Transboundary and Emerging Diseases
ORIGINAL ARTICLE
Cross-Protection Between Geographically Distinct Anaplasma marginale Isolates Appears to be Constrained by Limited Antibody Responses R. Kenneil1, V. Shkap2, B. Leibovich2, E. Zweygarth1, K. Pfister1, M. F. B. Ribeiro3 and L. M. F. Passos1,4 1 2 3 4
Institute for Comparative Tropical Medicine and Parasitology, Ludwig-Maximilians-Universit€at (LMU), Munich, Germany Parasitology Division, Kimron Veterinary Institute, Bet Dagan, Israel Departamento de Parasitologia, ICB-UFMG, Belo Horizonte, Minas Gerais, Brazil Departamento de Medicina Veterinaria Preventiva, INCT-Pecu aria, Escola de Veterinaria-UFMG, Belo Horizonte, Minas Gerais, Brazil
Keywords: Anaplasma marginale; Anaplasma centrale; UFMG1; anaplasmosis; vaccination Correspondence: L. M. F. Passos. Departamento de Medicina Veterinaria Preventiva, Escola de Veterina´riaUFMG, Av. Antonio Carlos, 6627, CP 567, Belo Horizonte 30123-970, Minas Gerais, Brazil. Tel.: +55(31)34092075; Fax: +55(31) 34092080 ; E-mail: [email protected] Work was carried out In vitro propagation of Anaplasma marginale: Departamento de Parasitologia, ICB-UFMG, Belo Horizonte, Brazil and Institute for Comparative Tropical Medicine and Parasitology, Ludwig-Maximilians-Universit€ at (LMU), Munich, Germany. Animal trial: Parasitology Division, Kimron Veterinary Institute, Bet Dagan, Israel ELISA: Institute for Comparative Tropical Medicine and Parasitology, LudwigMaximilians-Universit€ at, Munich, Germany.
Received for publication November 15, 2012 doi:10.1111/tbed.12125
Summary The rickettsia Anaplasma marginale causes the haemolytic disease bovine anaplasmosis, an economic problem in tropical and subtropical areas worldwide. The closely related but less pathogenic Anaplasma centrale is commonly used as a live vaccine to prevent anaplasmosis, but it can only be produced from infected blood. UFMG1 is a low pathogenic Brazilian strain of A. marginale, which has been shown to protect cattle against a high pathogenic Brazilian isolate. As UFMG1 can be grown in tick cells, the strain was proposed as a possible cell culturederived vaccine. We have evaluated whether UFMG1 could protect cattle against a geographically distant heterologous strain, using A. centrale vaccination as a standard for comparison. Trial calves were infected with UFMG1, A. centrale or PBS. UFMG1-infected animals were more symptomatic than those infected with A. centrale, but none required treatment. All calves were then challenged with the Israeli A. marginale Gonen strain (one of the most prevalent strain in Israel). The A. centrale group had the mildest symptoms, while UFMG1 and control groups both had a more severe response. Nevertheless, the challenge did not cause life-threatening disease in any group. Animals infected with A. centrale had a significantly higher IgG response than UFMG1, when measured in an ELISA against initial bodies from their homologous strain or Gonen. The level of crossreactivity of the response to initial infection correlated significantly with reduced symptoms after challenge. In conclusion, UFMG1 had limited effect in preventing disease by the geographically distant heterologous Gonen strain. While the low pathogenicity of the Gonen strain in this trial makes it impossible to conclusively state that UFMG1 would have given no protective effect against more serious disease, the comparatively low IgG response to UFMG1 suggests it would not have been as effective as A. centrale.
Introduction Anaplasma marginale (Anaplasmataceae family, Rickettsiales order) is an obligate intracellular bacterium which causes the haemolytic disease bovine anaplasmosis. Endemic worldwide in tropical and subtropical areas, A. marginale can infect a range of ruminant species, but appears to almost exclusively cause disease in cattle (Kocan et al., 2010). It can
be transmitted biologically by a range of tick species, mechanically by biting flies or surgical instruments (Kocan et al., 2010) and transplacentally (Passos and Lima, 1984; Zaugg and Kuttler, 1984). In cattle, it principally infects erythrocytes; this triggers high levels of erythrocyte phagocytosis, leading to anaemia and icterus. Other symptoms include fever, weight loss, lowered milk production and abortion. The disease can be fatal if untreated, particularly
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
97
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
in older animals (Jones and Brock, 1966; Kocan et al., 2010). Cattle that recover from acute disease remain persistently infected, and undergo cycles of recurrent rickettsemia (Palmer et al., 1999; Kocan et al., 2003). Therefore, although these persistently infected cattle are protected from subsequent challenge by homologous strains, they also act as reservoirs of infection for na€ıve cattle (Kocan et al., 2003). Immunity against A. marginale is mediated through a combination of humoral and cellular mechanisms. Antibodies against major surface proteins (MSPs) are associated with protection from infection (Kocan et al., 2003). However, antibody alone is not sufficient for protection (Kuttler and Adams, 1977; Gale et al., 1992), suggesting a requirement for cell-mediated mechanisms. Palmer et al. (1999) proposed a central role of antigen-specific CD4+ T cells, principally their production of IFNc. IFNc stimulates both macrophage activation and B-cell production of IgG2 (the most opsonizing subclass of IgG). The combination of activated macrophages and high levels of IgG2 would lead to increased opsonization and phagocytosis of the rickettsia. To corroborate this, Brown et al. (1998) found that protection after vaccination with outer membranes was associated with CD4+ proliferation and IFNc production and high IgG2 levels. Anaplasma marginale infection itself leads to the down-regulation of high pre-challenge, vaccinationinduced, CD4 T-cell responses. This is likely to contribute to bacterial persistence (Han et al., 2008). Multiple approaches to vaccination against A. marginale have been investigated, as reviewed by Kocan et al. (2003). The principal obstacle for effective vaccination is limited cross-protection between strains, as there is a high level of diversity worldwide, and even considerable variation within smaller areas (Almazan et al., 2008). An effective vaccine in one region may only offer limited protection in another (Palmer et al., 1994; Ocampo Espinoza et al., 2006). The most widespread method of disease prevention in active use is live vaccination with Anaplasma centrale, which is taxonomically closely related to A. marginale. It has been used as a live vaccine from the time it was first identified by Sir Arnold Theiler (1912) and is currently in use in Israel, Australia, South Africa and parts of South America (Shkap et al., 2002). Cattle infected with A. centrale develop a persistent infection with mild or no disease and are subsequently protected from disease caused by A. marginale. Even though it is generally very effective, there have been some reports of vaccine failure or high side effects (Brizuela et al., 1998; Payne et al., 1990; Turton et al., 1998). Despite many attempts, A. centrale has never been successfully cultivated in vitro, and therefore, the vaccine is derived from the blood of infected calves. This risks accidental transmission of other pathogens and requires using animals, which is laborious, expensive, and should be replaced where possible with more ethical alternatives. 98
R. Kenneil et al.
Unlike A. centrale, several A. marginale strains have been successfully established and propagated in tick cell lines (Bell-Sakyi et al., 2007; Passos, 2012). Cell culture-derived vaccines improve safety and reproducibility over those produced from blood (de la Fuente et al., 2002). Bastos et al. (2010) demonstrated that a naturally low pathogenic Brazilian strain (UFMG1) protected cattle from the heterologous high pathogenic Brazilian strain (UFMG2). As UFMG1 has been grown in IDE8 tick cell cultures (Bastos et al., 2009), it was proposed as a potential cell culturederived vaccine (Bastos et al., 2010). One of the main problems with controlling A. marginale is the high geographic variability of strains. Therefore, an essential step in characterizing UFMG1 as a potential vaccine was to investigate whether it could induce a more broadly cross-reactive response and could have potential use as a vaccine outside Brazil. Materials and Methods Anaplasma strains Anaplasma marginale UFMG1 (msp1a sequence GenBank EU676176), originally isolated from a naturally infected calf in Minas Gerais State, Brazil (Ribeiro et al., 1997), had been continuously maintained in IDE8 tick cells for approximately 5 years at the UFMG, using standard culture methods described by Bastos et al. (2010). The A. centrale vaccine strain (full genome sequence GenBank CP001759), routinely used as a live vaccine in Israel (Shkap et al., 2002), had been maintained through inoculation of splenectomized calves at the Kimron Veterinary Institute, Bet Dagan, Israel. Anaplasma marginale Gonen (msp1a sequence GenBank EU678755) originating from the Gonen farm in the north of Israel, was determined to have one of the most common MSP1a genotypes in Israel at the time of its isolation (Molad et al., 2009). Blood collected from the originally infected animal was inoculated into a splenectomized calf, resulting in minimum haematocrit of 9% and maximum rickettsemia of 55% before oxytetracycline treatment to prevent death (V. Shkap, written communication). This study used blood cryostabilate prepared from the splenectomized calf before antibiotic treatment. To obtain sufficient infective material for cattle inoculation, and in order to standardize all strains as fully viable and blood-derived, A. marginale UFMG1, Gonen and A. centrale were first inoculated individually into three splenectomized calves. For this step, the UFMG1 inoculate was initial bodies partially purified from infected IDE8 tick cells, and A. centrale and Gonen inoculates were blood cryostabilates in 15% DMSO. Once splenectomized calves reached high rickettsemia, infected blood was collected and used to inoculate trial cattle.
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
R. Kenneil et al.
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
Trial design The trial cattle were 12 Israeli Friesian calves (3–5 months of age at start of trial), kept under tick- and fly-free conditions, and confirmed negative for A. marginale and A. centrale infections by PCR and MSP-5 cELISA. They were randomly divided into three groups of four animals each and subcutaneously inoculated with 1 9 106 A. marginale UFMG1, 1 9 106 A. centrale or PBS alone (control). After 60 days (which was 1–2 weeks after the last point rickettsemia was >0.09%), animals were challenged with 1 9 107 Gonen strain initial bodies in 2 ml PBS. Progress of infection was monitored at least thriceweekly, and daily during acute infection, by rickettsemia, haematocrit and rectal temperature. Rickettsemia was measured in Giemsa-stained blood smears, with blood taken by ear vein puncture; percentage rickettsemia was calculated by the number of infected erythrocytes divided by total number of erythrocytes, counted over a minimum of 20 fields in 1009 oil immersion. Haematocrit was measured by microhaematocrit technique, again from capillary puncture of the ear vein. The threshold of disease severity at which cattle would be treated with antibiotics was set at a haematocrit remaining below 20% for 3 days. The Kimron Veterinary Institute Animals Welfare Committee and the Israeli Ministry of Health have approved all experiments in cattle (licence number 020_b1731).
500 g for 5 min at 4°C. The supernatant was retained and pelleted at 5000 g for 10 min at 4°C. This pellet containing purified initial bodies was resuspended in 1 ml PBS, and protein concentration was measured by Bio-Rad DC protein assay (Bio-Rad, Hercules, CA, USA). High protein-binding microtiter plates (Nunc, Rochester, NY, USA) were coated with 100 ll of 5 lg/ml initial body-derived protein, diluted in carbonate/bicarbonate buffer pH 9.5, and incubated overnight at 4°C. After washing four times with wash buffer (PBS with 0.05% Tween20), plates were blocked with 200 ll diluent (5% heat-inactivated horse serum in wash buffer), which was incubated shaking for 2 h at RT. One hundred microlitre of each sera sample was then added at a 1 : 90 dilution. After incubation and wash steps as before, 100 ll rabbit antibovine IgG conjugated to alkaline phosphatase (Thermo Scientific) was added at 1 : 5000 dilution, and incubated as previously. After washing, 100 ll p-nitrophenyl phosphate substrate in diethanolamine buffer (Thermo Scientific) was incubated for 30 min, and the reaction stopped by the addition of 50 ll 2 N NaOH. Absorbance values were read at 405 nm by LEDETECT 96 plate reader (Labexim Products, Lengau, Austria). Statistical analyses Statistical significances of differences of clinical parameters and ELISA ODs between the groups were calculated by ANOVA and Tukey HSD post-analysis.
Serological tests Blood samples were collected weekly in EDTA vacutainers (Becton Dickinson, Franklin Lakes, NJ, USA) for serological tests. A competitive ELISA against recombinant MSP5 (VMRD, Pullman, WA, USA) was used to confirm seroconversion; samples were tested according to manufacturers’ instructions. IgG levels against A. marginale strains or A. centrale were measured by an indirect ELISA using purified initial bodies as antigen, based on that described by Shkap et al. (1990). The initial bodies were partially purified from infected blood using a protocol similar to Palmer and McGuire (1984). Briefly, blood was collected in EDTA vacutainers, diluted into an equal volume of cold PBS, and then washed three times in cold PBS by centrifugation at 2600 g for 20 min at 4°C. The buffy coat was removed after each centrifugation. Packed erythrocytes were resuspended in an equal volume of PBS and frozen at 70°C to lyse the cells. They were then thawed at room temperature (RT), and washed in PBS eight times by centrifugation at 23 000 g for 15 min at 4°C (Sorvall SS-34 rotor; Thermo Scientific, Waltham, MA, USA), until no visible pink haemoglobin remained. The final pellet was resuspended in 20 ml PBS, sonicated at 100 W for 2 min, and spun at
Results Clinical response to initial infection with Anaplasma marginale UFMG1 and Anaplasma centrale Cattle infected with UFMG1 developed a clearly detectable rickettsemia (≥0.09%) on average 24 days post-infection; A. centrale infection was detectable after an average of 35 days. Once infection was detected, rickettsia remained measurable for around an average 32 days for UFMG1, and 13 days for A. centrale (Table 1). Rickettsemia peaked at an average 10% for the UFMG1 group and 5% for the A. centrale group. UFMG1 infection caused a more pronounced drop in haematocrit than A. centrale, with a minimum haematocrit of 18–21% in UFMG1-infected animals (average reduction of 13% points), compared with 21–33% for A. centrale (average reduction of 7.5% points) and 28–31% (average reduction of 5% points) for control animals. For both peak rickettsemia and minimum haematocrit, UFMG1 group values were significantly different from the uninfected control group (P < 0.01), whereas A. centrale group values were not significantly different (P > 0.05). The rate of decrease in haematocrit was twice as fast for UFMG1 as for
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
99
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
R. Kenneil et al.
Table 1. Summary of average clinical parameters for all groups, pre- and post-challenge
Peak rickettsemia (%)
Pre-challenge
UFMG1 Anaplasma centrale PBS UFMG1 A. centrale PBS
10 5 0 2.6 1.0 3.05
Post-challenge
3.1 3.6 0 3.2 1.4 3.1
Days of measurable rickettsemia (≥0.09%)
Minimum haematocrit (%)
32 4 13 1 00 15 11 44 20 17
19.8 26 29 21.5 28.3 22
A. centrale. All infections spontaneously resolved without requiring treatment (according to pre-determined threshold). Clinical response after challenge The Gonen strain caused relatively mild infection in all groups, and no animal showed sufficiently severe disease to require treatment. Rickettsemia and haematocrit are summarized and compared with pre-challenge parameters in Table 1. The level of rickettsemia after challenge was low in all groups – no individual calf exceeded 7%, and the group averages were 2.6% for UFMG1, 1.0% for A. centrale and 3.0% for the control group. There was high variability within each group, as shown in Fig. 1. No statistically significant difference was seen between any groups. The UFMG1 and control groups showed a similar level of red blood cell loss after challenge. Their minimum haematocrit ranged between 19–24% (average reduction of 12% points) and 20–26% (average reduction of 13% points), respectively. Animals in the A. centrale group were less affected, with minimum haematocrit between 22% and 32% (average reduction of 5% points). There was a significant difference between minimum haematocrit of UFMG1 and A. centrale-infected animals after challenge (P < 0.05); however, both UFMG1 and A. centrale groups were not significantly different from the control group.
Serum samples from the trial were tested with an antiMSP5 competition ELISA as a confirmation of infection and initial assessment of immune response. All animals were seronegative before the trial started. After infection with UFMG1 or A. centrale, they rapidly seroconverted, and antibody levels remained high until point of challenge and beyond (Fig. 2). The antibody response to UFMG1 infection had minimal cross-reactivity to the Gonen strain (Fig. 3). In contrast, the antibody response to A. centrale (in three out of four animals in the group) had a high level of binding to Gonen 100
1.5 5.3 1.2 2.4 4.3 2.8
39.4 20.3 5.4 34.3 15.3 37.2
12.2 25.6 11.9 12.7 13.1 3.5
8 7 6 5 4 3 2 1 0 –1 UFMG1
A.centrale
PBS
(b)
35 30 25 20
x2
15
UFMG1
Antibody response to infection
Percentage haematocrit reduction (%)
(a)
Peak rickettsemia (% iRBCs)
Group
Minimum haematocrit (%)
Period
A.centrale
PBS
Fig. 1. Peak rickettsemia (a) and minimum haematocrit (b) post-challenge. Horizontal bars represent average group value.
initial bodies (and to UFMG1, data not shown). This crossreactivity correlated significantly with lower red blood cell loss: the higher the level of cross-reactive IgG against the Gonen strain during initial infection, the lower the subsequent drop in haematocrit after challenge (P < 0.01, Pearson’s moment correlation coefficient r = 0.789, r2 = 0.624.
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
100
5
80
4.5
60
4
UFMG1
40
3.5
A.centrale
20
PBS
0 –40
–20
–20
0
20
40
60
80
100
120
140
Days after initial infection
Fig. 2. Total antibody levels against MSP5, measured by MSP5 cELISA. Serum samples were diluted 1 : 64 to prevent saturation of the assay (when measured neat, as in manufacturers’ instructions, all groups reached equal levels, around 90% inhibition). Time of challenge is indicated by the arrow.
OD 405nm
% inhibition
R. Kenneil et al.
3 2.5 2 1.5 1 0.5 0
Group
(a)
Animal number:
2.5
OD 405nm
UFMG1
1 2 3 4
2 1.5 1
0
0
20
40
60
80
100
120
Days post initial infection
(b)
2.5
Animal number: 5
OD 405nm
2
6 7
1.5
8
1 0.5
–20
(c)
0
0
20
40 60 80 Days post initial infection
2.5
100
Discussion
OD 405nm
9 10
1.5
11 12
1 0.5 0 0
20
40
60
80
post-challenge), tested against initial bodies from the homologous strain. The highest IgG levels were generally around 1 month post-infection. UFMG1 infection had a relatively low IgG response (average 3-fold increase above pre-immune levels, data not shown). Gonen infection led to a slightly higher response, an average 7-fold increase, with one very high-responding animal. Anaplasma centrale infection provoked considerably higher IgG levels in three of four animals (average 12-fold increase).
120
Animal number:
2
–20
Gonen
Fig. 4. Maximum IgG levels against initial bodies from homologous strains, after infection with UFMG1, Anaplasma centrale or Gonen alone. Significant difference between groups is labelled with an asterisk (P < 0.05); horizontal bars represent the group average.
0.5
–20
A.centrale
100
120
Days post initial infection
Fig. 3. IgG levels against A. marginale Gonen strain in animals from the UFMG1 (a), A.centrale (b), and control (c) groups. Time of challenge with Gonen strain is indicated by arrow.
To determine whether the difference in anti-Gonen IgG levels was due to a difference in strength of the overall IgG response, IgG levels against homologous initial bodies were measured. Figure 4 shows the maximum IgG response of cattle infected with UFMG1, A. centrale or Gonen alone (for the latter, samples were taken from control group
As reported here, UFMG1 had a negligible effect on disease caused by the Gonen strain. This correlated with a low level of cross-reactive IgG produced during UFMG1 infection, compared with considerably higher levels induced by the more protective A. centrale vaccine strain. The lack of protection from UFMG1 against the heterologous Israeli strain tested here contrasts with its complete protection against a highly pathogenic Brazilian strain, UFMG2 (Bastos et al., 2010). However, as the Israeli Gonen strain was not highly pathogenic in this trial, it cannot be conclusively stated that UFMG1 would have had no protective effect against fatal levels of disease. The level of disease caused by the A. marginale strains and A. centrale varied considerably within each group, both before and after challenge. For the latter, this correlated with very variable levels of Anaplasma-specific IgG production. Even in A. centrale-infected animals, one animal produced a very low antibody response and showed no reduction in symptoms after challenge. This is not
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
101
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
unexpected in animals not specifically inbred for laboratory studies, but does highlight the difficulties in inducing complete coverage across a diverse population. Results from the MSP5 cELISA were very different from the IgG ELISA against initial bodies, with total antibody levels against MSP5 lower in the A. centrale group compared with the UFMG1 group. This could have a number of possible explanations. First, the cELISA measures all antibody classes. Therefore, higher responses measured by this assay could be largely due to IgM. Although mainly produced in early infection, Klaus and Jones (1968) and Murphy et al. (1966) found IgM persisted until 60–80 days post-A. marginale infection. Second, the protocol for initial body preparation could damage some surface antigens, which would lower the IgG ELISA results. However, since initial bodies for A. centrale and both A. marginale strains were prepared in the same way, this would only skew the relative antibody production by between different groups if more A. marginale-specific antigens were damaged, which seems unlikely. Third, MSP5 is a highly conserved protein, demonstrated by the fact that this cELISA can be used to detect antibodies after A. marginale, A. centrale and Anaplasma phagocytophilum infections (Molloy et al., 1999; Dreher et al., 2005). As A. marginale immunity is highly strain-specific, antibodies against this conserved protein are unlikely to correlate well with protection – and as seen here, anti-MSP5 antibody levels do not appear to correlate with IgG levels against total surface protein. While the MSP5 cELISA assay is very useful for measuring seroconversion as an indicator of infection, it has limited application in estimating protective immunity. Measuring the IgG response against all MSPs, either by ELISA as used here, or by Western blot as in Brown et al. (1998), appears to give more information on the likely degree of cross-protection induced by infection or vaccination. There is known to be high cross-reactivity between A. marginale and A. centrale (Agnes et al., 2011). In addition, the considerably higher level of IgG production in response to A. centrale infection seen here is likely to contribute to its success as a vaccine. As previously mentioned, a model of immunity against A. marginale emphasizes the importance of antigen-specific activated CD4+ T cells and their IFNc production to stimulate antibody class switching of B cells, and cause high levels of IgG production (Palmer et al., 1999). Differences in CD4+ T-cell function and their level of IFNc production in response to stimuli would therefore be a likely candidate to correlate with differences in IgG levels between and within groups, and this will be tested in future work. Anaplasma marginale infection can down-regulate high pre-challenge, recombinant vaccineinduced CD4+ T-cell responses (Abbott et al., 2005; Han et al., 2008). Han et al. (2008) proposed this to be due to deletion of antigen-specific T cells after overstimulation 102
R. Kenneil et al.
during infection. It has not been investigated if this phenomenon also occurs in A. centrale infection – if it does not, this could be a central factor behind the higher antibody response seen here from this group. The smaller difference in IgG levels after UFMG1 versus Gonen infection may be due to UFMG1 being maintained in tick cell culture for a number of years before inoculation into the splenectomized calf. Tick cell culture stimulates different A. marginale protein expressions compared with mammalian cells (Garcia-Garcia et al., 2004), and it is possible that prolonged tick cell culture may have had an enduring effect on UFGM1 protein expression. In A. marginale, this is known to be the case for the major surface protein MSP1, a heterodimer composed of MSP1a and MSP1b. Both subunits are adhesins for bovine erythrocytes, but only MSP1a also binds to tick cells (Palmer and McGuire, 1984; de la Fuente et al., 2002). msp1a is more highly expressed in blood than in tick cell culture (Garcia-Garcia et al., 2004). The higher levels of MSP1a on the surface of A. marginale in cattle are proposed to improve the chances of successful transmission to feeding ticks. de la Fuente et al. (2002) found that higher levels of anti-MSPa IgG are stimulated by blood-derived A. marginale than by tick cellderived. MSP1a is an important antigen for both cellular and humoral responses: CD4 T-cells preferentially recognize MSP1a when immunized with the recombinant MSP1 complex (Brown et al., 2001), and antibodies against MSP1a inhibit the entry of rickettsia into red blood cells (Palmer and McGuire, 1984). Therefore, the lowered antiMSP1a response from tick cell-derived rickettsia could influence their effectiveness as vaccines. Bastos et al. (2010) found no significant difference in the level of protection provided by tick cell or blood-derived UFMG1, but in this case, tick cell cultures were under 1 year old after initialization from blood stabilate, as opposed to 5 years in this trial. UFMG1 infection itself gave rise to symptoms that were not severe enough to require intervention, but did come close to the threshold set for antibiotic treatment. Any pronounced pathogenicity could hinder the use of UFMG1 as a live vaccine. If A. marginale is transmitted from vaccinated calves to na€ıve adults by tick or mechanical transmission, the disease is very likely to be fatal – as seen in experimental infection of adult cattle with A. centrale (Pipano et al., 1985). The risk of this occurring after A. centrale vaccination is limited by its extremely restricted biological transmission – only reported to date with the African tick Rhipicephalus simus (from which it was initially isolated; Theiler, 1912; Potgieter and van Rensburg, 1987). Several other tick species that can act as A. marginale vectors cannot transmit A. centrale: Hyalomma excavatum, Rhipicephalus sanguineus, and Rhipicephalus (Boophilus) annulatus, tested by Shkap et al. (2009). Dermacentor
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
R. Kenneil et al.
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
andersoni was shown by Ueti et al. (2009) to transmit A. centrale, but with extremely low efficiency (over 400 infected ticks were used for one calf). Goncßalves-Ruiz et al. (2005) showed that A. marginale UFMG1 was not transmitted by Rhipicephalus (Boophilus) microplus ticks, which are the major biological vector of A. marginale in Brazil. However, as A. marginale is known to be transmissible by at least 20 tick species (Kocan et al., 2010), any further studies on UFMG1 as a live vaccine in regions with other vector species should investigate the effect of UFMG1 in adult cattle and its transmissibility by other species of tick vector. In conclusion, UFMG1 had a limited effect on disease caused by a geographically distant heterologous strain. The low pathogenicity of the Israeli strain in this trial makes it impossible to determine whether UFMG1 could have reduced fatal disease. However, the limited cross-reactivity of the IgG response to UFMG1 suggests that it would be less protective than A. centrale against a more pathogenic strain. The high levels of cross-reactive IgG induced by A. centrale infection are likely to be a critical factor in its success as a vaccine in a wide range of countries. However, as antibody responses alone are insufficient to protect cattle from bovine anaplasmosis, further work must be carried out to characterize other differences between A. marginale and A. centrale infection, and to determine whether these factors can be harnessed for future vaccines. Acknowledgements IDE8 cells were kindly provided by Dr U.G. Munderloh. R. Kenneil is a Marie Curie Early Stage Researcher (ESR) supported by the POSTICK ITN (Postgraduate training network for capacity building to control ticks and tickborne diseases) within the FP7- PEOPLE – ITN programme (EU Grant No. 238511). Conflicts of interest The authors have no conflicts of interest to declare. References Abbott, J. R., G. H. Palmer, K. A. Kegerreis, P. F. Hetrick, C. J. Howard, J. C. Hope, and W. C. Brown, 2005: Rapid and longterm disappearance of CD4 T lymphocyte responses specific for Anaplasma marginale major surface protein-2 (MSP2) in MSP2 vaccinates following challenge with live A. marginale. J. Immunol. 174, 6702–6715. Agnes, J. T., K. A. Brayton, M. LaFollett, J. Norimine, W. C. Brown, and G. H. Palmer, 2011: Identification of Anaplasma marginale outer membrane protein antigens conserved between A. marginale sensu stricto strains and the live A. marginale subsp. centrale vaccine. Infect. Immun. 79, 1311–1318.
Almazan, C., C. Medrano, M. Ortiz, and J. de la Fuente, 2008: Genetic diversity of Anaplasma marginale strains from an outbreak of bovine anaplasmosis in an endemic area. Vet. Parasitol. 158, 103–109. Bastos, C. V., L. M. F. Passos, M. M. C. Vasconcelos, and M. F. B. Ribeiro, 2009: In vitro establishment and propagation of a Brazilian strain of Anaplasma marginale with appendage in IDE8 (Ixodes scapularis) cells. Braz. J. Microbiol. 40, 399–403. Bastos, C. V., L. M. F. Passos, E. J. Facury-Filho, E. M. Rabelo, J. de la Fuente, and M. F. B. Ribeiro, 2010: Protection in the absence of exclusion between two Brazilian isolates of Anaplasma marginale in experimentally infected calves. Vet. J. 186, 374–378. Bell-Sakyi, L., E. Zweygarth, E. F. Blouin, E. A. Gould, and F. Jongejan, 2007: Tick cell lines: tools for tick and tick-borne disease research. Trends Parasitol. 23, 450–457. Brizuela, C. M., C. A. Ortellado, E. Sanabria, O. Torres, and D. Ortigosa, 1998: The safety and efficacy of Australian tickborne disease vaccine strains in cattle in Paraguay. Vet. Parasitol. 76, 27–41. Brown, W. C., V. Shkap, D. Zhu, T. C. McGuire, W. Tuo, T. F. McElwain, and G. H. Palmer, 1998: CD4+ T-Lymphocyte and immunoglobulin G2 responses in calves immunized with Anaplasma marginale outer membranes and protected against homologous challenge. Infect. Immun. 66, 5406–5413. Brown, W. C., G. H. Palmer, H. A. Lewin, and T. C. McGuire, 2001: CD4+ T lymphocytes from calves immunized with Anaplasma marginale major surface protein 1 (MSP1), a heteromeric complex of MSP1a and MSP1b, preferentially recognize the MSP1a carboxyl terminus that is conserved among strains. Infect. Immun. 69, 6853–6862. Dreher, U. M., J. de la Fuente, R. Hofmann-Lehmann, M. L. Meli, N. Pusterla, K. M. Kocan, Z. Woldehiwet, U. Braun, G. Regula, K. D. Staerk, and H. Lutz, 2005: Serologic cross-reactivity between Anaplasma marginale and Anaplasma phagocytophilum. Clin. Diagn. Lab. Immunol. 12, 1177–1183. de la Fuente, J., K. M. Kocan, J. C. Garcia-Garcia, E. F. Blouin, P. L. Claypool, and J. T. Saliki, 2002: Vaccination of cattle with Anaplasma marginale derived from tick cell culture and bovine erythrocytes followed by challenge-exposure with infected ticks. Vet. Microbiol. 89, 239–251. Gale, K. R., G. Leatch, M. Gartside, and C. M. Dimmock, 1992: Anaplasma marginale: failure of sera from immune cattle to confer protection in passive-transfer experiments. Parasitol. Res. 78, 410–415. Garcia-Garcia, J. C., J. de la Fuente, E. F. Blouin, T. J. Johnson, T. Halbur, V. C. Onet, J. T. Saliki, and K. M. Kocan, 2004: Differential expression of the msp1a gene of Anaplasma marginale occurs in bovine erythrocytes and tick cells. Vet. Microbiol. 98, 261–272. Goncßalves-Ruiz, P. M., L. M. F. Passos, and M. F. B. Ribeiro, 2005: Lack of infectivity of a Brazilian Anaplasma marginale isolate for Boophilus microplus ticks. Vet. Parasitol. 128, 325–331.
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
103
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
Han, S., J. Norimine, G. H. Palmer, W. Mwangi, K. K. Lahmers, and W. C. Brown, 2008: Rapid deletion of antigen-specific CD4+ T cells following infection represents a strategy of immune evasion and persistence for Anaplasma marginale. J. Immunol. 181, 7759–7769. Jones, E. W., and W. E. Brock, 1966: Bovine anaplasmosis: its diagnosis, treatment, and control. J. Am. Vet. Med. Assoc. 149, 1624–1633. Klaus, G. G., and E. W. Jones, 1968: The immunoglobulin response in intact and splenectomized calves infected with Anaplasma marginale. J. Immunol. 100, 991–999. Kocan, K. M., J. de la Fuente, A. A. Guglielmone, and R. D. Melendez, 2003: Antigens and alternatives for control of Anaplasma marginale infection in cattle. Clin. Microbiol. Rev. 16, 698–712. Kocan, K. M., J. de la Fuente, E. D. Blouin, J. F. Coetzee, and S. A. Ewing, 2010: The natural history of Anaplasma marginale. Vet. Parasitol. 167, 95–107. Kuttler, K. L., and L. G. Adams, 1977: Influence of dexamethasone on the recrudescence of Anaplasma marginale in splenectomized calves. Am. J. Vet. Res. 38, 1327– 1330. Molad, T., L. Fleidrovich, M. Mazuz, L. Fish, B. Leibovitz, Y. Krigel, and V. Shkap, 2009: Genetic diversity of major surface protein 1a of Anaplasma marginale in beef cattle. Vet. Microbiol. 136, 54–60. Molloy, J. B., P. M. Bowles, D. P. Knowles, T. F. McElwain, R. E. Bock, T. G. Kingston, G. W. Blight, and R. J. Dalgliesh, 1999: Comparison of a competitive inhibition ELISA and the card agglutination test for detection of antibodies to Anaplasma marginale and Anaplasma centrale in cattle. Aust. Vet. J. 77, 245–249. Murphy, F. A., J. W. Osebold, and O. Aalund, 1966: Kinetics of the antibody response to Anaplasma marginale infection. J. Infect. Dis. 116, 99–111. Ocampo Espinoza, V., J. E. Vazquez, M. D. Aguilar, M. A. Ortiz, G. J. Alarc on, and S. D. Rodrıguez, 2006: Anaplasma marginale: lack of cross-protection between strains that share MSP1a variable region and MSP4. Vet. Microbiol. 114, 34–40. Palmer, G. H., and T. C. McGuire, 1984: Immune serum against Anaplasma marginale initial bodies neutralizes infectivity for cattle. J. Immunol. 133, 1010–1015. Palmer, G. H., D. Munodzana, N. Tebele, T. Ushe, and T. F. McElwain, 1994: Heterologous strain challenge of cattle immunized with Anaplasma marginale outer membranes. Vet. Immunol. Immunopathol. 42, 265–273. Palmer, G. H., F. R. Rurangirwa, K. M. Kocan, and W. C. Brown, 1999: Molecular basis for vaccine development against the ehrlichial pathogen Anaplasma marginale. Parasitol. Today 15, 253–300.
104
R. Kenneil et al.
Passos, L. M. F., 2012: In vitro cultivation of Anaplasma marginale and A. phagocytophilum in tick cell lines: a review. Rev. Bras. Parasitol. Vet. 21, 81–86. Passos, L. M. F., and J. D. Lima, 1984: Diagnosis of congenital anaplasmosis in Minas Gerais. Arq. Bras. Med. Vet. Zootec. 36, 743–744. Payne, R. C., O. Osorio, and A. Yba~ nez, 1990: Tick-borne diseases of cattle in Paraguay. II. Immunisation against anaplasmosis and babesiosis. Trop. Anim. Health Prod. 22, 101– 108. Pipano, E., E. Mayer, and M. Frank, 1985: Comparative response of Friesian milking cows and calves to Anaplasma centrale vaccine. Br. Vet. J. 141, 174–178. Potgieter, F. T., and L. van Rensburg, 1987: Tick transmission of Anaplasma centrale. Onderstepoort J. Vet. Res. 54, 5–7. Ribeiro, M. F. B., L. M. F. Passos, and A. M. Guimar~aes, 1997: Ultrastructure of Anaplasma marginale with an inclusion appendage, isolated in Minas Gerais State, Brazil. Vet. Parasitol. 70, 271–277. Shkap, V., H. Bin, H. Ungar-Waron, and E. Pipano, 1990: An enzyme-linked immunosorbent assay (ELISA) for the detection of antibodies to Anaplasma centrale and Anaplasma marginale. Vet. Microbiol. 25, 45–53. Shkap, V., T. Molad, L. Fish, and G. Palmer, 2002: Detection of the Anaplasma centrale vaccine strain and specific differentiation from Anaplasma marginale in vaccinated and infected cattle. Parasitol. Res. 88, 546–552. Shkap, V., K. Kocan, T. Molad, M. Mazuz, B. Leibovich, Y. Krigel, A. Michoytchenko, E. Blouin, J. de la Fuente, M. Samish, M. Mtshali, E. Zweygarth, E. L. Fleiderovich, and L. Fish, 2009: Experimental transmission of field Anaplasma marginale and the A. centrale vaccine strain by Hyalomma excavatum, Rhipicephalus sanguineus and Rhipicephalus (Boophilus) annulatus ticks. Vet. Microbiol. 134, 254–260. Theiler, A., 1912: Gallsickness of imported cattle and the protective inoculation against this disease. Agric. J. Union S. Afr. 3, 7–46. Turton, J. A., T. C. Katsande, M. B. Matingo, W. K. Jorgensen, U. Ushewokunze-Obatolu, and R. J. Dalgliesh, 1998: Observations on the use of Anaplasma centrale for immunization of cattle against anaplasmosis in Zimbabwe. Onderstepoort J. Vet. Res. 65, 81–86. Ueti, M. W., D. P. Knowles, C. M. Davitt, G. A. Scoles, T. V. Baszler, and G. H. Palmer, 2009: Quantitative differences in salivary pathogen load during tick transmission underlie strain-specific variation in transmission efficiency of Anaplasma marginale. Infect. Immun. 77, 70–75. Zaugg, J. L., and K. L. Kuttler, 1984: Bovine anaplasmosis: in utero transmission and the immunologic significance of ingested colostral antibodies. Am. J. Vet. Res. 45, 440–443.
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
ORIGINAL ARTICLE
Cross-Protection Between Geographically Distinct Anaplasma marginale Isolates Appears to be Constrained by Limited Antibody Responses R. Kenneil1, V. Shkap2, B. Leibovich2, E. Zweygarth1, K. Pfister1, M. F. B. Ribeiro3 and L. M. F. Passos1,4 1 2 3 4
Institute for Comparative Tropical Medicine and Parasitology, Ludwig-Maximilians-Universit€at (LMU), Munich, Germany Parasitology Division, Kimron Veterinary Institute, Bet Dagan, Israel Departamento de Parasitologia, ICB-UFMG, Belo Horizonte, Minas Gerais, Brazil Departamento de Medicina Veterinaria Preventiva, INCT-Pecu aria, Escola de Veterinaria-UFMG, Belo Horizonte, Minas Gerais, Brazil
Keywords: Anaplasma marginale; Anaplasma centrale; UFMG1; anaplasmosis; vaccination Correspondence: L. M. F. Passos. Departamento de Medicina Veterinaria Preventiva, Escola de Veterina´riaUFMG, Av. Antonio Carlos, 6627, CP 567, Belo Horizonte 30123-970, Minas Gerais, Brazil. Tel.: +55(31)34092075; Fax: +55(31) 34092080 ; E-mail: [email protected] Work was carried out In vitro propagation of Anaplasma marginale: Departamento de Parasitologia, ICB-UFMG, Belo Horizonte, Brazil and Institute for Comparative Tropical Medicine and Parasitology, Ludwig-Maximilians-Universit€ at (LMU), Munich, Germany. Animal trial: Parasitology Division, Kimron Veterinary Institute, Bet Dagan, Israel ELISA: Institute for Comparative Tropical Medicine and Parasitology, LudwigMaximilians-Universit€ at, Munich, Germany.
Received for publication November 15, 2012 doi:10.1111/tbed.12125
Summary The rickettsia Anaplasma marginale causes the haemolytic disease bovine anaplasmosis, an economic problem in tropical and subtropical areas worldwide. The closely related but less pathogenic Anaplasma centrale is commonly used as a live vaccine to prevent anaplasmosis, but it can only be produced from infected blood. UFMG1 is a low pathogenic Brazilian strain of A. marginale, which has been shown to protect cattle against a high pathogenic Brazilian isolate. As UFMG1 can be grown in tick cells, the strain was proposed as a possible cell culturederived vaccine. We have evaluated whether UFMG1 could protect cattle against a geographically distant heterologous strain, using A. centrale vaccination as a standard for comparison. Trial calves were infected with UFMG1, A. centrale or PBS. UFMG1-infected animals were more symptomatic than those infected with A. centrale, but none required treatment. All calves were then challenged with the Israeli A. marginale Gonen strain (one of the most prevalent strain in Israel). The A. centrale group had the mildest symptoms, while UFMG1 and control groups both had a more severe response. Nevertheless, the challenge did not cause life-threatening disease in any group. Animals infected with A. centrale had a significantly higher IgG response than UFMG1, when measured in an ELISA against initial bodies from their homologous strain or Gonen. The level of crossreactivity of the response to initial infection correlated significantly with reduced symptoms after challenge. In conclusion, UFMG1 had limited effect in preventing disease by the geographically distant heterologous Gonen strain. While the low pathogenicity of the Gonen strain in this trial makes it impossible to conclusively state that UFMG1 would have given no protective effect against more serious disease, the comparatively low IgG response to UFMG1 suggests it would not have been as effective as A. centrale.
Introduction Anaplasma marginale (Anaplasmataceae family, Rickettsiales order) is an obligate intracellular bacterium which causes the haemolytic disease bovine anaplasmosis. Endemic worldwide in tropical and subtropical areas, A. marginale can infect a range of ruminant species, but appears to almost exclusively cause disease in cattle (Kocan et al., 2010). It can
be transmitted biologically by a range of tick species, mechanically by biting flies or surgical instruments (Kocan et al., 2010) and transplacentally (Passos and Lima, 1984; Zaugg and Kuttler, 1984). In cattle, it principally infects erythrocytes; this triggers high levels of erythrocyte phagocytosis, leading to anaemia and icterus. Other symptoms include fever, weight loss, lowered milk production and abortion. The disease can be fatal if untreated, particularly
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
97
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
in older animals (Jones and Brock, 1966; Kocan et al., 2010). Cattle that recover from acute disease remain persistently infected, and undergo cycles of recurrent rickettsemia (Palmer et al., 1999; Kocan et al., 2003). Therefore, although these persistently infected cattle are protected from subsequent challenge by homologous strains, they also act as reservoirs of infection for na€ıve cattle (Kocan et al., 2003). Immunity against A. marginale is mediated through a combination of humoral and cellular mechanisms. Antibodies against major surface proteins (MSPs) are associated with protection from infection (Kocan et al., 2003). However, antibody alone is not sufficient for protection (Kuttler and Adams, 1977; Gale et al., 1992), suggesting a requirement for cell-mediated mechanisms. Palmer et al. (1999) proposed a central role of antigen-specific CD4+ T cells, principally their production of IFNc. IFNc stimulates both macrophage activation and B-cell production of IgG2 (the most opsonizing subclass of IgG). The combination of activated macrophages and high levels of IgG2 would lead to increased opsonization and phagocytosis of the rickettsia. To corroborate this, Brown et al. (1998) found that protection after vaccination with outer membranes was associated with CD4+ proliferation and IFNc production and high IgG2 levels. Anaplasma marginale infection itself leads to the down-regulation of high pre-challenge, vaccinationinduced, CD4 T-cell responses. This is likely to contribute to bacterial persistence (Han et al., 2008). Multiple approaches to vaccination against A. marginale have been investigated, as reviewed by Kocan et al. (2003). The principal obstacle for effective vaccination is limited cross-protection between strains, as there is a high level of diversity worldwide, and even considerable variation within smaller areas (Almazan et al., 2008). An effective vaccine in one region may only offer limited protection in another (Palmer et al., 1994; Ocampo Espinoza et al., 2006). The most widespread method of disease prevention in active use is live vaccination with Anaplasma centrale, which is taxonomically closely related to A. marginale. It has been used as a live vaccine from the time it was first identified by Sir Arnold Theiler (1912) and is currently in use in Israel, Australia, South Africa and parts of South America (Shkap et al., 2002). Cattle infected with A. centrale develop a persistent infection with mild or no disease and are subsequently protected from disease caused by A. marginale. Even though it is generally very effective, there have been some reports of vaccine failure or high side effects (Brizuela et al., 1998; Payne et al., 1990; Turton et al., 1998). Despite many attempts, A. centrale has never been successfully cultivated in vitro, and therefore, the vaccine is derived from the blood of infected calves. This risks accidental transmission of other pathogens and requires using animals, which is laborious, expensive, and should be replaced where possible with more ethical alternatives. 98
R. Kenneil et al.
Unlike A. centrale, several A. marginale strains have been successfully established and propagated in tick cell lines (Bell-Sakyi et al., 2007; Passos, 2012). Cell culture-derived vaccines improve safety and reproducibility over those produced from blood (de la Fuente et al., 2002). Bastos et al. (2010) demonstrated that a naturally low pathogenic Brazilian strain (UFMG1) protected cattle from the heterologous high pathogenic Brazilian strain (UFMG2). As UFMG1 has been grown in IDE8 tick cell cultures (Bastos et al., 2009), it was proposed as a potential cell culturederived vaccine (Bastos et al., 2010). One of the main problems with controlling A. marginale is the high geographic variability of strains. Therefore, an essential step in characterizing UFMG1 as a potential vaccine was to investigate whether it could induce a more broadly cross-reactive response and could have potential use as a vaccine outside Brazil. Materials and Methods Anaplasma strains Anaplasma marginale UFMG1 (msp1a sequence GenBank EU676176), originally isolated from a naturally infected calf in Minas Gerais State, Brazil (Ribeiro et al., 1997), had been continuously maintained in IDE8 tick cells for approximately 5 years at the UFMG, using standard culture methods described by Bastos et al. (2010). The A. centrale vaccine strain (full genome sequence GenBank CP001759), routinely used as a live vaccine in Israel (Shkap et al., 2002), had been maintained through inoculation of splenectomized calves at the Kimron Veterinary Institute, Bet Dagan, Israel. Anaplasma marginale Gonen (msp1a sequence GenBank EU678755) originating from the Gonen farm in the north of Israel, was determined to have one of the most common MSP1a genotypes in Israel at the time of its isolation (Molad et al., 2009). Blood collected from the originally infected animal was inoculated into a splenectomized calf, resulting in minimum haematocrit of 9% and maximum rickettsemia of 55% before oxytetracycline treatment to prevent death (V. Shkap, written communication). This study used blood cryostabilate prepared from the splenectomized calf before antibiotic treatment. To obtain sufficient infective material for cattle inoculation, and in order to standardize all strains as fully viable and blood-derived, A. marginale UFMG1, Gonen and A. centrale were first inoculated individually into three splenectomized calves. For this step, the UFMG1 inoculate was initial bodies partially purified from infected IDE8 tick cells, and A. centrale and Gonen inoculates were blood cryostabilates in 15% DMSO. Once splenectomized calves reached high rickettsemia, infected blood was collected and used to inoculate trial cattle.
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
R. Kenneil et al.
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
Trial design The trial cattle were 12 Israeli Friesian calves (3–5 months of age at start of trial), kept under tick- and fly-free conditions, and confirmed negative for A. marginale and A. centrale infections by PCR and MSP-5 cELISA. They were randomly divided into three groups of four animals each and subcutaneously inoculated with 1 9 106 A. marginale UFMG1, 1 9 106 A. centrale or PBS alone (control). After 60 days (which was 1–2 weeks after the last point rickettsemia was >0.09%), animals were challenged with 1 9 107 Gonen strain initial bodies in 2 ml PBS. Progress of infection was monitored at least thriceweekly, and daily during acute infection, by rickettsemia, haematocrit and rectal temperature. Rickettsemia was measured in Giemsa-stained blood smears, with blood taken by ear vein puncture; percentage rickettsemia was calculated by the number of infected erythrocytes divided by total number of erythrocytes, counted over a minimum of 20 fields in 1009 oil immersion. Haematocrit was measured by microhaematocrit technique, again from capillary puncture of the ear vein. The threshold of disease severity at which cattle would be treated with antibiotics was set at a haematocrit remaining below 20% for 3 days. The Kimron Veterinary Institute Animals Welfare Committee and the Israeli Ministry of Health have approved all experiments in cattle (licence number 020_b1731).
500 g for 5 min at 4°C. The supernatant was retained and pelleted at 5000 g for 10 min at 4°C. This pellet containing purified initial bodies was resuspended in 1 ml PBS, and protein concentration was measured by Bio-Rad DC protein assay (Bio-Rad, Hercules, CA, USA). High protein-binding microtiter plates (Nunc, Rochester, NY, USA) were coated with 100 ll of 5 lg/ml initial body-derived protein, diluted in carbonate/bicarbonate buffer pH 9.5, and incubated overnight at 4°C. After washing four times with wash buffer (PBS with 0.05% Tween20), plates were blocked with 200 ll diluent (5% heat-inactivated horse serum in wash buffer), which was incubated shaking for 2 h at RT. One hundred microlitre of each sera sample was then added at a 1 : 90 dilution. After incubation and wash steps as before, 100 ll rabbit antibovine IgG conjugated to alkaline phosphatase (Thermo Scientific) was added at 1 : 5000 dilution, and incubated as previously. After washing, 100 ll p-nitrophenyl phosphate substrate in diethanolamine buffer (Thermo Scientific) was incubated for 30 min, and the reaction stopped by the addition of 50 ll 2 N NaOH. Absorbance values were read at 405 nm by LEDETECT 96 plate reader (Labexim Products, Lengau, Austria). Statistical analyses Statistical significances of differences of clinical parameters and ELISA ODs between the groups were calculated by ANOVA and Tukey HSD post-analysis.
Serological tests Blood samples were collected weekly in EDTA vacutainers (Becton Dickinson, Franklin Lakes, NJ, USA) for serological tests. A competitive ELISA against recombinant MSP5 (VMRD, Pullman, WA, USA) was used to confirm seroconversion; samples were tested according to manufacturers’ instructions. IgG levels against A. marginale strains or A. centrale were measured by an indirect ELISA using purified initial bodies as antigen, based on that described by Shkap et al. (1990). The initial bodies were partially purified from infected blood using a protocol similar to Palmer and McGuire (1984). Briefly, blood was collected in EDTA vacutainers, diluted into an equal volume of cold PBS, and then washed three times in cold PBS by centrifugation at 2600 g for 20 min at 4°C. The buffy coat was removed after each centrifugation. Packed erythrocytes were resuspended in an equal volume of PBS and frozen at 70°C to lyse the cells. They were then thawed at room temperature (RT), and washed in PBS eight times by centrifugation at 23 000 g for 15 min at 4°C (Sorvall SS-34 rotor; Thermo Scientific, Waltham, MA, USA), until no visible pink haemoglobin remained. The final pellet was resuspended in 20 ml PBS, sonicated at 100 W for 2 min, and spun at
Results Clinical response to initial infection with Anaplasma marginale UFMG1 and Anaplasma centrale Cattle infected with UFMG1 developed a clearly detectable rickettsemia (≥0.09%) on average 24 days post-infection; A. centrale infection was detectable after an average of 35 days. Once infection was detected, rickettsia remained measurable for around an average 32 days for UFMG1, and 13 days for A. centrale (Table 1). Rickettsemia peaked at an average 10% for the UFMG1 group and 5% for the A. centrale group. UFMG1 infection caused a more pronounced drop in haematocrit than A. centrale, with a minimum haematocrit of 18–21% in UFMG1-infected animals (average reduction of 13% points), compared with 21–33% for A. centrale (average reduction of 7.5% points) and 28–31% (average reduction of 5% points) for control animals. For both peak rickettsemia and minimum haematocrit, UFMG1 group values were significantly different from the uninfected control group (P < 0.01), whereas A. centrale group values were not significantly different (P > 0.05). The rate of decrease in haematocrit was twice as fast for UFMG1 as for
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
99
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
R. Kenneil et al.
Table 1. Summary of average clinical parameters for all groups, pre- and post-challenge
Peak rickettsemia (%)
Pre-challenge
UFMG1 Anaplasma centrale PBS UFMG1 A. centrale PBS
10 5 0 2.6 1.0 3.05
Post-challenge
3.1 3.6 0 3.2 1.4 3.1
Days of measurable rickettsemia (≥0.09%)
Minimum haematocrit (%)
32 4 13 1 00 15 11 44 20 17
19.8 26 29 21.5 28.3 22
A. centrale. All infections spontaneously resolved without requiring treatment (according to pre-determined threshold). Clinical response after challenge The Gonen strain caused relatively mild infection in all groups, and no animal showed sufficiently severe disease to require treatment. Rickettsemia and haematocrit are summarized and compared with pre-challenge parameters in Table 1. The level of rickettsemia after challenge was low in all groups – no individual calf exceeded 7%, and the group averages were 2.6% for UFMG1, 1.0% for A. centrale and 3.0% for the control group. There was high variability within each group, as shown in Fig. 1. No statistically significant difference was seen between any groups. The UFMG1 and control groups showed a similar level of red blood cell loss after challenge. Their minimum haematocrit ranged between 19–24% (average reduction of 12% points) and 20–26% (average reduction of 13% points), respectively. Animals in the A. centrale group were less affected, with minimum haematocrit between 22% and 32% (average reduction of 5% points). There was a significant difference between minimum haematocrit of UFMG1 and A. centrale-infected animals after challenge (P < 0.05); however, both UFMG1 and A. centrale groups were not significantly different from the control group.
Serum samples from the trial were tested with an antiMSP5 competition ELISA as a confirmation of infection and initial assessment of immune response. All animals were seronegative before the trial started. After infection with UFMG1 or A. centrale, they rapidly seroconverted, and antibody levels remained high until point of challenge and beyond (Fig. 2). The antibody response to UFMG1 infection had minimal cross-reactivity to the Gonen strain (Fig. 3). In contrast, the antibody response to A. centrale (in three out of four animals in the group) had a high level of binding to Gonen 100
1.5 5.3 1.2 2.4 4.3 2.8
39.4 20.3 5.4 34.3 15.3 37.2
12.2 25.6 11.9 12.7 13.1 3.5
8 7 6 5 4 3 2 1 0 –1 UFMG1
A.centrale
PBS
(b)
35 30 25 20
x2
15
UFMG1
Antibody response to infection
Percentage haematocrit reduction (%)
(a)
Peak rickettsemia (% iRBCs)
Group
Minimum haematocrit (%)
Period
A.centrale
PBS
Fig. 1. Peak rickettsemia (a) and minimum haematocrit (b) post-challenge. Horizontal bars represent average group value.
initial bodies (and to UFMG1, data not shown). This crossreactivity correlated significantly with lower red blood cell loss: the higher the level of cross-reactive IgG against the Gonen strain during initial infection, the lower the subsequent drop in haematocrit after challenge (P < 0.01, Pearson’s moment correlation coefficient r = 0.789, r2 = 0.624.
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
100
5
80
4.5
60
4
UFMG1
40
3.5
A.centrale
20
PBS
0 –40
–20
–20
0
20
40
60
80
100
120
140
Days after initial infection
Fig. 2. Total antibody levels against MSP5, measured by MSP5 cELISA. Serum samples were diluted 1 : 64 to prevent saturation of the assay (when measured neat, as in manufacturers’ instructions, all groups reached equal levels, around 90% inhibition). Time of challenge is indicated by the arrow.
OD 405nm
% inhibition
R. Kenneil et al.
3 2.5 2 1.5 1 0.5 0
Group
(a)
Animal number:
2.5
OD 405nm
UFMG1
1 2 3 4
2 1.5 1
0
0
20
40
60
80
100
120
Days post initial infection
(b)
2.5
Animal number: 5
OD 405nm
2
6 7
1.5
8
1 0.5
–20
(c)
0
0
20
40 60 80 Days post initial infection
2.5
100
Discussion
OD 405nm
9 10
1.5
11 12
1 0.5 0 0
20
40
60
80
post-challenge), tested against initial bodies from the homologous strain. The highest IgG levels were generally around 1 month post-infection. UFMG1 infection had a relatively low IgG response (average 3-fold increase above pre-immune levels, data not shown). Gonen infection led to a slightly higher response, an average 7-fold increase, with one very high-responding animal. Anaplasma centrale infection provoked considerably higher IgG levels in three of four animals (average 12-fold increase).
120
Animal number:
2
–20
Gonen
Fig. 4. Maximum IgG levels against initial bodies from homologous strains, after infection with UFMG1, Anaplasma centrale or Gonen alone. Significant difference between groups is labelled with an asterisk (P < 0.05); horizontal bars represent the group average.
0.5
–20
A.centrale
100
120
Days post initial infection
Fig. 3. IgG levels against A. marginale Gonen strain in animals from the UFMG1 (a), A.centrale (b), and control (c) groups. Time of challenge with Gonen strain is indicated by arrow.
To determine whether the difference in anti-Gonen IgG levels was due to a difference in strength of the overall IgG response, IgG levels against homologous initial bodies were measured. Figure 4 shows the maximum IgG response of cattle infected with UFMG1, A. centrale or Gonen alone (for the latter, samples were taken from control group
As reported here, UFMG1 had a negligible effect on disease caused by the Gonen strain. This correlated with a low level of cross-reactive IgG produced during UFMG1 infection, compared with considerably higher levels induced by the more protective A. centrale vaccine strain. The lack of protection from UFMG1 against the heterologous Israeli strain tested here contrasts with its complete protection against a highly pathogenic Brazilian strain, UFMG2 (Bastos et al., 2010). However, as the Israeli Gonen strain was not highly pathogenic in this trial, it cannot be conclusively stated that UFMG1 would have had no protective effect against fatal levels of disease. The level of disease caused by the A. marginale strains and A. centrale varied considerably within each group, both before and after challenge. For the latter, this correlated with very variable levels of Anaplasma-specific IgG production. Even in A. centrale-infected animals, one animal produced a very low antibody response and showed no reduction in symptoms after challenge. This is not
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
101
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
unexpected in animals not specifically inbred for laboratory studies, but does highlight the difficulties in inducing complete coverage across a diverse population. Results from the MSP5 cELISA were very different from the IgG ELISA against initial bodies, with total antibody levels against MSP5 lower in the A. centrale group compared with the UFMG1 group. This could have a number of possible explanations. First, the cELISA measures all antibody classes. Therefore, higher responses measured by this assay could be largely due to IgM. Although mainly produced in early infection, Klaus and Jones (1968) and Murphy et al. (1966) found IgM persisted until 60–80 days post-A. marginale infection. Second, the protocol for initial body preparation could damage some surface antigens, which would lower the IgG ELISA results. However, since initial bodies for A. centrale and both A. marginale strains were prepared in the same way, this would only skew the relative antibody production by between different groups if more A. marginale-specific antigens were damaged, which seems unlikely. Third, MSP5 is a highly conserved protein, demonstrated by the fact that this cELISA can be used to detect antibodies after A. marginale, A. centrale and Anaplasma phagocytophilum infections (Molloy et al., 1999; Dreher et al., 2005). As A. marginale immunity is highly strain-specific, antibodies against this conserved protein are unlikely to correlate well with protection – and as seen here, anti-MSP5 antibody levels do not appear to correlate with IgG levels against total surface protein. While the MSP5 cELISA assay is very useful for measuring seroconversion as an indicator of infection, it has limited application in estimating protective immunity. Measuring the IgG response against all MSPs, either by ELISA as used here, or by Western blot as in Brown et al. (1998), appears to give more information on the likely degree of cross-protection induced by infection or vaccination. There is known to be high cross-reactivity between A. marginale and A. centrale (Agnes et al., 2011). In addition, the considerably higher level of IgG production in response to A. centrale infection seen here is likely to contribute to its success as a vaccine. As previously mentioned, a model of immunity against A. marginale emphasizes the importance of antigen-specific activated CD4+ T cells and their IFNc production to stimulate antibody class switching of B cells, and cause high levels of IgG production (Palmer et al., 1999). Differences in CD4+ T-cell function and their level of IFNc production in response to stimuli would therefore be a likely candidate to correlate with differences in IgG levels between and within groups, and this will be tested in future work. Anaplasma marginale infection can down-regulate high pre-challenge, recombinant vaccineinduced CD4+ T-cell responses (Abbott et al., 2005; Han et al., 2008). Han et al. (2008) proposed this to be due to deletion of antigen-specific T cells after overstimulation 102
R. Kenneil et al.
during infection. It has not been investigated if this phenomenon also occurs in A. centrale infection – if it does not, this could be a central factor behind the higher antibody response seen here from this group. The smaller difference in IgG levels after UFMG1 versus Gonen infection may be due to UFMG1 being maintained in tick cell culture for a number of years before inoculation into the splenectomized calf. Tick cell culture stimulates different A. marginale protein expressions compared with mammalian cells (Garcia-Garcia et al., 2004), and it is possible that prolonged tick cell culture may have had an enduring effect on UFGM1 protein expression. In A. marginale, this is known to be the case for the major surface protein MSP1, a heterodimer composed of MSP1a and MSP1b. Both subunits are adhesins for bovine erythrocytes, but only MSP1a also binds to tick cells (Palmer and McGuire, 1984; de la Fuente et al., 2002). msp1a is more highly expressed in blood than in tick cell culture (Garcia-Garcia et al., 2004). The higher levels of MSP1a on the surface of A. marginale in cattle are proposed to improve the chances of successful transmission to feeding ticks. de la Fuente et al. (2002) found that higher levels of anti-MSPa IgG are stimulated by blood-derived A. marginale than by tick cellderived. MSP1a is an important antigen for both cellular and humoral responses: CD4 T-cells preferentially recognize MSP1a when immunized with the recombinant MSP1 complex (Brown et al., 2001), and antibodies against MSP1a inhibit the entry of rickettsia into red blood cells (Palmer and McGuire, 1984). Therefore, the lowered antiMSP1a response from tick cell-derived rickettsia could influence their effectiveness as vaccines. Bastos et al. (2010) found no significant difference in the level of protection provided by tick cell or blood-derived UFMG1, but in this case, tick cell cultures were under 1 year old after initialization from blood stabilate, as opposed to 5 years in this trial. UFMG1 infection itself gave rise to symptoms that were not severe enough to require intervention, but did come close to the threshold set for antibiotic treatment. Any pronounced pathogenicity could hinder the use of UFMG1 as a live vaccine. If A. marginale is transmitted from vaccinated calves to na€ıve adults by tick or mechanical transmission, the disease is very likely to be fatal – as seen in experimental infection of adult cattle with A. centrale (Pipano et al., 1985). The risk of this occurring after A. centrale vaccination is limited by its extremely restricted biological transmission – only reported to date with the African tick Rhipicephalus simus (from which it was initially isolated; Theiler, 1912; Potgieter and van Rensburg, 1987). Several other tick species that can act as A. marginale vectors cannot transmit A. centrale: Hyalomma excavatum, Rhipicephalus sanguineus, and Rhipicephalus (Boophilus) annulatus, tested by Shkap et al. (2009). Dermacentor
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
R. Kenneil et al.
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
andersoni was shown by Ueti et al. (2009) to transmit A. centrale, but with extremely low efficiency (over 400 infected ticks were used for one calf). Goncßalves-Ruiz et al. (2005) showed that A. marginale UFMG1 was not transmitted by Rhipicephalus (Boophilus) microplus ticks, which are the major biological vector of A. marginale in Brazil. However, as A. marginale is known to be transmissible by at least 20 tick species (Kocan et al., 2010), any further studies on UFMG1 as a live vaccine in regions with other vector species should investigate the effect of UFMG1 in adult cattle and its transmissibility by other species of tick vector. In conclusion, UFMG1 had a limited effect on disease caused by a geographically distant heterologous strain. The low pathogenicity of the Israeli strain in this trial makes it impossible to determine whether UFMG1 could have reduced fatal disease. However, the limited cross-reactivity of the IgG response to UFMG1 suggests that it would be less protective than A. centrale against a more pathogenic strain. The high levels of cross-reactive IgG induced by A. centrale infection are likely to be a critical factor in its success as a vaccine in a wide range of countries. However, as antibody responses alone are insufficient to protect cattle from bovine anaplasmosis, further work must be carried out to characterize other differences between A. marginale and A. centrale infection, and to determine whether these factors can be harnessed for future vaccines. Acknowledgements IDE8 cells were kindly provided by Dr U.G. Munderloh. R. Kenneil is a Marie Curie Early Stage Researcher (ESR) supported by the POSTICK ITN (Postgraduate training network for capacity building to control ticks and tickborne diseases) within the FP7- PEOPLE – ITN programme (EU Grant No. 238511). Conflicts of interest The authors have no conflicts of interest to declare. References Abbott, J. R., G. H. Palmer, K. A. Kegerreis, P. F. Hetrick, C. J. Howard, J. C. Hope, and W. C. Brown, 2005: Rapid and longterm disappearance of CD4 T lymphocyte responses specific for Anaplasma marginale major surface protein-2 (MSP2) in MSP2 vaccinates following challenge with live A. marginale. J. Immunol. 174, 6702–6715. Agnes, J. T., K. A. Brayton, M. LaFollett, J. Norimine, W. C. Brown, and G. H. Palmer, 2011: Identification of Anaplasma marginale outer membrane protein antigens conserved between A. marginale sensu stricto strains and the live A. marginale subsp. centrale vaccine. Infect. Immun. 79, 1311–1318.
Almazan, C., C. Medrano, M. Ortiz, and J. de la Fuente, 2008: Genetic diversity of Anaplasma marginale strains from an outbreak of bovine anaplasmosis in an endemic area. Vet. Parasitol. 158, 103–109. Bastos, C. V., L. M. F. Passos, M. M. C. Vasconcelos, and M. F. B. Ribeiro, 2009: In vitro establishment and propagation of a Brazilian strain of Anaplasma marginale with appendage in IDE8 (Ixodes scapularis) cells. Braz. J. Microbiol. 40, 399–403. Bastos, C. V., L. M. F. Passos, E. J. Facury-Filho, E. M. Rabelo, J. de la Fuente, and M. F. B. Ribeiro, 2010: Protection in the absence of exclusion between two Brazilian isolates of Anaplasma marginale in experimentally infected calves. Vet. J. 186, 374–378. Bell-Sakyi, L., E. Zweygarth, E. F. Blouin, E. A. Gould, and F. Jongejan, 2007: Tick cell lines: tools for tick and tick-borne disease research. Trends Parasitol. 23, 450–457. Brizuela, C. M., C. A. Ortellado, E. Sanabria, O. Torres, and D. Ortigosa, 1998: The safety and efficacy of Australian tickborne disease vaccine strains in cattle in Paraguay. Vet. Parasitol. 76, 27–41. Brown, W. C., V. Shkap, D. Zhu, T. C. McGuire, W. Tuo, T. F. McElwain, and G. H. Palmer, 1998: CD4+ T-Lymphocyte and immunoglobulin G2 responses in calves immunized with Anaplasma marginale outer membranes and protected against homologous challenge. Infect. Immun. 66, 5406–5413. Brown, W. C., G. H. Palmer, H. A. Lewin, and T. C. McGuire, 2001: CD4+ T lymphocytes from calves immunized with Anaplasma marginale major surface protein 1 (MSP1), a heteromeric complex of MSP1a and MSP1b, preferentially recognize the MSP1a carboxyl terminus that is conserved among strains. Infect. Immun. 69, 6853–6862. Dreher, U. M., J. de la Fuente, R. Hofmann-Lehmann, M. L. Meli, N. Pusterla, K. M. Kocan, Z. Woldehiwet, U. Braun, G. Regula, K. D. Staerk, and H. Lutz, 2005: Serologic cross-reactivity between Anaplasma marginale and Anaplasma phagocytophilum. Clin. Diagn. Lab. Immunol. 12, 1177–1183. de la Fuente, J., K. M. Kocan, J. C. Garcia-Garcia, E. F. Blouin, P. L. Claypool, and J. T. Saliki, 2002: Vaccination of cattle with Anaplasma marginale derived from tick cell culture and bovine erythrocytes followed by challenge-exposure with infected ticks. Vet. Microbiol. 89, 239–251. Gale, K. R., G. Leatch, M. Gartside, and C. M. Dimmock, 1992: Anaplasma marginale: failure of sera from immune cattle to confer protection in passive-transfer experiments. Parasitol. Res. 78, 410–415. Garcia-Garcia, J. C., J. de la Fuente, E. F. Blouin, T. J. Johnson, T. Halbur, V. C. Onet, J. T. Saliki, and K. M. Kocan, 2004: Differential expression of the msp1a gene of Anaplasma marginale occurs in bovine erythrocytes and tick cells. Vet. Microbiol. 98, 261–272. Goncßalves-Ruiz, P. M., L. M. F. Passos, and M. F. B. Ribeiro, 2005: Lack of infectivity of a Brazilian Anaplasma marginale isolate for Boophilus microplus ticks. Vet. Parasitol. 128, 325–331.
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
103
Antibody Responses in Cross-Protection Between Anaplasma marginale Isolates
Han, S., J. Norimine, G. H. Palmer, W. Mwangi, K. K. Lahmers, and W. C. Brown, 2008: Rapid deletion of antigen-specific CD4+ T cells following infection represents a strategy of immune evasion and persistence for Anaplasma marginale. J. Immunol. 181, 7759–7769. Jones, E. W., and W. E. Brock, 1966: Bovine anaplasmosis: its diagnosis, treatment, and control. J. Am. Vet. Med. Assoc. 149, 1624–1633. Klaus, G. G., and E. W. Jones, 1968: The immunoglobulin response in intact and splenectomized calves infected with Anaplasma marginale. J. Immunol. 100, 991–999. Kocan, K. M., J. de la Fuente, A. A. Guglielmone, and R. D. Melendez, 2003: Antigens and alternatives for control of Anaplasma marginale infection in cattle. Clin. Microbiol. Rev. 16, 698–712. Kocan, K. M., J. de la Fuente, E. D. Blouin, J. F. Coetzee, and S. A. Ewing, 2010: The natural history of Anaplasma marginale. Vet. Parasitol. 167, 95–107. Kuttler, K. L., and L. G. Adams, 1977: Influence of dexamethasone on the recrudescence of Anaplasma marginale in splenectomized calves. Am. J. Vet. Res. 38, 1327– 1330. Molad, T., L. Fleidrovich, M. Mazuz, L. Fish, B. Leibovitz, Y. Krigel, and V. Shkap, 2009: Genetic diversity of major surface protein 1a of Anaplasma marginale in beef cattle. Vet. Microbiol. 136, 54–60. Molloy, J. B., P. M. Bowles, D. P. Knowles, T. F. McElwain, R. E. Bock, T. G. Kingston, G. W. Blight, and R. J. Dalgliesh, 1999: Comparison of a competitive inhibition ELISA and the card agglutination test for detection of antibodies to Anaplasma marginale and Anaplasma centrale in cattle. Aust. Vet. J. 77, 245–249. Murphy, F. A., J. W. Osebold, and O. Aalund, 1966: Kinetics of the antibody response to Anaplasma marginale infection. J. Infect. Dis. 116, 99–111. Ocampo Espinoza, V., J. E. Vazquez, M. D. Aguilar, M. A. Ortiz, G. J. Alarc on, and S. D. Rodrıguez, 2006: Anaplasma marginale: lack of cross-protection between strains that share MSP1a variable region and MSP4. Vet. Microbiol. 114, 34–40. Palmer, G. H., and T. C. McGuire, 1984: Immune serum against Anaplasma marginale initial bodies neutralizes infectivity for cattle. J. Immunol. 133, 1010–1015. Palmer, G. H., D. Munodzana, N. Tebele, T. Ushe, and T. F. McElwain, 1994: Heterologous strain challenge of cattle immunized with Anaplasma marginale outer membranes. Vet. Immunol. Immunopathol. 42, 265–273. Palmer, G. H., F. R. Rurangirwa, K. M. Kocan, and W. C. Brown, 1999: Molecular basis for vaccine development against the ehrlichial pathogen Anaplasma marginale. Parasitol. Today 15, 253–300.
104
R. Kenneil et al.
Passos, L. M. F., 2012: In vitro cultivation of Anaplasma marginale and A. phagocytophilum in tick cell lines: a review. Rev. Bras. Parasitol. Vet. 21, 81–86. Passos, L. M. F., and J. D. Lima, 1984: Diagnosis of congenital anaplasmosis in Minas Gerais. Arq. Bras. Med. Vet. Zootec. 36, 743–744. Payne, R. C., O. Osorio, and A. Yba~ nez, 1990: Tick-borne diseases of cattle in Paraguay. II. Immunisation against anaplasmosis and babesiosis. Trop. Anim. Health Prod. 22, 101– 108. Pipano, E., E. Mayer, and M. Frank, 1985: Comparative response of Friesian milking cows and calves to Anaplasma centrale vaccine. Br. Vet. J. 141, 174–178. Potgieter, F. T., and L. van Rensburg, 1987: Tick transmission of Anaplasma centrale. Onderstepoort J. Vet. Res. 54, 5–7. Ribeiro, M. F. B., L. M. F. Passos, and A. M. Guimar~aes, 1997: Ultrastructure of Anaplasma marginale with an inclusion appendage, isolated in Minas Gerais State, Brazil. Vet. Parasitol. 70, 271–277. Shkap, V., H. Bin, H. Ungar-Waron, and E. Pipano, 1990: An enzyme-linked immunosorbent assay (ELISA) for the detection of antibodies to Anaplasma centrale and Anaplasma marginale. Vet. Microbiol. 25, 45–53. Shkap, V., T. Molad, L. Fish, and G. Palmer, 2002: Detection of the Anaplasma centrale vaccine strain and specific differentiation from Anaplasma marginale in vaccinated and infected cattle. Parasitol. Res. 88, 546–552. Shkap, V., K. Kocan, T. Molad, M. Mazuz, B. Leibovich, Y. Krigel, A. Michoytchenko, E. Blouin, J. de la Fuente, M. Samish, M. Mtshali, E. Zweygarth, E. L. Fleiderovich, and L. Fish, 2009: Experimental transmission of field Anaplasma marginale and the A. centrale vaccine strain by Hyalomma excavatum, Rhipicephalus sanguineus and Rhipicephalus (Boophilus) annulatus ticks. Vet. Microbiol. 134, 254–260. Theiler, A., 1912: Gallsickness of imported cattle and the protective inoculation against this disease. Agric. J. Union S. Afr. 3, 7–46. Turton, J. A., T. C. Katsande, M. B. Matingo, W. K. Jorgensen, U. Ushewokunze-Obatolu, and R. J. Dalgliesh, 1998: Observations on the use of Anaplasma centrale for immunization of cattle against anaplasmosis in Zimbabwe. Onderstepoort J. Vet. Res. 65, 81–86. Ueti, M. W., D. P. Knowles, C. M. Davitt, G. A. Scoles, T. V. Baszler, and G. H. Palmer, 2009: Quantitative differences in salivary pathogen load during tick transmission underlie strain-specific variation in transmission efficiency of Anaplasma marginale. Infect. Immun. 77, 70–75. Zaugg, J. L., and K. L. Kuttler, 1984: Bovine anaplasmosis: in utero transmission and the immunologic significance of ingested colostral antibodies. Am. J. Vet. Res. 45, 440–443.
© 2013 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 60 (Suppl. 2) (2013) 97–104
