Generation of Monkey Induced Pluripotent Stem Cell-Derived Cartilage Lacking Major Histocompatibility Complex Class I Molecules on the Cell Surface

Due to the poor capacity for articular cartilage to regenerate, its damage tends to result in progressively degenerating conditions such as osteoarthritis. To repair the damage, the transplantation of allogeneic human induced pluripotent stem cell (iPSC)-derived cartilage is being considered. However, although allogeneic cartilage transplantation is effective, immunological reactions can occur. One hypothetical solution is to delete the expression of major histocompatibility complex (MHC) class I molecules to reduce the immunological reactions. For this purpose, we deleted the β2 microglobulin (B2M) gene in a cynomolgus monkey (crab-eating monkey [Macaca fascicularis]) iPS cells (cyiPSCs) to obtain B2M−/− cyiPSCs using the CRISPR/Cas9 system. Western blot analysis confirmed B2M−/− cyiPSCs lacked B2M protein, which is necessary for MHC class I molecules to be transported to and expressed on the cell surface by forming multimers with B2M. Flow cytometry analysis revealed no B2M−/− cyiPSCs expressed MHC class I molecules on their surface. The transplantation of B2M−/− cyiPSCs in immunodeficient mice resulted in teratoma that contained cartilage, indicating that the lack of MHC class I molecules on the cell surface affects neither the pluripotency nor the chondrogenic differentiation capacity of cyiPSCs. By modifying the chondrogenic differentiation protocol for human iPSCs, we succeeded at differentiating B2M+/+ and B2M−/− cyiPSCs toward chondrocytes followed by cartilage formation in vitro, as indicated by histological analysis showing that B2M+/+ and B2M−/− cyiPSC-derived cartilage were positively stained with safranin O and expressed type II collagen. Flow cytometry analysis confirmed that MHC class I molecules were not expressed on the cell surface of B2M−/− chondrocytes isolated from B2M−/− cyiPSC-derived cartilage. An in vitro mixed lymphocyte reaction assay showed that neither B2M+/+ nor B2M−/− cyiPSC-derived cartilage cells stimulated the proliferation of allogeneic peripheral blood mononuclear cells. On the contrary, osteochondral defects in monkey knee joints that received allogeneic transplantations of cyiPSC-derived cartilage showed an accumulation of leukocytes with more natural killer cells around B2M−/− cyiPSC-derived cartilage than B2M+/+ cartilage, suggesting complex mechanisms in the immune reaction of allogeneic cartilage transplanted in osteochondral defects in vivo. Impact statement The transplantation of allogeneic induced pluripotent stem cell (iPSC)-derived cartilage is expected to treat articular cartilage damage, although the effects of major histocompatibility complex (MHC) in immunological reactions have not been well studied. We succeeded at creating B2M−/− cynomolgus monkey (cy)iPSCs and cyiPSC-derived cartilage that lack MHC class I molecules on the cell surface. B2M−/− cyiPSC-derived cartilage cells did not stimulate the proliferation of allogeneic peripheral blood mononuclear cells in vitro. On the contrary, the transplantation of B2M−/− cyiPSC-derived cartilage into osteochondral defects in monkey knee joints resulted in survival of transplants and accumulation of leukocytes, including natural killer cells, suggesting complex mechanisms for the immune reaction.

The transplantation of allogeneic induced pluripotent stem cell (iPSC)-derived cartilage is expected to treat articular cartilage damage, although the effects of major histocompatibility complex (MHC) in immunological reactions have not been well studied.We succeeded at creating B2M -/-cynomolgus monkey (cy)iPSCs and cyiPSC-derived cartilage that lack MHC class I molecules on the cell surface.B2M -/-cyiPSC-derived cartilage cells did not stimulate the proliferation of allogeneic 1 Cell Induction and Regulation Field, Department of Clinical Application, Center for iPS Cell Research and Application, Kyoto University, Kyoto, Japan. 2 Department of Orthopaedic Surgery, Graduate School of Medicine, Kyoto University, Kyoto, Japan.
peripheral blood mononuclear cells in vitro.On the contrary, the transplantation of B2M -/-cyiPSC-derived cartilage into osteochondral defects in monkey knee joints resulted in survival of transplants and accumulation of leukocytes, including natural killer cells, suggesting complex mechanisms for the immune reaction.

Introduction
A rticular cartilage resides at the end of bone, where it acts as a lubricant to ensure smooth joint motion.Cartilage has limited regenerative capacity, and its damage tends to result in a degenerative condition, impairing joint function.Cell therapies, including the implantation of chondrocytes or mesenchymal stem/stromal cells (MSCs), have been performed against the focal damage of articular cartilage, but, after expansion in culture, chondrocytes lose their chondrocytic character and MSCs lose their differentiation potential, compromising the formation of quality cartilage.In addition, expansion endows the senescent phenotype, which may affect the metabolism of these cells.
Cartilage is considered immunoprivileged, 1,2 because it is avascular and because chondrocytes are surrounded by the extracellular matrix (ECM).The ECM blocks immune cells from contacting chondrocytes, thus avoiding immunological reactions even in allogeneic conditions.Accordingly, allogeneic cartilage harvested from juveniles [2][3][4][5][6] have been transplanted into patients with articular cartilage damage without matching human leukocyte antigen (HLA) types or the use of immunosuppressive agents.However, donors are rare, and their allogeneic cartilage are heterogeneous in quality and risk the transmission of disease.
Induced pluripotent stem cells (iPSCs) 7,8 are a potential alternative source for allogeneic transplantation.0][11] Theoretically, iPSC-derived cartilage with uniform quality can be provided at almost infinite amounts due to the self-renew capacity of iPSCs, which overcomes the problems associated with current treatments for cartilage damage.Chondrocytes are segregated by the ECM in iPSC-derived cartilage, just like in native cartilage.Accordingly, human iPSC-derived cartilage does not stimulate the proliferation of allogeneic lymphocytes in mixed lymphocyte reaction assays. 12These findings suggest that the transplantation of allogeneic iPSCderived cartilage could have medical benefit.
However, although both cartilage and iPSC-derived cartilage appear to have low immunogenicity, chondrocytes express major histocompatibility complex (MHC; HLA in humans) class I when treated with interferon g (IFNg). 2,12In contrast, MHC class II are not expressed in chondrocytes regardless of IFNg treatment. 2,12Importantly, MHC class I risk inducing immune reactions.Therefore, when the cartilage is degraded and the chondrocytes are exposed, which is typical in arthritic conditions, immune reactions can occur.Furthermore, allogeneic cartilage transplantation into osteochondral defects, where the transplanted cartilage is exposed to abundant blood flow in the bone marrow, can further increase the risk of an immune reaction.To assess this risk, the effects of MHC class I on the cell surface in the transplantation of allogeneic cartilage or chondrocytes need to be investigated scientifically.
To evade immune reactions associated with the transplantation of allogeneic iPSC-derived cells, iPSCs with reduced immunogenicity, such as those made from HLA-homozygous individuals or genetically modified for HLA depletion, are under development. 13In addition, TALENs, 14 CRISPR/ Cas9, [15][16][17] and vector-mediated gene targeting technology 18 have been used to inactivate the B2M gene to deplete HLA-I.The simultaneous knockout of HLA-A/B/C has also been reported. 19Because natural killer (NK) cells reject HLAdepleted iPSC-derived allografts, the knock-in of HLA-E was done to prevent HLA-depleted ESCs from eliciting an immune response from either CD8 + T cells or NK cells. 18inally, the targeted disruption of HLA-A and -B alleles in iPSCs reduces the lysis caused by T cells and NK cells. 17efore moving to the application of HLA-depleted iPSCs in clinical settings to treat articular cartilage damage, the effects of these HLA-depleted iPSC-derived chondrocytes should be examined in appropriate animal models in preclinical tests.Monkeys have a similar MHC structure as HLAs, thus making a good model.Therefore, in the present study, we generated cynomolgus monkey (crab-eating monkey [Macaca fascicularis]) iPSC (cyiPSC)-derived cartilage lacking MHC class I molecules on the cell surface and studied their chondrogenic and immunogenic properties.

Ethics statement
All methods were carried out in accordance with relevant guidelines and regulations.Experiments using recombinant DNA were approved by the Recombinant DNA Experiments Safety Committee of Kyoto University.All animal experiments were approved by the Institutional Animal Committee of Kyoto University.

Monkey iPSCs and genome editing at the B2M locus
The wild-type cyiPSC line 1123C1 was previously described. 20cyiPSCs were maintained in AK-02N medium (Ajinomoto).
CRISPR-Cas9 constructs were designed based on a previous report. 21To inactivate B2M, we aimed at disrupting exon 2 of the cynomolgus monkey B2M gene.The guide RNA sequence was UAUGUUCCUCAGGUACUCCA, which corresponds to the sequence at the 5¢-end of exon 2 (Fig. 1A).The guide oligo DNAs (Table 1) corresponding to the guide RNA sequence were annealed and ligated to pSpCas9(BB)-2A-Puro. 21 BbsI was then used to digest pSpCas9(BB)-2A-Puro.DNA vectors (3.3 mg) encoding the guide RNA for B2M and Cas9 were introduced into 1.0 • 10 6 cyiPSCs in 100 mL OPTI-MEM by a nucleofection system following the manufacturer's instructions (Nepa21, Nepa Gene).Electric pulses (175 V, 5 ms) were used.Two days after the nucleofection, the cells were cultured with puromycin.Eight days after the nucleofection, 39 colonies were picked up, and the cells in each colony were expanded.Genomic DNAs were extracted from the cells and subjected to polymerase chain reaction (PCR) analysis.

PCR digestion with restriction enzymes and sequencing
We obtained and analyzed genomic DNA from cyiPSCs using the DNeasy Ò Blood & Tissue Kit (Qiagen, Hilden) and PCR, respectively.PCR was performed using KOD plus Neo (Toyobo).Electrophoresis was performed with 100 bp The PCR products were purified using NucleoSpin Ò Gel and PCR Clean-up (Macherey-Nagel) and digested by three restriction enzymes: HincII, BtsXI and MfeI (New England Biolabs).Table 1 lists the primers used.
For direct sequencing of the PCR products, we obtained DNA fragments from an agarose gel using NucleoSpin Gel and PCR Clean-up.The fragments were analyzed using a Big Dye Terminator Kit (Thermo Fisher Scientific) and Applied Biosystems Ò 3500 • L (Thermo Fisher Scientific).The sequence data were analyzed using SnapGene Ò ver 4.0.8(GSL Biotech LLC, Chicago).
mRNA expression cyiPSC-derived cartilage were frozen in liquid nitrogen and crushed by a Multi-Beads Shocker (Yasui Kikai, Japan).Total RNAs were extracted using RNeasy (Qiagen).For quantitative reverse transcription PCR (RT-PCR), 200 ng total RNA was reverse transcribed into first-strand cDNA using ReverTra Ace (Toyobo) and an oligo(dT)20 primer.The PCR amplification was performed using a KAPA SYBR FAST qPCR Master Mix ABI prism Kit (KAPA Biosystems, Wilmington) and StepOnePlus Real-Time PCR System (Thermo Fisher Scientific).Table 1 lists the PCR primers used. 23,24The RNA expression levels were normalized to the level of GAPDH or ACTB expression, and the results indicate the relative expression of the molecules.

Western Blotting
We obtained proteins from the cells using lysis buffer and performed western blotting using Nupage Novex Bis Tris Gel (Thermo Fisher Scientific).Table 2 lists the antibodies used.The signals were recorded using ImageQuant LAS 4000 (GE Healthcare).

Immunocytochemical analysis
cyiPSCs were cultured on a culture dish and fixed using 4% paraformaldehyde (Nacalai).After washing with phosphate-buffered saline (PBS), the cells were incubated at 4°C with primary antibodies overnight.After washing again with PBS, the cells were incubated at room temperature for 1 h and analyzed microscopically (Eclipse Ti, Nikon).

Flow cytometry analysis
The expression of cell surface markers was analyzed by flow cytometry using a FACS Aria II flow cytometer (BD Bioscience, California) and FlowJo version 10 software (FlowJo).cyiPSCs and cells isolated from cyiPSC-derived cartilage were stained with anti-human antibodies at 4°C for 30 min.Table 2 lists the antibodies used.The MHC expression was also analyzed after treating the cells with 25 ng/mL IFNg (Sigma) for 72 h.Isotype control antibodies were used (Table 2).

Histological and immunohistochemical analysis
Samples were fixed with 4% paraformaldehyde, processed and embedded in paraffin.Semiserial sections were prepared for staining with hematoxylin and eosin or Safranin O-fast green-iron hematoxylin (Safranin O) or immunostaining with specific antibodies.Anti-type II collagen antibodies were detected using a CSA II Biotin-free Tyramide Signal Amplification System Kit (Agilent Technologies, California) and DAB as the chromogen.For the anti-CD3 antibodies and anti-NKG2A antibodies, immune complexes were detected using secondary antibodies conjugated to Alexa Fluor 488.Antigens were unmasked by treatment with hyaluronidase and EDTA.Table 2 lists the antibodies used.

Teratoma formation
To form teratomas, cyiPSCs were injected into the testicular capsules of 4-week-old C.B-17/Icr-scid/scid Jcl male mice.Twelve weeks later, the tumors were recovered and subjected to histological analysis.
Mixed lymphocyte assay 5 • 10 5 cynomolgus monkey peripheral blood mononuclear cells (PBMCs; iQ Biosciences) were cocultured with 1 • 10 5 cells isolated from cyiPSC-derived cartilage in RPMI1640 supplemented with 10% FBS in 1 well of a 96well plate for 110 h in the presence or absence of 25 ng/mL IFNg.Then, the proliferation of PBMCs was analyzed using a CellTraceÔ CFSE Cell Proliferation Kit (Thermo Fisher Scientific).In brief, PBMCs were pretreated with carboxyfluorescein diacetate succinimidyl ester (CFSE) before the start of the coculture.The number of PBMC divisions was analyzed flow cytometry after the coculture.As for a positive control, PBMCs were cultured in the presence of 1 mg/mL anti-CD3 antibody (Thermo Fisher Scientific) for 110 h.

Transplantation of cyiPSC-derived cartilage into osteochondral defects in cynomolgus monkey
Cynomolgus monkeys (3-4 years old) were purchased from Ina Research (Nagano, Japan).We created osteochondral defects (1.5 mm in diameter and 1-2 mm in depth) in the trochlea of the distal femur in the right knee joints of four male cynomolgus monkeys.We transplanted B2M +/+ cyiPSC-derived cartilage in the defects in two monkeys and B2M -/-cyiPSC-derived cartilage in the defects in the other two monkeys.After surgery, antibiotics and buprenorphine hydrochloride (0.1 mg/body) were intramuscularly injected for 3 days.Monkeys were sacrificed 4 weeks later by injecting pentobarbital sodium (100 mg/kg) under deep anesthesia, and transplanted sites were harvested and analyzed histologically.Samples were decalcified with KCX (FALMA).

Statistical analysis
The data are shown as averages and standard deviations.In this study, we used two-tailed Student's t-tests.p-Values <0.05 were considered statistically significant.

B2M gene deletion in cyiPSCs
To be expressed on the cell surface, MHC class I molecules form multimers with b2 microglobulin (B2M). 25Therefore, we deleted the B2M gene (B2M) using the CRISPR-Cas9 system (Fig. 1A).A previous report on mice showed that deletion of exon 2 of B2M prevents the expression of B2M protein. 26Therefore, the target sequence for the guide RNA was designed at around the 5¢ end of exon 2 (Fig. 1A).After the nucleofection of 1123C1 cyiPSCs with the guide RNA and Cas9, we obtained 39 clone lines.
Genomic PCR analysis using primers 3 and 4 (Fig. 1B) revealed that the size of the PCR product generated from B2M -/-cyiPSCs was identical or slightly shorter than that from B2M +/+ cyiPSCs [Fig.1D, RE(-)].We then digested the PCR product with restriction enzymes.Digestion with BstXI showed a restriction pattern similar with that of B2M +/+ cyiPSCs except for one fragment (Fig. 1D).Digestion with MfeI also resulted in a restriction pattern similar with B2M +/+ cyiPSCs except for one fragment (Fig. 1D).In both the BstXI and MfeI digestions, the sizes of the one exceptional fragment (Fig. 1D, asterisks) was around 200 bp shorter than the fragment from B2M +/+ cyiPSCs.Digestion of the PCR product from B2M +/+ cyiPSCs with HincII produced expected fragments of 4696 and 374 bp, but that from B2M -/- cyiPSCs produced a single fragment of around 5 kbp (Fig. 1D).These results from the restriction enzyme analysis suggest that B2M lost an *200-bp fragment containing the HincII site (Fig. 1D) and locates between the sequences corresponding to primers 1 and 2 (Fig. 1B).Direct sequencing analysis of the PCR products using primers 1 and 2 revealed that the product from B2M -/-cyiPSCs lacked a 191 bp stretch (Fig. 1E) that includes the target sequence (Fig. 1A) of the guide RNA.The splice acceptor site of exon 2 and the 5¢ part of the 182 bp stretch of exon 2 were deleted from B2M in B2M -/-(Fig.1E).A single product from the PCR amplification and unique nucleotide signals in the direct sequencing analysis suggested that the identical 191 bp stretch was deleted from both alleles in B2M -/-cyiPSCs or that one allele lacks the large fragment.The deletion should cause either a frameshift mutation in exon 2 or exon 2 skipping due to the loss of the acceptor site for splicing.
Expression analysis of MHC class I molecules in B2M -/-cyiPSCs RT-PCR analysis using the forward primer, which corresponds to the junction of exon 1 and exon 2, indicated that amplification did not occur in B2M -/-cyiPSCs regardless of IFNg stimulation (Fig. 2A).Western blot analysis confirmed that B2M -/-cyiPSCs did not express B2M (Fig. 2B).In addition, the expression level of B2M protein was elevated by IFNg stimulation in B2M +/+ cyiPSCs, but not in B2M -/-cy-iPSCs.IFNg induced the expression of MHC class 1 proteins in B2M +/+ cyiPSCs and to a lesser extent in B2M -/-cyiPSCs.

MONKEY IPS CELL-DERIVED CARTILAGE LACKING MHC CLASS I MOLECULES
Flow cytometry analysis indicated that B2M +/+ cyiPSCs expressed MHC class I molecules on the cell surface (Fig. 2C, left, blue line) and that IFNg increased the expression (Fig. 2C, left, red line).On the contrary, no B2M -/-cyiPSCs expressed MHC class I molecules on the cell surface regardless of IFNg stimulation (Fig. 2C, right); thus, MHC class I protein indicated in the western blot (Fig. 2B) was present intracellularly.These results are consistent with the observed absence of B2M, which is needed for MHC class I proteins to form a complex and be expressed on the cell surface.
Differentiation of B2M -/-into cartilage Chondrogenic differentiation of B2M +/+ or B2M -/- cyiPSCs resulted in the formation of spherical particles 1-3 mm in diameter (Fig. 4A).RT-PCR expression analysis revealed the expression levels of genes associated with chondrogenesis, including COL2A1 and ACAN, were elevated, and the expression levels of genes associated with pluripotency were drastically decreased in the particles compared with cyiPSCs (Fig. 4B).Histological analysis revealed that the particles consisted of cells and ECM based on positive staining with safranin O and immunostaining with anti-type II collagen antibody (Fig. 4C, D).These results collectively suggest that both B2M +/+ and B2M -/-cyiPSCs can be chondrogenically differentiated to form cartilage.

Expression analysis of MHC class I molecules in B2M -/-cyiPSC-derived cartilage
As with cyiPSCs, flow cytometry analysis indicated that cells isolated from B2M +/+ cyiPSC-derived cartilage expressed MHC class I molecules weakly on the cell surface (Fig. 4E, left, blue line) and that treatment with IFNg increased the expression level (Fig. 4E, left, red line).On the contrary, no cells isolated from B2M -/-cyiPSC-derived cartilage expressed MHC class I molecules on the cell surface regardless of IFNg stimulation (Fig. 4E, right).

In vitro mixed lymphocyte reaction assay
Next, we performed mixed lymphocyte reaction assays by coculturing allogeneic PBMCs with B2M +/+ or B2M -/- Black lines represent data from the isotype control.Data are representative of two independent experiments.IFNg, interferon g; MHC, major histocompatibility complex; reverse transcription PCR (RT-PCR); SD, standard deviation.cyiPSC-derived cartilage.An analysis using CFSE showed that the proliferation of PBMCs was stimulated by the addition of anti-CD3 antibody, which served as a positive control (Fig. 5A, top row, left panel).In coculture, neither B2M +/+ nor B2M -/- cyiPSC-derived cartilage stimulated the proliferation of PBMCs (Fig. 5A, top row, middle, and right panel).The lack of proliferation may be because the ECM surrounds the chondrocytes in cyiPSC-derived cartilage, preventing the allogeneic PBMCs from directly contacting the chondrocytes.
Then we digested the ECM of B2M +/+ and B2M -/- cyiPSC-derived cartilage and isolated the cells.The cells were subsequently subjected to coculture with allogeneic PBMCs.In this condition, B2M +/+ and B2M -/-cyiPSCderived cartilage cells directly contacted lymphocytes.However, neither B2M +/+ nor B2M -/-cyiPSC-derived cartilage cells stimulated the proliferation of PBMCs regardless of IFNg treatment (Fig. 5A, bottom row).These results suggest that the immunogenicity of both B2M +/+ and B2M -/- cyiPSC-derived cartilage cells is below detectable levels in the in vitro mixed lymphocyte reaction assay.
In vivo allogeneic transplantation of B2M -/- cyiPSC-derived cartilage in osteochondral defects of monkey knee joints Finally, we analyzed the in vivo immunogenicity of B2M -/- cyiPSC-derived cartilage by transplanting B2M +/+ and B2M -/- cyiPSC-derived cartilage into osteochondral defects created in the joint surface of the distal femur in cynomolgus monkey knee joints.Four weeks after the transplantation, histological analysis revealed that both B2M +/+ and B2M -/- cyiPSC-derived cartilage survived and were not rejected In vitro mixed lymphocyte reaction assay using B2M -/-cyiPSC-derived cartilage and B2M -/-cyiPSC-derived cartilage cells.(A) PBMCs were pretreated with CFSE.Top, left, allogeneic PBMCs were cultured with anti-CD3 antibody, which served as a positive control.Top, middle, and right, allogeneic PBMCs were cocultured with B2M +/+ or B2M -/-cyiPSCderived cartilage.Bottom, B2M +/+ or B2M -/-cyiPSC-derived cartilage were digested with trypsin-EDTA and collagenase D, and the released cells were cocultured with allogeneic PBMCs in the absence or presence of IFNg.(B) B2M +/+ or B2M -/-cyiPSCs were cocultured with allogeneic PBMCs.The proliferation of CD3 + CD4 + T cells, CD3 + CD8 + T cells, and CD3 -CD159a + NK cells was analyzed by CFSE labeling.PBMCs were cultured with anti-CD3 antibody as a positive control (second column from left).CFSE, carboxyfluorescein diacetate succinimidyl ester; PBMCs, peripheral blood mononuclear cells; NK, natural killer.but that they also caused an accumulation of lymphocytes in the bone marrow (Fig. 6A).This preliminary study was performed in two monkeys for each B2M +/+ and B2M -/- cartilage transplantation, and the lymphocyte accumulation was reproducible in all four monkeys, suggesting that the lack of MHC class I is not sufficient for eliminating immune reactions against iPSC-derived cartilage transplanted in osteochondral defects.To characterize the cell types that accumulated around transplanted cyiPSC-derived cartilage, we performed immunohistochemical analysis (Fig. 6B).Immunoreactivity against anti-CD3 antibody tended to be weaker around B2M -/-cyiPSC-derived cartilage than around B2M +/+ cyiPSC-derived cartilage.On the contrary, immunoreactivity against anti-NKG2A antibody was stronger around B2M -/- cyiPSC-derived cartilage than around B2M +/+ cyiPSCderived cartilage, suggesting that NK cells rather than T cells accumulated around B2M -/-cyiPSC-derived cartilage.

Discussion
Allogeneic transplanted tissue elicits an immune response when recognized by the recipient's T lymphocytes as ''nonself.'' 27One mechanism involved in the rejection is the activation of lymphocytes by direct contact with intact MHC class I located on the cell surface of the transplanted allogeneic cells.In this study, we created a cyiPSC clone, in which B2M is deleted (B2M -/-).B2M -/-cyiPSCs expectedly lacked MHC class I expression on the cell surface, a property FIG. 6.In vivo allogeneic transplantation of B2M -/- cyiPSC-derived cartilage.(A) Allogeneic B2M +/+ cyiPSC-derived cartilage (top row) and B2M -/- cyiPSC-derived cartilage (bottom row) were transplanted into osteochondral defects created in the joint surface of the distal femur in the knee joints of monkeys.The transplanted sites were analyzed histologically 4 weeks after the transplantation.Left, Safranin O-fast green-iron hematoxylin staining.Middle, HE staining.Right, magnification of the boxed regions in the middle panels.(B) Immunohistochemical analysis of the semiserial sections in (A) using anti-CD3 antibody (top two rows) and anti NKG2A antibody (bottom two rows).Magnification of the boxed regions in the third column is shown in the fourth column.Scale bars, 100 mm.HE, hematoxylin and eosin.that did not affect the pluripotency of cyiPSCs.Indeed, we could generate cartilage efficiently from B2M -/-cyiPSCs as well as B2M +/+ cyiPSCs.B2M -/-cyiPSC-derived cartilage cells lack MHC class I expression on their surface even in the presence of IFNg.Considering the closeness in immune systems between human and monkey, the B2M -/-cyiPSCsand B2M -/-cyiPSC-derived chondrocytes/cartilage created in this study should be useful for estimating the immunogenicity of transplants derived from allogeneic iPSCs and ways to control immune rejection in human.
The in vitro mixed lymphocyte reaction assay revealed that lymphocytes were not stimulated by chondrocytes isolated from B2M +/+ cyiPSC-derived cartilage even in the presence of IFNg.This result can be explained by chondrocytes suppressing the activation of T cells partly through the constitutive expression of B7 family members, which transmit inhibitory signals to T cells. 2 Thus, in vitro mixed lymphocyte reaction assays may not be appropriate for demonstrating the immune evasion effects of B2M -/-cyiPSCderived cartilage cells.
Although allogeneic cartilage transplantation has been performed against chondral defects, in which subchondral bone is maintained, 4 it has not been well studied whether allogeneic cartilage in osteochondral defects, where transplants are exposed to abundant blood flow in the bone marrow, causes immune reactions.In this study, we found that transplanted B2M +/+ cyiPSC-derived cartilage survived in osteochondral defects and was not immune rejected, but it did cause an accumulation of leukocytes including CD3 + T cells around them to some extent.The survival of allogeneic cyiPSC-derived cartilage in osteochondral defects despite the T cell accumulation can be accounted for by a previous study that reported chondrocytes express negative regulators of immune responses, including B7 family members, chondromodulin-I.and indoleamine 2,3-dioxygenase. 2 On the contrary, transplanted B2M -/-cyiPSC-derived cartilage were surrounded by NKG2A + NK cells.This result is consistent with previous finding that cells not expressing HLA class I molecules (HLA-ABC) on their surface elicit an innate immune response by NK cells. 28Following that work, recent studies have developed human iPSCs that lack HLA class I molecules but retain HLA-C 17 or overexpress CD47 29 to suppress the activation of NK cells.On the contrary, chondrocytes constitutively express LLT1, a ligand of inhibitory NKR-P1A NK cell receptor. 30These findings could account for our results, in which B2M -/-cyiPSC-derived cartilage was not rejected after allogeneic transplantation into osteochondral defects in monkey despite the accumulation of NK cells.These findings raise the hypothesis that cartilage derived from iPSCs lacking HLA class I molecules but retaining HLA-C 17 or overexpressing CD47 29 can evade accumulation of NK cells and serve as effective transplants for osteochondral defects.Further studies are needed to understand the immune reactions during the allogeneic transplantation of cartilage.The B2M -/-cyiPSCs-and B2M -/- cyiPSCs-derived chondrocytes/cartilage described in this study will contribute to such research.

FIG. 1 .
FIG. 1. Inactivation of B2M gene in cyiPSCs using the CRISPR/Cas9 system.(A) Schematic representation of the target sequence recognized by the guide RNA.The target sequence for the guide RNA is underlined.The arrow indicates the site of the double strand break.(B) Locations of the sequences corresponding to the primers used for the PCR amplification.(C) PCR amplification with primers 1 and 2 using genomic DNA extracted from B2M +/+ or B2M -/-cyiPSCs.The PCR products were separated on an agarose gel.The expected size of the PCR product from B2M +/+ genomic DNA is 1205 bp.(D) PCR amplification with primers 3 and 4 using genomic DNA extracted from B2M +/+ or B2M -/-iPSCs.The PCR products were digested with restriction enzymes.Top, Schematic representation of the expected PCR products with primers 3 and 4 from the wild-type allele and locations of the sites recognized by the restriction enzymes BstXI, MfeI, and HincII.Bottom, PCR products digested with the restriction enzymes were separated on an agarose gel during electrophoresis.RE (-), without digestion.Asterisks indicate fragments of different size between B2M +/+ and B2M -/-cyiPSCs.(E) The PCR product from B2M -/-genomic DNA was subjected to direct sequencing.B2M, b2 microglobulin; iPSCs, induced pluripotent stem cells; polymerase chain reaction (PCR).

FIG. 2 .
FIG. 2. Expression analysis of B2M and MHC class I molecules in B2M+/+ and B2M -/-cyiPSCs.(A) RT-PCR expression analysis of B2M mRNA.The RNA expression levels were normalized to the level of ACTB expression.Error bars denote means -SD (n = 3 technical replicates).***p < 0.001 by the twotailed Student's t-tests compared with B2M +/+ cyiPSCs without IFNg treatment.Data are representative of three independent experiments.(B) Western blot analysis of MHC class I molecules with or without IFNg treatment.The data are representative of two independent experiments.(C) Flow cytometry analysis of cells without (blue line) or with (red line) IFNg stimulation using anti-MHC class I antibodies.Black lines represent data from the isotype control.Data are representative of two independent experiments.IFNg, interferon g; MHC, major histocompatibility complex; reverse transcription PCR (RT-PCR); SD, standard deviation.

FIG. 3 .
FIG. 3. Pluripotency of B2M -/-cyiPSCs.(A) RT-PCR expression analysis of POU5F1 and NANOG mRNAs.The RNA expression levels were normalized to the level of GAPDH expression.Error bars denote means -SD (n = 3 experiments).***p < 0.001, comparison with MK.P3 (cynomolgus monkey kidney epithelial-like cells, negative control; two-tailed Student's t-test).(B) Immunocytochemical analysis of POU5F1, NANOG, and GFP in B2M +/+ cyiPSCs, B2M -/-cyiPSCs, and MK.P3.The data are representative of four independent experiments.(C) B2M -/- cyiPSCs were inoculated on the testicular capsules of 6 immunodeficient mice.The mice were sacrificed 3 months later, and samples were subjected to histological analysis.Tumors were found in all six mice.Four tumors contained tissues of all three germ layers.Hematoxylin-eosin and safranin O-fast green-iron hematoxylin staining.Magnification of the boxed region in the third panel is shown in the right panel.Scale bars, 100 mm.B2M, b2 microglobulin.