CRISPR/Cas9-Mediated Deletion of Foxn1 in NOD/SCID/IL2rg−/− Mice Results in Severe Immunodeficiency

Immunodeficient mice engrafted with either normal or cancerous human cells are widely used in basic and translational research. In particular, NOD/SCID/IL2rg−/− mice can support the growth of various types of human cancer cells. However, the hairs of these mice interfere with the observation and imaging of engrafted tissues. Therefore, novel hairless strains exhibiting comparable immunodeficiency would be beneficial. Recently, the CRISPR/Cas9 system has been used for efficient multiplexed genome editing. In the present study, we generated a novel strain of nude NOD/SCID/IL2rg−/− (NSIN) mice by knocking out Foxn1 from NOD/SCID/IL2rg−/− (NSI) mice using the CRISPR/Cas9 system. The NSIN mice were deficient in B, T, and NK cells and not only showed impaired T cell reconstitution and thymus regeneration after allogeneic bone marrow nucleated cell transplantation but also exhibited improved capacity to graft both leukemic and solid tumor cells compared with NSI, NOG, and NDG mice. Moreover, the NSIN mice facilitated the monitoring and in vivo imaging of both leukemia and solid tumors. Therefore, our NSIN mice provide a new platform for xenograft mouse models in basic and translational research.

Various methods have been developed for genome modification, including designer zinc finger nucleases, transcription activator-like effector nucleases, and the type II bacterial CRISPR/Cas9 system. Recently, the CRISPR/Cas9 system has been shown to be suitable for multiplexed genome editing 28,29 . The ease of design, construction, and delivery of multiple sgRNAs by co-microinjection [30][31][32] of Cas9 mRNA suggest that this system can be used to generate a variety of novel immunodeficient mouse strains.
In the present study, we derived a Foxn1-mutated NOD/SCID/IL2rg −/− mouse (NSIN) strain using the CRISPR/Cas9 system. Cas9 mRNA and gRNA targeting Foxn1 were injected into the cytoplasm of pronuclear-stage NSI mouse embryos. The mutant offspring were mated to generate homozygous NSIN mice. The NSIN mice were hairless and deficient in B, T, and NK cells and exhibited an enhanced engrafting capacity for both leukemia and solid tumors compared with NSI, NOG, and NDG mice. Moreover, the hairlessness facilitated tumor observation and imaging. Our study shows that NSIN mice can be used to generate ideal models for basic and translational research.

Results
Efficient modification of Foxn1 in PL08 cells in vitro using CRISPR/Cas9. First, to test the targeting accuracy and efficiency of our CRISPR/Cas9 system, we designed gRNA targeting the first exon 33 of murine Foxn1 (Fig. 1A) and transfected plasmids expressing mammalian codon-optimized Cas9 and gRNA into a murine PL08 cell line 16 (Fig. 1B). Twenty-four hours later, transfected cells were selected via a72-h treatment with500 μg/ ml G418, and cell clones were then selected. DNA was extracted from twenty cell clones to determine their genotypes in each experiment. DNA sequencing revealed cell clones that carried the expected mutation at the target locus (Fig. 1C). The knock-out efficiency of Foxn1 in PL08 cells was approximately 20% (Fig. 1D). These data demonstrated the specific and efficient targeting of Foxn1 by our CRISPR/Cas9 system. Generation of NSIN mice by deletion of Foxn1. Next, using the CRISPR/Cas9 system with multiplexable genome engineering capabilities, we attempted to knock-out the Foxn1 gene from the NOD/SCID/IL2rg −/− background (NSI mice). Foxn1-deleted mice were generated through direct embryo manipulation ( Fig. 2A). After in vitro transcription, a mixture of Cas9 mRNA (20 ng/μl) and gRNA for Foxn1 (20 ng/μl) was microinjected into the cytoplasm of pronuclear-stage embryos of NSI mice 28 . Blastocysts derived from the injected embryos were transplanted into foster mothers, and 14 newborn pups were obtained (Table 1). Genomic DNA was extracted from the pups for PCR amplification. DNA sequencing revealed that one mouse carried the expected mutation at the target locus (Fig. 2B). A new AluI restriction enzyme recognition site was generated by deleting a thymidine (Fig. 2B). Then, a 123-bp fragment spanning the target site was amplified using PCR. The PCR products were digested with the AluI enzyme. AluI digestion of the PCR products of wild-type, heterozygous, and homozygous offspring generated fragments with lengths of 123 bp, 123 + 98 + 25 bp, and 98 + 25 bp, respectively (Fig. 2C). Using this restriction fragment length polymorphism (RFLP) assay, we could rapidly and accurately identify the genotypes of offspring mice. Due to the loss of Foxn1, homozygous NSIN female mice are poor breeders and fail to lactate. Heterozygous (Foxn1 +/− NSI) female mice breed well and were mated with homozygous male NSIN mice to generate the NSIN offspring. The average litter size was approximately 6 to 8 pups per litter.
Overall, approximately half of the offspring were NSIN mice. The NSIN mice were athymic and hairless (Fig. 2D). Furthermore, B, T, and NK cells were absent in the NSIN mice, similar to NSI, NOG, and NDG mice (Fig. 2E). In summary, we generated a novel strain of nude mice by knocking out the Foxn1 gene from NOD/SCID/IL2rg −/− (NSI) background mice using CRISPR/Cas9.

Loss of Foxn1 impairs T cell development and thymus regeneration in NSIN mice.
Foxn1 is a critical regulator of thymic and T cell development. To compare the capacities of adoptive T cell development between NSI and NSIN mice, we measured the reconstitution of allogeneic bone marrow nucleated cells (BMNCs) (Fig. 3A). After 4 weeks, the reconstitution of total donor cells was similar between NSI and NSIN mice (Fig. 3B); however, T cell development from BMNCs was markedly impaired in NSIN mice compared with NSI mice (Fig. 3C), as indicated by the percentages of donor T cells in the peripheral blood (PB) (Supplemental Fig. 1A), spleen (SP), and bone marrow (BM). Moreover, both CD4 and CD8 T cells were present in NSIN and NSI mice, but the CD4 − CD8 − compartment was increased in NSIN mice compared with NSI mice (Supplemental Fig. 1B). Strikingly, 4 weeks after BMNC transplantation, thymi were regenerated in NSI mice, but not in NSIN mice (Supplemental Fig. 1C), confirming that Foxn1 is a critical regulator of thymic development. In the regenerated thymus, most thymocytes were derived from the donor mice, and most T cells were CD4 + CD8 + (Supplemental Fig. 1D).

Foxn1 deletion results in improved engraftment of leukemic cells in NSIN mice.
Immunodeficiency is positively correlated with tumor engraftment capacity 12,34,35 . NOD/SCID/IL2rg −/− mice have been widely used in translational studies of human immunology and hematological malignancies 36 . To evaluate the effect of Foxn1 deletion on the engraftment capacity of hematological cells, we compared the engraftment of a human B cell acute lymphoblastic leukemia (B-ALL) cell line, Nalm6, in NDG, NOG, NSI, and NSIN mice. First, Nalm6 cells were labeled with green fluorescent protein (GFP)-luciferase (GL) using lentiviral transfection, followed by the FACS sorting of GFP + cells (Fig. 4A). Subsequently, Nalm6-GL cells were injected into NDG, NOG, NSI, and NSIN mice at low (n = 5, for each strain) (Fig. 4B), medium (n = 5, for each strain) (Fig. 4C), and high (n = 5, for each strain) (Fig. 4D) doses via the tail vein without preconditioning. Upon exhibiting symptoms of paralysis, the mice were euthanized. The proportions of GFP + Nalm6-GL cells in the PB, SP, and BM were higher in the NSIN mice than in the NDG, NOG, and NSI mice ( Fig. 4B-D, Supplementary Figure 2). To ensure  that GFP + cells were Nalm6-GL cells, we also analyzed human CD19 and found that all GFP + cells were tumor cells (Fig. 4E). Thus, the NSIN mice exhibited an enhanced capacity to engraft leukemic cells compared with the NDG, NOG, and NSI mice.

Foxn1 deletion in NSIN mice results in improved engraftment of solid tumors.
To evaluate the solid tumor engraftment capacity of NSIN mice, we compared the engraftment of a human lung adenocarcinoma cell line, A549, in NDG, NOG, NSI, and NSIN mice. A549-GL cells were generated similarly to Nalm6-GL cells (Fig. 5A). A549-GL cells were subcutaneously injected into NDG, NOG, NSI, and NSIN mice at low (n = 5, for  Table 1. CRISPR/Cas9-mediated Foxn1 gene targeting in NSI mice. Cas9 mRNA and sgRNA targeting Foxn1 were injected into fertilized eggs. The blastocysts derived from injected embryos were transplanted into foster mothers and newborn pups were obtained and genotyped. The number of total alleles mutated in each mouse is listed from 0 to 2.  group. Cells were gated from FSC/SSC. Error bars denote the standard error of the mean (SEM). Groups were compared using a two-tailed unpaired t-test. *P < 0.05; **P < 0.01; ***P < 0.001. each strain), medium (n = 5, for each strain), and high doses (n = 5, for each strain). The A549-GL tumors were significantly larger in the NSIN mice than in the NDG, NOG, and NSI mice at all three doses (Fig. 5B-D). NOG, NSI, and NSIN mice bearing subcutaneous A549-GL tumors in the medium-and high-dose groups are shown (Fig. 5E,F), and tumors were most clearly visible in the NSIN mice. In contrast to the NOG and NSI mice, the blood vessels of the tumors were also clearly visible in the NSIN mice.

NSIN mice exhibit higher tumor engraftment index (TEI) scores. Previously, we developed a TEI
and a statistical formula 15 for the simple and accurate quantification of the immunodeficiency of mouse strains. The TEI scores for NDG, NOG, NSI, and NSIN mice engrafting Nalm6-GL cells were calculated as previously described 15 (Fig. 6A). The differences between the TEI scores of these three strains were small, indicating significant engraftment of human acute leukemia cells by these strains, consistent with previous results 12 . However, the TEI score of the NSIN mice engrafting A549-GL cells was higher than those of the NDG, NOG, and NSI mice (Fig. 6B). The overall TEI scores for hematological and solid cancers were also calculated, which indicated an overall increase in the immunodeficiency of the NSIN mice compared with the NDG, NOG, and NSI mice (Fig. 6C). Groups were compared using a two-tailed unpaired t-test. *P < 0.05; **P < 0.01; ***P < 0.001.

NSIN mice facilitate live fluorescence imaging of xenografts.
The hairless phenotype of NSIN mice could be advantageous in tumor monitoring and imaging. Indeed, the cells of another human large cell lung cancer cell line, H460-GL, were injected into NSIN and NSI mice, and there were no differences in the subcutaneous tumors between the NSIN and NSI mice in terms of tumor size or luciferase-derived bioluminescence. However, tumors in NSIN mice showed more condensed and sharper GFP fluorescence than those in NSI mice (Fig. 7A). Likewise, CD123 + human acute myeloid leukemia (AML) Molm13-GL cells were intravenously injected into NSIN and NSI mice, and GFP signals emanating from the SP and BM of the NSIN mice were more distinct than those of the NSI mice. Moreover, the tumor burden of Molm13-GL cells was higher in the NSIN mice than in the NSI mice (Fig. 7B). These results indicated that NSIN mice are more suitable hosts for the live imaging of autofluorescence derived from xenografts.

Discussion
In basic and translational research, it is often desirable to analyze complex biological processes in vivo, which has led to a demand for new animal models. NSI mice have been used in translational biomedical research in our lab. However, a nude strain with comparable immunodeficiency has not been previously reported. In the present study, we generated nude NSI mice using the CRISPR/Cas9 system in the background of NSI mice via embryonic co-microinjection with Cas9 mRNA and gRNA. The mutant offspring were mated to generate homozygous NSIN mice. These mice were hairless and exhibited more severe immunodeficiency than NSI mice. Hence, NSIN mice provide a new platform for basic and translational research. Athymic nude mice with a deletion of Foxn1 have been widely used in cancer research. Foxn1 is required to maintain the postnatal thymus, and changes in Foxn1 expression in TECs may contribute to thymus involution during aging 24 . Our study also revealed that the loss of Foxn1 resulted in the blockade of thymic development following allogeneic BMNC transplantation, which confirmed a critical role for Foxn1 in thymic function. The postnatal thymus is the primary source of T cells in vertebrates, and thymocyte development requires interactions with TECs. Specifically, Foxn1 controls the transcription of genes involved in the attraction and lineage commitment of T cell precursors and regulates the expression of genes involved in antigen processing and thymocyte selection 25 . Consistently, T cell development from allogeneic BMNC in NSIN mice was also impaired by the deletion of Foxn1 in our study. Extrathymic T cell development is well established 37 and could be responsible for the reduced but still present level of T cells in NSIN mice, especially in the SP.
Nude mice are suitable for research requiring whole-body fluorescence-based imaging techniques [38][39][40] . The novel NSIN strain derived using the CRISPR/Cas9 system exhibited a hairless phenotype. This was advantageous for the observation and imaging of tumors by GFP or other fluorescence. For example, the blood vessels of subcutaneous tumors could be seen through the skin of NSIN mice, making NSIN mice a potentially ideal model for xenograft angiogenesis studies. In addition, NSIN mice exhibited higher immunodeficiency than NSI mice, as indicated by their TEI scores. The elevation of TEI scores caused by Foxn1 deletion was more significant for subcutaneous solid tumors than for hematological cancers (Fig. 5). In our study, NSI mice engrafted tumors more efficiently than NOG mice, which could be attributed to the presence of the extracellular domain of IL2rg in NOG mice but the complete deletion of IL2rg in NSI mice 3 . Foxn1 inactivity in nude mice maintains the skin in an immature state resembling neoteny [41][42][43] . Thus, NSIN mice may also be useful for generating humanized skin models to study human skin regeneration.
In sum, our study generated a new mouse strain and demonstrated the advantages of this strain in xenograft model generation and tumor surveillance.
Cell culture and transfection. PL08 is an immortalized murine fetal liver cell line that has been previously reported 16 . Millicell Hanging Cell Culture Inserts (Millipore, Darmstadt, Germany) were used to culture PL08 cells. PL08 cells (1 × 10 6 ) were transfected with U6-sgRNA vectors (10 μg) and pcDNA3.3-hCas9 plasmids (30 μg). Cells were selected using G418 (500 μg/ml; Sigma) after 24 h of transduction and were maintained in selection medium for 72 h. Single clones were collected and seeded into each well of a 96-well plate. Monoclonal cells were expanded. Then, monoclonal cell DNA was extracted to determine the cell genotypes by DNA sequencing. Nalm6, Molm13, A549, and H460 cells were cultured in RPMI-1640 media (Gibco, Grand Island, NY, USA) with 10% fetal bovine serum (FBS; Biochrom, Australia) and 1% penicillin/streptomycin. HEK-293T cells were used for lentivirus production and cultured in Dulbecco Scientific, Waltham, MA, USA) and used as the template for in vitro transcription (IVT) using an mMESSAGE mMACHINE T7 ULTRA Transcription Kit (Ambion, AM1345). Cas9 mRNA was purified using an RNeasy Mini Kit (Qiagen, 74104). The gRNA templates for the in vitro transcription of RNA were purified PCR products obtained from pUC57-T7-Foxn1-gRNA vectors using a primer pair (T7-F: 5′-gaaattaatacgactcactata-3′ and T7-R: 5′-aaaaccgactcggtgccacaaaagc-3′) and a high-fidelity enzyme (Takara). The T7-sgRNA PCR product was gel-purified and used as the template for IVT with a MEGAshortscript T7 Transcription Kit (Ambion, AM1354). The gRNA was purified using a MEGAclear Kit (Ambion, AM1908) and concentrated by alcohol precipitation.
Microinjection of single-cell embryos. NSI and ICR/HaJ mouse strains were used as zygote donors and foster mothers, respectively. Female NSI mice (aged 8-10 w) were super-ovulated using intraperitoneal injections of pregnant mare serum gonadotropin (5 IU; Sigma) and human chorionic gonadotropin (5 IU; Sigma) at 48-h intervals. The super-ovulated female mice were mated to NSI male mice, and the fertilized embryos were collected from the oviducts. Cas9 mRNA (20 ng/μl) and sgRNA (20 ng/μl) were injected into the cytoplasm of fertilized eggs with visible pronuclei in M2 medium (Sigma) using a piezo-driven micromanipulator. The injected zygotes were cultured in potassium simplex optimized medium with amino acids at 37 °C under 5% CO 2 in air until the blastocyst stage (3.5 d). The surviving embryos were selected and transferred into the uterus of pseudo-pregnant foster mothers. The resulting offspring were analyzed for edited Foxn1 genes. DNA extraction and genotyping. Monoclonal cells and genomic DNA from mouse tail samples were extracted using a Universal Genomic DNA Extraction Kit Ver.3.0 (DV811A, Takara, Japan) according to the manufacturer's instructions. Genomic DNA was subjected to PCR amplification, and mutations were identified by direct sequencing. For sequencing, the primer pair of Foxn1-F1 (5′-aatttctcaccttggctatc-3′)/Foxn1-R1 (5′-caggagtcccaaagtgacgg-3′) was used to amplify 644-bp PCR products spanning the target site. After identifying the mutation site, the primer pair Foxn1-F2 (5′-ttcgaggccaggactgggtg-3′)/Foxn1-R2 (5′-ttacgttctgtggggcaggg-3′) was used to amplify 123-bp PCR products spanning the mutation site. The PCR products were digested with the AluI enzyme (Fermentas; Thermo Fisher Scientific).

Engraftment of leukemia and solid tumors.
To directly compare the efficiency of cancer cell engraftment, 1 × 10 4 (L), 1 × 10 5 (M), or 1 × 10 6 (H) NALM6-GL cells suspended in phosphate buffer solution (PBS; 0.2 mL) were injected into the tail veins of NOG, NSI and NSIN mice. Similarly, 1 × 10 4 (L), 1 × 10 5 (M), or 1 × 10 6 (H) A549-GL cells suspended in PBS (0.2 mL) were injected subcutaneously into NOG, NSI, and NSIN mice. The end points were based on animal models in widespread use. The engraftment of leukemia was measured by analyzing the GFP + cells in the PB either weekly or when the mice were moribund after grafting.
Live fluorescence imaging. In vivo whole-body imaging of GFP-labeled cells was performed using a cooled CCD camera system (IVIS 100 Series Imaging System, Xenogen, Alameda, CA, USA). Before in vivo imaging, mice injected with H460-GL cells or Molm13-GL cells were anesthetized with isoflurane. Quantification of total and average emissions was performed using Living Image software (Xenogen).
Quantification of mouse immunodeficiencies (TEI). The hematologic TEI was calculated using the following equation: I strain−hematologic tumor−n = Gi/Di, where "strain" is the immunodeficient mouse strain, "hematologic tumor" is the type of tumor cells, n is the number of individuals, "Gi" is the sum of the percentages of tumor cells in the peripheral blood (GPB), bone marrow (GBM), and spleen (GSP) of an individual mouse either during morbidity or after death, and "Di" is the lifespan of the individual mouse after injection of the tumor cells. The solid TEI was calculated using the following equation: I strain−solid tumor−n = Wi/D, where "strain" is the immunodeficient mouse strain, "solid tumor" is the type of transplanted tumor cells, n is the number of individuals, "Wi" is the weight of the graft in an individual mouse either during morbidity or after death, and "D" is the survival time of the mouse after injection of the tumor cells. The total TEI score was determined as follows: I strain−xenograft−n = (I strain−Nalm6−n + I strain−A549−n )/2. Calculations were performed using tools on the website of In Vivo Biomedical (http://www.nsitei.com) 15 .
Statistical analysis. The data are presented as the mean ± standard deviation. Student's t-test was used to determine the statistical significance of differences between samples. P < 0.05 was considered statistically significant. All statistical analyses were performed using Prism software, version 5.0 (GraphPad, Inc., San Diego, CA, USA).