technical reports
Validated germline-competent embryonic stem cell lines from nonobese diabetic mice
© 2009 Nature America, Inc. All rights reserved.
Jennifer Nichols1,2, Kenneth Jones1,2, Jenny M Phillips3, Stephen A Newland3, Mila Roode1,2, William Mansfield1,4, Austin Smith1,4 & Anne Cooke3 Nonobese diabetic (NOD) mice provide an excellent model of type 1 diabetes. The genetic contribution to this disease is complex, with more than 20 loci implicated in diabetes onset. One of the challenges for researchers using the NOD mouse model (and, indeed, other models of spontaneous autoimmune disease) has been the high density of sequence variation within candidate chromosomal segments. Furthermore, the scope for analyzing many putative disease loci via gene targeting has been hampered by the lack of NOD embryonic stem (ES) cells. We describe here the derivation of NOD ES cell lines capable of generating chimeric mice after stable genetic modification. These NOD ES cell lines also show efficient germline transmission, with offspring developing diabetes. The availability of these cells will not only enable the dissection of closely linked loci and the role they have in the onset of type 1 diabetes but also facilitate the generation of new transgenics. Type 1 diabetes is an autoimmune disease in which pancreatic beta cells are selectively destroyed. The development of this disease is known to be governed by both genetic and environmental factors1. A robust spontaneous animal model of the human disease, the NOD mouse2, has enabled analyses of both the genetic and potential environmental influences on diabetes onset3. Analyses of congenic mouse lines has led to the identification of more than 20 genetic regions potentially involved in the disease process4. Among the candidate genes identified, those encoding major histocompatibility complex (MHC) class II and insulin have been shown to play a part in governing diabetes susceptibility, but the formal validation of other genes has been hampered by gene linkages and extensive polymorphisms. NOD ES cells would provide the ideal tool to dissect and analyze the genetic contribution to type 1 diabetes, as they would permit targeted replacement of potential candidate alleles to be carried out individually and selectively. Such analyses could lead to identification of pathways with a role in the human disease and disclose targets for therapeutic manipulation. Despite many attempts to obtain robust ES cell lines capable of germline transmission from the NOD mouse strain, it has proved difficult5–7. Alteration of cell culture conditions seemed to provide increased
germline transmission of genetically marked NOD ES cells, but only when the cells were cultured in certain batches of fetal calf serum (FCS) on a layer of feeder cells, with subsequent injection of NOD blastocysts8. No germline transmission was observed when these cells were injected into C57BL/6 blastocysts. Therefore, genetic studies in such cells would be extremely laborious. As an alternative, ES cells have been derived from F1 NOD embryos, but creation of useful transgenic mice with targeted genetic modifications from these embryos would require extensive backcrossing5. Our recent description of ground-state culture conditions for ES cell derivation and maintenance9 provided the opportunity for a fresh attempt at NOD ES cell derivation. We report here the successful generation of multiple lines of NOD ES cells, as well as their ability to form chimeras and be genetically manipulated. We also show that these cells are able to achieve germline transmission and give rise to offspring in which diabetes develops. This means that, to our knowledge, for the first time stable, germline-competent NOD ES cells are available for immunological, physiological and genetic studies of type 1 diabetes. RESULTS NOD ES cells can be derived and are stable for multiple passages We have previously described a serum- and feeder-free system for generating ES cell lines10. The essence of this system is the use of smallmolecule inhibitors to prevent differentiation and promote survival and expansion. We postulate that this constitutes a generic culture condition for maintaining authentic pluripotency11. Consistent with this idea, we have demonstrated ES cell derivation from blastocysts of not only the permissive 129 strain but also from the CBA and MF1 strains, which had previously proved refractory to ES cell derivation using standard protocols9. Furthermore, this approach using small-molecule inhibitors has enabled the first derivations of rat ES cells9,12. We therefore cultured NOD embryos from the eight-cell stage to the expanded blastocyst stage in medium supplemented with the mitogen-activated protein kinase kinase inhibitor PD0325901, the glycogen synthase kinase inhibitor Chir99021 and the cytokine leukemia inhibitory factor (LIF). We then plated inner cell masses isolated by immunosurgery13 in the same conditions. We subsequently disaggregated and expanded the macroscopic colonies formed to derive NOD ES cells. From 30 NOD embryos, we
1Wellcome Trust Centre for Stem Cell Research, 2Department of Physiology, Development and Neuroscience, 3Department of Pathology and 4Department of Biochemistry, University of Cambridge, Cambridge, UK. Correspondence should be addressed to A.C. (
[email protected]).
Received 10 March; accepted 27 May; published online 02 June 2009; doi:10.1038/nm.1996
814
volume 15 | number 7 | july 2009 nature medicine
technical reports
© 2009 Nature America, Inc. All rights reserved.
a
b
c
d
e
f
Figure 1 Characterization of NOD ES cells. (a) Phase contrast micrograph example of NOD ES cells clone 2 passage 8 growing in ground state culture conditions. (b) Karyotype of clone 1 passage 14, showing 40 chromosomes. (c,d) DAPI (c) and Oct4 (d) immunostaining of NOD ES cells clone 1 passage 14. (e,f) DAPI (e) and Nanog (f) immunostaining of NOD ES cells clone 1 passage 14. Scale bars, 20 µm.
virtually pure albino, indicating a massive contribution of the ES cells. As NOD mice express the MHC class I and class II molecules, Kd and 1-Ag7, respectively, whereas C57BL/6 mice express Kb and 1-Ab respectively, it is possible to determine the extent of chimerism on the basis of specific antibody staining. Two of the mice showed extensive hematopoietic contribution from the NOD-derived cells, on the basis of Kd and 1-Ag7 expression on B cells in peripheral blood (Fig. 2b–d). Further confirmation of NOD genotype can be provided by examining microsatellite markers that distinguish between NOD and C57BL/6 mouse strains. Therefore, we subsequently analyzed the same four mice in detail for tissue expression of one of these polymorphic markers (D17Mit28). We found that mouse 1, which, in terms of its hematopoietic system, seemed to be wholly NOD derived, also showed only the NOD D17Mit28 variant in ear, brain, gut, muscle, spleen and pancreas tissue (Fig. 2e). In contrast, mouse 4, which showed variegated coat color chimerism but a C57BL/6 hematopoietic system, expressed the C57BL/6 polymorphic variant in all other tissues examined (Fig. 2e). Immunohistochemical analysis of the pancreas of chimeric mice revealed that several had pancreatic infiltrates (data not shown). Such infiltrates are characteristically
a derived 16 ES cell lines. These all showed typical ES cell morphology (Fig. 1a). We expanded 11 of the 16 NOD ES cell lines to passage 14 and prepared metaphase spreads. Seven, including both male and female cell lines (sexed by Y chromosome–specific PCR14), possessed the diploid complement of 40 chromosomes in 80–90% of spreads (Fig. 1b). These cells were positive for the pluripotency markers Oct4 and Nanog by immunostaining (Fig. 1c–f). These cells are capable of differentiation in vitro into neuronal cells and beating muscle (data not shown). NOD ES cells contribute extensively to chimeric mice We injected four of the NOD ES cell lines (three male and one female) into C57BL/6 blastocysts, transferred the injected blastocysts into pseudopregnant mice and scored the offspring for evidence of coat color chimerism. The injected cells from all four lines yielded chimeric mice at a high frequency, with very extensive contribution from the injected ES cells (Table 1). We assessed this chimerism by several parameters including coat color, cell surface marker expression and microsatellite markers (Fig. 2). Examples of four of the chimeric mice with a range of coat color chimerism are shown (Fig. 2a). Several of the mice were
103
103
99.83
102
H-2K
d
101 100
101
102
103
100
13.58
102
101 100
103
103
87.45
102
101
102
100
103
0.28
102 101
101 100
Mouse 4
100
101
102
103
100
100
101
102
103
102
103
CD19
c
103
103
H-2Kb
103
102
1.13
102
102
14.37
101
101 100
100
101
102
103
103
87.66
100 100
101
102
100
103
99.93
102
101
101 100
101
102
103
100
100
101
CD19
d
0.00
I-Ab
11.28
0.11
0.31
89.20
0.00
98.83
103
103
103
103
102
102
102
102
101
101
101
98.15
1.75
100
100
101
102
103
I-Ag7 Mouse 1
e
3.13
100 100
101
85.28 102
103
100
Mouse 2
0.06
101
0.44 100
10.36
101
102
103
1.02
100 100
101
0.09 102
103
Mouse 4
Mouse 3
{ {
nature medicine volume 15 | number 7 | july 2009
b
Mouse 3
{ {
Figure 2 NOD ES cells generate chimeric mice. (a) Four different chimeric mice, showing a range of coat color. (b–d) Hematopoietic chimerism shown by flow cytometry analysis of blood samples from the four individual mice. Samples were stained for and gated on CD19. The percentages of the CD19 cells expressing the various markers is shown. (b) Expression of H-2Kd (NOD) on CD19+ cells. (c) Expression of H-2Kb (C57BL/6) on CD19 cells. (d) I-Ab (C57BL/6) and I-Ag7 (NOD) MHC class II expression on CD19+ cells. (e) Allele-specific PCR of microsatellite markers specific for NOD (bottom band) and C57BL/6 (top band) were used to identify the contributions of NOD or C57BL/6 genomic DNA from the following tissues: A, ear; B, brain; C, gut; D, muscle; E, spleen; and F, pancreas. bp, base pairs. (f) Flow cytometry analysis of cells infiltrating the pancreas of the four chimeric mice. Cells were stained for CD4 and CD8. The percentages of the cells from the pancreas expressing the various markers is shown.
Mouse 2
Mouse 1
A B C D E F A B C DE F
A B C D E F A B C DE F
100 bp
f
6.19
0.07
0.06
1.04
0.62
0.06
103
103
103
102
102
102
102
CD4 10
101
101
1
18.04
75.70
100 100
101
102
103
95.75
100 100
101
3.15 102
103
0.05
0.19
103
101
0.87
98.46
100 100
101
102
103
99.08
100 100
101
0.69 102
103
CD8
815
technical reports antigen expression on peripheral blood. Of these chimeras, 28 (97%) expressed NOD MHC to some extent. Microsatellite analyses of tissues from eight chimeric mice showed that only one mouse (mouse 4 in Fig. 2e) did not express NOD polymorphic variants in any of the tissues examined.
Cell line
No. of embryos injected
No. born
Male
Female
Male 1
39
14
14
0
Male 2
40
29
28
0
4/5
Male 3
20
5
3
0
3/3
No. of chimeras
Germline
4/4
The NOD ES cell lines mediate germline transmission Nearly all chimeras produced with XY NOD ES cells were male. This is indicative of sex conversion, which occurs when Male 1 (DsRED) 37 23 17 1 6/8 XY ES cell derivatives predominate over host embryo XX cells Female 1 60 15 1 11 Not tested in the developing genital ridge15. Sex conversion is a feature of high-quality ES cells that favors production of ES cell–derived seen in the NOD pancreas before diabetes development but are never progeny because host XX germ cells cannot complete spermatogenesis. seen in the pancreas of non–diabetes-prone mouse strains. Flow cytom- To establish whether the NOD ES cells do indeed colonize the germ line etry analysis of the pancreatic infiltrating cells in the four chimeric mice and produce gametes, we crossed the male chimeric mice generated from revealed the presence of CD4+ T cells, CD8+ T cells (Fig. 2f) and B cells each of the male NOD ES cell lines with NOD female mice. Generation (data not shown) in mice 1 and 2. By these analyses, we were able to of albino progeny would indicate homozygosity for the NOD genome, establish that chimeras were efficiently generated from all three male possible only if the sperm in the male chimeric parent were derived from the injected NOD ES cells. Sperm of the C57BL/6 host blastocyst, NOD ES cell lines that we tested. We have provided detailed data for four chimeras, but we analyzed in contrast, would produce agouti offspring. On the basis of coat color many more. Of 71 mice, 63 (89%) showed coat color chimerism (Table of the offspring, all NOD ES cell lines were capable of highly efficient 1). We examined 29 of the male chimeras for expression of NOD MHC germline transmission (Table 1 and Fig. 3a). There is not always correlation between the various tissues and the a b H-2Kd 0.00 99.55 0.00 99.71 0.00 99.76 0.00 99.69 degree of chimerism. For example, of four 10 10 10 10 mice (Fig. 2), mouse 1 had 22/22 albino off10 10 10 10 spring, mouse 2 had 21/21 albino offspring and 10 10 10 10 0.24 0.00 0.45 0.29 0.00 0.00 0.31 0.00 mouse 3 had 20/20 albino offspring, although 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 CD19 he showed only 13% NOD-derived B cells. 0.00 0.43 0.36 0.31 0.50 0.00 0.00 0.00 10 10 10 Mouse 4, who had less than 1% NOD-derived H-2Kb 10 10 10 10 10 B cells, produced mixed litters with thirteen 10 10 10 10 albino and three agouti pups. In summary, of 0.00 0.00 0.00 99.64 99.69 0.00 99.50 99.57 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 the initial 12 male chimeras that we mated to CD19 female NOD mice, nine produced only albino 3
3
3
3
2
2
2
2
1
1
1
0
Mouse A
1
2
3
103
102
102
101
102
103
0.00
101
102
102
10
10
2
3
CD19
2
1
1
0
1
2
3
100.00 Mouse B
0.00
100 0 10
101
0.00 102
103
g NOD CAM (n = 34)
80 60
NOD ES– cell derived (n = 36)
40 20 0
4
8
12
16
Age (weeks)
816
0
0
0
1
2
20
24
28
3
0
1
2
3
0
1
2
3
1
0
1
2
0
3
Offspring
Offspring
NOD
BL/6
Offspring
BL/6
Ch17
Ch6 NOD
NOD
Ch3 BL/6
BL/6
Ch1
103
100
0
0
3
1
101
99.04
0.00 10
2
3
0.00 102
0.00 103
1
1
2
100.00
0.00 100
0.96
101
f
100
103
100 0 10
1
0
3
101
0.50
0.00
H-2Kb
0.00
99.50
101
0
3
2
NOD
0.00
CD19
2
Mouse B
103
100 0 10
1
3
Mouse A
H-2Kd
0
2
e
d
0
Offspring
c
0
3
0
Diabetes (%)
© 2009 Nature America, Inc. All rights reserved.
Table 1 NOD ES cells generate chimeras and effect germline transmission
Figure 3 NOD ES cells support efficient germline transmission. (a) Example of NOD offspring generated when a male chimera was mated with a NOD female. (b) Expression of H2-Kd (NOD) (top row) and H2-Kb (C57BL/6) (bottom row) on CD19+ cells of the four offspring. (c) Examples of two offspring from a mixed litter generated when a male chimera was mated with a NOD female. Mouse A, albino (NOD); mouse B, agouti (NOD × C57BL/6). (d) Flow cytometry analysis of expression of H2-Kd (NOD) (top row) and H2-Kb (C57BL/6) (bottom row) on CD19 cells in blood samples from the two mice in c. We stained the samples for and gated on CD19. Mouse A, left plots; mouse B, right plots. (e) Allele-specific PCR of microsatellite markers used to identify NOD and C57BL/6 at four different polymorphic sites. (f) Incidence of diabetes in female NOD mice from the colony in the Pathology Department, Cambridge (NOD CAM) and female NOD ES cell progeny of the chimeras deriving from all three ES cell lines tested. (g) Immunohistochemical analysis of pancreas sections from 12-week-old female mice stained for CD3 (green) and insulin (red). Top left, section of pancreas from a female NOD CAM mouse. Top right, bottom left and bottom right, sections of pancreas from progeny of chimeras generated using ES cell lines 1, 2 and 3, respectively. Scale bars, 50 µm.
volume 15 | number 7 | july 2009 nature medicine
technical reports a
0.19
60.51 103
102
102
101
Gated on CD19
13.96
101
34.95
100 100
101
4.35 102
0.58
103
83.82
100 100
101
1.17 102
0.25
13.65
103
103
102
102
101
dsRED
1.05
103
103
21.07
101
100 83.80 100 101
1.97 102
103
100 76.84 100 101
1.84 102
103
H-2Kd
b
Bright field
Fluorescence
c
© 2009 Nature America, Inc. All rights reserved.
Hair
Tail
Kidney
Spleen and pancreas
Figure 4 DsRED expression in chimeric mice and F1 germline progeny. (a) Flow cytometry analysis of blood samples from chimeric mice stained for and gated on B cell CD19 expression and showing DsRED and H2-Kd (NOD) expression. The percentages of the CD19+ cells expressing the various markers is shown. (b) Expression of DsRed in various tissues from chimeric mice: hair, tail, kidney, and spleen and pancreas. Scale bars, 5 mm. (c) Hairs showing DsRED from four individual offspring generated when male chimeric mice were mated with NOD females. Top image is a control NOD without DsRED. Scale bars, 50 µm.
offspring, two produced mostly albino offspring with a few agouti pups, and one produced only agouti offspring. To confirm that the albino progeny were indeed NOD derived, we analyzed spleen and blood samples for expression of NOD MHC class I and class II markers as well as for the presence of any detectable C57BL/6 MHC markers (Fig. 3b and data not shown). Additional confirmation of NOD genotype was provided by analysis of four selected polymorphic markers on several chromosomes that distinguish between NOD and C57BL/6 strain backgrounds. These clearly distinguish the germline transmitted NOD genotype of mouse A from the agouti mouse B (Fig. 3c–e). Flow cytometry analysis of blood samples from the two individual mice show only NOD class 1 (Kd) in mouse A and expression of both NOD (Kd) and C57BL/6 (Kb) class 1 in mouse B (Fig. 3d). This is further confirmed by the use of microsatellite markers (D1Mit387, D3Mit182, D6Mit 254 and D17Mit 28) that distinguish between NOD and C57BL/6 on chromosomes 1, 3, 6 and 17 (Fig. 3e and Supplementary Table 1). We chose these markers because they are located near the prominent diabetes susceptibility loci Idd5d, Idd3, Idd19 and Idd1. Diabetes in offspring after germline transmission Female NOD mice spontaneously develop a mononuclear cell infiltrate in the pancreas around the islet area by around 5 weeks of age and type 1 diabetes from around 12 weeks of age2. In most NOD mouse colonature medicine volume 15 | number 7 | july 2009
nies, this incidence reaches around 80–90% by the time the mice are 30 weeks of age2. Diabetes onset in NOD mice can, however, be accelerated by administration of high doses of cyclophosphamide to mice when they have developed a pancreatic infiltrate2. In most studies, diabetes is induced in this way in around 70–80% of mice2. To determine whether the NOD ES cell–derived mice could develop diabetes, we injected eight of the female albino offspring with 250 mg per kg body weight cyclophosphamide when they were 5–6 weeks of age. Six out of the eight cyclophosphamide-treated mice (75%) developed type 1 diabetes. This incidence is comparable to that obtained when we gave cyclophospha mide to NOD mice in our colony16. Of more importance is the spontaneous incidence of type 1 diabetes in the germline-derived mice. A cohort of 36 ES cell–derived female NOD mice generated from the three NOD ES cell lines developed diabetes with similar incidence and kinetics to female NOD mice in the Cambridge Pathology Department colony (Fig. 3f). As expected, immunohistochemical analysis of the pancreas of 12-week-old nondiabetic ES cell–derived mice revealed the presence of CD3+ T cell infiltrates (Fig. 3g). Genetically modified NOD ES cells retain germline competence For the NOD ES cells to be useful, it is crucial to show that they can be genetically modified and still mediate germ line transmission. We stably transfected one of the NOD ES cell lines with a DsRED expression cassette using PiggyBac transposition17,18. After selection in hygromycin and expansion, we injected the transfected ES cells into C57BL/6 blastocysts, which gave rise to chimeras. Flow cytometry analysis of peripheral blood showed the presence of hematopoietic chimerism and expression of DsRED in cells expressing NOD MHC class I (Fig. 4a). Evidence of DsRED expression could be seen in skin, kidney, liver and pancreas (Fig. 4b). Furthermore, these genetically modified cells were capable of efficient germline transmission (Table 1), and germline offspring showed DsRED fluorescence (Fig. 4c). DISCUSSION Inability to generate NOD ES cells capable of robust chimerism and germline transmission has been a longtime frustration to researchers using this key model. Definition of the ground-state culture requirements for sustaining ES cell pluripotency has enabled generation of ES cell lines from mouse strains previously found to be recalcitrant10 and also from rats9,12. We have applied this approach which uses small-molecules to inhibit differentiation together with LIF to generate multiple lines of NOD ES cells. These lines are stable, retaining normal karyotypes and expression of pluripotency markers, until at least passage 14. We injected several of the male lines into C57BL/6 blastocysts, and all generated chimeras. The majority of the chimeric mice were male, indicating efficient sex conversion of the host blastocyst. Most of the chimeric mice showed extensive contribution from the injected ES cells, as detected by coat color, the extent of hematopoietic chimerism and the presence of NOD polymorphic markers. Notably, we detected islet infiltrates in several of the male chimeric mice, and, indeed, two spontaneously developed type 1 diabetes. Our data furthermore show that these ES cells are clearly capable of robust germline transmission. Female germline offspring from these chimeras could be induced to develop diabetes by injection of cyclophosphamide and spontaneously developed type 1 diabetes at a tempo and incidence characteristic of the NOD mouse strain. One of the major reasons that cells such as these are desirable is that they provide the means to identify definitively alleles of genes contributing to diabetes development. Genetic manipulation via NOD ES cells will enable geneticists to determine whether candidate gene loci identified by mapping studies are indeed the genes involved in predisposing to diabetes development or whether other nearby genes in linkage
817
technical reports
© 2009 Nature America, Inc. All rights reserved.
disequilibrium are responsible. We have shown that our NOD ES cells can be genetically modified and still retain their capacity to contribute to the germline. These results indicate that our NOD ES cells will indeed provide a means of carrying out detailed genetic analyses. These cells can be differentiated in vitro, and it will be crucial to establish whether they can be differentiated into beta cells capable of sustained insulin production and glucose responsiveness that can be retained in vivo following transplantation. Such studies will enable researchers to determine whether NOD ES cells can be used to provide a source of replacement beta cells and assess the efficacy of tolerogenic strategies to prevent their autoimmune destruction. They will also allow investigators to assess whether there is any genetic factor influencing beta cell development in NOD mice that may render their beta cells more susceptible to autoimmune destruction and diabetes development. In summary, we have applied ground-state culture conditions to derive NOD ES cells that should prove useful for a range of studies that will enable the diabetes research community to dissect the processes leading to development of type 1 diabetes. METHODS Methods and any associated references are available in the online version of the paper at http://www.nature.com/nm/ Note: Supplementary information is available on the Nature Medicine website. ACKNOWLEDGMENTS We would like to thank G. Guo (Wellcome Trust Centre for Stem Cell Research) for providing the PiggyBac DsRED vector used in our studies, R. Zamoyska (University of Edinburgh) for the FITC-labeled antibody to H-2Kb and K. Tomonari (Fukui Medical School) for the KT3 antibody. We thank N. Holmes and J. Cooke for careful reading of our manuscript. We thank C. Bland and Y. Sawyer for technical assistance. We are grateful to the Wellcome Trust and the Medical Research Council for supporting this research. AUTHOR CONTRIBUTIONS J.N. contributed to the design of the experiments, supervised and conducted the experiments and was involved in writing the manuscript;, K.J. helped conduct the experiments and generated the genetically modified NOD ES cells; J.M.P. helped conduct experiments, conducted all of the in vivo and ex vivo NOD work and generated the manuscript figures; S.A.N. conducted all of the microsatellite
818
experiments in Figures 2 and 3; M.R. helped in the conduct of experiments in Figure 1; W.M. performed the blastocyst injections; A.S. contributed to the design of the experiments and the review of the manuscript; and A.C. contributed to the design of the experiments and supervised and was involved in writing the manuscript. Published online at http://www.nature.com/naturemedicine/ Reprints and permissions information is available online at http://npg.nature.com/ reprintsandpermissions/
1. Redondo, M.J., Fain, P.R. & Eisenbarth, G.S. Genetics of type 1A diabetes. Recent Prog. Horm. Res. 56, 69–89 (2001). 2. Kikutani, H. & Makino, S. The murine autoimmune diabetes model: NOD and related strains. Adv. Immunol. 51, 285–322 (1992). 3. Wicker, L.S., Todd, J.A. & Peterson, L.B. Genetic control of autoimmune diabetes in the NOD mouse. Annu. Rev. Immunol. 13, 179–200 (1995). 4. Serreze, D.V. & Leiter, E.H. Genes and cellular requirements for autoimmune diabetes susceptibility in nonobese diabetic mice. Curr. Dir. Autoimmun. 4, 31–67 (2001). 5. Brook, F.A. et al. The derivation of highly germline-competent embryonic stem cells containing NOD-derived genome. Diabetes 52, 205–208 (2003). 6. Nagafuchi, S. et al. Establishment of an embryonic stem (ES) cell line derived from a non-obese diabetic (NOD) mouse: in vivo differentiation into lymphocytes and potential for germ line transmission. FEBS Lett. 455, 101–104 (1999). 7. Yang, W. et al. Pluripotin combined with LIF greatly promotes the derivation of ES cell lines from refractory strains. Stem Cells 27, 383–389 (2009). 8. Arai, S., Minjares, C., Nagafuchi, S. & Miyazaki, T. Improved experimental procedures for achieving efficient germ line transmission of nonobese diabetic (NOD)-derived embryonic stem cells. Exp. Diabesity Res. 5, 219–226 (2004). 9. Ying, Q.L. et al. The ground state of embryonic stem cell self-renewal. Nature 453, 519–523 (2008). 10. Buehr, M. et al. Capture of authentic embryonic stem cells from rat blastocysts. Cell 135, 1287–1298 (2008). 11. Silva, J. & Smith, A. Capturing pluripotency. Cell 132, 532–536 (2008). 12. Li, P. et al. Germline competent embryonic stem cells derived from rat blastocysts. Cell 135, 1299–1310 (2008). 13. Solter, D. & Knowles, B.B. Immunosurgery of mouse blastocyst. Proc. Natl. Acad. Sci. USA 72, 5099–5102 (1975). 14. McClive, P.J. & Sinclair, A.H. Rapid DNA extraction and PCR-sexing of mouse embryos. Mol. Reprod. Dev. 60, 225–226 (2001). 15. Bradley, A., Evans, M., Kaufman, M.H. & Robertson, E. Formation of germ-line chimaeras from embryo-derived teratocarcinoma cell lines. Nature 309, 255–256 (1984). 16. Brode, S., Raine, T., Zaccone, P. & Cooke, A. Cyclophosphamide-induced type-1 diabetes in the NOD Mouse is associated with a reduction in CD4+CD25+Foxp3+ regulatory T cells. J. Immunol 177, 6603–6612 (2006). 17. Guo, G. et al. Klf4 reverts developmentally programmed restriction of ground state pluripotency. Development 136, 1063–1069 (2009). 18. Wang, W. et al. Chromosomal transposition of PiggyBac in mouse embryonic stem cells. Proc. Natl. Acad. Sci. USA 105, 9290–9295 (2008).
volume 15 | number 7 | july 2009 nature medicine
ONLINE METHODS
© 2009 Nature America, Inc. All rights reserved.
Mice. We maintained NOD, C57BL/6 and C57BL/6 × CBA F1 mice under barrier conditions in the Biological Services Facilities of the School of Biological Sciences at the University of Cambridge. They received standard laboratory food and water ad libitum. All animal experiments were approved by the Ethical Review Committee of the University of Cambridge and authorized by the UK Home Office. Derivation of embryonic stem cells. We flushed NOD embryos at the eight-cell stage from female mice at 2.5 d post coitum and incubated them in organ culture dishes containing KSOM 2 medium (Millipore) supplemented with 1 µΜ of the mitogen-activated protein kinase kinase inhibitor PD0325901 and 3 µΜ of the glycogen synthase kinase-3 inhibitor CHIR99021 (2i, synthesized in the Division of Signal Transduction Therapy, University of Dundee)10. After 2 d, we transferred them to N2B27 medium19 with LIF plus 2i for 1 d. We removed the zona pellucida by brief incubation in acid tyrodes and incubated blastocysts for 1 h in N2B27 supplemented with 20% rabbit antiserum raised against mouse (Sigma). We rinsed them in N2B27 alone and transferred them to N2B27 supplemented with 20% non–heat-inactivated rat serum as a source of complement. After about 1 h, we removed the lysed trophectoderm and plated the isolated inner cell masses individually into gelatinized wells of a 96-well plate containing N2B27 supplemented with 2i plus LIF. After 5–7 d, we trypsinized clumps and replated them into new wells. We then expanded the ES cell lines. Karyotyping, sexing and immunostaining of embryonic stem cells. We treated subconfluent embryonic stem cells with 100 ng ml–1 colchemid (Karyomax, Invitrogen) for 2 h at 37 °C before trypsinization. We centrifuged the cells, resuspended them in 5 ml of hypotonic solution (75 mM KCl) and incubated them for 10 min at room temperature (20 °C). We added 100 µl of freshly prepared fixative (methanol:acetic acid (3:1)) dropwise and repelleted the cells. We then fixed the cells in methanol:acetic acid, pelleted them and resuspended them in 1 ml of fresh fixative. To generate metaphase spreads, we dropped the fixedcell suspension onto a clean microscope slide and mounted it in Vectashield with DAPI (Vector Labs). We stained ES cells for Nanog (rabbit polyclonal IgG; Abcam) and Oct 4 (mouse monoclonal antibody, Santa Cruz), both at 1 in 200 dilutions, and visualized them with donkey antibody to rabbit IgG (Molecular Probes) or goat antibody to mouse IgG (Molecular Probes) respectively, both at 1 in 4,000 dilutions. We extracted genomic DNA from NOD ES clones with a DNeasy Blood and Tissue kit (Qiagen), and we determined the sex by PCR for the Y chromosome– specific Sry gene14. Establishment of embryonic stem cell clones constitutively expressing DsRED. We transfected a PiggyBac vector together with the DsRED reporter gene under the control of the constitutive CAG promoter17 with pPBase18 into NOD ES line 1 using Lipofectamine 2000 (Invitrogen). We selected stable clones in N2B27
nature medicine
plus 2i plus LIF containing 200 µg ml–1 hygromycin. We picked resistant colonies emerging after 10–14 d and expanded them initially with hygromycin. Generation of chimeras. We injected 10–15 trypsinized NOD ES cells into each C57BL/6 blastocyst and then transferred the blastocysts to the uterus of a pseudopregnant F1 female mouse at 2.5 d post coitum. We assessed the resulting chimeras for germline transmission by mating them with NOD mice. Flow cytometric analysis of blood and pancreatic infiltrates. We collected blood samples into heparin (Sigma). We lysed red blood cells with deionized water. To examine pancreatic infiltrates, we harvested each pancreas into cold PBS containing 5% FCS, 56 mM glucose (Sigma) and Complete Mini Protease inhibitors (Roche). We incubated the tissues in 1 ml PBS containing 15% FCS, 0.3 mg ml–1 Liberase CI (Boehringer Mannheim) and 10 µg ml–1 DNAse (Sigma). After digesting for 10 min at 37 °C, we washed the tissues and prepared cell suspensions by mashing the tissues through a 70-µm cell strainer. We left the suspensions to settle for 10 min and collected the supernatants. We blocked Fcγ receptor binding by incubating the cells with 2.4G2, and then we stained the cells in FACS buffer containing 2% FCS. We used combinations of antibodies to CD4 (RM4-5, PerCP conjugated), CD19 (1D3, phycoerythrin (PE) conjugated), CD8 (53-6.7, PE conjugated), H-2Kd (FITC conjugated) and I-Ab (PE conjugated) all from BD Biosciences. Antibody to I-Ag7 (OX-6) was FITC conjugated and was from Serotec. We analyzed cells with a FACScalibur (BD Biosciences) and FlowJo (Tree Star) software. Immunohistochemistry. We snap-froze pancreases in isopentane. We air-dried 5-µm cryostat sections and fixed them in acetone for 10 min. We detected pancreatic beta cells by preblocking sections with 20% normal mouse serum followed by incubating with guinea pig antibody to porcine insulin (DAKO) in 10% normal mouse serum, and we visualized them by rhodamine-conjugated goat antibody to guinea pig IgG (ICN Pharmaceuticals). We detected CD3+ T cells using KT3 and visualized them with FITC-conjugated rabbit antibody to rat IgG (Serotec). Microsatellite analysis of tissue samples. We took selected tissues and purified genomic DNA as described above. We chose microsatellite markers spanning Idd susceptibility loci able to distinguish NOD mice from C57BL/6 mice using UniSTS (US National Center for Biotechnology Information), and we designed appropriate PCR primers (Supplementary Table 1). We performed Taq PCR (Biotaq Bioline) with 2 ng of starting genomic DNA and 1.5 mM MgCl2 and cycles as follows: 95 °C for 5 min followed by 40 cycles of 95 °C for 30 s, 60 °C for 30 s and 72 °C for 1 min. We resolved PCR products by electrophoresis on a 4% ethidium bromide agarose gel. 19. Ying, Q.L., Stavridis, M., Griffiths, D., Li, M. & Smith, A. Conversion of embryonic stem cells into neuroectodermal precursors in adherent monoculture. Nat. Biotechnol. 21, 183–186 (2003).
doi:10.1038/nm.1996
