Retroviral Overexpression of CXCR4 on Murine B-1a Cells and Adoptive Transfer for Targeted B-1a Cell Migration to the Bone Marrow and IgM Production

Summary

Here we describe a method for retroviral overexpression and adoptive transfer of murine B-1a cells to examine in vivo B-1a cell migration and localization. This protocol can be extended for diverse downstream functional assays including quantification of donor B-1a cell localization or analysis of donor cell-derived secreted factors post-adoptive transfer.

Abstract

As cell function is influenced by niche-specific factors in the cellular microenvironment, methods to dissect cell localization and migration can provide further insight on cell function. B-1a cells are a unique B cell subset in mice that produce protective natural IgM antibodies against oxidation-specific epitopes that arise during health and disease. B-1a cell IgM production differs depending on B-1a cell location, and therefore it becomes useful from a therapeutic standpoint to target B-1a localization to niches supportive of high antibody production. Here we describe a method to target B-1a cell migration to the bone marrow by retroviral-mediated overexpression of the C-X-C motif chemokine receptor 4 (CXCR4). Gene induction in primary murine B cells can be challenging and typically yields low transfection efficiencies of 10-20% depending on technique. Here we demonstrate that retroviral transduction of primary murine B-1a cells results in 30-40% transduction efficiency. This method utilizes adoptive cell transfer of transduced B-1a cells into B cell-deficient recipient mice so that donor B-1a cell migration and localization can be visualized. This protocol can be modified for other retroviral constructs and can be used in diverse functional assays post-adoptive transfer, including analysis of donor cell or host cell phenotype and function, or analysis of soluble factors secreted post B-1a cell transfer. The use of distinct donor and recipient mice differentiated by CD45.1 and CD45.2 allotype and the presence of a GFP reporter within the retroviral plasmid could also enable detection of donor cells in other, immune-sufficient mouse models containing endogenous B cell populations.

Introduction

Recent studies have demonstrated considerable immune cell, and specifically B cell, phenotypic and functional heterogeneity depending on cell localization^1,2,3,4,5. B-1a cells are one such population with heterogeneous capacity to produce protective IgM antibodies; bone marrow B-1a cells secrete IgM constitutively and contribute significantly to plasma IgM titers⁶, while peritoneal B-1a cells have low-level IgM secretion at homeostasis and instead can be activated through innate toll-like receptor (TLR) or cytokine-mediated signaling to rapidly proliferate, migrate, and secrete IgM^7,8,9,10. B-1a cell IgM antibodies recognize oxidation-specific epitopes (OSE) that are present on pathogens, apoptotic cells, and oxidized LDL, and IgM binding to OSE can prevent inflammatory downstream signaling in diseases like atherosclerosis¹¹. Therefore, strategies to increase IgM production via increasing peritoneal B-1a cell migration to sites like the bone marrow may be therapeutically useful. However, it is important for such strategies to be targeted and cell-type specific, as off-target effects may negatively impact immune function or health.

Here we describe a method for targeted and long-term overexpression of CXCR4 in primary murine B-1a cells and subsequent adoptive transfer to visualize cell migration and functional IgM antibody production (Figure 1). Genetic manipulation of primary B cells is limited by low transfection efficiencies compared to transfection of transformed cell lines. However, as transformed cell lines can significantly deviate from primary cells^12,13, the use of primary cells is likely to provide results that more closely align to normal physiology. Several techniques have been described for gene transfer in primary murine B cells, including retroviral transduction, adenoviral transduction, lipofection, or electroporation-based transfection, which have varying levels of efficiency, transience, and impact on cell health^13,14,15. The following method utilized retroviral transduction as it yielded adequate gene transfer efficiency of >30% while minimally impacting cell viability. The CXCR4-expressing retrovirus was generated using the previously described retroviral construct murine stem cell virus-internal ribosomal entry site-green fluorescent protein (MSCV-IRES-GFP; MigR1)¹⁶, into which the mouse CXCR4 gene was sub-cloned⁴. MigR1 (control(Ctl)-GFP) and CXCR4-GFP retroviral particles were generated using calcium phosphate transfection as described in previously published protocols^4,14.

Successfully transduced B-1a cells were then intravenously transferred into lymphocyte-deficient Rag1^-/- mice. Both donor and recipient mice additionally contained knockout of the apolipoprotein E (ApoE) gene, which results in increased OSE accumulation and atherosclerosis, thereby providing a model for in vivo B-1 cell activation and IgM production. Moreover, donor and recipient mice differed in CD45 allotype; donor B-1 cells came from CD45.1+ ApoE^-/- mice and were transferred into Rag1^-/- CD45.2+ ApoE^-/- recipients. This allowed differentiation of donor CD45.1 from recipient CD45.2 B cells post-transfer without the need to additionally stain for B cell markers during flow cytometry analysis. The results provided here demonstrate that targeted CXCR4 overexpression on B-1a cells associates with increased ability of B-1a cells to migrate to the bone marrow, which associates with increased plasma anti-OSE IgM. We additionally provide a method for the enrichment of peritoneal B-1 cells through negative selection and demonstrate the requirement of B-1 cell activation for efficient transduction. This method can be adapted for other retroviral constructs to study the effect of protein overexpression on B-1a cell migration, phenotype, or function. Moreover, the use of CD45.1 versus CD45.2 allotype distinction could theoretically allow transfer into other immune-sufficient murine models containing endogenous B cells.

Protocol

All animal protocols were approved by the Animal Care and Use Committee at the University of Virginia.

1. Magnetic separation and enrichment of peritoneal B-1 cells

1. Euthanize a 12−14-week-old, male, CD45.1⁺ApoE^-/- mouse using CO₂.
2. Make a superficial cut in the abdomen using straight surgical scissors and peel back skin using curved scissors to expose the peritoneal wall. Flush peritoneal cavity with 10 mL of 37 °C RPMI-1640 medium using a 10 mL syringe and 25 G needle. Shake mouse to disengage cells by grasping the tail and moving the mouse side to side thoroughly for 15−20 s.
   NOTE: Massaging the lavaged peritoneum can also maximize cell recovery.
3. Collect peritoneal fluid using a 10 mL syringe and 25 G needle by drawing up fluid at the lower right side of the peritoneum just above the level of the hip, near the intestines. Avoid disrupting epidydimal fat depots and underlying organs. Avoid drawing fluid from the mouse’s left side as the omental fat can easily be drawn into the syringe.
4. Once ~6−7 mL of fluid is collected, dispense into a 50 mL conical tube placed on ice. Next, elevate the mouse vertically by holding the peritoneal wall above the diaphragm using forceps so that any remaining fluid remains at the bottom of the peritoneal cavity. Make a small cut in the peritoneal wall using surgical scissors above the liver (make sure not to cut the liver) and collect any remaining peritoneal fluid using a glass pipet and bulb.
5. Pool peritoneal washout cells from all CD45.1⁺ApoE^-/- mice (n = 15−20) into 50 mL conical tubes, and store on ice.
   NOTE: For calculating the number of mice needed: B-1a cells comprise 5−10% of the total peritoneal population and yield roughly 2.5−5 x 10⁵ B-1a cells per mouse in authors’ hands. Transduction efficiency is ~30−40%, as shown below.
6. Count live cells using a viability dye such as trypan blue and a hemocytometer (dilute samples 1:5 in trypan blue and load 10 µL into the hemocytometer chamber). Pool cells at up to 1 x 10⁸ cells per tube. Centrifuge cells at 400 x g for 5 min at 4 °C, then aspirate supernatant.
7. Resuspend up to 1 x 10⁸ cells in 1 mL of anti-CD16/CD32 antibody (Table of Materials) diluted 1:50 in assay buffer (1x phosphate buffered saline [PBS], 0.5% bovine serum albumin [BSA], 2 mM EDTA) in order to block Fc receptors. Scale as necessary based on cell count. Incubate on ice for 10 min at 4 °C.
8. Prepare a 2x master mix of the biotinylated antibodies (Table 1) in assay buffer for depletion. The 2x master mix accounts for the volume of the liquid the cells are already in when incubating with anti-CD16/CD32. For example, if cells are incubating in 500 µL of diluted anti-CD16/CD32, then add 500 µL of 2x master mix containing 10 µL of biotinylated Ter119, Gr-1, CD23, and NK1.1 antibodies, and 25 µL of biotinylated F4/80 antibody to achieve the final concentrations given in Table 1. Add 2x master mix to cells and stain for 20 min at 4 °C.
9. Wash cells with 5 mL of assay buffer and centrifuge at 400 x g for 5 min at 4 °C and aspirate supernatant.
10. Resuspend and incubate cells with anti-biotin microbeads (Table of Materials) diluted in assay buffer as per the concentration and protocol recommended by the manufacturer.
11. Wash with 5 mL of assay buffer and centrifuge at 400 x g for 5 min at 4 °C. Aspirate supernatant and resuspend up to 1 x 10⁸ cells in 500 µL of assay buffer.
12. Prime magnetic selection columns (Table of Materials) with 3 mL of assay buffer. Transfer cells onto primed magnetic selection columns and collect the eluent containing enriched B-1 cells in a 15 mL conical tube on ice. Wash the magnetic selection column with additional assay buffer till the overall volume collected is 10 mL.
    NOTE: An aliquot of the pre-purified and post-purified cell fractions should be separately analyzed by flow cytometry for purification efficiency by staining for CD19+ B cells, F480+ macrophages, and CD5+ T cells as shown in Figure 2.
13. Count the post-purified cell fraction (the eluent containing enriched B-1 cells) by counting live cells as in step 1.6, and resuspend at 1 x 10⁶ cells/mL in B cell culture medium (RPMI-1640, 10% heat-inactivated fetal bovine serum [FBS], 10 mM HEPES, 1x non-essential amino acids, 1 mM sodium pyruvate, 50 µg/mL gentamicin, 55 µM β-mercaptoethanol).

2. Peritoneal B-1 cell stimulation

1. Split pooled cells for the two transduced conditions (Ctl-GFP and CXCR4-GFP), while setting aside and plating at least 10 x 10⁶ cells for a non-transduced control.
2. Plate up to 150 µL (150,000 cells) per well into 96-well round-bottom plates.
3. Add 100 nM TLR9 agonist ODN1668 to all wells to stimulate cell proliferation.
4. Incubate for 16−18 h at 37 °C, 5% CO₂.

3. Retroviral transduction of peritoneal B cells

1. Generate Ctl-GFP (MigR1) and CXCR4-GFP retroviral particles using calcium phosphate transfection as described in previously published protocols¹⁴.
   NOTE: This will take several days, so prepare and titer retroviral stocks and store aliquots at -80 °C prior to starting this protocol.
2. Thaw retrovirus stocks on ice. Use immediately and do not re-freeze as virus titer significantly diminishes during freeze-thaw cycles. Add Ctl-GFP or CXCR4-GFP retroviral supernatants at 20:1 multiplicity of infection (MOI) to cells, in the presence of 8 µg/mL polybrene and fresh β-mercaptoethanol at 55 µM final concentration. Do not add viral stocks to the cells set aside for the non-transduced control.
3. Perform spinfection by centrifuging plates at 800 x g for 90 min at room temperature.
4. Incubate plates with retrovirus at 37 °C, 5% CO₂ for an additional 3 h.
5. Harvest and replate cells in fresh B cell medium and incubate at 37 °C, 5% CO₂ overnight.

4. Cell sorting of transduced peritoneal B-1a cells

1. Harvest cultured cells and pool into three separate 50 mL conical tubes for each condition: non-transduced, Ctl-GFP transduced, and CXCR4-GFP transduced.
2. Count live cells as in step 1.6, then centrifuge at 400 x g for 5 min at 4 °C and aspirate supernatant.
3. Resuspend cells at 100,000 cells/µL in sort buffer (PBS + 1% BSA) containing 1:50 anti-CD16/CD32 antibody (Table of Materials) in order to block Fc receptors. Incubate on ice for 10 min at 4 °C.
4. Aliquot cells for compensation controls (~30,000 cells per compensation control): for unstained and single stain controls aliquot from non-transduced sample and for GFP single stain control aliquot from transduced samples.
   NOTE: Commercially available compensation beads can alternatively be used for single stain controls if non-transduced cell number is low.
5. Prepare a 2x antibody master mix containing the fluorophore-conjugated antibodies in sort buffer (Table 1). Add 2x master mix to non-transduced, CTL-GFP transduced, and CXCR4-GFP transduced samples. Add individual antibodies to single stain controls. Incubate for 20 min at 4 °C in the dark.
   NOTE: The 2x master mix accounts for the volume of liquid the cells are already in when incubating with anti-CD16/CD32, as in step 1.8. An aliquot of cells from each condition can separately be stained for CXCR4 to confirm CXCR4 overexpression.
6. Wash samples with 1 mL of sort buffer and strain through 70 µm filters into polypropylene tubes.
7. Centrifuge at 400 x g for 5 min at 4 °C and aspirate supernatant. Resuspend samples at 50,000 cells/µL in sort buffer.
8. Prior to cell sorting prepare labeled fluorescence-activated cell sorting (FACS) collection tubes containing 1 mL of collection medium (RPMI-1640, 20% heat-inactivated FBS, 10 mM HEPES, 1x non-essential amino acids, 1 mM sodium pyruvate, 50 µg/mL gentamicin, 55 µM β-mercaptoethanol) for each population to be sorted.
9. Prior to running samples on the cell sorter add 2x DAPI (prepared as 1:5000 dilution in sort buffer) for dead cell discrimination.
10. Sort GFP+ B-1a cells into FACS tubes containing collection medium as DAPI^- CD19⁺ GFP⁺ B220^mid-lo CD23^- IgM⁺ CD5⁺ cells from the CTL-GFP and CXCR4-GFP samples. Use the non-transduced sample to set the GFP+ gate and to sort DAPI^- CD19⁺ GFP^- B220^mid-lo CD23^- IgM⁺ CD5⁺ non-transduced B-1a cells.
    NOTE: Alternatively, non-transduced cells can also be sorted from the CTL-GFP and CXCR4-GFP samples by separately gating B-1a cells within the GFP- fraction. Transduced or non-transduced B-1b cells can also be sorted using this sort strategy as DAPI^- CD19⁺ GFP^+/- B220^mid-lo CD23^- IgM⁺ CD5^- cells.

5. Adoptive transfer

1. After cell sorting, centrifuge cells at 400 x g for 5 min at 4 °C and aspirate supernatant carefully.
2. Resuspend cells in cold sterile 1x PBS at 1,000 cells/µL.
3. Anesthetize male Rag1^-/- ApoE^-/- mice using isoflurane and inject 100 µL (100,000 cells) per mouse via intravenous retro-orbital or tail vein injection using an ultra-fine insulin syringe. Inject a few mice with 1x PBS as a control.

6. Quantification of donor cells and plasma IgM

1. At the desired time post-adoptive transfer, analyze transferred cells in recipient mice by bone marrow and spleen tissue harvest⁴ and flow cytometry. Quantify donor cell localization and CXCR4 overexpression by staining for CD45.1, CD45.2, and CXCR4 and analyze flow cytometry results using a software such as Flowjo.
   NOTE: See Table 1 for antibodies used in this step. Ensure not to use an antibody conjugate that fluoresces in the FITC channel as GFP will be present on transduced donor cells.
2. Isolate plasma by adding 10 µL of 0.5 M EDTA to whole blood collected via cardiac puncture at the time of animal sacrifice. Centrifuge whole blood at 7,000 x g and aliquot plasma into separate 1.5 mL centrifuge tubes. Store at -80 °C until performing analysis of secreted factors such as IgM by ELISA⁴.

Results

An overview of the protocol is given in Figure 1. Figure 2 displays enrichment of peritoneal B-1a cells after magnetic depletion of other peritoneal cell types. Live singlet cells in the post-depletion fraction have a greater proportion of CD19+ B cells compared to F4/80⁺ macrophages, lack CD5^hi CD19^- T cells, and contain an increased frequency of CD19⁺ CD5^mid B-1a cells compared to the pre-depletion fractio...

Discussion

The method provided here enables stable and relatively efficient primary B-1a cell gene delivery, in vivo adoptive transfer, and identification and localization of injected cells. Cells were able to be detected 17 weeks post-cell transfer and retained increased CXCR4 expression. Retrovirus-mediated delivery yielded 30-40% transduction efficiency of primary murine B-1a cells with minimal impact on cell viability in our hands (Figure 4e). This is in line with results from a previous study by M...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
70 micron filter caps	Falcon	352235
anti-biotin microbeads	Miltenyi Biotec	130-090-485
anti-CD16/CD32, or Fc block	Life Technologies	MFCR00
B220 APC	eBioscience	17-0452-83	Clone: RA3-6B2
Beta-mercaptoethanol	Gibco	21985-023
CD19 APCef780	eBioscience	47-0193-82	Clone: eBio1D3
CD23 biotin	eBioscience	13-0232-81	Clone: B3B4
CD23 PECy7	eBioscience	25-0232-82	Clone: B3B4
CD3e biotin	eBioscience	13-0033-85	Clone: eBio500A2
CD45.1 ApoE-/- mice	N/A	N/A	Bred in house
CD45.1 PerCP-Cy5.5	BD Biosciences	560580	Clone: A20
CD45.2 BV421	BD Biosciences	562895	Clone: 104
CD45.2 Rag1-/- ApoE-/- mice	N/A	N/A	Bred in house
CD5 PE	eBioscience	12-0051-83	Clone: 53-7.3
Ctl-GFP retrovirus	N/A	N/A	Generated in house using GFP-expressing retroviral plasmid MigR1 provided by Dr. T.P. Bender
CXCR4 APC	eBioscience	17-9991-82	Clone: 2B11
CXCR4-GFP retrovirus	N/A	N/A	Generated in house by cloning mouse CXCR4 into MigR1 retroviral plasmid
F4/80 biotin	Life Technologies	MF48015	Clone: BM8
Flowjo Software v. 9.9.6	Treestar Inc.	License required
Gentamicin	Gibco	15710-064
Gr-1 biotin	eBioscience	13-5931-82	Clone: RB6-8C5
heat-inactivated fetal bovine serum	Gibco	16000-044
HEPES	Gibco	15630-080
IgM PECF594	BD Biosciences	562565	Clone: R6-60.2
Insulin syringes	BD Biosciences	329461
Isoflurane	Henry Schein Animal Health	029405
Live/Dead Yellow	Life Technologies	L34968
LS columns	Miltenyi Biotec	130-042-401
NK1.1 biotin	BD Biosciences	553163	Clone: PK136
Non-essential amino acids	Gibco	11140-050
ODN 1668	InvivoGen	tlrl-1668
PBS	Gibco	14190-144
RPMI-1640	Gibco	11875-093
Sodium pyruvate	Gibco	11360-070
Ter119 biotin	eBioscience	13-5921-82	Clone: Ter119