This protocol describes the isolation of double-negative thymocytes from the mouse thymus followed by retroviral transduction and co-culture on the delta-like 4-expressing bone marrow stromal cell line co-culture system (OP9-DL4) for further functional analysis.
Transduced mouse immature thymocytes can be differentiated into T cells in vitro using the delta-like 4-expressing bone marrow stromal cell line co-culture system (OP9-DL4). As retroviral transduction requires dividing cells for transgene integration, OP9-DL4 provides a suitable in vitro environment for cultivating hematopoietic progenitor cells. This is particularly advantageous when studying the effects of the expression of a specific gene during normal T cell development and leukemogenesis, as it allows researchers to circumvent the time-consuming process of generating transgenic mice. To achieve successful outcomes, a series of coordinated steps involving the simultaneous manipulation of different types of cells must be carefully performed. Although these are very well-established procedures, the lack of a common source in the literature often means a series of optimizations are required, which can be time-consuming. This protocol has been shown to be efficient in transducing primary thymocytes followed by differentiation on OP9-DL4 cells. Detailed here is a protocol that can serve as a quick and optimized guide for the co-culture of retrovirally transduced thymocytes on OP9-DL4 stromal cells.
The OP9 bone marrow stromal cell line provides a useful in vitro system for the induction of lymphopoiesis from several sources of progenitors1. The first studies that used OP9 cells demonstrated that the lack of macrophage colony-stimulating factor (MCSF) expression contributed to the ability of the OP9 cell line to support hematopoiesis and B cell differentiation from bone marrow-derived hematopoietic stem cells (HSCs), as was later also shown for embryonic stem cells (ESCs)2,3,4,5. In previous studies, the generation of delta-like 1/4-expressing OP9 cells (OP9-DL1/OP9-DL4) enabled the induction of T cell lineage commitment6 and demonstrated the ability to successfully recapitulate thymic maturation7,8. Briefly, T cell development has been described by the sequential expression of CD4 and CD8 molecules. Immature thymocytes are "double-negative" (DN, CD4− CD8−) and can be further subdivided according to the surface expression of CD44 and CD25. DN thymocytes differentiate through the immature single-positive (ISP) stage, characterized by CD8 expression in mice and CD4 in humans, followed by the double-positive (DP) stage, characterized by the co-expression of CD4 and CD8, and, finally, the mature single-positive stage, characterized by the expression of CD4 or CD89. HSCs express the Notch1 receptor, which normally interacts with the delta-like 4 (DL4) expressed on thymic epithelial cells to induce T lineage differentiation10. Hence, interest in the OP9-DL1/4 model has progressively increased, leading to the extensive use of this approach in a wide variety of applications in the past two decades8,11,12,13. Although DL1 and DL4 are both able to support T cell differentiation in vitro, they show differential requirements in vivo, and a few studies have suggested that OP9_DL4 is more efficient than OP9_DL1 in recapitulating the mouse thymic environment7,14.
Among the potential applications of the OP9-DL1/4 system, there is particular interest in the combination of this system with the transduction of DN cells or HSCs with retroviral vectors. This combination is an effective way to manipulate gene expression during normal T cell development and leukemogenesis and has been shown to be an efficient method to induce or inhibit the function of a gene of interest15,16,17. This model has been particularly successfully used to study the collaboration between the oncogenes driving leukemia15 as it is flexible and enables the examination of the effects of multiple gene combinations in a reasonable time, in contrast to generating transgenic mice. Moreover, similar models have been used previously to assess the effects of introducing oncogenes into normal cells15,16,17. Additionally, retroviral transduction requires dividing cells for transgene integration18, and while lentiviral transduction would overcome this limitation by eliminating the need for dividing cells for transgene integration, we have been unable to achieve the successful transduction of DN thymocytes using lentiviral vectors. Thus, OP9-DL1/DL4 is a suitable tool for growing hematopoietic progenitor cells.
The standard protocol for the lymphopoiesis of transduced thymocytes on OP9-DL4 involves a series of coordinated steps that must be carefully performed to achieve a successful outcome. Although these are techniques that have been serving the community well for many years, often the protocols available in the literature are fragmented. As a result, each laboratory is forced to adapt and optimize different stages of the procedure, which can be time-consuming. Here, this protocol describes the isolation of DN thymocytes from the mouse thymus, followed by retroviral transduction and co-culture on OP9-DL4 stromal cells for further functional analysis. This established protocol has been shown to be efficient and reproducible in transducing primary thymocytes, followed by differentiation on OP9-DL4 cells or the induction of T cell acute lymphoblastic leukemia15.
All animal experiments described were approved by the NIH Institutional Biosafety Committee (IBC) and Animal Care and Use Committee (ACUC). See the Table of Materials for details related to all the reagents and materials used in this protocol. Refer to published guidelines19 for more details on retrovirus-producer cell culture and maintenance procedures. See Figure 1 for an overview of the protocol.
1. Starting the culture of OP9-DL4 cells (Day 1)
2. Starting the culture of the retrovirus-producer cell line (Day 1)
3. Beginning the culture of the retrovirus-producer cells in 6-well plates (Days 4–5)
4. Transfecting the retrovirus-producer cells to generate the retrovirus containing the genes of interest (Days 5–6)
5. Changing the retrovirus-producer cell medium (Days 6–7)
6. Thymocyte preparation and depletion of CD4+ and CD8+ cells (Days 6–7)
7. Culturing thymocytes on OP9-DL4 cells (Days 6–7)
8. Harvesting the retrovirus from the supernatant and using it to transduce the thymocytes (Days 8–9)
9. Maintaining transduced thymocytes on OP9-DL4 culture for 2–5 days or freezing as needed (Day 9+)
The depletion efficiency can be assessed flow cytometrically by labeling the magnetically unlabeled cell fraction for CD4 and CD8 after immunomagnetic cell separation (MACS) and analyzing this on a two-dimensional bivariate dot-plot (Figure 3). A good yield of double negative (CD4−, CD8−) cells is 95% or above, as represented in Figure 3. Two of the most common causes of lower yield are the miscalculation of the microbeads based on the number of cells and the number of labeled cells exceeding the column capacity. It is recommended to choose the correct number of MACS columns according to the number of labeled cells. When working with thymocytes, the number of labeled cells (DP and SP cells) is almost equal to the number of total cells (more than 96%). Cell counting can be done in an automatic cell counter, a Neubauer chamber, or with any alternative method. As a result, the volumes allocated for cell counting may need to be adjusted depending on the specific machine and the chosen counting method.
It is important to determine the number of cells present before and after depletion. These cell counts are needed to calculate the number of LD depletion columns needed and to evenly distribute the DN thymocytes into the appropriate number of OP9-DL4 flasks. The controls for flow cytometry (unlabeled cells, cells labeled for CD4, and cells labeled for CD8) can be performed with the reserved pre-depletion sample, as this contains more cells. However, as most of the cells are expected to be retained within the column, the post-depletion sample to be labeled will require a higher volume. Consequently, adjustments may need to be made according to the labeling protocol selected.
When using vectors that express screenable markers, such as a fluorescence gene, the transfection and transduction can be roughly and empirically assessed by fluorescence microscopy (Figure 2). The transduction efficiency can be analyzed by harvesting the thymocyte from the OP9-DL4 monolayer, as described in step 8.3, and looking at the expression of a fluorescence gene by flow cytometry. The efficiency of transduction using an empty retroviral vector with GFP as the reporter gene was 84.2% (Figure 4).
T cell differentiation on OP9-DL4 cells can be observed 4 days after transduction. Flow cytometry is usually performed to assess the cell differentiation induced by the co-culture on OP9-DL4 cells and/or the transgene expression, as represented in Figure 5, where the cells were labeled for CD4, CD8, CD44, and CD25. There are several possible combinations of cell surface molecule labeling that have been shown to be useful for investigating the molecular and cellular mechanisms of T cell development in mice24,25,26,27. Therefore, the panels of fluorescent antibodies may vary according to the question of interest being addressed. The transduction of DN thymocytes with the empty retroviral vector pMIG, shown in panel B, presented approximately the same proportions of single positives (CD4+ or CD8+), double positives (CD4+/CD8+), double negatives (CD4−/CD8−), and its substages double-negative 1–4 (DN1–DN4) as the untransduced thymocytes, shown in panel A, indicating that T cell development was not affected by the transduction process.
Figure 1: Diagram of the steps of thymocyte isolation, transduction, and co-culture. Abbreviation: DN = double-negative. Please click here to view a larger version of this figure.
Figure 2: Co-culture fluorescence microscopy of GFP-transduced thymocytes or untransduced thymocytes and OP9-DL4 cells. (A) Untransduced thymocytes and (B) stable GFP-expressing murine thymocytes on OP9-DL4 at day 3 after the second transduction. An Olympus-IX71 fluorescence microscope with a 40x lens and a 480/30 filter suitable for GFP detection was used. Scale bar = 40 µm. Abbreviations: GFP = green fluorescent protein; DN = double-negative. Please click here to view a larger version of this figure.
Figure 3: Flow cytometry of thymocyte depletion. Representative flow cytometry plots analyzing the expression of CD4 and CD8 on thymocytes obtained from 7-8 week old C57BL/6J female mice (A) pre-depletion and (B) post-depletion of CD4+ and CD8+ using microbeads and LD columns according to the manufacturer's instructions. The dot plots on the left show gates based on the size and complexity of the event (FCS-A and SSC-A, respectively). The middle panel shows FSC-H versus FSC-A to gate single cells and exclude doublets. In the plots on the right, cells were defined as CD4 and CD8 from the single-cell gate. Abbreviations: FSC-A = forward scatter-peak area; SSC-A = side scatter-peak area; FSC-H = forward scatter-peak height; FITC = fluorescein isothiocyanate; PE = phycoerythrin. Please click here to view a larger version of this figure.
Figure 4: Flow cytometry of the thymocyte retroviral transduction efficiency. (A) Untransduced thymocytes co-cultured on OP9-DL4; (B) retrovirally transduced thymocytes on OP9-DL4 at day 3 post transduction. Abbreviations: FSC-A = forward scatter-peak area; SSC-A = side scatter-peak area; FSC-H = forward scatter-peak height; FITC = fluorescein isothiocyanate; PE = phycoerythrin; DN = double-negative. Please click here to view a larger version of this figure.
Figure 5: Flowcytometry of transduced thymocytes after 4 days of OP9-DL4 co-culture. (A) Untransduced and (B) transduced with pMIG retrovirus. The cells were first gated on live cells and then gated based on size and complexity (FCS-A and SSC-A, respectively), followed by plotting FSC-H versus FSC-A to gate on single cells and exclude doublets. For untransduced cells, we used the following gate strategy. From the single-cell gate, the cells were defined as single positives (CD4+ or CD8+), double positives (CD4+/CD8+), and double negatives (CD4−/CD8−). Then, from the double negatives (CD4−/CD8−) gate, the cells were defined as CD44+, CD25+, CD44+/CD25+, and CD44−/CD25−, as shown in panel A. For the pMIG transduced cells, shown in panel B, from the single-cell gate, the cells were first defined as GFP+/GFP−, and then from GFP+ cells, the cell population distribution into the major T cell development stages, as defined by CD4, CD8, CD44, and CD25 expression, was determined using the same gate strategy as for the untraduced cells. Abbreviations: FSC-A = forward scatter-peak area; SSC-A = side scatter-peak area; FSC-H = forward scatter-peak height; FITC = fluorescein isothiocyanate; GFP = green fluorescent protein. Please click here to view a larger version of this figure.
The protocol described here was developed specifically for thymus-derived DN (CD4−/CD8−) T cell studies with retroviral transfection followed by an OP9-DL4 differentiation model. However, it is likely that the target cells that will be subjected to this protocol of transduction followed by cell differentiation will have a wider interdisciplinary utility. Thus, in addition to immature thymocytes, hematopoietic stem cells, such as cells derived from either fetal liver or bone marrow, could potentially be used in this protocol.
The OP9-DL4 system has proven to be an effective model to study gene function in a variety of aspects, including cell differentiation17 and oncogenesis15. While the retroviral modification of hematopoietic progenitors is a well-established technique that enables stable genetic modification, combining the induction of cell differentiation on the OP9-DL4 system and retroviral transduction requires care and skill. The critical aspect for achieving success with this protocol is ensuring all the steps are well-coordinated, since the protocol involves using three different cell types that need to be kept healthy and at the ideal confluence required for each specific stage. With that in mind, it is important to perform all the quality control checkpoint analyses after the execution of each step before proceeding to the next step. This will ensure that all the steps are working. Therefore, checking the depletion, transfection, and transduction efficiencies is important for the successful execution of this protocol (see a typical result for transduction efficiency in Figure 4). Good primary cell transduction efficiency is linked to a high viral titer. Typically, larger inserts result in lower virus titers18. For training purposes, we use an empty vector to represent the results that can be obtained with this protocol. In our experience, the transduction and transfection efficiencies vary according to the insert size, especially considering the viral backbone-expressing reporter genes, such as GFP. One strategy that can be used when studying the interaction of more than one gene is to clone each gene into a different transfer vector, followed by individual virus production, and finally, co-transduction of the target cell. In that case, a selection step may be applied to eliminate the singly transduced cells and retain only the co-transduced cells.
It is worth noting that the majority of OP9-DL1/DL4 stromal feeder layer cell lines have been genetically engineered to express GFP as part of the DL1 or DL4 constructs7. In this protocol, we used a retroviral vector that also expresses GFP; however, it is brighter than the GFP protein expressed by OP9-DL4 cells and does not interfere with the transduction inspection when visualizing the co-culture under the fluorescence microscope.
OP9 cells differentiate into adipocytes after many passages, long periods in culture, or under conditions of over-confluency19. This is evidenced by the development of large vacuoles. Thus, OP9 cells presenting these characteristics should not be used in this protocol. Transfecting overly confluent retrovirus-producer cells will result in a low virus titer. Indeed, the subconfluent stage is when the cells are most transfectable. Furthermore, transfecting low-confluence retrovirus-producer cells will decrease the cell stress in the transfection process and give the highest virus titer.
While, in this protocol, we do not titrate the virus supernatant, virus supernatant titration must be considered in some cases, such as in the absence of a reporter gene in the retroviral vector, which would prevent the indirect determination of viral production, or in cases where the experimental design requires a more accurate number of the transgene copies to be integrated into the target cell genome. However, it is important to note that the titer of retroviral vector supernatant decreases significantly when stored at −80 °C or 4 °C until the titration results are obtained. Therefore, using freshly prepared virus supernatant for transduction will yield better transduction efficiency.
The thymus contains a large number of double-positive (DP) thymocytes (more than 85%) and about 10% single-positive15 cells (CD4 or CD8), which are the post-DN stage thymocytes. The DP cells cannot survive in vitro for retroviral manipulation, while the SP cells are untransducible. Therefore, this protocol can be applied to generate retroviral vector transducible DN thymocytes.
The authors have nothing to disclose.
This work was supported by the intramural program of the National Cancer Institute, project ZIABC009287. OP9-DL4 was obtained from Dr. Juan Carlos Zúñiga-Pflücker (Sunnybrook Health Sciences Centre, Toronto, ON, Canada). The authors thank the NCI-Frederick Laboratory Animal Sciences Program for their continued technical assistance and experimental advice and input, as well as Jeff Carrel, Megan Karwan, and Kimberly Klarmann for flow cytometry assistance. We are grateful to Howard Young for critical advice and input.
2-mercaptoethanol | Sigma | M3148 | |
ACK Lysis buffer | Lonza | 10-548E | |
BSA | Cell Signaling Technology | 9998S | |
CD4 Microbeads | Miltenyi | 130-049-201 | |
CD8 Microbeads | Miltenyi | 130-049-401 | |
Centrifuge 5910R | eppendorf | 5942IP802353 | |
DMEM | Corning | 10-013-CV | |
EDTA | Invitrogen | 15575-038 | |
Fetal calf serum HyClone FBS | ThermoScientific | SH30910.03 | |
LD columns | Miltenyi | 130-042-901 | |
L-glutamine | Sigma | G7513 | Freeze glutamine in aliquots and use freshly-thawed glutamine |
Lipofectamine 2000 | Invitrogen | P/N 52887 | |
MEM-alpha Medium | Gibco | 12561-072 | |
OPTI-MEM I Reducing Serum Medium | Invitrogen | 31985-062 | |
PBS pH 7.2 | Corning | 21-040-CV | |
pcL-Eco Plasmid | Addgene | 12371 | |
penicillin/streptomycin | Gibco | 15140-122 | |
pMIG Plasmid | Addgene | 6492 | |
Polybrene | Chemicon | TR-1003-G | |
Pre-Separation Filters | Miltenyi | 130-041-407 | |
recombinant hFLT-3L | PeproTech | 300-19 | |
recombinant IL-7 | Peprotech | 217-17 | |
Retrovial packaging cell line Phoenix-Eco | Orbigen | RVC-10002 | |
Syringe filter (0.45 µm) | Millipore | SLHV033RS |