Derivation and characterization of mouse embryonic stem cells (mESCs) from permissive and non-permissive strains

1 Genetic Resource Science, The Jackson Laboratory, Bar Harbor, ME, USA.

Candice Byers

1 Genetic Resource Science, The Jackson Laboratory, Bar Harbor, ME, USA.

Ian Greenstein

1 Genetic Resource Science, The Jackson Laboratory, Bar Harbor, ME, USA.

Nadine Schrode

2 Developmental Biology Program, Sloan-Kettering Institute, New York, New York, USA.

Leah Rae Donahue

1 Genetic Resource Science, The Jackson Laboratory, Bar Harbor, ME, USA.

Anna-Katerina Hadjantonakis

2 Developmental Biology Program, Sloan-Kettering Institute, New York, New York, USA.

Laura Reinholdt

1 Genetic Resource Science, The Jackson Laboratory, Bar Harbor, ME, USA.

Abstract

Mouse embryonic stem cells (mESCs) are critical tools for genetic engineering, development of stem cell based therapies, and basic research on pluripotency and early lineage commitment. However, successful derivation of germline-competent embryonic stem cell lines has, until recently, been limited to a small number of inbred mouse strains. Recently, there have been significant advances in the field of embryonic stem cell biology, particularly in the area of pluripotency maintenance in the epiblast from which mESCs are derived. Here we describe a protocol for efficient derivation of germline competent mESCs from any mouse strain, including strains previously deemed non-permissive. We provide a primary method that is generally applicable to most inbred strains, as well as an alternative method for non-permissive strains. Using this protocol, mESCs can be derived in 3 weeks and fully characterized after an additional 12 weeks, at efficiencies as high as 90% and in any strain background.

Introduction

Embryonic stem cells (ESCs) are the ex vivo equivalent of the epiblast lineage of the blastocyst and therefore, share the same developmental potential to differentiate into any one of the three primary germ layers, mesoderm, definitive endoderm and ectoderm ( Figure 1 ). This developmental pluripotency combined with a high capacity for self-renewal in vitro are defining features of ESCs. Mouse embryonic stem cells (mESCs) are derived from pre-implantation stage embryos 1,2 . The progenitor cells that give rise to mESCs reside in the epiblast of the late blastocyst (~4 days post coitum) and express several pluripotency-associated factors, including Oct3/4 (Pou5f1) and Nanog 3 . In addition to their capacity for self-renewal and stable pluripotency in vitro, mESCs have the defining capacity to populate the germline after microinjection into, or aggregation with, host embryos, making mESCs essential tools for genetic engineering 4 . The Nobel prize-winning discovery that genes could be genetically modified in mice using mESCs was published over 30 years ago and since, nearly 50,000 genetically modified alleles have been created by individual investigators around the world and by the International Knockout Mouse Consortium (IKMC, www.knockoutmouse.org), which endeavors to create null and/or conditional null alleles for every gene in the mouse genome 5,6 . mESCs are also used for basic research on pluripotency and for development of stem cell based therapies as the starting material for directed differentiation of enriched, defined cell types in vitro.

The pre-epiblast lineage in the early embryo is defined by lineage-restricted expression of the Oct3/4, Nanog and Gata6 genes. As early lineage specification proceeds, the pluripotent epiblast lineage is defined by Nanog expression. The epiblast lineage will give rise to all three definitive germ layers of the embryo-proper, namely all somatic cells and germ cells, and is the population from which mESCs are derived. mESCs cell lines retain the developmental potential of the epiblast lineage and as such, can contribute to all three germ layers and the germline of host blastocyst or morula stage embryos.

Derivation of mESCs

Despite knowledge of the basic requirements for mESCs to maintain pluripotency, derivation of mESCs remained inefficient and was limited to just a few mouse strains for many years 7 . These, so-called permissive strains included 129 sub-strains, as well as the most commonly used inbred mouse strain, C57BL6. Early protocols demonstrated the requirement of leukemia inhibitory factor (LIF) to activate STAT3, bone morphogenic protein (BMP) (or serum), and mitotically inactivated feeder layers, preferably mouse embryonic fibroblasts (MEFs), to prevent differentiation of mESCs in vitro. However, derivation efficiency in permissive strains was at best 30%, as determined by the percentage of embryos giving rise to stable mESC lines 8,9 . Moreover, in non-permissive mouse strains, like CBA, NOD, DBA and others derivation efficiency was either extraordinarily low or non-existent 7 . Therefore, for mammalian geneticists who rely on specific inbred strain backgrounds for human disease modeling, genetically engineered alleles created in 129, C57BL6 or hybrid ES cells required 10 to 20 backcross generations (up to 2 years) to create the desired genetic background. Moreover, for mammalian species other than the mouse, genetic engineering was simply not possible due to an inability to derive legitimate ES cells despite considerable effort over many years. Recent advances in site-specific nuclease technologies (e.g ZFN, TALEN, CRISPR/Cas) are enabling direct, targeted deletion and targeted, sequential gene modifications via pronuclear injection of mouse embryos 10 and other species, including rat 11,12 . For genetic engineering, these technologies circumvent the need for ES cells. However, the applicability for multifaceted genomic modifications via homologous recombination with large inserts, across a variety of strains has not yet been demonstrated.

Derivation of mESCs from non-permissive strains

To overcome strain and species limitations to ES cell derivation, a variety of approaches have been used. For example, based on the premise that the presence of primitive endoderm caused loss of pluripotency in mES cell progenitors within the inner cell mass, careful excision of the epiblast by biopsy or immunosurgery was shown to improve derivation efficiency 3,13 . In addition, for many years, delayed implantation or diapause induction by ovariectomy or tamoxifen injection was also used to promote derivation efficiency possibly via developmental stasis during which epiblast have an opportunity to expand. Finally, as knowledge of the genes required for early lineage specification and pluripotency has grown, protocols for efficient derivation of mES cells by promoting or inhibiting expression of specific genetic pathways were developed. Oct4 (Pou5f1) is a transcription factor that is essential for the maintenance of pluripotency in cells of the inner cell mass (ICM), the epiblast and in mES cell lines. Importantly, loss of Oct4 was shown to be a feature of cultured embryos that failed to give rise to stable ES cell lines 14 . Based on this discovery, culture conditions that promote Oct4 expression, namely inhibition of the MAP kinase pathway, were introduced. However, successful derivation of mES cells from the recalcitrant strain background, CBA, still required a combination of diapause induction, epiblast excision and inhibition of MEK kinase via PD98059 14 . In the context of these modifications to traditional ES cell derivation protocols, derivation efficiency in CBA was ~25%, a significant advance for a non-permissive strain 14 .

The pluripotent ground state and overcoming barriers to mESC derivation

The discovery that self-renewal and pluripotency are intrinsic properties of mESCs was later demonstrated by Austin Smith and colleagues 14 , who showed that inhibition of MEK/ERK and glycogen synthase kinase-3 (GSK3) signaling (3i: PD184353, PD173074 / SU5402 and CHIR99021 respectively) were together sufficient, combined with activation of STAT3 by LIF (3i/LIF), to promote the pluripotent ground state of emergent ESCs from mice and from rats 15–17 . These laboratories went on to show that inhibition of FGF receptor signaling is dispensible in the context of more potent inhibition of MEK signaling (2i: CHIR99021 to inhibit GSK3β and PD0325901 to inhibit MEK1/2) 16 . Both 3i/LIF and, subsequently, 2i/LIF culture conditions have since been successfully applied for efficient (50–70%) derivation of germline competent mESCs from recalcitrant strains like NOD, CBA and DBA 18–21 . Moreover, these culture conditions have been used to successfully derive germline competent rESCs from rat embryos 16,17 , an accomplishment that quickly led to the creation of the first rat gene knockout by homologous recombination in rESCs 22 . Successful derivation of ESCs from recalcitrant strains and from rat using 2i/LIF culture conditions suggests that emergent ESCs from these strains / species are unable to maintain a pluripotent ground state under traditional ESC culture conditions (serum +LIF). In fact, it was later shown that unlike emergent ESCs from permissive strain background (e.g. 129), emergent ESCs from non-permissive strain backgrounds (e.g. NOD) are unstable and differentiate to a more advanced, EpiSC (post-implantation, epiblast stem cell) state, which has been termed a primed pluripotent state, in the absence of exogenously provided inhibitors of ERK signaling 23 .

Although the basis of strain and species recalcitrance to ESC derivation is not yet fully understood, these results suggest that inhibition of the pathways responsible for differentiation of inner cell mass epiblast cells to post-implantation epiblast cells might be sufficient to overcome barriers to mESC derivation in all inbred strain backgrounds. This new model of the pluripotent, ground state of ESCs is an important advance in our understanding of early lineage commitment and has informed our mESC derivation protocol, which is highly efficient, regardless of strain background.

Experimental Design

We previously published efficient derivation of germ line competent mESC lines from the recalcitrant strain DBA/2J 20 . Crucial to the success of this protocol was the exclusion of serum during the outgrowth phase, combined with inhibition of MEK / ERK (1i: PD98059) signaling during the outgrowth phase and during subsequent culture of emergent ES cell lines (3i: CHIR99021, PD173074 and PD032901). Since published data later showed the FGF receptor inhibitor, PD173074, to be dispensible and the MEK inhibitor, PD98059, redundant, in the context of the more potent MEK inhibitor, PD032901 16 , our current protocol utilizes the now standard 2i combination (CHIR99021 and PD032901) to achieve the same exogenous inhibition with simpler media formulae.

Our protocol begins with the harvest and culture of late blastocyst stage embryos, which can be generated by natural mating or by in vitro fertilization. These embryos are then cultured in derivation medium to allow for ICM outgrowth. Unlike traditional ESC derivation medium, which contains serum, our derivation medium utilizes serum replacement in the form of an artificial serum replacement or, in the case of non-permissive strain backgrounds, we use defined serum free medium 24 . The exclusion of serum from the derivation medium was previously shown to promote mESC derivation efficiency 9,25 and we have found that it promotes NANOG expression in ICM outgrowths ( Figure 2 ). NANOG, which is essential for acquisition and maintenance of pluripotency, is a biomarker of mESC progenitor cells and is a more reliable readout of cell potency than OCT3/4 14,26 . Upon ICM disaggregation, incipient mESC lines are cultured either in traditional ESC medium +2i/LIF (variant A, step 10A) or in defined serum free media +2i/LIF (as described by Silva et al 18 , Ying and Smith 24 and this protocol, variant B step 10B). We, and others, have found the latter culture conditions to be essential for robust derivation of ES cell lines from the recalcitrant NOD and it’s derivative strains (NSG and NRG) ( Table 1 ) 19 . Although the original defined, serum free +2i/LIF conditions were created for feeder free derivation and culture, feeder layers improve derivation efficiency, promote colony attachment (which is favorable for sub-cloning, a key manipulation in gene targeting experiments) and may provide enhanced karyotypic stability. Therefore, we employ feeder layers of mitotically inactivated embryonic fibroblasts. Importantly, this protocol provides all of the steps necessary to derive and establish low passage (P3) ES cell lines from any inbred strain background, as well as detailed instructions for quality assurance and characterization. Regardless of the culture conditions, pluripotency and euploidy degrade with increasing passage number in mESCs (P20 and higher). Therefore, newly established mESC lines should be maintained with an eye towards maximizing and preserving low passage stocks.

NANOG immunolabeling in inner cell mass (ICM) outgrowths grown in the presence of 2i/LIF with or without serum or KOSR. ICM outgrowths grown in traditional derivation medium containing serum (A) have minimal NANOG expression (B) and exhibit differentiation (orange arrows). (B) Outgrowths grown in the presence of KOSR (knockout serum replacement) contain NANOG expressing cells (white arrows) but also exhibit differentiation (B, orange arrows). (C) Outgrowths grown in serum free media exhibit NANOG expressing cells (white arrows) with minimal differentiation. Scale bars, 50 µm. All procedures involving mice were approved by The Jackson Laboratorys and Sloan-Kettering Institutes Institutional Animal Care and Use Committees and were performed in accordance with the National Institutes of Health guidelines for the care and use of animals in research.

Table 1

Derivation efficiencies in selected strain backgrounds using traditional derivation conditions and this protocol.

Sub-strains are provided if known / reported. Efficiency is the percentage of disaggregated embryos that gave rise to stable mESC lines, as defined by morphology and survival through passaging. In the literature, efficiency is defined by the number of stable mESCs generated from the number of embryos collected. With this protocol, the vast majority (90%) of embryos give rise to an outgrowth, but not every outgrowth is disaggregated due to budget constraints. Therefore an efficiency calculation based on number of embryos harvested would be a gross underestimate.

b Required serum free culture conditions (as descried by Ying and Smith 24 and this protocol, variant B).

MATERIALS

REAGENTS

Derivation of mES cells

Pregnant mouse, 3.5 days post coitum (d.p.c) CAUTION: Experiments involving rodents must conform to all relevant governmental and institutional (IACUC) standard operating procedures and regulations

2-Mercaptoethanol 55mM (Invitrogen, cat. no. 21985-023) CAUTION: This reagent is a biohazard; adequate safety instructions should be taken when handling.

Dulbecco’s Modified Eagle Medium (DMEM), high glucose, no glutamine, no sodium pyruvate (Invitrogen, cat. no. 11960069)

Dimethyl sulfoxide, DMSO (Sigma-Aldrich, cat. no. D2650) CAUTION: This reagent is a biohazard; adequate safety instructions should be taken when handling

Ethanol, 70% (v/v) (Sigma Aldrich, cat. no. E7023) CAUTION: This reagent is a highly flammable and a skin and eye irritant.

Fetal bovine serum (FBS), ES grade (Lonza, cat. no. 14-501F) CRITICAL: FBS should be lot tested for maximum ES cell viability and growth. Filter through 0.45 µm filter and store 25 ml single-use aliquots at −20°C for up to 1 year.

Karyotyping

Glacial Acetic Acid (Sigma Aldrich, cat. no. A6283) CAUTION: This reagent is a biohazard and severe irritant; adequate safety instructions should be taken and all work with this chemical should be conducted in a fume hood.

Methanol (Sigma Aldrich, cat. no. 494437) CAUTION: This reagent is a biohazard; adequate safety instructions should be taken when handling

Immunolabeling

4',6-diamidino-2-phenylindole (DAPI) (Invitrogen, cat. no. D1306) (see reagent setup) or Hoechst (Invitrogen, cat. no. 33342).

Paraformaldehyde, 4% (PFA) (Electron Microscopy Sciences, cat. no. 157-4) CAUTION: This reagent is a biohazard; adequate safety instructions should be taken when handling

SNP Panel/Genotyping

Pathogen testing

EQUIPMENT

Calibrated, glass micropipets, 50 µl (Drummond Scientific, cat. no. 2-000-050) note: aspirator apparatus required for mouth-controlled embryo transfer pipet is supplied with the micropipets (see equipment setup). Alternatively, 9” glass Pasteur pipets, pulled over a flame can be assembled with a plastic mouthpiece and P1000 pipet tip as shown in Figure 3 .

Overview of blastocyst stage embryo harvest. Embryos are flushed from the uteri at 3.5 – 3.75 days post coitum (d.p.c.) using M2 media. Embryos are pooled and washed through through a series of M2 drops using a mouth-controlled pipet prior to plating for ES cell derivation. It is important to get rid of tissue debris prior to embryo plating. It also provides an opportunity to assess the stage and quality of embryos recovered for stem cells derivation. All procedures involving mice were approved by The Jackson Laboratorys and Sloan-Kettering Institutes Institutional Animal Care and Use Committees and were performed in accordance with the National Institutes of Health guidelines for the care and use of animals in research.

Cell counter, automatic (e.g Nexcelom Cellometer Auto T4) or manual haemocytometer (e.g. Reichert Bright-Line, Fisher Scientific, cat. no. 02-671-5)