Terumo Spectra Optia leukapheresis of cynomolgus macaques for hematopoietic stem cell and T cell collection

Abstract

Macaques are physiologically relevant animal models of human immunology and infectious disease that have provided key insights and advanced clinical treatment in transplantation, vaccinology, and HIV/AIDS. However, the small size of macaques is a stumbling block for studies requiring large numbers of cells, such as hematopoietic stem cells (HSCs) for transplantation, antigen-specific lymphocytes for in-depth immunological analysis, and latently-infected CD4+ T-cells for HIV cure studies. Here, we provide a detailed protocol for collection of large numbers of HSCs and T-cells from cynomolgus macaques as small as 3 kilograms using the Terumo Spectra Optia apheresis system, yielding an average of 5.0 x 10⁹ total nucleated cells from mobilized animals and 1.2 x 10⁹ total nucleated cells from non-mobilized animals per procedure. This report provides sufficient detail to adapt this apheresis technique at other institutions, which will facilitate more efficient and detailed analysis of HSCs and their progeny blood cells.

INTRODUCTION

Macaque models of human disease continue to yield major biological discoveries and serve as the gold-standard in pre-clinical testing of novel therapeutics and vaccines. Among these, macaques serve as physiologically relevant pre-clinical models of hematopoietic stem cell transplant (HSCT) and as the primary animal models for HIV/AIDS research.^1,2 However, the small size of macaques is a significant hurdle for studies requiring collection of large leukocyte numbers and/or rare leukocyte populations, including HSCT graft collection, in-depth analysis of antigen-specific lymphocytes, and interrogation of HIV reservoirs in CD4+ T cells. Here, we utilized apheresis to collect sufficient numbers of HSCs from mobilized cynomolgus macaque donors for allogeneic HSC transplantation (HSCT),^3,4 as well as large numbers of T cells from non-mobilized cynomolgus macaques for use in T cell assays. Although this procedure is described in detail for the collection of HSCs and T cells from cynomolgus macaques, it can be adapted to isolate any circulating blood cell of interest from both cynomolgus and rhesus macaques.

MATERIALS & METHODS

Animals

A total of 35 Mauritian-origin cynomolgus macaques (Macaca fascicularis, 20 males and 15 females) between 3 and 13 years of age and weighing between 3 and 10 kilograms (kg) underwent apheresis for this study, 1-4 apheresis procedures per macaque. All animals were housed at the Oregon National Primate Research Center. The ONPRC houses over 4,500 nonhuman primates (NHPs), primarily rhesus macaques, and over 2,500 other animals, primarily rats and mice. Animal facilities for NHPs and animal care programs are accredited by the AAALAC. Expert health care and husbandry are provided by the veterinarians and veterinary technicians in the ONPRC’s Division of Comparative Medicine. Animals were maintained in the ONPRC Animal Services Building, where the animals were pair-housed whenever possible in USDA-approved caging. Health checks were performed twice a day (during feeding times) by the animal care staff. Animals also had access to an extensive environmental enrichment program (play structures, toys, treats) designed by the ONPRC Psychological Wellbeing program. Animal care and the surgical procedures were performed by veterinary staff of the Division of Comparative Medicine at the ONPRC. All animal procedures were conducted at this facility in accordance with the Public Health Services Policy on Humane Care and Use of Laboratory Animals and approved by the IACUC. A physical examination and medical records review were completed by a veterinarian for each macaque prior to the study. All macaques were deemed healthy, with no evidence of preexisting disease. Serum chemistries and complete blood counts were taken throughout mobilization and leukapheresis procedures. Standard reference intervals for these analytes (displayed as gray boxes on graphs) were calculated from a reference sample group including 96 clinically healthy cynomolgus macaques ages 6 to 16 years. During leukapheresis procedures, i-STAT CG8+ readings were taken approximately every 30 minutes, providing Hct, Hgb, iCa, Glu, Na, K, pH, PCO2, HCO3, TCO2, Base excess, PO2 and sO2 readings.

Blood typing

Macaques were blood typed by PCR (Zoologix, Chatsworth, CA; GP100 assay). In rare cases, PCR yielded inconclusive results, and thus these macaques were typed by agglutination assays instead. For agglutination assays, RBCs from human donors were used due to the low levels of antigen expression on macaque RBCs.^5–7 Briefly, ABO-typed human blood samples (BioIVT, Westbury, NY) were washed with PBS and prepared as 50% and 5% RBC-containing suspensions. Serum from the macaques to be ABO-typed was absorbed to human type O 50% RBC suspension at 37°C for 30 minutes, then centrifuged (900 x g, 5 minutes) to re-isolate serum free of human-reactive, but not AB antigen-specific, antibodies. For each agglutination test, 50 μL of 5% human RBC suspension was combined with 50 μL pre-absorbed macaque serum and incubated at 37°C for 30 minutes prior to observing for agglutination. Each assay included controls using plasma or serum from ABO-typed human donors and PCR-typed macaques. Of note, we found evidence supporting identification of a type O cynomolgus macaque -- multiple blood and PBMC samples from this macaque resulted in inconclusive typing by PCR (no amplification), and pre-absorbed serum from this macaque agglutinated type A, B, and AB human RBCs (Supporting information file 1). For apheresis procedures on this macaque, the Optia was primed with autologous blood.

Mobilization of HSCs

Some macaques were mobilized to increase hematopoietic stem cell progenitor numbers in the peripheral blood prior to undergoing leukapheresis. Mobilized macaques received 10 or 50 μg/kg G-CSF (Amgen, Thousand Oaks, CA) for four days prior to leukapheresis and 1 mg/kg AMD3100 (Sigma-Aldrich, St. Louis, MO) two to twelve hours prior to leukapheresis. Some macaques receiving 10 μg/kg G-CSF also received a dose immediately prior to the apheresis procedure. G-CSF dosing was increased from 10 to 50 μg/kg G-CSF because a recent report demonstrated higher blood CD34+ cell counts in rhesus macaques mobilized with 50 μg/kg G-CSF dosing compared to 10 μg/kg or 100 μg/kg.⁸ Daily small blood draws were taken for complete blood count (CBC) and flow cytometry staining to determine frequencies and absolute numbers of CD34+ hematopoietic stem cell progenitors. Briefly, EDTA-treated whole blood (50-100 μL) was washed twice with PBS and stained for CD45 (PE-Cy7, clone D058-1283, BD, Franklin Lakes, NJ), CD3 (AF700, clone SP34-2, BD), CD34 (PE, clone 561, Biolegend, San Diego, CA), CD45RA (V450, clone 5H9, BD), CD90 (FITC, clone 5E10, Biolegend), and viability (Live/dead Yellow Fixable, Invitrogen, Carlsbad, CA) at room temperature for 30 minutes. After staining, whole blood was resuspended in 1mL FACS Lysing Solution (BD) to lyse red blood cells and fix remaining cells, incubated at room temperature for 8 minutes, washed three times with FACS buffer (PBS with 10% bovine growth serum), and collected on a BD LSRII instrument. Samples were analyzed with FlowJo (Tree Star, Ashland, OR). Cell subsets were identified by the following progressive gating: CD34+ HSCs = singlets, live, CD45+, CD34+; CD45RA-CD90+CD34+ HSCs = singlets, live, CD45+, CD34+, CD45RA-CD90+. Absolute counts were calculated by multiplying the subset frequency within the live, singlet, CD45+ population by the white blood cell count (WBC), as determined by CBC performed on an additional aliquot of EDTA-treated whole blood from the same blood draw.

Anesthesia and patient preparation

Staff for each apheresis procedure included one surgeon, one anesthetist, and 1 Optia operator. Each subject was sedated with an intramuscular injection of ketamine (5 to 10 mg/kg; Ketathesia, Butler Schein Animal Health, Tualatin, OR), and then anesthesia was maintained with isoflurane (Piramal Healthcare Limited, Boston, MA) 1 to 2 Vol% in 100% oxygen administered through an endotracheal tube. Subjects were placed in dorsal recumbency and 2 intravenous catheters (20 gauge intracath) were inserted: one in each cephalic vein. One intravenous catheter was used as the blood return line from the leukapheresis machine and calcium delivery (see below) and the other was used as a blood sampling port. Baseline blood samples (0.15 mL) were taken for i-STAT CG8+ analysis (Abaxis, Union City, CA) prior to beginning leukapheresis procedures. Blood samples were taken for i-STAT analysis approximately every 30 minutes throughout the leukapheresis, which typically lasted 2-3 hours.

Blood withdrawal for apheresis

While it is common to collect blood for apheresis in humans via a large venous catheter line, we found it difficult to maintain adequate machine inflow rates in NHPs, resulting in delays due to apheresis machine alarms and flow disruptions. This problem was most likely due to the small luminal diameter of the vascular catheters used in NHP subjects relative to those used in human patients. To maximize the catheter size used while reducing vascular trauma induced by catheterization, we placed femoral arterial or venous catheters (16 gauge intracath) via small surgical cut down to the femoral artery/vein at the femoral triangle. The catheter was advanced through a 2 mm longitudinal arterotomy or venotomy and then secured with silk ligatures. The decision to use artery or vein for blood withdrawal was made at the discretion of the surgeon based on vessel accommodation – typically an artery was utilized for macaques <4.0 kg. The catheter was then connected to a 3-way stopcock connected to the Optia collection (inlet) line of the apheresis machine. Upon completion of the apheresis procedure, the catheter was removed and the arterotomy/venotomy closed using 6-0 Prolene suture, thus preserving the vessel. In smaller subjects, the flow rate from venous catheters was consistently inadequate for the machine to function. In these cases, the femoral artery was catheterized to ensure high flow rates throughout the procedure.

Optia MNC collection

Leukapheresis was performed on a Spectra Optia Apheresis System using the mononuclear cell (MNC) exchange kit (Terumo BCT, Lakewood, CO) and MNC program (maximum extracorporeal volume 191 mL) according to manufacturer’s instructions, with modifications described in detail in Supporting information file 2. Acid citrate dextrose (ACD-A, Terumo BCT) was used as an anticoagulant in the apheresis circuit. Several adjustments were made in running the Optia apheresis machine. First, small cynomolgus macaques (<4.7 kg) have a total blood volume (TBV) lower than 300 mL (TBV calculated as 64 mL per kg), the minimum value allowed when entering patient data into the Optia. In these cases, “300 mL” was entered for the TBV in the patient data section of the Optia. Second, the Optia requires a minimum inlet flow rate of 10 mL/min. In order to achieve this flow rate with small macaques, the Optia’s anticoagulant (AC) infusion rate was maximized to 2.5 mL/min/L TBV. If this was insufficient to achieve 10 mL/min inlet flow rate, the inlet:AC ratio was increased incrementally until an inlet flow rate of 10mL/min was achieved. Third, to reduce the risk of clotting in the system, particularly blood between the femoral catheter (inlet line) and AC infusion point, supplemental ACD was infused into the Optia collection (inlet) line via a syringe pump. This is particularly important for small macaques when a higher Optia inlet:ACD ratio is required. An extension line from the syringe of supplemental ACD was connected to the 3-way stop-cock between the femoral catheter and Optia collection (inlet) line. Supplemental ACD was infused at 2-5 mL/hour and adjusted based on the macaque’s blood ionized calcium level (see below). Prior to adding the external ACD syringe pump at the port nearest blood collection, we observed clotting problems in about 50% of the animals. The clotting issues typically happened early on in the procedure, prior to the animals becoming systemically anticoagulated via the ACD in the apheresis return line. Once we began administering the supplemental ACD (i.e. the animals reported here), we did not observe any clotting issues. Finally, due to the small TBVs of macaques undergoing leukapheresis and to maintain sufficient hematocrit levels, supplemental ABO-matched red blood cells (RBCs) were added to the system via a custom RBC prime (see below). As a result of RBC priming, when the apheresis procedure begins and blood is pulled from the patient into the machine, ABO-matched RBCs, rather than saline, are returned to the patient.

Priming blood

Approximately 120 mL of autologous or ABO-matched heterologous cynomolgus macaque blood was collected in ACD or CPDA anticoagulant for Optia custom prime. Priming blood was leukoreduced using a BPF High Efficiency Leukocyte Reduction Filtration System (Haemonetics, Braintree, MA), which removes both leukocytes and platelets, according to manufacturer’s instructions. To assess efficiency of leukoreduction, priming blood was sampled before and after leukoreduction for CBC and flow cytometric staining. Blood was stained and analyzed for surface CD45 (APC, clone D-58-1283, BD), CD3 (AF700, clone SP34-2, BD), CD20 (APC-H7, clone 2H7, BD), and viability (Live/dead Yellow Fixable, Invitrogen) as described above.

Thermoregulation

Thermoregulation is one of the most challenging aspects of maintaining the homeostasis of NHPs undergoing leukapheresis. Hypothermia results from additive effects of anesthesia and the leukapheresis procedure. Radiant heat loss from blood traversing through the Optia circuit results in the continuous re-infusion of blood to the animal at a lower temperature. Additionally, heat loss may be associated with intravenous fluid administration, the anesthesia circuit, and cool ambient temperature. Lower metabolic rate associated with anesthesia also results in impaired thermoregulation. A multimodal approach to patient warming was necessary to prevent hypothermia in this study. We used a water-recirculating blanket placed beneath and a forced-air warming blanket positioned over NHPs undergoing leukapheresis. Additionally, infant stockings are placed on the hands and feet and an infant cap is placed on the head. Warm water bottles (500mL IV fluid bags) are also placed along each side of the body, beneath the armpits. Administered fluids are warmed using an inline IV fluid warmer, not including the Optia return line. Body temperature is continuously monitored using an esophageal temperature probe. Through these mechanisms, we were able to maintain core temperatures of subjects in the range of 36-38°C (98-101°F). To monitor for hypoglycemia, plasma glucose was checked by i-STAT analysis approximately every 30 minutes throughout the leukapheresis procedure. Hypoglycemia was not observed, and thus no supplemental glucose was administered.

Maintaining calcium homeostasis

Acid citrate dextrose (ACD) was used as an anticoagulant in the apheresis circuit. Citrate exerts its anticoagulant effect through reversible chelation of circulating divalent cations, including Ca²⁺, and sequestration of these ions from their normal physiological function. Ionized calcium (iCa²⁺) is the physiologically active form of calcium necessary for hemostasis, regulation of muscle contraction, and the stabilization of cellular membranes.⁹ Normal ionized calcium levels range from 1.1-1.4 mmol/L. ACD is mixed with the blood in the apheresis circuit. As citrate-containing blood is returned to the subject, chelation of cations continues in the systemic circulation resulting in hypocalcemia, metabolic alkalosis, and other electrolyte derangements. Ionized calcium levels that fall below 0.8 mmol/L are associated with tetany and fatal arrhythmias in humans.¹⁰ Due to the large volume of these apheresis procedures, life-threatening hypocalcemia is highly likely without continuous intravenous calcium supplementation. During apheresis, the higher the inlet:AC ratio, the less citrate is present in the circuit which may result in platelet clumping and coagulation. The lower the inlet:AC ratio, the greater the concentration of citrate delivered to the patient, which can increase symptoms of toxicity. A balance to maintain the subjects between these outcomes must be sustained throughout the procedure through constant monitoring of patient physiological parameters and apheresis machine setting. Supplemental calcium was infused intravenously to subjects throughout the leukapheresis procedures. A 4 mg/mL elemental calcium solution was made by adding 44 mL of 23% calcium borogluconate to 194 mL 0.9% NaCl. Initial infusion rate was set to 40-45 mL per hour, which was adjusted based upon blood iCa levels determined approximately every 30 minutes throughout the procedure by i-STAT CG8+. Blood iCa levels were generally maintained between 1.0 and 1.5 mmol/L. Low iCa levels were corrected by increasing the calcium infusion rate, reducing ACD infusion rate, or both.

Apheresis product and blood processing

Apheresis product and peripheral blood was processed into mononuclear cell preparations using Ficoll-Paque (GE Healthcare, Chicago, IL). Briefly, apheresis product was slowly layered onto Ficoll-paque diluted to 93.5% with PBS and centrifuged at 1860 x g for 30 minutes with the brake off. Buffy coats were removed into HBSS and washed twice more with HBSS prior to counting. To determine HSC and T cell yields from each apheresis procedure, an aliquot of apheresis mononuclear cells was stained for surface CD45 (PE-Cy7, clone D058-1283, BD), CD3 (AF700, clone SP34-2, BD), CD34 (PE, clone 561, Biolegend), CD45RA (V450, clone 5H9, BD), CD90 (FITC, clone 5E10, Biolegend), and viability (Live/dead Yellow Fixable, Invitrogen) at room temperature for 30 minutes. After staining, cells were washed once with FACS buffer, fixed with 2% paraformaldehyde for 10 minutes, and collected on a BD LSRII instrument. Samples were analyzed with FlowJo. Cell subsets were identified by the following progressive gating: CD34+ HSCs = singlets, live, CD45+, CD34+; CD45RA-CD90+CD34+ HSCs = singlets, live, CD45+, CD34+, CD45RA-CD90+; T cells = singlets, live, CD45+, CD3+. Yields were calculated by multiplying the subset frequency within the live, singlet, CD45+ population by the total number of apheresis mononuclear cells.

T cell Assays

Intracellular cytokine staining assays were performed as previously described.^11–15 Briefly, PBMC or apheresis mononuclear cell preparations were incubated with co-stimulatory molecules CD28 and CD49D (BD), with or without positive control antigen rhesus CMV (RhCMV) lysate, for 1 hour, followed by addition of Brefeldin A (Sigma-Aldrich) for an additional 8 hours. Stimulated cells were surface stained for CD4 (PE-Cy7, clone OKT-4, Biolegend) and CD8α (eTruRed, clone SK1, BD) for 30 minutes, washed, fixed with 2% PFA, permeabilized with 0.1% saponin buffer, and stained intracellularly for CD3 (PacBlu, clone SP34-2, BD), TNF-α (FITC, clone Mab11, BD), and IFN-γ (APC, clone 25723.11, BD) for 45 minutes, and washed. Data was collected on an LSR-II instrument (BD) and analyzed with FlowJo software.

Statistical analyses

Changes in cell numbers pre- and post-mobilization were evaluated by non-parametric Mann-Whitney test. Correlation between pre-apheresis HSC blood counts and apheresis HSC yield was assessed by non-parametric Spearman test. Serum and CBC analyte levels at various mobilization timepoints were compared using non-parametric Kruskalis-Wallis test followed by Dunn’s multiple comparison correction.

RESULTS

Large numbers of HSCs and T cells are collected by apheresis

A total of 76 apheresis procedures were performed on 35 Mauritian-origin cynomolgus macaques ranging in body weight from 3 to 10 kilograms (average 6 kilograms). Fifty-three procedures were performed on macaques after HSC mobilization to increase HSC numbers in the blood for HSCT graft collection. The remaining 23 non-mobilized procedures were performed on untreated macaques. Figure 1 shows a detailed schematic of the apheresis setup, described in Materials & Methods and Supporting information file 2. We made a few key adaptations in order to safely perform apheresis and collect cells of sufficient quality for HSCT and T cell assays, including performing surgical femoral cut down to maximize inlet flow rates, providing supplemental ACD via an external syringe pump, and priming the Optia apheresis machine with leukoreduced, ABO-matched heterologous blood. Table I summarizes peripheral blood CD34+ HSC counts prior to apheresis and apheresis mononuclear cell yields from apheresis procedures where data were collected. On average, a single mobilized apheresis procedure collected on the order of 10⁹ mononuclear cells, including 10⁷ CD34+ HSCs and 10⁹ T cells. Importantly, a substantial fraction of collected HSCs were CD45RA-CD90+, a critical subset of HSCs demonstrated to be responsible for rapid multilineage hematopoiesis in vivo in macaques.¹⁶ No significant correlation between animal weight and CD34+ HSC or T cell yield was observed (data not shown).

Figure 1. Schematic of cynomolgus macaque apheresis.

Illustration depicts apheresis procedure setup for macaques, described in detail in Materials & Methods and Supporting information file 1. Blood is drawn into the Optia from a femoral catheter, which is placed via surgical cut down (boxed inset). An external pump is used to provide supplemental ACD to the system via a three-way stop-cock connecting the inlet line to the femoral catheter. Blood products are returned to the macaque via a cephalic catheter, which is also used to deliver 4 mg/ml elemental calcium in 0.9% NaCl. A second cephalic catheter is placed for blood sampling to monitor the macaque via iStat Cg8+ readings. Created with BioRender.com.

Table I.

Summary of apheresis procedures performed in cynomolgus macaques.

	Total cell #		Cell # per kilogram b.w.		n	Average b.w.(kg)
MOBILIZED	Average	Range	Average	Range
CD34/μL blood pre-apheresis	124	38 - 425	22	4 - 99	43	6.4
Mononuclear cell yield	4.96 x 109	2.5 x 109 - 1.4 x 1010	8.58 x 108	3.6 x 108 - 2.3 x 109	35	5.9
CD34+ cell yield	3.83 x 107	1.1 x 107 - 1.1 x 108	6.81 x 106	1.2 x 106 - 2.5 x 107	33	6.1
CD45RA-CD90+CD34+ cell yield	8.39 x 106	9.9 x 105 - 1.5 x 107	1.55 x 106	1.4 x 105 - 6.2 x 106	29	6.0
CD3+ cell yield	1.44 x 109	5.6 x 108 - 3.7 x 109	2.43 x 108	9.2 x 107 - 5.8 x 108	33	6.1
NON-MOBILIZED
Mononuclear cell yield	1.18 x 109	3.0 x 108 - 2.1 x 109	2.28 x 108	6.2 x 107 - 5.3 x 108	23	5.8

b.w. = body weight

Mobilization with G-CSF and AMD3100 increases leukocyte and CD34+ HSC numbers in peripheral blood

In order to collect sufficient CD34+ HSCs for engraftment into cynomolgus macaques,^3,4 donor macaques underwent a mobilization protocol similar to those used in clinical apheresis of human HSCT donors. Donor macaques received G-CSF for 4-5 days leading up to apheresis and AMD3100 2-12 hours prior to the start of the apheresis procedure (Figure 2A). During the mobilization period, white blood cell counts (WBC) increased, peaking immediately prior to the apheresis procedure (Figure 2B). As expected, significantly elevated numbers of CD34+ HSCs, including CD45RA-CD90+CD34+ HSCs, were present after mobilization, immediately prior to apheresis, compared to baseline levels prior to mobilization (Figure 2C–D). In addition, the yield of CD34+ HSCTs, both bulk and the CD45RA-CD90+ subset, was directly correlated with the absolute count of that cell subset in blood prior to apheresis (Figure 3A–B).

Figure 2. Mobilization of cynomolgus macaques.

A, Timeline of mobilization procedures in cynomolgus macaques. Macaques received 4 doses of G-CSF (10 or 50 μg/kg per dose, subcutaneous) on days −4, −3, −2, and −1 relative to the apheresis procedure, and a single dose of AMD3100 (1 mg/kg, subcutaneous) 2-12 hours prior to the apheresis procedure. Some macaques receiving 10 μg/kg doses of G-CSF were also given a G-CSF dose immediately prior to the apheresis procedure (day 0, gray arrow). Macaque art from BioRender.com B, Longitudinal white blood cell counts as determined by complete blood counts on blood taken prior to any drug administration for that day. Graph shows mean±SD for n=56 apheresis procedures. Standard WBC reference interval for cynomolgus macaques is indicated by the gray box. C, Absolute counts of CD34+ cells in peripheral blood prior to mobilization (n=32) and after mobilization (n=42, blood taken immediately before the apheresis procedure). D, Absolute counts of CD45RA- CD90+ CD34+ cells in peripheral blood prior to mobilization (n=24) and after mobilization (n=31, blood taken immediately before the apheresis procedure). Graph in B-D displays p-value calculated by Mann-Whitney test: *p ≤ 0.05, ** p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. X-axes in B-D show timepoints relative to apheresis (d = day). For some procedures, there is no data available for certain timepoints.

Figure 3. HSC yields correlate with absolute counts in blood prior to apheresis procedure.

A-B, Graphs show correlations between blood absolute counts (day 0, pre-apheresis) and apheresis yield for bulk CD34+ cells (A, n=30) and CD45RA- CD90+ CD34+ cells (B, n=29). Graphs display and r and p-values calculated by Spearman tests: *p ≤ 0.05, ** p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

Mobilized apheresis is associated with temporary increases in serum creatinine and temporary decreases in hematocrit and platelet count

To assess the macaque tolerance for mobilized apheresis procedures, we monitored macaques by serum chemistry and CBC throughout mobilization and apheresis (Figure 4). Mobilized apheresis was associated with temporary increases in serum creatinine (CRE), which eventually resolved back to normal range (Figure 4A–B). As expected, temporary decreases in hematocrit (HCT) were observed after the apheresis procedure, but resolved to pre-apheresis levels by the day after apheresis (d+1) (Figure 4C). Finally, consistent decreases in platelet count (PLT) occurred after mobilized macaque apheresis, but subsequently returned to normal range (Figure 4D). No post-procedure bleeding complications were observed. Macaques successfully recovered from this apheresis protocol, and we performed multiple apheresis procedures for individual macaques, separated by as little as 3 weeks.

Figure 4. Serum chemistry and CBC analytes during mobilization and apheresis.

A, Serum blood urea nitrogren (BUN, n=45). B, Serum creatinine (CRE, n=45). C, Peripheral blood hematocrit (HCT, n=53). D, Peripheral blood platelet count (PLT, n=53). X-axes show timepoint relative to apheresis (d = day). Standard reference intervals for cynomolgus macaques are indicated with gray boxes. Graph displays p-value calculated by Dunn’s Kruskali-Wallis test: *p ≤ 0.05, ** p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. For some procedures, there is no data available for certain timepoints.

Apheresis collection for antigen-specific T cell analysis requires leukoreduction of Optia priming blood

Aside from HSCT, the ability to collect large numbers of leukocytes via apheresis is useful for other studies requiring large cell numbers. In particular, we aimed to use apheresis to collect large numbers of T cells from macaques for detailed epitopic analysis via intracellular cytokine staining (ICS),^11–15 which requires ex vivo stimulation of T cells with various antigens and identifying antigen-responding cells via cytokine production. When initially developing our apheresis protocol, we did not leukoreduce the heterologous ABO-matched blood used to prime the Optia, and thus foreign leukocytes from the blood donor(s) were mixed into the apheresis system. Apheresis mononuclear cell preparations from these procedures resulted in a high level of non-specific, background T cell activation in ICS, which made identifying antigen-specific activation challenging or impossible (Figure 5A). To solve this issue, priming blood was leukoreduced prior to each apheresis procedure, eliminating virtually all leukocytes (Figure 5B) and thus avoiding introduction of foreign leukocytes into the apheresis system during priming. Indeed, ICS with apheresis mononuclear cell preparations from procedures using leukoreduced priming blood resulted in minimal background T cell activation, similar to levels observed with peripheral blood mononuclear cells (PBMC) (Figure 5C).

Figure 5. Leukoreduction of heterologous priming blood reduces non-specific T cell activation in intracellular cytokine staining.

A, Flow cytometric plots display intracellular staining (ICS) of apheresis mononuclear cells (MNC) derived from an apheresis procedure where the Optia was primed with ABO-matched whole blood from a heterologous donor (no leukoreduction). Cells were either left unstimulated (“No stim”, left) or stimulated with positive control CMV lysate antigen (“CMV lysate”, right). Plots are gated on CD3+CD4+ singlets. B, Flow cytometric plots display surface CD45 staining of blood pre- (left) and post-leukoreduction (right) with Haemonetics BPF High Efficiency Leukocyte Reduction Filtration System. Plots are gated on live cells. Gates denote granulocyte (“Gran”) and lymphocyte (“Lymph”) frequencies. Box at the top of each plot shows white blood cell count (WBC, x10³ per μL) and red blood cell count (RBC, x10⁶ per μL) in each sample, as measured by complete blood count. C, Flow cytometric plots display ICS of apheresis MNC derived from an apheresis procedure where the Optia was primed with leukoreduced, ABO-matched whole blood from a heterologous donor (top) or ICS of peripheral blood mononuclear cells (PBMC) isolated from a blood draw taken prior to leukapheresis (bottom). As in panel A, cells were either left unstimulated (“No stim”, left) or stimulated with positive control CMV lysate antigen (“CMV lysate”, right). Plots are gated on CD3+CD4+ singlets.

DISCUSSION

In this report, we detail the safe apheresis of mobilized and non-mobilized cynomolgus macaques weighing 3 to 10 kilograms. While apheresis of macaques has been previously reported,^8,17–19 this is the first report describing apheresis of cynomolgus macaques and of macaques weighing as low as 3 kilograms using Spectra Optia. Indeed, these previous descriptions of macaque apheresis were insufficient to facilitate successful leukapheresis of small cynomolgus macaques. Here, we demonstrate collection of large numbers of CD34+ HSCs, including significant frequencies of CD45RA- CD90+ CD34+ cells previously demonstrated to singlehandedly provide complete multilineage engraftment in transplanted pigtailed macaques.¹⁶ Indeed, cells collected via the apheresis procedures described here have been used to successfully engraft new immune systems in transplanted cynomolgus macaques.^3,4 In addition, we demonstrate collection of large numbers of mononuclear cells from non-mobilized macaques. We found that Optia priming blood must be leukoreduced in order to eliminate undesired T cell activation in the apheresis product. This non-specific T cell activation likely represents an allogeneic response due to the introduction of foreign, priming blood-derived leukocytes into the Optia system. Elimination of allogeneic T cell activation via priming blood leukoreduction is critical for accurate interpretation of T cell activation assays, such as ICS, as well as studies of latently HIV-infected CD4+ T cells. In addition, elimination of priming blood donor leukocytes is critical for graft collection, in order to avoid introduction of non-HSCT donor leukocytes into the transplant recipient.

We found mobilization and apheresis to be well-tolerated, but associated with temporary increases in serum creatinine and decreases in hematocrit and platelet numbers, consistent with previous reports of macaque apheresis.^8,17 Platelet loss also occurs with human leukapheresis procedures^20–24. Of note, two early apheresis procedures not included here resulted in euthanasia of two macaques with prolonged and severe thrombocytopenia, likely resulting from forced collection of Optia chambers. The Optia chamber contains large numbers of contaminating platelets that are pushed out of the chamber as it fills with the desired MNCs. Thus, forced collection of a chamber before it is completely filled with MNCs results in platelet loss to the patient, and this loss is increased with each subsequent chamber collection. Thus, while filling the chamber completely can be challenging with small macaques, we recommend avoiding forced chamber collections, as well as closely monitoring platelet counts after apheresis procedures. In addition, employing a method of leukoreduction that preserves platelets in Optia priming blood may be beneficial.

We were able to perform multiple apheresis procedures on a single macaque donor, with time between procedures varying from weeks to months. Of note, for the vast majority of animals mobilized more than once (7 of 9 for which we have complete data), we observed higher numbers of CD34+ HSCs in pre-apheresis blood after the second mobilization compared to the first mobilization. However, in many cases the increase was mild, and this observation did not reach statistical significance. Of animals that underwent a third mobilization (8), we did not observe a trend toward higher pre-apheresis CD34+ blood counts for the third mobilization compared to the second mobilization.

In conclusion, we offer a detailed protocol to perform safe and effective apheresis of macaques as small as 3 kilograms. The ability to collect large numbers of leukocytes via a single apheresis procedure has numerous advantages, allowing for detailed studies of rare cell populations (e.g. HSCs, antigen-specific lymphocytes), harvest of large numbers of cells for cellular infusion therapies (e.g. transplantation, donor lymphocyte infusions, CAR T cells), and in-depth analysis of cells derived from a single timepoint.