Differences between immunodeficient mice generated by classical gene targeting and CRISPR/Cas9-mediated gene knockout

Differences between immunodeficient mice generated by classical gene targeting and... Transgenic Res (2018) 27:241–251 https://doi.org/10.1007/s11248-018-0069-y ORIGINAL PAPER Differences between immunodeficient mice generated by classical gene targeting and CRISPR/Cas9-mediated gene knockout . . . . Jae Hoon Lee Jong-Hyung Park Tae-Wook Nam Sun-Min Seo . . . . Jun-Young Kim Han-Kyul Lee Jong Hyun Han Song Yi Park Yang-Kyu Choi Han-Woong Lee Received: 6 November 2017 / Accepted: 19 March 2018 / Published online: 28 March 2018 The Author(s) 2018 Abstract Immunodeficient mice are widely used for mature B cells and T cells and an increase in the ? ? pre-clinical studies to understand various human number of CD45 DX-5 natural killer cells. How- -/- diseases. Here, we report the generation of four ever, B6-Il2rg mice had a unique phenotype, with a immunodeficient mouse models using CRISPR/Cas9 lack of mature B cells, increased number of T cells, system without inserting any foreign gene sequences and decreased number of natural killer cells. Addi- such as Neo cassettes and their characterization. By tionally, serum immunoglobulin levels in all four eliminating any possible effects of adding a Neo immunodeficient mouse models were significantly cassette, our mouse models may allow us to better reduced when compared to those in wild-type mice -/- elucidate the in vivo functions of each gene. Our FVB- with the exception of IgM in B6-Il2rg mice. These -/- -/- -/- Rag2 , B6-Rag2 , and BALB/c-Prkdc mice results indicate that our immunodeficient mouse showed phenotypes similar to those of the earlier models are a robust tool for in vivo studies of the immunodeficient mouse models, including a lack of immune system and will provide new insights into the variation in phenotypic outcomes resulting from different gene-targeting methodologies. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s11248-018-0069-y) con- Keywords Il2rg  Prkdc  Rag2  CRISPR/Cas9 tains supplementary material, which is available to authorized Immunodeficient mouse models users. Jae Hoon Lee and Jong-Hyung Park have contributed equally to this work. Introduction J. H. Lee  T.-W. Nam  J. H. Han  S. Y. Park H.-W. Lee (&) Department of Biochemistry, College of Life Science and Since severe combined immunodeficiency (SCID) Biotechnology, Laboratory Animal Research Center, mice lacking functional B and T lymphocytes were Yonsei University, Seoul 03722, Republic of Korea first spontaneously discovered in a colony of C.B-17 e-mail: hwl@yonsei.ac.kr mice (Bosma et al. 1983), many immunodeficient J.-H. Park  S.-M. Seo  J.-Y. Kim  H.-K. Lee  mice have served as invaluable model organisms in Y.-K. Choi (&) both biological and clinical studies (Ito et al. 2002; Department of Laboratory Animal Medicine, College of Ohbo et al. 1996; Shinkai et al. 1992). Among them, Veterinary Medicine, Konkuk University, Seoul 05029, Rag2-deficient mice fail to generate mature B and T Republic of Korea cells due to the impairment of the V(D)J e-mail: yangkyuc@konkuk.ac.kr 123 242 Transgenic Res (2018) 27:241–251 rearrangements that are crucial to initiating the Importantly, our models lack PGK-neo cassettes that development of functional immunoglobulin (Ig) and are usually included in existing knockout mouse T cell receptors (TCR) (Shinkai et al. 1992). Similarly, models. We evaluated whether CRISPR/Cas9-medi- deficiency of a DNA-dependent protein kinase cat- ated knockout mice exhibit phenotypes different from alytic subunit (DNA-PKcs; encoded by Prkdc gene) those of previous immunodeficient mouse models that causes a SCID phenotype that is characterized by an were generated by classical gene targeting using ES absence of functional B and T cells, lymphopenia, cells. hypogammaglobulinemia, but a normal hematopoietic microenvironment (Buckley et al. 1997; Puck et al. 1997). Therefore, SCID mice are acceptable for both Materials and methods allogeneic and xenogeneic cell transfer experiments. Unlike other immunodeficient mice lacking B and T Animals and ethics statement cell development, however, the absence of IL2rc in mice disrupts not only B and T cells but also natural BALB/cAnNTac (BALB/c), C57BL/6JBomTac (B6), killer (NK) cells (Belizario 2009). This permits the FVB/NTac (FVB), and IcrTac:ICR (ICR) mice were engraftment with human hematopoietic cells to estab- purchased from Taconic Biosciences (Dae Han lish a human immune system in this mouse model. Biolink Co., Ltd., Chungbuk, Republic of Korea). Humanized mice containing human cells or tissues are All mice were housed in the specific pathogen-free becoming increasingly important as animal models for (SPF) facility of the Yonsei Laboratory Animal studying basic and applied human diseases (Honeycutt Research Center. All efforts were made to minimize et al. 2015; Ito et al. 2012; Walsh et al. 2017). To animal suffering, and all animal experiments were improve the utility of immunodeficient mouse models, conducted in accordance with the Korean Food and we generated four different models using the CRISPR/ Drug Administration (KFDA) guidelines. Experimen- Cas9 system and evaluated whether different back- tal protocols were reviewed thoroughly and approved ground strains or targeted genes affect the immunod- by the Institutional Animal Care and Use Committees eficiency phenotypes. (IACUC) at Yonsei University (Permit Number: The CRISPR/Cas9 system can generate improved 201506-322-02). All immunodeficient mouse models genetically modified mouse models, because it does presented in this study will be made readily available not require the inclusion of a drug resistance marker to the research community. such as a PGK-neo cassette which is required to generate ES cell-based knockout mice. The Cas9 Preparation of CRISPR/Cas9 mRNA derived from Streptococcus pyogenes recognizes 5 - NGG-3 as the protospacer adjacent motif (PAM), and The mMESSAGE mMACHINE T7 Ultra kit (Am- the sgRNA consists of a sequence complementary to bion) was used to obtain the Cas9 mRNA, which was the 20 nucleotides upstream of the PAM to identify the diluted in diethyl pyrocarbonate (Sigma)-treated target site (Wright et al. 2016). The double-strand injection buffer (0.25 mM EDTA, 10 mM Tris, pH breaks (DSBs) generated by the Cas9/sgRNA ribonu- 7.4) to obtain the working concentration. Additionally, cleoprotein complex lead to either non-homologous the MEGAshortscript T7 Transcription kit (Ambion) end joining (NHEJ) or homologous recombination was used to synthesize sgRNAs from PCR-generated (HR) repair—mechanisms that are used to generate templates. Plasmids encoding S. pyogenes Cas9 knockout and knock-in mice, respectively (Lieber (SpCas9) protein (Cho et al. 2013) were obtained 2010; Vasquez et al. 2001). Without a template for from ToolGen, Inc. (Seoul, Republic of Korea). HR-mediated DNA repair, the DSB will be repaired through the error-prone NHEJ pathway, forming Microinjection insertion and deletion (indels) mutations in the target gene that may lead to frameshift mutations (Lieber To generate immunodeficient mice using CRISPR/ 2010). Loss of gene function is often induced by Cas9, microinjection of fertilized embryos was per- premature stop codons downstream of the frameshift formed: initially, 6–8-week-old BALB/c, B6, and mutation using the CRISPR/Cas9 system approach. FVB mice were super-ovulated by intra-peritoneal 123 Transgenic Res (2018) 27:241–251 243 injections of 5 IU pregnant mare serum gonadotropin Santa Cruz Biotechnology, Santa Cruz, TX, USA), (Sigma) and 5 IU human chorionic gonadotropin Alexa Fluor 488-anti-CD4 (RM4-5; BD Biosciences), (Sigma) at 48-h intervals. The fertilized embryos were APC-anti-CD8a (53-6.7; BioLegend), PE-anti-CD45 then collected from the super-ovulated mice crossed (30-F11; BD Biosciences), and APC-anti-CD49b with stud males. A mixture of 50 ng/lLof Cas9 (DX-5; BD Biosciences) antibodies. At least 10,000 mRNA and 250 ng/lL of sgRNA was microinjected live cells were analyzed with the FACSCalibur system into the cytoplasm of zygotes, using a piezo-driven (BD Biosciences). manipulator (Prime Tech) to induce mutations, and the resulting embryos were transferred into the oviducts of Enzyme-linked immunosorbent assay (ELISA) ICR pseudo-pregnant foster mothers to produce live mice. Serum samples were collected from the caudal veins of both homozygous mutant and wild-type mice and Founder screening and genotyping PCR then separated in a refrigerated centrifuge at 4 C. Serum immunoglobulin levels of mice were measured To screen founder mice for endonuclease-mediated using the mouse Ready-SET-Go Kits plate (eBio- mutations such as indels, PAGE-PCR assays were science, San Diego, CA, USA) following the manu- performed using genomic DNA samples from tail facturer’s protocol. Briefly, 96-well ELISA plates biopsies (Zhu et al. 2014). In brief, the genomic were coated with the purified anti-mouse IgG1, IgG2a, regions spanning the sgRNA target site were amplified IgG2b, IgG3, IgA, or IgM monoclonal antibody in by PCR. By simply denaturing and annealing, PCR coating buffer at 4 C overnight and then blocked with products containing a mixture of mutant and wild-type diluted assay buffer for 2 h at room temperature, alleles form heteroduplex DNA and homoduplex followed by washing two times. Serial dilutions of pre- DNA as described previously (Zhu et al. 2014). Since titrated standards (IgG1, IgG2a, IgG2b, IgG3, IgA, heteroduplex DNA migrates slower than homoduplex and IgM) and 50 lL of diluted mouse sera were added DNA under a nondenaturing condition, PCR products and the plates incubated for 1 h at room temperature. carrying indels were analyzed by acrylamide gel After washing four times, avidin-horseradish peroxi- electrophoresis. The PCR products from founder mice dase-conjugated antibodies were added and the plates were also cloned using the T-Blunt PCR Cloning Kit incubated again for 1 h at room temperature. After (SolGent Co., Ltd., Republic of Korea) for sequence another four washes, the samples were incubated with analysis, and the mutations were identified by direct the substrate 3,3,5,5—tetramethylbenzidine (TMB) sequencing analysis (Cosmobiotech Co., Ltd., Repub- for 15 min followed by the addition of TMB stop lic of Korea). Those that exhibited the earliest solutions. The absorbance was read in a Tecan premature stop codons were selected as founder mice. spectrometer plate reader (Ma¨nnedorf, Switzerland) at 450 nm with a background subtraction of 570 nm. Antibodies and flow cytometry Immunohistochemical staining Eight-week-old mice were sacrificed and single cells from the spleen, thymus, and bone marrow were Formalin-fixed, paraffin-embedded samples including prepared for flow cytometry analysis. To obtain the spleen, thymus, and lymph nodes, of 8-week-old homozygous null and wild-type mice were used for single-cell suspensions, tissues were cut into small pieces with scissors and passed through a 70-lm cell immunohistochemical staining. Anti-B220 (RA3- strainer (BD Biosciences, Franklin Lakes, NJ, USA) 6B2; Abcam, Cambridge, UK) and anti-CD4 by pressing with a plunger. The remaining red blood (EPR19514; Abcam) antibodies were used for B cell cells were eliminated by using 1 9 RBC lysis buffer and T cell immunostaining, respectively. (eBioscience, San Diego, CA, USA). Prepared cells were stained with FITC-anti-B220 (RA3-6B2; BD Statistical analysis Biosciences), PerCP/Cy5.5-anti-IgM (RMM-1; BioLegend, San Diego, CA, USA), APC-anti-CD3e Statistically significant differences between groups (145-2C11; BD Biosciences), PE-anti-TCR (H57-597; were calculated using the unpaired t-test (GraphPad 123 244 Transgenic Res (2018) 27:241–251 Software, Inc., La Jolla, USA). The results were termination of each target protein (Fig. S1). These considered statistically significant at P \ 0.05, mutations introduced premature termination codons P \ 0.01, or P \ 0.001. (PTCs) into the open reading frames of the target genes, leading to loss of gene function. All immunodeficient mouse strains were then validated in offspring derived Results from each founder mouse by sequence analysis. The homozygous knockout mice (F2) were generated from Generation of immunodeficient mice carrying out- an intercross between F1 heterozygous knockout mice of-frame alleles of Rag2, Prkdc, or Il2rg and confirmed by PCR genotyping (Fig. S1 and Table S2). Since mRNAs with PTCs are targeted by the To generate immunodeficient mice carrying disrup- nonsense-mediated decay (NMD) pathway, RT-PCR tions of the Rag2, Prkdc,or Il2rg genes, we used analysis showed significant reduction of each mRNA in CRISPR/Cas9 system. Each sgRNA was designed to the immune system tissues of the homozygous knock- cleave downstream of the start codon of the genes of out mice compared with the wild-type mice, except for -/- -/- interest to induce frameshift mutations caused by non- the thymus of B6-Rag2 and BALB/c-Prkdc mice homologous end joining (NHEJ)-mediated DNA (Fig. S2). repair. To avoid potential off-target effects of each -/- sgRNA, we designed several sgRNAs using CRIPSR Immunophenotypes of Rag2 mice Design (http://crispr.mit.edu) and sgRNA Design Tool -/- (https://portals.broadinstitute.org/gpp/public/analysis- Gross examination of Rag2 mice (both FVB and tools/sgrna-design), and then intentionally chose the B6) revealed that its spleen and thymus were smaller sgRNAs to recognize the specific sequence that con- than those of wild-type mice. Furthermore, the spleen- tains no apparent homology with the mouse genome to-body weight ratios of both types and both sexes of -/- sequence. We directly injected Cas9 mRNA with the Rag2 mice were significantly lower than that of target-specific sgRNA(s) into one-cell embryos and wild-type mice (P \ 0.05; Fig. 1). Flow cytometric screened the mutants by PAGE-PCR assay using analyses showed a markedly decreased population of ? ? genomic DNAs obtained from the newborns. We suc- mature B cells, especially B220 IgM cells in the cessfully generated several mutant mice with high spleen (P \ 0.001; Figs. 2a, 3a), suggesting that Rag2 efficiency (Table S1) and selected founder mice (F0) deficiency leads to impaired B cell differentiation -/- harboring nucleotide excision leading to premature (Shinkai et al. 1992). Both FVB-Rag2 and B6- a b * ** 2.5 -/- Wild-type B6-Il2rg Wild 2.0 Homo Hemizygote 1.5 1.0 ** *** **** *** 0.5 0.0 ♂♀♂♀♂♀♂♀ -/- -/- -/- FVB-Rag2 B6-Rag2 B6-Il2rg BALB/c- -/- Prkdc -/- Fig. 1 Spleen-to-body weight ratio in immunodeficient mice. Unlike other mutant strains, B6-Il2rg mice showed severe -/- -/- -/- Spleens of FVB-Rag2 , B6-Rag2 , and BALB/c-Prkdc splenomegaly. Each value represents the mean ± SD (n = 4–5 mice were significantly smaller than those of wild-type mice. mice per group). *P \ 0.05, **P \ 0.01, and ***P \ 0.001 Spleen/body weight ratio (%) BM Thymus Spleen Transgenic Res (2018) 27:241–251 245 -/- -/- Wild-type FVB-R Rag2 Wild-type FVB-Rag2 0.2 52.4 0.2 5.1 0.5 8.8 0.2 1.0 11.1 11.5 18.8 7.5 B220 B220 b e 11.7 17.7 0.4 3.3 11.0 16.8 0.3 1.8 0.9 0.7 4.0 1.1 TCR TCR 0.0 10.4 0.0 41.9 2.8 77.9 1.2 1.9 10.1 1.6 87.6 50.7 CD45 CD4 Fig. 2 Flow cytometric analysis of lymphoid cells in 8-week- f Analysis of T cells in the thymus. Each value represents the -/- old FVB-Rag2 mice. a–c analysis of B, T, and NK cells in mean ± SD. (n = 10 mice per group) the spleen. d Analysis of B cells in the bone marrow. e, -/- ? ? Rag2 mice exhibited a significantly decreased Although the ratio of TCR CD3e double-positive ? ? number of B220 IgM B cells in the bone marrow cells in the spleen was higher in wild-type B6 mice where B lymphocytes mature (P \ 0.001; Figs. 2d, (30.1%) than that in wild-type FVB mice (17.7%), -/- -/- 3d). both FVB-Rag2 and B6-Rag2 mice exhibited ? ? ? ? Early T cells differentiate from CD4 CD8 dou- similar decreases in the TCR CD3e double-positive ? ? ble-positive T cells into CD4 or CD8 single- thymocyte population compared to wild-type mice. positive mature T cells, with the latter also expressing The NK cell population was confirmed by exami- ? ? -/- T-cell receptors (TCR) and the CD3 complex (Ger- nation of CD45 DX-5 marker expression in Rag2 ? ? ? ? main 2002). The proportion of CD4 CD8a double- mice. Compared to the wild-type, CD45 DX-5 ? ? positive thymocytes, mature CD4 or CD8a single- double-positive NK cells in the spleen were signifi- ? ? positive thymocytes, and TCR CD3e double-posi- cantly increased by more than fourfold in both FVB- -/- tive mature thymocytes was markedly decreased in Rag2 (wild-type, 10.4%; homozygous null, 41.9%) -/- both Rag2-deficient mouse strains. Interestingly, B6- and B6-Rag2 mice (wild-type, 8.8%; homozygous -/- Rag2 mice (6.8%; Fig. 3f) exhibited slightly null, 41.3%) (P \ 0.001; Figs. 2c, 3c). ? ? higher populations of CD4 CD8a double-positive -/- thymocytes than FVB-Rag2 mice (1.9%; Fig. 2f). DX-5 IgM CD3ε CD8α CD3ε IgM BM Thymus Spleen 246 Transgenic Res (2018) 27:241–251 -/- -/- Wild-type B6-R Rag2 Wild-type B6-Rag2 2.6 52.2 8.0 6.0 3.4 13.6 5.9 2.9 13.9 2.0 12.9 12.2 B220 B220 b e 13.0 30.1 7.1 3.8 30.7 18.4 10.8 3.0 12.9 13.1 2.3 2.1 TCR TCR c f 0.0 8.8 0.0 41.3 2.8 86.8 23.1 6.8 1.6 9.7 88.1 48.8 CD45 CD4 Fig. 3 Flow cytometric analysis of lymphoid cells in 8-week- T cells in the thymus. Each value represents the mean ± SD. -/- old B6-Rag2 mice. a–c analysis of B, T, and NK cells in the (n = 10 mice per group) spleen. d Analysis of B cells in the bone marrow. e, f Analysis of -/- Immunophenotypes of Il2rg mice homozygous null, 13.3%, Fig. 4b). Not only were T cells found in the spleen, but a large number of ? ? The spleen-to-body weight ratio of both sexes of B6- CD4 CD8a double-positive T lymphocytes were -/- Il2rg mice was greatly increased compared to that located in the thymus as well (wild-type, 91.5%; of wild-type mice (P \ 0.01; Fig. 1). Similar to homozygous null, 72.6%; Fig. 4f). Interestingly, the -/- ? ? Rag2 mice, immunophenotyping analysis of B number of TCR CD3e double-positive T thymo- lymphocytes in the spleen revealed a sharp reduction cytes was increased by more than twofold when ? ? in the number of B220 IgM double-positive B compared to those in the wild-type (wild-type, 15.9%; -/- lymphocytes (P \ 0.001; Fig. 4a). B6-Il2rg mice homozygous null, 36.9%; Fig. 4e). ? ? also displayed a lack of functional mature The cell population of CD45 DX-5 double-pos- ? ? -/- B220 IgM B lymphocytes in the bone marrow itive NK cells in B6-Il2rg mice was decreased by (Fig. 4d). 50% in the spleen (P \ 0.05; wild-type, 8.5%; -/- Unlike the immunophenotyping results of Rag2 homozygous null, 4.2%; Fig. 4c). ? ? mice which confirmed TCR CD3e double-positive T cell deficits, a group of remnant T lymphocytes were -/- observed in B6-Il2rg mice (wild-type, 29.3%; DX-5 CD3ε IgM CD8α CD3ε IgM BM Thymus Spleen Transgenic Res (2018) 27:241–251 247 -/- -/- Wild-type B6-IIl2rg Wild-type B6-Il2rg 1.8 54.5 0.6 2.3 1.0 11.8 0.9 0.2 1.9 1.8 12.3 1.8 B220 B220 b e 12.2 29.3 6.2 13.3 34.5 15.9 26.2 36.9 13.7 15.5 0.4 2.0 TCR TCR c f 0.0 8.5 0.0 4.2 1.9 91.5 1.2 72.6 4.3 13.5 86.3 81.5 CD45 CD4 Fig. 4 Flow cytometric analysis of lymphoid cells in 8-week- T cells in the thymus. Each value represents the mean ± SD. -/- old B6-Il2rg mice. a–c analysis of B, T, and NK cells in the (n = 10 mice per group) spleen. d Analysis of B cells in the bone marrow. e, f Analysis of -/- Immunophenotypes of Prkdc mice Additionally, there was a threefold increase in the ? ? number of CD45 DX-5 double-positive spleen NK -/- -/- The spleen and thymus in BALB/c-Prkdc mice cells in BALB/c-Prkdc mice (36.0%) compared to were smaller than those of wild-type mice. Further- that in wild-type mice (10.7%; P \ 0.001; Fig. 5c). more, the spleen-to-body weight ratio of female, but -/- not male, BALB/c-Prkdc mice was much smaller Serum immunoglobulin levels than that of wild-type mice (P \ 0.001; Fig. 1). Flow cytometric analysis showed marked differences Because immunodeficient mice are defective in B cell between homozygous null and wild-type mice: pop- maturation, the serum immunoglobulin levels of IgG1, ? ? ulations of both B220 IgM mature B lymphocytes in IgG2a, IgG2b, IgG3, IgA, and IgM were assessed by ? ? the spleen and bone marrow and TCR CD3e double- ELISA in our four immunodeficient mouse strains -/- positive T lymphocytes in the thymus and spleen were (Fig. 6). All immunoglobulin levels in FVB-Rag2 , -/- -/- significantly decreased (P \ 0.001; Fig. 5a–b, d–e). B6-Rag2 , and BALB/c-Prkdc mice were signif- ? ? -/- CD4 CD8a double-positive T lymphocytes could icantly decreased, particularly those in FVB-Rag2 barely be detected in the thymus (Fig. 5f). mice, which were hardly detectable (P\ 0.05 to P\ 0.001). Interestingly, the IgM level was not DX-5 CD3ε IgM CD3ε IgM CD8α BM Thymus Spleen 248 Transgenic Res (2018) 27:241–251 -/- -/- Wild-type BALB/c-P Prkdc Wild-Type BALB/c-Prkdc 1.1 44.7 3.5 2.5 1.2 14.1 1.8 0.7 2.2 11.7 17.1 4.4 B220 B220 b e 15.6 31.0 3.8 1.2 36.2 26.0 9.0 4.6 9.0 15.8 1.7 4.6 TCR TCR 0.0 10.7 0.0 35.9 2.5 93.1 5.9 3.8 2.7 8.2 82.6 48.5 CD4 CD45 Fig. 5 Flow cytometric analysis of lymphoid cells in 8-week- f Analysis of T cells in the thymus. Each value represents the -/- old BALB/c-Prkdc mice. a–c analysis of B, T, and NK cells mean ± SD. (n = 8–10 mice per group) in the spleen. d Analysis of B cells in the bone marrow. e, -/- significantly changed in B6-Il2rg mice (wild type, Splenic nodular hypoplasia was observed in both -/- 674 lg/lL; homozygote, 597.8 lg/lL; Fig. 6f), and all Rag2 mice, and a noticeable decrease in the -/- ? immunoglobulin levels in B6-Il2rg mice were the number of CD4 T cells was observed in the spleen, highest among those in the four immunodeficient mouse lymph nodes, and thymus (Fig. S3). However, the strains (IgG1, 3.6 lg/lL; IgG2a, 1.4 lg/lL; IgG2b, number of residual B220 cells was slightly higher in -/- -/- 3.7 lg/lL; IgG3, 586.8 lg/lL; IgA, 2.7 lg/lL; IgM, B6-Rag2 mice than in FVB-Rag2 mice. 597.8 lg/lL). Immunohistochemical findings Discussion Immunohistochemical analysis of organs from the two Numerous immunodeficient mouse models have been Rag2-deficient mouse strains by flow cytometry and generated for biomedical research involving engraft- ELISA demonstrated that they had severe deficits in B ment and transplantation of hematopoietic cells and and T lymphocytes. B220 staining of the spleen and xenografting of tumor cells or tissues (Greiner et al. -/- lymph nodes of Rag2 mice showed a considerable 1998; Majeti et al. 2007; Shultz et al. 2007). We reduction in the number of positive cells (Fig. S3). developed four immunodeficient mouse models using DX-5 CD3ε IgM CD8α IgM CD3ε Transgenic Res (2018) 27:241–251 249 a b c d e f -/- Fig. 6 Levels of serum immunoglobulins in FVB-Rag2 , (e), and IgM (f) were measured by ELISA. Each value -/- -/- -/- B6-Rag2 , B6-Il2rg , and BALB/c-Prkdc mice. The represents the mean ± SD. (n = 8–10 mice per group). serum levels of IgG1 (a), IgG2a (b), IgG2b (c), IgG3 (d), IgA *P \ 0.05, **P \ 0.01, and ***P \ 0.001 ? ? CRISPR/Cas9 and analyzed the immunophenotype of cells: the numbers of B220 IgM mature B cells in ? ? each homozygous null mouse to determine the degree the spleen and bone marrow and TCR CD3e mature of immune deficiency, particularly the levels of B, T, T cells in the spleen and thymus were decreased and NK cells and immunoglobulins. Our FVB- dramatically in homozygous null mice. Serum -/- em1Hwl -/- em2Hwl -/- Rag2 (Rag2 ), B6-Rag2 (Rag2 ), immunoglobulin levels in B6-Rag2 and BALB/c- -/- em1Hwl -/- and BALB/c-Prkdc (Prkdc ) mice exhibited Prkdc mice were undetectable, particularly in -/- phenotypes slightly different from those of conven- FVB-Rag2 mice. Flow cytometry and serum -/- tional models (Table S3). Rag2-deficient mice gener- immunoglobulin levels showed that B6-Rag2 mice ated in 1992 were reported to have a smaller thymus had a slightly increased leakage compared to FVB- -/- -/- but spleen size was not affected (Shinkai et al. 1992). Rag2 or BALB/c-Prkdc mice. This leakiness -/- In contrast, our Rag2 mice showed both smaller varies by background strain and age (Nonoyama et al. spleen and thymus. Furthermore, we found that both 1993). Consistent with these findings, effector CD4 T -/- -/- Rag2 and Prkdc mice lacked mature B and T cell differentiation was enhanced in the mesenteric 123 250 Transgenic Res (2018) 27:241–251 -/- -/- lymph nodes of B10-Rag2 compared to that in our Il2rg mice was also decreased. The difference -/- B6-Rag2 mice (Valatas et al. 2013). between our mutant strains and the conventional -/- em1Hwl Distinctively, our B6-Il2rg (Il2rg ) mice immunodeficient mouse models may be due to the exhibited phenotypes different from those of conven- difference in the production methods. The conven- tional Il2rg-deficient mouse models (Table S3), in tional models generated by ES cell-based gene particular demonstrating much more abundant pres- targeting contain a drug resistance marker such as a ? ? ence of mature T cells. The number of TCR CD3e T PGK-neo cassette in order to enable screening of lymphocytes was decreased in the spleen of homozy- correctly targeted clones. Thus, expression of the gote knockout mice, but there was a 2.3-fold increase neomycin-resistance gene can have an effect on gene ? ? in the number of TCR CD3e mature thymocytes. expression and cell physiology (Scacheri et al. 2001; ? ? Interestingly, more CD4 CD8a thymocytes Valera et al. 1994). On the other hand, our models -/- remained in B6-Il2rg mice compared to those in generated using the CRISPR/Cas9 system lacked any -/- -/- both Rag2 and Prkdc mice. Serum insertion of exogenous DNA sequences. Therefore, -/- immunoglobulin levels in B6-Il2rg mice were these immunodeficient mouse models may be a useful dramatically decreased, except for that of IgM. Similar tool for gaining further insight into the immune to our findings, an X-SCID rat model showed total IgG system. We can consider using our mouse models to and IgA levels that were severely decreased except for improve current humanized mouse models such as the the IgM level (Mashimo et al. 2010). In fact, the serum NOD SCID gamma (NSG) mice that are being vastly -/- IgM level was found to be higher in Il2rg mice than utilized presently at the cutting edge of this field. in control wild-type mice, although the number of ? ? Acknowledgements This research was supported by a grant CD45R sIgM B cells was severely decreased as from the Korea Food and Drug Administration reported (Ohbo et al. 1996). (14182MFDS978), grants from the National Research ? ? We found that the number of CD45 DX-5 NK Foundation of Korea (NRF) funded by the MEST -/- cells in B6-Il2rg mice was twofold lower than that (2015R1A2A1A01003845 and 2010-0020878) and a Korean -- Healthcare Technology R&D Project from the Ministry of in wild-type mice and confirmed that TCRab Health and Welfare (A085136). NK1.1 NK cells were not detectable in the spleen -/- of Il2rg mice (Ohbo et al. 1996). However, the Compliance with ethical standards C.B-17 SCID mouse reported in 1983 exhibited Conflict of interest The authors declare that they have no unaffected NK cell function (Bosma et al. 1983; conflict of interest. Dorshkind et al. 1985). Consistent with previous reports, all of the homozygous Rag2 and Prkdc Open Access This article is distributed under the terms of the knockout mice generated in this study expressed Creative Commons Attribution 4.0 International License (http:// ? ? creativecommons.org/licenses/by/4.0/), which permits unre- three- to four-fold elevated CD45 DX-5 NK cell stricted use, distribution, and reproduction in any medium, populations compared to wild-type mice. These results provided you give appropriate credit to the original may reflect immune compensation for the lack of author(s) and the source, provide a link to the Creative Com- mature B and T lymphocytes and should be further mons license, and indicate if changes were made. examined to determine the accurate functions of the Rag2, Prkdc, and Il2rg genes. References In summary, our newly generated immunodeficient mice exhibited some differences from conventional Belizario JE (2009) Immunodeficient mouse models: an over- immunodeficient mouse strains, particularly the B6- view. Open Immunol J 2:79–85 -/- -/- Il2rg mice. The homozygous FVB-Rag2 , B6- Bosma GC, Custer RP, Bosma MJ (1983) A severe combined -/- -/- Rag2 , and BALB/c-Prkdc mice showed similar immunodeficiency mutation in the mouse. Nature 301:527–530 phenotypes including a lack of mature B cells and Buckley RH, Schiff RI, Schiff SE, Markert ML, Williams LW, mature T cells as well as similar serum immunoglob- Harville TO, Roberts JL, Puck JM (1997) Human severe -/- ulin levels. However, B6-Il2rg mice showed combined immunodeficiency: genetic, phenotypic, and unique phenotypes, lacking mature B cells and functional diversity in one hundred eight infants. J Pediatr 130:378–387 possessing increased mature T cell numbers and high serum IgM levels. The NK cell population in B6- 123 Transgenic Res (2018) 27:241–251 251 Cho SW, Kim S, Kim JM, Kim JS (2013) Targeted genome Small TN et al (1997) Mutation analysis of IL2RG in engineering in human cells with the Cas9 RNA-guided human X-linked severe combined immunodeficiency. endonuclease. Nat Biotechnol 31:230–232 Blood 89:1968–1977 Dorshkind K, Pollack SB, Bosma MJ, Phillips RA (1985) Nat- Scacheri PC, Crabtree JS, Novotny EA, Garrett-Beal L, Chen A, ural killer (NK) cells are present in mice with severe com- Edgemon KA, Marx SJ, Spiegel AM, Chandrasekharappa bined immunodeficiency (scid). J Immunol 134:3798–3801 SC, Collins FS (2001) Bidirectional transcriptional activity Germain RN (2002) T-cell development and the CD4-CD8 of PGK-neomycin and unexpected embryonic lethality in lineage decision. Nat Rev Immunol 2:309–322 heterozygote chimeric knockout mice. Genesis Greiner DL, Hesselton RA, Shultz LD (1998) SCID mouse 30:259–263 models of human stem cell engraftment. Stem Cells Shinkai Y, Rathbun G, Lam KP, Oltz EM, Stewart V, Men- 16:166–177 delsohn M, Charron J, Datta M, Young F, Stall AM et al Honeycutt JB, Sheridan PA, Matsushima GK, Garcia JV (2015) (1992) RAG-2-deficient mice lack mature lymphocytes Humanized mouse models for HIV-1 infection of the CNS. owing to inability to initiate V(D)J rearrangement. Cell J Neurovirol 21:301–309 68:855–867 Ito M, Hiramatsu H, Kobayashi K, Suzue K, Kawahata M, Hioki Shultz LD, Ishikawa F, Greiner DL (2007) Humanized mice in K, Ueyama Y, Koyanagi Y, Sugamura K, Tsuji K et al translational biomedical research. Nat Rev Immunol (2002) NOD/SCID/gamma(c)(null) mouse: an excellent 7:118–130 recipient mouse model for engraftment of human cells. Valatas V, He J, Rivollier A, Kolios G, Kitamura K, Kelsall BL Blood 100:3175–3182 (2013) Host-dependent control of early regulatory and Ito R, Takahashi T, Katano I, Ito M (2012) Current advances in effector T-cell differentiation underlies the genetic sus- humanized mouse models. Cell Mol Immunol 9:208–214 ceptibility of RAG2-deficient mouse strains to transfer Lieber MR (2010) The mechanism of double-strand DNA break colitis. Mucosal Immunol 6:601–611 repair by the nonhomologous DNA end-joining pathway. Valera A, Perales JC, Hatzoglou M, Bosch F (1994) Expression Annu Rev Biochem 79:181–211 of the neomycin-resistance (neo) gene induces alterations Majeti R, Park CY, Weissman IL (2007) Identification of a in gene expression and metabolism. Hum Gene Ther hierarchy of multipotent hematopoietic progenitors in 5:449–456 human cord blood. Cell Stem Cell 1:635–645 Vasquez KM, Marburger K, Intody Z, Wilson JH (2001) Mashimo T, Takizawa A, Voigt B, Yoshimi K, Hiai H, Kur- Manipulating the mammalian genome by homologous amoto T, Serikawa T (2010) Generation of knockout rats recombination. Proc Natl Acad Sci USA 98:8403–8410 with X-linked severe combined immunodeficiency (X- Walsh NC, Kenney LL, Jangalwe S, Aryee KE, Greiner DL, SCID) using zinc-finger nucleases. PLoS ONE 5(1):e8870 Brehm MA, Shultz LD (2017) Humanized mouse models Nonoyama S, Smith FO, Bernstein ID, Ochs HD (1993) Strain- of clinical disease. Annu Rev Pathol 12:187–215 dependent leakiness of mice with severe combined Wright AV, Nunez JK, Doudna JA (2016) Biology and appli- immune deficiency. J Immunol 150:3817–3824 cations of CRISPR systems: harnessing nature’s toolbox Ohbo K, Suda T, Hashiyama M, Mantani A, Ikebe M, Miyakawa for genome engineering. Cell 164:29–44 K, Moriyama M, Nakamura M, Katsuki M, Takahashi K Zhu X, Xu Y, Yu S, Lu L, Ding M, Cheng J, Song G, Gao X, Yao et al (1996) Modulation of hematopoiesis in mice with a L, Fan D et al (2014) An efficient genotyping method for truncated mutant of the interleukin-2 receptor gamma genome-modified animals and human cells generated with chain. Blood 87:956–967 CRISPR/Cas9 system. Sci Rep 4:6420 Puck JM, Pepper AE, Henthorn PS, Candotti F, Isakov J, Whitwam T, Conley ME, Fischer RE, Rosenblatt HM, http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Transgenic Research Springer Journals

Differences between immunodeficient mice generated by classical gene targeting and CRISPR/Cas9-mediated gene knockout

Free
11 pages
Loading next page...
 
/lp/springer_journal/differences-between-immunodeficient-mice-generated-by-classical-gene-sX0zAdHtZK
Publisher
Springer International Publishing
Copyright
Copyright © 2018 by The Author(s)
Subject
Life Sciences; Animal Genetics and Genomics; Plant Genetics and Genomics; Transgenics; Biomedical Engineering/Biotechnology; Genetic Engineering; Molecular Medicine
ISSN
0962-8819
eISSN
1573-9368
D.O.I.
10.1007/s11248-018-0069-y
Publisher site
See Article on Publisher Site

Abstract

Transgenic Res (2018) 27:241–251 https://doi.org/10.1007/s11248-018-0069-y ORIGINAL PAPER Differences between immunodeficient mice generated by classical gene targeting and CRISPR/Cas9-mediated gene knockout . . . . Jae Hoon Lee Jong-Hyung Park Tae-Wook Nam Sun-Min Seo . . . . Jun-Young Kim Han-Kyul Lee Jong Hyun Han Song Yi Park Yang-Kyu Choi Han-Woong Lee Received: 6 November 2017 / Accepted: 19 March 2018 / Published online: 28 March 2018 The Author(s) 2018 Abstract Immunodeficient mice are widely used for mature B cells and T cells and an increase in the ? ? pre-clinical studies to understand various human number of CD45 DX-5 natural killer cells. How- -/- diseases. Here, we report the generation of four ever, B6-Il2rg mice had a unique phenotype, with a immunodeficient mouse models using CRISPR/Cas9 lack of mature B cells, increased number of T cells, system without inserting any foreign gene sequences and decreased number of natural killer cells. Addi- such as Neo cassettes and their characterization. By tionally, serum immunoglobulin levels in all four eliminating any possible effects of adding a Neo immunodeficient mouse models were significantly cassette, our mouse models may allow us to better reduced when compared to those in wild-type mice -/- elucidate the in vivo functions of each gene. Our FVB- with the exception of IgM in B6-Il2rg mice. These -/- -/- -/- Rag2 , B6-Rag2 , and BALB/c-Prkdc mice results indicate that our immunodeficient mouse showed phenotypes similar to those of the earlier models are a robust tool for in vivo studies of the immunodeficient mouse models, including a lack of immune system and will provide new insights into the variation in phenotypic outcomes resulting from different gene-targeting methodologies. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s11248-018-0069-y) con- Keywords Il2rg  Prkdc  Rag2  CRISPR/Cas9 tains supplementary material, which is available to authorized Immunodeficient mouse models users. Jae Hoon Lee and Jong-Hyung Park have contributed equally to this work. Introduction J. H. Lee  T.-W. Nam  J. H. Han  S. Y. Park H.-W. Lee (&) Department of Biochemistry, College of Life Science and Since severe combined immunodeficiency (SCID) Biotechnology, Laboratory Animal Research Center, mice lacking functional B and T lymphocytes were Yonsei University, Seoul 03722, Republic of Korea first spontaneously discovered in a colony of C.B-17 e-mail: hwl@yonsei.ac.kr mice (Bosma et al. 1983), many immunodeficient J.-H. Park  S.-M. Seo  J.-Y. Kim  H.-K. Lee  mice have served as invaluable model organisms in Y.-K. Choi (&) both biological and clinical studies (Ito et al. 2002; Department of Laboratory Animal Medicine, College of Ohbo et al. 1996; Shinkai et al. 1992). Among them, Veterinary Medicine, Konkuk University, Seoul 05029, Rag2-deficient mice fail to generate mature B and T Republic of Korea cells due to the impairment of the V(D)J e-mail: yangkyuc@konkuk.ac.kr 123 242 Transgenic Res (2018) 27:241–251 rearrangements that are crucial to initiating the Importantly, our models lack PGK-neo cassettes that development of functional immunoglobulin (Ig) and are usually included in existing knockout mouse T cell receptors (TCR) (Shinkai et al. 1992). Similarly, models. We evaluated whether CRISPR/Cas9-medi- deficiency of a DNA-dependent protein kinase cat- ated knockout mice exhibit phenotypes different from alytic subunit (DNA-PKcs; encoded by Prkdc gene) those of previous immunodeficient mouse models that causes a SCID phenotype that is characterized by an were generated by classical gene targeting using ES absence of functional B and T cells, lymphopenia, cells. hypogammaglobulinemia, but a normal hematopoietic microenvironment (Buckley et al. 1997; Puck et al. 1997). Therefore, SCID mice are acceptable for both Materials and methods allogeneic and xenogeneic cell transfer experiments. Unlike other immunodeficient mice lacking B and T Animals and ethics statement cell development, however, the absence of IL2rc in mice disrupts not only B and T cells but also natural BALB/cAnNTac (BALB/c), C57BL/6JBomTac (B6), killer (NK) cells (Belizario 2009). This permits the FVB/NTac (FVB), and IcrTac:ICR (ICR) mice were engraftment with human hematopoietic cells to estab- purchased from Taconic Biosciences (Dae Han lish a human immune system in this mouse model. Biolink Co., Ltd., Chungbuk, Republic of Korea). Humanized mice containing human cells or tissues are All mice were housed in the specific pathogen-free becoming increasingly important as animal models for (SPF) facility of the Yonsei Laboratory Animal studying basic and applied human diseases (Honeycutt Research Center. All efforts were made to minimize et al. 2015; Ito et al. 2012; Walsh et al. 2017). To animal suffering, and all animal experiments were improve the utility of immunodeficient mouse models, conducted in accordance with the Korean Food and we generated four different models using the CRISPR/ Drug Administration (KFDA) guidelines. Experimen- Cas9 system and evaluated whether different back- tal protocols were reviewed thoroughly and approved ground strains or targeted genes affect the immunod- by the Institutional Animal Care and Use Committees eficiency phenotypes. (IACUC) at Yonsei University (Permit Number: The CRISPR/Cas9 system can generate improved 201506-322-02). All immunodeficient mouse models genetically modified mouse models, because it does presented in this study will be made readily available not require the inclusion of a drug resistance marker to the research community. such as a PGK-neo cassette which is required to generate ES cell-based knockout mice. The Cas9 Preparation of CRISPR/Cas9 mRNA derived from Streptococcus pyogenes recognizes 5 - NGG-3 as the protospacer adjacent motif (PAM), and The mMESSAGE mMACHINE T7 Ultra kit (Am- the sgRNA consists of a sequence complementary to bion) was used to obtain the Cas9 mRNA, which was the 20 nucleotides upstream of the PAM to identify the diluted in diethyl pyrocarbonate (Sigma)-treated target site (Wright et al. 2016). The double-strand injection buffer (0.25 mM EDTA, 10 mM Tris, pH breaks (DSBs) generated by the Cas9/sgRNA ribonu- 7.4) to obtain the working concentration. Additionally, cleoprotein complex lead to either non-homologous the MEGAshortscript T7 Transcription kit (Ambion) end joining (NHEJ) or homologous recombination was used to synthesize sgRNAs from PCR-generated (HR) repair—mechanisms that are used to generate templates. Plasmids encoding S. pyogenes Cas9 knockout and knock-in mice, respectively (Lieber (SpCas9) protein (Cho et al. 2013) were obtained 2010; Vasquez et al. 2001). Without a template for from ToolGen, Inc. (Seoul, Republic of Korea). HR-mediated DNA repair, the DSB will be repaired through the error-prone NHEJ pathway, forming Microinjection insertion and deletion (indels) mutations in the target gene that may lead to frameshift mutations (Lieber To generate immunodeficient mice using CRISPR/ 2010). Loss of gene function is often induced by Cas9, microinjection of fertilized embryos was per- premature stop codons downstream of the frameshift formed: initially, 6–8-week-old BALB/c, B6, and mutation using the CRISPR/Cas9 system approach. FVB mice were super-ovulated by intra-peritoneal 123 Transgenic Res (2018) 27:241–251 243 injections of 5 IU pregnant mare serum gonadotropin Santa Cruz Biotechnology, Santa Cruz, TX, USA), (Sigma) and 5 IU human chorionic gonadotropin Alexa Fluor 488-anti-CD4 (RM4-5; BD Biosciences), (Sigma) at 48-h intervals. The fertilized embryos were APC-anti-CD8a (53-6.7; BioLegend), PE-anti-CD45 then collected from the super-ovulated mice crossed (30-F11; BD Biosciences), and APC-anti-CD49b with stud males. A mixture of 50 ng/lLof Cas9 (DX-5; BD Biosciences) antibodies. At least 10,000 mRNA and 250 ng/lL of sgRNA was microinjected live cells were analyzed with the FACSCalibur system into the cytoplasm of zygotes, using a piezo-driven (BD Biosciences). manipulator (Prime Tech) to induce mutations, and the resulting embryos were transferred into the oviducts of Enzyme-linked immunosorbent assay (ELISA) ICR pseudo-pregnant foster mothers to produce live mice. Serum samples were collected from the caudal veins of both homozygous mutant and wild-type mice and Founder screening and genotyping PCR then separated in a refrigerated centrifuge at 4 C. Serum immunoglobulin levels of mice were measured To screen founder mice for endonuclease-mediated using the mouse Ready-SET-Go Kits plate (eBio- mutations such as indels, PAGE-PCR assays were science, San Diego, CA, USA) following the manu- performed using genomic DNA samples from tail facturer’s protocol. Briefly, 96-well ELISA plates biopsies (Zhu et al. 2014). In brief, the genomic were coated with the purified anti-mouse IgG1, IgG2a, regions spanning the sgRNA target site were amplified IgG2b, IgG3, IgA, or IgM monoclonal antibody in by PCR. By simply denaturing and annealing, PCR coating buffer at 4 C overnight and then blocked with products containing a mixture of mutant and wild-type diluted assay buffer for 2 h at room temperature, alleles form heteroduplex DNA and homoduplex followed by washing two times. Serial dilutions of pre- DNA as described previously (Zhu et al. 2014). Since titrated standards (IgG1, IgG2a, IgG2b, IgG3, IgA, heteroduplex DNA migrates slower than homoduplex and IgM) and 50 lL of diluted mouse sera were added DNA under a nondenaturing condition, PCR products and the plates incubated for 1 h at room temperature. carrying indels were analyzed by acrylamide gel After washing four times, avidin-horseradish peroxi- electrophoresis. The PCR products from founder mice dase-conjugated antibodies were added and the plates were also cloned using the T-Blunt PCR Cloning Kit incubated again for 1 h at room temperature. After (SolGent Co., Ltd., Republic of Korea) for sequence another four washes, the samples were incubated with analysis, and the mutations were identified by direct the substrate 3,3,5,5—tetramethylbenzidine (TMB) sequencing analysis (Cosmobiotech Co., Ltd., Repub- for 15 min followed by the addition of TMB stop lic of Korea). Those that exhibited the earliest solutions. The absorbance was read in a Tecan premature stop codons were selected as founder mice. spectrometer plate reader (Ma¨nnedorf, Switzerland) at 450 nm with a background subtraction of 570 nm. Antibodies and flow cytometry Immunohistochemical staining Eight-week-old mice were sacrificed and single cells from the spleen, thymus, and bone marrow were Formalin-fixed, paraffin-embedded samples including prepared for flow cytometry analysis. To obtain the spleen, thymus, and lymph nodes, of 8-week-old homozygous null and wild-type mice were used for single-cell suspensions, tissues were cut into small pieces with scissors and passed through a 70-lm cell immunohistochemical staining. Anti-B220 (RA3- strainer (BD Biosciences, Franklin Lakes, NJ, USA) 6B2; Abcam, Cambridge, UK) and anti-CD4 by pressing with a plunger. The remaining red blood (EPR19514; Abcam) antibodies were used for B cell cells were eliminated by using 1 9 RBC lysis buffer and T cell immunostaining, respectively. (eBioscience, San Diego, CA, USA). Prepared cells were stained with FITC-anti-B220 (RA3-6B2; BD Statistical analysis Biosciences), PerCP/Cy5.5-anti-IgM (RMM-1; BioLegend, San Diego, CA, USA), APC-anti-CD3e Statistically significant differences between groups (145-2C11; BD Biosciences), PE-anti-TCR (H57-597; were calculated using the unpaired t-test (GraphPad 123 244 Transgenic Res (2018) 27:241–251 Software, Inc., La Jolla, USA). The results were termination of each target protein (Fig. S1). These considered statistically significant at P \ 0.05, mutations introduced premature termination codons P \ 0.01, or P \ 0.001. (PTCs) into the open reading frames of the target genes, leading to loss of gene function. All immunodeficient mouse strains were then validated in offspring derived Results from each founder mouse by sequence analysis. The homozygous knockout mice (F2) were generated from Generation of immunodeficient mice carrying out- an intercross between F1 heterozygous knockout mice of-frame alleles of Rag2, Prkdc, or Il2rg and confirmed by PCR genotyping (Fig. S1 and Table S2). Since mRNAs with PTCs are targeted by the To generate immunodeficient mice carrying disrup- nonsense-mediated decay (NMD) pathway, RT-PCR tions of the Rag2, Prkdc,or Il2rg genes, we used analysis showed significant reduction of each mRNA in CRISPR/Cas9 system. Each sgRNA was designed to the immune system tissues of the homozygous knock- cleave downstream of the start codon of the genes of out mice compared with the wild-type mice, except for -/- -/- interest to induce frameshift mutations caused by non- the thymus of B6-Rag2 and BALB/c-Prkdc mice homologous end joining (NHEJ)-mediated DNA (Fig. S2). repair. To avoid potential off-target effects of each -/- sgRNA, we designed several sgRNAs using CRIPSR Immunophenotypes of Rag2 mice Design (http://crispr.mit.edu) and sgRNA Design Tool -/- (https://portals.broadinstitute.org/gpp/public/analysis- Gross examination of Rag2 mice (both FVB and tools/sgrna-design), and then intentionally chose the B6) revealed that its spleen and thymus were smaller sgRNAs to recognize the specific sequence that con- than those of wild-type mice. Furthermore, the spleen- tains no apparent homology with the mouse genome to-body weight ratios of both types and both sexes of -/- sequence. We directly injected Cas9 mRNA with the Rag2 mice were significantly lower than that of target-specific sgRNA(s) into one-cell embryos and wild-type mice (P \ 0.05; Fig. 1). Flow cytometric screened the mutants by PAGE-PCR assay using analyses showed a markedly decreased population of ? ? genomic DNAs obtained from the newborns. We suc- mature B cells, especially B220 IgM cells in the cessfully generated several mutant mice with high spleen (P \ 0.001; Figs. 2a, 3a), suggesting that Rag2 efficiency (Table S1) and selected founder mice (F0) deficiency leads to impaired B cell differentiation -/- harboring nucleotide excision leading to premature (Shinkai et al. 1992). Both FVB-Rag2 and B6- a b * ** 2.5 -/- Wild-type B6-Il2rg Wild 2.0 Homo Hemizygote 1.5 1.0 ** *** **** *** 0.5 0.0 ♂♀♂♀♂♀♂♀ -/- -/- -/- FVB-Rag2 B6-Rag2 B6-Il2rg BALB/c- -/- Prkdc -/- Fig. 1 Spleen-to-body weight ratio in immunodeficient mice. Unlike other mutant strains, B6-Il2rg mice showed severe -/- -/- -/- Spleens of FVB-Rag2 , B6-Rag2 , and BALB/c-Prkdc splenomegaly. Each value represents the mean ± SD (n = 4–5 mice were significantly smaller than those of wild-type mice. mice per group). *P \ 0.05, **P \ 0.01, and ***P \ 0.001 Spleen/body weight ratio (%) BM Thymus Spleen Transgenic Res (2018) 27:241–251 245 -/- -/- Wild-type FVB-R Rag2 Wild-type FVB-Rag2 0.2 52.4 0.2 5.1 0.5 8.8 0.2 1.0 11.1 11.5 18.8 7.5 B220 B220 b e 11.7 17.7 0.4 3.3 11.0 16.8 0.3 1.8 0.9 0.7 4.0 1.1 TCR TCR 0.0 10.4 0.0 41.9 2.8 77.9 1.2 1.9 10.1 1.6 87.6 50.7 CD45 CD4 Fig. 2 Flow cytometric analysis of lymphoid cells in 8-week- f Analysis of T cells in the thymus. Each value represents the -/- old FVB-Rag2 mice. a–c analysis of B, T, and NK cells in mean ± SD. (n = 10 mice per group) the spleen. d Analysis of B cells in the bone marrow. e, -/- ? ? Rag2 mice exhibited a significantly decreased Although the ratio of TCR CD3e double-positive ? ? number of B220 IgM B cells in the bone marrow cells in the spleen was higher in wild-type B6 mice where B lymphocytes mature (P \ 0.001; Figs. 2d, (30.1%) than that in wild-type FVB mice (17.7%), -/- -/- 3d). both FVB-Rag2 and B6-Rag2 mice exhibited ? ? ? ? Early T cells differentiate from CD4 CD8 dou- similar decreases in the TCR CD3e double-positive ? ? ble-positive T cells into CD4 or CD8 single- thymocyte population compared to wild-type mice. positive mature T cells, with the latter also expressing The NK cell population was confirmed by exami- ? ? -/- T-cell receptors (TCR) and the CD3 complex (Ger- nation of CD45 DX-5 marker expression in Rag2 ? ? ? ? main 2002). The proportion of CD4 CD8a double- mice. Compared to the wild-type, CD45 DX-5 ? ? positive thymocytes, mature CD4 or CD8a single- double-positive NK cells in the spleen were signifi- ? ? positive thymocytes, and TCR CD3e double-posi- cantly increased by more than fourfold in both FVB- -/- tive mature thymocytes was markedly decreased in Rag2 (wild-type, 10.4%; homozygous null, 41.9%) -/- both Rag2-deficient mouse strains. Interestingly, B6- and B6-Rag2 mice (wild-type, 8.8%; homozygous -/- Rag2 mice (6.8%; Fig. 3f) exhibited slightly null, 41.3%) (P \ 0.001; Figs. 2c, 3c). ? ? higher populations of CD4 CD8a double-positive -/- thymocytes than FVB-Rag2 mice (1.9%; Fig. 2f). DX-5 IgM CD3ε CD8α CD3ε IgM BM Thymus Spleen 246 Transgenic Res (2018) 27:241–251 -/- -/- Wild-type B6-R Rag2 Wild-type B6-Rag2 2.6 52.2 8.0 6.0 3.4 13.6 5.9 2.9 13.9 2.0 12.9 12.2 B220 B220 b e 13.0 30.1 7.1 3.8 30.7 18.4 10.8 3.0 12.9 13.1 2.3 2.1 TCR TCR c f 0.0 8.8 0.0 41.3 2.8 86.8 23.1 6.8 1.6 9.7 88.1 48.8 CD45 CD4 Fig. 3 Flow cytometric analysis of lymphoid cells in 8-week- T cells in the thymus. Each value represents the mean ± SD. -/- old B6-Rag2 mice. a–c analysis of B, T, and NK cells in the (n = 10 mice per group) spleen. d Analysis of B cells in the bone marrow. e, f Analysis of -/- Immunophenotypes of Il2rg mice homozygous null, 13.3%, Fig. 4b). Not only were T cells found in the spleen, but a large number of ? ? The spleen-to-body weight ratio of both sexes of B6- CD4 CD8a double-positive T lymphocytes were -/- Il2rg mice was greatly increased compared to that located in the thymus as well (wild-type, 91.5%; of wild-type mice (P \ 0.01; Fig. 1). Similar to homozygous null, 72.6%; Fig. 4f). Interestingly, the -/- ? ? Rag2 mice, immunophenotyping analysis of B number of TCR CD3e double-positive T thymo- lymphocytes in the spleen revealed a sharp reduction cytes was increased by more than twofold when ? ? in the number of B220 IgM double-positive B compared to those in the wild-type (wild-type, 15.9%; -/- lymphocytes (P \ 0.001; Fig. 4a). B6-Il2rg mice homozygous null, 36.9%; Fig. 4e). ? ? also displayed a lack of functional mature The cell population of CD45 DX-5 double-pos- ? ? -/- B220 IgM B lymphocytes in the bone marrow itive NK cells in B6-Il2rg mice was decreased by (Fig. 4d). 50% in the spleen (P \ 0.05; wild-type, 8.5%; -/- Unlike the immunophenotyping results of Rag2 homozygous null, 4.2%; Fig. 4c). ? ? mice which confirmed TCR CD3e double-positive T cell deficits, a group of remnant T lymphocytes were -/- observed in B6-Il2rg mice (wild-type, 29.3%; DX-5 CD3ε IgM CD8α CD3ε IgM BM Thymus Spleen Transgenic Res (2018) 27:241–251 247 -/- -/- Wild-type B6-IIl2rg Wild-type B6-Il2rg 1.8 54.5 0.6 2.3 1.0 11.8 0.9 0.2 1.9 1.8 12.3 1.8 B220 B220 b e 12.2 29.3 6.2 13.3 34.5 15.9 26.2 36.9 13.7 15.5 0.4 2.0 TCR TCR c f 0.0 8.5 0.0 4.2 1.9 91.5 1.2 72.6 4.3 13.5 86.3 81.5 CD45 CD4 Fig. 4 Flow cytometric analysis of lymphoid cells in 8-week- T cells in the thymus. Each value represents the mean ± SD. -/- old B6-Il2rg mice. a–c analysis of B, T, and NK cells in the (n = 10 mice per group) spleen. d Analysis of B cells in the bone marrow. e, f Analysis of -/- Immunophenotypes of Prkdc mice Additionally, there was a threefold increase in the ? ? number of CD45 DX-5 double-positive spleen NK -/- -/- The spleen and thymus in BALB/c-Prkdc mice cells in BALB/c-Prkdc mice (36.0%) compared to were smaller than those of wild-type mice. Further- that in wild-type mice (10.7%; P \ 0.001; Fig. 5c). more, the spleen-to-body weight ratio of female, but -/- not male, BALB/c-Prkdc mice was much smaller Serum immunoglobulin levels than that of wild-type mice (P \ 0.001; Fig. 1). Flow cytometric analysis showed marked differences Because immunodeficient mice are defective in B cell between homozygous null and wild-type mice: pop- maturation, the serum immunoglobulin levels of IgG1, ? ? ulations of both B220 IgM mature B lymphocytes in IgG2a, IgG2b, IgG3, IgA, and IgM were assessed by ? ? the spleen and bone marrow and TCR CD3e double- ELISA in our four immunodeficient mouse strains -/- positive T lymphocytes in the thymus and spleen were (Fig. 6). All immunoglobulin levels in FVB-Rag2 , -/- -/- significantly decreased (P \ 0.001; Fig. 5a–b, d–e). B6-Rag2 , and BALB/c-Prkdc mice were signif- ? ? -/- CD4 CD8a double-positive T lymphocytes could icantly decreased, particularly those in FVB-Rag2 barely be detected in the thymus (Fig. 5f). mice, which were hardly detectable (P\ 0.05 to P\ 0.001). Interestingly, the IgM level was not DX-5 CD3ε IgM CD3ε IgM CD8α BM Thymus Spleen 248 Transgenic Res (2018) 27:241–251 -/- -/- Wild-type BALB/c-P Prkdc Wild-Type BALB/c-Prkdc 1.1 44.7 3.5 2.5 1.2 14.1 1.8 0.7 2.2 11.7 17.1 4.4 B220 B220 b e 15.6 31.0 3.8 1.2 36.2 26.0 9.0 4.6 9.0 15.8 1.7 4.6 TCR TCR 0.0 10.7 0.0 35.9 2.5 93.1 5.9 3.8 2.7 8.2 82.6 48.5 CD4 CD45 Fig. 5 Flow cytometric analysis of lymphoid cells in 8-week- f Analysis of T cells in the thymus. Each value represents the -/- old BALB/c-Prkdc mice. a–c analysis of B, T, and NK cells mean ± SD. (n = 8–10 mice per group) in the spleen. d Analysis of B cells in the bone marrow. e, -/- significantly changed in B6-Il2rg mice (wild type, Splenic nodular hypoplasia was observed in both -/- 674 lg/lL; homozygote, 597.8 lg/lL; Fig. 6f), and all Rag2 mice, and a noticeable decrease in the -/- ? immunoglobulin levels in B6-Il2rg mice were the number of CD4 T cells was observed in the spleen, highest among those in the four immunodeficient mouse lymph nodes, and thymus (Fig. S3). However, the strains (IgG1, 3.6 lg/lL; IgG2a, 1.4 lg/lL; IgG2b, number of residual B220 cells was slightly higher in -/- -/- 3.7 lg/lL; IgG3, 586.8 lg/lL; IgA, 2.7 lg/lL; IgM, B6-Rag2 mice than in FVB-Rag2 mice. 597.8 lg/lL). Immunohistochemical findings Discussion Immunohistochemical analysis of organs from the two Numerous immunodeficient mouse models have been Rag2-deficient mouse strains by flow cytometry and generated for biomedical research involving engraft- ELISA demonstrated that they had severe deficits in B ment and transplantation of hematopoietic cells and and T lymphocytes. B220 staining of the spleen and xenografting of tumor cells or tissues (Greiner et al. -/- lymph nodes of Rag2 mice showed a considerable 1998; Majeti et al. 2007; Shultz et al. 2007). We reduction in the number of positive cells (Fig. S3). developed four immunodeficient mouse models using DX-5 CD3ε IgM CD8α IgM CD3ε Transgenic Res (2018) 27:241–251 249 a b c d e f -/- Fig. 6 Levels of serum immunoglobulins in FVB-Rag2 , (e), and IgM (f) were measured by ELISA. Each value -/- -/- -/- B6-Rag2 , B6-Il2rg , and BALB/c-Prkdc mice. The represents the mean ± SD. (n = 8–10 mice per group). serum levels of IgG1 (a), IgG2a (b), IgG2b (c), IgG3 (d), IgA *P \ 0.05, **P \ 0.01, and ***P \ 0.001 ? ? CRISPR/Cas9 and analyzed the immunophenotype of cells: the numbers of B220 IgM mature B cells in ? ? each homozygous null mouse to determine the degree the spleen and bone marrow and TCR CD3e mature of immune deficiency, particularly the levels of B, T, T cells in the spleen and thymus were decreased and NK cells and immunoglobulins. Our FVB- dramatically in homozygous null mice. Serum -/- em1Hwl -/- em2Hwl -/- Rag2 (Rag2 ), B6-Rag2 (Rag2 ), immunoglobulin levels in B6-Rag2 and BALB/c- -/- em1Hwl -/- and BALB/c-Prkdc (Prkdc ) mice exhibited Prkdc mice were undetectable, particularly in -/- phenotypes slightly different from those of conven- FVB-Rag2 mice. Flow cytometry and serum -/- tional models (Table S3). Rag2-deficient mice gener- immunoglobulin levels showed that B6-Rag2 mice ated in 1992 were reported to have a smaller thymus had a slightly increased leakage compared to FVB- -/- -/- but spleen size was not affected (Shinkai et al. 1992). Rag2 or BALB/c-Prkdc mice. This leakiness -/- In contrast, our Rag2 mice showed both smaller varies by background strain and age (Nonoyama et al. spleen and thymus. Furthermore, we found that both 1993). Consistent with these findings, effector CD4 T -/- -/- Rag2 and Prkdc mice lacked mature B and T cell differentiation was enhanced in the mesenteric 123 250 Transgenic Res (2018) 27:241–251 -/- -/- lymph nodes of B10-Rag2 compared to that in our Il2rg mice was also decreased. The difference -/- B6-Rag2 mice (Valatas et al. 2013). between our mutant strains and the conventional -/- em1Hwl Distinctively, our B6-Il2rg (Il2rg ) mice immunodeficient mouse models may be due to the exhibited phenotypes different from those of conven- difference in the production methods. The conven- tional Il2rg-deficient mouse models (Table S3), in tional models generated by ES cell-based gene particular demonstrating much more abundant pres- targeting contain a drug resistance marker such as a ? ? ence of mature T cells. The number of TCR CD3e T PGK-neo cassette in order to enable screening of lymphocytes was decreased in the spleen of homozy- correctly targeted clones. Thus, expression of the gote knockout mice, but there was a 2.3-fold increase neomycin-resistance gene can have an effect on gene ? ? in the number of TCR CD3e mature thymocytes. expression and cell physiology (Scacheri et al. 2001; ? ? Interestingly, more CD4 CD8a thymocytes Valera et al. 1994). On the other hand, our models -/- remained in B6-Il2rg mice compared to those in generated using the CRISPR/Cas9 system lacked any -/- -/- both Rag2 and Prkdc mice. Serum insertion of exogenous DNA sequences. Therefore, -/- immunoglobulin levels in B6-Il2rg mice were these immunodeficient mouse models may be a useful dramatically decreased, except for that of IgM. Similar tool for gaining further insight into the immune to our findings, an X-SCID rat model showed total IgG system. We can consider using our mouse models to and IgA levels that were severely decreased except for improve current humanized mouse models such as the the IgM level (Mashimo et al. 2010). In fact, the serum NOD SCID gamma (NSG) mice that are being vastly -/- IgM level was found to be higher in Il2rg mice than utilized presently at the cutting edge of this field. in control wild-type mice, although the number of ? ? Acknowledgements This research was supported by a grant CD45R sIgM B cells was severely decreased as from the Korea Food and Drug Administration reported (Ohbo et al. 1996). (14182MFDS978), grants from the National Research ? ? We found that the number of CD45 DX-5 NK Foundation of Korea (NRF) funded by the MEST -/- cells in B6-Il2rg mice was twofold lower than that (2015R1A2A1A01003845 and 2010-0020878) and a Korean -- Healthcare Technology R&D Project from the Ministry of in wild-type mice and confirmed that TCRab Health and Welfare (A085136). NK1.1 NK cells were not detectable in the spleen -/- of Il2rg mice (Ohbo et al. 1996). However, the Compliance with ethical standards C.B-17 SCID mouse reported in 1983 exhibited Conflict of interest The authors declare that they have no unaffected NK cell function (Bosma et al. 1983; conflict of interest. Dorshkind et al. 1985). Consistent with previous reports, all of the homozygous Rag2 and Prkdc Open Access This article is distributed under the terms of the knockout mice generated in this study expressed Creative Commons Attribution 4.0 International License (http:// ? ? creativecommons.org/licenses/by/4.0/), which permits unre- three- to four-fold elevated CD45 DX-5 NK cell stricted use, distribution, and reproduction in any medium, populations compared to wild-type mice. These results provided you give appropriate credit to the original may reflect immune compensation for the lack of author(s) and the source, provide a link to the Creative Com- mature B and T lymphocytes and should be further mons license, and indicate if changes were made. examined to determine the accurate functions of the Rag2, Prkdc, and Il2rg genes. References In summary, our newly generated immunodeficient mice exhibited some differences from conventional Belizario JE (2009) Immunodeficient mouse models: an over- immunodeficient mouse strains, particularly the B6- view. Open Immunol J 2:79–85 -/- -/- Il2rg mice. The homozygous FVB-Rag2 , B6- Bosma GC, Custer RP, Bosma MJ (1983) A severe combined -/- -/- Rag2 , and BALB/c-Prkdc mice showed similar immunodeficiency mutation in the mouse. Nature 301:527–530 phenotypes including a lack of mature B cells and Buckley RH, Schiff RI, Schiff SE, Markert ML, Williams LW, mature T cells as well as similar serum immunoglob- Harville TO, Roberts JL, Puck JM (1997) Human severe -/- ulin levels. However, B6-Il2rg mice showed combined immunodeficiency: genetic, phenotypic, and unique phenotypes, lacking mature B cells and functional diversity in one hundred eight infants. J Pediatr 130:378–387 possessing increased mature T cell numbers and high serum IgM levels. The NK cell population in B6- 123 Transgenic Res (2018) 27:241–251 251 Cho SW, Kim S, Kim JM, Kim JS (2013) Targeted genome Small TN et al (1997) Mutation analysis of IL2RG in engineering in human cells with the Cas9 RNA-guided human X-linked severe combined immunodeficiency. endonuclease. Nat Biotechnol 31:230–232 Blood 89:1968–1977 Dorshkind K, Pollack SB, Bosma MJ, Phillips RA (1985) Nat- Scacheri PC, Crabtree JS, Novotny EA, Garrett-Beal L, Chen A, ural killer (NK) cells are present in mice with severe com- Edgemon KA, Marx SJ, Spiegel AM, Chandrasekharappa bined immunodeficiency (scid). J Immunol 134:3798–3801 SC, Collins FS (2001) Bidirectional transcriptional activity Germain RN (2002) T-cell development and the CD4-CD8 of PGK-neomycin and unexpected embryonic lethality in lineage decision. Nat Rev Immunol 2:309–322 heterozygote chimeric knockout mice. Genesis Greiner DL, Hesselton RA, Shultz LD (1998) SCID mouse 30:259–263 models of human stem cell engraftment. Stem Cells Shinkai Y, Rathbun G, Lam KP, Oltz EM, Stewart V, Men- 16:166–177 delsohn M, Charron J, Datta M, Young F, Stall AM et al Honeycutt JB, Sheridan PA, Matsushima GK, Garcia JV (2015) (1992) RAG-2-deficient mice lack mature lymphocytes Humanized mouse models for HIV-1 infection of the CNS. owing to inability to initiate V(D)J rearrangement. Cell J Neurovirol 21:301–309 68:855–867 Ito M, Hiramatsu H, Kobayashi K, Suzue K, Kawahata M, Hioki Shultz LD, Ishikawa F, Greiner DL (2007) Humanized mice in K, Ueyama Y, Koyanagi Y, Sugamura K, Tsuji K et al translational biomedical research. Nat Rev Immunol (2002) NOD/SCID/gamma(c)(null) mouse: an excellent 7:118–130 recipient mouse model for engraftment of human cells. Valatas V, He J, Rivollier A, Kolios G, Kitamura K, Kelsall BL Blood 100:3175–3182 (2013) Host-dependent control of early regulatory and Ito R, Takahashi T, Katano I, Ito M (2012) Current advances in effector T-cell differentiation underlies the genetic sus- humanized mouse models. Cell Mol Immunol 9:208–214 ceptibility of RAG2-deficient mouse strains to transfer Lieber MR (2010) The mechanism of double-strand DNA break colitis. Mucosal Immunol 6:601–611 repair by the nonhomologous DNA end-joining pathway. Valera A, Perales JC, Hatzoglou M, Bosch F (1994) Expression Annu Rev Biochem 79:181–211 of the neomycin-resistance (neo) gene induces alterations Majeti R, Park CY, Weissman IL (2007) Identification of a in gene expression and metabolism. Hum Gene Ther hierarchy of multipotent hematopoietic progenitors in 5:449–456 human cord blood. Cell Stem Cell 1:635–645 Vasquez KM, Marburger K, Intody Z, Wilson JH (2001) Mashimo T, Takizawa A, Voigt B, Yoshimi K, Hiai H, Kur- Manipulating the mammalian genome by homologous amoto T, Serikawa T (2010) Generation of knockout rats recombination. Proc Natl Acad Sci USA 98:8403–8410 with X-linked severe combined immunodeficiency (X- Walsh NC, Kenney LL, Jangalwe S, Aryee KE, Greiner DL, SCID) using zinc-finger nucleases. PLoS ONE 5(1):e8870 Brehm MA, Shultz LD (2017) Humanized mouse models Nonoyama S, Smith FO, Bernstein ID, Ochs HD (1993) Strain- of clinical disease. Annu Rev Pathol 12:187–215 dependent leakiness of mice with severe combined Wright AV, Nunez JK, Doudna JA (2016) Biology and appli- immune deficiency. J Immunol 150:3817–3824 cations of CRISPR systems: harnessing nature’s toolbox Ohbo K, Suda T, Hashiyama M, Mantani A, Ikebe M, Miyakawa for genome engineering. Cell 164:29–44 K, Moriyama M, Nakamura M, Katsuki M, Takahashi K Zhu X, Xu Y, Yu S, Lu L, Ding M, Cheng J, Song G, Gao X, Yao et al (1996) Modulation of hematopoiesis in mice with a L, Fan D et al (2014) An efficient genotyping method for truncated mutant of the interleukin-2 receptor gamma genome-modified animals and human cells generated with chain. Blood 87:956–967 CRISPR/Cas9 system. Sci Rep 4:6420 Puck JM, Pepper AE, Henthorn PS, Candotti F, Isakov J, Whitwam T, Conley ME, Fischer RE, Rosenblatt HM,

Journal

Transgenic ResearchSpringer Journals

Published: Mar 28, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off