The goal of this protocol is to isolate nonhuman primate CD34+ cells from primed bone marrow, to gene-modify these cells with lentiviral vectors, and to prepare a product for infusion into the autologous host. The total protocol length is approximately 48 h.
Hematopoietic stem and progenitor cell (HSPC) transplantation has been a cornerstone therapy for leukemia and other cancers for nearly half a century, underlies the only known cure of human immunodeficiency virus (HIV-1) infection, and shows immense promise in the treatment of genetic diseases such as beta thalassemia. Our group has developed a protocol to model HSPC gene therapy in nonhuman primates (NHPs), allowing scientists to optimize many of the same reagents and techniques that are applied in the clinic. Here, we describe methods for purifying CD34+ HSPCs and long-term persisting hematopoietic stem cell (HSC) subsets from primed bone marrow (BM). Identical techniques can be employed for the purification of other HSPC sources (e.g., mobilized peripheral blood stem cells [PBSCs]). Outlined is a 2 day protocol in which cells are purified, cultured, modified with lentivirus (LV), and prepared for infusion back into the autologous host. Key readouts of success include the purity of the CD34+ HSPC population, the ability of purified HSPCs to form morphologically distinct colonies in semisolid media, and, most importantly, gene modification efficiency. The key advantage to HSPC gene therapy is the ability to provide a source of long-lived cells that give rise to all hematopoietic cell types. As such, these methods have been used to model therapies for cancer, genetic diseases, and infectious diseases. In each case, therapeutic efficacy is established by enhancing the function of distinct HSPC progeny, including red blood cells, T cells, B cells, and/or myeloid subsets. The methods to isolate, modify, and prepare HSPC products are directly applicable and translatable to multiple diseases in human patients.
Stem cell gene therapy is a powerful means to address a wide range of human pathologies. HSPC gene therapy is a particularly attractive approach, due to i) the relative ease of collecting these cells from patients, ii) the wealth of knowledge that is available regarding cell surface phenotypes and ex vivo culture parameters, and, as the field expands, because iii) it presents scientists with an ever-increasing toolbox of gene modification strategies tailored to various diseases of interest. We are actively investigating HSPC gene therapy approaches from multiple angles, including the basic science of HSPC biology, the engraftment of gene-modified HSPCs in preclinical in vivo models, and the application to relevant patient populations. We and others have characterized the cell surface phenotype of functionally distinct HSPC subsets1,2,3, the mobilization and conditioning regimens that maximize HSPC yield and engraftment while minimizing toxicity4,5, and the gene modification and gene-editing strategies that have been tailored to a wide range of malignant, genetic, and infectious diseases6,7,8,9,10. The function and engraftment of gene-modified HSPCs can be evaluated in a number of small- and large-animal models, including mice, dogs, and NHPs. In particular, NHP models are advantageous because many reagents, for example, antibodies specific for HSPC cell surface proteins like CD34 and CD90, can be used interchangeably in human and NHP cells. Furthermore, in contrast to mice, large animals such as NHPs allow a closer approximation of the scale of gene modification necessary for clinical efficacy. Finally, NHPs are the gold standard for the modeling of human pathologies such as HIV-1 infection11 and are an emerging model system for candidate anticancer and anti-HIV immunotherapies12,13.
The purpose of this protocol is to outline methods for purifying, genetically modifying, and preparing NHP HSPC infusion products. Although outside the scope of this protocol, we have previously shown that these products engraft in autologous NHP hosts, give rise to all hematopoietic lineages, and provide therapeutic efficacy in a broad range of disease models1. We have also characterized the clonality of engrafting HSPCs and built a platform to track the kinetics, trafficking, and phenotype of individual HSPCs and their progeny, following autologous transplantation1,14. The methods presented here have been developed with the following goals: i) to isolate highly pure HSPCs and long-term engrafting HSC subsets, ii) to maintain primitive HSCs during ex vivo culture, and iii) to efficiently gene-modify either bulk HSPCs or long-term engrafting HSC subsets. We employ magnetic-assisted cell-sorting (MACS), as well as fluorescence-activated cell sorting (FACS), to isolate phenotypically/functionally distinct HSPC populations, consistent with the methods of many groups2,15,16. The maintenance of primitive HSCs in culture (i.e., minimizing the differentiation of these cells into committed progenitors that give rise to fully differentiated lymphoid and myeloid subsets) is an essential facet of the protocol described here. Although we have previously characterized approaches to expand HSPCs while retaining a primitive phenotype17,18, here, we describe a protocol that focuses on maintaining HSCs via a minimal (48 h) and defined ex vivo culture.
The efficient modification of HSPCs and HSC subsets is a central goal of this protocol. Among several approaches we have reported, two are by far the most investigated in clinical trials: LV-mediated gene modification and nuclease-mediated gene editing1,6,19. Gene-editing strategies use one of a number of nuclease platforms to specifically modify a targeted gene of interest, for example, C-C chemokine receptor type 5 (CCR5) for the treatment of HIV infection7,19 or Bcl11A for the treatment of hemoglobinopathies6. Here, we focus on LV-mediated gene modification, in which transgenic cargoes integrate semirandomly into the genome1,8,20. A key advantage of LV approaches is the ability to deliver large amounts of genetic material (up to 8 or 9 kilobases). Although gene-editing strategies are being developed to target a transgene of interest to integrate only at a specified locus by homologous donor recombination (HDR), these methods require further development in vitro and in small animal models. In contrast, LV vectors have been used extensively in NHPs and in patients21,22. Importantly, the protocol described here, which uses primed BM as a starting HSPC source, can be easily and broadly adapted, for example, to isolate PBSCs. As described above, we take advantage of the high degree of genetic similarity between NHPs and humans to use reagents that are applicable to both species. Finally, this approach has been adapted to modify other hematopoietic subsets, namely T cells12,23,24; the advent of efficacious T-cell immunotherapy approaches has relied heavily on the same LV platform utilized in this protocol. These methods are appropriate for any researcher interested in either HSPC biology or LV-mediated gene modification. For example, the HSPC purification protocol presented here could be used to characterize novel HSC-enriched subsets, as described previously1,15,25. Likewise, the LV transduction methods presented here could similarly be applied and further developed for numerous other cell types and experimental questions, both in in vitro and in vivo models.
In summary, we present methods to isolate and genetically modify NHP HSPCs. These methods can be easily adapted for other species and other sources of HSPCs. This thoroughly vetted protocol shows great promise in the modeling of efficacious therapies for numerous human diseases.
Autologous NHP transplants, priming (mobilization), the collection of cells, and gene modification are conducted consistent with previously published protocols26. All experimental procedures are reviewed and approved by the Institutional Animal Care and Use Committee of the Fred Hutchinson Cancer Research Center and the University of Washington (Protocol #3235-01). All studies are carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health (“The Guide”); animals were randomly assigned to the studies.
1. Enrichment of CD34+ HSPCs and Overnight Culture (Day -1)
2. Quality Control of CD34-enriched Cells (Day -1)
3. Gene Modification of CD34+ HSPCs and Overnight Recovery (Day 0)
4. Cell Harvest and Preparation for Infusion (Day 1)
5. Quality Control
NOTE: Flow cytometry and cell sorting are performed after the cells are infused into the animals as part of the follow-up described in step 4.4.2. Flow cytometric data is used immediately after transplantation to determine the composition of phenotypically defined stem and progenitor cell subsets in the infusion product (steps 5.1 and 5.3), whereas the analysis of CFC assays is performed 12 – 14 days postinfusion to determine the gene-modification efficiency by colony PCR (step 5.4).
The protocol described above is designed to isolate and gene-modify NHP CD34+ HSPCs, which can subsequently be infused back into the autologous host (Figure 1 and Figure 2). When following this protocol, we usually obtain up to 8 x 109 total WBCs from primed BM from pigtail macaques and, sometimes, double that amount from rhesus macaques. In both species, the number of CD34+ HSPCs that we enrich is proportional to the input of the total WBC count (Figure 3). Previous findings demonstrate that the total CD34+ HSPC product includes cells that are not true, long-term engrafting HSCs. Hence, we have developed flow-cytometry-based techniques (Figure 4) and use CFC assays (Figures 5-6) to identify long-term HSCs and committed progenitor subsets in cultures. Unlike true HSCs, committed progenitors will persist for a relatively short time period in vivo. Finally, the LV-mediated gene modification strategy results in robust gene marking in cultured CD34+ HSPCs. These cells are monitored over up to 2 weeks in culture, in part to reduce the number of "false positive" cells that express GFP from nonintegrated LV vectors. Because these cells do not carry a stably integrated copy of the LV vector in the cellular genome, they will dilute out over the 12-day liquid culture assay (Figure 7).
Figure 1: Production of an autologous, gene-modified NHP HSPC product. The protocol isolates CD34+ HSPCs (day -1), gene-modifies these cells (day 0), and prepares an infusion product (day 1) over a 48 h time course. Colony formation, liquid culture, and related ex vivo assays continue for an additional 2 weeks, in order to characterize these products. Please click here to view a larger version of this figure.
Figure 2: Timeline of events. The average time to perform each individual step of the protocol, over approximately 2 days. The timing of individual steps varies depending on the stem cell source (step 1 of the protocol) and/or the type of gene modification (step 3 of the protocol). Please click here to view a larger version of this figure.
Figure 3: White blood cell and CD34 yield from primed pigtail and rhesus macaque bone marrow. Enriched CD34+ HSPC counts (step 1.4 of the protocol) as a function of total white blood cell counts (step 1.3.3 of the protocol) from 10 pigtail macaques (squares) and six rhesus macaques (triangles). Please click here to view a larger version of this figure.
Figure 4: Gating strategy for the quality control of CD34-enriched and gene-modified products. Total white blood cells ("pre-enrichment") and subsequently CD34-enriched HSPCs are stained with antibodies specific for CD34, CD45, CD90, and CD45RA, in order to quantify the number of HSC-, MPP-, EMP-, and LMP-enriched CD34 subsets. Please click here to view a larger version of this figure.
Figure 5: Morphology of CD34+ cells before and after gene modification. Top panels: Three representative brightfield images from HSPC colony assays. Bottom panels: GFP fluorescence corresponding to each brightfield image above. Scale bar = 1 mm. Please click here to view a larger version of this figure.
Figure 6: Colony-forming cell (CFC) potential of CD34+ cells and sort-purified subsets. Following the sort purification for CFC assays from HSPC subsets on day -1 (sections 1 and 2 of the protocol) and day 1 (section 5 of the protocol), single colonies are scored based on morphological characteristics. CFU-MIX = mix of myeloid (white) and erythroid (red) cells; BFU-E = only erythroid cells; CFU-G = granulocytes; CFU-GM = granulocytes and macrophages/monocytes; CFU-M = macrophages/monocytes. Please click here to view a larger version of this figure.
Figure 7: Representative flow cytometric data from gene-modified cells in liquid culture. Percentage of CD34+ HSPCs following transduction with a GFP-expressing LV vector. Cells are cultured for up to 2 weeks following transduction. A decrease in GFP+ events over time reflects a loss of GFP signal from cells carrying nonintegrated LV vectors. Please click here to view a larger version of this figure.
Name | Contents |
Hemolytic Buffer | 150 mM Ammonium Chloride, 12 mM Sodium Bicarbonate, 0.1 mM EDTA in double-distilled water (ddH2O) |
Commercial Buffer | Phosphate-buffered saline (1X) pH 7.2, 0.5% BSA, 2 mM EDTA |
FACS buffer | Phosphate-buffered saline (1X) pH 7.2, 2% Fetal Bovine Serum |
HBSS + 2% BSA | Hank's Balanced Salt Solution, 2% Bovine Serum Albumin |
HSPC Media | StemSpan SFEM II, 1% Penicillin/Streptomycin, 100 ng/mL each recombinant human TPO, SCF, FLT-3 |
Transduction Media | StemSpan SFEM II, 1% Penicillin/Streptomycin, 100 ng/mL each recombinant human TPO, SCF, FLT-3, 1 µg/mL Cyclosporine, 4 ug/mL Protamine Sulfate |
Table 1: Buffer and media formulations.
ID | PE | PECF594 | APC-Cy7 | V450 | Description |
1 | CD90 | Compensation Beads | |||
2 | CD34 | Compensation Beads | |||
3 | CD45RA | Compensation Beads | |||
4 | CD45 | Compensation Beads | |||
5 | Unstained WBCs before CD34-enrichment | ||||
6 | CD90 | CD34 | CD45RA | CD45 | Stained WBCs before CD34-enrichment |
7 | Unstained WBCs of flow-through (FT) | ||||
8 | CD90 | CD34 | CD45RA | CD45 | Stained WBCs of flow-through (FT) |
9 | Unstained WBCs of CD34-enriched product | ||||
10 | CD90 | CD34 | CD45RA | CD45 | Stained WBCs of CD34-enriched product |
Table 2: Representative flow panel for the quality control of CD34-enriched and gene-modified cell products.
Antigen | Clone | Fluorochrome | Laser | Filter |
CD45 | D058-1284 | V450 | 395 nm | 450/40 nm |
CD90 | 5.00E+10 | PE | 488 nm or 532 nm | 585/42 nm |
CD34 | 563 | PE-CF594 | 488 nm or 532 nm | 610/20 nm |
CD45RA | 5H9 | APC-Cy7 | 633 nm | 780/60 nm |
V450: violet 450 nm; PE: Phycoerythrin; PE-CF594: trademark name from Biotium; APC: Allophycocyanin-cyanine 7 |
Table 3: Antibody-staining panel for quality control by flow cytometry and cell sorting.
LV vector engineering is the best-characterized method to gene-modify cell types such as CD34+ HSPCs, for subsequent transplantation in vivo. The protocol described here is designed to maximize the number of gene-modified HSPCs that persist long-term in vivo, and provide clinical benefits to patients with various malignant, infectious, and genetic diseases. Although gene-editing strategies have emerged over the last decade, LV-modified cells are the best studied in vitro, in animal models, and in patients1,8,20,21,22.
Based on our extensive experience with this protocol, the enrichment of pure CD34+ HSPCs (i.e., >80% of the CD34+ cells in the sorted cell product) is a critical aspect of success. Because these cells are derived from a mixed population of total BM WBCs, low-purity cultures may include cells that will not engraft long-term, in turn lowering the dose of true stem cells that are infused into the autologous host. Additionally, high-quality LV VCM will ensure the highest efficiency of gene modification.
To address shortcomings in the purity of enriched CD34+ HSPC products, it is often useful to validate the quality of the reagents used to isolate these cells, namely the anti-CD34 antibody (which our group purifies in-house from a hybridoma cell line), and magnetic beads which bind antibody-labeled cells. We recommend an MOI of 10, with an option to repeat this transduction (MOI 10 x 2). Importantly, the MOI should be determined empirically, following a quality assessment of VCM, including the titration of each vector on a standardized titering cell line such as HT108029. Importantly, different VCM-producing laboratories may use distinct titering cell lines, including HOS30, HeLA31, and 293T32, which may limit the ability to compare titers between facilities. We prefer to retiter a vector with the assay, in order to calculate the most accurate amount of VCM to efficiently gene-modify CD34+ HSPC target cells. Once high-quality CD34+ HSPCs and VCM have been obtained, transduction efficiency and engraftment are further enhanced in three distinct steps. First, the addition of cyclosporine to transduction media aids in the early steps of vector transduction and integration into target cells35,36, while protamine sulfate decreases repulsive forces between the lentiviral vector particles and the cell surface33,37. Second, treating plates with a recombinant fibronectin fragment (e.g., RetroNectin, CH-296) increases transduction efficiency and also improves the in vivo engraftment potential of gene-modified HSPCs 33,38,39. Notably, previous studies suggest that CH-296 is only important during, but not prior to, transduction33,38. Finally, we pulse gene-modified cell products with PGE2 to enhance engraftment and persistence in vivo, as has previously been shown for human and nonhuman primate CD34+ HSPC products40,41.
LVs integrates randomly into the genome, as do their predecessor, gammaretroviral vectors (RVs). Although less clinical trial data is available for LVs than for RVs, current findings suggest that LV approaches carry substantially lower risks of cell transformations due to LV's insertional mutagenesis; RV strategies are less often used due to this risk42. A further limitation is that the efficiency of gene modification is less than 100%, and a proportion of gene-modified cells will not persist in vivo. For example, a 60% gene-modified HSPC product may result in 30% long-term engrafting, gene-modified cells. Whenever possible, the gene therapy strategies presented here are designed to overcome this limitation by introducing transgenes that provide therapeutic efficacy, even when expressed in a minority of hematopoietic-origin cells8,43.
The future is bright for LV-based HSPC modification approaches. We routinely use this protocol to "gene-mark" cells, enabling tracking in vivo1. We have also adapted this strategy to compare LV variants within the same animal. Such "competitive" transplants allow a comparison of different vectors (e.g., to identify those with better efficiency) and transgenes (e.g., to track different HSPC subsets or compare them in disease models). Moving forward, we believe that LV gene therapy in HSPCs will remain an essential tool alongside gene-editing approaches, especially in situations where large genetic cargoes must be stably expressed for the lifetime of an individual.
The authors have nothing to disclose.
The authors thank Helen Crawford for preparing this manuscript, Jim Woolace for graphic design, and Veronica Nelson and Devikha Chandrasekaran for participating in the development of the protocol. The development of this protocol was supported by grants from the NIH National Institute of Allergy and Infectious Diseases (R01 AI135953 and AI138329 to H.P.K.) and the National Heart, Lung, and Blood Institute (R01 HL136135, HL116217, P01 HL122173, and U19 HL129902 to H.P.K.), as well as NIH P51 OD010425 and, in part through the NIH/NCI Cancer Center, Support Grant P30 CA015704. H.P.K. is a Markey Molecular Medicine Investigator and received support as the inaugural recipient of the José Carreras/E. Donnall Thomas Endowed Chair for Cancer Research and the Fred Hutch Endowed Chair for Cell and Gene Therapy.
Stemspam SFEM II ("HSPC") Media | StemCell | 09655 | |
Hank's Balanced Salt Solution | Gibco | 14175095 | |
Phosphate-Buffered Saline | Gibco | 14190-144 | |
Penicillin/Streptomycin | Gibco | 15140-122 | |
Dimethyl Sulfoxide | Sigma Aldrich | D2650-100 | |
100% Ethanol | Decon labs | M18027161M | |
Cyclosporine | Sigma | 30024-25MG | |
500 mM EDTA | Invitrogen | 15575-038 | |
Heat-Inactivated Fetal Bovine Serum | Sigma Aldrich | PS-0500-A | |
CH-296/ RetroNectin (2.5 mL, 1 µg/µL) | TaKaRA | T100B | |
Bovine Serum Albumin | Sigma | A7906-100g | |
HEPES | Sigma | H9897 | |
Rat anti-mouse IgM magnetic beads | Miltenyi Biotec | 130-047-301 | |
Recombinant HumanStem Cell Factor (SCF) | Peprotech | 300-07 | |
Recombinant Human Thrombopoietin (TPO) | Peprotech | 300-18 | |
Recombinant Human FMS-like tyrosine kinase 3 (FLT-3) | Peprotech | 300-19 | |
Protamine sulfate | Sigma | P-4505 | |
14 mL Polypropylene Round-Bottom Tube | Corning | 352059 | |
Colony Gel 1402 | ReachBio | 1402 | |
QuadroMACS Separators | Miltenyi Biotec | 130-090-976 | |
MACS L25 Columns | Miltenyi biotec | 130-042-401 | |
10 mM PGE2 | Cayman Chemical | 14753-5mg | |
TC-treated T-75 flasks | Bioexpress | T-3001-2 | |
Non-TC-treated T-75 flasks | Thermo-Fisher | 13680-57 | |
20 ml syringes | BD Biosciences | 302830 | |
16.5 G needles | BD Precision | 305198 | |
Syringe Tip Cap | BD Biosciences | 305819 | |
QuickExtract DNA Solution | Epicentre | QE09050 | |
8-tube strip cap PCR Tubes | USA scientific | 1402-2708 | |
96-well Thermocycler | Thermo-Fisher | 4375786 | |
Pre-Separation filters | Miltenyi Biotec | 130-041-407 | |
Strainer, Cell; BD Falcon; Sterile; Nylon mesh; Mesh size: 70um; Color: white; 50/CS | fisher scientific | 352350 | |
Ultracomp ebeads | eBioscience | 01-2222-42 | |
MACSmix Tube Rotator | Miltenyi | 130-090-753 | |
3 mL Luer-Lock Syringes | Thermo-Fisher | 14823435 | |
35 mm x 10 mm cell culture dish | Corning | 430165 | |
60 mm x 15 mm cell culture dish | Corning | 430196 | |
150 mm x 25 mm cell culture dish | Corning | 430599 | |
Non TC treated flasks | Falcon | 353133 | |
Qiagen DNA extraction | Qiagen | 51104 | |
PE Anti-Human CD90 (Thy1) Clone:5E10 | Biolegend | 328110 | |
PE-CF594 Mouse Anti-Human CD34 Clone:563 | BD horizon | 562449 | |
APC-H7 Mouse Anti-Human CD45RA Clone: 5H9 | BD Pharmingen | 561212 | |
V450 Mouse Anti-NHP CD45 Clone:d058-1283 | BD Biosciences | 561291 | |
Autologous Serum | Collected from autologous host and cryopreserved prior to mobilization and collection of CD34+ HSPCs | N/A | Beard, B. C. et al. Efficient and stable MGMT-mediated selection of long-term repopulating stem cells in nonhuman primates. Journal of Clinical Investigation. 120 (7), 2345-2354, (2010). |
Virus-Conditioned Media (VCM) | Kiem Lab, FHCRC Co-operative Center for Excellence in Hematology (CCEH) | N/A | Beard, B. C. et al. Efficient and stable MGMT-mediated selection of long-term repopulating stem cells in nonhuman primates. Journal of Clinical Investigation. 120 (7), 2345-2354, (2010). |
Anti-CD34 antibody, Clone 12.8 | Kiem Lab | N/A | Beard, B. C. et al. Efficient and stable MGMT-mediated selection of long-term repopulating stem cells in nonhuman primates. Journal of Clinical Investigation. 120 (7), 2345-2354, (2010). |
Lenti F primer: AGAGATGGGTGCGAGAGCGTCA | Integrated DNA Technologies | N/A | Peterson, C. W. et al. Multilineage polyclonal engraftment of Cal-1 gene-modified cells and in vivo selection after SHIV infection in a nonhuman primate model of AIDS. Mol Ther Methods Clin Dev. 3 16007, (2016). |
Lenti R primer: TGCCTTGGTGGGTGCTACTCCTAA | Integrated DNA Technologies | N/A | Peterson, C. W. et al. Multilineage polyclonal engraftment of Cal-1 gene-modified cells and in vivo selection after SHIV infection in a nonhuman primate model of AIDS. Mol Ther Methods Clin Dev. 3 16007, (2016). |
Actin F primer: TCCTGTGGCACTCACGAAACT | Integrated DNA Technologies | N/A | Peterson, C. W. et al. Multilineage polyclonal engraftment of Cal-1 gene-modified cells and in vivo selection after SHIV infection in a nonhuman primate model of AIDS. Mol Ther Methods Clin Dev. 3 16007, (2016). |
Actin R primer: GAAGCATTTGCGGTGGACGAT | Integrated DNA Technologies | N/A | Peterson, C. W. et al. Multilineage polyclonal engraftment of Cal-1 gene-modified cells and in vivo selection after SHIV infection in a nonhuman primate model of AIDS. Mol Ther Methods Clin Dev. 3 16007, (2016). |