Absence of Replication of Porcine Endogenous Retrovirus and Porcine Lymphotropic Herpesvirus Type 1 with Prolonged Pig Cell Microchimerism after Pig-to-Baboon Xenotransplantation

Abstract

Porcine endogenous retrovirus (PERV), porcine cytomegalovirus (PCMV), and porcine lymphotropic herpesvirus (PLHV) are common porcine viruses that may be activated with immunosuppression for xenotransplantation. Studies of viral replication or transmission are possible due to prolonged survival of xenografts in baboon recipients from human decay-accelerating factor transgenic or α-1,3-galactosyltransferase gene knockout miniature swine. Ten baboons underwent xenotransplantation with transgenic pig organs. Graft survival was 32 to 179 days. Recipient serial samples of peripheral blood mononuclear cells (PBMC) and plasma were analyzed for PCMV, PERV, and PLHV-1 nucleic acids and viral replication using quantitative PCR assays. The PBMC contained PERV proviral DNA in 10 animals, PLHV-1 DNA in 6, and PCMV in 2. PERV RNA was not detected in any PBMC or serum samples. Plasma PLHV-1 DNA was detected in one animal. Pig cell microchimerism (pig major histocompatibility complex class I and pig mitochondrial cytochrome c oxidase subunit II sequences) was present in all recipients with detectable PERV or PLHV-1 (85.5%). Productive infection of PERV or PLHV-1 could not be demonstrated. The PLHV-1 viral load did not increase in serum over time, despite prolonged graft survival and pig cell microchimerism. There was no association of viral loads with the nature of exogenous immune suppression. In conclusion, PERV provirus and PLHV-1 DNA were detected in baboons following porcine xenotransplantation. Viral detection appeared to be due to persistent pig cell microchimerism. There was no evidence of productive infection in recipient baboons for up to 6 months of xenograft function.

Xenotransplantation using porcine organs has the potential to provide an unlimited supply of organs for the treatment of organ failure or as a “bridge” to allotransplantation. The development of the α-1,3-galactosyltransferase gene knockout (GalT-KO) miniature swine and other genetically modified donor animals has reduced immunologic barriers to xenotransplantation (21, 51). However, concerns persist regarding the potential risk of interspecies transmission of infectious pathogens with xenografts when clinical trials are initiated (13). Porcine cytomegalovirus (PCMV) and porcine lymphotropic herpesvirus (PLHV) are pathogens of swine that undergo accelerated viral replication and cause clinical syndromes in immunosuppressed xenograft recipients (3, 11, 15, 28, 29, 31). PCMV is associated with consumptive coagulopathy in baboons after porcine xenotransplantation, and PLHV, a gamma herpesvirus, has been associated with a form of posttransplant lymphoma in immunosuppressed swine after allogeneic hematopoietic stem cell or splenic transplantation (4, 10, 14, 17, 28, 31). Porcine endogenous retroviruses (PERV) are members of a ubiquitous family of proviral elements of swine, which are capable of replicating in human cells, and thus carry the theoretical risk of transmission to humans in the setting of xenotransplantation (12, 36). PERV-A and -B and recombinant PERV-AC have been shown to infect human and pig cells in vitro, while PERV-C is restricted largely to porcine cells (22, 24, 33, 37, 41, 42, 44, 50). No productive infections due to PERV in vivo have been described for humans exposed to porcine tissues or for nonhuman primates following xenotransplantation (6-9, 16, 34, 35, 38, 41). The transmission and detection of PERV proviral DNA attributed to pig cell microchimerism have been described for guinea pigs (1), mice (20), and baboons (27, 48) but without evidence of viral replication in host cells. Reported PERV infection of human cells in mice was shown to be due to the pseudotyping of PERV by endogenous murine retroviruses rather than direct viral infection (52). In this study we examine baboons for evidence of the replication of PCMV, PERV, or PLHV-1 during prolonged posttransplant graft survival after xenotransplantation from inbred miniature swine with the GalT-KO modification or expressing human decay-accelerating factor (hDAF).

MATERIALS AND METHODS

Animals.

Large white/Landrace crossbreed pigs transgenic for hDAF (n = 4) (Novartis Pharmaceuticals/Harlan, Madison, WI) or Massachusetts General Hospital miniature swine homozygous for GalT-KO (n = 6) were used as donors. Baboons (Papio anubis) (n = 10), weighing between 8 and 20 kg were purchased from Biological Resources Foundation (Houston, TX) or Mannheimer Foundation (Homestead, FL). The care of animals was in accordance with the Guide for the Care and Use of Laboratory Animals prepared by the National Academy of Sciences and published by the National Institutes of Health (32a). Animal protocols were approved by the Massachusetts General Hospital Subcommittee on Research Animal Care.

Ten baboons underwent xenotransplantation with hDAF or GalT-KO transgenic pig organs at the Massachusetts General Hospital Transplant Biology Research Center in Boston, MA, from September 2002 until November 2007 and were followed longitudinally until death or euthanasia. Xenografts included vascularized thymic lobes, kidneys, hearts, and composite thymokidney grafts, with some variation in the immunosuppressive regimens applied. The characteristics of the baboon recipients, the nature of immunosuppression, and the xenografts are described in Table 1. All baboons received 5 mg/kg/day ganciclovir intravenously for baboon cytomegalovirus (BCMV) prophylaxis. Serially collected samples of plasma and peripheral blood mononuclear cells (PBMC) from all 10 baboons were studied. The number of specimens analyzed from each animal ranged from 7 to 19, with an average of 12 per animal. Time intervals between specimens from each animal ranged from 3 to 21 days (Table 2).

TABLE 1.

Characteristics of the baboon recipients, xenografts, immunosuppressive regimens, and graft survival

Baboon	Xenograft	Donor	Immunosuppressiona	Graft survival (days)b	Time to euthanasia or death (days)	Outcomec
B107	Thymocyte infusion	hDAF	MMF + FK506	N/A	92	Anti-donor response day 36
B109	Vascularized thymic lobe	hDAF	FN18-CRM9 + LoCD2b + steroids + anti-CD154 MAb + MMF + FK506 + CVF	>64*	103	Donor-specific unresponsiveness POD 64
B110	Vascularized thymic lobe	hDAF	FN18-CRM9 + LoCD2b + steroids + anti-CD154 MAb + MMF + FK506 + CVF	>70*	104	Donor-specific unresponsiveness day 70; regained anti-donor response day 97
B112	Vascularized thymic lobe	hDAF	Horse ATG + cyclophosphamide + LoCD2b + steroids + anti-CD154 MAb + MMF + FK506 + CVF	34	34	Klebsiella pneumonia
B113	Vascularized thymic lobe and kidney	GalT-KO	Horse ATG + LoCD2b + MMF + anti-CD154 MAb + FK506 + CVF + steroid induction	>68	68	Died with functioning graft and anesthetic complication
B114	Kidney	GalT-KO	700-rad TI + horse ATG and MMF + anti-CD154 MAb + FK 506 + CVF + steroid induction	33	33	Kidney rejection
B194	Thymokidney	GalT-KO	Rabbit ATG + MMF + anti-CD154 MAb + FK 506 + CVF	>49	49	Died with functioning graft and pulmonary hemorrhage
B216	Heterotopic heart	GalT-KO	700-rad TI + horse ATG + LoCD2b + anti-CD154 MAb + CVF + MMF + steroid induction and then tapered	>56	56	Died from intra-abdominal hemorrhage with beating heart
B223	Heterotopic heart	GalT-KO	700-rad TI + horse ATG + LoCD2b + anti-CD154 MAb + MMF + steroid induction and then tapered	110	110	Graftectomy POD 110 with development of anti-pig Ab
B228	Heterotopic heart	GalT-KO	Horse ATG + LoCD2b + anti-CD154 MAb + CVF + MMF + steroid induction and then tapered	179	179	Graftectomy POD 179 with rejection

^aFN18-CRM9, anti-CD3 monoclonal antibody linked to a mutated diphtheria toxin; LoCD2b, anti-CD2 monoclonal antibody; MMF, mycophenolate mofetil; ATG, antithymocyte globulin; CVF, cobra venom factor; TI, thymic irradiation; MAb, monoclonal antibody; FK506, tacrolimus.

^b*, minimum graft survival based on limited biopsies of thymic grafts; N/A, not applicable.

^cPOD, postoperative day; Ab, antibody.

TABLE 2.

Details of blood sampling

Baboon	Total no. of samples	Bleed dates after transplantation (POD)a
B107	10	25, 28, 32, 35, 39, 47, 53, 66, 80, 90
B109	14	2, 5, 8, 11, 14, 18, 21, 25, 28, 32, 39, 42, 46, 53
B110	14	1, 4, 7, 11, 18, 21, 25, 32, 39, 46, 49, 52, 59, 62
B112	9	1, 5, 8, 12, 19, 21, 25, 28, 32
B113	19	−2, 1, 5, 8, 14, 16, 17, 21, 24, 28, 31, 35, 38, 42, 44, 48, 51, 55, 58
B114	10	0, 4, 7, 10, 13, 17, 20, 24, 27, 32
B194	7	8, 12, 19, 26, 34, 40, 46
B216	12	2, 5, 7, 12, 13, 14, 18, 21, 28, 33, 46, 53
B223	11	−1, 6, 13, 25, 49, 96, 115, 122, 129, 150, 157
B228	11	81, 87, 94, 103, 108, 115, 132, 146, 159, 173, 179

^aPOD, postoperative days.

PERV proviral DNA and PLHV-1 DNA detection by PCR in PBMC of xenograft recipients.

DNA extraction from PBMC and plasma was performed using the Puregene DNA isolation kit (Gentra Systems, Minneapolis, MN) per the manufacturer's instructions. The PERV polymerase gene (PERV pol) and PLHV-1 polymerase gene (PLHV-1 pol) were amplified and detected using a quantitative PCR TaqMan assay carried out on the Stratagene Mx Pro 3000P multiplex quantitative PCR system (Stratagene, LaJolla, CA). The PERV polymerase gene-specific primers used were 5′-AGCTCCGGGAGGCCTACTC-3′ (forward) and 5′-ACAGCCGTTGGTGTGGTCA-3′ (reverse) (34). The PERV polymerase TaqMan probe used was 6-carboxyfluorescein (FAM)-CCACCGTGCAGGAAACCTCGAGACT-6-carboxytetramethylrhodamine (TAMRA). The PLHV-1 polymerase gene-specific forward primer was derived from GenBank (http://www.ncbi.nlm.nih.gov/), accession number AF118399.1, and was 5′-TCAGAAAGGAATTAGCAGCAT-3′ (forward) and 5′-TGCAATCTTGAGACAGGGCA-3′ (reverse) (31). The PLHV-1 polymerase TaqMan probe was FAM-TGGGTTCACTGGTGTTGCATCTGGTATG-TAMRA (31). The PERV gag gene was also amplified to confirm intact PERV by using the same quantitative TaqMan PCR assay protocol described above. The PERV gag-specific primers and probe used were derived from GenBank accession number AY056035.1. The forward primer was 5′-TTGGAAAACTAACCATCCCCC-3′, and the reverse primer was 5′-AAGGGACTCCACCAACCCC-3′. The PERV gag TaqMan probe was FAM-TCTCGGAGGATCCCCAACGCCT-TAMRA.

PCR was carried out in a mixture containing a final concentration of 0.8 μM PERV pol or PLHV-1 pol forward and reverse primers, 0.2 μM PERV pol or PLHV-1 TaqMan probe, 1× mM Stratagene Brilliant QPCR master mix (Stratagene, Cedar Creek, TX), 0.03 μM Rox dye, and 200 ng of DNA over 40 cycles (95°C for 30 s, 60°C for 30 s, and 72°C for 30 s). The quantitative detection limit for this assay was 5 to 10 DNA copies/200 ng of total cellular DNA. Positive and negative (water) controls were used in each PCR. Positive controls were generated from PERV-infected PK15 cells and PLHV-1-infected lymph node cells from an animal with PLHV-1-associated posttransplantation lymphoproliferative disorder. PCR results were confirmed by running 10 μl of PCR products on a 2% agarose gel with ethidium bromide staining. Representative products were recovered from agarose, cloned using a Topo XL PCR cloning kit (Invitrogen, Carlsbad, CA) and sequenced at the Massachusetts General Hospital DNA Core Sequencing laboratory.

PCR detection of BCMV and PCMV in PBMC of xenograft recipients.

DNA from PBMC was amplified and studied for the presence of BCMV and PCMV. The TaqMan PCR assay protocol described previously was used with the following primers and probes: BCMV forward primer 5′-GTTTAGGGAACCGCCATTCTG-3′ and reverse primer 5′-GTATCCGCGTTCCAATGCA-3′ (28). The BCMV probe was FAM-TCCAGCCTCCATAGCCGGGAAGG-TAMRA (28). The PCMV primers were forward primer 5′-GTTCTGGGATTCCGAGGTTG-3′ and reverse primer 5′-ACTTCGTCGCAGCTCATCTGA-3′ (28). The PCMV probe was FAM-CAGGGCGGCGGTCGAGCTC-TAMRA (28).

Detection of porcine nucleic acids in circulation (microchimerism).

The pig major histocompatibility complex class I (p-MHC-I) gene and the pig mitochondrial cytochrome c oxidase subunit II (p-mtCOII) gene were amplified from recipient PBMC DNA by quantitative PCR using the same protocol and platform described above, along with positive and negative controls. The p-MHC-I gene-specific primers were 5′-GCCCTGGGCTTCTACCCTAA-3′ (forward) and 5′-TCTCAGGGTGAGTGGCTCCT-3′) (reverse) (28). The p-MHC-I TaqMan probe was FAM-CCAGGACCAGAGCCAGGACATGGAGCTCGT-TAMRA (28). The p-mtCOII gene-specific primers and probe were newly designed and were derived from GenBank accession number AJ002189.1 and were 5′-CGTATTCTTAATCAGCTCTTTAGTG-3′ (forward) and 5′-GCGGGTAGGATTGTTCAAATTGT-3′ (reverse). The p-mtCOII TaqMan probe was FAM-ACACACACTAGCACAATGGATGCCCAA-TAMRA. PCR results were confirmed by running 10 μl of PCR products on a 2% agarose gel with ethidium bromide staining with positive controls.

PCR detection of PLHV-1 in plasma of xenograft recipients.

DNA was extracted from all baboon plasma samples (50 μl) using the Puregene DNA isolation kit (Gentra Systems, Minneapolis, MN) according to the manufacturer's instructions. Extracted DNA was amplified and detected using the same quantitative TaqMan PCR assay protocol and PLHV-1-specific forward and reverse primers and probe described above, with appropriate positive and negative (water) controls.

RT-PCR detection of PERV in plasma of xenograft recipients.

Total RNA was extracted from all baboon plasma samples (100 μl) using RNA STAT-60 total RNA/mRNA isolation reagent (Tel-Test, Inc., Friendswood, TX). Total RNA was precipitated using 100% isopropanol in the presence of linear acrylamide (20-μg/ml final concentration) and 300 mM (final concentration) of lithium chloride, washed with 75% ethanol, air-dried, and then resuspended in 100 μl RNase-free water. RNA (60 to 108 ng) was used in the two-step reverse transcription-PCR (RT-PCR) using the Qiagen Omniscript RT kit (Qiagen, Valencia, CA) according to the manufacturer's instructions. In the first step, RT-PCR was performed on total RNA extracted from baboon plasma and on RNA extracted from the PK15 cell line infected with PERV (positive control for the RT) using the following protocol: 37°C for 60 min and then 4°C for 5 min. In the second step, cDNA was amplified and detected using TaqMan quantitative PCR with PERV pol-specific primers and probe using the same platform and protocol described above, with appropriate positive and negative controls.

RT-PCR detection of PERV in PBMC of xenograft recipients.

Total RNA was extracted from all PBMC samples stored in RNA STAT-60 total RNA/mRNA isolation reagent (Tel-Test, Inc., Friendswood, TX) and per the RNA isolation protocol described above. The processing of specimens was biased toward detection. RNA (39.6 ng to 2,000 ng) was used in the two-step RT-PCR, and then cDNA amplification and detection were carried out using the TaqMan quantitative PCR protocol with the PERV pol primers and probe described previously.

RESULTS

The duration of the function of porcine xenografts ranged from 32 to 179 days (mean was 76 ± 53 days). Microchimerism, or circulating cells or cellular components from the xenografts as reflected by detection of p-mtCOII and p-MHC-I DNA, was detected in most samples throughout the postxenotransplant course. PERV proviral DNA and PLHV-1 DNA were detected in PBMC samples from xenograft recipients (Table 3). The lower limit of detection for each virus was 5 to 10 copies per 200 ng of cellular DNA. The identification of PERV DNA and PLHV-1 DNA in PBMC was confirmed by sequencing of the amplified reaction products. PERV proviral DNA and PLHV-1 DNA were demonstrable in recipient PBMC only in the presence of pig cell microchimerism. Viral loads in blood fluctuated over time (Fig. 1 and 2). The type or intensity of immunosuppression as measured by serum levels of mycophenolate mofetil (MMF) and tacrolimus did not correlate with the quantitative levels of PERV (Fig. 3A, B, C, and D) or BCMV (Fig. 4A and B).

TABLE 3.

Detection of PLHV-1 pol, PERV pol, p-MHC-I, p-mtCOII, and BCMV DNA in PBMC of baboon xenograft recipients^a

Baboon	PLHV-1 pol			PERV pol			p-MHC-I		p-mtCOII		BCMV
	Result	No. of positive samples/total no. of samples (%)	VL	Result	No. of positive samples/total no. of samples (%)	VL	Result	No. of positive samples/total no. of samples (%)	Result	No. of positive samples/total no. of samples (%)	Result	No. of positive samples/total no. of samples (%)
B107	ND			Pos	6/10 (60)	1,221 ± 3,583	Pos	2/10 (20)	Pos	8/10 (80)	Pos	7/10 (70)
B109	Pos	4/14 (28.6)	107 ± 267	Pos	11/14 (78.6)	14,239 ± 25,444	Pos	5/14 (35.7)	Pos	14/14 (100)	Pos	10/14 (71.4)
B110	Pos	3/14 (21.4)	2 ± 4	Pos	6/14 (42.9)	50 ± 123	Pos	3/14 (21.4)	Pos	9/14 (64.3)	Pos	7/14 (50)
B112	Pos	1/9 (11.1)	6 ± 17	Pos	5/9 (55.5)	5,742 ± 12,409	Pos	4/9 (44.4)	Pos	8/9 (88.8)	Pos	9/9 (100)
B113	Pos	3/19 (15.8)	0 ± 1	Pos	16/19 (84.2)	3,106 ± 12,147	Pos	15/19 (78.9)	Pos	17/19 (89.5)	Pos	8/19 (42.1)
B114	Pos	6/10 (60)	16 ± 17	Pos	8/10 (80)	22 ± 28	Pos	6/10 (60)	Pos	10/10 (100)	Pos	6/10 (60)
B194	ND			Pos	4/7 (57.1)	108 ± 175	Pos	6/7 (85.7)	Pos	6/7 (85.7)	Pos	3/7 (42.9)
B223	ND			Pos	5/11 (45.4)	18 ± 28	Pos	1/11 (9)	Pos	6/11 (54.5)	Pos	3/11 (27.3)
B228	ND			Pos	7/11 (63.6)	841 ± 1724	Pos	6/11 (54.5)	Pos	11/11 (100)	Pos	6/11 (54.5)

^aND, No DNA detected; Pos, positive; VL, mean viral load. Viral loads are expressed as the mean numbers of copies/200 ng of DNA ± standard deviations.

FIG. 1.

Viral loads of PLHV-1, PERV pol, MHC-I, mtCOII, and BCMV in baboon B113. POD, postoperative day.

FIG. 2.

Viral loads of PLHV-1, PERV pol, MHC-I, mtCOII, and BCMV in baboon B114. POD, postoperative day.

FIG. 3.

MMF levels, tacrolimus (FK) levels, and PERV pol viral loads in baboons. (A) FK (FKB113) levels and PERV pol (PERVB113) viral loads (number of copies/200 ng of DNA) (secondary y axis) in baboon B113. (B) PERV pol (PERVB113) viral loads (number of copies/200 ng of DNA) (primary y axis) and MMF (MMFB113) levels in baboon B113. (C) FK (FKB114) levels and PERV pol (PERVB114) viral loads in baboon B114. (D) PERV pol (PERVB114) viral loads (number of copies/200 ng of DNA) (primary y axis) and MMF (MMFB114) levels in baboon B114. POD, postoperative day.

FIG. 4.

PERV pol and BCMV viral loads and tacrolimus (FK506) levels in baboons. (A) PERV pol and BCMV viral loads (number of copies/200 ng of DNA) (primary y axis) and FK506 (secondary y axis) in baboon B109; (B) PERV pol and BCMV viral loads (number of copies/200 ng of DNA) (primary y axis) and FK506 (secondary y axis) in baboon B110. POD, postoperative days.

PERV RNA was not detected in any plasma or cellular samples from xenograft recipients at the time of blood sampling. PLHV-1 DNA was detected at a low titer (1 to 10 copies per 50 μl of plasma) in the plasma of one animal with detectable microchimerism. BCMV was detected in the PBMC of all recipients (1 to 38,610 copies/200 ng of DNA). The BCMV viral load increased over time (from 118 copies/200 ng of DNA at the beginning of immunosuppression to 38,610 copies/200 ng of DNA at the time of death) in only one animal. PCMV was detected at a low titer (1 to 374 copies/200 ng of DNA) in the PBMC of 2 out of 10 animals. The xenograft donors for these two baboons carrying PCMV were GalT-KO swine that were not derived by early weaning. P-mtCOII and p-MHC-I were detected in 85.5% and 49.6% of recipient PBMC, respectively. There was no statistical difference in the mean PERV viral loads in the group of baboon recipients of GalT-KO organs and the group of recipients of hDAF organs (mean viral loads of 5,774 copies/200 ng of DNA versus 21,251 copies/200 ng of DNA, respectively; P = 0.27).

DISCUSSION

This study was designed to seek evidence of viral transmission or productive viral infection with the improved survival of porcine xenografts in immunosuppressed baboon recipients, as well as in light of more recent observations of the biology of PERV and other porcine viruses. Persistent pig cell microchimerism has been demonstrated in patients treated with various living pig tissues for over 12 years without evidence of PERV infection (34). This study also showed persistent microchimerism up to 6 months posttransplantation. While the identity of these cells could not be determined in the present study, microchimerism is likely due to graft-derived porcine PBMC or endothelial cells, both of which are known to contain functional PERV proviral elements (5, 18, 45, 46). No active replication of either PERV or PLHV-1 could be detected in baboon recipients of a variety of porcine xenografts receiving a variety of intensive and immunosuppressive regimens.

Most previous studies of PERV infectivity were performed prior to the recognition of recombinant PERV-AC as a major species causing infection of human cells in vitro (49). In addition, the use of GalT-KO miniature swine as donors might increase the persistence of PERV in circulation. Natural, preformed antibodies in higher primates target the Gal epitope and provide a barrier against the transmission of PERV in vivo and in vitro (25, 26, 39). PERV from GalT-KO swine lack the Gal epitope and might be expected to persist longer in circulation in the face of reduced opsonization by natural antibodies. However, despite these potential risks, we were unable to detect PERV RNA in any samples of PBMC or serum from any xenograft recipients using pan-PERV (A, B, and C) probes that detect all strains of PERV as well as PERV A-C. In our study, there was no statistical difference in the mean PERV viral loads in baboon recipients of hDAF organs and in GalT-KO organ recipients. This likely reflects similarities in the numbers of active PERV loci in the hDAF and GalT-KO swine (J. Fishman and L. Scobie, preliminary data). Previous studies using the depletion of anti-α-Gal antibodies in 27 baboon recipients of hDAF or human membrane cofactor protein/hDAF transgenic pig organs, which were followed for over 2 months, also failed to show evidence of PERV replication (27). These data are consistent with the failure to demonstrate PERV replication in xenograft recipients (27, 48). Apparent PERV infection of human cells in a pig-to-SCID mouse model following islet cell transplantation was shown to be due to the pseudotyping of PERV by endogenous murine retroviruses rather than PERV infection of human cells (52). While active PERV infection cannot be completely excluded by such studies, viral replication has not been demonstrated in baboon recipients of porcine xenografts to date.

The use of nonhuman primate models, including baboons, to assess the risk of PERV transmission is controversial. While some studies have shown that cell lines derived from some primate species could not be infected by PERV (23, 37, 43, 47, 49), others have demonstrated that baboon B lymphocytes were permissive for PERV, although without productive infection (2, 43). Inefficient cell entry and replication (40) and expression of APOBEC3G protein (19) may play a role in rendering susceptible baboon cells relatively resistant to productive infection. Even in permissive human cells, the demonstration of productive infection in vitro is often technically difficult. Thus, it is important to continue to utilize all available preclinical models to assess the potential risk for xenograft-associated infections.

In the presence of likely pig microchimerism, PERV proviral DNA and PLHV-1 DNA were found in the circulations of baboon recipients of hDAF or GalT-KO transgenic pig organs. There was no evidence of productive infection with either virus. The DNA viral loads of PERV provirus and PLHV-1 fluctuated over time, possibly reflecting the variable shedding of porcine cells from the xenografts into the circulation. The quantitative level of PERV, PLHV-1, or BCMV DNA did not correlate with the nature or intensity of immunosuppression. This surprising result suggests that other factors, such as ongoing graft rejection, may impact viral replication and/or contribute to the variable release of pig cells from xenografts.

Human cytomegalovirus is a major pathogen in clinical allotransplantation. In immunosuppressed pigs and baboons, PCMV and BCMV, respectively, cause systemic infections and contribute to allograft injury (14, 28-30). PCMV was excluded from our miniature swine colony by early weaning of piglets (30). In this series, PCMV was detected only in two baboons after transplantation with grafts obtained from swine raised prior to the institution of an early weaning strategy for GalT-KO animals. PCMV has not been observed in xenograft donors or recipients subsequently. BCMV was, however, activated by the immunosuppression utilized in these studies, despite ganciclovir prophylaxis. BCMV has previously been shown to have reduced susceptibility to the ganciclovir used for prophylaxis (32).

In summary, PERV DNA and PLHV-1 DNA can be detected in the circulations of baboons following porcine xenotransplantation. However, this observation appears to be due to persistent pig cell microchimerism. There was no evidence of productive infection due to these viruses or to PCMV in recipient baboons. Further studies with prolonged follow-up are warranted with each strain of pig intended as a source species for clinical trials to continue to ensure the absence of infection associated with clinical xenotransplantation.