Determination of the Copy Number of Porcine Endogenous Retroviruses (PERV) in Auckland Island Pigs Repeatedly Used for Clinical Xenotransplantation and Elimination of PERV-C

1. Introduction

The Auckland Islands are a group of subantarctic islands which lie some 560 kilometers south of New Zealand. Pigs were originally introduced onto these islands in 1807 by Captain Abraham Bristow and in 1840 James Ross brought more pigs onto the island [1,2]. From the time of the last release the pig population on Auckland Island remained isolated for the next hundred years. Pigs were removed from Auckland Island in 1999 to help restore the environment to its former state and were brought to the mainland of New Zealand for conservation. 17 animals were housed, initially in quarantine, in a special, purpose-built facility in Invercargill, where they have been successfully bred [3]. The pigs are black or white to brown with black markings, have a long snout, high hair and bristle coverage for the cold conditions [1]. When the mitochondrial D-loop DNA sequences of the Auckland Island pigs were compared with sequences from domestic and wild boar, the sequence from the Auckland Island pigs clustered with domestic European breeds and they represent a single breeding population [1,2]. Phylogenetic analyses of microsatellites also showed that they are more closely related to European pigs than Chinese pigs, which is consistent with the mitochondrial analyses [3].

These pigs have been monitored for 8 years and were found negative for all tested porcine infectious agents [4,5,6,7]. Among the tested viruses potentially posing a risk for xenotransplantation were the porcine endogenous retrovirus (PERV), porcine cytomegalovirus (PCMV), a porcine roseolovirus (PCMV/PRV), porcine lymphotropic herpesvirus (PLHV), hepatitis E virus (HEV), porcine teschovirus (PTV), porcine rotaviruses A, B, C (PRVA, B, C), porcine hemagglutinating encephalomyelitis virus (PHEV), porcine parvovirus (PPV), encephalomyocarditis virus (EMCV), porcine enterovirus B (PEVB), porcine reproductive and respiratory syndrome virus (PRRSV), and porcine circovirus (PCV). For testing diagnostic PCR and serology methods were applied. In comparison with other pig breeds, the Auckland Island pigs are the “cleanest” pigs concerning virological safety. This was the reason to use pig islets from Auckland Island pigs for the treatment of humans with diabetes. In a prospective preclinical trial, transplanting islet cells from Auckland Island pigs into non-human primates, no transmission of pig viruses was observed [6]. Most importantly, no transmission of pig viruses including PERVs was observed in two clinical trials performed in New Zealand [7] and Argentina [8].

Here we analyze the number of PERV proviruses in the genome of adult Auckland Island pigs, in cells from young piglets as well as in animals obtained by somatic cell nuclear transfer (SCNT) using these piglet cells. Furthermore, we screened these animals for an additional 8 xenotransplantation-relevant viruses, among them PCMV/PRV. PCMV/PRV is of special interest since it is known to significantly reduce the survival time of non-human primates after transplantation of PCMV/PRV-positive kidneys or hearts [9,10,11,12]. PCMV/PRV was also transmitted in the first clinical trial transplanting a pig heart into a patient in Baltimore and contributed to the death of the patient [13,14]. We confirmed the absence of PCMV/PRV and other potentially zoonotic viruses in the Auckland Island pigs.

It is known that PERV-C can recombine with PERV-A in living pigs and the resulting PERV-A/C recombinant viruses are characterized by higher virus titers compared with the paternal PERV-A [15,16,17]. Release of human tropic PERV-A/C from mitogen-triggered peripheral blood mononuclear cells (PBMCs) was observed in the case of different minipigs as well as ill pigs [18,19,20,21,22]. Since PERV-C is in contrast to PERV-A and PERV-B not present in all pigs, we screened the Auckland Island pigs for PERV-C and selected PERV-C animals for further breeding. The absence of PERV-C will prevent a recombination with PERV-A.

2. Materials and Methods

2.1. Animals and cells

Auckland Island pigs were bred by the NZ Xeno Ltd, Auckland, New Zealand as described [4,5]. DNA and blood from two groups of animals were sent to Berlin and the PERV copy numbers were analyzed. Later, primary kidney cells from four animals (two male, two female) at an age of about 4 weeks were sent to Germany and the PERV copy numbers were analyzed. Somatic cell nuclear transfer (SCNT) was performed using in vitro matured oocytes which were collected from abattoir derived ovaries from German Landrace/Piétrain hybrid gilts. The SCNT was performed at the Chair for Molecular Animal Breeding and Biotechnology, and Center for Innovative Medical Models (CiMM) of the Ludwig Maximilian University Munich. Prior to SCNT, matured oocytes were washed in 0.25% trypsin medium to avoid transmission of pathogens from abattoir. The embryos were transferred laparoscopically to estrus-synchronized recipient sows. 7 piglets were obtained from 373 reconstructed embryos. All piglets were clinically healthy and developed normally. Blood samples taken from 2 weeks old piglets and the foster mothers were examined for the presence of PERV-C. In addition, DNA was obtained from ear punches of the animals of the F0 generation at age 416 days and from kidney cells of the F1 generation at age of 70 days and analyzed for the PERV copy number. As an internal standard the porcine embryonic kidney cell line PK15 (ACC 640, from the Leibniz Institute DSMZ German Collection of Microorganisms and Cell lines, Braunschweig, Germany) was used.

2.2. DNA isolation

DNA was extracted at XenoNZ and CiMM, DNA was isolated from blood using two DNA extraction methods: DNeasy Blood and Tissue kit (Qiagen GmbH, Hilden, Germany) or phenol chloroform extraction using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer′s instructions. DNA was quantified using a Qubit 3.0 Fluorometer (Thermo Fischer), and the 260 nm/280 nm ratio was determined using a NanoDrop ND-1000 (Thermo Fisher Scientific Inc., Worcester, MA, USA).

2.3. Droplet digital PCR

Droplet digital PCR (ddPCR) was performed according to the manufacturer's instructions (Bio-Rad, Hercules, CA, USA, [http://www.bio-rad.com/de-de/applications-technologies/droplet-digital-pcr-ddpcr-technology?ID=MDV31M4VY]) using a QX200 droplet generator and a QX100 droplet reader (Bio-Rad) [23,24]. Purified genomic DNA (100 ng genomic DNA) was digested with MseI (New England Biolabs, Ipswich, MA, USA) (20U) at 37°C for 1 hour, and the restriction enzyme was heat-inactivated. The DNA digest was diluted to 5-10 ng/μL for the ddPCR. The ddPCR mix consisted of 10 μL 2X ddPCR Master mix, 1.8 μL of each 10 μmol/L target primers (Table 1), 0.5 μL of each 10 μmol/L fluorescent probes (FAM/HEX) (Table 1), 2.5-10 ng digested DNA, and water to a total volume of 20 μL. Samples were placed into the QX200 Droplet Generator, which utilizes proprietary reagents and microfluidics to partition the samples into 10,000-20,000 nanoliter-sized droplets. Droplets were transferred to a 96-well plate for PCR amplification in an Eppendorf Mastercycler X50 (Eppendorf, Germany). The following cycling conditions were used: 10-minute initial enzyme activation at 95°C, 40x [30 second denaturation at 94°C, 30 seconds annealing] followed by an 10 minute final elongation step at 98°C. Following PCR amplification, the samples were placed in the QX200 Droplet Reader, which analyzes each droplet individually using a two-color detection system (set to detect FAM and either HEX), enabling multiplexed analysis. The droplet reader and the QuantaSoft software count the PCR-positive and PCR-negative droplets. The fraction of positive droplets is then fitted to a Poisson distribution to determine the absolute initial copy number of the target DNA molecule in the input reaction mixture in units of copies/µL

2.3. Spleen and liver tissues from non-Auckland Island pigs

Frozen samples of spleen and liver tissues from dead born (day 0), 9-30 days old and older than 12 weeks piglets from the CiMM, Munich, were used. DNA was isolated as described above and the PERV copy number was measured in both organs.

2.4. Tests for PERV-C

DNA from the Auckland Island pigs were used to analyze the presence of PERV-C using a conventional PCR. Primers and conditions used for were described (PCR1) [25,26] (Table 1). In addition, a real-time PCR was established using specific primers and probe (Table 1) [26]. 100 ng DNA and the SensiFAST Probe No-ROX kit (Meridian Bioscience Cincinnati, OH, USA) in a 20 µL reaction volume. The cycling conditions used were initial denaturation 5 minutes at 95°C, followed by 45 amplification cycles of 95°C for 15 seconds, and annealing at 58°C for 30 seconds, and extension at 72°C for 30 seconds in a qTOWER3 G qPCR cycler (Analytik Jena, Jena, Germany). A standard curve was produced using the 510 bp amplicon of PCR6 [26] (Table 1) as template. This PCR was performed as follows: 100 ng of DNA template (PERV-C positive pig #6249), PCR buffer I containing MgCl₂, 0.2 mM dNTPs, and 1 unit of AmpliTaq DNA polymerase (Applied Biosystems, Inc., Waltham, MA, USA). Each sample was subjected to an initial

Table 1. Primers and probes used for the estimation of the PERV copy number and the detection of PERV-C.

Name	Sequence	Location (nucleotid number)	Accession number	Reference
PERV pol1-forward	CGACTGCCCCAAGGGTTCAA	3568-3587	HM159246	Yang et al., 2015 [27]
PERV pol2-reverse	TCTCTCCTGCAAATCTGGGCC	3803-3783
PERV pol probe	/56FAM/CACGTACTGGAGGAGGGTCACCTG	3678-3655
Pig actin forward	TAACCGATCCTTTCAAGCATTT			Krüger et al., 2020 [23]
Pig actin reverse	TGGTTTCAAAGCTTGCATCATA
Pig actin probe	/5HEX/CGTGGGGATGCTTCCTGAGAAAG
Pig GAPDH forward Pig GAPDH reverse	CCGCGATCTAATGTTCTCTTTC TTCACTCCGACCTTCACCAT	3951-3970 4022-4001	NC_010447.5 (396823)	Krüger et al., 2020 [23]
Pig GAPDH probe	/5HEX/CAGCCGCGTCCCTGAGACAC	3991-3972
PCR1* PERV-C forward PERV-C reverse	CTGACCTGGATTAGAACTGG ATGTTAGAGGATGGTCCTGG	6606-6625 6867-6886	AM229312	Takeuchi et al. [25], Kaulitz et al. [26]
PCR6* PERV-C 2 forward PERV-C 2 reverse	CCAGGACCACCAAATAATGG AAGTTTTGCCCCCATTTTAGT	6435-6454 6924-6944		Kaulitz et al. [26]
Real-time PCR* PERV-C 3 forward PERV-C 3 reverse PERV-C probe	CCCCAACCCAAGGACCAG AAGTTTTGCCCCCATTTTAGT FAM-CTCTAACATAACTTCTGGATCAGACCC- BHQ1	6853-6870 6924-6944 6878-6904

*Designation as in Kaulitz et al. [26] denaturation of 10 min at 95°C, followed by 45 amplification cycles (95°C for 15 s, 58°C for 30 s, and 72°C for 40 s) and a final extension at 72°C for 5 min The sensitivity of the real-time PCR was 10 copy (Supplemental Figure 1).

Table 1. Primers and probes used for the estimation of the PERV copy number and the detection of PERV-C.

Name	Sequence	Location (nucleotid number)	Accession number	Reference
PERV pol1-forward	CGACTGCCCCAAGGGTTCAA	3568-3587	HM159246	Yang et al., 2015 [27]
PERV pol2-reverse	TCTCTCCTGCAAATCTGGGCC	3803-3783
PERV pol probe	/56FAM/CACGTACTGGAGGAGGGTCACCTG	3678-3655
Pig actin forward	TAACCGATCCTTTCAAGCATTT			Krüger et al., 2020 [23]
Pig actin reverse	TGGTTTCAAAGCTTGCATCATA
Pig actin probe	/5HEX/CGTGGGGATGCTTCCTGAGAAAG
Pig GAPDH forward Pig GAPDH reverse	CCGCGATCTAATGTTCTCTTTC TTCACTCCGACCTTCACCAT	3951-3970 4022-4001	NC_010447.5 (396823)	Krüger et al., 2020 [23]
Pig GAPDH probe	/5HEX/CAGCCGCGTCCCTGAGACAC	3991-3972
PCR1* PERV-C forward PERV-C reverse	CTGACCTGGATTAGAACTGG ATGTTAGAGGATGGTCCTGG	6606-6625 6867-6886	AM229312	Takeuchi et al. [25], Kaulitz et al. [26]
PCR6* PERV-C 2 forward PERV-C 2 reverse	CCAGGACCACCAAATAATGG AAGTTTTGCCCCCATTTTAGT	6435-6454 6924-6944		Kaulitz et al. [26]
Real-time PCR* PERV-C 3 forward PERV-C 3 reverse PERV-C probe	CCCCAACCCAAGGACCAG AAGTTTTGCCCCCATTTTAGT FAM-CTCTAACATAACTTCTGGATCAGACCC- BHQ1	6853-6870 6924-6944 6878-6904

*Designation as in Kaulitz et al. [26] denaturation of 10 min at 95°C, followed by 45 amplification cycles (95°C for 15 s, 58°C for 30 s, and 72°C for 40 s) and a final extension at 72°C for 5 min The sensitivity of the real-time PCR was 10 copy (Supplemental Figure 1).

2.5. Tests for additional porcine viruses

DNA, RNA and blood from the Auckland Island pigs were used to analyze the presence of different DNA viruses and the RNA virus HEV. DNA was isolated as described above, and RNA was isolated using the RNeasy Mini kit (Qiagen, Hilden, Germany). Primers and condition used for the PCR or real-time PCR assays have been described for PCMV/PRV [28,29], PCV2 [30], PCV3 [31], PLHV-1, PLHV-2, and PLHV-3 [32] as well as for HEV [32]. The Western blot and ELISA to detect antibodies against HEV using recombinant proteins were described as well [32].

3. Results

3.1. PERV copy number in the genome of adult Auckland Island pigs

To determine the copy number of PERVs, a ddPCR was developed which used primers binding to a highly conserved region of the polymerase gene (pol) of PERV [23,24]. DNA from 30 Auckland Island pigs, bred in New Zealand, were analyzed using the ddPCR and a PERVpol copy number of around 65 was measured (Figure 1). This copy number was similar to the copy number of Göttingen minipigs and slightly lower compared with Aachen minipigs, investigated in parallel at the same time [24]. Since the copy number of Auckland Island pigs were determined in blood cells, the copy number of Aachen minipigs in spleen tissue, and the copy number of Göttingen minipigs in kidney tissue, it is difficult to compare these figures because the PERV copy number may differ in dependence of the organ (see below and [23,24,33]). Furthermore, the cellular composition of the different tissues was different and all tissues contained blood cells. DNA from kidneys of Göttingen minipigs revealed between 45 and 78 PERVpol copies using GAPDH as a reference and in a second study analyzing 16 Göttingen minipigs a median number of 64 (45-93) was found [24]. When DNA samples from Aachen minipigs were analyzed, a median number of 69 (34-97) was determined [24].

At a later point in time, DNA from another 14 Auckland Island pigs, also bred in New Zealand, were analyzed and copy numbers between 45 and 63 were estimated (Figure 2). This experiment confirmed the previous results shown in Figure 1, but showed a broader range of copy numbers.

3.2. Selection of PERV-C negative animals

When the 14 animals tested for the PERV copy number above (Figure 2) and another pig (494) were tested using a conventional PCR, eight out of 15 animals (53%) were PERV-C negative (Table 2). These investigations had the goal to select four PERV-C-free animals from which kidney cells were obtained, which could later be used for the SCNT. PERV-C-negative cells were used for the SCNT (Table 2) and consequentially all cloned Auckland Island piglets were negative for PERV-C in the germline as tested by PCR. It is important to note that the foster mothers were positive for PERV-C and the oocytes were of unknown PERV-C status, most likely PERV-C positive. Therefore, PERV-C was neither transmitted through recipient ooplasm during SCNT procedure nor by the embryo transfer recipient sows.

When Auckland Island pigs of the F0 and F1 generation as well as of later generations were tested by conventional PCR, all animals were negative (not shown). To confirm this, a real-time PCR was established with a limit of detection of 10 copies (Supplementary Figure 1) and all tested animals were PCR negative (no ct, not detected).

3.3. PERV copy number in the genome of cell lines from very young piglets

DNA from four kidney cell lines from young piglets four weeks old which had been selected as PERV-C negative were analyzed for the PERV copy number. A very low copy number of 20-22 copies was found (Figure 3).

3.4. PERV copy number in the genome of Auckland Island pigs obtained by SCNT

PERV-C-negative kidney cells from four weeks old piglets were used for an SCNT into in vitro matured oocytes collected from abattoir derived ovaries from German Landrace/Piétrain hybrid gilts. Ear punches from the animals of the F0 generation at age 416 days generation and kidney cells of the F1 generation at age of 70 days were analyzed for the PERVpol copy number. The copy numbers of the individual pigs ranged from 32 to 57 (Figure 3), but were lower compared with the numbers of the Auckland Island pigs analyzed in the past (Figure 1, Figure 2), and lower than other pig breeds such as the Göttingen minipigs, and the Aachen minipigs [24]. However, since the copy numbers were determined in tissues of different organs, a direct comparison is not possible. PK15 cells were used as internal control in order to be able to compare the results obtained here with previous results measuring PERV copy numbers. Since different copy numbers of integrated PERV were found in different PK15 cell lines [24], here the PK15 cells ordered from the Leibniz Institute DSMZ German Collection of Microorganisms and Cell lines, Braunschweig, Germany (ACC 640), were used and the measured copy number was 32.12 +/- 1.7 (Figure 3), which was in agreement with our previous findings [23,24].

3.5. Increase of the PERV copy number with age in non-Auckland Island pigs

In order to obtain additional evidence for the increase of the PERV copy number with age, the copy numbers were determined in two different organs of dead born (day 0), 9-30 days old and older than 12 weeks piglets from non-Auckland Island pigs (German landrace pigs), obtained from the CiMM, Munich. The copy number was determined in the spleen and liver and the difference between the copy number was depicted depending on the age of the animals (Figure 4). This way of presentation best illustrates the result: Differences in the copy number between spleen and liver were well described previously [24], and here it was shown that they increase with age. This indicates that in addition to the copy number in the germ line de novo integrations took place in one or the other organ. The de novo integrations were not tissue-specific, in some animals the higher copy number was in the spleen, in others in the liver. The difference was larger in older piglets, indicating that the copy number increased in dependence on the age of the animals (Figure 4).

3.6. Further virological characterization of the Auckland Island pigs

In order to characterize the Auckland Island pigs further, PCR-based methods for the detection of PERV-C, as well as the DNA genome of PCMV/PRV, PCV1, PCV2, PCV3, PLHV-1, PLHV-2, and PLHV-3 were performed (Table 2). In addition, a reverse transcriptase (RT) PCR for the RNA genome of HEV were performed (Table 2). The tests for all these viruses were negative, with exception of animal 476. In this animal PLHV-3 was found by PCR. Negative was a Western blot assay and an ELISA screening for antibodies against HEV, confirming that these animals are free of HEV as shown also by PCR. In addition to the Auckland Island pigs, the four kidney cell lines from young piglets four weeks old were tested in addition to PERV-C also for PCMV/PRV, and PCV3 and were found negative (Table 2).

4. Discussion

For the first time, a determination of the number of PERV copies in Auckland Island pigs was performed using the ddPCR method. The Auckland Island pigs had been used in the first xenotransplantation trials in New Zealand and Argentina [7,8], they are virologically well characterized [4,5,6] and are the “cleanest” pigs compared with other pig breeds. Theoretically endogenous retroviruses behave like cellular genes, they are fixed and should be identical in all tissues and cells. However, since PERVs are still active [33] and de novo integrations in somatic cells of different organs were observed leading to different copy numbers in different organs (Figure 4), it remains unclear whether all copies are already integrated.

The animals were free of eight porcine viruses, only in one animals PLHV-3 was detected. PLHV-3 does not harm pigs [34]. To note, the closely related PLHV-1, and -2 were not transmitted in preclinical trials transplanting pig hearts into baboons despite the viruses were detected in the donor pigs [9].

In the past it had been shown that the PERV copy number differs in dependence on the pig strain, on the individual animal tested, on the tissue tested, and on the method used (for review see [33]). Auckland Island pigs had been reported previously to contain three to 37 copies using a real-time PCR assay and a light cycler [5] and four to 40 copies were found using a PCR-based limited dilution assay (PLDA) [7]. In another study, relative gene dosages of PERV in Auckland Island pigs ranged from 2.8 ± 0.1 to 68 ± 1.6 at an average of 17.6 ± 10.4 using a real-time PCR [35]. This indicates a generally lower copy number compared to the numbers found here, but some animals had even a higher copy number, up to 68 copies. This was higher than the copy number found in this study. In our study we found that the PERV copy number is much lower in cells from very young piglets used for the SCNT (20-22 copies) compared with the copy number in adult animals (ranging from 32 to 59 copies in the animals obtained from these cells by SCNT) (Figure 3). Unfortunately, in most studies the age of the animals was unknown.

This data confirms that the copy number depends on the method used and probably on the age of the animals. Due to de novo integrations the number of integrated PERV proviruses differs from organ to organ (for review of numerous publications see [24]) and even in different parts of a single organ [23] in living pigs. It is possible that not all detected copies are integrated. The finding that the copy number is lower in cells from very young animals correlates with findings in another study: When analyzing 14-21-day-old neonatal and 1-2-year-old adult Belgian landrace pigs a significant increase in the copy number over time was described [36]. In these animals, a difference between least five copies and 80 copies was reported, indicating that the viruses are active and replicating in the neonatal pig. In Figure 5 a schematic presentation of these events is given. Indicated are the copy numbers found in the somatic cells of the animals and in the cells from young animals used for SCNT measured by ddPCR. Considering that copy numbers in adult animals are the result of de novo integrations in somatic cells, it logically follows that the number of proviruses in the germline can be at most 20 (Figure 5).

The fact that the number of integrated proviruses increased by age was also supported by a comparison of the copy number in two different tissues, spleen and liver, of non-Auckland Island pigs (Figure 4). Differences in the number of integrated copies means that in one or the other organ additional copies appeared in addition to the copy number in the germ line. The fact that the difference increases with age indicates that more new copies were integrated with age.

As shown above, the differences in the copy number of PERV in Auckland Island pigs when measuring by different methods ([5,7,35] and this paper) did not only depend on the method. Based on our findings the copy number increases due to de novo integrations and it depends on the age of the animal which was not always known to us and therefore cannot be taken into account. It would be highly interesting to test the PERV copy number in organs of a single animal over time. Furthermore, the PERV copy number may also depend on the type of tissue and the amount of PBMCs present in this tissue. The probability that de novo integrated copies can be found in the PBMCs is high since these cells are proliferating, a prerequisite for integration of gammaretroviruses. PBMCs are prone for release of different PERVs [37] and de novo integration evidenced by the presence of integration and release of infectious recombinant PERV-A/C from PBMCs [18,20,21,37]. PERV-A/C have never been found in the germ line of pigs [38]. Future investigations using separated cell populations or in situ hybridization experiments may answer this question. Furthermore, treatment of pigs during ontogenesis with antiretroviral drugs may prevent amplification of PERVs.

Auckland Island pigs are descendants of Western pigs brought to the island [1,2]. This was confirmed by phylogenetic analyses of microsatellites and mitochondrial analyses showing that they are more closely related to European pigs than Chinese pigs [3]. The PERV copy number in Western pigs is higher compared with Chinese pigs. By analyzing 63 high-quality porcine whole-genome resequencing data, it was found that the PERV copy numbers in Chinese pigs were lower (32.0±4.0) than in Western pigs (49.1±6.5) [39].

From all methods ever used to count PERV copies in the DNA of pigs, such as Southern blot [40,41], PCR titration [42], real-time PCR [43,44,45,46,47,48], fluorescence in situ hybridization (FISH) [49], PCR-based limited dilution assay [5], and genome wide sequencing [50], the ddPCR seems to be the most accurate method [33]. For example, using Southern blot analysis about 50 PERV copies were found in certain pig breeds [40,41], but PCR titration [42] or real-time PCR [43,44,45,46,47,48] detected from one to 98 copies. By using FISH, 19 copies of PERV-A and 13 copies of PERV-B were found in Westran pigs [49]. Using genome-wide sequencing, 20 gammaretrovirus copies and 4 betaretrovirus copies were detected in Duroc pigs [50], indicating that sequencing does not detect all integrated viruses.

DdPCR was found to have numerous favorable features in comparison with all other methods including real-time PCR (for review see [33]). One major advantage is that reference standard curves are not required, and bias effects arising from amplification efficiency and PCR inhibitors were reduced. The ddPCR technique provides absolute quantification based on the principles of sample partitioning and Poisson statistics, thus overcoming the normalization and calibrator issues; it has shown increased precision and sensitivity; it is relatively insensitive to PCR inhibitors and directly provides the result of the analysis expressed as number of copies of target per microliter of reaction [51].

Using ddPCR, fifty-nine copies were found when analyzing kidney fibroblasts from multi-transgenic pigs expressing the human transgenes CD46, CD55, CD59, HO1, and A20 [24]. When wild boars in and around Berlin were analyzed using droplet digital PCR, three to 69 PERV copies were found. A lower copy number, but a higher variability of this number was found in wild boars compared to domestic pigs, including minipigs [23].

All PCR-based methods used to estimate the PERV copy number have a common limitation, in that they used primers mostly for the highly conserved pol gene. This allows to detect all PERVs, i.e. PERV-A, PERV-B and PERV-C. However, the detection of this sequence does not mean that there is an intact provirus able to produce infectious virus particles. Therefore, the estimated number of integrated proviruses measured with PCR-based methods tells us nothing concerning the safety of xenotransplantation, e.g., the risk posed by PERVs [52]. Only sequencing the pig genome could give more reliable information. However, sequencing of repetitive, nearly identical sequences is always associated with problems as has been seen with hidden sequences of the human endogenous retrovirus HERV-K [53,54]. In these investigations, in addition to 91 full-length HERV-K proviruses previously found in the human genome draft sequence, hundreds more proviruses were detected. Sequencing the genome of a domestic Duroc pig, 212 endogenous retrovirus sequences were detected, among them 9 PERV-A, 11 PERV-B and 4 betaretrovirus copies [49], a lower number of PERVs compared with other methods [33].

One major result of the work was the elimination of PERV-C. Among the Auckland Island pigs analyzed in the first experiment approximately 53% of the animals were PERV-C-negative (Table 2). PERV-C is an ecotropic virus, infecting only pig cells and therefore does not pose a direct risk for xenotransplantation. However, recombinants between the human-tropic PERV-A and PERV-C have been observed, which are able to infect human cells and which are characterized by higher replication rates [15,17,55]. At present, PERV-A/C were found mainly in minipigs, either released from their PBMCs [18,20,21,38] or present in some organs of minipigs with melanomas [56] or even apparently healthy pigs [57]. Miniature swine may be a unique pig breed as they possess more copies of PERV-C sequence (typically five to nine copies) than many other pig herds (one to three copies [41]), suggesting that PERV-C may be more active in miniature swine [19]. One exception was the finding of PERV-A/C in large farm animals with infectious diseases [22]. However, in all cases PERV-A/C was only found in somatic cells, never in the germ line [58,59,60].

Due to our selection, there is no risk of PERV-A/C release from Auckland Island pigs generated by SCNT using nuclei from PERV-C-free Auckland Island pigs. These PERV-C free pigs will be a valuable foundation for additional genetic modification in terms of future clinical xenotransplantation.

5. Conclusions

Auckland Island pigs had been used repeatedly as donors for pig islet cell transplantation into non-human primates and diabetic patients. In all preclinical and clinical trials no porcine viruses including PERV had been transmitted.

Auckland Island pigs are well characterized, they are free of many potentially pathogenic viruses and have a PERV copy number comparable to other Western pig breeds. The copy number is much lower in cells from very young piglets and in animals obtained by SCNT using nuclei from Auckland Island pig cells. To enhance the virus safety, PERV-C negative animals were selected. In addition, it was shown that eight other xenotransplantation-relevant pig viruses were absent in these animals. Auckland Island pigs have numerous advantages compared with other pig breeds including the size, the health status, the absence of many known viruses and may be, due to the long isolation, also the absence of still unknown potentially xenozoonotic microorganisms.