Developing Xenostandards for Microbiological Safety: New Zealand Experience

Human patients have the potential to become infected with animal viruses following xenotransplantation. Concerns have also been raised that in a worst case scenario, pathogens originating from pig donors may adapt and then propagate to the wider public resulting in a new epidemic (Fiane, Mollnes, & Degre, 2000; Fishman, 2001a; Onions et al., 2000; Patience, Wilkinson, & Weiss, 1997; Weiss, 2003). The likelihood that cross-species infection may occur is enhanced in a xenotransplantation setting because normal host defences such as skin and mucosal surfaces are bypassed and direct contact between donor and recipient cells is maintained for extended periods of time (O’Rourke, 2000). Similarly for applications that use immunosuppression to prevent xenograft rejection, host complementmediated immunity is circumvented (Takeuchi, Magre & Patience, 2005). Currently pigs are recognised as the most popular choice as donor animals due in part to their ostensibly lower infectious risk (compared with non-human primates), excellent breeding potential, comparable organ size, physiological similarity, amenability to genetic modification, non contentious public perception and the relatively moderate costs associated with their maintenance (Sachs, Sykes, Robson, & Cooper, 2001). In terms of pig pathogens it has been determined that the preponderance of fungi, bacteria and parasites can be excluded as major risk factors simply by the use of good animal husbandry practices in Specific Pathogen Free (SPF) herds (Ye, Niekrasz, Kosanke, Welsh, Jordan et al, 1994). Consequently viruses are recognised as the predominant infectious agent for zoonosis owing to their rapid rate of evolution and excellent adaptive competence within new hosts. Precedents for cross-species infection of viruses and adaptation in humans are numerous and include several notorious examples, notably AIDS (Gao, Bailes, Robertson, Chen, Rodenburg et al, 1999) and avian influenza [reviewed in (Alexander & Brown, 2000)]. Although pigs are not always a natural reservoir for exogenous viruses, it has been hypothesised that swine may act as “mixing vessels” for adaptation to human hosts, as is certainly the case for avian viruses such as Severe Acute Respiratory Syndrome (SARS) (Bush, 2004).


Introduction
Human patients have the potential to become infected with animal viruses following xenotransplantation. Concerns have also been raised that in a worst case scenario, pathogens originating from pig donors may adapt and then propagate to the wider public resulting in a new epidemic (Fiane, Mollnes, & Degre, 2000;Fishman, 2001a;Onions et al., 2000;Patience, Wilkinson, & Weiss, 1997;Weiss, 2003). The likelihood that cross-species infection may occur is enhanced in a xenotransplantation setting because normal host defences such as skin and mucosal surfaces are bypassed and direct contact between donor and recipient cells is maintained for extended periods of time (O'Rourke, 2000). Similarly for applications that use immunosuppression to prevent xenograft rejection, host complementmediated immunity is circumvented (Takeuchi, Magre & Patience, 2005). Currently pigs are recognised as the most popular choice as donor animals due in part to their ostensibly lower infectious risk (compared with non-human primates), excellent breeding potential, comparable organ size, physiological similarity, amenability to genetic modification, non contentious public perception and the relatively moderate costs associated with their maintenance (Sachs, Sykes, Robson, & Cooper, 2001). In terms of pig pathogens it has been determined that the preponderance of fungi, bacteria and parasites can be excluded as major risk factors simply by the use of good animal husbandry practices in Specific Pathogen Free (SPF) herds (Ye, Niekrasz, Kosanke, Welsh, Jordan et al, 1994). Consequently viruses are recognised as the predominant infectious agent for zoonosis owing to their rapid rate of evolution and excellent adaptive competence within new hosts. Precedents for cross-species infection of viruses and adaptation in humans are numerous and include several notorious examples, notably AIDS (Gao, Bailes, Robertson, Chen, Rodenburg et al, 1999) and avian influenza [reviewed in (Alexander & Brown, 2000)]. Although pigs are not always a natural reservoir for exogenous viruses, it has been hypothesised that swine may act as "mixing vessels" for adaptation to human hosts, as is certainly the case for avian viruses such as Severe Acute Respiratory Syndrome (SARS) (Bush, 2004).
Recommendations to minimize cross-species infection stipulate that donor pigs should be maintained in quarantined facilities and monitored for the presence of exogenous pathogens. This type of program requires significant resources so there is considerable incentive to monitor only those viruses that are relevant. Several comprehensive reviews have been published describing zoonotic agents in pigs that may potentially cause disease in www.intechopen.com Xenotransplantation 78 transplant recipients and theoretically the population at large (Fishman, 2001a(Fishman, , 2001bMueller, Takeuchi, Mattiuzzo, & Scobie, 2011;Scobie & Takeuchi, 2009;Takeuchi et al, 2005). A particularly valuable resource is the report in the journal Xenotransplantation by Shuurman (2009) that offers a consensus view of organisms that should be excluded from donor pigs. The most important pathogens are reported to be swine influenza viruses belonging to the influenza A genus, Nipah virus, Marburgvirus and Ebolavirus of the Filoviridae family, rotaviruses, parvoviruses, hepatitis E and herpesviruses such as porcine circovirus type 2 (PCV2) and porcine lymphotrophic herpes virus type 2 (PLHV2).
In New Zealand a very deliberate strategy was employed in order to maximise patient safety by ensuring that these viruses and other pathogens were absent in a donor herd destined for clinical trials. Two important goals were considered integral to the success of the New Zealand approach. The first goal was to ensure the safety of donor material and was achieved by utilising the following steps: (1) investigating the health status of the New Zealand pig population; (2) developing and implementing an algorithm for herd selection; (3) establishing a program to characterise pig endogenous retrovirus; (4) defining a testing schedule to maintain the specific pathogen-free (SPF) status of the donor herd; (5) certifying the health of all donor animals and (6) certifying the safety of the final product. The second goal required the implementation of a robust monitoring program for patient follow-up post-transplant. In this case specific emphasis was placed upon the development of reliable and sensitive assays to detect pig pathogens in humans as well as developing a network of collaborative reference laboratories. Important also was the establishment of a xeno-microbiology laboratory and its accreditation as a medical diagnostic laboratory to guarantee the highest diagnostic standards. Using this strategy regulatory approval was obtained to begin the first clinical trials using insulinproducing cells (DIABECELL®) for the treatment of Type I Diabetes (http://www.lctglobal.com/lctdiabecell-diabetes-treatment.php).

Determining infections relevant to xenotransplantation in NZ
The Public Health Service Guidelines on Infectious Disease Issues in Xenotransplantation states that monitoring programs for source animals should be tailored to specific geographical areas (U.S Food and Drug [FDA], 2001). This recommendation implies knowledge of the local pig infection profile and in particular, knowledge of virus infections potentially relevant to xenotransplantation. The health status of the NZ pig population is considered to be favourable thanks largely to its geographic remoteness and strict animal health control policies. Unlike many countries NZ remains free from infectious vesicular diseases such as foot and mouth, vesicular stomatitis and vesicular exanthema. Notable infections like rabies, Brucella suis, swine fever, pseudorabies and spongiform encephalopathy are also absent. Routine screening protocols for a source herd intended for xenotransplantation may encompass up to 45 different pathogens (H. J. Schuurman, 2009) many of which are common to veterinary practices. However, for the NZ pig population very limited data was initially available regarding viruses such as Porcine Cytomegalovirus (PCMV), Porcine Lymphotrophic Herpesvirus (PLHV), Encephalomyocarditis Virus (EMCV), Porcine Circovirus Virus (PCV) and pig Hepatitis E Virus (HEV). In order to extend the data regarding the prevalence of these potentially zoonotic viruses an investigatory screening program was performed that tested representatives from herds www.intechopen.com Developing Xenostandards for Microbiological Safety: New Zealand Experience 79 throughout NZ (Garkavenko, Elliott, & Croxson, 2005;Garkavenko, Muzina, et al., 2004;Garkavenko et al., 2001). This information led to a significantly improved understanding of the infection profile within NZ pig breeds and facilitated the development of a screening protocol for viruses that were particularly relevant to xenotransplantation. Numerous NZ pig herds intended as a source of islet cells for transplantation were comprehensively assessed according to this screening protocol.
Having established the epidemiology of our test panel of viruses in NZ pigs, a further examination was performed in one-week old piglets (the donor age group from which islet cells are harvested). All one-week-old piglets were free of the majority of tested viruses. This finding implies that infection with PCMV, PLHV and HEV takes place perior post-natal and that despite the presence of infection within a herd, new-born piglets and their tissues may remain virus-free. This conclusion is in contrast with another study in which the PLHV virus was shown to be vertically transmitted although it should be noted that this type of transmission phenomenon is considered to be a rare event (Tucker et al., 2003). The presence of PCV2 in tissues and faecal samples of one-week-old piglets was an unexpected result. Although it was shown that this virus can be transmitted vertically Ladekjaer-Mikkelson et al., 2001;O'Connor et al., 2001;Sanchez, Nauwynck, McNeilly, Allan, & Pensaert, 2001), there is a common view that this virus is associated with severe pathology in new-born piglets, and its vertical transmission is also a rare event.
Three factors must be considered when assessing the infectious risk of PCV2 associated with xenotransplantation: (1) the ubiquity of the PCV type 2 virus (Celera & Carasova, 2002;Garkavenko et al., 2005;Kim et al., 2002;Labarque, Nauwynck, Mesu, & Pesaert, 2000;Trujano, Iglesias, Segales, & Palacios, 2001;Wattrang et al., 2002); (2) the potential that the virus might be transmitted vertically without showing any sign of abnormality  and (3) transmission of the virus occurs in human cells in vitro (Hattermann, 2002, unpublished data), and in experimental mice in vivo (Kiupel et al., 2001). Despite evidence that caesarean section and barrier rearing techniques are effective in excluding PCV and PLHV from pig populations, these interventions do not guarantee the exclusion of all viruses from a pig herd (Tucker et al., 2003). While each of these factors requires further intensive study, it seems reasonable to suggest that any source herd intended for xenotransplantation must be free from PCV and PLHV.

Selection criteria for xenotransplantation donors
The microbiological 'xenostandard' demands the absence of identifiable infectious agents in the source herd. A NZ donor herd that fulfilled this standard was found through a monitoring program that required the completion of several key tasks. This included a viral profile assessment of the pig population within the country, identification of a specific source herd, development of safety criteria including a plan to manage positive animals and finally, implementation of a multilevel monitoring schedule encompassing pig founders through to the final product. For the New Zealand herd an algorithm to determine whether source animals were suitable for xenotransplantation was developed (see Figure 1). This algorithm stipulates that if an agent capable of infecting human cells cannot be eradicated from the herd then an assessment of risk versus benefit must be made. Using this algorithm it became apparent that pigs from the Auckland Islands (AI) represented the ideal donor herd.

The Auckland Island donor herd
The Auckland Islands are a group of islands to the south of mainland New Zealand. In 1807 individual pigs were first introduced to the island as a source of food for whalers and shipwrecked sailors. Over subsequent decades these animals were reported to be thriving leading to further releases in 1840, 1842 and 1890. By the end of the nineteenth century a strong feral pig population had been established that was to remain isolated for the next hundred years. By the mid 1990's these animals were marked for eradication by the New Zealand Department of Conservation as part of a program to restore the natural ecosystems on the island. Consequently the Rare Breeds Conservation Society of New Zealand (RBCSNZ) attempted to preserve the breed in captivity. In 1999 a RBCSNZ expedition caught and removed seventeen pigs, including several pregnant sows, from Auckland Island, transferring them to Invercargill, New Zealand. These animals were subsequently acquired by the private biotech company Living Cell Technologies for the purposes of providing donor material for cell transplantation in humans for the treatment of disease. Currently these animals are maintained within quarantined facilities with one founder herd of approximately 40 animals kept in Invercargill, New Zealand and a second smaller donor herd of 15 animals located in Kumeu, Auckland, New Zealand. These animals as well as the cellular products derived from them are subject to a rigorous testing schedule for an extensive list of pathogens (Table 1).

Multilevel monitoring
A key facet of the 'selection algorithm' is that a multilevel monitoring process is adopted that begins with the regular screening (see table 1) of every individual pig within the donor herd. For certain pathogens such as PCV2, PLHV2 and PCMV sampling maybe performed from multiple tissues including pancreas and peripheral blood leukocytes (PBL). In the case of HEV both blood serum and faeces are processed. This is followed by the screening of all the sows' pre and post farrowing as well as the screening of each piglet donor (see table 2 for tests performed). Prior to surgery and removal of organs each piglet is examined for signs of infection. Any abnormalities are noted and in certain cases further consultation maybe sort from veterinary specialists. A strict policy is enforced to exclude even slightly suspect piglets from contributing donor material. Following removal of donor material each piglet is subject to a full post mortem by veterinary pathologists. Lastly the transplant product itself is screened immediately prior to transplantation. This final clearance is mandatory before the product can be released for transplantation. Such a comprehensive schedule of testing was judged necessary to assure the microbiological safety of xenografts and to satisfy NZ regulatory authorities.  Table 2. List of porcine microorganisms for multi-level screening program. *PCR is performed on DNA extracted from both peripheral blood leukocytes and pancreas tissue.

Porcine Endogenous Retrovirus (PERV) -Developing a standard approach
In comparison to exogenous viruses a unique problem is posed by endogenous retroviruses (ERV). In pigs this pathogen is known as Porcine Endogenous Retrovirus (PERV) and is ubiquitous to all pig species studied to date. Endogenous retroviruses can be distinguished from exogenous types by their presence as genomic components that do not present pathology in their natural host . Moreover, endogenous retroviruses behave like normal cellular genes in that they are inherited by the offspring and not acquired by infection. Accordingly PERV cannot be removed by conventional barrier methods instead requiring more elaborate strategies such as the selective breeding of pigs with favourable PERV genetic characteristics Garkavenko, Wynyard, Nathu, Simond, Muzina et al, 2008;Stoye et al, 1998), or the development of vaccines to protect against PERV transmission or the inhibition of PERV expression by RNA interference using PERV-specific short hairpin RNA (shRNA) and retroviral vectors [reviewed in (Denner, 2008)]. Like most retroviruses PERV has an element of unpredictability in regards to its transmission potential. The concern exists that retroviral transmission may occur 'silently' by means of an undetectable recombination event, oncogenesis or alteration in gene regulation resulting in pathology that does not manifest until decades later. Such transmission is typical for gammaretroviruses and xenotransplantation could provide the right environment for selection of variants that can efficiently infect the human population. Numerous studies have been performed examining the risk of PERV transmission as it relates to xenotransplantation with the consensus that PERV must be considered an endogenous agent to be reduced or excluded where possible in animals destined for clinical use (H.J. Schuurman & Pierson, 2008;Taylor, 2008;Wilson, 2008).

PERV infectivity in vitro -The gold standard
With regard to PERV infectious characteristics, a standard infectivity method can be applied to check the infectivity of primary tissues such as peripheral blood mononuclear cells (PBMC) and islets derived from SPF pigs (Patience, Wilkinson, et al., 1997;Takeuchi et al., 1998). Such data is necessary because PCR for either the PERV pol or PERV env region does not always provide information on the transmission characteristics of PERV. In fact recombinant PERV is most often detected in transmission studies by co-culturing primary pig cells with target human cell lines as was performed in the NZ SPF herd. To ensure the release of infectious PERV particles (if any), tested cells were mitogenically stimulated prior to co-culture with standard susceptible cell lines 293 (to elucidate xeno-tropic PERV) and St-Iowa (to elucidate eco-tropic PERV). Although mitogenic stimulation enhanced proliferation of all tested cells, no evidence of PERV transmission was detected in the infectivity test with both human and pig target cells using PBMC or islet cells isolated from the New Zealand SPF pigs Garkavenko, Wynyard, Nathu, Simond, et al., 2008). Thus it was concluded that the New Zealand SPF pigs could be classified as non-transmitters for PERV, possessing the "null" transmission phenotype. The term "null", as coined by (Wood et al, 2004), simply means that despite the presence of PERV-A, PERV-B and PERV-C sequences in the genome of these pigs they lack the ability to infect human or pig cells in vitro. This method must be considered the gold standard for determining PERV infectious risk and should be considered compulsory when evaluating new donor herds.
One aspect that can complicate the issue of transmission phenotype is the potential for the transmission status of an animal to change over time. Indeed, data has shown that miniature swine do not maintain their transmission status. For example Wood et al demonstrated that during one 14 month interval, several adolescent miniature swine that were initially capable of infecting pig cells only (non-transmitters) eventually infected human cells thus converting to the transmitter phenotype (Wood et al., 2004). It is unclear if this phenomenon is applicable for other pig breeds, nevertheless any monitoring program of donor SPF herds should include a periodic screening for infectious PERV. Founder animals in New Zealand are tested annually or biennially for PERV infectivity by co-culturing with target cell lines and data has shown that these animals have retained their PERV null phenotype for more than 6 years (Wynyard, 2011).

PERV expression and RT activity
In addition to studying in vitro co-culture infectivity it is recommended that the PERV viral load and RT activity is measured in donor blood plasma. During the characterisation of the www.intechopen.com New Zealand SPF herd PERV viral load in donor blood plasma was measured by real-time PCR using a methodology similar to that described by Dieckhoff et al (2009). RT activity that would indicate retroviral activity and therefore the presence of retroviral particles was also measured in the blood of the donors using a C-Type RT kit Garkavenko, Wynyard, Nathu, Simond, et al., 2008). No evidence of PERV expression or RT activity in pig donors' blood plasma or from stimulated PBMC was found.

PERV gene dosage (copy number)
Pigs that possess a low number of PERV provirus within their genome will theoretically pose less infectious risk than those with a high copy number. This is because the majority of PERV sequences appear to be either defective or deleted, i.e., they contain only one or two intact retro-viral open reading frames of gag, pol and env (Bosch, Arnauld, & Jestin, 2000;Czauderna, Fischer, Boller, Kurth, & Tonjes, 2000;Herring et al., 2001;Niebert, Rogel-Gaillard, Chardon, & Tonjes, 2002). Consequently PERV proviral copy number is another important characteristic that should be investigated when assessing a donor herd for infectious risk. This data can be obtained using a real-time PCR absolute quantification methodology. If this is not feasible then a laboratory may employ the simpler PCR based limited dilution assay (PLDA). PLDA has been widely used to quantify different target molecules, including human immunodeficiency virus copy number (Rodrigo, Goracke, Rowhanian, & Mullins, 1997). The New Zealand SPF herd has been screened using both methods (Wynyard, 2011;Wynyard, Garkavenko, & Elliott, 2011) for which a high concordance was observed. PERV copy number in NZ SPF pigs varied from 3 to 68 copies per cell with an average copy number of 17.6 and were found to be not statistically different from the NZ landrace breed or more interestingly the Miniature Swine breed that are known to infect human cells in vitro . These techniques provide a sensitive and reliable method to specifically identify animals with low PERV copy number (<10 copies per cell) that are suitable for further selective breeding. Numerous such individual pigs have been identified in the AI pig herd and are currently being bred to reduce PERV infectious risk.

PERV recombinants
Another important aspect to consider is the ability of PERV-A and PERV-C to recombine as there is evidence that PERV A/C recombinants show higher titres when cultured in human cells in vitro (Bartosch et al., 2004;Oldmixon et al., 2002;Wilson, Wong, VanBrocklin, & Federspiel, 2000;Wood et al., 2004). As discussed earlier, using the co-culture infectivity test with human HEK293 and swine St Iowa target cell lines, it was established that primary cells (PBMCs) from NZ donor pigs do not release either xeno-or ecotropic infective viruses Garkavenko, Wynyard, Nathu, Simond, et al., 2008). To support the lack of transmission from this herd, animals were tested for and found to lack the genomic presence of high titre recombinant PERV A/C. A potential PERV-C locus that may contribute to recombination and the generation of transmissible PERV sequences in miniature swine was also found to be absent. These animals also appear to be transcriptionally inactive for PERV-C as PERV-C RNA could not be detected despite possessing PERV-C proviral sequences.

PERV infectivity in vivo
An assessment of PERV infectious risk can be made at multiple levels beginning with the ability to release virus from the cells of a xenografted organ. Subsequent analysis would need to characterise the virus's ability to cause viremia by revealing productive infection within the recipient cells. Final recognition as a public health hazard would require demonstration of the infectious virus circulating within the patients' bodily fluids. While the ability of PERV to infect human cells in vitro is well documented, less is actually known about PERV infectious potential in vivo and its capacity to cause disease. To address this question numerous animal models have been investigated. The most common models have been concentrated around the use of non-human primates (NHP) or small animals such as severe combined immunodeficiency (SCID) mice. Unfortunately both models have proven problematic and in the case of SCID mice somewhat contentious. In NHP, conflicting results have been reported as regards the resistance of NHP cells to PERV infection, with some studies reporting non-permissiveness (Martin, Steinhoff, Kiessig, Chikobava, Anssar et al, 1999;Takeuchi et al, 1998;Wilson et al, 2000) and others suggesting susceptibility (Blusch, Patience, Takeuchi, Templin, Von Der Helm et al, 2000;Specke, Tacke, Boller, Schwendemann & Denner, 2001;Templin, Schroder, Simon, Laaff, Kohl et al, 2000). For mice the situation is more complicated. Originally it was discovered that PERV, produced from pig pancreatic cells and transplanted into SCID mice, could infect mouse tissues in vivo. However, the virus produced appeared to be transcriptionally inactive, signifying a non-productive infection (van der Laan, Lockey, Griffeth, Frasier, Wilson et al, 2000). Subsequent analysis revealed that this was not a true infection but rather evidence of pseudotyping involving the collaboration of mouse (endogenous xenotropi c M LV ) and PE RV r etro v ira l e lem ents (Martina, Kurian, Cherqui, Evanoff, Wilson et al, 2005). In terms of overall success, despite several studies demonstrating the transmission of PERV in vivo (Argaw, Colon-Moran & Wilson, 2004;Martina, Marcucci, Cherqui, Szabo, Drysdale et al, 2006;Popp, Mann, Milburn, Gibbs, McCullagh et al, 2007), no report has conclusively demonstrated productive infection (Denner, Specke, Karlas, Chodnevskaja, Meyer et al, 2008;Hermida-Prieto, Domenech, Moscoso, Diaz, Ishii et al, 2007;Levy, Argaw, Wilson, Brooks, Sandstrom et al, 2007;Moscoso, Hermida-Prieto, Manez, Lopez-Pelaez, Centeno et al, 2005;Paradis, Langford, Long, Heneine, Sandstrom et al, 1999;Specke, Schuurman, Plesker, Coulibaly, Ozel et al, 2002). In terms of in vivo transmission from the AI pig herd no evidence of PERV infection was found in non-human primates following transplantation of islet cells (Garkavenko, Dieckhoff, Wynyard, Denner, Elliott et al, 2008) or in twelve human patients sampled from the New Zealand clinical trial (Wynyard, 2011).
Such a predominance of data against PERV infection in vivo begs the question as to whether PERV is of real importance. On the other hand it may just be a case of having not yet found a suitable animal model. It is worth noting that the difficulties associated with PERV in vivo infection may also be attributed to the inconsistency of methods used for PERV detection in recipients. Indeed the detection of all PERV subtypes, especially PERVA/C, may not be adequately covered by currently employed PCR and serological methodologies. Establishing quality testing practices therefore becomes vital. www.intechopen.com

Patient follow up and laboratory standards
Approval for clinical trials in xenotransplantation requires comprehensive regulation and approval by government authorities. Crucial to any successful application is demonstration that the methodologies employed to test both donor animals and xenograft recipients are current and accurate. The Molecular Diagnostic Laboratory responsible for testing the New Zealand SPF donor herd and transplant patients employs both serological (ELISA, Late Agglutination Test) and molecular techniques (PCR, real-time PCR) as part of its testing program. However the preferred methodology is a multiplex High Resolution Melting (HRM) real-time PCR similar to that described in (Wynyard, 2011). In this study the superior melting properties of the HRM chemistry enable the simultaneous amplification of sequences for PERV pol, Cytochrome Oxidase II (a pig cell marker) and a heterologous internal control in a single multiplexed reaction. This assay has been employed successfully to screen 12 xenograft recipients of porcine islets to exclude PERV infection whilst simultaneously checking for template integrity, PCR inhibition and microchimerism. Research groups expecting to perform xeno-testing must be prepared to show competencies with nucleic acid methodologies preferably with expertise in both veterinary and medical diagnostic fields. For pathogens that lack commercially available tests or where in-house capabilities are insufficient then a network of reference laboratories and collaborators maybe employed to perform the tests instead. All assays should be suitably validated to ensure accuracy, reproducibility, specificity and sensitivity (both analytical and where possible diagnostic) as expected from any medical diagnostic laboratory (American Association of Veterinary Laboratory Diagnosticians, 2010; Raymaekers, Bakkus, Boone, de Rijke, El Housni et al, 2011). It is important that assays for pathogens that are screened in donor animals show equal efficacy when tested from human tissues and validation protocols should account for this. Controls are mandatory and it is recommended that suitable internal controls are incorporated to improve assay reliability (Hoorfar, Malorny, Abdulmawjood, Cook, Wagner et al, 2004). For example in the New Zealand diagnostic laboratory tissues destined for real-time PCR analysis are routinely spiked with a heterologous internal control that can be used to infer nucleic acid integrity and PCR inhibition (Wynyard, 2011; assays in preparation for publication).
To achieve the highest diagnostic standards it is advisable that laboratories expecting to perform xenomicrobiology testing are accredited to international standards. This will facilitate the use of suitable methods and that laboratory staff are trained and competent to perform each assay. Moreover, accreditation ensures that a suitable QA/QC program is put in place to guarantee that results are reliable and can be trusted. The NZ testing laboratory is audited annually and registered with International Accreditation New Zealand (IANZ). Accreditation is maintained to the ISO15189:2007 standard. It is important to note that for the screening of both the herd and recipients that robust protocols for managing positive results are established. This encompasses the confirmatory testing needed to exclude false positives and describes clearly defined communication lines for reporting. Reporting contacts may include the animal facilities, executive management, pig handlers or in the case of positive results for non-endemic pathogens, the involvement of government departments responsible for biosecurity, animal welfare and public health. Full documentation (integral to the accreditation process) is essential for providing traceability when subject to auditing by regulatory bodies.

Conclusions
Any xenotransplantation project should consider a comprehensive safety program that includes two key aspects -donor monitoring and patients' follow-up. Initial identification of donor herds suitable for xenotransplantation requires a comprehensive understanding of viral epidemiology within pig populations and an understanding of the relevant viruses as dictated by the geographical area. A clear decision making process is required to exclude animals that carry infectious agents that ultimately requires a risk versus benefit analysis. Upon the selection of suitable pig donor herd a multilevel screening program is required that tests donor material from breeding animals at the herd level through to the final product. Strategies for management of positive animals must be in place to minimize the risk to the donor herd and transplant recipients. PERV remains the central safety consideration in xenotransplantation. It is very important to develop a standard approach towards the characterization of PERV infectious potential within a potential donor herd. Such a benchmark allows for the comparison of PERV characteristics from pig donors of different backgrounds and ensures the selection of techniques and methods that are reliable and practical for the industry. To effectively characterize PERV in source animals requires a rigorous PERV screening program as has been implemented in a New Zealand specific pathogen free (SPF) pig herd. The key elements of this program should consist of: (a) testing for in vitro infectivity of both eco-and xenotropic viruses using standard cell culture infectivity methods (Patience, Takeuchi, & Weiss, 1997); (b) measuring reverse transcriptase (RT) activity and PERV viral expression in donors' blood plasma (Dieckhoff et al., 2009;Garkavenko, Wynyard, Nathu, Simond, et al., 2008); (c) measuring the PERV proviral copy number per cell (Wynyard, 2011) and (d) testing for the presence or absence of PERV recombinants and a potential PERV-C loci that may contribute to recombination and the generation of highly transmissible PERV sequences Garkavenko, Wynyard, Nathu, Simond, et al., 2008). These data provide the basis for a selective pig breeding program with the ultimate goal of enhancing the safety of donors for cell transplantation by minimising PERV infectious risk. It should be evident that patient safety relies heavily upon a robust program of monitoring in transplant recipients. It is expected that laboratory practices meet international standards to ensure technical competence and result validity.
The strategies described in this chapter have been successfully applied in New Zealand and proven crucial towards facilitating clinical trials using porcine islets. It is expected that the same strategies will find broad application outside of New Zealand and provide sufficient guidelines to benefit interested parties looking to enter the xenotransplantation field.