Available online at www.sciencedirect.

com

Vaccine 25 (2007) 82988305

The NSP immune response of vaccinated animals after in-eld exposure to FMDV
H. Yadin a , J. Brenner a, , D. Chai a , Z. Oved b , Y. Hadany b , A. Kusak b , M. Haimovich b
a

FMD Laboratory Kimron Veterinary Institute, 50250 Bet Dagan, Israel b Field Regional Veterinary Services, 50250 Bet Dagan, Israel

Received 4 July 2007; received in revised form 11 September 2007; accepted 20 September 2007 Available online 22 October 2007

Abstract The aim was to examine the immune response (IR) to non-structural proteins (NSPs), in order to assess the validity of the detection of antibodies to NSPs as a means of diagnosing foot and mouth disease (FMD infection) infection when vaccinated populations are in close contact with clinically sick animals. The study was performed during FMD outbreaks in Israel in January 2004; the IR was examined in vaccinated dairy and feedlot cattle herds under natural eld exposure to FMDV, and in vaccinated and unvaccinated sheep ocks. During the 2004 outbreaks, clinical signs were age-related and were noted only among imported calves, although they had been vaccinated; such signs were not found among the local dairy cattle populations. The NSP IR among the feedlot cattle that had been vaccinated more than 4 months prior to the in-eld exposure was 86%, compared with only 30% among those feedlot cattle that had received one dose of vaccine less than 4 months before the eld exposure. The prevalence of NSP IR indicates that animals vaccinated once, less than 4 months prior to exposure, were clinically resistant to FMDV infection, although possibly still susceptible to subclinical infections, whereas those vaccinated more than 4 months prior to the in-eld exposure presented clinical manifestations. This situation is unlikely to occur among repeatedly vaccinated livestock; these remained refractory to FMD exposure, as reected in the absence of clinical manifestations and a relatively low prevalence of NSP IR compared with that in imported calves. 2007 Elsevier Ltd. All rights reserved.
Keywords: FMDV; Vaccination; NSP

1. Introduction Foot and mouth disease (FMD) is the most contagious disease of mammals, and is not only expensive to the owners of the infected animals [1] but may also be expensive for the whole economy of a country. FMD is caused by a virus of the genus Aphtovirus, family Picornaviridae; it cannot be differentiated clinically from other vesicular diseases, including swine vesicular disease, vesicular stomatitis, and vesicular exantema [2]; therefore, laboratory diagnosis of any suspected FMD case is a matter of urgency.

Corresponding author. Tel.: +972 3 9681668; fax: +972 3 9681788. E-mail address: yakovb@moag.gov.il (J. Brenner).

A rational approach to study recent or previous FMDV infections is by detecting viral antigens or specic antibody responses. The tests generally used are virus neutralization (VN) and ELISA [38]. Detection of antibodies to the non-structural proteins (NSPs) has been used to identify past or ongoing infections with any of the existing serotypes of FMDV [35,912]. The modern FMD vaccines undergo a step of antigen purication, which makes them free of NSPs; therefore, NSPs should not be present in the inactivated FMD vaccines so that nearly every NSP IR that is detected should be due to eld infection. The use of antibodies to anti-virus infection-associated antigen (anti-VIAA) as an indicator to distinguish between animals that had undergone eld exposure and those that had been vaccinated [11], which was the practice a few decades

0264-410X/$ see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.vaccine.2007.09.071

H. Yadin et al. / Vaccine 25 (2007) 82988305

8299

ago, are to be replaced with assays capable of detecting anti-NSP antibodies. Animals that are seropositive to the polyproteins 3AB or 3ABC or to one or more of the NSPs, including the L, 2C, 3A, 3B or 3D, are (or were considered) likely to be infected by eld viruses [5,6]. However, vaccine purity is an important issue, as the presence of trace amount of NSPs in the vaccine could result in false positive reactions in animals that had been repeatedly vaccinated [5,10,12]. The validity of NSP responses or of their interpretation has never been assessed under natural conditions in which vaccinated populations are in close contact with clinically sick animals so that they are exposed to natural challenge in addition to their vaccination. An opportunity to make such an assessment arose during the FMD outbreak that occurred in Israel in January 2004. Although Israel is located in the Middle East, in an FMDendemic area, no FMD outbreak has been recorded since August 1999. To minimize FMD episodes, all livestock either ruminants or pig are vaccinated conscientiously at least once a year, and young animals are vaccinated twice before reaching the age of 1214 months. Additionally, beef cattle (male calves) that are imported to Israel at the age of 12 months for slaughter at 1014 months of age are vaccinated upon arrival and are then liable to receive a repeat vaccination during the annual vaccination campaign. In January 2004, Israel experienced an FMD outbreak; clinical signs suspected to be due to FMD were observed in seven farms. Studying this episode has placed us in a position to address the question of whether it is possible to distinguish between vaccinated protected and vaccinated partially protected animals under natural eld exposure. Moreover, this episode enabled us to evaluate the solidity of protection imparted to repeatedly immunized animals and to those immunized only once. Additionally, we tried to relate the individual immune status, as indicated by whether an animal was diseased or not, to the elapsed time between the single vaccination and the eld exposure. The aim of the present study was to examine the NSP immune response (IR) in vaccinated dairy and beef cattle herds and in vaccinated and unvaccinated sheep ocks, under natural eld exposure to FMDV. We adopted the rational approach that detection of NSP is a good indicator of FMDV replication in the host after eld exposure.

Picture 1. Outbreaks of FMD in Israel in January 2004.

Table 1 The populations tested after the January 2004 episode Episodea Farm Herd type Number of animals with FMD lesions/no. of animals on farm 0/460 46/750 0/331 177/360 43/600 39/300 0/40 0/40 80/400 (estimated) 0/580 0/120 0/350 4/80 No. of tested animals 141 59 152 134 43 69 40 40 23 223 80 131 69 1204

1B 1A 2A 2B 3

EHb EHb G(a)b G(a)b G(b)b KV Collection Collection AZb E SN VC NSFb Total

Dairy Feedlot Dairy Feedlot Feedlot Dairy Feedlot Dairy Feedlot Dairy Beef Sheep Sheep

2. Materials and methods 2.1. The epidemiological background (consult also the map, Picture 1) On January 18, 2004, in a mixed cattle farm (dairy and feedlot), located in the inner southern coastal plain (G(a); Table 1), many beef calves aged between 10 and 12 months were observed to be depressed, with massive sciallorea and to be emitting kissing sounds; they were restless and trampled the ground.

5
a b

Consult the map. Farms located in the affected sites.

8300

H. Yadin et al. / Vaccine 25 (2007) 82988305

An FMDV outbreak was conrmed by antigen detection (Hagai, personal communication). Upon detection of the outbreak, the animals in the infected site, as well as those in all other farms situated in the same village, i.e., the rst southern village received a booster vaccination. Animal movements were banned, and the sick animals remained within their original premises. Those that died during the epidemic were buried on the spot. The number of the animal sample represents the result of the physical effort of dealing with groups of growing calves kept in yards without adequate means of constraint (feedlot cattle), or in the open eld (beef cattle). However, almost all the feedlot and beef groups were sampled, regardless of their accessibility, whereas the dairy herds are presented in the samples almost in their entirety. Feedlot calves of about 10 months are presented in very small numbers, although they were the most affected population, because of the constraint difculties. The dairy cattle were of the Israeli Holstein-Friesian type, whereas the feedlot cattle were imported calves, together with a collection of young male calves that had been gathered from neighboring dairy herds and introduced into the farm at approximately 6 weeks of age. All the animals designated for fattening were maintained in yards, each containing a group of 5060 animals of closely similar ages, i.e., within a range of no more than 1 month. There was a close intimate contact between one of the groups of calves and one of the groups of 210-month-old dairy heifers (G(a); Table 1). The distances between the dairy sheds and feedlot yards in all the mixed farms were not more then a few dozens of meters. One day after the appearance of clinical signs in the rst farm (G(a)), similar signs appeared in a neighboring feedlot farm (G(b); Table 1), and an additional infectious focus was reported from a feedlot farm about 150 km to the north (AZ; Table 1). This third farm housed an estimated 400 imported beef calves aged between 4 and 10 months. An investigation revealed that a cattle transporter truck visited feedlot farms in the West Bank, and the third (north, AZ) and the rst (south, G(a)) farms on the same day, some days before onset of the clinical signs. Moreover, the son of the owner of the second feedlot farm (south, G(b)) owned the suspected cattle transporter (Picture 1). A week after the primary outbreaks, two additional farms within the 10-km restricted zone around the northern infection site reported FMD outbreaks; one in a mixed cattle farm (EH) and the other in an unvaccinated sheep ock (NSF) (Picture 1). The clinical affected animals were observed only among the imported feedlot calves and the unvaccinated sheep ock. Moreover, only imported calves that had been vaccinated only once experienced illness, and there was an age-related pattern. The clinical signs were observed only in calves aged more than 7 months whereas among the dairy cattle, the younger calves and the local calves were spared.

2.2. Experimental design and selection of the target populations The target populations, i.e., both those from the infected sites and the controls that were tested for the prevalence of NSP are listed in Table 1. According to the cattle type (dairy, feedlot or beef) and the schedule of routine FMD vaccinations, the studied populations were subdivided into age-related categories as follows. 2.2.1. Dairy cattle The animals that had been vaccinated at least three times, were selected from among the cows aged more than 24 months. This subpopulation was designated old animals. The younger cows, aged less than 24 months, which had been vaccinated only once or twice, were designated young animals. The subdivisions of the dairy cattle populations are summarized in Table 4. 2.2.2. Feedlot calves As mentioned above, this population had been vaccinated only once; therefore, the rationale of subdivision was different from that applied to the dairy cattle. Among the calves with clinical disease manifestation, those aged 7 months or more (feedlot, not grazing animals) were designated old animals and those aged less than 7 months as young animals. The subdivisions of the feedlot calves populations are summarized in Table 6. 2.2.3. Beef cattle A sample of 80 animals was taken in a herd of 120 adult beef (Charolais) dams that grazed in the Northern Golan Heights (farm SN), an area as far as 200 km from the northern FMD outbreak. The NSP responders of this herd were used as control for comparison with FMD-free herds in infected and uninfected sites: EH and E, respectively. All those cattle populations had been vaccinated at least three times; therefore, these grazing beef cattle were classied as adult beef cattle, i.e., repeatedly vaccinated population. The differences between these subpopulations, in prevalence of NSP IR, were evaluated by the 2 test and the 95% coefcient interval (CI) was calculated. Two distinct subpopulations were selected and their prevalence of NSP responders was compared with that among the adult dairy cattle from the infective sites. The rst subpopulation comprised 223 adult dairy cattle out of 580 from the Negev desert (herd E), which had never experienced FMDV infection, 69 adult lactating cows from small villages situated more than 20 km from the infected site (KV), and a collection of additional 40 lactating cows from several small holdings located more than 20 km from the southern infection site; the second group comprised 80 out of 120 adult beef cattle (SN).

H. Yadin et al. / Vaccine 25 (2007) 82988305

8301

2.2.4. Sheep Two sheep ocks were investigated and the ndings are summarized in Table 8. One ock was located within the restricted infected perimeter; the control, uninfected ock was located on a sheep farm located 100 and 50 km, from the northern and southern infective foci, respectively. 2.3. NSP surveillance The blood of animals on all the affected farms and in the sheep ocks was sampled 3080 days after disappearance of the clinical signs. Details of the uninfected dairy populations are presented in Tables 1, 35. NSP antibodies were evaluated with commercial ELISA kits: the FMDV 3ABC-Ab (SVANOVATM Uppsala, Sweden) and the Ceditest FMDV-NS (Cedi Diagnostics Lelystad, the Netherlands) were used for bovine and sheep sera, respectively, according to the manufacturers instructions. 2.4. Statistics The SPSS-13 statistical package (SPSS Inc., Chicago, USA) was used to evaluate the correlation between the age subgroups and the prevalence of NSP-IR in the various cattle populations. Non-parametric tests were used to compare the proportions of NSP responders among the various subpopulations; the comparisons are shown in Tables 3, 4, 6 and 7. 2.5. Clinical considerations Because of the difculty of performing individual clinical examination, especially of the heavier beef calves, changes of behavior such as loneliness/isolation (standing alone) and reluctance to eat, together with abnormal salivation and/or nervous movements were considerer to be signs of illness, and mortality was reported immediately.

3. Results 3.1. Clinical manifestations No morbidity or mortality was observed among the dairy herds. Moreover, no clinical signs were observed, even in the yard of dairy heifers located next to the infected calves (farm G(a)). The data on the prevalence of clinical cases diseased or deceased among the feedlot calves are presented in Table 2. In the two mixed-cattle farms (G(a) and EH), only the feedlot calves presented clinical manifestations (Table 2). In the southern outbreak (G(a)), 177 out of 210 (84%) calves aged 7 months or more became sick, and 1 of them died (Table 2). Among the young calves, aged between 6 weeks and 7 months, no sick animals have been observed but 1 calf out of 150 died during the epidemic. No post-mortem examination was performed since the carcass was destroyed; however, the owner claimed that this death was not related to the FMD episode. In the northern mixed cattle farm (EH) out of 48 calves aged more than 7 months, 43 showed clinical signs and another 2 died, i.e., 94% infected, whereas only 1 out of 702 young calves was depressed (Table 2). In the other feedlot farm, located in the southern focus of infectious (G(b)), 35 out of 40 calves aged over 7 months became sick, and an additional three calves died, i.e., a total of 95% infected, whereas only 1 out of 560 younger calves died, probably related to FMDV infection (Table 2). No postmortem exanimation was performed. The northern feedlot (AZ) housed only a stock of older calves. An estimate 20% of the animals (several hundred, which the owner refused to declare) presented clinical manifestations that were observed in the eld. No mortality was observed in the affected sheep ock but 4 out of 80 (5.0%) were sick (data not presented).

Table 2 The number of diseased and deceased ( ) animals among the feedlot cattle populations originating from the infected (*) and uninfected sites Older approximately 79 months old EH* G(a)* G(b*) SN Collection AZ* 43/48 (94%) (2 ) 177/210 (84%) (1 ) 35/40 (95%) (3 ) Younger less than 7 months old 1/702 (0.1%) 1 /150 (0.7%) 1 /560 (0.2%) 7/80 (8.8%; 417%) 40 (0%) 20% (estimated) Adult beef 24 months old or older

Table 3 Prevalence of NSP responders in the dairy and feedlot/beef cattle populations tested in this study NSP+ (%; 95% CI) from infected sites Feedlot/beef Dairy cattle**
* **

NSP+ (%; 95% CI) from non-infected sites 8/120 (6.7%; 2/332 (0.8%; 03.0%)** 313%)*

Total 379 625

cattle*

134/259 (52.1%; 12/293 (2.3%; 14%)**

4658%)*

P < 0.001. P = 0.02.

8302

H. Yadin et al. / Vaccine 25 (2007) 82988305 Table 6 Comparison of the prevalence of the NSP responders between the young and old feedlot cattle subpopulations NSP+ (%) Less than 7 months old More than 7 months old Total 36 (26.3%) 98 (86.0%) 134 (51.7%) NSP 109 16 125 Total 145 114 259

Table 4 Comparison of the prevalence of the NSP responders between the young and old dairy subpopulations NSP+ Less than 24 months old More than 24 months old Total 3 (1.1%) 12 (2.9%) 15 (2.4%) NSP 212 398 610 Total 215 410 625

Correlation: 0.109; P = 0.18, not signicant.

Correlation 0.424; P > 103 ; Chi-square test; P > 103 . Table 7 The distribution of the NSP responders among the various feedlot farms from the infected sites Older approximately 79 months EH G(a) G(b) AZ Total 37/38 (97%) 38/53 (72%) 23/23 (100%) 98/114 (86.0%)* 36/145 (26.3%)* Younger less than <7 months 8/21 (38.1%) 18/81 (22.2%) 10/43 (23%)

3.2. Immune response to FNDV NSPs Out of 1004 cattle samples that were tested, 149 (16.7%) were NSP reactors; their distribution between the dairy and the feedlot/beef subpopulations is presented in Table 3. The comparisons between the NSP IR rates among the infected and uninfected sites revealed statistically signicant differences, at P < 0.001 and P = 0.02 in the feedlot/beef and dairy cattle populations, respectively. The comparison between the proportions of NSP responders among the young and the old dairy cattle are summarized in Table 4; the difference between the young and the older dairy cattle populations was not statistically signicant, although a trend (P = 0.18) was observed, and the correlation between the age of animals and their NSP-IR was very low. However, there was a statistically signicant difference between the prevalence of NSP responders among the entire dairy cattle population and that among the adult beef cattle population chosen for this study (Table 4). However, comparison between the prevalence of NSP responders among the adult dairy cattle from the infected focus (EH) and among those from the uninfected control herds (E, KV, and the collection of adult dairy cattle) (Table 5) also yielded a statistically signicant (P < 0.0001) difference, whereas the prevalence of NSP responders among the adult dairy population of EH showed no statistically signicant difference (P = 0.6) from that among the adult beef cattle of the control farm SN (Table 3). That the adult dairy population of EH behaved differently is also reected in Table 5, where the distribution of NSP responders among the various dairy farms is presented.
Table 5 The distribution of the NSP immune responders among the different dairy farms (%; 95% C. I) Lactating cows EHa G(a)a E KV Collection of dairy cattle Total 9/66 (13.6%; 624%) 1/52 (2%; 010%) 2/223 (1.0%; 03%) 0/69 (0%; 05%) Heifers (less than 24 months old) 1/75 (1.3%; 07%) 1/100 (1.0%; 05%)

Reactors/number tested. * P < 0.0001. Table 8 The prevalence of NSP responders in the two sheep farms V.C. Baladi sheep ocka
a

1/131 (0.7%) 69/69 (100%)

Infected sites; reactors/number tested.

The proportions of NSP responders among the adult dairy cow populations ranged from 0 to 13.6%, a range that seems to be a little greater than that among the young populations (Table 5), but the difference is not statistically signicant. The proportions of NSP responders among the young and the old feedlot calves are summarized in Table 6, for comparison. The differences between the proportions of NSP responders among the young and the old subpopulations of feedlot calves, and the correlation between the age of the population and the NSP-IR rates were both statistically highly signicant. The correlation between the age group and NSP-IR was relatively high, and the probability of exhibiting anti-NSP antibodies was 18.5 times higher (95% condence interval between 9.7 and 35.5) in an adult animal than in an animal aged 7 months or less. Among 80 out of 120 adult beef cattle tested (farm SN; Table 3) there were seven NSP reactors (8.8%). The distribution of NSP responders in the various feedlot farms is presented in Table 7. Table 8 shows that only one sheep from the control farm reacted against NSP, whereas all the 69 sheep in the infected ock reacted positive. 4. Discussion It was previously noted that FMD vaccination of young animals in endemic areas, where adult ruminants are period-

0/80 (0%; 05%) 12/410 (2.7%; 15%); 3/255 (1.1%; 03%)

NSP-IR of lactating cows adult of KV + E + G(a) and of EH is statistically signicant (P < 0.0001) (reactors/number tested). a Infected sites.

H. Yadin et al. / Vaccine 25 (2007) 82988305

8303

ically vaccinated, resulted in a high level of herd protection [13,14]. The same was noted in the outbreak described in the present paper: all the young local dairy calves remained refractory to clinical manifestations during the outbreak. The colostrum-derived antibodies protect calves up to 3 months of age, the age of the rst vaccination [6,7,1316], after which the rst vaccination imparts active immunity that replaces the passive lactogenic immunity, and the repeated vaccination regime has proved its solidity. Our present study of this FMD outbreak found the same trend: no clinical signs were observed among the local dairy cattle populations. Clinical signs were observed only among the imported calves, and their occurrence was age-related. The older animals above 7 months exhibited clinical signs, whereas the younger imported stock, which would be expected to be more likely to show clinical manifestations than the older animals, showed no clinical signs related to FMDV infection. Protection of the young imported stock and the loss of the immune protection among the older stock of imported calves have been reported in controlled studies [14,17], which found that repeat vaccinations at 4-month intervals were necessary to impart full clinical protection in endemic areas where such vaccinations are administered routinely. Indeed, in the presently described outbreak, more than 4 months had passed between the vaccination of the older imported animals and the onset of the epidemic, whereas among the younger imported stock this interval was 4 months or less. On the assumption that the antibody NSP response reects the immune status of a group of animals, it is possible to categorize the immune status of the group, under FMDV challenge in the eld, according to the prevalence of NSP responders in each group of animals. Not only did the younger imported stocks not show clinical signs, but also the proportion of NSP responders among these groups ranged between 22.2 and 38.1%, whereas among the older stock of animals, the average proportion of NSP responders reached 86% (Table 6). Moreover, mortality occurred only among calves belonging to the older stock of imported animals, whereas, in contrast, the proportion of NSP responders among the young dairy stock indicated that these animals did not support FMDV multiplication following eld-borne infection. According to the same principle, all the animals in the imported group, treated as an epidemiological unit, should be regarded as potential carriers of FMDV. One dose of vaccine confers clinical protection if it is given less than 4 months before the onset of an epidemic but, from the epidemiological point of view, these animals would pose a threat to the entire susceptible animal population if they were not revaccinated 4 months later, because their clinical protection could mask potential carriers. The prevalence of NSP responders among the clinically affected animals indicates that almost all the animals originating from the imported stocks harbored FMDV, even if for a very short time. Therefore, in Israel, to minimize the spread of FDMV from an infected site, quarantine is imposed for the entire period during which clinical manifestations are exhib-

ited, and for an additional 21 days after the disappearance of the last clinical manifestation. Only if no additional signs are observed in the quarantine is lifted, and movement of animals from the infected site is permitted, but only to the slaughterhouse. Bergman et al. [5] suggested, that the freedom from outbreaks in endemic areas for many years after suspension of vaccination indicated that the re-appearance of FMD was not triggered by carriers, but most probably resulted from commercial activities. This suggestion was given after a large-scale investigation that assessed various diagnostic tools for epidemiological surveillance in endemic areas in South America. This phenomenon resembles the situation that led to the January 2004 outbreak in Israel, which occurred after an FMD-free period of 3 years. It seemed to be trigged by an infected commercial lorry that visited an infected site in the West Bank, before returning back to his village where the rst infected foci was observed. The relatively high NSP prevalence among the adult beef cattle on the Golan Highs (SN) and in the dairy herd at the infected site (EH) can be partially explained by the fact that these two zones contain a dense population of wild gazelles, and also wild boars, especially around the SN herd. It is plausible that FMDV circulating undetected among the wild animal populations might be the trigger for the greater prevalence of NSP responders among these herds than among dairy herds situated in other areas. In fact, an outbreak of FMD occurred in a grazing beef herd in December 2005, conrming the possibility of FMDV circulation near the Syrian/Lebanese border area with Israel. Bergmann et al. [3] found that for 90900 days following the FMD outbreak, 3ABC positive and negative populations co-existed. In light of these revelations, investigators tried to classify these subpopulations according to their herddisease status, but the classication was based on an FMD-positive population, part of which comprised unvaccinated cattle that had been experimentally infected. In contrast, our present investigation addressed only vaccinated subpopulations kept in endemic premises, and found that only a few of the animals became naturally infected. Moreover, in our study the infected population could be easily characterized by age, number of vaccinations, and the time elapsed between the last vaccination and the time of exposure. In another study, also carried out by Bergmann et al. [4], the studied populations comprised huge numbers of animals from FMD-free areas that had been free of clinical signs of FMD under an intensive vaccination regime for several years. Only one herd of 400 animals (no age or other stratication details were presented, regarding the prole of the sick animals) that the authors claimed to have been repeatedly vaccinated contained 4% of diseased animals, following eld exposure. In the present study, a similar situation occurred in Israel, where repeatedly vaccinated animals have never showed clinical FMD manifestations. To control the last epidemic in the UK, stamping-out measures were taken as the ultimate means of eradication. This

8304

H. Yadin et al. / Vaccine 25 (2007) 82988305

measure angered the public and the media because of opposition to such mass killing of animals. Therefore, the European Community issued a new directive regarding FMD control (EU Council Directive 85/511/EEC on Community measures for the control of FMDE). The EU solution is essentially a version of the Israeli approach to withstanding epidemics after eliminating the infectious sites. The Israeli approach could be a reasonable alternative for the EU; it forms the basis of schemes that have been adopted in other FMD-endemic areas including, in particular, the endemic countries immediately outside the EU. Israeli practice is based on the establishment of local quarantine around the infective focus, restriction of animal movements, and the creation of protective zone by FMDV vaccination. This trend emerged from the hope that the newly developed commercial kits will enter common use and prove reliable. These kits are designed to use the NSP antibodies response to distinguish between resistant vaccinated (refractory) animals and those that are clinically healthy animals but not completely resistant. Currently, the detection of antibodies to NSPs is the preferred diagnostic method to distinguish among animals that are virus-infected, unvaccinated, or vaccinated. However, this method cannot be applied at the individual level. Moreover, the cutoff point for the distinction between FMDV-free herds and those containing carriers, on the basis of the prevalence of NSP responders has not yet been established [5]. Analysis of situations such as the 2004 outbreak in Israel might provide the clues that are necessary for the calculation of this important cutoff point. Epidemiologists are currently seeking models that might be accepted by the policy makers [18]. Thus, they should include in their theoretical (mathematical) models the fact that animals originating from FMDV-free countries, if they had been vaccinated subsequently to expose to the eld strains of FMDV, would probably remain asymptomatic carriers, and should, therefore, be banned from international or even national trade. Srensen et al. [12], in a controlled experimental study, vaccinated cattle that subsequently became carriers after challenge with homologous FMD. Bruderer et al. [9], in trying to evaluate ELISA-based assays for detecting a 3ABC response, found that anti-3ABC antibodies could be detected in vaccinated animals upon experimental challenge. This experimental model should mimic natural infection, but the eld challenge poses different and uncontrolled situations. Orsel et al. [16] determined the reproductive ratio (R0 ) of a group of vaccinated calves and of a group of unvaccinated calves under experimental challenge. Their ndings conrmed that in a group of vaccinated animals the probability of virus transmission was very low; they generated an SIR model without using the time as predictive factor. If we combine our present data with that nding of Orsel et al. [16], we can suggest that if less than 25% of the members of a group of repeatedly vaccinated animals without clinical sings that were vaccinated less than 4 months before challenge or

exposure are NSP responders, it can be considered that the group would not pose any peril to other animals. In contrast, a similar group of animals that had been vaccinated more than 4 months prior to FMDV exposure must be considered potentially hazardous, especially if any animal, even only one, was ill. The most dangerous groups of animals are those that receive only one dose of vaccine. Such a group, even if it presents no sick individual, should remain in quarantine pending destruction or controlled slaughtering.

Acknowledgement The EUFMD committee supported this work. References

[1] Breeze R. Agroterrorism: betting far more than the farm. Biosecurity Bioterrorism Strategy Pract Sci 2004;2:114. [2] Holliman A. Differential diagnosis of diseases causing oral lesions in cattle. Practice 2005;7:213. [3] Bergmann IE, Malirat V, Neitzert E, Peck E, Panizzutti N, S anchez C, et al. Improvement of a serological strategy for foot-and-mouth diseases virus surveillance in cattle under systematic vaccination: a combined system of an indirect ELISA-3ABC with an enzyme-linked immunoelectrotransfer blot assay. Arch Virol 2000;145:47389. [4] Bergmann IE, Neitzert E, Malirat V, Ortiz S, Colling A, S anchez C, et al. Rapid serological proling by enzyme-linked immunosorbent assay and its use as an epidemiological indicator of foot-and-mouth disease viral activity. Arch Virol 2003;148:891901. [5] Bergmann IE, Malirat V, Neitzert E. Diagnostic tools for epidemiological surveillance in South America. Appendix 75. In: Report of the Session of Research Group of the Standing Technical Committee of the European Commission for the Control of Foot-and-Mouth Disease, 75, Leylystad, the Netherlands, September 1983. Rome: Food and Agriculture Organization of the United Nations; 2004. p. 523 [Appendix]. [6] Pay TWF, Hingley PH, Radelett PJ, Black L, OReilly KJ. The correlation of 140S antigen dose with the serum neutralizing antibody response and with protection from challenge induced by FMD vaccines. In: Report of the Session of Research Group of the Standing Technical Committee of the European Commission for the Control of Foot-and-Mouth Disease, Leylystad, Netherlands, September 1983. Rome: Food and Agriculture Organization of the United Nations; 1983. p. 523 [Appendix IX]. [7] Sutmoller P, Viera A. The relationship of neutralizing antibody titers for foot-and-mouth disease virus and the protection of cattle. Boletin del Centro Panamerican de Fiebre Aftosa 1980;30/40:5762. [8] Van Maanen C. A complex-trapping-blocking (CTB) ELISA, using monoclonal antibodies and detecting specically antibodies directed against foot and mouth disease types A, O and C. 1. Method and characteristics. Vet Microbiol 1990;24:1718. [9] Bruderer U, Swam H, Haas B, Visser N, Brocchi E, Grazioli S, et al. Differentiating infection from vaccination in foot-and-mouth disease: evaluation of an ELISA based on recombinant 3ABC. Vet Microbiol 2004;101:18797. [10] Clavijo A, Wright P, Kitching P. Development in diagnostic techniques for differentiating infection from vaccination in foot-and-mouth disease. Vet J 2004;167:922. [11] McVicar JW, Sutmoller P. Foot-and-mouth disease: the agar gel immunodiffusion test for antibody to virus-infection-associated (VIA) antigen as a tool for epizootiologic survey. Am J Epidemiol 1970;92: 2738.

H. Yadin et al. / Vaccine 25 (2007) 82988305 [12] Srensen KJ, de Stricker K, Dyrting KC, Grazioli S, Hass B. Differentiation of foot-and-mouth disease virus infected animals from vaccinated animals using a blocking ELISA based on baculovirus expressed FMDV 3ABC antigen and a 3ABC monoclonal antibody. Arch Virol 2005;150:80514. [13] Riverson S, Sadir AM, Gaggino OP, Marcovecchio FE, Zabal O, Laporte O. Estudio comparativo en bovinos de dos vacunas antiaftosa: oleosa e hidroxido saponinada. Revista de Medicina Veterinaria 1982;63:36470. [14] Sadir AM, Schudel AA, Laporte O, Braun M, Margni A. Response to foot-and-mouth disease vaccines in newborn calves. Inuence of age, colostral antibodies and adjuvants. Epidemiol Inf 1988;100: 13544.

8305

[15] Kitching RP, Salt JS. The interference by maternally derived antibody with active immunization of farm animals against foot-and-mouth disease. Br Vet J 1995;15:37989. [16] Orsel K, Dekker A, Bouma A, Stegeman JA, de Jong MCM. Vaccination against foot and mouth disease reduces virus transmission in group of calves. Vaccine 2005;23:488797. [17] INTA, PIADC. Instituto National de Tecnologia AgropecuariaPlum Island Animal Disease Center. Foot-and-mouth disease: a vaccine study. Development in Biological Standardization 1977;35:12333. [18] Hamblin C, Kitching RP, Donaldson AI, Crowther JR, Barnett ITR. Enzyme-linked immunosorbent assay (ELISA) for the detection of antibodies against foot-and-mouth disease virus. 3. Evaluation of antibodies after infection and vaccination. Epidemiol Infect 1987;99:73344.