Search

Inference on data with missingness can be challenging, particularly if the knowledge that a measurement was unobserved provides information about its distribution. Our work is motivated by the Commit to Quit II study, a smoking cessation trial that measured smoking status and weight change as weekly outcomes. It is expected that dropout in this study was informative and that patients with missed measurements are more likely to be smoking, even after conditioning on their observed smoking and weight history. We jointly model the categorical smoking status and continuous weight change outcomes by assuming normal latent variables for cessation and by extending the usual pattern mixture model to the bivariate case. The model includes a novel approach to sharing information across patterns through a Bayesian shrinkage framework to improve estimation stability for sparsely observed patterns. To accommodate the presumed informativeness of the missing data in a parsimonious manner, we model the unidentified components of the model under a nonfuture dependence assumption and specify departures from missing at random through sensitivity parameters, whose distributions are elicited from a subject-matter expert. Supplementary materials for this article are available online.

Inference on data with missingness can be challenging, particularly if the knowledge that a measurement was unobserved provides information about its distribution. Our work is motivated by the Commit to Quit II study, a smoking cessation trial that measured smoking status and weight change as weekly outcomes. It is expected that dropout in this study was informative and that patients with missed measurements are more likely to be smoking, even after conditioning on their observed smoking and weight history. We jointly model the categorical smoking status and continuous weight change outcomes by assuming normal latent variables for cessation and by extending the usual pattern mixture model to the bivariate case. The model includes a novel approach to sharing information across patterns through a Bayesian shrinkage framework to improve estimation stability for sparsely observed patterns. To accommodate the presumed informativeness of the missing data in a parsimonious manner, we model the unidentified components of the model under a non-future dependence assumption and specify departures from missing at random through sensitivity parameters, whose distributions are elicited from a subject-matter expert.

The nearly complete, PCR-amplified, 16S rRNA gene sequences have been determined from the representative type strains of eight xanthomonad phena, including six validly described species of the genus Xanthomonas and Stenotrophomonas maltophilia. Pairwise sequence comparisons and phylogenetic analysis demonstrated that the xanthomonads comprise a monophyletic lineage-within the gamma-subclass of the Proteobacteria. Although the genus Xanthomonas was observed to comprise a cluster of very closely related species, the observed species-specific primary sequence differences were confirmed through sequencing additional strains belonging to the respective species.

Livestock feed is susceptible to contamination from wildlife excreta during on farm storage. Pathogens associated with diseases such as paratuberculosis, salmonella and cryptosporidiosis are present in wild rodent and bird excreta. Feed stores on four farms in the east of Scotland were monitored monthly over the winter of 1998/9 to quantify the levels of wildlife faecal contamination. A mean of 79·9 rodent (95% confidence interval: 37·5–165·9) and 24·9 (14·3–41·7) bird faeces were deposited per m2 of stored feed per month. It was estimated that individual cattle and sheep could encounter 1626 and 814 wildlife faeces over the winter.A model based on the numbers of infected faeces consumed per annum was used to estimate ‘infectious probabilities’ (Pinf) required to account for the reported prevalence of paratuberculosis, salmonella and cryptosporidiosis in sheep and cattle in the east of Scotland in 1998. Based on empirical data for input variables [the number of faeces encountered (Fe), the number ingested (Fi) and the prevalence of infection in wildlife species (Ip)], Pinf estimates ranged from 1·6×10−8 for cryptosporidiosis in sheep to 8·2×10−6 for paratuberculosis in cattle. The model suggested that ingestion of feed contaminated by wildlife faeces could account for the prevalence of all three diseases. Wildlife faecal contamination of stored feed should be given serious consideration as a potential source of infection to livestock.

Understanding the factors that influence the extent of mammalian browsing of young trees can facilitate the development of management strategies that permit successful woodland establishment. The effect of tree density, tree species, relative mountain hare ( Lepus timidus L.) abundance and ground vegetation height on the extent and severity of browsing damage by mountain hares on Scots pine ( Pinus sylvestris L.) and silver birch ( Betula pendula L.) saplings was investigated between May 1998 and May 2000. Browsing damage by mountain hares was recorded annually on 300 trees ( Utilisation of the study sites by mountain hares was highest in spring (May) and the young woodland at each site was utilised less than the adjacent control areas of moorland. A mean of 5.8% of trees were browsed annually per site, although the percentage of trees browsed at each site differed significantly between years. Tree density, relative hare abundance, tree species and ground vegetation height did not significantly influence the percentage of trees browsed. The extent of recorded browsing suggests that the impact of mountain hares on the establishment of Scots pine or birch woodlands in the uplands of Scotland would be minimal where hares occur at the moderate densities such as those indicated by pellet-counts in this study.

Paratuberculosis (Johne's disease) was long considered only a disease of ruminants. Recently non-ruminant wildlife species have been shown to harbor Mycobacterium avium subsp. paratuberculosis, the causative organism of paratuberculosis. We review the known non-ruminant wildlife host range of M. avium subsp. paratuberculosis and consider their role in the epidemiology of paratuberculosis in domestic ruminant livestock. Mycobacterium avium subsp. paratuberculosis has been isolated from lagomorph, canid, mustelid, corvid, and murid species. In agricultural environments domestic ruminants may contact wildlife and/or their excreta when grazing or feeding on farm-stored feed contaminated with wildlife feces, opening up the possibility of inter-species transmission. Of the wildlife species known to harbor M. avium subsp. paratuberculosis in Scotland, the rabbit is likely to pose the greatest risk to grazing livestock. Paratuberculosis in domestic ruminants is a notoriously difficult disease to control; the participation of non-ruminant wildlife in the epidemiology of the disease may partially account for this difficulty.

On farms where rats and mice are present, unprotected livestock feed may become contaminated with rodent faeces, thereby creating a possible source of infection for cattle and sheep. Livestock unable to avoid contaminated feed may choose to eat it entirely, reject it completely or attempt to reject faeces selectively while consuming some of the feed. Two experiments were conducted to investigate which of these three responses were demonstrated by livestock. Ten cattle and ten sheep were presented individually with three repeats of ten feed treatments. Treatments were based on two feed types (meal and pelleted compound), with three levels of contamination (none, 'low' and 'high'), from one of two rodent species (rat and mouse). Avoidance behaviour was greater for feed contaminated with rat faeces compared to feed contaminated with mouse faeces. At low levels of rat contamination there was evidence that livestock actively rejected faeces whilst consuming feed. At higher levels of contamination animals rejected faeces and feed. Livestock could not actively discriminate against mouse faeces and thus rejection of feed was used to avoid faeces ingestion. Despite rejection of contaminated feed and some discrimination against faeces, significant numbers of rodent faeces were ingested illustrating that livestock feeding behaviour cannot prevent ingestion of rodent faeces. Feed contaminated with rodent faeces therefore poses a significant risk of disease infection to livestock.

Transmission of Mycobacterium avium subspecies paratuberculosis , the organism responsible for paratuberculosis (or Johne’s disease) in ruminants, occurs through the faecal–oral route. As M. a. paratuberculosis has been isolated from rabbit faeces, cattle grazing rabbit faecal contaminated pasture may thus be at risk. A herd of 57 beef cattle was monitored on a farm in Perthshire, throughout the 1999 ‘grazing year’, to investigate whether the cattle avoided rabbit faecal contaminated pasture and thus the potential for disease transmission. Grazing was measured every two days over eight rotations by sward heights on 40 marked treatment plots (0.5 m × 0.5 m) to which 0, 10, 50 and 250 rabbit faecal pellets were added. Cattle were also monitored by an active transponder system which enabled individual animals contacting two plots per field rotation (one control and one contaminated) to be recorded. During the monitored grazing year, grazing pressure was low with a net mean sward offtake of 18% of sward height per rotation. There were no significant differences between rabbit faecal treatments (0, 10, 50 and 250 pellets) with respect to the height or proportion of sward removed, or between the numbers of contacts made by cattle on contaminated and uncontaminated plots. Over 90% of all the cattle contacted contaminated plots, indicating that the potential for disease transmission was widespread among the herd. To our knowledge, this is the first reported instance of a lack of avoidance by grazing cattle towards swards contaminated with faeces, and implies that the potential for transmission of paratuberculosis from rabbit contaminated pasture is high.

Recent reports of natural paratuberculosis (or Johne's disease) in rabbits, foxes, and stoats has focused debate on the presence and importance of wildlife reservoirs in the epidemiology of this disease. This paper describes an extensive study investigating 18 nonruminant wildlife species for evidence of paratuberculosis. Using both culture and histopathological analysis, fox, stoat, weasel, crow, rook, jackdaw, rat, wood mouse, hare, and badger were found to harbor Mycobacterium avium subsp. paratuberculosis , the causative organism of paratuberculosis, suggesting that the epidemiology of this disease is more complex than previously realized.

This paper describes a genetic analysis of wild-living cats in Scotland. Samples from 230 wild-living Scottish cats (including 13 museum skins) and 74 house cats from England and Scotland were surveyed for nine microsatellite loci. Pelage characteristics of the wild-living cats were recorded, and the cats were then grouped into five separate categories depending on the degree to which they conformed to the characteristics attributed to Felis silvestris Schreber, 1775. Allele frequency differences between the morphological groups are greater than those among the three house cat samples. Analysis of genetic distances suggests that more of the differences between individuals can be explained by pelage than geographical proximity, and that pelage and geographical location are not confounded. Ordination of the genetic distances suggests two main groups of wild-living cats, with intermediates, and one group is genetically very similar to the house cats, while the other group contains all cats taxonomically identified as wildcat based on morphology. A genetic mixture analysis gives similar results to the ordination, but also suggests that the genotypes of a substantial number of cats in the wildcat group are drawn from a gene pool with genotypes in approximately equilibrium proportions. We argue that this is evidence that these cats do not have very recent domestic ancestry. However, from the morphological data it is highly likely that this gene pool also contains a contribution from earlier introgression of domestic cat genes.

Xylella fastidiosa, a pathogen of citrus, is the first plant pathogenic bacterium for which the complete genome sequence has been published. Inspection of the sequence reveals high relatedness to many genes of other pathogens, notably Xanthomonas campestris. Based on this, we suggest that Xylella possesses certain easily testable properties that contribute to pathogenicity. We also present some general considerations for deriving information on pathogenicity from bacterial genomics.

Xylella fastidiosa, a pathogen of citrus, is the first plant pathogenic bacterium for which the complete genome sequence has been published. Inspection of the sequence reveals high relatedness to many genes of other pathogens, notablyXanthomonas campestris. Based on this, we suggest thatXylellapossesses certain easily testable properties that contribute to pathogenicity. We also present some general considerations for deriving information on pathogenicity from bacterial genomics.

The indigenous wildcat, Felis silvestris Schreber, 1775, and the introduced domestic cat, F. catus L., have been sympatric in Britain for more than 2000 years. As a result of interbreeding, any distinction between these two forms has become obscured, although a range of morphological criteria (pelage patterns, body measurements, gut lengths, skull morphometrics) and genetic techniques (immunological distances, electrophoresis, DNA hybridization) have been used previously to distinguish between them. A sample of 333 wild-living cats in Scotland was assessed for coloration and markings of pelage, standard body measurements and weights, and (for carcasses only) limb bone lengths, intestine lengths, and skull measurements. These cats were also classified as wildcat, hybrid, or domestic cat according to traditional pelage criteria. Multivariate analyses on these variables, for adult cats, failed to show any clearly distinct groups. When each of the variables was analysed separately, only the distribution of limb bone and intestine length measurements suggested the possibility that two groups might exist. Group 1 cats had short intestines and long limb bones. Group 2 consisted of cats with long intestines and short limb bones. Although the characteristics defining cats in Group 1 were similar to those traditionally associated with wildcats, they exhibited a much broader range of pelage and coloration than traditionally described. The groups exhibited a degree of geographical separation. The distribution of Group 1 cats was found to be related to certain environmental variables, namely mean annual temperature and land with poor potential for forestry and agriculture, suggesting that there may be a biological basis for the separation. The implications of these results on the identification and taxonomy of the wildcat are significant. The concept of the wildcat and the domestic cat as separate species can be challenged. The paper highlights the complexity and difficulties for conventional taxonomy when used as a means for distinguishing between a wild type and its domesticated form where there is interbreeding.

A panel of monoclonal antibodies (MAC204, MAC236, MAC265) which recognise extracellular matrix glycoproteins implicated in plant-microbe interactions has been used to study glycoprotein antigens in petioles of turnip (Brassica campestris L.). While MAC204 recognised two glycoproteins (gp120 and gp45) with apparent M(r) 120,000 and 45,000 in petiole extracts made with 2-amino-2-(hydroxymethyl)-1,3-propanediol (Tris) buffer containing sodium dodecyl sulfate, MAC236 recognised gp120 but not gp45, and MAC265 gave no or only weak reactivity. Tissue dissection studies established that gp120 was predominantly associated with the vascular bundle whereas gp45 was largely associated with the pith. This was consistent with results from tissue prints probed with MAC204 and MAC236 which also suggested a vascular localisation for gp120. Immunoelectronmicroscopy showed that MAC204 and MAC236 both labelled three-way junctions between cells of the phloem and sclerid fibres. Both gp120 and gp45 were shown to carry epitopes in common with known hydroxyproline-rich glycoproteins. Unlike gp45, gp120 could be extracted from petioles with Tris buffer alone and then isolated from this extract by trichloroacetic acid treatment (which left gp120 soluble), followed by size-exclusion and ion-exchange chromatography. Amino acid analysis revealed gp120 to be a novel glycoprotein, particularly rich in proline, lysine, valine and threonine but relatively poor in hydroxyproline. The most abundant sugars were arabinose and galactose. The potential role of this very basic cell surface glycoprotein in plant defence against microbes is discussed.

In order to make a missing at random (MAR) or ignorability assumption realistic, auxiliary covariates are often required. However, the auxiliary covariates are not desired in the model for inference. Typical multiple imputation approaches do not assume that the imputation model marginalizes to the inference model. This has been termed "uncongenial" [Meng (1994, Statistical Science 9, 538-558)]. In order to make the two models congenial (or compatible), we would rather not assume a parametric model for the marginal distribution of the auxiliary covariates, but we typically do not have enough data to estimate the joint distribution well non-parametrically. In addition, when the imputation model uses a non-linear link function (e.g., the logistic link for a binary response), the marginalization over the auxiliary covariates to derive the inference model typically results in a difficult to interpret form for the effect of covariates. In this article, we propose a fully Bayesian approach to ensure that the models are compatible for incomplete longitudinal data by embedding an interpretable inference model within an imputation model and that also addresses the two complications described above. We evaluate the approach via simulations and implement it on a recent clinical trial.

The membranes of plant and animal cells contain aquaporins, proteins that facilitate the transport of water. In plants, aquaporins are found in the vacuolar membrane (tonoplast) and the plasma membrane. Many aquaporins are mercury sensitive, and in AQP1, a mercury-sensitive cysteine residue (Cys-189) is present adjacent to a conserved Asn-Pro-Ala motif. Here, we report the molecular analysis of a new Arabidopsis aquaporin, delta-TIP (for tonoplast intrinsic protein), and show that it is located in the tonoplast. The water channel activity of delta-TIP is sensitive to mercury. However, the mercury-sensitive cysteine residue found in mammalian aquaporins is not present in delta-TIP, or in gamma-TIP, a previously characterized mercury-sensitive tonoplast aquaporin. Site-directed mutagenesis was used to identify the mercury-sensitive site in these two aquaporins as Cys-116 and Cys-118 for delta-TIP and gamma-TIP, respectively. These mutations are at a conserved position in a presumed membrane-spanning domain not previously known to have a role in aquaporin mercury sensitivity. Comparing the tissue expression patterns of delta-TIP with gamma-TIP and alpha-TIP showed that the TIPs are differentially expressed.

Plant cells contain proteins that are members of the major intrinsic protein (MIP) family, an ancient family of membrane channel proteins characterized by six membrane-spanning domains and two asparagine-proline-alanine (NPA) amino acid motifs in the two halves of the protein. We recently demonstrated that [gamma]-TIP, one of the MIP homologs found in the vacuolar membrane of plant cells, is an aquaporin or water channel protein (C. Maurel, J. Reizer, J.I. Schroeder, M.J. Chrispeels [1993] EMBO J 12: 2241–2247). RD28, another MIP homolog in Arabidopsis thaliana, was first identified as being encoded by a turgor-responsive transcript. To find out if RD28 is a water channel protein, rd28 cRNA was injected into Xenopus laevis oocytes. Expression of RD28 caused a 10- to 15-fold increase in the osmotic water permeability of the oocytes, indicating that the protein creates water channels in the plasma membrane of the oocytes and is an aquaporin just like its homolog [gamma]-TIP. Although RD28 has several cysteine residues, its activity is not inhibited by mercury, and in this respect it differs from [gamma]-TIP and all but one of the mammalian water channels that have been described. Introduction of a cysteine residue next to the second conserved NPA motif creates a mercury-sensitive water channel, suggesting that this conserved loop is critical to the activity of the protein. Antibodies directed at the C terminus of RD28 were used in combination with a two-phase partitioning method to demonstrate that RD28 is located in the plasma membrane. The protein is present in leaves and roots of well-watered plants, suggesting that its presence in plants does not require a specific desiccation regime. These results demonstrate that plant cells contain constitutively expressed aquaporins in their plasma membranes (RD28), as well as in their tonoplasts ([gamma]-TIP).

Sows subjected to prenatal stress have been found to produce offspring that have altered responses to stress. Our objective was to determine if exposing a sow to stress would alter the response of the offspring to lipopolysaccharide (LPS) at 2 mo of age or their response to mixing stress at 4 mo of age. Sow treatments consisted of intravenous injections of ACTH (1 IU/kg of BW), exposure to rough handling for a 10-min duration (rough), or no treatment (control) once per week from d 42 to 77 of gestation. At 2 mo of age, pigs from each treatment, 1 per litter (n = 21, 17, and 15 for the ACTH, rough, and control treatments, respectively), were challenged with 2 μg of LPS/kg of BW or saline, or served as a noninjected control. Their behavioral response to a human approach test and salivary cortisol were measured. At 4 mo of age, 1 pig from each treatment (n = 14, 14, and 15 for the ACTH, rough, and control treatments, respectively) was taken from its home pen and placed in a pen of unfamiliar pigs. At this time, a punch biopsy wound (6 × 6 mm) was created to measure the ability of the pig to heal the wound. At this same time, each pig received a 1-mL intramuscular injection of 20% ovine red blood cells (oRBC), and then a second injection of oRBC at 21 d postmixing. Blood samples were collected 3 times per week for 2 wk and then once a week for 4 more weeks. Blood samples were analyzed for cortisol, porcine corticosteroid-binding globulin, antibody response to oRBC, and nitric oxide production by macrophages. Behavior was recorded during the first 5 d after mixing. All pigs in the LPS challenge responded with characteristic sickness behavior; however, pigs in the rough treatment showed less sickness behavior than those in the other 2 treatments (P0.05). Maternal stress treatment did not affect (P0.43) salivary cortisol. Pigs from all treatments responded similarly to mixing stress with regard to cortisol, porcine corticosteroid-binding globulin, antibody titers, nitric oxide production, and hematology measures, and all pigs experienced the same amount of aggression in response to mixing. Without altering peripheral measures of stress responsivity, prenatal stress enhanced the ability of pigs to cope with a simulated immune challenge, which could prove to be an adaptation to challenging environments.

Molecular genetic approaches were adopted in the model crucifer, Arabidopsis thaliana, to unravel components of RPP5- and RPP1-mediated disease resistance to the oomycete pathogen, Peronospora parasitica. The products of RPP5 and three genes comprising the RPP1 complex locus belong to a major subclass of nucleotide-binding/leucine-rich repeat (NB-LRR) resistance (R) protein that has amino-terminal homology to the cytoplasmic domains of Drosophila and mammalian Toll and interleukin-1 family receptors (the so called 'TIR' domain). Similarities in the domain architecture of these proteins and animal regulators of programmed cell death have also been observed. Mutational screens revealed a number of genes that are required for RPP5-conditioned resistance. Among these are EDS1 and PAD4. Both EDS1 and PAD4 precede the function of salicylic acid-mediated plant responses. The EDS1 and PAD4 genes were cloned and found to encode proteins with similarity to the catalytic site of eukaryotic lipases, suggesting that they may function by hydrolysing a lipid-based substrate.

A spiroplasma isolated from citrus with little-leaf disease was grown in a cell-free medium and injected into leafhoppers (Euscelis plebejus). Injected leafhoppers, but not those fed on infected plants, transmitted the spiroplasma to white clover (Trifolium repens cv. S100) and sweet orange (Citrus sinensis cv. Valencia). Infected clover plants were severely stunted; infected sweet orange plants showed typical symptoms of citrus little-leaf disease. The spiroplasma was detected in clover and sweet orange plants by electron microscopy; the helical morphology of the organisms was most easily recognizable in sections 150-200 nm thick. The organism was re-isolated in cell-free media both from infected plants and from injected E. plebejus. The original isolate and those re-isolated from experimentally infected clover and sweet orange appeared by morphological, cultural, biochemical and serological criteria to be identical to each other and to the R8-A2 (type) and C-189 strains of Spiroplasma citri. Serological tests and electrophoretic analysis of protein preparations indicated no relationship to Acholeplasma laidlawii, although this organism survived for at least 10 wk after injection into E. plebejus. Our results show that the causal agent of little-leaf disease is related to S. citri.

Exposing a pregnant sow to stress has been shown to affect the resulting offspring. Our objective was to determine if rough handling of pregnant sows altered the physiology of her offspring and if these alterations were different from an experimentally induced model of prenatal stress. Sow treatments consisted of i.v. injections of ACTH (1 IU/kg of BW), exposure to rough handling for 10 min (Rough), or no treatment (Control) once a week during d 42 to 77 of gestation. To determine the plasma cortisol response to treatments, blood (5 mL) was collected from 30 sows after treatment administration. To conduct the prenatal stress study, a separate group of 56 sows was used in 1 of 4 replicates. At birth, production data were collected for each litter, including birth weight, number born, anogenital distance, and pig viability. At weaning, pigs were blocked by BW and sex, and placed in a nursery pen of 6 pigs, with 2 pigs from each treatment group. To assess the effect of treatments on cortisol, corticosteroid-binding globulin (CBG), and hematological cell profiles, blood was collected every other day for 10 d after weaning. Application of treatments caused plasma cortisol concentrations to be greatest in ACTH sows compared with Control sows (P0.001), with Rough sows having intermediate values (P = 0.07). Treatments did not affect the number of pigs born, number of stillborn, or pig viability (P0.40). The ratio of cortisol to CBG did not differ between treatments (P = 0.09). Hematological variables did not differ between treatments (P0.19). Pigs born to ACTH sows had a smaller anogenital distance compared with controls (P0.03), with pigs from Rough sows being intermediate. Our data indicate that swine exposed to prenatal stress (ACTH injection) can have alterations in sexual morphology without effects on growth or the immune cell populations measured in this study.

Monitoring the health of wildlife populations is important for understanding and controlling the risk of infections to livestock, humans and/or other wildlife. In this paper, we analyse the results of surveys of parasites and non-specific signs of diseases carried out on organs from 638 red and 107 sika deer culled in four regions of Scotland between 1991 and 1997. Infections of the lung by Elaphostrongylus spp. were significantly greater in red than sika deer. Older animals were more heavily infected with Elaphostrongylus spp. and Sarcocystis spp., and infections with Sarcocystis spp. tended to be heavier in more recent years. The results suggest that a combination of key indicator parasite species and non-specific signs of disease may be useful for monitoring the health of wildlife populations at a national scale. However, they also demonstrate that such monitoring needs to be long-term, carried out according to standard protocols and at an appropriate resolution to enable integration with data on other potentially influential environmental factors.

Livestock feed is susceptible to contamination from wildlife excreta during on farm storage. Pathogens associated with diseases such as paratuberculosis, salmonella and cryptosporidiosis are present in wild rodent and bird excreta. Feed stores on four farms in the east of Scotland were monitored monthly over the winter of 1998/9 to quantify the levels of wildlife faecal contamination. A mean of 79.9 rodent (95% confidence interval: 37.5-165.9) and 24.9 (14.3-41.7) bird faeces were deposited per m2 of stored feed per month. It was estimated that individual cattle and sheep could encounter 1626 and 814 wildlife faeces over the winter. A model based on the numbers of infected faeces consumed per annum was used to estimate 'infectious probabilities' (Pinf) required to account for the reported prevalence of paratuberculosis, salmonella and cryptosporidiosis in sheep and cattle in the east of Scotland in 1998. Based on empirical data for input variables [the number of faeces encountered (Fe), the number ingested (Fi) and the prevalence of infection in wildlife species (Ip)], Pinf estimates ranged from 1.6 x 10(-8) for cryptosporidiosis in sheep to 8.2 x 10(-8) for paratuberculosis in cattle. The model suggested that ingestion of feed contaminated by wildlife faeces could account for the prevalence of all three diseases. Wildlife faecal contamination of stored feed should be given serious consideration as a potential source of infection to livestock.

The reported incidence of Johne's disease has been increasing in the east of Scotland since 1993. A postal questionnaire survey was sent to 127 farms to identify potential risk factors for Johne's disease in relation to wildlife and farm management practices, and 86 returns were obtained. Of 22 farms which had been assumed to be free of the disease, on the basis of information held by local veterinary centres, seven (32 per cent) reported cases of Johne's disease in the 1990s, indicating that the disease is under-reported. Logistic regression analyses showed that eight of 63 potentially explanatory variables were significant at the 5 per cent level in affecting the likelihood of farms reporting Johne's disease. Of these, large numbers of livestock and rabbits, and access of wildlife to feed stores were the clearest and most consistent risk factors associated with the disease. The application of manure to grazing pasture, the type of water supply for the cattle and the numbers of crows were also related to the presence of Johne's disease but the nature of these relationships was less clear. Only 38 per cent of the farms reported taking any control measures to combat Johne's disease, but three of the control measures were relevant to risk factors identified as significant by the survey, namely maintaining a clean water supply, controlling rabbits and not spreading manure on to grazing pasture.

Natural paratuberculosis infection of rabbits ( Oryctolagus cuniculus ) was recently diagnosed in Scotland, and an investigation into the pathology of the disease in wild rabbits is reported in this paper. Evidence of Mycobacterium avium subsp. paratuberculosis ( M.a. paratuberculosis ) infection was detected in 22% of 110 rabbits; the organism was cultured from 17 of 110 rabbits, and histopathological lesions consistent with M.a. paratuberculosis infection were noted in 18 of 98 rabbits examined. No macroscopical lesions suggestive of M.a. paratuberculosis infection were observed. The histopathological lesions were either severe or mild. Severe lesions consisted of extensive macrophage granulomata and numerous giant cells, with many intracellular acid-fast bacteria in the small intestine. For the examination of formalin-fixed, paraffin wax-embedded tissues, neither immunohistochemistry nor the polymerase chain reaction was as sensitive a method of diagnosis as histopathology.

The synthesis of extracellular enzymes and extracellular polysaccharide (EPS) in Xanthomonas campestris pv. campestris (Xcc) is regulated by a cluster of genes called rpf (for regulation of pathogenicity factors). Two of the genes, rpfF and rpfB, have previously been implicated in the synthesis of a diffusible regulatory molecule, DSF. Here, we describe a screen of transposon insertion mutants of Xcc that identified two DSF-overproducing strains. In each mutant, the gene disrupted is rpfC, which encodes a hybrid two-component regulatory protein in which the sensor and regulator domains are fused and which contains an additional C-terminal phosphorelay (HPt) domain. We show that rpfC is in an operon with rpfH and rpfG. The predicted protein RpfG has a regulatory input domain attached to a specialized version of an HD domain, previously suggested to function in signal transduction. The predicted protein RpfH is structurally related to the sensory input domain of RpfC. We show that RpfC and RpfG act positively to regulate the synthesis of extracellular enzymes and EPS, but that RpfC acts negatively to regulate the synthesis of DSF. We propose that RpfGHC is a signal transduction system that couples the synthesis of pathogenicity factors to sensing of environmental signals that may include DSF itself.

Ultraviolet radiation (UVR) causes systemic immune suppression, decreasing the delayed type and contact hypersensitivity responses in animals and humans and enhancing certain mycobacterial, parasitic and viral infections in mice. This study tests the hypothesis that prior exposure to UVR enhances influenza infections in mice. BALB/c female mice were exposed to 0-8.2 kJ/m2 of UVR. Exposed and unexposed mice were infected intranasally three days later with 150-300 plaque-forming units/mouse (lethal dose (LD)20-LD40) of mouse-adapted Hong Kong Influenza A/68 (H3N2) virus or sham infected with 50 microL Hanks' balanced salt solution/mouse. Mortality from viral infection ranged from 25-50%. UVR exposure increased virus-associated mortality in a dose-dependent manner (up to a two-fold increase at 8.2 kJ/m2). The increased mortality was not associated with bacterial pneumonia. The highest dose of UVR also accelerated the body weight loss and increased the severity and incidence of thymic atrophy associated with influenza infection. However, UVR treatment had little effect on the increase in lung wet weight seen with viral infection, and, to our surprise, did not cause an increase in virus titers in the lung or dissemination of virus. The mice died 5-6 days after infection, too early for adaptive immune responses to have much impact. Also, UVR did not interfere with the development of protective immunity to influenza, as measured by reinfection with a lethal challenge of virus. Also, cells adoptively transferred from UVR or untreated mice were equally protective of recipient mice challenged with a lethal dose of virus. The mice resemble mice succumbing to endotoxin, and influenza infection increased the levels of tumor necrosis factor alpha (TNF-alpha) in bronchoalveolar lavage fluid and serum cortisol levels; however, UVR preexposure did not increase either of these responses to the virus. The results show that UVR increased the morbidity, mortality and pathogenesis of influenza virus in mice without affecting protective immunity to the virus, as measured by resistance to reinfection. The mechanism of enhanced mortality is uncertain, but the data raises concerns that UVR may exacerbate early responses that contribute to the pathogenesis of a primary viral infection.

In Arabidopsis ecotype Landsberg erecta (Ler), RPP5 confers resistance to the pathogen Peronospora parasitica. RPP5 is part of a clustered multigene family encoding nucleotide binding-leucine-rich repeat (LRR) proteins. We compared 95 kb of DNA sequence carrying the Ler RPP5 haplotype with the corresponding 90 kb of Arabidopsis ecotype Columbia (Col-0). Relative to the remainder of the genome, the Ler and Col-0 RPP5 haplotypes exhibit remarkable intraspecific polymorphism. The RPP5 gene family probably evolved by extensive recombination between LRRs from an RPP5-like progenitor that carried only eight LRRs. Most members have variable LRR configurations and encode different numbers of LRRs. Although many members carry retroelement insertions or frameshift mutations, codon usage analysis suggests that regions of the genes have been subject to purifying or diversifying selection, indicating that these genes were, or are, functional. The RPP5 haplotypes thus carry dynamic gene clusters with the potential to adapt rapidly to novel pathogen variants by gene duplication and modification of recognition capacity. We propose that the extremely high level of polymorphism at this complex resistance locus is maintained by frequency-dependent selection.

Few data are available on the prevalence of feline viruses in European wildcats (Felis silvestris). Previous surveys have indicated that wildcats may be infected with the common viruses of domestic cats, apart from feline immunodeficiency virus (FIV). In the present study, 50 wildcats trapped throughout Scotland (UK) between August 1992 and January 1997 were tested for evidence of viral infection. All were negative for FIV by several serological or virological methods. By contrast, 10% of the cats were positive for feline leukemia virus (FeLV) antigen and infectious virus was isolated from 13% of a smaller subset. Of the wildcats tested for respiratory viruses, 25% yielded feline calicivirus (FCV) and although no feline herpesvirus was isolated, 16% of the samples had neutralizing antibodies to this virus. Antibodies to feline coronavirus (FCoV) were found in 6% of samples. Feline foamy virus (FFV) was an incidental finding in 33% of samples tested. This study confirms that wildcats in Scotland are commonly infected with the major viruses of the domestic cat, except for FIV.

In mammalian metallothionein Zn2+ is exclusively coordinated to Cys-thiolate to form clusters in which the metal is thermodynamically stable but also kinetically labile. By contrast, little is known about coordination to prokaryotic metallothionein, SmtA. 3 nmol of Zn2+ nmol-1 SmtA were displaced by 8 nmol of p-(hydroxymercuri)phenylsulfonate implicating eight of the nine Cys in the coordination of three metal ions. None of the Zn2+ associated with SmtA was accessible to 4-(2-pyridylazo)resorcinol prior to the addition of p-(hydroxymercuri)phenylsulfonate. An unusual feature of SmtA is the presence of three His residues, and we have investigated whether these contribute to metal coordination. Less Zn2+ was associated with purified SmtA(H40R/H49R/H55R), in which all three His residues were substituted with Arg, and approximately one equivalent of Zn2+ was immediately accessible to 4-(2-pyridylazo)resorcinol. Following incubation of SmtA with 111Cd, three 111Cd resonances were detected, two in a range expected for CdS4 and the third indicative of either CdNS3 or CdN2S2 coordination. Two-dimensional TOCSY 1H NMR and 111Cd-edited 1H NMR showed two His residues bound to 111Cd, confirming CdN2S2 coordination. The pH of half-dissociation of Zn2+ increased from 4.05 for SmtA to 5.37 for SmtA(H40R/H49R/H55R). Equivalent values for single His mutants SmtA(H40R), SmtA(H49R), and SmtA(H55R) were 4.62, 4.48, and 3.81, respectively, revealing that conversion of His40 or His49 to Arg impairs Zn2+ binding at the CdN2S2 and CdS4 sites. Only approximately two equivalents of Zn2+ were associated with purified SmtA(H49R). The appearance of a fourth 111Cd resonance at lower pH suggests that an alternative CdN2S2 site also exists.

Antimicrobial saponins are found in many plant species and it has been suggested that these compounds may contribute to the protection of plants against pathogen attack. The ability of a number of phytopathogenic fungi to produce saponin-detoxifying enzymes suggests that detoxification of these compounds may have widespread significance in determining the outcome of plant-fungus interactions. The contribution of saponins to plant defence is being investigated by studies of mechanisms of saponin resistance in pathogens, and also by the isolation of plant mutants which lack saponins.

The analysis of variations is an important area of interest in health services and outcomes research and has two main goals: to identify and quantify variability across units, such as geographic regions or health care providers, in terms of procedure utilization and outcomes, and to explore the links between process, such as regional or hospital practice patterns, and outcomes, such as patient mortality and functional status. Hierarchical regression models are well suited for this type of analysis. In this paper we formulate a hierarchical polytomous regression model and apply it to the analysis of variations in the utilization of alternative cardiac procedures in a national cohort of elderly Medicare patients who had an acute myocardial infarction during 1987. The model is designed to accommodate clustered multinomial data with covariate vectors available on individual cases and on clusters. We present a Bayesian approach to fitting and checking the model using simulated values from the posterior distribution of the parameters. The simulation algorithms are based on Gibbs sampling in combination with Metropolis steps. Using the hierarchical polytomous regression model, we examine how the rates of cardiac procedures depend on patient-level characteristics, including age, gender and race, and whether there exist interstate differences and regional patterns in the use of these procedures.

We describe a meta-analysis approach for the evaluation of a potential surrogate marker. Surrogate markers are useful in helping to identify therapeutic mechanisms of action and disease pathogenesis, and for selecting therapies to take forward from phase II to phase III clinical trials. They have also become increasingly important for regulatory purposes by providing a basis for preliminary approval of drugs pending clinical outcome studies. Methodology for evaluating surrogate markers has focused on determining the difference in the effects of two treatments on clinical outcome in an individual clinical trial, and then estimating the proportion of this difference explained by the treatment's effects on the potential marker. Studies are, however, frequently underpowered or cease before they accumulate sufficient evidence to draw strong conclusions about the value of a potential surrogate marker using this approach, and there are also some technical difficulties with the approach. Consideration of the association between the difference in treatment effects on the clinical outcome and the difference in treatment effects on the potential marker over a range of trials provides an alternative means to evaluate a potential marker. We describe a meta-analysis approach using Bayesian methods to model this association. Importantly, this approach enables one to obtain prediction intervals for the true difference in clinical outcome for a given estimated treatment difference in the effect on the potential marker. We illustrate the methodology by applying it to results from studies of the AIDS Clinical Trials Group to assess the value of CD4 T-lymphocyte cell count as a potential surrogate marker for the treatment effects on the development of AIDS or death.

A panel of monoclonal antibodies that recognize plant extracellular matrix glycoproteins previously implicated in plant-microbe interactions was used to study the effects of pathogen inoculation and wounding on glycoproteins in petioles of Brassica campestris. The panel of monoclonals comprised two sets: JIM11, JIM12, and JIM20 recognize epitopes carried on hydroxyproline-rich glycoproteins (HRGPs) (M. Smallwood, A. Beven, N. Donovan, S. J. Neill, J. Peart, K. Roberts, and J. P. Knox, Plant J. 5:237–246, 1994); MAC204 and MAC265 recognize glycoproteins of the Rhizobium infection thread (K. A. VandenBosch, D. J. Bradley, S. Perotto, G. W. Butcher, and N. J. Brewin, EMBO J. 8:335–342, 1989). Wounding or inoculation of petioles with avirulent strains of pathovars of Xanthomonas campestris induced the synthesis of two new groups of antigens: gp160 ran as a smear on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) with apparent molecular mass from 120 to 200 kDa and was recognized by JIM20 and MAC204; gpS remained in the stacking gel on SDS-PAGE and was recognized by JIM11, JIM20, and MAC204. The response to virulent strains of pathovars of X. campestris was either less pronounced or absent. gpS comprised several components that were resolved by cation-exchange chromatography. Some of these components were characterized as extensin-like HRGPs. The level of induction of the gpS group of antigens by virulent strains was not altered by mutation of a number of genes required for basic pathogenicity or by heat-killing the bacteria.

Publisher Summary This chapter discusses aquaporins and water transport across the tonoplast. The membranes of living cells are differentially permeable to water and solutes: water passes more easily through a membrane than solutes, whose passage depends on the presence of specific channels and carriers. It was assumed for a long time that osmotic water movement occurs strictly by diffusion of water molecules across the lipid bilayer. When it was noticed that certain membranes permitted a water flux that was greater than expected, the existence of water–filled pores was proposed as an explanation. The nature of these membrane pores has long been a mystery, and it was postulated that water could pass either through proteinaceous pores (water channel protcins) or through locally disorganized regions of the phospholipid bilayer. These water channel proteins are referred to as aquaporins. Demonstration of aquaporin–mediated water permeability was first accomplished by the injection of Xenopus oocytes with aquaporin cRNA, followed by measurements of the osmotic permeability of the oocytes. Proof that a membrane protein has arrived at the plasma membrane of the oocyte can be obtained by immunofluorescence microscopy if antibodies to the protein are available, or by fractionating the oocyte membranes. New methods are now being developed to assay the activity of aquaporins by expressing them in yeast or in E. coli.

Xanthomonas campestris pv. campestris, the causal agent of black-rot disease of cruciferous plants, and an important industrial microbe, was able to express the Escherichia coli beta-glucuronidase reporter gene (uidA) when fused to the E. coli lactose operon promoter on a wide-host-range plasmid vector. The gene fusion is expressed constitutively at high levels in both complex and defined media using a wide range of carbon sources, and is not repressible by glucose or inducible by the gratuitous lac inducer isopropyl beta-D-thiogalactoside. An X. campestris campestris strain with a lesion in the clp (catabolite-repressor-like protein) locus, and containing the placluidA fusion, was tested for beta-glucuronidase activity. We found that the expression of the placluidA fusion gene is dependent on the presence of catabolite-repressor-like protein, with an approximately 75% reduction of expression in the clp -deficient mutant.

The plasma membrane and the vacuolar membrane (tonoplast) contain water-selective channel proteins called aquaporins. These 27 kDa proteins function as monomers and facilitate osmotic pressure-driven water transport through the membranes. The presence of aquaporins in a membrane increases its water permeability 5 to 10-fold.

There is strong evidence that corticosteroids contribute to the objective and subjective response rate observed following treatment with several cytotoxic chemotherapy agents, and that there is a dose response effect for treatment of multiple myeloma with alkylating agents. Therefore, the Eastern Cooperative Oncology Group (ECOG) studied cyclophosphamide 600 mg/M2 given for four consecutive days intravenously combined with prednisone 100 mg orally daily in 57 patients who had progressed following or failed to respond to standard doses of these drugs. Forty eight patients met the eligibility criteria for evaluation of response and toxicity. Fourteen patients (29%) had an objective response (OR) and an additional 2 (4%) had a subjective response (SR) only. The median duration of objective response was 3.1 months and estimated median survival was 8.6 months. These results are identical to our prior experience with high dose cyclophosphamide alone. The addition of prednisone does not appear to enhance results either through increased remissions or greater survival. Therefore, this study indicates that the preferred form of high dose cyclophosphamide for multiple myeloma is as a single agent given in intravenous four day courses.

Xanthomonas campestris pathovar campestris (hereafter X.c. campestris), the causal agent of black rot of crucifers, is a phytopathogenic bacterium of great economic importance (WILLIAMS 1980; ONSANDO 1992; SCHAAD and ALVAREZ 1993). Molecular genetic methods have now allowed a number of genes encoding pathogenicity determinants of X.c. campestris and related pathovars to be identified. In this chapter we will briefly review some of these and discuss our current understanding of the regulation of pathogenicity in this bacterium.

The title of this chapter is not wholly accurate. Our primary purpose is not to describe Xanthomonas genetics as a topic per se but rather to illustrate the use of genetic approaches to analyse aspects of Xanthomonas biology. Most of the studies reported so far have been concerned either with phytopathogenicity or with the production of polysaccharide. One of the most striking features of modern science is the impact which genetics (including molecular genetics) has had upon the development of almost all aspects of biology. In the 1950s and 1960s classical bacterial genetics consisted mainly of isolation and mapping of mutations (i.e. genetic manipulation in vivo) and was applied with great success to investigating the properties of so-called ‘metabolic black boxes’, pathways or processes about which little was known. Frequently the genetics suggested models which could be tested biochemically. Most of the genetics of this era was carried out with Escherichia coli and its relatives and a small number of other bacteria (e.g. Bacillus subtilis, Streptomyces coelicolor) (Hayes, 1968). The main limitation of working with other bacteria was the need for methods of transferring DNA between strains, a problem which persists to this day.

The three-dimensional structure of spinach ferredoxin-NADP+ reductase (NADP+, nicotinamide adenine dinucleotide phosphate) has been determined by x-ray diffraction at 2.6 angstroms (A) resolution and initially refined to an R factor of 0.226 at 2.2 A resolution. The model includes the flavin-adenine dinucleotide (FAD) prosthetic group and the protein chain from residue 19 through the carboxyl terminus at residue 314 and is composed of two domains. The FAD binding domain (residues 19 to 161) has an antiparallel beta barrel core and a single alpha helix for binding the pyrophosphate of FAD. The NADP binding domain (residues 162 to 314) has a central five-strand parallel beta sheet and six surrounding helices. Binding of the competitive inhibitor 2'-phospho-AMP (AMP, adenosine monophosphate) places the NADP binding site at the carboxyl-terminal edge of the sheet in a manner similar to the nucleotide binding of the dehydrogenase family. The structures reveal the key residues that function in cofactor binding and the catalytic center. With these key residues as a guide, conclusive evidence is presented that the ferredoxin reductase structure is a prototype for the nicotinamide dinucleotide and FAD binding domains of the enzymes NADPH-cytochrome P450 reductase, NADPH-sulfite reductase, NADH-cytochrome b5 reductase, and NADH-nitrate reductase. Thus this structure provides a structural framework for the NADH- or NADPH-dependent flavoenzyme parts of five distinct enzymes involved in photosynthesis, in the assimilation of inorganic nitrogen and sulfur, in fatty-acid oxidation, in the reduction of methemoglobin, and in the metabolism of many pesticides, drugs, and carcinogens.

A DNA fragment from Xanthomonas campestris pv. campestris that partially restored the carbohydrate fermentation pattern of a cya crp Escherichia coli strain was cloned and expressed in E. coli. The nucleotide sequence of this fragment revealed the presence of a 700-base-pair open reading frame that coded for a protein highly similar to the catabolite activation factor (CAP) of E. coli (accordingly named CLP for CAP-like protein). An X. campestris pv. campestris clp mutant was constructed by reverse genetics. This strain was not affected in the utilization of various carbon sources but had strongly reduced pathogenicity. Production of xanthan gum, pigment, and extracellular enzymes was either increased or decreased, suggesting that CLP plays a role in the regulation of phytopathogenicity.