Abstract
Footnotes
†Present address: Parco Tecnologico Padano, Via Einstein, Polo Universitario, Lodi 26900, Italy.The transmissible spongiform encephalopathies (TSEs) are a group of fatal, neurodegenerative diseases that have been described in animals and humans. These diseases include CreutzfeldtJakob disease in humans, scrapie in sheep and goats, chronic wasting disease in deer and elk, and bovine spongiform encephalopathy (BSE) in cattle. The causative agent of TSEs remains unknown, but it is generally accepted that an abnormal form (PrPSc) of the cellular prion protein (PrPC) is the major component of the infectious prion agent (Prusiner, 1982). BSE was first observed as a disease in cattle in the UK in 1986 (Wells et al., 1987) and has subsequently affected more than 184 000 cattle. The introduction of a ruminant feed ban in 1988 was only partially effective. A second, more stringent feed ban was introduced on 1 August 1996, which prohibited the feeding of mammal-derived meat and bonemeal to all farmed livestock. Until the end of 2005, there had been 120 cases of BSE in the UK in cattle born after the more stringent feed ban. These are referred to as born-after-the-reinforced-ban (BARB) cases. BSE cases have arisen in cattle born as recently as May 2002, nearly 6 years after the introduction of the reinforced ban. Whilst the use of BSE-contaminated meat and bonemeal in animal feed has been recognized as the means by which the disease was disseminated and sustained as an epidemic within the cattle population (Wilesmith et al., 1992), the origin of the disease, including in the BARB cases, has remained unclear. It has been postulated that BSE was a strain of sheep scrapie that crossed the species barrier to cattle via animal feed (Wilesmith et al., 1988). Alternatively, it has been proposed that BSE was a pre-existing disease of cattle, albeit at very low levels, prior to the BSE epidemic or could have arisen as a novel TSE agent in the early 1970s, possibly due to a spontaneous mutation in the PrP gene of a single animal (BSE Inquiry, 1999), leading to the misfolding of the PrP protein. This hypothesis led to suggestions that BARB BSE cases could be direct descendants of such an animal (Ferguson-Smith, 2003) and, therefore, carry a common mutation. The BARB cases, consisting of cattle theoretically not exposed to the BSE agent, are also the most appropriate British cohort for identifying a rare, spontaneous PrP mutation event should it exist.
No variation in the coding region of the bovine PrP gene has been associated with BSE susceptibility in pre-BARB BSE cattle (Hunter et al., 1994; Sander et al., 2004); however, the (probably) substantially lowered prion levels of infectious agent in the environment associated with the post-feed ban period might increase the power to detect alleles that are strongly predisposing to disease. In this study, DNA sequences of the PrP coding regions or open reading frames (ORFs) of 100 BARB BSE cases from Great Britain were compared with those of 66 healthy controls. Statistical analysis was used to evaluate the possibility of an association between PrP ORF polymorphisms and susceptibility to BSE in BARB cases.
Blood, brainstem tissue (obex region) or 10 % brain (obex) homogenates (where no alternative samples were available) were obtained from 100 of the 120 BARB animals reported in Great Britain. For 66 BARB cases, blood samples were obtained from contemporary control animals, which were matched for farm, sex, breed and date of birth. The breed was nominated by the farmer, but breeds have been grouped together to form broader gene pools for analysis. The BARB BSE cases examined in this study were identified by various surveillance routes: clinical suspects (28 %), casualty or emergency slaughtered animals (38 %), fallen stock (12 %), cattle slaughtered under the over 30 month scheme (OTMS) (15 %) or slaughtered cohorts of BSE cattle (7 %). BARB cases were diagnosed as BSE-positive by at least two contemporary European Commission-approved diagnostic tests (i.e. histology, immunohistochemistry, ELISA or Western blot, depending on surveillance route and date of diagnosis).
DNA was extracted from samples and purified by three rounds of phenol/chloroform extraction, followed by chloroform extraction and ethanol precipitation (Sambrook et al., 1989). Several DNA preparations that were highly degraded or difficult to amplify were treated with an additional purification step (QIAamp micro DNA kit; Qiagen). Fifty nanograms of purified genomic DNA was used to amplify each of the two PCR products described in Fig. 1. Amplification products were sequenced in both directions by using the same primer pairs (Qiagen). Polymorphisms were identified by comparison to a PrP gene reference sequence from a Jersey cow (Hills et al., 2001; GenBank accession no. AJ298878[GenBank] ) using GAP4 software (Staden package, version 1.4.1).
|
In this study, four silent single nucleotide polymorphisms (SNPs), all published previously, were detected within the PrP gene ORF (summarized in Fig. 1 and Table 1). These corresponded to codon positions L23 (leucine, CTC→CTT), Q78 (glutamine, CAG→CAA), P113 (proline, CCC→CCT) and N192 (asparagine, AAC→AAT), where the polymorphic base is underlined and the most commonly occurring allele for each codon is shown preceding the rarer, minor allele. There was no evidence of departure of genotypic frequencies from HardyWeinberg equilibrium.
Table 1. PrP gene ORF polymorphisms by allele in BARB BSE and control cattle of different breeds Sequences are compared with the reference sequence (GenBank accession no. AJ298878; Hills et al., 2001). Holstein and Friesian breeds include Friesian, Holstein and crosses of these breeds. X, Cross-breed; B, BARB; C, control. Counts (n) refer to numbers of alleles (2x number of animals).
The frequencies of the variants observed in this study were broadly similar to those in the 96 healthy US cattle reported by Heaton et al. (2003), with minor allele frequencies of 2, 26, 14 and 10 % for L23, Q78, P113 and N192, respectively. The major difference lies in the higher frequency of the minor allele at P113 in the US animals (14 versus 0.9 % of all alleles in this study), which may be due to the predominance of beef breeds in the US study, compared with mainly dairy cattle breeds reported here. In both studies, the Q78 synonymous change displayed the greatest heterozygosity. Also shown in Table 1 is the distribution of polymorphisms among BARB BSE and control cattle breed groups. In the Holstein and Friesian gene pool, the most highly represented, the polymorphism at L23 was not observed.
Further coding variations detected in this study were located within the PrP octapeptide region (Goldmann et al., 1991). Genotypes of the N-terminal octapeptide repeats were identified as 6 : 6, 6 : 5, 5 : 5 and 6 : 7, where the numbers refer to the number of 24 bp repeats. The rare seven-octapeptide-repeat allele has been reported previously in healthy cattle (Naharro et al., 2003; Schläpfer et al., 1999), but not in BSE cases. The wild-type 6 : 6 genotype was most common in both BARB and control cases (89 and 85 %, respectively). The 6 : 5 genotype was seen in 9.0 % of BARB cases and 13.6 % of controls, with one BARB and one control animal being homozygous for five octapeptide repeats (5 : 5). As with the SNPs, there was no evidence of departure from HardyWeinberg equilibrium. In studies by Hunter et al. (1994) and Neibergs et al. (1994), the frequencies of octapeptide-repeat PrP genotypes corresponded very closely to our findings, although no 6 : 7 genotypes were found. Our findings were also consistent with prior sequence information suggesting that the polymorphic Q78 codon was associated with the six-octapeptide-repeat allele (Heaton et al., 2003).
To test for associations between PrP genotype and BARB BSE cases, a sign test was applied to examine differences in genotype (measured by the number of alleles) between two members of each farm pair. Pairs that shared the same genotype at a particular polymorphic site were excluded. This method has the benefit of using the pair structure to deal with variable, undefined and uncontrollable factors across the pairs (Curnow et al., 1997), including between-farm differences in allele frequency and exposure to BSE. However, no significant associations were observed. All polymorphisms were tested, but because of the combination of a small dataset (partly due to the fact that not all BARB cattle had a matched control) and the skewed allele frequencies at most of the loci, only the test of the Q78 site had reasonable power of detection. Differences in the numbers of alleles between all of the controls and all BARB cases for each of the polymorphic sites were also examined, disregarding the pair structure. This latter procedure allowed more data to be used, but was more open to bias. Regardless, no significant associations were observed in these tests. An alternative approach was examined in which the number of heterozygous observations among all four SNPs and the octapeptide repeat for each individual were compared within pairs. However, this comparison also showed no association between BARB BSE cases and the degree of heterozygosity. In summary, this study has found no relationship between the occurrence of BARB BSE cases and PrP gene polymorphisms. Other studies have also looked for, and failed to identify, an association between BSE and PrP gene ORF polymorphisms: neither Hunter et al. (1994) nor Sander et al. (2004) found associations with PrP octapeptide genotype or the N192 SNP, and the latter study also failed to find associations with Q78 or P113.
This study revealed cases of BSE among genotype 5 : 5 and 6 : 7 cattle. There was no statistical association between disease and the number of octapeptide repeats; however, the failure to see such cases previously, and possibly the lack of an association, could be a function of the low frequency of these genotypes. Recent studies in transgenic mice have indicated that an increase in the number of octapeptide repeats in bovine PrP can enhance susceptibility to BSE (Castilla et al., 2005). However, there were too few examples of rare genotypes in this study to test this hypothesis.
In Table 2, the surveillance routes were used to classify the BARBs provisionally into two groups according to clinical status, in order to examine the hypothesis that a subclinical form of BSE, more readily detectable through active surveillance, may be associated with a PrP gene polymorphism. One group contained animals that were clinical suspects, casualty or emergency slaughtered animals or fallen stock, based on the assumption that these animals were unhealthy, although it is not certain that this was due to clinical symptoms of BSE. The second group contained apparently healthy BARB cattle, slaughtered under the OTMS or as cohorts of BSE cattle. The distribution of the N192 allele differed significantly (P<0.05; Fishers exact test) between the clinical-status groupings, with the rarer N192 T allele being more frequent in the apparently healthy group (Table 2). In addition, all four L23 heterozygotes were in the unhealthy BARB group; however, the difference in allele frequencies between the two groups was not significant.
Table 2. Allelic frequencies in unhealthy BARB cattle versus apparently healthy BARB cattle and in isolated versus multiple or clustered BARB BSE cases Allele 1 is the most common allele, and allele 2 the minor allele, at each locus, as described in the text. n, No. cattle; OR, octapeptide repeat (six or five). The single seven-OR allele in the study occurred in the unhealthy and isolated BARB groups with a frequency of 0.01 in each group and is omitted from the table.
The comparison of N192 frequencies in the clinical-status group may be confounded by breed association (see Table 2). For example, two Highland BARB cases were apparently healthy and carried a total of three N192 T alleles, but the single Highland control also carried two copies of the N192 T allele at this locus. Further analysis was made by using only those BARB cases in the combined Holstein and Friesian (n=68 animals) or combined Simmental/Limousin (n=16) groups and, in these subsets, the association was not significant. Therefore, we conclude that the current data show no evidence for an association between PrP genotype and clinical status.
Further stratification of the BARB cases was undertaken into those originating from farms with multiple cases or associated with the south-west Wales cluster (n=18 with sequence data), making the assumption that contaminated feed was the most probable cause of the multiple cases (Anonymous, 2005a, b) and that another cause, such as a PrP mutation, might be responsible for the remaining isolated BARB cases (n=82). Again, in terms of PrP ORF sequence, the distribution of variants was similar between the two groups, with a few minor and statistically non-significant exceptions (Table 2). Only the rare, silent L23 polymorphism was associated with BARB and not control animals in this study and was distributed similarly with respect to clinical status or geographical occurrence (Table 2); however, this finding was not statistically significant.
This study did not identify an association between the PrP gene coding region and an increase in susceptibility of BARB cattle to BSE through environmental exposure to the BSE agent. Furthermore, the results do not support the suggestion that a particular mutation in the PrP ORF led to an inherited or a spontaneous form of BSE in this group of animals. It should be considered that increased susceptibility could also be caused by polymorphisms in the regulatory regions of the PrP gene or by other, as-yet-unidentified genes elsewhere in the genome (Hernandez-Sanchez et al., 2002; Zhang et al., 2004). If BARB BSE cases continue to arise, it will be important to identify the origin of the disease and route(s) of transmission among affected cattle in order to contribute to the eradication of BSE.
References
Anonymous (2005b). Final report on BSE confirmed in an animal born on 3 October 2001 & two cohort animals born on 28 September 2001 and 1 May 2002 in a Pembrokeshire herd.
BSE Inquiry (1999). The BSE Inquiry: the report.
Castilla, J., Gutierrez-Adan, A., Brun, A., Pintado, B., Salguero, F. J., Parra, B., Segundo, F. D., Ramirez, M. A., Rabano, A. & other authors (2005). Transgenic mice expressing bovine PrP with a four extra repeat octapeptide insert mutation show a spontaneous, non-transmissible, neurodegenerative disease and an expedited course of BSE infection. FEBS Lett 579, 62376246.[CrossRef][Medline]
Curnow, R. N., Hodge, A. & Wilesmith, J. W. (1997). Analysis of the bovine spongiform encephalopathy maternal cohort study: the discordant case-control pairs. Appl Stat 46, 345350.
Ferguson-Smith, M. A. (2003). Continuing anxiety about BSE. Vet Rec 153, 723[Medline]
Goldmann, W., Hunter, N., Martin, T., Dawson, M. & Hope, J. (1991). Different forms of the bovine PrP gene have five or six copies of a short, G-C rich element within the protein-coding region. J Gen Virol 72, 201204.
Heaton, M. P., Leymaster, K. A., Freking, B. A., Hawk, D. A., Smith, T. P., Keele, J. W., Snelling, W. M., Fox, J. M., Chitko-McKown, C. G. & Laegreid, W. W. (2003). Prion gene sequence variation within diverse groups of US sheep, beef cattle, and deer. Mamm Genome 14, 765777.[CrossRef][Medline]
Hernandez-Sanchez, J., Waddington, D., Wiener, P., Haley, C. S. & Williams, J. L. (2002). Genome-wide search for markers associated with bovine spongiform encephalopathy. Mamm Genome 13, 164168.[CrossRef][Medline]
Hills, D., Comincini, S., Schlaepfer, J., Dolf, G., Ferretti, L. & Williams, J. L. (2001). Complete genomic sequence of the bovine prion gene (Prnp) and polymorphism in its promoter region. Anim Genet 32, 231233.[CrossRef][Medline]
Hunter, N., Goldmann, W., Smith, G. & Hope, J. (1994). Frequencies of PrP gene variants in healthy cattle and cattle with BSE in Scotland. Vet Rec 135, 400403.[Abstract]
Naharro, G., Yugueros, J., Temprano, A., del Rio, M. L., Rodriguez-Ferri, E. F. & Luengo, J. M. (2003). Prion protein gene polymorphisms in a population of Spanish cows. Vet Rec 152, 212213.
Neibergs, H. L., Ryan, A. M., Womack, J. E., Spooner, R. L. & Williams, J. L. (1994). Polymorphism analysis of the prion gene in BSE-affected and unaffected cattle. Anim Genet 25, 313317.[Medline]
Prusiner, S. B. (1982). Novel proteinaceous particles cause scrapie. Science 216, 136144.
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Sander, P., Hamann, H., Pfeiffer, I., Wemheuer, W., Brenig, B., Groschup, M. H., Ziegler, U., Distl, O. & Leeb, T. (2004). Analysis of sequence variability of the bovine prion protein gene (PRNP) in German cattle breeds. Neurogenetics 5, 1925.[CrossRef][Medline]
Schläpfer, I., Saitbekova, N., Gaillard, C. & Dolf, G. (1999). A new allelic variant in the bovine prion protein gene (PRNP) coding region. Anim Genet 30, 386405.[CrossRef][Medline]
Wells, G. A. H., Scott, A. C., Johnson, C. T., Gunning, R. F., Hancock, R. D., Jeffrey, M., Dawson, M. & Bradley, R. (1987). A novel progressive spongiform encephalopathy in cattle. Vet Rec 121, 419420.[Medline]
Wilesmith, J. W., Wells, G. A. H., Cranwell, M. P. & Ryan, J. B. M. (1988). Bovine spongiform encephalopathy: epidemiological studies. Vet Rec 123, 638644.[Abstract]
Wilesmith, J. W., Ryan, J. B. & Hueston, W. D. (1992). Bovine spongiform encephalopathy: case-control studies of calf feeding practices and meat and bonemeal inclusion in proprietary concentrates. Res Vet Sci 52, 325331.[Medline]
Zhang, C., de Koning, D. J., Hernandez-Sanchez, J., Haley, C. S., Williams, J. L. & Wiener, P. (2004). Mapping of multiple quantitative trait loci affecting bovine spongiform encephalopathy. Genetics 167, 18631872.
Received 29 August 2006; accepted 22 December 2006.