Transplantation of human cells into Interleukin-2 receptor gamma gene knockout pigs under several conditions.

Introduction: Previously, we performed gene knockout (KO) of interleukin-2 receptor gamma (IL2RG) in porcine fetal fibroblasts using zinc finger nuclease-encoding mRNAs, subsequently generating IL2RG KO pigs using these cells through somatic cell nuclear transfer. The IL2RG KO pigs lacked a thymus and were deficient in T lymphocytes and natural killer cells, similar to human X-linked severe combined immunodeficiency (SCID) patients. The present study aimed to evaluate whether pigs can support the growth of xenografted human cells and have the potential to be an effective animal model.
Methods: The IL2RG XKOY pigs used in this study were obtained by mating IL2RG XKOX females with wild-type boars. This permitted the routine production of IL2RG KO pigs via natural breeding without complicated somatic cell cloning procedures; therefore, a sufficient number of pigs could be prepared. We transplanted human HeLa S3 cells expressing the tandem dimer tomato into the ears and pancreas of IL2RG KO pigs. Additionally, a newly developed method for the aseptic rearing of SCID pigs was used in case of necessity.
Results: Tumors from the transplanted cells quickly developed in all pigs and were verified by histology and immunohistochemistry. We also transplanted these cells into the pancreas of designated pathogen-free pigs housed in novel biocontainment facilities, and large tumors were confirmed. Conclusions: IL2RG KO pigs have the potential to become useful animal models in a variety of translational biology fields.

Introduction

Pigs are large animals with similar anatomy and physiology to humans and are effective models for biomedical research [[1], [2], [3], [4]]. The high sequence homology of pigs with human xenobiotic receptors may allow the accurate prediction of the pharmacodynamic and pharmacokinetic properties of drugs compared to rodents [5]. For surgical and clinical monitoring purposes, rodents are too small to be practical [6,7]. Thus, there is a demand for sophisticated pig animal models, which may improve the translation efficiency from preclinical to clinical studies.

IL2RG, the common gamma chain, is an indispensable subunit of functional receptor complexes for the following interleukins (ILs); IL-2, IL-4, IL-7, IL-9, IL-15, and IL-21 [[8], [9], [10], [11], [12], [13], [14]]. The common gamma chain plays a key role in the differentiation of T lymphocytes and natural killer (NK) cells [15,16]. The porcine common gamma chain gene, assigned to swine chromosome Xq13, comprises eight exons, spans approximately 3.7 kb and encodes a 368-amino acid polypeptide [13,15]. A mutation in the IL2RG gene induces a marked decrease or depletion of T and NK cells and loss of production of antibodies from B cells. This is seen in human X-linked SCID patients [17], which is a genetic disorder of the functions of major lymphocytes, such as B and T cell impairment [18,19].

SCID pigs arose spontaneously in a standard breeding program [20,21]. They were unable to produce antibodies, had atrophied lymph nodes, and lacked a thymus and B and T cells. In addition, SCID-like pigs are generated using genetically modified techniques. So far, Suzuki et al. [22] and our group [23] have disrupted the X-linked porcine gene encoding IL2RG and obtained SCID-like pigs through various strategies. Suzuki et al. targeted IL2RG in somatic cells via homologous recombination and used serial somatic cell nuclear transfer (SCNT). Subsequently, they performed further breeding to generate heterozygous IL2RG+/− females and IL2RG−/Y males. The males were athymic and had impaired immune cell production. Our group used the more efficient ZFN-encoding mRNA technology to knock out the IL2RG gene in porcine fetal fibroblasts. IL2RG KO pigs were generated using KO cells via SCNT. The pigs completely lacked a thymus and were deficient in T and NK cells. The use of ZFNs is promising for creating genetically engineered pigs [[23], [24], [25]]

Human xenotransplantation studies were not possible in pigs until the identification [20,21,26] or creation [[22], [23], [24], [25],[27], [28], [29], [30]] of SCID-like pigs. SCID pigs have been reported to accept grafts of human melanoma (A375SM) and pancreatic carcinoma (PANC-1) cancer cell lines [20], OSPC-ARK1 primary ovarian cell lines [31], and human induced pluripotent stem (iPS) cells [32,33]. However, several issues are constantly raised in obtaining SCID-like pigs: 1. Additional cloning and breeding steps are required; 2. Pigs require 7 months to reach sexual maturity; 3. There is a high possibility of severe infection due to disruption of the immune system during rearing [25]; and 4. They often suffer from reproductive problems. Therefore, the establishment of an efficient reproductive system is key for acquiring IL2RG KO pigs. Previously, we showed that the valuable traits of IL2RG KO pigs can be maximized by generating unique chimeric boars composed of mutant and normal cells [25]. The chimeric boars exhibited a cured phenotype with fertility when carrying and transmitting the X-linked genetic disease (XLGD) genotype. Male-based breeding with wild-type (WT) female pigs permits the routine production of IL2RG KO pigs in a Mendelian fashion.

The present study aimed to evaluate whether the generated IL2RG KO pigs can host and support the growth of xenografted human cells under several conditions. Currently, we have developed a feasible and economical method to produce designated pathogen-free (DPF) pigs, which involves a uterectomy procedure on a full-term sow, the recovery of fetuses from the uterus, and the rearing of neonatal piglets under aseptic conditions in specially designed isolator units [34]. We termed this method the uterectomy-isolated rearing (U-iR) method. Considering the vulnerability of IL2RG KO pigs to infection, the method was attempted for this study because it allows for long-term follow-up if the system is functional.

Recently, Itoh et al. [35] developed an immunodeficient pig model, the operational immunodeficient pig (OIDP), by surgically removing the thymus and spleen and creating a controlled immunosuppressive protocol with a combination of drugs commonly used in the clinical setting. This model allowed for the long-term accommodation of artificial human vascular grafts. Therefore, this study also comparatively investigated the differences between IL2RG KO pigs and OIDPs.

Materials and methods

Animals

All animal experiments performed in this study, including the creation of immunodeficient pigs, xenotransplantation, and the production of DPF pigs, were approved by the Institutional Animal Care and Use Committee of Meiji University (IACUC, IACUC15-0005, IACUC17-0006, IACUC18-0004, and MUIACUC2020-05). All experiments were performed in accordance with the relevant guidelines and regulations. All animals were maintained under conventional conditions in an air-conditioned room, bred, and observed daily by the animal husbandry personnel. All efforts were made to minimize suffering.

Production of IL2RG KO pigs

The IL2RG XKOY pigs used in this study were obtained by mating IL2RG XKOX females with WT boars. We have previously produced an IL2RG XKOY↔XWTXWT chimeric boar via blastocyst complementation with IL2RG XKOY and WT cloned embryos [25] (Supplementary Fig. 1). This chimeric boar was used to breed WT females which could generate IL2RG XKOX progeny. IL2RG XKOX progeny (IL2RG mutation carriers) were bred for approximately 7–15 months before the first mating. IL2RG XKOY pigs were surgically obtained from the uteri of pregnant females at 112 days of gestation using a previously reported procedure [36]. Briefly, the uterus of a pregnant female was exenterated and placed in an isolation chamber sterilized with 0.1% peracetic acid vapor. Piglets were removed from the uterus through an incision in the uterine wall and, once breathing spontaneously, were moved to a rearing isolator for artificial nursing.

Genotyping

Genotyping was performed as previously described [23]. Briefly, genomic DNA was extracted from a tail biopsy and/or the umbilical cord specimens of newborn piglets using a DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany). The target region of IL2RG was amplified from the cell clones by direct PCR using MightyAmp DNA polymerase (Takara Bio, Shiga, Japan) and the appropriate primers (5-ATACCCAGCTTTCGTCTCTGC and 5-TTCCAGAATTCTATACGACC). The PCR products were separated by agarose gel electrophoresis to determine whether they were WT, heterozygous (female carrier; IL2RG +/−), or SCID (IL2RG -/Y).

Production of DPF pigs using U-iR method

The DPF pigs were treated in different ways. Each excised uterus was transferred to an aseptic isolator (recovery unit) via a disinfecting tub filled with 0.05% peracetic acid, and fetuses were recovered from the uterus within 5 min. The piglets were then transferred to a second isolator unit (rearing unit, 1–5 piglets/unit) for artificial nursing with γ-irradiated milk substitute and kept in the novel biocontainment rearing unit (Supplementary Fig. 2) until the study was complete.

Microbiological testing

Swab samples from the body surface, oral mucosa, and feces of the pigs were obtained using the U-iR method, while samples from the internal surface of the rearing unit were collected weekly. The samples were cultured on thioglycolate and potato dextrose agar and examined using a standard protocol developed and recommended by the Japanese Association for Laboratory Animal Science [37]. The samples were analyzed for sterility and the presence of viruses.

Production of different immunodeficient pigs

We prepared three different forms of immunodeficient pig in this study. The OIDP was produced as described by Itoh et al. [35], with some modifications. One male Large White/Landrace x Duroc pig (34 days after birth) was anesthetized by isoflurane inhalation. A midline incision was made in the ventral region of the neck and three lobes of the thymus were excised. Laparotomy was performed by making an incision in the middle of the abdomen to remove the spleen. Simultaneously, the upper portion of the stomach was exposed and a catheter (16 F, JMS hydrophilic Foley catheter, JMS) was inserted, after which the stomach and catheter insertion site were ligated and fixed (K302: K1). The second pig went through the same procedure, without a splenectomy (K303: K2) and the third pig underwent gastrostomy only (K304: K3). All three pigs underwent surgery 16 days before transplantation and were treated with immunosuppressive drugs according to the following protocol: Tacrolimus hydrate was administered at 0.5 mg/kg/day (Pfizer, New York, NY, USA); mycophenolate mofetil (MMF) was administered at 60 mg/kg/day (Pfizer); and prednisolone was administered at 20 mg/pig/day (Teva Takeda Pharma, Tokyo, Japan). All drugs were administered through the gastric tube 6 days before transplantation. The day of transplantation was defined as day 0. The blood tacrolimus and MMF concentrations were measured at SRL International, Inc., Japan, and the dose was adjusted so that the trough levels of these drugs could be measured.

Preparation of donor human cells

For human donor cells, we selected the human cervix carcinoma cell line HeLa S3 (National Institutes of Biomedical Innovation, Health, and Nutrition, Osaka, Japan) for this study and modified the cells to express the tandem dimer Tomato (tdTomato). HeLa S3 cells were grown in F12 medium with 10% FBS (Sigma–Aldrich, St. Louis, MO, USA) and antibiotic-antimycotic solution (Thermo Fisher Scientific) until 80% confluence and were passaged several times for preparation. The tdTomato expression vector used in the present study was prepared using our previously reported method [38]. The vector consisted of a chicken beta-actin promoter with a cytomegalovirus enhancer (CAG promoter), tdTomato cDNA, a rabbit beta-globin 3′-flanking sequence including a polyadenylation signal, and the puromycin N-acetyltransferase gene driven by the phosphoglycerate kinase promoter. For transfection, HeLa S3 cells were cultured to 70–90% confluence, washed twice with Dulbecco's phosphate-buffered saline, and collected after treatment with 0.05% trypsin-ethylenediaminetetraacetic acid (trypsin–EDTA, Life Technologies). The collected cells (6.0 × 105) were then resuspended in 60 μL of resuspension buffer supplied as part of a NeonTransfection System kit (Life Technologies) and 1.5 μg of the tdTomato expression vector was added. The cells were then electroporated under the following conditions: Pulse voltage, 1200 V; pulse width, 20 ms; pulse number, 2. After 48 h of electroporation, the cells were transferred to a medium containing 2.5 μg/mL puromycin. After 12 days in culture, puromycin-resistant cells were collected and a limiting dilution was performed to obtain single cell-derived clones. tdTomato-positive cells were grown to confluence and cryopreserved for later use as donor cells for transplantation.

Cell transplantation

For transplantation, the cells were brought to a concentration of 3 × 106 cells/0.2 mL or 1 × 105 cells/0.05 mL. The cells were placed on ice until immediately before the transplantation.

All pigs in the following experiment were anesthetized by administration of isoflurane gas vaporized into oxygen and delivered via a face mask. The transplanted cells were monitored for tumor development.

Experiments

Experiment 1: evaluation of transplanted human cells in IL2RG KO and WT pigs in conventional rearing

Seven IL2RG KO and six WT pigs were prepared, and donor cells were transplanted subcutaneously into one or both ears one day after birth. Two IL2RG KO pigs and three WT pigs received transplants into both ears, while all others received transplants in one ear only. Donor cells (3 × 106) were suspended in 200 μL phosphate-buffered saline (PBS) for each transplant.

Experiment 2: evaluation of human cells transplanted into the pancreas of DPF IL2RG KO pigs housed in novel biocontainment facilities

Two 7-day-old IL2RG KO pigs underwent laparotomy in an isolator unit under sterile conditions. Pigs were transplanted with 1 × 105 cells in 50 μL of PBS-cell suspension into the tail of the pancreas.

Experiment 3: comparison between the IL2RG KO pigs and other immunodeficient models

This study also comparatively investigated the differences between IL2RG KO pigs and other immunodeficient models. Three immunodeficient pigs were produced through three different operations including OIDP 16 days before transplantation. The day of transplantation was defined as day 0. Cells were transplanted into both ears in the same manner as in Experiment 1 and these pigs were reared in conventional housing.

Size of the tumor at the transplanted site

First, the tumor from the transplanted cells was verified to determine whether it emitted red tdTomato signals using a fluorescent lamp and filter. Once confirmed, the short and long diameters of the tumor at the transplant site were measured. Tumors were visually inspected daily for tumor growth in all experiments. Slide calipers were used to measure the tumor volume and calculated using the following formula: Short diameter2 × long diameter/2.

Histology

Samples from each transplantation site were collected at every time point. The tissue samples were immediately fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned at a thickness of 4 μm. Hematoxylin-eosin (HE) staining and analysis were performed to establish the presence of tumor architecture, following previously described immunohistochemical methods [39,40]. Briefly, after deparaffinization and blocking, sections were incubated with diluted primary antibodies overnight at 4 °C. The following primary antibodies were used: Common lymphocyte marker [35] rabbit anti-CD45 (Abcam plc, Cambridge, UK), T lymphocyte marker [35] anti-CD3 (DAKO, Santa Clara, CA, USA), B lymphocyte marker [35] anti-CD20 (Thermo Fisher Scientific), NK cell marker [41] anti-CD56 (LSBio, Seattle, WA), monocyte/macrophage lineage and dendritic cell marker [42] anti-Iba1 (Wako Pure Industries), cytotoxic T cell and NK cell marker [43,44] anti-granzyme B (Abcam plc), and major histocompatibility complex class II molecule marker [45,46] anti-HLA-DR (DAKO). Horseradish peroxidase (Rabbit/Mouse, DAKO) was used as the secondary antibody. A liquid DAB + Substrate Chromogen System (DAKO) was used as the substrate for colorization. The negative controls included sections incubated with secondary antibodies only. The number of CD45+ cells 1 week after transplantation was quantified using the Hybrid Cell Count Module (H4A; Keyence, Osaka, Japan). To evaluate the CD45+ cells, we counted them in three randomly selected areas of six samples from the transplant sites of two different IL2RG KO and WT pigs (four pigs in total). The mean value for each lobe was calculated on every day of assessment. The sections were examined under an all-in-one fluorescence microscope (BZ-X800; Keyence, Osaka, Japan) and the built-in software was used for data analysis (BZ-H4A; Keyence).

Statistical analyses

Statistical analyses were performed with SPSS software (version 26; IBM Corporation, New York, USA). Data were averaged and expressed as mean ± standard deviation (SD). Unpaired Student's t-tests, repeated measures one-way analysis of variance, and Fisher's protected least significant difference tests were used to compare groups. Results with p values of less than 0.05 compared with the WT were considered statistically significant.

Results

Repetition of phenotypic symptoms in offspring

Four IL2RG XKOX females grew into healthy fertile adults. No deaths occurred during rearing. From four females, 13 IL2RG KO males were obtained from 35 siblings (37.1%). They harbored the same IL2RG mutation as the nuclear donor cells.

Pig observations and development of tumors

After transplantation, both IL2RG KO and WT pigs exhibited normal growth and development over time (Fig. 1A). In both groups, the transplanted cells were chronologically followed (Fig. 1B and C). In IL2RG KO pigs, transplanted cells were identified visually and by palpation one week after transplantation. The transplanted cells developed into tumors with small, firm, raised, and elongated masses (Fig. 1B).

Fig. 1

Transplantation of HeLa S3 cells into subcutis of the ear in IL2RG KO and WT pigs. (A) Body weight gain of pigs over time. No differences were observed in the growth curves between the two groups. Blue line: IL2RG KO pigs (n = 7); Green line: WT pigs (n = 6). (B, C) Transplanted cells were monitored for tumor development. Photographs show an external view of tumor growth in the ear. (B) IL2RG KO pigs. Tumors derived from transplanted cells were identifiable in pig ears and their size increased with time. (C) WT pigs. Palpable tumors were observed in the ears of pigs 1 week after transplantation; however, no obvious tumors were observed at the transplantation site from 2 weeks after transplantation. (D) Broad photograph of the dissected tumor derived from an IL2RG KO pig. For the donor cells in this study, we selected the HeLa S3 cell line, which was modified to express tdTomato. The tumor was inspected with a fluorescent lamp and filter to verify whether the tumor emitted tdTomato. (D, left) Bright-field stereomicroscopic images of dissected tumors at 3 weeks after transplantation. (D, right) Fluorescence-stereomicroscopic image (left panel). The tumor exhibited a strong red signal. Scale bar: 5 mm. (E) Tumor volume in both groups. Two IL2RG KO pigs and three WT pigs received cell transplants in both ears, while all others received them in one ear only. Blue line: IL2RG KO pigs (n = 9); Green line: WT pigs (n = 9). The time after birth is indicated as weeks. Quantitative data are presented as mean ± SD.

At 3 weeks, the pigs were euthanized, and the tumors in the ears were dissected. Broad photographs of the tumors were taken (Fig. 1D, left) and it was confirmed that the cells were strongly red (Fig. 1D, right). The tumors were substantially larger than those at 1 week after transplantation, measuring in at 6930.3 ± 918.0 mm3 (Fig. 1B and E).

In WT pigs, 1 week after transplantation, the transplanted cells were visually identified as tumors, but the size was smaller; IL2RG KO 1135.4 ± 556.5 mm3 vs WT 528.3 ± 250.8 mm3 (p < 0.05; Fig. 1C and E). No obvious tumors were observed at 2 weeks after transplantation (Fig. 1C and E).

Histology and immunohistochemistry revealed the presence of transplanted cells in all transplanted sites in IL2RG KO pigs 1 week after transplantation (Fig. 2A and A′). The tumors were well demarcated and formed nests (Fig. 2A, left). The transplanted cells clonally expanded at the transplanted site (Fig. 2A, right) and scarce CD45+ cells were confirmed (Fig. 2A).

Fig. 2

Histological evaluation of tumors within the subcutis of the ear of pigs at 1 week after transplantation. (A, A′) IL2RG KO pigs. (B, B′) WT pigs. (A) HE staining. Tumors were visible, well demarcated, and formed nests in the ears (A, left). The transplanted cells expanded clonally at the transplantation site (A, right). (A′) CD45 staining. Scarce CD45+ cells were detected in the tumor. (B) HE staining. Within the subcutis of the ears of WT pigs, there were moderately demarcated tumors (B, left); however, they were not as dense as the tumors of IL2RG KO pigs (B, right). (B′) CD45 staining. CD45+ cells were evident and abundant within the tumor. Scale bars: black, 10 mm; red, 100 μm. (C) Lymphocyte infiltration. The number of CD45+ lymphocytes was counted and compared between the IL2RG KO and WT groups as the rate of CD45+ lymphocytes in the tumor area. In WT pigs, CD45+ cells comprised 46.7% of the tumor area, which was significantly higher than that in the IL2RG KO pigs (∗p < 0.05). Data are presented as the mean ± SD of six independent experiments.

In WT pigs, 1 week after transplantation, transplanted cells were also confirmed (Fig. 2B). Within the subcutis of the ear, moderately demarcated tumors were observed (Fig. 2B, left) but they lacked the same density as the tumors in the IL2RG KO pigs (Fig. 2B, right). The number of infiltrating CD45+ cells in the tumor area was confirmed as 46.7 ± 9.1 and was significantly higher than that of the IL2RG KO pigs (2.0 ± 1.4) (Fig. 2B’ and C). Lymphocyte markers were verified by positive immunostaining (Fig. 3B–F). CD56+ cells were not abundant; however, numerous lymphoid cells infiltrated the transplanted tissue. CD3+ cells were strongly positive and many HLA-DR + cells were also identified (Fig. 3G). This indicates that lymphocytes were activated and that pigs with intact immune systems rejected the transplanted cells at the transplantation site. These phenomena were confirmed at all transplantation sites in all WT pigs.

Fig. 3

Histological analysis of the transplanted cells in WT pigs 1 week after transplantation. (A–G) Tumors derived from transplanted cells within the subcutis of the ear. The right panels show an enlarged image of each stain. (A) HE staining. Moderately encapsulated and infiltrative tumors were visible. (B–F) Lymphocyte markers were verified by immunostaining. (B) T lymphocyte marker, CD3. (C) B lymphocyte marker, CD20. (D) NK cell marker, CD56. (E) Monocyte/macrophage lineage and dendritic cell marker, Iba1. (F) Cytotoxic T cell and NK cell marker, granzyme B. (G) Major histocompatibility complex class II molecule marker, HLA-DR. Numerous lymphoid cells infiltrated the tumor and lymphocyte activation was confirmed; therefore, rejection of transplanted cells by pigs with an intact immune system occurred at the transplantation site. These phenomena were confirmed in all WT pigs. Scale bar: 100 μm.

In IL2RG KO pigs, the presence of tumors was confirmed 3 weeks after transplantation. There was a focal, moderately demarcated area composed of nests and packets of transplanted cells within the subcutis of the ears. Scarce lymphoid-related markers were detected in all transplanted sites (Fig. 4B–F). HLA-DR + cells were not detected (Fig. 4G), indicating that the immune system, including lymphocytes, was not activated and that there was no rejection or expansion of the transplanted cells at the transplanted site.

Fig. 4

Histological analysis of the transplanted cells in IL2RG KO pigs 3 weeks after transplantation. (A–G) Tumors derived from transplanted cells within the subcutis of the ears. The right panels show the enlargement of each staining. (A) HE staining. There is a focal, moderately demarcated area composed of nests and packets of transplanted cells within the subcutis of the ears. (B–F) Lymphocyte markers were verified by immunostaining. (B) T lymphocyte marker, CD3. (C) B lymphocyte marker, CD20. (D) NK cell marker, CD56. (E) Monocyte/macrophage lineage and dendritic cell marker, Iba1. (F) Cytotoxic T cell and NK cell marker, granzyme B. (G) Major histocompatibility complex class II molecule marker, HLA-DR. In IL2RG KO pigs, scarce lymphoid-related markers were confirmed at all transplanted sites. The transplanted cells were not rejected and formed tumors at the transplanted sites. Scale bar: 100 μm.

The average body weight of the pigs did not differ significantly from that of the IL2RG KO pigs under conventional rearing. At 3 and 4 weeks post-transplantation (446–4 and 446–2, respectively), each pig was euthanized and the transplanted cells were confirmed (Fig. 5A and B). The tumors consisted of transplanted cells and strongly emitted red under fluorescent light (Fig. 5A and B, lower panels). The presence of a tumor with central necrosis was verified by HE staining (Fig. 5C, upper). At 3 and 4 weeks after transplantation, a small number of CD45+ cells were observed at the tumor boundary, but they did not infiltrate the tumor (Fig. 5C, lower, 446–4). Additionally, after the rearing period, swab samples were negative for bacteria, fungi, and protozoa (Table 1).

Fig. 5

Transplantation of HeLa S3 cells into the pancreas of DPF IL2RG KO pigs. Two 7-day-old IL2RG KO pigs underwent laparotomy in an isolator unit under sterile conditions. tdTomato-expressing HeLa S3 cells were transplanted into the tail of the pancreas in the pigs. (A and B) Broad bright-field photographs of tumors at 3 (Y446-4) and 4 weeks (Y446-2) after transplantation. (A and B, left panels) Macroscopic appearance of the pancreas. The yellow lines show the outline of each pancreas. (A and B, middle panels) Macroscopic appearance of a dissected pancreas. A large tumor was observed in the pancreatic tail. (A and B, right panels) Dissected tumors. (A′ and B′) Fluorescence-stereomicroscopic image of the tumor in IL2RG KO pigs. The tumor was inspected with a fluorescent lamp and filter to verify whether it emitted tdTomato. The transplanted cells in the pancreas appeared bright red. (A′ and B′, left and middle panels). The dissected tumor exhibited a strong red signal (A′ and B′, right panels). (C, C′) Histological appearance of a tumor (Y446-4). The tumor was HE stained and labeled with CD45. Each right lower panel is an enlargement of the images in C and C′. C and C′ are lower magnifications of the dissected tumor. (B) HE staining. The tumors exhibited central necrosis. (C′) A small number of CD45+ cells were verified at the boundary of the tumor, but they did not infiltrate. The time after birth is indicated weeks. Scale bars: black = 10 mm; yellow = 100 μm.

Table 1

Microbiological testing.

No	sample	0 weeks				3 weeks				4 weeks
		aerobic bacteria	anaerobic bacteria	fungi		aerobic bacteria	anaerobic bacteria	fungi		aerobic bacteria	anaerobic bacteria	fungi
Y446-4	oral mucosa	–	–	–		–	–	–
	Skin	–	–	–		–	–	–
	Stool	–	–	–		–	–	–
Y446-2	oral mucosa	–	–	–		NT	NT	NT		–	–	–
	Skin	–	–	–						–	–	–
	Stool	–	–	–						–	–	–
Isolator	Isolator	–	–	–		–	–	–		–	–	–

NT; Not tested.

For DPF pigs produced using the U-iR method, swab samples of the body surface, the oral mucosa, the feces, and the internal surface of the rearing unit were collected weekly during the study. The samples were cultured on thioglycolate and potato dextrose agar and examined using a standard protocol developed and recommended by the Japanese Association for Laboratory Animal Science.

Operations to induce immunodeficiency were performed without complications, including infections and liver or renal dysfunction after transplantation (Suppl. Table 1). The weight gain was unstable in two pigs (Fig. 6A, K302 and K304). The transplanted cells were verified by red tdTomato color using a fluorescent lamp and filter (Fig. 6B–D). In two pigs, K302 and K304, red signals were observed at 2 weeks after transplantation (Fig. 6B and D); however, the body weights of these two pigs did not increase from day 3 after transplantation (Fig. 6A). The K303 pig did not have a visible tumor at 1 week post-transplantation (Fig. 6C) but showed healthy bodyweight gains (Fig. 6A). This pig did not reach the therapeutic effective blood concentration level of MMF from day 0–2 weeks post-transplantation (Supplementary Fig. 3).

Fig. 6

Comparison between IL2RG KO pigs and other immunodeficient models. Three different immunodeficient pigs were produced by the following operations: K302 underwent thymectomy, splenectomy, and gastrostomy. K303 underwent thymectomy and gastrostomy. K304 underwent gastrostomy. (A) Body weights at specified times. The body weight gain was unstable in two pigs after transplantation (K302 and K304). Yellow line: K302; red line: K303; purple line: K304. The time after birth is indicated as days and weeks. ∗ Production of different immunodeficient pigs, ∗∗ Treated with immunosuppressive drugs, ∗∗∗ Transplantation. (B–D) The transplanted cells were verified by the red tdTomato color at the transplanted sites using a fluorescent lamp and filter. In two pigs, K302 and K304, red signals were observed 2 weeks after transplantation. No tumor was identified in the ears of K303 at 1 week post-transplantation. Scale bar: 10 mm.

Discussion

We have described the successful formation of human HeLa S3 cells in IL2RG KO pigs under several conditions. Pig models are highly effective and are employed for a plethora of translational medical research. SCID-like pigs, for example, comprise two independent mutations in the Artemis gene, which is a critical component of the V(D)J recombination pathway required for T and B cell receptor development. The pigs have a T− B– NK+ cellular phenotype [26]. One drawback of this model is that the NK cells in the pigs are functional [47], which could have antitumor activity in human cancer cells [31]. Another model, the RAG2 KO pig, has been used for studies on human cell xenotransplantation [32,33]. Although the pigs also had a T− B– NK+ cellular phenotype, they did not reject iPS cell lines or porcine cells with a trophoblast phenotype [33]. Our SCID-like model, the IL2RG KO pig, is similar to human X-linked SCID patients with a T− B+ NK− cellular phenotype [23]; however, the ability to obtain large numbers of these genetically engineered animals is greatly limited by expense. Additionally, the use of this model is associated with other critical issues. X-linked SCID patients typically die from severe infections in the first year of life if they do not receive any treatment [[48], [49], [50]]. We found that IL2RG KO pigs were also prone to systemic infection and suffered from severe complex infections by three weeks after birth under conventional housing conditions [25]. Furthermore, the generation of cloned gene KO pigs via SCNT is prone to be influenced by epigenetic dysregulation [51]. Thus, it is difficult to generate healthy, fertile boars and these hurdles should be carefully considered when implementing the IL2RG KO pig model in future studies.

Rescue experiments on embryos carrying lethal genetic mutations through chimerism with normal cells have given rise to individuals with normalized traits [[52], [53], [54]]. In our previous study, we generated chimeric boars composed of IL2RG KO cells and normal cells, because SCID is an X-linked disease. The chimeric boars exhibited a cured fertile phenotype while carrying and transmitting the SCID-like genotype. They appeared physiologically normal and remained healthy until up to 3 years of age under conventional rearing conditions [25]. This allowed for the routine production of IL2RG KO pigs through male-based breeding. All IL2RG KO pigs utilized in this study had a healthy reproductive system.

In the current study, we investigated whether human cells could ectopically develop tumors in IL2RG KO pigs. The transplanted cells developed in all of our SCID-like pigs and lymphatic markers were negative for the transplanted cells. Preliminary results indicated that B lymphocytes were present in IL2RG KO pigs, but disappeared by 12 weeks after birth [23,36]. It is known that B cells need T cell-associated cytokine stimulation to be functional [55,56]. In this study, CD3, CD8, CD45+, and the B cell marker CD20+ cells were not confirmed 3 weeks after transplantation, suggesting that B cells were not functional at the transplanted sites. Killer T cells were also not produced due to the negative expression of lba1 and granzyme B. We focused on the analysis of human cells in the present study because we plan to develop a human cancer study using IL2RG KO pigs in the near future. However, the pigs have also been previously engrafted with non-cancer cells; mouse whole pancreases were transplanted into the omentum. The transplanted pancreatic cells were also confirmed at 14 days after transplantation (data not shown).

These results may indicate support of the immune system in IL2RG KO pigs, because lymphocytes, including B cells, are not functional in the early stages of their lives.

For Experiment 1, we chose an easily monitored tumor location, in the ears of pigs. However, we will explore more challenging cancer studies in the future, especially pancreatic cancer, using orthotopic models. Pancreatic cancer, a particularly devastating human cancer, should be explored in pig models to determine their suitability. One of the major problems with current animal models of human tumors is their low translatability to clinical settings [31]. Once a pig tumor model is generated with IL2RG KO pigs, it could be chiefly employed for testing chemotherapy, radiotherapy, and surgical reduction strategies with more realistic results. To elucidate the effectiveness of anticancer treatments, studies require long-term follow-up. Therefore, pigs should be maintained in an environment where pathogen exposure is minimized and the risk of premature death due to SCID-related complications is reduced. Thus, establishing an efficient system to produce DPF pigs is essential for studies of this nature.

In Experiment 2, human cells were transplanted into the pancreas of IL2RG KO pigs under aseptic conditions. The fluorescent marker tdTomato was used to identify HeLa S3 cells, which were traceable by human-specific imaging systems, even outside the pig's body. We attempted this study with a prototype of a small isolator unit; hence, we were limited to a maximum study length of 5 weeks, after which the pigs would be too large. We are in the process of constructing a larger unit, which will allow us to perform long-term follow-up studies. Ultimately, we confirmed that human cells could successfully develop in the pancreas of IL2RG pigs, providing a basis for the refinement of a pancreatic cancer model in pigs. We have subsequently commenced further pancreatic cancer studies on human pancreatic cell lines in IL2RG KO pigs.

This study also comparatively investigated whether IL2RG KO pigs are superior xenotransplantation models to other immunodeficient models, in Experiment 3. In WT pigs, the blood concentrations of immunosuppressive drugs fluctuate according to their dosage, making it difficult to uniformly induce immunodeficiency. All three pigs in this experiment exceeded the therapeutic range of tacrolimus, while two of them exhibited lower therapeutic levels of MMF after transplantation. Furthermore, we used neonatal WT animals because of their quick development and, consequently, immunosuppressive drugs could not be monitored promptly. Indeed, the drug concentration levels of MMF for K302 and K303 were not within the therapeutic range, even when administered according to the protocol [35]. The therapeutic range for neonatal pigs was narrow due to their light body weight and, as such, the concentration of tacrolimus was much higher. In addition, the timing of immunosuppressive drug monitoring is critical and, in our study, the results could only be returned with a 3–4 day delay. Thus, without knowing the blood concentration levels of drugs on a daily basis, such as in human organ transplantation, it was difficult to control the concentration within the expected ranges.

Conclusions

We demonstrate that IL2RG KO pigs could be an effective host for human cells and that pig models should be further explored in various biomedical fields, such as stem cell, cancer, and translational research.

Declaration of competing interest

The authors have no conflicts of interest.