Notice: Wiley Online Library will be unavailable on Saturday 30th July 2016 from 08:00-11:00 BST / 03:00-06:00 EST / 15:00-18:00 SGT for essential maintenance. Apologies for the inconvenience.
This work was funded by the National Institutes of Health (grant RO1-DK56733), the Marriott Foundation, the Wallace H. Coulter Foundation, the American Society of Transplant Surgeons/Pfizer Collaborative Scientist Grant, and the American Society of Transplant Surgeons/National Kidney Foundation Folkert Belzer Award.
Address reprint requests to Scott L. Nyberg, M.D., Ph.D., Division of Transplant Surgery, Department of Surgery, Mayo Clinic, 200 First Street Southwest, Rochester, MN 55905. E-mail: email@example.com
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
There currently exists a high demand for an abundant, routinely available, high-quality source of human hepatocytes for both therapeutic and diagnostic applications. Current methodologies do not permit the growth and expansion of primary human hepatocytes; thus, new techniques are required to overcome this shortage.
To address this need, we have genetically engineered a porcine model of human hereditary tyrosinemia type 1 (HT1). HT1 is an autosomal-recessive inborn error of metabolism resulting from a deficiency in fumarylacetoacetate hydrolase (FAH), which is the catalyst of the last step in tyrosine metabolism. In mice and humans, FAH deficiency results in tyrosinemia, hepatic failure, cirrhosis, and hepatocellular carcinoma. Recent efforts by Azuma et al. have led to nearly complete (>90%) hepatocyte replacement in the livers of mutant (Fah−/−) mice. Because of the accumulation of the toxic substrate fumarylacetoacetate within mutant hepatocytes, healthy FAH+ cells have a strong selective growth advantage and can repopulate extensively. In a novel approach, we used the synthetic, chimeric adeno-associated virus DJ serotype to target and disrupt the porcine Fah gene by homologous recombination. To date, both Fah-null heterozygote and homozygote offspring have been produced.
The expansion of primary human hepatocytes in FAH-deficient mice occurs because native hepatocytes are metabolically defective and allow the selective growth of FAH+ human hepatocytes. In addition, the immune system of these mice has been altered genetically to prevent the rejection of transplanted human hepatocytes. However, we have not genetically altered the immune system of our FAH-deficient pigs.
Rather, to overcome the pig immune system, we have considered in utero cell transplantation (IUCT) leading to a state of immune hyporesponsiveness. The idea of acquired tolerance was first described by Billingham et al. in 1953 when they showed that a fetal chicken receiving blood in utero from another chicken could engraft skin from the blood-donating chicken. Since this first description, additional animal models have been developed and have demonstrated the utility of allogeneic IUCT, including the correction of hemoglobinopathies and some immunological disorders.[4-7] Additionally, others have shown similar results in the setting of the xenogeneic transplantation of human cells into fetal sheep. In the sheep model, IUCT of human embryonic stem cell–derived hematopoietic cells led to stable engraftment at low levels for more than a year after birth.
In utero transplantation and potential tolerization are based on the immunological immaturity of the early developing fetus, which leads to the possibility of donor- or species-specific tolerance of xenogeneic cells. In this study, we explored the possibility of producing a state of hyporesponsiveness in pigs to human hepatocytes via the transplantation of human hepatocytes into fetal pig livers. We have developed an IUCT procedure by which piglets are stably engrafted with human hepatocytes when they are first transplanted during early gestation before CD3+ lymphocyte maturation in the pig thymus.
MATERIALS AND METHODS
Animals and Animal Care
All animals received humane care in compliance with the regulations of the Institutional Animal Care and Use Committee at the Mayo School of Graduate Medical Education. Gilts were obtained from Midwest Research Swine (Gibbon, MN) and were acclimated for at least 1 week before IUCT. All gilts received 150 mg of medroxyprogesterone (Greenstone, Peapack, NJ) intramuscularly on gestational day 37. On gestational day 40, all gilts underwent general anesthesia and lower midline laparotomy. Both uterine horns were exposed. All fetuses in the right uterine horn received a direct intrahepatic injection under ultrasound guidance with a 1.5-inch 25-gauge needle (Fig. 1). The injections consisted of 300 μL of lactated Ringer's solution containing 1 × 107 human hepatocytes (Yecuris, Inc., Portland, OR). Later, from gestational day 100 to gestational day 115, gilts received 20 mg of altrenogest (Merck, Millsboro, DE) daily mixed with chow. On gestational day 114, surrogate sows of the same gestational age received 10 mg of Lutalyse (Pfizer, New York, NY) intramuscularly in the morning and 12 hours later. On gestational day 115, surrogate sows received 40 US Pharmacopeia units of oxytocin (VetTek, Blue Springs, MO) intramuscularly to induce labor, whereas gilts underwent cesarean section under general anesthesia. All live piglets were identified as right horn (hyporesponsive) or left horn (normoresponsive). Live piglets were then allowed to nurse from the surrogate sows for 3 weeks and were subsequently transitioned to a regular feed.
Priming to Assess Immune Responsiveness
Normoresponsive and hyporesponsive piglets were split into 2 groups: saline-primed and hepatocyte-primed. Saline-primed piglets had blood drawn, and they received subcutaneous (SQ) injections of normal saline (1 mL) at 2, 4, and 6 weeks of life. Hepatocyte-primed piglets had blood drawn, and they received 1 × 107 human hepatocytes in 1 mL of normal saline via SQ injection at 2, 4, and 6 weeks of life. Both groups had blood drawn at 8 weeks of life and were euthanized.
At 1 week of life, piglets received an ultrasound-guided percutaneous injection into the right lobe of the liver. Normoresponsive piglets received a percutaneous injection of 1 mL of normal saline or 5 × 107 human hepatocytes with 20 mg of tantalum dust (Bal-Tec, Los Angeles, CA) in 1 mL of normal saline. Hyporesponsive piglets received a percutaneous injection of 5 × 107 human hepatocytes in 1 mL of normal saline with 20 mg of tantalum dust. All piglets had blood drawn 2, 4, and 6 weeks after engraftment and were euthanized 6 weeks after engraftment. Livers were harvested from euthanized piglets. Samples were obtained from all livers. From piglets injected with tantalum, samples were obtained with X-ray localization at the site of liver injection.
Serum Cytotoxicity Assay
A Terasaki plate was lined with Roswell Park Memorial Institute 1640 (RPMI-1640)–moistened filter paper strips. Two-microliter aliquots of RPMI-1640 plus 2% fetal bovine serum (FBS), 2 -μL aliquots of piglet sera diluted 1:5 with RPMI-1640 plus 2% FBS, and 2 -μL aliquots of RPMI-1640 plus 2% FBS containing human peripheral blood lymphocytes (PBLs; 1 × 106 cells/mL) were added to each well of the Terasaki plate and were allowed to incubate at 37°C for 30 minutes. RPMI-1640 (10 μL) was added to each well, and they were incubated at room temperature for 10 minutes. The supernatant from each well was removed. Two-microliter aliquots of a rabbit complement (Pel-Freez, Rogers, AR) diluted 1:10 in RPMI-1640 were added to each well, and they were incubated at 37°C for 30 minutes. A fluorescence-activated cell sorting buffer containing formalin (20 μL) was added to each well. Live and dead cells were counted with a phase contrast microscope.
Human albumin was measured with an enzyme-linked immunosorbent assay quantification kit from Bethyl Laboratories (Montgomery, TX) per the manufacturer's instructions. All assays were done in triplicate. The serum albumin concentration (C; ng/mL) was then used to determine the concentration change per day (dC/dT; ng/mL/day) with a power trend line. The production rate of human albumin per day (P; ng/day) was determined with the following equation:
where V is the blood volume (mL) and k is equal to ln 2/half-life of albumin in pigs. V was based on the total body weight of the piglets, with 8% of the total body weight represented by the intravascular volume. The average half-life of pig albumin is 8 days. We determined the engraftment percentage by taking the total number of viable human hepatocytes transplanted and dividing the observed human albumin production by the expected albumin production.
Slides containing fresh frozen piglet liver were fixed for 10 minutes at −20°C in 100% ethanol. The slides were washed for 2 minutes in 1× phosphate-buffered saline (PBS) plus 0.05% Tween 20, and this was repeated twice. The slides were incubated with mouse anti-human beta-2 microglobulin antibody (Santa Cruz Biotechnology, Santa Cruz, CA) for 24 hours at 4°C. The slides were washed for 2 minutes in 1× PBS plus 0.05% Tween 20, and this was repeated twice. The slides were incubated with donkey anti-mouse immunoglobulin G antibody (Invitrogen, Grand Island, NY) for 1 hour at room temperature. The slides were washed for 2 minutes in 1× PBS plus 0.05% Tween 20 (this was repeated once), and then they were dipped 10 times in 1× PBS. The slides were covered with 20 μL of 1× PBS plus Hoechst stain (Invitrogen, Grand Island, NY) and a cover slip and were immediately viewed with a Leica DMI 4000B inverted microscope at ×20.
Fluorescence In Situ Hybridization (FISH)
Slides were probed with a cocktail of a homebrewed whole genome human probe labeled in Spectrum Orange deoxyuridine triphosphate (dUTP; Abbott Molecular/Vysis Products, Abbott Park, IL) and a chromosome 1 pig probe labeled in Spectrum Green dUTP (Abbott Molecular/Vysis Products). FISH was performed according to standard methods for the hybridization of labeled DNA probes to paraffin-embedded specimens as described elsewhere. Visualization of the FISH signals was performed with a Zeiss fluorescent microscope fitted with a triple bandpass filter. Pictures were captured with CytoVision imaging software.
RNA and Genomic DNA Amplification
RNA was isolated from liver tissue with the RNeasy Plus kit (Qiagen, Valencia, CA) according to the manufacturer's protocol. Reverse transcripts were created with the SuperScript III first-strand synthesis system (Invitrogen). Human albumin was amplified from the resulting complementary DNA with AmpliTaq Gold polymerase chain reaction (PCR) master mix (Applied Biosystems, Carlsbad, CA) per the manufacturer's protocol along with forward primer CCCTCCGTTT GTCCTAGCTTTTCTC and reverse primer CCTCACTCT TGTGTGCATCTCGACG. The PCR conditions were set as follows: 10 minutes at 95°C followed by 40 cycles of 1 minute at 95°C, 1 minute at 65°C, and 1 minute at 72°C followed by 10 minutes at 72°C. Genomic DNA was isolated from liver tissue with the DNeasy blood and tissue kit (Qiagen) according to the manufacturer's protocol. Human beta-2 microglobulin was amplified from the resulting genomic DNA with the Sure-Pol DNA polymerase kit (Denville Scientific, Metuchen, NJ) according to the manufacturer's protocol along with forward primer GGTTGCTCCACAGGTAGCTC and reverse primer TTTCCCCCAAATTC TAAGCA. The PCR conditions were set as follows: 5 minutes at 95°C followed by 35 cycles of 1 minute at 95°C, 1 minute at 52°C, and 1 minute at 68°C followed by 10 minutes at 68°C.
Evidence of Postnatal Immune Hyporesponsiveness
On day 40 of gestation, fetal pigs received either human hepatocytes or normal saline in utero by direct injection into the fetal liver. To assess whether early prenatal exposure to human hepatocytes would lead to immune hyporesponsiveness in pigs in the postnatal environment, we examined the serum for cytotoxic antibody production. Serum isolated from fetal pigs injected with human hepatocytes or normal saline at 2, 4, 6, and 8 weeks of life was analyzed for the development of cytotoxic antibodies against human PBLs (Fig. 2). As expected, piglets primed with normal saline did not develop antibodies against human cells. Normoresponsive piglets primed with human hepatocytes developed a strong cytotoxic response after the second and third priming injections. In contrast, serum collected from hyporesponsive piglets primed postnatally with human hepatocytes showed little to no cytotoxic activity toward human PBLs; this was similar to the situation for animals primed with only normal saline.
Histological and PCR Evidence of Human Hepatocyte Engraftment in Hyporesponsive Pigs
Fresh-frozen liver samples were obtained from 7-week-old hyporesponsive piglets that underwent human hepatocyte transplantation 1 week after birth via direct hepatic injection. We also took liver samples from 7-week-old normoresponsive piglets that were transplanted with human hepatocytes by direct liver injection 1 week after birth. These experimental samples were compared to fresh-frozen liver samples from normal human and piglet controls. Samples from the hyporesponsive and postnatally transplanted piglets demonstrated large clusters of cells that stained positive for human-specific beta-2 microglobulin (Fig. 3A). In addition, this group of samples showed abundant fluorescent in situ hybridization with a whole genome human probe labeled in Spectrum Orange dUTP (Fig. 3B). All samples were evaluated by PCR for the presence of human-specific beta-2 microglobulin. Human liver samples and hyporesponsive piglets receiving human hepatocyte transplantation after birth were the only samples that generated PCR-specific products related to the presence of human-specific beta-2 microglobulin (Fig. 4). A control piglet liver did not amplify human beta-2 microglobulin. Samples obtained from the livers of normoresponsive piglets that underwent human hepatocyte transplantation 1 week after birth did not show any histological or PCR evidence of human cells (data not shown).
Presence of Human Albumin
Complementary DNA was created from RNA that was isolated from liver samples obtained from humans, normal piglets, 7-week-old hyporesponsive piglets (n=4) that underwent human hepatocyte transplantation 1 week after birth via direct hepatic injection, and 7-week-old normoresponsive piglets (n=3) that underwent human hepatocyte transplantation 1 week after birth via direct hepatic injection. Serum samples were also obtained from all animals 2, 4, and 6 weeks after engraftment. Hyporesponsive piglets undergoing human hepatocyte transplantation after birth showed strong expression of human albumin 6 weeks after transplantation (Fig. 5). Additionally, hyporesponsive piglets that underwent human hepatocyte transplantation after birth had measurable amounts of human albumin in their serum 2, 4, and 6 weeks after engraftment (piglet 1, 723.1±173.4, 379.0±47.8, and 157.7±11.8 ng/mL; piglet 2, 2377.7±171.4, 335.5± 28.1, and 114.4±9.5 ng/mL; piglet 3, 1459.8±418.9, 641.5±107.1, and 267.1±29.8 ng/mL; and piglet 4, 227.2±35.2, 88.3±16.1, and 55.8±9.2 ng/mL). On the basis of these serum concentrations, daily albumin production was extrapolated (piglet 1, 11.5±1.5 μg/day; piglet 2, 2.1±1.4 μg/day; piglet 3, 18.0±2.1 μg/day; and piglet 4, 4.6±0.4 μg/day), and this was further used to determine the percentage of hepatocytes stably engrafted 6 weeks after engraftment (piglet 1, 38.8%±5.1%; piglet 2, 7.2%±4.9%; piglet 3, 60.8%±7.1%; and piglet 4, 15.4%±1.4%; Fig. 6). Piglets not undergoing transplantation in utero but undergoing engraftment in the postnatal environment did not demonstrate expression or production of human albumin.
IUCT can result in the long-term engraftment of allogeneic and xenogeneic cells.[14-17] Previously, the in utero transplantation of hematopoietic cells led to stable long-term engraftment in the postnatal environment.[14, 16-18] In this study, we show for the first time that IUCT in piglets with adult human hepatocytes allows the stable postnatal engraftment of human hepatocytes. The effectiveness of our IUCT step was determined by a cytotoxicity assay that measured the change in piglet serum reactivity to human PBLs after 3 postnatal SQ priming injections using human hepatocytes. Piglets that underwent IUCT with human hepatocytes and were primed postnatally with human hepatocytes showed cytotoxicity profiles similar to those of piglets primed with injections of normal saline alone. Additionally, we showed that human hepatocytes engrafted postnatally in animals undergoing transplantation in utero and that these hepatocytes were present, viable, and functionally active 6 weeks after transplantation.
In this study, we chose day 40 of gestation for IUCT on the basis of previous work by Sinkora et al., who demonstrated that pre-T cell maturation does not occur in fetal pigs before day 45 of gestation. The fetal thymus is the primary site at which pre-T cells determine self-recognition and responsiveness to foreign antigens. In a series of maturation steps, pre-T cells undergo positive and negative selection in the fetal pig thymus after day 45 of gestation. During T cell maturation in the fetal environment, clonal T cell deletion occurs for clones that exhibit a high affinity to self-antigens and to any foreign antigens introduced before this gestational check point. Immune hyporesponsiveness is induced by the clonal deletion of T cells reactive to human hepatocyte-specific antigens. However, this proposed mechanism of acquired in utero immune hyporesponsiveness requires further validation because it has been reported that T cells may escape thymic deletion and persist in the postnatal environment, with immune hyporesponsiveness relying on regulatory T cells.[19, 20]
In our model, we injected human hepatocytes into the fetal liver. In the setting of tolerization acquired by the in utero transplantation of hematopoietic cells, researchers have previously targeted the peritoneal cavity, liver, or vitelline vein.[14, 21-23] The optimal site for in utero transplantation remains unclear, especially when human hepatocytes are being used. In addition, concern for fetal receptivity has been raised with respect to IUCT because of the available space and homing ability of the transplanted cell.[24, 25] We chose the liver because of its relatively large size at 40 days’ gestation and because of our uncertainty about whether hepatocytes would home to the liver effectively in utero if they were injected elsewhere.
In our current model, we use IUCT of human hepatocytes to induce an immune-hyporesponsive state allowing the postnatal engraftment of a larger number of cells. A number of potential barriers to our IUCT and engraftment procedure still exist. In humans, for example, the fetal environment has recently been shown to have immunologically active T cell populations early in gestation. The human fetal liver contains natural killer cells and T cells that may undergo T cell receptor rearrangements to make them alloreactive against major histocompatibility complex in vitro.[26, 27] Moreover, recent work has demonstrated mature maternally derived immune cells that track to fetal tissue, and this raises concerns about their impact on the engraftment of transplanted foreign cells.
Additionally, the issue of competition in the wild-type fetal and neonatal liver for transplanted human hepatocytes is a concern. In this study, the human hepatocytes did not have any selective advantage. They persisted but did not appear to have any degree of robust repopulation because only 30% of the postnatally transplanted cells demonstrated engraftment at 6 weeks. To address this concern, we recently developed a porcine model of HT1. In this model, native hepatocytes build up toxic metabolites secondary to a defective tyrosine metabolism. This environment in the mouse model of HT1 provides a selective advantage to transplanted normal primary adult human hepatocytes to expand and nearly replace the mouse liver with human hepatocytes, with repopulation numbers exceeding 90%. Similar results have occurred with the competitive advantage of transplanted hematopoietic cells in anemic mouse models. The degree of engraftment in these competitive hematopoietic cell transplant models appears to be directly related to the degree of deficiency that exists in the recipient animal. However, it is possible that the engraftment of transplanted human hepatocytes may all be possible in utero when our studies progress to homozygote FAH-deficient pigs, in which the recipient environment favors the transplanted human hepatocytes and results in an active and extensive repopulation of human hepatocytes.
Future work in this area will focus on potential barriers to IUCT and on the optimal cell type and site to be used for IUCT. With the evidence that exists, it is our belief that the greater competitive advantage of the transplanted cell in the host environment, along with a defective immune response, will lead to adequate engraftment and expansion of the transplanted human hepatocytes. In the setting of primary human hepatocytes, there exists a great demand for these cells in medicine and in the research community. Our porcine model of HT1 may provide a highly selective competitive advantage to transplanted human hepatocytes and, in conjunction with IUCT, may serve as the first system capable of the large-scale expansion of primary human hepatocytes.
The authors thank Yecuris, Inc., for providing human hepatocytes from FRG mice and Raymond Hickey for preliminary work on this project.