laboratory tests discriminate between severe acute respiratory syndrome and other causes of community-acquired pneumonia? Clin Infect Dis. 40(8):1079-86. https://doi.org/10.1086/428577][Park H E., Park, H-T., Jung Y H., Yoo, H S. (2017). Establishment a real-time reverse transcription PCR based on host biomarkers for the detection of the sub-clinical cases of Mycobacterium avium subsp. Paratuberculosis. PLoS One. 25; 12(5):e0178336. doi: 10.1371/journal.pone.0178336] as follows: AUC scores≥0.9 were considered to have excellent discriminatory power; 0.8≤AUC <0.9 good discriminatory power; 0.7≤AUC<0.8 fair discriminatory power; and AUC<0.7, poor discriminatory power (Muller et al., 2005). Higher AUC scores were considered as showing better discriminatory powers.
Multivariate binary logistic regression models (Caret package of R) were used to assess the diagnostic capacity of the simultaneous use of several biomarkers, providing the AUC, and values like sensitivity and specificity for the different biomarker combinations. Comparison of ROC curves to test the statistical significance of the difference between the areas under ROC curves (derived from the same cases) was performed with the method of DeLong et al. (1988) [DeLong E R, DeLong D M, Clarke-Pearson D L (1988): Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44:837-845] for the biomarkers with fair, good and excellent AUC values (AUC≥0.7). Statistical significance of differences in quantitative variables (for instance: age) between two histological groups were studied using the t Student test (normality) or Wilcoxon test (not normality).
All the data was analyzed using the pROC, OptimalCutpoints and Caret [Max Kuhn (2020). caret: Classification and Regression Training. R package version 6.0-85. https://CRAN.R-project.org/package=caret] packages of R Statistical environment version 3.6.0 [R Core Team (2018). R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna. Austria. URL https://www.R-projectorg/], with confidence intervals stated at 95%.
The histological, immunological and microbiological characteristics of all the animals included in this study (slaughtered animals n=93 and control animals from a PTB-free farm n=61) are summarized in Table 2.
Thus, Table 2 shows the assessment of MAP infection status in 154 Holstein Friesian cows included in the study. Pathological examination of intestinal tissue sections allowed the classification of the animals in four groups: focal (n=55, 59.14%), multifocal (n=17, 18.28%), diffuse (n=15, 16.12%) and no lesions (n=6, 6.45%).
100%
In the group of animals with focal lesions, 76.36% were positive by one or more diagnostic methods (Ziehl-Neelsen, fecal and tissue real-time PCR, fecal and tissue bacteriological culture and serum ELISA) and the remaining 23.64% were negative by all the tests assayed (n=13). Specifically, 49.09% were positive by Ziehl-Neelsen, 9.09% by fecal real-time PCR, 27.27% by tissue real-time PCR and by tissue bacteriological culture, and 5.45% by fecal bacteriological culture, and serum ELISA. None of the animals in the focal group showed clinical signs associated with PTB, so these animals with focal lesions were considered to be sub-clinically infected.
In the group of animals with multifocal lesions, 100% of the samples were positive for at least one of the following 6 techniques: Ziehl-Neelsen, fecal and tissue real-time PCR, fecal and tissue bacteriological culture and serum ELISA. In particular, 94.11% of the animals were positive by Ziehl-Neelsen, 23.52% by serum ELISA, 29.41% by fecal real-time PCR, 41.17% by tissue real-time PCR, 17.64% by fecal culture and 41.17% by tissue culture. In this group a higher percentage of animals (15.38%) showed clinical signs, including the youngest animal with clinical signs (2.72 years old). In the diffuse group, 100% of the animals were positive by Ziehl-Neelsen, 73.33% were positive by serum ELISA and by fecal and tissue real-time PCR, 20% by fecal culture and 66.66% by tissue culture. In this group, 64.28% of the animals had PTB-associated clinical signs.
Only 6 out of the 94 animals analyzed by histopathology did not show histological lesions in their intestinal tissues. These animals were negative by fecal and tissue MAP-specific PCR and bacteriological culture, and by serum ELISA. Given the difficulty to find animals without lesions it was decided to use as control group 61 animals coming from a PTB-free farm in Asturias in order to increase the size of the sample and the statistical significance of the results. The PTB-free status of this control farm had been verified yearly four times by IDEXX serum ELISA and absence of clinical cases in the period 2016-2019, as well as by bacteriological culture and specific real-time PCR of feces in 2019. However, none of them had been examined histologically since this is a post-mortem test and these animals are still alive.
Significant differences were found between the age of the focal (5.72±2.16 years) and the PTB-free control (3.62±2.41 years) group (p<0.001); and between the age of the multifocal (5.50±1.85 years) and PTB-free control (3.62±2.41 years) group (p=0.009). However, no significant differences were found between the age of the diffuse (4.98±1.58 years) and control (3.62±2.41) groups (p=0.141), focal and diffuse groups (p=0.65), multifocal and diffuse groups (p=0.905), and multifocal and focal groups (0.982).
The diagnostic accuracies of each biomarker-based ELISA to discriminate between the different histological groups and the control group were calculated by ROC analysis (
Three ELISAs for the detection of the biomarkers ABCA13, MMP8 and SPARC had good discriminatory power between the focal and control groups (0.8≤AUC<0.9). However, the ABCA13-based ELISA showed the most accurate diagnostic performance (AUC value of 0.866, p<0.001), with a sensitivity of 79.25% and a high specificity of 93.44%. Combination of ABCA13 and SPARC-based ELISAs gave an AUC value slightly higher 0.951, with a sensitivity of 93.44% and a specificity (80.77%). Diagnostic performance of ABCA13-based ELISA for the detection of animals with focal lesions was better than that of the IDEXX ELISA (AUC value of 0.536), which had a sensitivity of 14.50% and a specificity of 100.00%. No significant differences were observed between the ROC curves of the ABCA13, MMP8 and SPARC-based ELISAs (ABCA13 vs MMP8, p=0.386; ABCA13 vs. SPARC, p=0.372 and MMP8 vs. SPARC, p=0.781). Table 3 shows the diagnostic performance of selected biomarker-based ELISAs for diagnosis of cattle with focal histological lesions in the intestinal tissues. The ELISA with the best diagnostic performance is shown in bold face.
ABCA13
0.866
<0.001
1.74
79.2
93.4
0.863
athe diagnostic accuracy of the IDEXX ELISA was calculated by ROC curve analysis;
bthe cut off value used to calculate the sensitivity and specificity of the assay was calculated using the cut off value established by the supplier.
In the multifocal group the SPARC-based ELISA showed the most accurate diagnostic performance. The SPARC-based ELISA had good discriminatory power (AUC value of 0.865, p<0.001) between the multifocal and control groups, with a sensitivity of 66.70% and a specificity of 95.10%. Combination of the results using the ABCA13 and SPARC-based ELISAs gave an AUC of 0.900 with a sensitivity of 98.36% and a specificity of 64.71%. The diagnostic performance of the ELISA based on detection of SPARC was also compared to that of the IDEXX ELISA. Diagnostic performance of SPARC-based ELISA for detection of animals with multifocal lesions was better than that of the IDEXX ELISA (AUC value of 0.594), which has a sensitivity of 27.80% and a specificity of 100.00%. There were no significant differences between their ROC curves of the MMP8 and SPARC-based ELISAs (MMP8 vs. SPARC, p=0.771).
Table 4 shows the diagnostic performance of selected biomarkers-based ELISAs for diagnosis of cattle with multifocal histological lesions in their intestinal tissues. The ELISA with the best diagnostic performance is shown in bold face.
SPARC
0.865
<0.001
354.94
66.70
95.10
0.809
athe diagnostic accuracy of the IDEXX ELISA was calculated by ROC curve analysis;
bthe cut off value used to calculate the sensitivity and specificity of the assay was calculated using the cut off value established by the supplier.
Two ELISAs for the detection of ABCA13 and MMP8 had good discriminatory power between the diffuse and control groups (0.8≤AUC<0.9). However, MMP8-based ELISA showed the most accurate diagnostic performance (AUC value of 0.866, <0.001) with a sensitivity of 100.00% and a specificity of 77.10%. In this case, combination of biomarkers-based ELISAs does not improve the results obtained individually by the MMP8-based ELISA. The diagnostic performance of the ELISA based on detection of MMP8 was also compared to that of the IDEXX ELISA. The discriminatory power of the IDEXX ELISA between the diffuse and control groups was excellent with an AUC value of 0.917 (p<0.001). The diagnostic performance of the IDEXX ELISA for the detection of animals with diffuse lesions was better than that of the biomarker-based ELISAs, with a sensitivity of 86.70% and a specificity of 96.70%. Comparison of the ROC curves of ABCA13, MMP8 and IDEXX ELISAs showed that there were no significant differences between the three ROC curves (ABCA13 vs MMP8, p=0.843; ABCA13 vs. IDEXX ELISA, p=0.482 and MMP8 vs. IDEXX ELISA, p=0.444).
Table 5 shows the diagnostic performance of selected biomarkers-based ELISAs for diagnosis of cattle with diffuse histological lesions in their intestinal tissues. The ELISA with the best diagnostic performance is shown in bold face.
IDEXX ELISA
a
0.917
<0.001
19.6
86.70
96.70
0.917
athe diagnostic accuracy of the IDEXX ELISA was calculated by ROC curve analysis;
bthe cut off value used to calculate the sensitivity and specificity of the assay was calculated using the cut off value established by the supplier of the kit.
We compared the ability of the biomarker-based ELISAs to discriminate infected cows with any type of histological lesion (focal, multifocal or diffuse) from the control group. It must be taken into account that the three different histological groups are not equally represented (focal n=55, multifocal n=17 and diffuse n=15), however, this is a reflection of the real situation in the farms. The ABCA13 and MMP8-based ELISAs had good discriminatory power (0.8≤AUC<0.9) between the animals with lesions and the control group. However, ABCA13-based ELISA showed the most accurate diagnostic performance (AUC value of 0.825, p<0.001) with a sensitivity of 70.60% and a specificity of 91.80%. The combination of the results obtained with the ABCA13 and SPARC-based ELISAs gave an AUC value slightly higher (0.933) with a sensitivity of 88.52% and a specificity of 79.76%. The diagnostic performance of the ELISA based on detection of ABCA13 was also compared to that of the IDEXX ELISA. Diagnostic performance of ABCA13-based ELISA for detection of infected animals was better than that of the IDEXX ELISA (AUC value of 0.613), which had a sensitivity of 28.40% and a specificity of 100.00%.
Table 6 shows the diagnostic perfoi nance of selected biomarkers-based ELISAs for diagnosis of cattle with histological lesions in their intestinal tissues. The ELISA with the best diagnostic performance is shown in bold face.
ABCA13
0.825
<0.001
1.29
70.60
91.80
0.812
athe diagnostic accuracy of the IDEXX ELISA was calculated by ROC curve analysis;
bthe cut off value used to calculate the sensitivity and specificity of the assay was calculated using the cut off value established by the supplier.
A summary of the ELISAs with the best diagnostic performance for each histological group is provided below. ABCA13-based ELISA showed the most accurate diagnostic performance (AUC value of 0.866, p<0.001) for detection of animals with focal lesions. SPARC-based ELISA showed the most accurate diagnostic performance (AUC value of 0.865, p<0.001) for the detection of animals with multifocal lesions. MMP8-based ELISA showed the most accurate diagnostic performance (AUC value of 0.866, p<0.001) for the detection of animals with diffuse lesions, however, the diagnostic performance of the IDEXX ELISA (AUC value of 0.917, p<0.001) was better than that of the biomarker-based ELISAs with a sensitivity of 86.70% and a specificity of 96.70%. Finally, for overall detection of animals with any type of histological lesions ABCA13-based ELISA showed the most accurate diagnostic performance (AUC value of 0.825, p<0.001) with a sensitivity of 70.60% and a specificity of 91.80%. Moreover, the combination of biomarkers for detection of animals with focal, multifocal or any type of lesions could improve the sensitivity of the diagnosis. Taken together our results show that the ABCA13-based ELISA is a very accurate method for the discrimination of animals with focal lesions and for overall detection of animals with any type of histological lesion.
Table 7 shows the diagnostic performance of the best biomarker-based ELISAs and IDEXX ELISA for the detection of the different histological animal groups. ROC analysis of MMP8 and SPARC-based ELISAs and the IDEXX ELISA was performed using 87 serum samples from 55 animals with focal, 17 with multifocal and 15 with diffuse lesions. ROC analysis of ABCA13-based ELISA was performed using 83 serum samples from 53 animals with focal, 16 with multifocal and 14 with diffuse lesions:
athe diagnostic accuracy of the IDEXX ELISA was calculated by ROC curve analysis
The diagnostic performance of the ELISA based on detection of ABCA13 was also compared to that of conventional PTB diagnosis methods such as specific anti-MAP antibody ELISA (IDEXX ELISA), and specific fecal and tissue real-time PCR and bacteriological culture.
Table 8 shows the diagnostic performance of conventional and novel biomarker-based diagnostic assays for the detection of animals with focal lesions in their intestinal tissues. ABCA13-based ELISA showed a better diagnostic value than the other diagnostic methods tested for the detection and discrimination of animals with focal lesions. It was able to detect as positive 44 out of 55 animals with focal lesions (80% sensitivity) and as negative 57 out of the 61 controls (93% specificity). The diagnostic value or sensitivities of the other conventional methodologies was lower. Specifically, the IDEXX ELISA was 100% specific but only detected as positive 3 out of 55 animals (5.45% sensitivity) when the cut off used (55%) was the one established by the supplier or 8 out of 55 (14.54%) when we used the cut-off point (28.39%) calculated by ROC analysis.
aindicate that the number of animals with focal, multifocal, diffuse or with any type of lesions analysed with theABCA13-based ELISA was 53, 16, 14 and 83, respectively;
bthe cut off value used to estimate the sensitivity and specificity of the assay was calculated by ROC analysis;
cthe cut off value used to estimate the sensitivity and specificity of the assay was the one established by the supplier;
dthe control group used to calculate the specificity consisted of 61 animals from a PTB-free farm.
Diagnostic models based on the combination of some of the five selected biomarkers slightly improve the diagnostic value of the single biomarker-based ELISAs (see Table 9).
Table 9 shows the diagnostic performance of models based on the combined use of various biomarkers compared with the performance obtained for single biomarker-based ELISAs for diagnosis of cattle with different histological lesions in the intestinal tissues. This analysis was carried out using 55 focal, 18 multifocal, 15 diffuse animals and 67 control animals
We had shown that the ELISA for the detection of ABCA13 is an accurate method for detection of animals with focal lesions and for overall detection of animals with any type of histological lesion.
To further confirm these results a large-scale validation of the ABC13-based ELISA (n=661) was performed using 556 serum samples from 445 infected animals with focal, 55 with multifocal and 58 with diffuse lesions. The non-infected control group consisted of 105 animals from two PTB-free farms in Asturias.
Table 10 shows the diagnostic performance of the ABCA13-based ELISA for diagnosis of cattle with different types of histological lesions in their intestinal tissues. The diagnostic accuracy of the ABCA13-based ELISA was calculated by ROC curve analysis.
The ELISA for detection of ABCA13 has a good discriminatory power between the focal and the control group (AUC value of 0.843, p<0.0001) and between the all-lesion group and the control (AUC value of 0.883, p<0.0001). These results further confirm that the ELISA for the detection of ABCA13 is an accurate diagnostic method for detection of animals with focal lesions and for overall detection of animals with any type of histological lesion.
As we have mentioned the IDEXX ELISA is nowadays widely used for the diagnosis of PTB although this method is associated with a high rate of false negative results (low sensitivity). For this reason, we have compared the results obtained with the ABCA13-based ELISA and the IDEXX ELISA for the 661 samples used in the large-scale validation study.
Table 11 shows the sensitivity, specificity, and diagnostic value of the novel ABCA13-based ELISA and the conventional IDEXX ELISA for the detection of the different histological groups.
The cut-off value used to estimate the sensitivity and the specificity of the ABCA13-based ELISA and IDEXX ELISA were 3,66 ng/mL and 55% (relative % ODsample/ODpositive control), respectively. The estimation of the sensitivity was based on the analysis of 445 animals with focal lesions, 55 with multifocal, and 58 with diffuse. The control group used to calculate the specificity consisted of 105 animals from two PTB-free farms.
These findings support our previous results indicating that the ABCA13-based ELISA is a very accurate method for the discrimination of animals with focal lesions and for overall detection of animals with any type of histological lesion. Specifically, the ABC13-based ELISA has a higher sensitivity for detection of animals with focal lesions representing subclinical infections difficult to detect by conventional diagnostic methods.
| Number | Date | Country | Kind |
|---|---|---|---|
| 20382291.1 | Apr 2020 | EP | regional |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2021/059316 | 4/9/2021 | WO |
