Host Infectio

Paratuberculosis has historically been consid­ered an infection of ruminants. First indica­tions of a wider host range date to the 19 70s, but it was not until the late 1990s that non­ruminant wildlife hosts of paratuberculosis were confirmed and further investigated in relation to their significance in the epidemi­ology of livestock infection.

15.1.1 Known host range

Mycobacterium avium subsp. paratuberculosis (MAP) has been identified in a broad range of non-ruminant wildlife species. Suspected isola­tion of MAP was first reported from a European brown hare (Lepus europaeus) in England in 19 77 (Matthews and Sargent, 1977), although the organism responsible was not fully charac­terized. Lesions attributed to paratuberculosis were subsequently described in a wild rabbit (Oryctolagus cuniculus) from Scotland (Angus, 1990), and presence of MAP has since been confirmed in rabbits (Greig et al., 199 7, 1999; Beard et al., 2001a) by a polymerase chain re­action (PCR) assay based on the species-specific IS900 insertion sequence (Vary et al., 1990). Following detection of MAP in rabbits, studies were extended to investigate other wildlife spe­cies associated with infected farms in the same region. MAP was isolated from foxes (Vulpes vulpes) and stoats (Mustela erminea) (Beard et al.,

‘Corresponding author: Naomi.Fox@sruc.ac.uk

© CAB International 2020.

(eds M.A. Behr et al.)

1999), then subsequently from weasels (Mustela nivalis), badgers (Meles meles), wood mice (Apodemus sylvaticus), rats (Rattus norvegicus), brown hares (Lepus europaeus), jackdaws (Corvus monedula), rooks (Corvus frugilegus) and crows (Corvus corone) (Beard et al., 2001a). Following these studies researchers have looked for MAP in non-ruminant species around the world, and MAP has been isolated from a diverse range of non-ruminant species, from marsupials in New Zealand (Nugent et al., 2011) to birds in the USA (Corn et al., 2005). MAP has also been detected in the intestinal tracts of a number of inverte­brates including cockroaches (Fischer et al., 2003a), earthworms (Fischer et al., 2003b) and a number of diptera species (Fischer etal., 2001). Table 15.1 contains a complete list of the known non-r uminant wildlife species in which MAP has been detected. However, a species should not be considered a competent reservoir host based solely on isolation of the bacteria.

The MAP strains that infect non-ruminants are principally of the ‘Cattle-type’ (MAP-C). However, ‘Sheep-type’ strains (MAP-S) are oc­casionally isolated from wildlife, with reports of MAP-S being isolated from the house mouse (Mus musculus) (Florou et al., 2008), western grey kangaroo (Macropus fuliginosus) and tam- mar wallabies (Macropus eugenii decres) (Cleland et al., 2010). More detailed descriptions of the different types of MAP are given elsewhere (see Chapter 6, this volume).

15.1.2 Pathology

In wild ruminants, gross lesions and clinical signs have been reported as similar to those in infected cattle and sheep, where the disease is ultimately fatal (Williams et al., 19 79; Buergelt et al., 2000). In contrast, macroscopic lesions are rare in non-ruminant wildlife. To date, re­ports of clinical cases of paratuberculosis in non-ruminant species are limited.

The histopathology of non-ruminant MAP infections has been studied most in naturally in­fected rabbits, in which both severe and mild le­sions have been observed in the intestines (Greig et al., 1997; Beard et al., 2001b).

Disease-induced mortality has been ob­served in primates, with clinical and pathologi­cal features similar to those in ruminant hosts (McClure et al., 1987; Zwick et al., 2002). In addition to high mortality rates, high preva­lence was reported within an infected colony of stumptail macaques (Macaca arctoides) (76%, n = 38), with up to 10⁸ colony-forming units (CFU)Zg intestinal tissue in clinical hosts, and high shedding (up to 2 ? 10⁶ CFUZg faeces) in non-clinical hosts (McClure et al., 1987).

The pathology of MAP infections in other non-ruminants is much more subtle. In foxes, weasels and stoats, small numbers of single, large macrophage-like cells or granulomata consisting of 10 or fewer cells have been ob­served in the mesenteric lymph nodes and mucosa-associated lymphoid tissue of the gut (Beard et al., 2001a). Only small numbers of AFOs have been detected in the cytoplasm of these macrophage-like cells. No histopathologi­cal lesions have been observed in the intestines or liver of these host species. Similar mild histo­pathological lesions have also been observed in a crow and a wood mouse. Beard et al. (2001b) observed cells containing fewer than five AFOs scattered throughout the lamina propria of a crow intestine. Multiple granulomata were ob­served in the liver but did not contain AFOs.

15.1.3 Prevalence and excretion rates

The contribution made by a species to the amount of MAP in the environment is a function

Continued

of both the numbers of infected animals (preva­lence ? population size) and the rates of MAP excretion. In the UK, prevalences of MAP are generally higher in carnivores such as the fox compared with prey species such as lagomorphs and rodents.

High prevalences have been found in rab­bits (26%, n = 113), possums (25%, n = 73), and hedgehogs (36% n = 42) in New Zealand, and it has been postulated that these species could contribute to the persistence of MAP at the wildlife-livestock interface (Nugent et al., 2011). High prevalence rates and levels of infection in non-r uminant wildlife as found in the UK and New Zealand have not been reported in Europe or the USA. The prevalence of MAP in wildlife in the USA ranged from 1.7­25%, although the sample sizes for those with a prevalence of 25% were small (n = 4) (Corn et al., 2005). Most species with a prevalence of greater than 10% were predators or scavengers (including armadillo, feral cat, opossum and raccoon) (Corn et al., 2005), which follows the patterns of prevalence found in the UK. Faecal samples were culture-positive from raccoons, armadillos, an opossum and a feral cat, sug­gesting that these animals shed the bacteria in their faeces (Corn et al., 2005) and therefore have the potential to play a part in the onward transmission of the disease.

Published data on MAP shedding rates in non-ruminant wildlife are limited, and primarily restricted to lagomorph hosts. The mean number of CFU from infected rabbit faeces was 7.6 ? 10⁵ ± 5.2 ? 10⁵ CFU/g (Daniels et al., 2003a), which is lower than the 108 CFU/g reported in faeces from clinically affected cattle (Cranwell, 1997; Whittington et al., 2000). Infected rabbits may also shed MAP in their urine (n = 2/17), although the levels of shedding are thought to be far lower than those in faeces (Daniels et al., 2003a). However, as rabbits are asymptomatic, the excretion rates in rabbit faeces is a mean across different levels of in­fection and as such cannot be compared with the clinically affected cattle excretion rates that are often cited in the literature. Although not directly quantified, pathological comparisons suggest that the shedding rates of other non-ruminant wildlife would be expected to be far lower than for rab­bits. A relative estimate of the input of MAP on to pasture suggested that sheep and cattle poten­tially contributed 4 and 125 times more organ­isms per hectare per day, respectively, than rabbits. None the less, rabbits were estimated to contribute >10⁶ CFU of MAP per hectare per day (Daniels et al., 2003a).

15.2