Veterinary Immunology and Immunopathology 166 (2015) 125–131
Contents lists available at ScienceDirect
Veterinary Immunology and Immunopathology journal homepage: www.elsevier.com/locate/vetimm
Specific faecal antibody responses in sheep infected with Mycobacterium avium subspecies paratuberculosis D.J. Begg, K. de Silva, K.M. Plain, A.C. Purdie, N. Dhand, R.J. Whittington ∗ Farm Animal and Veterinary Public Health, Faculty of Veterinary Science, The University of Sydney, Australia
a r t i c l e
i n f o
Article history: Received 19 December 2014 Received in revised form 18 June 2015 Accepted 22 June 2015 Keywords: Mycobacterium avium subspecies paratuberculosis Johne’s disease Sheep Antibody Faeces
a b s t r a c t Many studies have examined the serum antibody response to Mycobacterium avium subspecies paratuberculosis (MAP) infection in cases of Johne’s disease (JD), but there are no reports on the mucosal antibody response. Faecal immunoglobulin (Ig) G and IgA ELISA responses were examined from sheep experimentally inoculated with MAP for up to 23 months post inoculation (PI). Corresponding serum IgG responses and the presence of viable MAP shed in faeces were also examined. The sheep were divided into three groups: (i) “un-inoculated controls” (n = 10), (ii) “clinical cases” (n = 8) which were inoculated animals that developed clinical disease and had moderate to high levels of MAP shedding and (iii) “survivors” (n = 11) which were inoculated animals from which MAP could not be cultured from tissues at the conclusion of the trial. Serum IgG responses gradually increased in all inoculated animals, peaking at 12–16 months PI. A significant increase in the levels of MAP-specific faecal IgG and IgA was measured in the survivors at 16 and 17 months PI, while levels in the un-inoculated controls and clinical cases remained at baseline levels. The detection of faecal Ig in the survivors coincided with the removal of sheep that developed clinical disease. The data suggest that some sheep produced MAP-specific IgG and IgA in the intestinal mucosa, which was released into their faeces. We hypothesise that the survivors produced faecal Ig as a direct response to ingestion of MAP associated with environmental contamination from clinical cases. Thus MAP specific mucosal antibodies may play a previously unreported role as part of a protective response triggered by environmental exposure. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Infection of ruminants with Mycobacterium avium subspecies paratuberculosis (MAP) can result in a chronic enteritis called Johne’s disease (JD). On farm, the disease represents a considerable economic cost to farmers (Bush et al., 2006). The time from infection to appearance of disease may be several years with diagnostic tests not accurate until later periods. Culture of MAP is currently considered to be the most sensitive test for ante mortem detection of MAP infection in sheep, although this technique is time consuming and expensive. Detection of the host’s immune response to MAP is most frequently performed via antibody detection ELISA carried out on serum or milk samples; these assays are high throughput and relatively low cost. However, the sensitivity of MAP-specific ELISAs
∗ Corresponding author at: University of Sydney, Faculty of Veterinary Science, 425 Werombi Road, Camden 2570, NSW, Australia. Tel.: +61 2 93511619; fax: +61 2 93511618. E-mail address: [email protected] (R.J. Whittington). http://dx.doi.org/10.1016/j.vetimm.2015.06.011 0165-2427/© 2015 Elsevier B.V. All rights reserved.
are generally well below 50%, depend on the stage of disease and vary by animal species and test used (Gumber et al., 2006; Nielsen, 2008). The specificities of MAP ELISAs are high (85–100%), although considered not to be perfect (Nielsen and Toft, 2008). Most of the commercially available MAP ELISA tests including the IDEXX Pourquier serum antibody ELISA detect the IgG isotype of antibody. Studies have been done to examine other antibody isotypes such as IgG1, IgM and IgA in various ruminant species; the findings indicate that MAP-specific serum antibodies are of the IgG isotype, specifically IgG1 (Abbas and Riemann, 1988; Griffin et al., 2005). Recently, studies on tuberculosis in humans have suggested that using a combination of antigens to detect multiple antibody isotypes may improve the predictive diagnostic outcome (Baumann et al., 2014; Feng et al., 2014). IgA is thought to play an important role in the immune protection of mucosal surfaces and is the predominant isotype in the intestinal mucosa. In ruminants IgG is transferred into the intestinal mucosa, but is thought to be degraded by proteases which may limit its protective efficacy (Cripps et al., 1974). Interestingly antibody isotypes IgG, IgA and IgM specific to viral infections can be recovered from the faeces of cattle (Heckert et al., 1991; Parreno
126
D.J. Begg et al. / Veterinary Immunology and Immunopathology 166 (2015) 125–131
et al., 2004). Examination of mice indicated that there is a correlation between the concentration of IgA in intestinal lavage fluid compared to a matched faecal sample (Grewal et al., 2000). These findings indicate that a faecal sample from a MAP infected animal might contain immunoglobulins specific for MAP. Abbas and Riemann (1988) found very little MAP-specific IgA in the serum of MAP infected cattle but these authors did not test faecal samples. There are no published studies in which faeces from MAP-infected animals have been examined for antibodies that may be specific to the infection. In the context of mycobacterial infections, intestinal aspirates, but not serum, of patients with lepromatous leprosy were deficient in IgA (Saha et al., 1978), indicating that a reduction in the amount of mucosal IgA may be associated with disease development. Secretory Igs play an important role in the intestines, protecting against pathogenic organisms and modulating gut inflammatory homeostasis (Campos-Rodriguez et al., 2013). Examination of faecal IgA and IgG responses may provide information regarding JD pathogenesis, in particular mucosal immunity, throughout the course of the disease. Additionally, the diagnostic potential of detecting these antibody isotypes should be determined. This longitudinal study examined MAP-antigen specific IgG and IgA responses in faecal samples collected from sheep experimentally exposed to MAP compared to unexposed control animals. Faecal samples were also cultured to detect the level of MAP shedding, and matching serum samples were tested for antigen-specific IgG. 2. Methods 2.1. Animals Thirty Merino wether lambs aged 3 months were sourced from a flock in Armidale, New South Wales (NSW), an area that has no prior history of JD. Absence of MAP infection was confirmed through extensive whole flock faecal tests and serum antibody ELISA (Begg et al., 2010). The animals were moved to a JD – free quarantine farm at the University of Sydney, Camden, NSW and maintained under conventional Australian sheep farming conditions by grazing on open pasture, with unexposed control sheep kept in separate paddocks to the inoculated sheep. 2.2. Ethical considerations All animal experiments were conducted with the approval of the University of Sydney Animal Ethics Committee. 2.3. Experimental inoculations The 30 Merino lambs were systematically randomised into two groups. The first 10 sheep were used as the un-inoculated, unexposed controls. The remaining 20 lambs were inoculated orally as described by Begg et al. (2010) with a total of 3.18 × 109 viable MAP strain Telford 9.2, a clonal culture at passage level 5, isolated from sheep faeces and characterised as IS1311 S strain (Marsh et al., 2006; Marsh and Whittington, 2007). 2.4. Ante-mortem sampling Faecal samples were collected directly from the rectum and blood samples via jugular venipuncture. Sample collection was performed prior to inoculation with MAP then repeated every 1–3 months post inoculation (PI) until necropsy to monitor the progress of the infection, and the controls were also sampled. Faecal samples were stored at −80 ◦ C until required. Serum from the blood samples was stored at −20 ◦ C until required.
2.5. Identification of clinically diseased sheep The sheep were culled if they lost more than 10% body weight over a 1-month period. All animals were necropsied as previously described (Begg et al., 2010). Animals culled for weight loss were confirmed to have clinical JD by detection of gross and histopathological lesions consistent with the disease (Perez et al., 1996) and were shown to be infected with MAP as determined by culture of intestinal tissues as described by Begg et al. (2010) and Whittington et al. (1999). Twelve gut tissues, including ileum, jejunum and associated lymph nodes were collected from each sheep for analysis. 2.6. Faecal culture for MAP Faecal samples were processed and cultured in modified BACTEC 12B medium (Becton Dickinson) containing egg yolk, PANTA-PLUS and mycobactin J as described previously (Whittington et al., 1999). Briefly, after hexadecylpyridinium chloride (HPC), vancomycin, amphotericin B, and nalidixic acid (VAN) decontamination to reduce overgrowth by other bacteria, samples were inoculated into modified BACTEC 12B media and incubated for 12 weeks at 37 ◦ C. Growth index positive samples were confirmed by IS900 PCR. The weeks to maximal growth (999) were recorded. The weeks to maximal growth can be used as an indicator of the concentration of MAP in the faeces, in that greater numbers of viable MAP in the faeces require fewer weeks to reach maximal growth in culture (Reddacliff et al., 2003). 2.7. Serum antibody ELISA Serum antibody ELISA (Institute Pourquier (now IDEXX), France) was conducted according to the manufacturer’s instructions following the method described previously (Gumber et al., 2006). Results were expressed as signal of the test sample as a proportion of the positive control, corrected for the negative control (S/P%). 2.8. Preparation of faecal samples for ELISA Faecal samples were removed from the −80 ◦ C freezer and thawed at room temperature. Using aseptic technique, 0.5 g (±0.05 g) of the faecal sample was removed and placed into a 5 mL tube. A 1/10 dilution of the faeces was made by adding 4.5 mL of PBS to each half gram of faeces. The samples were then mixed vigorously on a vortex mixer, incubated at 4 ◦ C overnight and mixed again. A sterile wooden stick was used to break up any clumps in the sample and the sample was mixed vigorously. Using a transfer pipette, 600 L was then placed in a 1.5 mL tube. The samples then were either frozen at −20 ◦ C until required or processed immediately. 2.9. Faecal IgG and IgA assays ELISA plates (Nunc Maxisorb) were coated with 50 L per well of 2.5 g/mL MAP 316v French pressed antigen (EMAI, NSW Department of Primary Industries) diluted in carbonate buffer (0.1 M, pH 9.6) and stored at 4 ◦ C overnight. The ELISA plates were then machine washed 5 times (Tecan, Austria) using wash buffer (reverse osmosis water with 0.05% v/v Tween 20). The 1/10 diluted faecal samples were thawed if required and centrifuged at 3000 × g for 5 min. An aliquot of the supernatant (10 L) was removed and diluted into 790 L of PBS from which duplicate 50 L aliquots were added to wells of the ELISA plate as required. A dilution of 1/800 for the faecal samples was chosen after a set of faecal sample dilutions were tested between 1/60 and 1/16,000. The curves developed indicated that the optimal dilution was 1/800. This dilution gave the best differentiation between faecal samples for both IgA and IgG
D.J. Begg et al. / Veterinary Immunology and Immunopathology 166 (2015) 125–131
determinations. All samples were tested at 1/400, 1/800 and 1/1600 dilutions with similar results seen for all dilutions, therefore only the results for 1/800 are presented. The plates were then incubated at 37 ◦ C for 1 h. The plates were washed as described earlier and the conjugated antibody, either anti-ovine IgG or anti-ovine IgA was added at 50 L per well. The conjugated antibody for the detection of IgG was a monoclonal mouse anti-goat/ovine IgG conjugated to HRP, clone GT34 (Sigma) diluted 1/20,000. The conjugated antibody for the detection of IgA was a polyclonal rabbit anti-ovine IgA conjugated to HRP (Serotec) diluted 1/10,000 to 0.1 g/mL. The plates were incubated for 1 h at 37 ◦ C and machine-washed 5 times. TMB substrate (100 L) (Pierce) was added to each well and the plates were incubated at room temperature in the dark for 20 min. To stop the reaction, 50 L of 2 M sulphuric acid was added to each well and the plates were read using an ELISA plate reader (Multiskan Ascent, Thermo Electric Corporation) to determine the optical density (OD) at 450 nm. Mean OD results for duplicate samples are presented for the MAP antigen binding capacity test. Data for the longitudinal animal trial were standardised by calculating the S/P% (as for the serum ELISA) using known positive and negative controls for IgG or IgA on all plates. The controls were sourced from a MAP exposed sheep with high levels of IgG and IgA specific to MAP in its faeces and from an unexposed animal. Aliquots of these faeces were stored and used on every test plate. 2.10. MAP antigen binding capacity test To gain an understanding of the MAP antigen binding capacity of the IgG and IgA antibodies present within faeces, a titration experiment was carried out. This experiment examined how much MAP needs to be added to a sample to quench the faecal IgG and IgA response. A faecal sample from a MAP inoculated animal, with a high faecal antibody load (IgG S/P% 121 and IgA S/P% 106) and a faecal sample from an unexposed control sheep with low levels of faecal antibodies (IgG S/P% 9.8 and IgA S/P% 4.3) were compared. Both samples were collected at 16 months PI and no MAP could be detected by faecal culture. These faecal samples were diluted 1/400, 1/800 and 1/1600 in PBS according to the ELISA method and then spiked with varying amounts of suspensions of heat killed MAP (70 ◦ C for 2 h), incubated over night at 4 ◦ C and then processed in the ELISA. Two replicates were tested for each sample. Only data from the 1/800 dilutions are presented as the results were reproduced at the other dilutions. 2.11. Statistical analysis Initially, descriptive analyses were conducted to evaluate distributions of variables. These included calculation of summary statistics and creation of histograms and boxplots for numeric variables and frequency tables for categorical variables. Serum antibody, faecal IgG and faecal IgA values were log transformed for further analyses to meet the assumptions of normality and equal variance for statistical models. A small value was added to all serum antibody, faecal IgG and faecal IgA values before log transformation to avoid the error due to calculation of the log of zero values. Three linear mixed model analyses were then built to evaluate the differences in means of serum antibody, faecal IgG and faecal IgA values between experimental groups over time. Log serum antibody, faecal IgG and faecal IgA values were specified as outcomes. Experimental group, time and their interaction were fixed effects, while animal was a random effect. In addition, an autoregressive correlation structure was imposed on the model to account for correlation of values over time. Similar analyses were conducted to evaluate the association between faecal culture status and log serum antibody, faecal IgG and faecal IgA over time. Predicted means and their standard errors were estimated and
127
then back-transformed for presentation. Residual diagnostics were used to evaluate the assumptions of statistical models. The analyses were conducted in SAS statistical program (version 9.3© 2002–2010 by SAS Institute Inc., Cary, NC, USA). All P-values reported in the manuscript are two sided. 3. Results 3.1. Experimental inoculation outcomes An experimental inoculation trial was performed including 10 aged-matched unexposed “control” sheep (N◦ 1–10) and 20 MAPinoculated sheep (N◦ 11–30) (Table 1). The animals were followed for 2½ years PI unless clinical disease was evident (10% body weight loss in 1 month), at which stage the individuals were necropsied (Begg et al., 2010). Eight of the 20 MAP-inoculated sheep developed clinical disease, confirmed by gross and microscopic lesions associated with JD and positive culture for MAP from intestinal tissues. This group is referred to as “clinical cases”. One animal (N◦ 29) at necropsy 30 months PI, while not having clinical disease, did have gross and microscopic lesions consistent with JD and MAP was cultured from its tissues and faeces; this animal with subclinical disease was removed from the analysis of the surviving animals. The remaining 11 “survivors” were all negative by tissue culture and did not have gross pathology consistent with JD. One of these 11 animals (N◦ 27) had a grade 2 microscopic lesion (Perez et al., 1996) in an anterior to middle mesenteric lymph node, although all the other 11 gut associated tissues sampled in this sheep had no significant histopathological lesions associated with JD. The first of the clinically affected sheep (N◦ 24) developed clinical disease just after the 12 months PI sampling (Table 1). A further three animals (N◦ 13, 16 and 21) were necropsied due to clinical JD development one week before the 16 month sampling. One week after the 16 month sampling a further 3 animals (N◦ 15, 22 and 25) developed clinical disease and were necropsied. The final sheep (N◦ 28) to develop clinical disease was necropsied at 21 months PI. 3.2. Faecal shedding from the inoculated sheep Throughout the trial, 16 of the inoculated sheep had detectable faecal shedding (Table 1). Of the animals that developed clinical disease, all had detectable persistent faecal shedding for 6–11 months prior to necropsy. These animals had varying amounts of detectable MAP in their faeces as demonstrated by the weeks to maximal growth in culture, an indirect determinant of the amount of MAP, which varied from 3 weeks (∼105 viable MAP/g) to 7 weeks (∼1.5 viable MAP/g) (Reddacliff et al., 2003) at the last sampling for the clinically affected sheep. Thus MAP faecal contamination of the environment from the inoculated sheep increased from 7 to 16 months PI because of the increase in number of sheep shedding and the amount of MAP shed per sheep. Seven of the 8 clinical animals were most likely shedding the highest amount of MAP during the 12–16 months PI period. This was supported by testing of faecal samples collected at necropsy from animals N◦ 13, 16 and 21. These samples had time to MAP maximal growth of 3, 6 and 4 weeks, respectively, which was lower than the 14 month PI sampling (Table 1). Beyond 23 months PI, only one animal, N◦ 29, was intermittently faecal shedding until the end of the trial at 30 months (data not shown). 3.3. Serum antibody responses MAP-specific serum antibody levels in the inoculated sheep increased gradually throughout the trial until 12 months PI. When the sheep that developed clinical disease were compared to the sheep that survived until the end of the trial, there appeared to be a
128
D.J. Begg et al. / Veterinary Immunology and Immunopathology 166 (2015) 125–131 Table 1 Longitudinal faecal culture results for individual animals.
Inoculated
Unexposed controls
Months post-inoculation Animal 0 N 1 2 3 4 5 6 7 8 9 10
0
1
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29* 30
2
3
5
6
4
5
6
>12
7
8
10
12
14
16
6 6 6
5
5
4
3
3
>12
5 5
12 6
4 5
4 7
4 6
7 6 6
5 5
6 6
4 4
7 5
4 3
4 6
3 4
3 6
3 4
5
5
6
5 6
6 6
5 5
6
12 6
6
17
19
21
23
30
6 6
4
3 6
5
5
5 6
>12
5
5 6
5
6
>12
5 6
5
5 6
White boxes indicate a negative culture result. Grey boxes with a number inside indicate a faecal sample from which MAP was cultured; the number indicates the weeks to maximal growth in BACTEC 12B culture. Black boxes indicate the animal was culled due to onset of clinical JD. White numbers in the black boxes indicate weeks to maximal growth from faecal samples collected at necropsy. Faecal samples collected at necropsy from the remaining animals were not tested. Animals 1–10 were unexposed controls while 11–30 were inoculated with MAP. After 23 months PI only animal 29 was observed to be intermittently faecal shedding MAP (data not shown). Red shaded boxes indicate the faecal IgG SP% detected from the sample was >50%. No clinical cases generated a faecal IgG SP% >50%. * Animal 29 was not included in the faecal antibody analysis of clinical cases, survivors or controls.
trend for the clinical cases to have lower serum antibody levels until 7 months PI (Fig. 1). At later time points the MAP-specific serum antibody responses tended to be higher in clinical cases compared to the surviving sheep and this effect was greatest at 12 months PI, however the differences did not reach significance due to variation in the responses of individual sheep, particularly those within the clinical group. The high antibody level from the survivors in the first 6 months of the trial was mainly due to 4 sheep which had S/P%
values greater than 80 within this period. For example, animal N◦ 12 had an S/P% of 126 at 1 month PI; this remained high until 12 months and then gradually dropped to S/P% of 46 at 23 months. 3.4. Faecal IgG and IgA responses The faecal IgG antibody levels specific to MAP were similar between the un-inoculated controls, the clinical cases and the
Geometric mean serum anbody SP%
180 Control
160
Survivors 140 Clinicals cases 120
#
100 80 60 40 20 0 0
2
4
6
8
10
12
14
16
18
20
22
24
Months post inoculaon Fig. 1. Geometric mean serum MAP-specific IgG responses from clinical, survivor and unexposed control sheep. Serum antibodies were measured by Institute Pourquier ELISA (now IDEXX). The inoculated sheep were further categorised as those that developed clinical disease and those that survived until the end of the trial and were tissue culture negative. Error bars are the standard error of the geometric means. # After 17 months only one animal remained from the sheep that developed clinical disease, therefore error bars are not presented for this treatment from 17 to 21 months post inoculation.
D.J. Begg et al. / Veterinary Immunology and Immunopathology 166 (2015) 125–131
129
120
Geometric Mean IgG SP%
100
*
80
Control
60
Survivors
#
Clinical cases
40 20 0 0
2
4
6
8
10 12 14 16 Months post inoculaon
18
20
22
24
Fig. 2. Geometric mean MAP-specific faecal IgG responses from clinical, survivor and unexposed control sheep. Arrows indicate times at which multiple sheep were removed from the trial due to the onset of clinical disease; three sheep on each occasion. * Survivors are significantly different from clinical cases and controls P < 0.001. Error bars are the standard error of the geometric means. # After 17 months only one animal remained from the sheep that developed clinical disease, therefore error bars are not presented for this treatment from 17 to 21 months post inoculation.
survivors until the sampling point at 16 months PI (Fig. 2). At this time the survivors had a significant (P < 0.001) spike in the faecal IgG level which remained high at 17 months PI. However, by 19 months PI the response had returned to baseline levels, similar to the control animals. The MAP-specific faecal IgA antibody levels showed a similar pattern to the MAP-specific faecal IgG responses (Fig. 3). A significant (P < 0.001) peak was seen at the 16 month PI time point and by 19 months, the response was similar to the level observed from the faecal samples of the unexposed controls.
of the unexposed animal, indicating all available antibodies had been adsorbed by binding to MAP antigens.
4. Discussion The results from this study show for the first time that faecal samples from ruminants in a MAP exposed flock contain MAPspecific IgG and IgA, although it is only measurable by ELISA at certain times. The MAP inoculated sheep with increased amounts of faecal antibodies at 16 months PI were not clinical cases, shedding no detectable or low levels of MAP in their faeces and did not develop clinical disease in the next 3–4 months. Faecal antibodies were detected in MAP inoculated animals soon after removal of sheep that had been shedding high numbers of MAP onto pasture and had to be culled due to clinical disease. The faecal antibody response dropped to background levels several months after removal of these clinical cases. The detection of mucosal antibodies in these sheep may be indicative of a protective response triggered by environmental exposure, that had the effect of sequestering MAP passing through the gut to prevent or limit the infection. This theory is supported by the experiment conducted
3.5. Binding capacity of the IgG and IgA in faecal samples to MAP To determine the amount of MAP that could be bound by a faecal sample containing a high amount MAP specific IgG and IgA a titration experiment was performed. The results showed that 1.6–4 × 1011 MAP/g of faeces was required to bind all of the detectable anti-MAP IgG and IgA from the faeces based on replicate dilution curves (Fig. 4). This was determined by the point at which the level of IgG and IgA antibodies detected by ELISA in the faecal sample from the inoculated animal had dropped to the level 70
Geometric mean IgA SP%
60 50
* #
40
Control Survivors
30
Clinical cases 20 10 0 0
2
4
6
8
10
12
14
16
18
20
22
24
Months post inoculaon Fig. 3. Geometric mean MAP-specific faecal IgA responses from clinical, survivor and unexposed control sheep. Arrows indicate times at which multiple sheep were removed from the trial due to the onset of clinical disease; three sheep on each occasion. * Survivors are significantly different from clinical cases and controls P < 0.001. Error bars are the standard error of the geometric means. # After 17 months only one animal remained from the sheep that developed clinical disease, therefore error bars are not presented for this treatment from 17 to 21 months post inoculation.
130
D.J. Begg et al. / Veterinary Immunology and Immunopathology 166 (2015) 125–131
A
1.4
OD at 450nm
1.2 1.0 0.8 0.6
Unexposed
0.4
Inoculated
0.2 8.0E+11
4.0E+11
8.0E+10
4.0E+10
8.0E+09
4.0E+09
8.0E+08
0.0E+00
0.0
MAP added per gram of faeces
B
1.6 1.4
OD at 450nm
1.2 1.0 0.8 0.6
Unexposed
0.4
Inoculated
0.2 8.0E+11
4.0E+11
8.0E+10
4.0E+10
8.0E+09
4.0E+09
8.0E+08
0.0E+00
0.0
MAP added per gram of faeces Fig. 4. MAP binding capacity of faecal antibodies measured by titration of MAP suspension in faeces from an unexposed sheep and an inoculated sheep. The later had high levels of faecal IgG/IgA but was culture negative for MAP. (A) IgG ELISA. (B) IgA ELISA. Results from one of two replicate assays are shown with similar results obtained for both experiments. Data shown are mean OD values of the 1/800 faecal dilution ± the standard deviation of two technical replicates.
to test the MAP antigen binding capacity of faecal antibodies. The addition of exogenous MAP to a faecal sample containing a high level of IgG and IgA antibodies resulted in a reduction in the level of free antibodies in a dose responsive manner. Exposure to non-pathogenic environmental mycobacteria in field situations is common and in this study it is possible the sheep were exposed to these. A small increase in the faecal IgG response at 14 months post inoculation from all the groups including controls not inoculated with MAP may have been related to exposure to environmental mycobacteria or other unknown factors causing biological variation at this sampling point. After this time point, the faecal IgG responses returned to base line for the control animals, compared to the responses in the inoculated animals which were correlated to the clinical outcome, indicating MAP exposure as the driver of the later faecal Ig responses observed. The exact role that antibody plays in protection against mycobacterial infections such as JD is unknown. The importance of mucosal (secretory) antibodies in protection against mycobacterial infection in general is poorly understood. A study using a mouse model for tuberculosis indicated that secretory antibodies in the mucosa are important in reducing bacterial load (Alvarez et al., 2013). Treatment of mice with secretory IgA prior to inoculation resulted in a significant reduction in recoverable bacteria up to 60 days later and fewer lesions (Alvarez et al., 2013). Significant increases in IgA levels have also been associated with enhanced protection with novel vaccines against tuberculosis (Ai et al., 2013) and conversely an intestinal IgA deficiency has been observed in lepromatous leprosy patients (Saha et al., 1978). These studies,
taken in combination with the data presented here indicate that there may be an association between mucosal antibodies and disease outcome in systemic mycobacterial infections. The amount of MAP required to quench the IgA/IgG faecal ELISAs in the binding capacity assays was higher than expected, up to 4 × 1011 MAP/g of faeces. This is most likely related to the very high faecal IgG and IgA SP% of the sample, unknown efficiency of antibody–MAP binding within the faeces. The high antibody levels in the faeces might be due to heavy contamination of viable and non-viable MAP by the clinically affected animals which were later removed. Most likely an elevated level of faecal antibodies might be an immune mechanism to cope with high numbers of ingested MAP. Clinically affected animals can excrete 109 viable MAP per gram of faeces and sheep can produce a kilogram of faeces each day (Whitlock et al., 2005; Whittington et al., 2000). Accidental ingestion of 1 gram of MAP contaminated faeces by a sheep could expose the animal to a large amount of viable and non-viable MAP. The results lead to speculation: (i) that MAP ingestion leads to stimulation of mucosal IgG and IgA, (ii) the more MAP ingested the more that is adsorbed by the mucosal Ig and that (iii) the amount of faecal Igs detected by ELISA is determined by the level of stimulation and the level of adsorption. Ways to examine this hypothesis in more detail could involve using experimentally or naturally MAP exposed animals that are not shedding MAP in faeces, and remove them from further MAP exposure. Later the animals could be exposed to different sized oral boluses of viable or non-viable MAP. This would determine if antigen or viable MAP exposure is required to stimulate the mucosal Ig response, and whether a
D.J. Begg et al. / Veterinary Immunology and Immunopathology 166 (2015) 125–131
measurable Ig response is abolished by adsorption and further our understanding of immune protection against MAP infection. The high serum IgG antibody response from the inoculated animals that survived the infection was also a surprising result, suggesting a problem with assay specificity. The published specificity of this assay is 98.8% (Gumber et al., 2006), but 4 surviving animals had more than 7 positive antibody tests determined by using a cut-off point of 70 S/P% throughout the trial. None of these surviving animals had histological lesions associated with JD and were tissue culture negative at necropsy. Two did not have viable MAP in faecal samples collected at any time throughout the trial. The study design used by Gumber et al. (2006) to develop specificity data may have resulted in work up and case control biases (Nielsen et al., 2011). Although costly and time consuming, a cross sectional study with full pathological and tissue culture examination maybe be required on a proportion of animals from random flocks to gain a better estimate of specificity (Nielsen et al., 2011). The findings show that sheep exposed to MAP have a mucosal immune response, detectable as Ig in the faeces. The assays developed to detect faecal IgG and IgA may not be of use in the diagnosis of JD as the MAP-specific faecal antibodies were only detected under certain conditions. The results lead to speculation that MAP ingestion leads to stimulation of mucosal IgG and IgA, but detection of the faecal antibody response could not be seen until the source of the ingested adsorptive load was suddenly removed. The amount of faecal Igs detected by ELISA maybe determined by the level of stimulation and the level of adsorption by recently ingested MAP, suggesting that mucosal immunity may play a role in protecting against re-infection. Conflict of interest statement This work was supported by Meat and Livestock Australia and by Cattle Council of Australia, Sheepmeat Council of Australia and WoolProducers Australia through Animal Health Australia. These supporting groups had no involvement in the study design, in the collection, analysis and interpretation of data; in the writing of the manuscript and in the decision to submit the manuscript for publication. Acknowledgements The authors would like to thank Nicole Carter, Ann Michele Whittington, and Anna Waldron for providing laboratory assistance with the ELISAs and cultures. Craig Kristo, Nobel Toribio, Lee White and James Dalton who assisted with the field work. Dr. Graeme Eamens, Department of Primary Industry, Elizabeth Macarthur Agricultural Institute who supplied the MAP 316v antigen. This work was supported by Meat and Livestock Australia and by Cattle Council of Australia, Sheepmeat Council of Australia and WoolProducers Australia through Animal Health Australia. References Abbas, B., Riemann, H.P., 1988. IgG, IgM and IgA in the serum of cattle naturally infected with Mycobacterium paratuberculosis. Comp. Immunol. Microbiol. Infect. Dis. 11, 171–175. Ai, W., Yue, Y., Xiong, S., Xu, W., 2013. Enhanced protection against pulmonary mycobacterial challenge by chitosan-formulated polyepitope gene vaccine is associated with increased pulmonary secretory IgA and gamma-interferon(+) T cell responses. Microbiol. Immunol. 57, 224–235. Alvarez, N., Otero, O., Camacho, F., Borrero, R., Tirado, Y., Puig, A., Aguilar, A., Rivas, C., Cervantes, A., Falero-Diaz, G., Cadiz, A., Sarmiento, M.E., Norazmi, M.N., Hernandez-Pando, R., Acosta, A., 2013. Passive administration of purified
131
secretory IgA from human colostrum induces protection against Mycobacterium tuberculosis in a murine model of progressive pulmonary infection. BMC Immunol. 14 (Suppl. 1), S3. Baumann, R., Kaempfer, S., Chegou, N.N., Oehlmann, W., Loxton, A.G., Kaufmann, S.H., van Helden, P.D., Black, G.F., Singh, M., Walzl, G., 2014. Serologic diagnosis of tuberculosis by combining Ig classes against selected mycobacterial targets. J. Infect. 69, 581–589. Begg, D.J., de Silva, K., Di Fiore, L., Taylor, D.L., Bower, K., Zhong, L., Kawaji, S., Emery, D., Whittington, R.J., 2010. Experimental infection model for Johne’s disease using a lyophilised, pure culture, seedstock of Mycobacterium avium subspecies paratuberculosis. Vet. Microbiol. 141, 301–311. Bush, R.D., Windsor, P.A., Toribio, J.A., 2006. Losses of adult sheep due to ovine Johne’s disease in 12 infected flocks over a 3-year period. Aust. Vet. J. 84, 246–253. Campos-Rodriguez, R., Godinez-Victoria, M., Abarca-Rojano, E., Pacheco-Yepez, J., Reyna-Garfias, H., Barbosa-Cabrera, R.E., Drago-Serrano, M.E., 2013. Stress modulates intestinal secretory immunoglobulin A. Front. Integr. Neurosci. 7, 86. Cripps, A.W., Husband, A.J., Lascelle, A.K., 1974. Origin of immunoglobulins in intestinal secretion of sheep. Aust. J. Exp. Biol. Med. Sci. 52, 711–716. Feng, X., Yang, X., Xiu, B., Qie, S., Dai, Z., Chen, K., Zhao, P., Zhang, L., Nicholson, R.A., Wang, G., Song, X., Zhang, H., 2014. IgG, IgM and IgA antibodies against the novel polyprotein in active tuberculosis. BMC Infect. Dis. 14, 336. Grewal, H.M., Karlsen, T.H., Vetvik, H., Ahren, C., Gjessing, H.K., Sommerfelt, H., Haneberg, B., 2000. Measurement of specific IgA in faecal extracts and intestinal lavage fluid for monitoring of mucosal immune responses. J. Immunol. Methods 239, 53–62. Griffin, J.F., Spittle, E., Rodgers, C.R., Liggett, S., Cooper, M., Bakker, D., Bannantine, J.P., 2005. Immunoglobulin G1 enzyme-linked immunosorbent assay for diagnosis of Johne’s disease in red deer (Cervus elaphus). Clin. Diagn. Lab. Immunol. 12, 1401–1409. Gumber, S., Eamens, G., Whittington, R.J., 2006. Evaluation of a Pourquier ELISA kit in relation to agar gel immunodiffusion (AGID) test for assessment of the humoral immune response in sheep and goats with and without Mycobacterium paratuberculosis infection. Vet. Microbiol. 115, 91–101. Heckert, R.A., Saif, L.J., Mengel, J.P., Myers, G.W., 1991. Isotype-specific antibody responses to bovine coronavirus structural proteins in serum, feces, and mucosal secretions from experimentally challenge-exposed colostrum-deprived calves. Am. J. Vet. Res. 52, 692–699. Marsh, I.B., Bannantine, J.P., Paustian, M.L., Tizard, M.L., Kapur, V., Whittington, R.J., 2006. Genomic comparison of Mycobacterium avium subsp. paratuberculosis sheep and cattle strains by microarray hybridization. J. Bacteriol. 188, 2290–2293. Marsh, I.B., Whittington, R.J., 2007. Genomic diversity in Mycobacterium avium: single nucleotide polymorphisms between the S and C strains of M. avium subsp. paratuberculosis and with M. a. avium. Mol. Cell. Probes 21, 66–75. Nielsen, S.S., 2008. Transitions in diagnostic tests used for detection of Mycobacterium avium subsp. paratuberculosis infections in cattle. Vet. Microbiol. 132, 274–282. Nielsen, S.S., Toft, N., 2008. Ante mortem diagnosis of paratuberculosis: a review of accuracies of ELISA, interferon-gamma assay and faecal culture techniques. Vet. Microbiol. 129, 217–235. Nielsen, S.S., Toft, N., Gardner, I.A., 2011. Structured approach to design of diagnostic test evaluation studies for chronic progressive infections in animals. Vet. Microbiol. 150, 115–125. Parreno, V., Bejar, C., Vagnozzi, A., Barrandeguy, M., Costantini, V., Craig, M.I., Yuan, L., Hodgins, D., Saif, L., Fernandez, F., 2004. Modulation by colostrum-acquired maternal antibodies of systemic and mucosal antibody responses to rotavirus in calves experimentally challenged with bovine rotavirus. Vet. Immunol. Immunopathol. 100, 7–24. Perez, V., Garcia Marin, J.F., Badiola, J.J., 1996. Description and classification of different types of lesion associated with natural paratuberculosis infection in sheep. J. Comp. Pathol. 114, 107–122. Reddacliff, L.A., Nicholls, P.J., Vadali, A., Whittington, R.J., 2003. Use of growth indices from radiometric culture for quantification of sheep strains of Mycobacterium avium subsp. paratuberculosis. Appl. Environ. Microbiol. 69, 3510–3516. Saha, K., Agarwal, S.K., Misra, R.C., 1978. Gut-associated IgA deficiency in lepromatous leprosy. Scand. J. Immunol. 8, 397–402. Whitlock, R.H., Sweeney, R.W., Fyock, T., Smith, J., 2005. MAP super-shedders: another factor in the control of Johne’s disease. In: Manning, E.J., Nielsen, S.S. (Eds.), 8th International Colloquium on Paratuberculosis. Copenhagen, Denmark. Whittington, R.J., Marsh, I., McAllister, S., Turner, M.J., Marshall, D.J., Fraser, C.A., 1999. Evaluation of modified BACTEC 12B radiometric medium and solid media for culture of Mycobacterium avium subsp. paratuberculosis from sheep. J. Clin. Microbiol. 37, 1077–1083. Whittington, R.J., Reddacliff, L.A., Marsh, I., McAllister, S., Saunders, V., 2000. Temporal patterns and quantification of excretion of Mycobacterium avium subsp. paratuberculosis in sheep with Johne’s disease. Aust. Vet. J. 78, 34–37.
Contents lists available at ScienceDirect
Veterinary Immunology and Immunopathology journal homepage: www.elsevier.com/locate/vetimm
Specific faecal antibody responses in sheep infected with Mycobacterium avium subspecies paratuberculosis D.J. Begg, K. de Silva, K.M. Plain, A.C. Purdie, N. Dhand, R.J. Whittington ∗ Farm Animal and Veterinary Public Health, Faculty of Veterinary Science, The University of Sydney, Australia
a r t i c l e
i n f o
Article history: Received 19 December 2014 Received in revised form 18 June 2015 Accepted 22 June 2015 Keywords: Mycobacterium avium subspecies paratuberculosis Johne’s disease Sheep Antibody Faeces
a b s t r a c t Many studies have examined the serum antibody response to Mycobacterium avium subspecies paratuberculosis (MAP) infection in cases of Johne’s disease (JD), but there are no reports on the mucosal antibody response. Faecal immunoglobulin (Ig) G and IgA ELISA responses were examined from sheep experimentally inoculated with MAP for up to 23 months post inoculation (PI). Corresponding serum IgG responses and the presence of viable MAP shed in faeces were also examined. The sheep were divided into three groups: (i) “un-inoculated controls” (n = 10), (ii) “clinical cases” (n = 8) which were inoculated animals that developed clinical disease and had moderate to high levels of MAP shedding and (iii) “survivors” (n = 11) which were inoculated animals from which MAP could not be cultured from tissues at the conclusion of the trial. Serum IgG responses gradually increased in all inoculated animals, peaking at 12–16 months PI. A significant increase in the levels of MAP-specific faecal IgG and IgA was measured in the survivors at 16 and 17 months PI, while levels in the un-inoculated controls and clinical cases remained at baseline levels. The detection of faecal Ig in the survivors coincided with the removal of sheep that developed clinical disease. The data suggest that some sheep produced MAP-specific IgG and IgA in the intestinal mucosa, which was released into their faeces. We hypothesise that the survivors produced faecal Ig as a direct response to ingestion of MAP associated with environmental contamination from clinical cases. Thus MAP specific mucosal antibodies may play a previously unreported role as part of a protective response triggered by environmental exposure. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Infection of ruminants with Mycobacterium avium subspecies paratuberculosis (MAP) can result in a chronic enteritis called Johne’s disease (JD). On farm, the disease represents a considerable economic cost to farmers (Bush et al., 2006). The time from infection to appearance of disease may be several years with diagnostic tests not accurate until later periods. Culture of MAP is currently considered to be the most sensitive test for ante mortem detection of MAP infection in sheep, although this technique is time consuming and expensive. Detection of the host’s immune response to MAP is most frequently performed via antibody detection ELISA carried out on serum or milk samples; these assays are high throughput and relatively low cost. However, the sensitivity of MAP-specific ELISAs
∗ Corresponding author at: University of Sydney, Faculty of Veterinary Science, 425 Werombi Road, Camden 2570, NSW, Australia. Tel.: +61 2 93511619; fax: +61 2 93511618. E-mail address: [email protected] (R.J. Whittington). http://dx.doi.org/10.1016/j.vetimm.2015.06.011 0165-2427/© 2015 Elsevier B.V. All rights reserved.
are generally well below 50%, depend on the stage of disease and vary by animal species and test used (Gumber et al., 2006; Nielsen, 2008). The specificities of MAP ELISAs are high (85–100%), although considered not to be perfect (Nielsen and Toft, 2008). Most of the commercially available MAP ELISA tests including the IDEXX Pourquier serum antibody ELISA detect the IgG isotype of antibody. Studies have been done to examine other antibody isotypes such as IgG1, IgM and IgA in various ruminant species; the findings indicate that MAP-specific serum antibodies are of the IgG isotype, specifically IgG1 (Abbas and Riemann, 1988; Griffin et al., 2005). Recently, studies on tuberculosis in humans have suggested that using a combination of antigens to detect multiple antibody isotypes may improve the predictive diagnostic outcome (Baumann et al., 2014; Feng et al., 2014). IgA is thought to play an important role in the immune protection of mucosal surfaces and is the predominant isotype in the intestinal mucosa. In ruminants IgG is transferred into the intestinal mucosa, but is thought to be degraded by proteases which may limit its protective efficacy (Cripps et al., 1974). Interestingly antibody isotypes IgG, IgA and IgM specific to viral infections can be recovered from the faeces of cattle (Heckert et al., 1991; Parreno
126
D.J. Begg et al. / Veterinary Immunology and Immunopathology 166 (2015) 125–131
et al., 2004). Examination of mice indicated that there is a correlation between the concentration of IgA in intestinal lavage fluid compared to a matched faecal sample (Grewal et al., 2000). These findings indicate that a faecal sample from a MAP infected animal might contain immunoglobulins specific for MAP. Abbas and Riemann (1988) found very little MAP-specific IgA in the serum of MAP infected cattle but these authors did not test faecal samples. There are no published studies in which faeces from MAP-infected animals have been examined for antibodies that may be specific to the infection. In the context of mycobacterial infections, intestinal aspirates, but not serum, of patients with lepromatous leprosy were deficient in IgA (Saha et al., 1978), indicating that a reduction in the amount of mucosal IgA may be associated with disease development. Secretory Igs play an important role in the intestines, protecting against pathogenic organisms and modulating gut inflammatory homeostasis (Campos-Rodriguez et al., 2013). Examination of faecal IgA and IgG responses may provide information regarding JD pathogenesis, in particular mucosal immunity, throughout the course of the disease. Additionally, the diagnostic potential of detecting these antibody isotypes should be determined. This longitudinal study examined MAP-antigen specific IgG and IgA responses in faecal samples collected from sheep experimentally exposed to MAP compared to unexposed control animals. Faecal samples were also cultured to detect the level of MAP shedding, and matching serum samples were tested for antigen-specific IgG. 2. Methods 2.1. Animals Thirty Merino wether lambs aged 3 months were sourced from a flock in Armidale, New South Wales (NSW), an area that has no prior history of JD. Absence of MAP infection was confirmed through extensive whole flock faecal tests and serum antibody ELISA (Begg et al., 2010). The animals were moved to a JD – free quarantine farm at the University of Sydney, Camden, NSW and maintained under conventional Australian sheep farming conditions by grazing on open pasture, with unexposed control sheep kept in separate paddocks to the inoculated sheep. 2.2. Ethical considerations All animal experiments were conducted with the approval of the University of Sydney Animal Ethics Committee. 2.3. Experimental inoculations The 30 Merino lambs were systematically randomised into two groups. The first 10 sheep were used as the un-inoculated, unexposed controls. The remaining 20 lambs were inoculated orally as described by Begg et al. (2010) with a total of 3.18 × 109 viable MAP strain Telford 9.2, a clonal culture at passage level 5, isolated from sheep faeces and characterised as IS1311 S strain (Marsh et al., 2006; Marsh and Whittington, 2007). 2.4. Ante-mortem sampling Faecal samples were collected directly from the rectum and blood samples via jugular venipuncture. Sample collection was performed prior to inoculation with MAP then repeated every 1–3 months post inoculation (PI) until necropsy to monitor the progress of the infection, and the controls were also sampled. Faecal samples were stored at −80 ◦ C until required. Serum from the blood samples was stored at −20 ◦ C until required.
2.5. Identification of clinically diseased sheep The sheep were culled if they lost more than 10% body weight over a 1-month period. All animals were necropsied as previously described (Begg et al., 2010). Animals culled for weight loss were confirmed to have clinical JD by detection of gross and histopathological lesions consistent with the disease (Perez et al., 1996) and were shown to be infected with MAP as determined by culture of intestinal tissues as described by Begg et al. (2010) and Whittington et al. (1999). Twelve gut tissues, including ileum, jejunum and associated lymph nodes were collected from each sheep for analysis. 2.6. Faecal culture for MAP Faecal samples were processed and cultured in modified BACTEC 12B medium (Becton Dickinson) containing egg yolk, PANTA-PLUS and mycobactin J as described previously (Whittington et al., 1999). Briefly, after hexadecylpyridinium chloride (HPC), vancomycin, amphotericin B, and nalidixic acid (VAN) decontamination to reduce overgrowth by other bacteria, samples were inoculated into modified BACTEC 12B media and incubated for 12 weeks at 37 ◦ C. Growth index positive samples were confirmed by IS900 PCR. The weeks to maximal growth (999) were recorded. The weeks to maximal growth can be used as an indicator of the concentration of MAP in the faeces, in that greater numbers of viable MAP in the faeces require fewer weeks to reach maximal growth in culture (Reddacliff et al., 2003). 2.7. Serum antibody ELISA Serum antibody ELISA (Institute Pourquier (now IDEXX), France) was conducted according to the manufacturer’s instructions following the method described previously (Gumber et al., 2006). Results were expressed as signal of the test sample as a proportion of the positive control, corrected for the negative control (S/P%). 2.8. Preparation of faecal samples for ELISA Faecal samples were removed from the −80 ◦ C freezer and thawed at room temperature. Using aseptic technique, 0.5 g (±0.05 g) of the faecal sample was removed and placed into a 5 mL tube. A 1/10 dilution of the faeces was made by adding 4.5 mL of PBS to each half gram of faeces. The samples were then mixed vigorously on a vortex mixer, incubated at 4 ◦ C overnight and mixed again. A sterile wooden stick was used to break up any clumps in the sample and the sample was mixed vigorously. Using a transfer pipette, 600 L was then placed in a 1.5 mL tube. The samples then were either frozen at −20 ◦ C until required or processed immediately. 2.9. Faecal IgG and IgA assays ELISA plates (Nunc Maxisorb) were coated with 50 L per well of 2.5 g/mL MAP 316v French pressed antigen (EMAI, NSW Department of Primary Industries) diluted in carbonate buffer (0.1 M, pH 9.6) and stored at 4 ◦ C overnight. The ELISA plates were then machine washed 5 times (Tecan, Austria) using wash buffer (reverse osmosis water with 0.05% v/v Tween 20). The 1/10 diluted faecal samples were thawed if required and centrifuged at 3000 × g for 5 min. An aliquot of the supernatant (10 L) was removed and diluted into 790 L of PBS from which duplicate 50 L aliquots were added to wells of the ELISA plate as required. A dilution of 1/800 for the faecal samples was chosen after a set of faecal sample dilutions were tested between 1/60 and 1/16,000. The curves developed indicated that the optimal dilution was 1/800. This dilution gave the best differentiation between faecal samples for both IgA and IgG
D.J. Begg et al. / Veterinary Immunology and Immunopathology 166 (2015) 125–131
determinations. All samples were tested at 1/400, 1/800 and 1/1600 dilutions with similar results seen for all dilutions, therefore only the results for 1/800 are presented. The plates were then incubated at 37 ◦ C for 1 h. The plates were washed as described earlier and the conjugated antibody, either anti-ovine IgG or anti-ovine IgA was added at 50 L per well. The conjugated antibody for the detection of IgG was a monoclonal mouse anti-goat/ovine IgG conjugated to HRP, clone GT34 (Sigma) diluted 1/20,000. The conjugated antibody for the detection of IgA was a polyclonal rabbit anti-ovine IgA conjugated to HRP (Serotec) diluted 1/10,000 to 0.1 g/mL. The plates were incubated for 1 h at 37 ◦ C and machine-washed 5 times. TMB substrate (100 L) (Pierce) was added to each well and the plates were incubated at room temperature in the dark for 20 min. To stop the reaction, 50 L of 2 M sulphuric acid was added to each well and the plates were read using an ELISA plate reader (Multiskan Ascent, Thermo Electric Corporation) to determine the optical density (OD) at 450 nm. Mean OD results for duplicate samples are presented for the MAP antigen binding capacity test. Data for the longitudinal animal trial were standardised by calculating the S/P% (as for the serum ELISA) using known positive and negative controls for IgG or IgA on all plates. The controls were sourced from a MAP exposed sheep with high levels of IgG and IgA specific to MAP in its faeces and from an unexposed animal. Aliquots of these faeces were stored and used on every test plate. 2.10. MAP antigen binding capacity test To gain an understanding of the MAP antigen binding capacity of the IgG and IgA antibodies present within faeces, a titration experiment was carried out. This experiment examined how much MAP needs to be added to a sample to quench the faecal IgG and IgA response. A faecal sample from a MAP inoculated animal, with a high faecal antibody load (IgG S/P% 121 and IgA S/P% 106) and a faecal sample from an unexposed control sheep with low levels of faecal antibodies (IgG S/P% 9.8 and IgA S/P% 4.3) were compared. Both samples were collected at 16 months PI and no MAP could be detected by faecal culture. These faecal samples were diluted 1/400, 1/800 and 1/1600 in PBS according to the ELISA method and then spiked with varying amounts of suspensions of heat killed MAP (70 ◦ C for 2 h), incubated over night at 4 ◦ C and then processed in the ELISA. Two replicates were tested for each sample. Only data from the 1/800 dilutions are presented as the results were reproduced at the other dilutions. 2.11. Statistical analysis Initially, descriptive analyses were conducted to evaluate distributions of variables. These included calculation of summary statistics and creation of histograms and boxplots for numeric variables and frequency tables for categorical variables. Serum antibody, faecal IgG and faecal IgA values were log transformed for further analyses to meet the assumptions of normality and equal variance for statistical models. A small value was added to all serum antibody, faecal IgG and faecal IgA values before log transformation to avoid the error due to calculation of the log of zero values. Three linear mixed model analyses were then built to evaluate the differences in means of serum antibody, faecal IgG and faecal IgA values between experimental groups over time. Log serum antibody, faecal IgG and faecal IgA values were specified as outcomes. Experimental group, time and their interaction were fixed effects, while animal was a random effect. In addition, an autoregressive correlation structure was imposed on the model to account for correlation of values over time. Similar analyses were conducted to evaluate the association between faecal culture status and log serum antibody, faecal IgG and faecal IgA over time. Predicted means and their standard errors were estimated and
127
then back-transformed for presentation. Residual diagnostics were used to evaluate the assumptions of statistical models. The analyses were conducted in SAS statistical program (version 9.3© 2002–2010 by SAS Institute Inc., Cary, NC, USA). All P-values reported in the manuscript are two sided. 3. Results 3.1. Experimental inoculation outcomes An experimental inoculation trial was performed including 10 aged-matched unexposed “control” sheep (N◦ 1–10) and 20 MAPinoculated sheep (N◦ 11–30) (Table 1). The animals were followed for 2½ years PI unless clinical disease was evident (10% body weight loss in 1 month), at which stage the individuals were necropsied (Begg et al., 2010). Eight of the 20 MAP-inoculated sheep developed clinical disease, confirmed by gross and microscopic lesions associated with JD and positive culture for MAP from intestinal tissues. This group is referred to as “clinical cases”. One animal (N◦ 29) at necropsy 30 months PI, while not having clinical disease, did have gross and microscopic lesions consistent with JD and MAP was cultured from its tissues and faeces; this animal with subclinical disease was removed from the analysis of the surviving animals. The remaining 11 “survivors” were all negative by tissue culture and did not have gross pathology consistent with JD. One of these 11 animals (N◦ 27) had a grade 2 microscopic lesion (Perez et al., 1996) in an anterior to middle mesenteric lymph node, although all the other 11 gut associated tissues sampled in this sheep had no significant histopathological lesions associated with JD. The first of the clinically affected sheep (N◦ 24) developed clinical disease just after the 12 months PI sampling (Table 1). A further three animals (N◦ 13, 16 and 21) were necropsied due to clinical JD development one week before the 16 month sampling. One week after the 16 month sampling a further 3 animals (N◦ 15, 22 and 25) developed clinical disease and were necropsied. The final sheep (N◦ 28) to develop clinical disease was necropsied at 21 months PI. 3.2. Faecal shedding from the inoculated sheep Throughout the trial, 16 of the inoculated sheep had detectable faecal shedding (Table 1). Of the animals that developed clinical disease, all had detectable persistent faecal shedding for 6–11 months prior to necropsy. These animals had varying amounts of detectable MAP in their faeces as demonstrated by the weeks to maximal growth in culture, an indirect determinant of the amount of MAP, which varied from 3 weeks (∼105 viable MAP/g) to 7 weeks (∼1.5 viable MAP/g) (Reddacliff et al., 2003) at the last sampling for the clinically affected sheep. Thus MAP faecal contamination of the environment from the inoculated sheep increased from 7 to 16 months PI because of the increase in number of sheep shedding and the amount of MAP shed per sheep. Seven of the 8 clinical animals were most likely shedding the highest amount of MAP during the 12–16 months PI period. This was supported by testing of faecal samples collected at necropsy from animals N◦ 13, 16 and 21. These samples had time to MAP maximal growth of 3, 6 and 4 weeks, respectively, which was lower than the 14 month PI sampling (Table 1). Beyond 23 months PI, only one animal, N◦ 29, was intermittently faecal shedding until the end of the trial at 30 months (data not shown). 3.3. Serum antibody responses MAP-specific serum antibody levels in the inoculated sheep increased gradually throughout the trial until 12 months PI. When the sheep that developed clinical disease were compared to the sheep that survived until the end of the trial, there appeared to be a
128
D.J. Begg et al. / Veterinary Immunology and Immunopathology 166 (2015) 125–131 Table 1 Longitudinal faecal culture results for individual animals.
Inoculated
Unexposed controls
Months post-inoculation Animal 0 N 1 2 3 4 5 6 7 8 9 10
0
1
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29* 30
2
3
5
6
4
5
6
>12
7
8
10
12
14
16
6 6 6
5
5
4
3
3
>12
5 5
12 6
4 5
4 7
4 6
7 6 6
5 5
6 6
4 4
7 5
4 3
4 6
3 4
3 6
3 4
5
5
6
5 6
6 6
5 5
6
12 6
6
17
19
21
23
30
6 6
4
3 6
5
5
5 6
>12
5
5 6
5
6
>12
5 6
5
5 6
White boxes indicate a negative culture result. Grey boxes with a number inside indicate a faecal sample from which MAP was cultured; the number indicates the weeks to maximal growth in BACTEC 12B culture. Black boxes indicate the animal was culled due to onset of clinical JD. White numbers in the black boxes indicate weeks to maximal growth from faecal samples collected at necropsy. Faecal samples collected at necropsy from the remaining animals were not tested. Animals 1–10 were unexposed controls while 11–30 were inoculated with MAP. After 23 months PI only animal 29 was observed to be intermittently faecal shedding MAP (data not shown). Red shaded boxes indicate the faecal IgG SP% detected from the sample was >50%. No clinical cases generated a faecal IgG SP% >50%. * Animal 29 was not included in the faecal antibody analysis of clinical cases, survivors or controls.
trend for the clinical cases to have lower serum antibody levels until 7 months PI (Fig. 1). At later time points the MAP-specific serum antibody responses tended to be higher in clinical cases compared to the surviving sheep and this effect was greatest at 12 months PI, however the differences did not reach significance due to variation in the responses of individual sheep, particularly those within the clinical group. The high antibody level from the survivors in the first 6 months of the trial was mainly due to 4 sheep which had S/P%
values greater than 80 within this period. For example, animal N◦ 12 had an S/P% of 126 at 1 month PI; this remained high until 12 months and then gradually dropped to S/P% of 46 at 23 months. 3.4. Faecal IgG and IgA responses The faecal IgG antibody levels specific to MAP were similar between the un-inoculated controls, the clinical cases and the
Geometric mean serum anbody SP%
180 Control
160
Survivors 140 Clinicals cases 120
#
100 80 60 40 20 0 0
2
4
6
8
10
12
14
16
18
20
22
24
Months post inoculaon Fig. 1. Geometric mean serum MAP-specific IgG responses from clinical, survivor and unexposed control sheep. Serum antibodies were measured by Institute Pourquier ELISA (now IDEXX). The inoculated sheep were further categorised as those that developed clinical disease and those that survived until the end of the trial and were tissue culture negative. Error bars are the standard error of the geometric means. # After 17 months only one animal remained from the sheep that developed clinical disease, therefore error bars are not presented for this treatment from 17 to 21 months post inoculation.
D.J. Begg et al. / Veterinary Immunology and Immunopathology 166 (2015) 125–131
129
120
Geometric Mean IgG SP%
100
*
80
Control
60
Survivors
#
Clinical cases
40 20 0 0
2
4
6
8
10 12 14 16 Months post inoculaon
18
20
22
24
Fig. 2. Geometric mean MAP-specific faecal IgG responses from clinical, survivor and unexposed control sheep. Arrows indicate times at which multiple sheep were removed from the trial due to the onset of clinical disease; three sheep on each occasion. * Survivors are significantly different from clinical cases and controls P < 0.001. Error bars are the standard error of the geometric means. # After 17 months only one animal remained from the sheep that developed clinical disease, therefore error bars are not presented for this treatment from 17 to 21 months post inoculation.
survivors until the sampling point at 16 months PI (Fig. 2). At this time the survivors had a significant (P < 0.001) spike in the faecal IgG level which remained high at 17 months PI. However, by 19 months PI the response had returned to baseline levels, similar to the control animals. The MAP-specific faecal IgA antibody levels showed a similar pattern to the MAP-specific faecal IgG responses (Fig. 3). A significant (P < 0.001) peak was seen at the 16 month PI time point and by 19 months, the response was similar to the level observed from the faecal samples of the unexposed controls.
of the unexposed animal, indicating all available antibodies had been adsorbed by binding to MAP antigens.
4. Discussion The results from this study show for the first time that faecal samples from ruminants in a MAP exposed flock contain MAPspecific IgG and IgA, although it is only measurable by ELISA at certain times. The MAP inoculated sheep with increased amounts of faecal antibodies at 16 months PI were not clinical cases, shedding no detectable or low levels of MAP in their faeces and did not develop clinical disease in the next 3–4 months. Faecal antibodies were detected in MAP inoculated animals soon after removal of sheep that had been shedding high numbers of MAP onto pasture and had to be culled due to clinical disease. The faecal antibody response dropped to background levels several months after removal of these clinical cases. The detection of mucosal antibodies in these sheep may be indicative of a protective response triggered by environmental exposure, that had the effect of sequestering MAP passing through the gut to prevent or limit the infection. This theory is supported by the experiment conducted
3.5. Binding capacity of the IgG and IgA in faecal samples to MAP To determine the amount of MAP that could be bound by a faecal sample containing a high amount MAP specific IgG and IgA a titration experiment was performed. The results showed that 1.6–4 × 1011 MAP/g of faeces was required to bind all of the detectable anti-MAP IgG and IgA from the faeces based on replicate dilution curves (Fig. 4). This was determined by the point at which the level of IgG and IgA antibodies detected by ELISA in the faecal sample from the inoculated animal had dropped to the level 70
Geometric mean IgA SP%
60 50
* #
40
Control Survivors
30
Clinical cases 20 10 0 0
2
4
6
8
10
12
14
16
18
20
22
24
Months post inoculaon Fig. 3. Geometric mean MAP-specific faecal IgA responses from clinical, survivor and unexposed control sheep. Arrows indicate times at which multiple sheep were removed from the trial due to the onset of clinical disease; three sheep on each occasion. * Survivors are significantly different from clinical cases and controls P < 0.001. Error bars are the standard error of the geometric means. # After 17 months only one animal remained from the sheep that developed clinical disease, therefore error bars are not presented for this treatment from 17 to 21 months post inoculation.
130
D.J. Begg et al. / Veterinary Immunology and Immunopathology 166 (2015) 125–131
A
1.4
OD at 450nm
1.2 1.0 0.8 0.6
Unexposed
0.4
Inoculated
0.2 8.0E+11
4.0E+11
8.0E+10
4.0E+10
8.0E+09
4.0E+09
8.0E+08
0.0E+00
0.0
MAP added per gram of faeces
B
1.6 1.4
OD at 450nm
1.2 1.0 0.8 0.6
Unexposed
0.4
Inoculated
0.2 8.0E+11
4.0E+11
8.0E+10
4.0E+10
8.0E+09
4.0E+09
8.0E+08
0.0E+00
0.0
MAP added per gram of faeces Fig. 4. MAP binding capacity of faecal antibodies measured by titration of MAP suspension in faeces from an unexposed sheep and an inoculated sheep. The later had high levels of faecal IgG/IgA but was culture negative for MAP. (A) IgG ELISA. (B) IgA ELISA. Results from one of two replicate assays are shown with similar results obtained for both experiments. Data shown are mean OD values of the 1/800 faecal dilution ± the standard deviation of two technical replicates.
to test the MAP antigen binding capacity of faecal antibodies. The addition of exogenous MAP to a faecal sample containing a high level of IgG and IgA antibodies resulted in a reduction in the level of free antibodies in a dose responsive manner. Exposure to non-pathogenic environmental mycobacteria in field situations is common and in this study it is possible the sheep were exposed to these. A small increase in the faecal IgG response at 14 months post inoculation from all the groups including controls not inoculated with MAP may have been related to exposure to environmental mycobacteria or other unknown factors causing biological variation at this sampling point. After this time point, the faecal IgG responses returned to base line for the control animals, compared to the responses in the inoculated animals which were correlated to the clinical outcome, indicating MAP exposure as the driver of the later faecal Ig responses observed. The exact role that antibody plays in protection against mycobacterial infections such as JD is unknown. The importance of mucosal (secretory) antibodies in protection against mycobacterial infection in general is poorly understood. A study using a mouse model for tuberculosis indicated that secretory antibodies in the mucosa are important in reducing bacterial load (Alvarez et al., 2013). Treatment of mice with secretory IgA prior to inoculation resulted in a significant reduction in recoverable bacteria up to 60 days later and fewer lesions (Alvarez et al., 2013). Significant increases in IgA levels have also been associated with enhanced protection with novel vaccines against tuberculosis (Ai et al., 2013) and conversely an intestinal IgA deficiency has been observed in lepromatous leprosy patients (Saha et al., 1978). These studies,
taken in combination with the data presented here indicate that there may be an association between mucosal antibodies and disease outcome in systemic mycobacterial infections. The amount of MAP required to quench the IgA/IgG faecal ELISAs in the binding capacity assays was higher than expected, up to 4 × 1011 MAP/g of faeces. This is most likely related to the very high faecal IgG and IgA SP% of the sample, unknown efficiency of antibody–MAP binding within the faeces. The high antibody levels in the faeces might be due to heavy contamination of viable and non-viable MAP by the clinically affected animals which were later removed. Most likely an elevated level of faecal antibodies might be an immune mechanism to cope with high numbers of ingested MAP. Clinically affected animals can excrete 109 viable MAP per gram of faeces and sheep can produce a kilogram of faeces each day (Whitlock et al., 2005; Whittington et al., 2000). Accidental ingestion of 1 gram of MAP contaminated faeces by a sheep could expose the animal to a large amount of viable and non-viable MAP. The results lead to speculation: (i) that MAP ingestion leads to stimulation of mucosal IgG and IgA, (ii) the more MAP ingested the more that is adsorbed by the mucosal Ig and that (iii) the amount of faecal Igs detected by ELISA is determined by the level of stimulation and the level of adsorption. Ways to examine this hypothesis in more detail could involve using experimentally or naturally MAP exposed animals that are not shedding MAP in faeces, and remove them from further MAP exposure. Later the animals could be exposed to different sized oral boluses of viable or non-viable MAP. This would determine if antigen or viable MAP exposure is required to stimulate the mucosal Ig response, and whether a
D.J. Begg et al. / Veterinary Immunology and Immunopathology 166 (2015) 125–131
measurable Ig response is abolished by adsorption and further our understanding of immune protection against MAP infection. The high serum IgG antibody response from the inoculated animals that survived the infection was also a surprising result, suggesting a problem with assay specificity. The published specificity of this assay is 98.8% (Gumber et al., 2006), but 4 surviving animals had more than 7 positive antibody tests determined by using a cut-off point of 70 S/P% throughout the trial. None of these surviving animals had histological lesions associated with JD and were tissue culture negative at necropsy. Two did not have viable MAP in faecal samples collected at any time throughout the trial. The study design used by Gumber et al. (2006) to develop specificity data may have resulted in work up and case control biases (Nielsen et al., 2011). Although costly and time consuming, a cross sectional study with full pathological and tissue culture examination maybe be required on a proportion of animals from random flocks to gain a better estimate of specificity (Nielsen et al., 2011). The findings show that sheep exposed to MAP have a mucosal immune response, detectable as Ig in the faeces. The assays developed to detect faecal IgG and IgA may not be of use in the diagnosis of JD as the MAP-specific faecal antibodies were only detected under certain conditions. The results lead to speculation that MAP ingestion leads to stimulation of mucosal IgG and IgA, but detection of the faecal antibody response could not be seen until the source of the ingested adsorptive load was suddenly removed. The amount of faecal Igs detected by ELISA maybe determined by the level of stimulation and the level of adsorption by recently ingested MAP, suggesting that mucosal immunity may play a role in protecting against re-infection. Conflict of interest statement This work was supported by Meat and Livestock Australia and by Cattle Council of Australia, Sheepmeat Council of Australia and WoolProducers Australia through Animal Health Australia. These supporting groups had no involvement in the study design, in the collection, analysis and interpretation of data; in the writing of the manuscript and in the decision to submit the manuscript for publication. Acknowledgements The authors would like to thank Nicole Carter, Ann Michele Whittington, and Anna Waldron for providing laboratory assistance with the ELISAs and cultures. Craig Kristo, Nobel Toribio, Lee White and James Dalton who assisted with the field work. Dr. Graeme Eamens, Department of Primary Industry, Elizabeth Macarthur Agricultural Institute who supplied the MAP 316v antigen. This work was supported by Meat and Livestock Australia and by Cattle Council of Australia, Sheepmeat Council of Australia and WoolProducers Australia through Animal Health Australia. References Abbas, B., Riemann, H.P., 1988. IgG, IgM and IgA in the serum of cattle naturally infected with Mycobacterium paratuberculosis. Comp. Immunol. Microbiol. Infect. Dis. 11, 171–175. Ai, W., Yue, Y., Xiong, S., Xu, W., 2013. Enhanced protection against pulmonary mycobacterial challenge by chitosan-formulated polyepitope gene vaccine is associated with increased pulmonary secretory IgA and gamma-interferon(+) T cell responses. Microbiol. Immunol. 57, 224–235. Alvarez, N., Otero, O., Camacho, F., Borrero, R., Tirado, Y., Puig, A., Aguilar, A., Rivas, C., Cervantes, A., Falero-Diaz, G., Cadiz, A., Sarmiento, M.E., Norazmi, M.N., Hernandez-Pando, R., Acosta, A., 2013. Passive administration of purified
131
secretory IgA from human colostrum induces protection against Mycobacterium tuberculosis in a murine model of progressive pulmonary infection. BMC Immunol. 14 (Suppl. 1), S3. Baumann, R., Kaempfer, S., Chegou, N.N., Oehlmann, W., Loxton, A.G., Kaufmann, S.H., van Helden, P.D., Black, G.F., Singh, M., Walzl, G., 2014. Serologic diagnosis of tuberculosis by combining Ig classes against selected mycobacterial targets. J. Infect. 69, 581–589. Begg, D.J., de Silva, K., Di Fiore, L., Taylor, D.L., Bower, K., Zhong, L., Kawaji, S., Emery, D., Whittington, R.J., 2010. Experimental infection model for Johne’s disease using a lyophilised, pure culture, seedstock of Mycobacterium avium subspecies paratuberculosis. Vet. Microbiol. 141, 301–311. Bush, R.D., Windsor, P.A., Toribio, J.A., 2006. Losses of adult sheep due to ovine Johne’s disease in 12 infected flocks over a 3-year period. Aust. Vet. J. 84, 246–253. Campos-Rodriguez, R., Godinez-Victoria, M., Abarca-Rojano, E., Pacheco-Yepez, J., Reyna-Garfias, H., Barbosa-Cabrera, R.E., Drago-Serrano, M.E., 2013. Stress modulates intestinal secretory immunoglobulin A. Front. Integr. Neurosci. 7, 86. Cripps, A.W., Husband, A.J., Lascelle, A.K., 1974. Origin of immunoglobulins in intestinal secretion of sheep. Aust. J. Exp. Biol. Med. Sci. 52, 711–716. Feng, X., Yang, X., Xiu, B., Qie, S., Dai, Z., Chen, K., Zhao, P., Zhang, L., Nicholson, R.A., Wang, G., Song, X., Zhang, H., 2014. IgG, IgM and IgA antibodies against the novel polyprotein in active tuberculosis. BMC Infect. Dis. 14, 336. Grewal, H.M., Karlsen, T.H., Vetvik, H., Ahren, C., Gjessing, H.K., Sommerfelt, H., Haneberg, B., 2000. Measurement of specific IgA in faecal extracts and intestinal lavage fluid for monitoring of mucosal immune responses. J. Immunol. Methods 239, 53–62. Griffin, J.F., Spittle, E., Rodgers, C.R., Liggett, S., Cooper, M., Bakker, D., Bannantine, J.P., 2005. Immunoglobulin G1 enzyme-linked immunosorbent assay for diagnosis of Johne’s disease in red deer (Cervus elaphus). Clin. Diagn. Lab. Immunol. 12, 1401–1409. Gumber, S., Eamens, G., Whittington, R.J., 2006. Evaluation of a Pourquier ELISA kit in relation to agar gel immunodiffusion (AGID) test for assessment of the humoral immune response in sheep and goats with and without Mycobacterium paratuberculosis infection. Vet. Microbiol. 115, 91–101. Heckert, R.A., Saif, L.J., Mengel, J.P., Myers, G.W., 1991. Isotype-specific antibody responses to bovine coronavirus structural proteins in serum, feces, and mucosal secretions from experimentally challenge-exposed colostrum-deprived calves. Am. J. Vet. Res. 52, 692–699. Marsh, I.B., Bannantine, J.P., Paustian, M.L., Tizard, M.L., Kapur, V., Whittington, R.J., 2006. Genomic comparison of Mycobacterium avium subsp. paratuberculosis sheep and cattle strains by microarray hybridization. J. Bacteriol. 188, 2290–2293. Marsh, I.B., Whittington, R.J., 2007. Genomic diversity in Mycobacterium avium: single nucleotide polymorphisms between the S and C strains of M. avium subsp. paratuberculosis and with M. a. avium. Mol. Cell. Probes 21, 66–75. Nielsen, S.S., 2008. Transitions in diagnostic tests used for detection of Mycobacterium avium subsp. paratuberculosis infections in cattle. Vet. Microbiol. 132, 274–282. Nielsen, S.S., Toft, N., 2008. Ante mortem diagnosis of paratuberculosis: a review of accuracies of ELISA, interferon-gamma assay and faecal culture techniques. Vet. Microbiol. 129, 217–235. Nielsen, S.S., Toft, N., Gardner, I.A., 2011. Structured approach to design of diagnostic test evaluation studies for chronic progressive infections in animals. Vet. Microbiol. 150, 115–125. Parreno, V., Bejar, C., Vagnozzi, A., Barrandeguy, M., Costantini, V., Craig, M.I., Yuan, L., Hodgins, D., Saif, L., Fernandez, F., 2004. Modulation by colostrum-acquired maternal antibodies of systemic and mucosal antibody responses to rotavirus in calves experimentally challenged with bovine rotavirus. Vet. Immunol. Immunopathol. 100, 7–24. Perez, V., Garcia Marin, J.F., Badiola, J.J., 1996. Description and classification of different types of lesion associated with natural paratuberculosis infection in sheep. J. Comp. Pathol. 114, 107–122. Reddacliff, L.A., Nicholls, P.J., Vadali, A., Whittington, R.J., 2003. Use of growth indices from radiometric culture for quantification of sheep strains of Mycobacterium avium subsp. paratuberculosis. Appl. Environ. Microbiol. 69, 3510–3516. Saha, K., Agarwal, S.K., Misra, R.C., 1978. Gut-associated IgA deficiency in lepromatous leprosy. Scand. J. Immunol. 8, 397–402. Whitlock, R.H., Sweeney, R.W., Fyock, T., Smith, J., 2005. MAP super-shedders: another factor in the control of Johne’s disease. In: Manning, E.J., Nielsen, S.S. (Eds.), 8th International Colloquium on Paratuberculosis. Copenhagen, Denmark. Whittington, R.J., Marsh, I., McAllister, S., Turner, M.J., Marshall, D.J., Fraser, C.A., 1999. Evaluation of modified BACTEC 12B radiometric medium and solid media for culture of Mycobacterium avium subsp. paratuberculosis from sheep. J. Clin. Microbiol. 37, 1077–1083. Whittington, R.J., Reddacliff, L.A., Marsh, I., McAllister, S., Saunders, V., 2000. Temporal patterns and quantification of excretion of Mycobacterium avium subsp. paratuberculosis in sheep with Johne’s disease. Aust. Vet. J. 78, 34–37.
