Specific Applications of Culture

18.11.1 Faecal samples

Decontamination of samples

The first attempts to cultivate MAP from clini­cal samples utilized the methods that were available at the time for tuberculosis.

Faecal samples present a particular problem due to their high load of enteric bacteria, some of which break through decontamination rou­tines and multiply in culture media. Common contaminants from bovine faeces in Chile

Table 18.4. Common methods for cultivation of Mycobacterium avium subsp. paratuberculosis (MAP) from faeces.

Medium and Notes and

Method Decontamination inoculum Incubation references^a

North American methods

(Whittington, 2009)

Continued

^aThe first reference is the earliest mention of the original method; VAN, vancomycin 100 μg∕ml, amphotericin B 50 μg∕ ml, naladixic acid 100 μg∕ml; BHI, brain heart infusion.

OA, oxalic acid; NaOH, sodium hydroxide; HPC, hexadecylpyridinium chloride; BAC, benzalkonium chloride.

were Paenibacillus sp., Enterobacteriaceae and Pseudomonas aeruginosa (Steuer et al., 2015). Most protocols require removal of the particu­late matter with which many microbes associ­ate. This is achieved by suspending faeces in water or decontamination solution followed by filtration through cheesecloth (Cameron, 1956), light centrifugation (Merkal et al., 1968) or more commonly by sedimentation. The filtrate or supernatant is removed, and if it is already a decontamination solution it can be inoculated directly into culture media. In contrast, a water suspension is added first to a decontamination solution and the sediment from this is inoculated (Merkal et al., 1968).

Concerns about the failure to inactivate contaminants that were not in a vegetative growth phase led to incubation of the faecal suspension in antimicrobial solutions made up in nutrient media, in the hope that spores would germinate and then be killed. This approach has been used for preparation of milk samples and is commonly applied to faecal samples where HPC and antibiotics are constituted in half-strength brain heart infusion (BHI) broth (Shin, 1989; Whitlock and Rosenberger, 1990; Ellingson et al., 2005).

Culture protocols have evolved indepen­dently in different laboratories over the years (Table 18.4). In some countries in Europe, the NaOH-OA decontamination protocol is used in conjunction with LJ medium, while in the US and most other countries, sedimentation methods with HPC are combined with HEYM or liquid media. In a small study, 20 AFB smear-positive bovine faecal and tissue sam­ples were processed using either NaOH-OA or HPC and inoculated on to LJ media with result­ing sensitivity of 70 and 85%, respectively; all samples were positive by culture in modified BACTEC 12B medium after HPC decontamina­tion (Collins et al., 1990b). In the absence of data from a larger study it is unclear whether the decontamination protocol or medium can be interchanged without loss of sensitivity or an increase in the contamination rate. An attempt was made to compare the two most popular types of solid media - HEYM and LJ- in a large study in Denmark but using only the NaOH-OA decontamination proto­col. Ironically the HEYM medium appeared to have slightly greater sensitivity (Nielsen et al., 2004).

Decontamination of other types of samples is discussed below under each sample type.

Concentration of MAP from the faecal sample

To increase the analytical sensitivity of faecal culture, MAP may be concentrated by a variety of methods during or after decontamination of the sample. This was first tried by Merkal, who allowed coarse particles to settle then removed and centrifuged the supernatant of the faecal water suspension; the pellet was then decon­taminated (Merkal et al., 1964). Others modi­fied this by centrifuging the decontamination solution (Kim et al., 1989). Another modifica­tion combined both approaches by centrifuging all of the water supernatant, resuspending its pellet in decontamination solution and after incubation, centrifuging the entire solution (Stabel, 199 7). In a similar approach using three decontaminants, centrifugation was un­dertaken after the first two steps (Kalis et al., 2000).

Pooled faecal culture

The pooling of faecal samples from more than one animal for culture is a logical way to reduce the cost of detection of MAP at herd or flock level. Initial attempts were not encouraging due to loss of sensitivity associated with the dilution effect of pooling (Vialard et al., 1993). However, pooled culture was shown to be highly sensitive in sheep provided that an animal with multi- bacillary disease was present, as these cases shed about 10⁸ MAP/gram of faeces (Whittington et al., 2000b). Pooling rates of greater than 50 sheep per culture were clearly possible with­out losing sensitivity. For practical applica­tion a pooling rate of 50 was selected and was shown to provide higher flock-level sensitivity than serological examination of the same ani­mals (Whittington et al., 2000a; Sergeant et al., 2002). By 2002 in Australia this test together with abattoir surveillance for gross lesions of paratuberculosis had replaced other screening tests. Research in cattle showed similar benefits, but a lower pooling rate was required, gener­ally five (Kalis et al., 2000; Wells et al., 2002, 2003; Eamens et al., 2007b, 2008). Both com­puter modelling and empiric results showed that pooled faecal culture is practical and cost effec­tive for cattle (van Schaik et al., 2003; Kalis et al., 2004). The same approach is applicable in goats and probably most other species (Eamens et al., 2007a). Pooled faecal culture provides aggre­gate results but the data can be used to estimate the prevalence of infection within a population provided that some pools yield negative results.

There are several published methods for calcu­lating within-herd prevalence from pooled sam­ples (Toribio and Sergeant, 2007). Apart from ensuring the thorough mixing of the pooled sample prior to selecting an aliquot to culture, there are no additional technical considerations for the laboratory. The individual samples that were used to create a pool can be stored if it is desired to determine later which individual ani­mals contributed to the positive pooled culture result.

18.11.2 Tissue samples

MAP can be isolated readily from the intesti­nal tissues of infected animals and has some­times been cultured from the tissues of humans (Chiodini, 1989) (see Chapter 3, this volume). Less effort has been devoted to the development of protocols for culture of tissues than there has been for faeces, probably because contamination rates have not caused concern (Taylor, 1950; Smith, 1953). Tissues are usually collected in bulk at post-mortem, but small surgical biopsy samples from animals and humans can also be cultured (Kirkwood et al., 2009; Dennis et al., 2011; Begg et al., 2018). Methods described for tissues are simpler than those for faeces, usu­ally involving only one decontamination step. Incubation of homogenized tissues overnight or for up to 3 days in HPC is sufficient to remove most contaminants from the intestinal wall or lymph nodes (Cousins et al., 1995; Whittington et al., 1999). Tissues may be prepared using a variety of methods. Usually fat is trimmed away, and the remaining intestine or lymph node is finely divided with scissors, ground, blended mechanically or disrupted using a stomacher machine in solutions that may include proteases to disrupt tissue structure (Merkal et al., 1964; Merkal, 19 73; Sweeney et al., 1992; Aduriz et al., 1995). The homogenate is decontaminat­ed by suspension in a solution of antimicrobials, and the sediment is used as the inoculum. As for faeces, centrifugation of the suspension may be used to concentrate MAP and improve ana­lytical sensitivity, but this tends to increase the contamination rate (Reddacliff et al., 2003b). An alternative approach involved the use of the Zwitterionic detergent CB-18 to release MAP, combined with a cocktail of lytic enzymes to de­stroy contaminants; this was tried on intestinal tissues from cattle and bison but has not been compared independently with traditional meth­ods (Thornton et al., 2002).

18.11.3 Meat

MAP has been cultured from red meat in sev­eral countries using solid (HEYM, LJ) and liquid (BACTEC, MGIT and Trek) media. Alonso-Hearn et al. (2009) homogenized 2 g of diaphragm muscle and decontaminated it in 0.75% HPC before culture on HEYM and LJ slopes using a protocol for faecal culture. Mutharia et al. (2010) obtained colonies of MAP on HEYM after inoculation of stomacher fluid from a 5-g meat sample, without any decontamination. Others used 3-g samples of ground beef in an HPC de­contamination routine and obtained isolates of MAP in Trek liquid media from two of 140 sam­ples (Savi et al., 2015).

To overcome potentially very low concen­trations of the organism in muscle, an acid enzy­matic digestion method was developed to enable processing of large samples (20 g) (Reddacliff et al., 2010). This method was cumbersome, and a modification of a routine faecal culture proto­col (centrifugation at 3000 ? g for 20 min instead of 900 ? g for 30 min, decontamination in HPC for 24 h instead of 72h, inoculum volume 0.3 m instead of 0.1 ml) using a 2-g sample, although less sensitive, was more practical, and had a simi­lar rate of isolation of MAP from field samples in BACTEC medium due to there being fewer con­taminated samples (Reddacliff et al., 2010).

Processed meats are another potential source of MAP for humans. Akineden et al. (2011) evaluated methods for raw sausage using spiked samples and recommended decontamina­tion in HPC followed by culture on HEYM, but they did not evaluate liquid media.

18.11.4 Milk and processed dairy products

The shedding of MAP into the milk of infected cows was recognized many years ago (Alexejeff- Goloff, 1935; Taylor et al., 1981), while its isolation from the milk of two Crohn's disease patients has fuelled the debate about its role in disease in humans (Naser et al., 2000). There has been a great deal of research on ways to detect the organism in samples of milk from individual cows, in bulk tank milk and in pas­teurized milk, including retail samples (Millar et al., 1996; Grant et al., 2002). Pragmatically, reliable culture from milk was vital to study the effects of pasteurization on the viability of MAP (Stabel et al., 199 7). Most culture protocols in­volve centrifugation of milk at about 2500 ? g for 15 min to produce a pellet containing MAP; to increase the chance of detection, the cream layer - where MAP also partitions - may also be collected and pooled with the pellet for culture (Gao et al., 2005). Sample volumes of 40-50 ml of milk are necessary as the efficiency of re­covery is decontaminated for 15 min using N-acetyl-L-cysteine (NALC) with 1.5% NaOH; recovery of viable MAP was greater and con­tamination rates were lower compared with de­contamination with 0.75% HPC for 5 h (Bradner et al., 2013a). These recommendations were val­idated by the same authors in subsequent work using natural milk samples from seven dairies (Bradner et al., 2014).

Culture of MAP from yoghurt was achieved on 7H10 agar with antibiotics (Van Brandt et al., 2011a). For hard cheese, Donaghy et al. (2003) found that a decontamination step inhibited all cheddar cheese starter and nonstarter microor­ganisms, and that Middlebrook 7H10 agar con­taining polymyxin, amphotericin B, nalidixic acid, trimethoprim and azlocillin (PANTA) and to a lesser degree HEYM containing vancomy­cin, amphotericin B and nalidixic acid (VAN) inhibited growth of these microflora when no decontamination step was employed. MAP was subsequently cultured from cheddar cheese on HEYM agar medium containing VAN without a decontamination step (Donaghy et al., 2004). MAP was isolated recently from a soft goat cheese that had been decontaminated in 0.75% HPC and cultured on 7H11 agar (Galiero et al., 2015).

18.11.5 Blood

Blood culture for MAP is not undertaken routinely and there have been no systematic studies on suitable methodologies or on the prevalence of bacteraemia. In the first study involving blood culture from livestock (Koenig et al., 1993), heparinized buffy coat-plasma preparations were decontaminated in 0.75% HPC and cultured on HEYM, with positive re­sults from one of seven cows that had advanced disease. After first optimizing the methods for liquid culture in BACTEC medium (Bower et al.,

2010), the occurrence of mycobacteraemia was investigated in 111 sheep. Dissemination of infection to the liver and hepatic lymph node was detected in 18 of 53 infected sheep while the bacterium was isolated from the blood of only 4 of these; prolonged incubation periods prior to growth were consistent with inhibition of growth or dormancy of MAP in some cul­tures (Bower et al., 2011). The problem of con­tamination of blood cultures was dealt with by incubation of blood lysates in 0.75% (w/v) HPC for 72 h (Bower et al., 2010, 2011). There have been at least two studies looking for MAP in the blood of Crohn's disease patients, with various methods, different levels of stringency in microbiological assessment and conflicting results (Naser et al., 2004; Parrish et al., 2008).

18.11.6 Environmental samples

MAP tends to associate with soil particles, espe­cially organic matter and certain clays (Dhand et al., 2009a, b). It tends to remain on grass and in the upper layers of soil where it is available to grazing livestock and can contaminate the runoff after heavy rain (Dhand et al., 2009a; Salgado et al., 2011). To determine the status of grazing livestock at herd/flock level, MAP can be cultured from environmental samples includ­ing soil, water, pasture, faeces collected from the environment, manure and slurry collected from drains, dust from barns, boot swabs and collec­tion socks (Whittington etal., 2003b; Khol et al., 2009; Pillars et al., 2009; Eisenberg et al., 2011, 2013; Arango-Sabogal et al., 2016a; Ritter et al., 2016; Noll et al., 2017). The entire approach was reviewed recently (Wolf et al., 2017). Culture methods for environmental samples are usually based on those for faeces (Whittington et al., 1998, 2004, 2005; Pickup et al., 2005; Salgado et al., 2011).

The sensitivity of culture of faecal samples collected from the environment is less than that of individual or pooled culture of faecal samples collected directly from animals; for example, see Lavers et al. (2013).

Environmental culture is most applicable in dairy cattle where manure collects in drains adjacent to the dairy parlour. Faeces from most cows will collect there over time and is readily available for sampling. It is the cheap­est method of sampling for determination of herd infection status (Lombard et al., 2006; Tavornpanich et al., 2008). As the number of animals contributing to a particular sample is unknown, environmental sampling cannot be used to reliably estimate within-herd preva­lence but can be useful for regional prevalence estimation (Kruze et al., 2013). The method has also been applied in extensively grazed sheep but few positive cultures were obtained from pastures on affected farms (Whittington et al., 2003a).

Other applications of environmental cul­ture include the search for possible transport vectors of MAP such as blowflies, earthworms and parasitic nematode larvae (Whittington et al., 2001; Fischer et al., 2003, 2004).

Observations of MAP in the environment using culture revealed an important phenom­enon - dormancy - whereby the organism was able to survive for an extended period in a non-cultivable state (Whittington et al., 2004). Further evidence of dormancy was an extended lag phase when stressed MAP were inoculated into liquid media (Gumber et al., 2009). MAP possesses genes and proteins that are known from homology searches with other mycobacte­ria to be dormancy-associated. It may be possible to revive dormant MAP cells to enhance culture success rates, but further studies are required (Rock et al., 2016).

Concentration of water samples by centrif­ugation or filtration is required prior to culture (Aboagye and Rowe, 2011). Decontamination methods suitable for river water and municipal drinking water have received relatively little attention. Recent work to understand the in­teraction between decontamination protocols and culture media employed for water sediment samples has confirmed the need for further stud­ies leading to optimized methods (Aboagye and Rowe, 2018).

18.11.7 Animal wastes, compost and manufactured pasture by-products

Stall waste consisting of bedding straw, faeces and the slurry of washings may be converted into useful products such as biogas, compost and fertilizer. A question remains about the viability of MAP after such industrial processes (Gobec et al., 2009; Khol et al., 2010; Slana et al., 2011; Mazzone et al., 2018; Donat et al., 2019). Similar questions arise about the risk of transmission of MAP when straw or silage is made from pas­tures on which livestock have grazed or to which their manure has been applied (Katayama et al., 2001; Fecteau et al., 2013) and the ultimate fate of the pathogen when manure is sprayed on to grasslands is also of interest (Salgado et al.,

2011). Protocols based on those for faeces can be used to culture MAP from such samples to in­form risk assessments.

18.11.8 Evaluation of antimicrobials

Treatment of MAP infection in animals has not been attempted very often due to the adverse economics and the practical constraints of ad­ministering drugs over a prolonged period; for a zoological collection example, see Bryant et al. (2012). However, there may be applications for antimicrobials in neonatal prophylaxis (Fecteau et al., 2011; Ali et al., 2019). Antimicrobials are also relevant in the context of Crohn's disease in humans, where individual patients may be treat­ed and clinical trials undertaken (Selby et al., 2007; Krishnan et al., 2009a, b). To assess an­timicrobial sensitivity of MAP isolates, growth inhibition assays are possible in liquid or solid media, but standards for such tests are needed (Krishnan et al., 2009a, b; Fecteau et al., 2014; Nowotarska et al., 2017; Steuer et al., 2018; Ali et al., 2019).

18.11.9 In vitro propagation and stability of MAP

There are many applications that require bulk propagation of MAP. These include for pro­duction of DNA for non-PCR molecular typing methods such as restriction fragment length polymorphism (RFLP) analysis, antigen pro­duction for immunological tests such as delayed type hypersensitivity (Johnin PPD) or produc­tion of live bacterial pellets or suspensions for inclusion in vaccines and for use as inoculum in experimental animal models. There has been an implicit assumption when MAP is cul­tivated for each of these applications that the organism is genetically stable. In fact, most au­thors do not report the passage history (media, growth conditions, number of subcultures) of isolates used in their studies. While there have been few studies to investigate genetic changes following laboratory propagation of MAP, the findings have been enlightening. Kasnitz et al. (2013) showed that alterations can occur in IS900 BstE11 RFLP profiles in some isolates of MAP after subculture to different media and af­ter repeated subculture, due to factors such as activity of transposase within IS900 and point mutations at specific recognition sequences for restriction endonucleases. Furthermore, some short sequence repeats were unstable. Bull et al. (2013) showed that different culture histories over decades in different laboratories was associated with substantial genome diver­sity in the vaccine strain MAP 316, including large tandem genomic duplications, deletions and transposable element mobility. This mir­rored findings with BCG, which has a similar history of unregulated, multicentric, labora­tory passage (Behr et al., 1999). Evidence was presented recently to show that in vitro cul­ture can lead to a reduction in the virulence of MAP (Fernandez et al., 2015), which is consistent with a genetic change in the organ­ism through selection. Bryant et al. (2016) in­vestigated genome stability in two MAP field strains that were passaged several times on a 7H11 agar medium and found mutations in genes responsible for nitrogen assimilation. Authors should record the provenance of MAP strains and their passage history when reporting their findings, regardless of the type of study.

18.12