The evaluation of novel biomarkers and antigens for the diagnosis of Mycobacterium bovis infection in African buffaloes (Syncerus caffer).

Goosen, Wynand Johan (2016-12)

Thesis (PhD)--Stellenbosch University, 2016.

Thesis

ENGLISH SUMMARY: Mycobacterium bovis (M. bovis) forms part of the Mycobacterium tuberculosis complex (MTC), a group of genetically related bacteria that causes tuberculosis in humans and animals. African buffaloes (Syncerus caffer) are maintenance hosts of M. bovis, the causative organism of bovine tuberculosis (bTB). Since this species acts as a bTB reservoir for a wide range of domestic and wildlife species, the detection of M. bovis-infected animals is essential to control spread of the disease. However, diagnostic assays used for bTB management and control programmes, such as the single intradermal comparative tuberculin test (SICTT) and commercially available interferon-gamma release assays (IGRAs), are still believed to be sub-optimal for the diagnosis of bTB in bovids. A potential approach to improving detection of M. bovis infection could be the use of novel diagnostic antigens or the identification of alternative or ancillary biomarkers to interferon-gamma (IFN-γ). The studies presented in this dissertation aim to identify, develop and evaluate novel approaches for improving the detection of M. bovis infection in African buffaloes. The first objective was to evaluate the performance of two new commercially available IGRAs, the Bovigam® PC-EC assay and the Bovigam® PC-HP assay, for the first time in buffaloes and compare their performance to that of two versions of an adapted human IGRA, the modified QuantiFERON® TB-Gold (mQFT) assay. In addition, the effect of increased blood incubation time on sensitivity of the mQFT assay was assessed along with whether centrifugation was a necessary step prior to harvesting the plasma fraction. Furthermore, the relative sensitivities of Bovigam® assays, a modified Bovigam assay that contains an additional stimulation criteria with Mycobacterium fortuitum tuberculin and the SICTT were compared in identified M. bovis-infected buffaloes. Combinations of these assays were evaluated to identify the optimal test algorithm (i.e., highest sensitivity) for detection of bTB in African buffaloes. Commercially available bovine ELISAs as well as a human IP-10 ELISA were used to identify selected candidate biomarkers of M. bovis-specific immune activation and evaluate their diagnostic utility. Lastly, this study investigated what effect storing stimulated whole blood plasma under various conditions would have on the diagnostic performance of IP-10 assays in African buffaloes. Agreement between the Bovigam® PC-EC and the Bovigam® PC-HP assays was high (κ = 0.86, 95% CI 0.75–0.97) and these detected the greatest number of test-positive animals suggesting that they were the most sensitive assays. Agreement between two versions of the mQFT assay was also high (κ=0.88, 95% CI 0.77-0.98); however, all buffaloes with discordant mQFT results (n=6), including 3 confirmed M. bovis-infected animals, were positive at 30 hours incubation and negative at 20 hours. These results suggest that the mQFT assay is more sensitive using the longer incubation period (i.e., 30 hours). There were no significant differences in IFN-γ concentrations in plasma samples harvested from QFT tubes prior to and after centrifugation, a step which may facilitate plasma sampling over multiple incubation times to improve sensitivity. When the test performance of IFN-γ assays was compared in buffaloes, the Bovigam PPD assay had a relative sensitivity between 91-93% while the sensitivity of the modified PPD assay was between 90-91%. Diagnostic sensitivity was improved by combining one or more IGRA together with the SICTT (95-100%). Investigation of alternative biomarkers to IFN-γ found that IP-10 levels were significantly increased in antigen-stimulated blood samples from M. bovis-infected buffaloes (p < 0.0001). In addition, IP-10 was produced in far greater abundance than IFN-γ, demonstrating its potential as a novel biomarker of bTB in buffaloes. Moreover, using IP-10, agreement between the mQFT assay and the Bovigam assays was increased while the excellent agreement between the Bovigam assays was retained. Since transport and storage of buffalo blood samples are important considerations for development of diagnostic tests for bTB, the effects of heat-inactivation and storage on protein saver cards (PSCs) on IP-10 performance were assessed. Incubation of plasma at 65 °C for 20 min caused no statistically significant loss of IP-10 and this protein could be quantified in plasma stored on PSCs for 2 and 8 weeks. Moreover, for all storage conditions, IP-10 retained its excellent diagnostic characteristics. Diagnostic tests for bTB in wildlife are limited by the lack of species-specific immunological reagents. Logistical constraints make validation of tests, using novel biomarkers such as IP-10, more difficult. These limitations may preclude definitive conclusions about the diagnostic utility of IP-10 in buffaloes. Some of the factors that may have influenced our conclusions include: 1) limited sample size which could affect calculation of appropriate IP-10 test cut-off values; 2) use of antigen-specific whole-blood incubation assay protocols which had been optimized for measurement of IFN-γ rather than IP-10, and 3) lack of post-mortem samples to confirm M. bovis infection status in Bovigam-negative IP-10-positive buffaloes. In conclusion, both the Bovigam® PC-EC assay and the Bovigam® PC-HP assay were shown to be more sensitive than either the SICTT or mQFT assay in its current format. Plasma collected from the QFT tubes prior to centrifugation could be reliably utilized in this assay. Moreover, increasing the blood incubation time from 20h to 30h increased the mQFT assay’s sensitivity. In an additional study, both the Bovigam PPD assay and modified PPD assay displayed greatest sensitivity for the detection of M. bovis-infected buffaloes. The SICTT detected additional IGRA-negative animals and maximum sensitivity was attained when these assays were used in combination. In addition to IFN-γ, IP-10 appears to be a useful marker of immune activation in buffaloes when using a commercially available IP-10 bovine ELISA. IP-10 shows promise as a diagnostic biomarker in M. bovis-infected buffaloes and measurement of IP-10 increased the sensitivity over conventional IGRAs. IP-10 could be measured in plasma stored on PSCs; however, the sensitivity of tests utilizing such samples was reduced with increased storage time. Plasma samples could be heated to 65 ºC for 20 min with no degradation of IP-10, demonstrating the thermal stability of this cytokine. These findings are supported by previous cattle studies that advocate the parallel use of the SICTT and the Bovigam PPD assay. Moreover, these findings also highlight the potential application of IP-10 for the diagnosis of bTB in African buffaloes. Improved sensitivity of M. bovis-specific IGRAs is a significant advantage of using IP-10 as a preferred biomarker in this species. Other advantages of IP-10 are its thermal tolerance and stability on PSCs. These characteristics facilitate movement of diagnostic samples by permitting heat-inactivation of potential pathogens in plasma, and transport of samples by conventional delivery methods, respectively.

AFRIKAANSE OPSOMMING: Mycobacterium bovis (M. bovis) vorm deel van die Mycobacterium tuberculosis-kompleks (MTK), ǹ groep genetiesverwante bakterieë verantwoordelik vir tuberkulose in mense en diere. Die Kaapse buffel (Syncerus caffer) is ǹ reservoir (instandhoudingsgasheer) vir M. bovis, die bakterie verantwoordelik vir beestuberkulose (bTB). Omdat hierdie spesie as reservoir van infeksie vir ǹ verskeidenheid van plaasmak- en wilde diere dien, is die suksesvolle diagnose van M. bovis-geïnfekteerde diere noodsaaklik om die verspreiding van die siekte te kan beheer. Daar word egter nog steeds geglo dat die diagnostiese toetse wat gebruik word gedurende bTB-beheerprogramme, soos die enkele intradermale vergelykende tuberkulien-toets (“single intradermal comparative tuberculin test”; SICTT) en kommersieel beskikbare interferon-gamma-vrystellingstoetse (“interferon gamma release assays”; IGRA’s), nog steeds suboptimaal is vir die diagnose van bTB in beesverwante hoefdiere. ǹ Moontlike benadering tot die verbetering van sulke toetse mag dalk die gebruik wees van nuwe antigene of die identifisering van alternatiewe of aanvullende biomerkers anders as interferon-gamma (IFN-γ). Hierdie proefskrif bespreek studies wat ontwerp is om nuwe en oorspronklike benaderings te identifiseer, te ontwikkel en te evalueer met die doel om die diagnose van M. bovis-infeksie in Kaapse buffels te verbeter. Die eerste doelwit was om die diagnostiese potensiaal van twee nuwe, beskikbare, kommersiële IGRA’s, die Bovigam® PC-EC-toets en die Bovigam® PC-HP-toets, vir die eerste keer in buffels te evalueer, asook om hulle prestasie te vergelyk met diè van ǹ gemodifieerde mens-IGRA (“modified QuantiFERON® TB-Gold (mQFT) assay”). Verder is die uitwerking van ǹ verlengde inkubasietyd van bloed op die sensitiwiteit van die mQFT-toets bepaal, asook die noodsaaklikheid van sentrifugering voor isolering van die plasmafraksie van die bloed. ǹ Opvolgstudie het die relatiewe sensitiwiteit van die Bovigam-toetse, ǹ gemodifiseerde Bovigam-toets en die SICTT in geïdentifiseerde M. bovis-geïnfekteerde buffels vergelyk. Die gebruik van hierdie toetse in kombinasie is ook geëvalueer in ǹ poging om die optimale toetsalgoritme vir die diagnose van M. bovis-geïnfekteerde buffels te identifiseer. Kommersieel beskikbare bees-ELISA’s, asook ǹ mens-IP-10-ELISA is verder gebruik om geselekteerde kandidaatbiomerkers van M. bovis-spesifieke immuunaktivering te identifiseer, en om hulle diagnostiese potensiaal te evalueer. Laastens het hierdie studie ook bepaal wat die uitwerking van die stoor van bloedplasma onder verskillende omstandighede op die diagnostiese potensiaal van die gebruik van IP-10-toetse in buffels sal wees. Ooreenstemming tussen die Bovigam® PC-EC- en die Bovigam® PC-HP-toetse was hoog (κ = 0.86, 95% CI 0.75–0.97) en hierdie toetse het die hoogste aantal toetspositiewe diere geïdentifiseer, wat aandui dat hulle die mees sensitiewe toetse was. Ooreenstemming tussen die mQFT-toetse was ook hoog (κ=0.88, 95% CI 0.77-0.98). Alle buffels met verskillende mQFT-resultate (n=6), asook 3 definitiewe M. bovis-geïnfekteerde diere, was egter positief na 30 uur inkubasie, maar negatief na 20 uur. Hierdie resultate impliseer dat die mQFT-toets die mees sensitiewe van die twee is wanneer langer inkubasieperiodes gebruik word. Verder is daar ook geen betekenisvolle verskil in IFN-γ-konsentrasies waargeneem tussen plasmafraksies geïsoleerd vanaf QFT-buise voor en na sentrifugering nie. Hierdie bevinding beteken dat verskillende inkubasieperiodes in ǹ poging om die toetssensitiwiteit te verbeter, aanvaarbaar is. Wanneer die betroubaarheid van die IFN-γ-toetse vergelyk is in buffels, het die Bovigam PPD-toets ǹ relatiewe sensitiwiteit van tussen 91-93% getoon, terwyl die sensitiwiteit van die gemodifieerde PPD-toets tussen 90-91% was. Diagnostiese sensitiwiteit is verder verbeter deur een of meer IGRA’s met die SICTT te kombineer (95-100%). Verdere navorsing oor alternatiewe biomerkers anders as IFN-γ in ǹ poging om die sensitiwiteit van IGRA’s te verhoog, het getoon dat IP-10-vlakke ǹ betekenisvolle verhoging in antigeen-gestimuleerde bloedmonsters van M. bovis-geïnfekteerde buffels toon (p < 0.0001). Daar is ook aangetoon dat IP-10 in groter hoeveelhede as IFN-γ geproduseer is, ǹ eienskap wat grootliks kan bydra tot IP-10 se potensiaal as ǹ nuwe biomerker vir die diagnose van bTB in buffels. In aansluiting hierby is vasgestel dat deur IP-10 te gebruik, daar ǹ verbetering in ooreenstemming waargeneem is tussen die mQFT-toets en die Bovigam-toetse, terwyl die uitstekende ooreenstemming tussen die Bovigam-toetse onveranderd gebly het. Aangesien die vervoer en stoor van buffelbloed ǹ belangrike aspek is om by die ontwikkeling van diagnostiese toetse vir bTB in ag te neem, is die uitwerking van hitte-inaktivering en die stoor van plasma op proteïenstoorkaarte (“protein saver cards”; PSC’s) op IP-10 se betroubaarheid as biomerker ondersoek. Daar is vasgestel dat die inkubasie van bloedplasma teen 65 °C vir 20 min geen betekenisvolle vernietiging van IP-10 getoon het nie en dat IP-10 nog steeds suksesvol gemeet kon word in plasma nà storing op PSC’s vir 2 weke en 8 weke onderskeidelik. Verder het IP-10 uitstekende diagnostiese potensiaal onder alle stoortoestande behou. Die onbeskikbaarheid van spesiesspesifieke immunologiese reagense beperk diagnostiese toetse vir bTB in wild grootliks. Verder is die validering van diagnostiese toetse, deur die gebruik van nuwe biomerkers soos IP-10, bemoeilik deur sekere logistieke beperkings. Sulke beperkings sluit in: 1) beperkte dieregetalle in ǹ studie wat die akkurate berekeninge van ǹ toepaslike afsnypunt vir IP-10 kan benadeel; 2) die gebruik van antigeenspesifieke toetsprotokolle vir bloedinkubasie wat alleenlik geoptimiseer is vir die meting van IFN-γ en nie noodwendig vir IP-10 nie, en 3) die gebrek aan nadoodse monsters vir die bevestiging van M. bovis-infeksie in Bovigam-negatiewe IP-10-positiewe buffels. Ten slotte, die Bovigam® PC-EC-toets en die Bovigam® PC-HP-toets was meer sensitief as albei die SICTT- of die mQFT-toetse in hulle huidige formaat. Bloedplasma verkry vanaf die QFT-buise voor sentrifugering kon met vertroue gebruik word in die mQFT-toets. Verlenging van die inkubasietyd van bloed in die QFT-buise van 20 uur tot 30 uur, het gelei tot die aansienlike verbetering van die mQFT-toets se sensitiwiteit. Volgens die resultate van ǹ aanvullende studie het die Bovigam PPD-toets, asook ǹ gemodifiseerde PPD-toets die beste sensitiwiteit vir die identifisering van M. bovis-geïnfekteerde buffels getoon. Die SICTT het addisionele IGRA-negatiewe diere geïdentifiseer en die hoogste sensitiwiteit is bereik wanneer hierdie toetse in kombinasie gebruik is. Addisioneel tot IFN-γ, wil dit ook voorkom of IP-10 as ǹ nuttige merker van immuunaktivering in buffels kan dien deur gebruik te maak van ǹ IP-10-bees-ELISA, wat kommersieel beskikbaar is. IP-10 toon verder goeie potensiaal as ǹ diagnostiese biomerker in M. bovis-geïnfekteerde buffels en die gebruik daarvan dui ook op ǹ verhoging in die sensitiwiteit van konvensionele IGRA’s. IP-10 kon suksesvol gemeet word in plasma gestoor op PSC’s, alhoewel die sensitiwiteit verminder het met ǹ verlengde stoortydperk in die toetse op hierdie monsters. Plasma verhit tot 65 ºC vir 20 min het geen degradering van IP-10 getoon nie en hierdie eienskap is ǹ verdere demonstrasie van die hittestabiliteit van hierdie sitokien. Hierdie bevindinge word ondersteun deur vorige studies in beeste wat die gesamentlike gebruik van die SICTT- en die Bovigam PPD-toets sterk aanbeveel. Die huidige studie het ook grootliks die potensiaal van IP-10 vir die diagnose van bTB in die Kaapse buffel beklemtoon. Die verbetering in sensitiwiteit van M. bovis-spesifieke IGRA’s, deur die gebruik van IP-10, is ǹ betekenisvolle voordeel van die gebruik van hierdie sitokien as voorkeurbiomerker in hierdie spesie. ǹ Verdere voordeel van diè sitokien is sy hittetoleransie en stabiliteit op PSC’s. Hierdie nuttige eienskappe mag daarop dui dat diagnostiese monsters veilig vervoer kan word deur potensiële patogene in die plasma deur hitte te inaktiveer, of deur alternatiewelik gebruik te maak van konvensionele vervoer- of afleweringsmetodes.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/100384
This item appears in the following collections: